Special Session Vena Cava Pulmonary Interventions 1.1 Vena cava filters H. P. Rousseau, V. M. Chabbert, P. P. Otal, G. P. Canevet, F. G. Joffre; Radiology, CHU Rangueil, Toulouse, France. Table of Contents

o

VENA CAVAL FILTERS.

VENA CAVAL FILTERS. Venous thromboembolic disease is a significant cause of morbidity and mortality in Occidental countries. Pulmonary embolism (PE), the most deadly form of venous thromboembolic disease, results in as many as 240 000 deaths per year in the United States. Although anticoagulation remains the primary therapy for venous thromboembolism, the implantation in the inferior vena cava (IVC) of a permanent vena cava filter (PVCF) can be regarded as a therapy modality for the prevention of pulmonary embolism in the management of venous thromboembolic disease. Since the first IVC filter in 1967, a wide range of filters with varying configurations have been developed, approved and subsequently improved in line with increased clinical application and attained knowledge. The purpose of this paper is to summarize and critique the published data on vena caval filters and thus facilitate informed clinical decisionmaking. Indications for vena caval filter placement Clearly, it is appropriate to consider placing a vena caval filter when an absolute contraindication to anticoagulation exists or when a lifethreatening complication from anticoagulation arises. Failure of anticoagulation is considered by many to be a reason for vena caval filter placement. Given their adverse effects, filters should be used judiciously for this indication. Numerous other indications for vena caval filter placement have been endorsed; however, definitive evidence of their superiority to conventional care for these situations is lacking. A number of investigators have suggested that vena caval filters should be considered as first-line therapy for DVT and PE in patients with cancer as well as prophylaxis for PE in patients after trauma and with orthopedic problems. Although a considerable amount of literature has been devoted to examining the utility of filters in these settings, none of the studies are randomized and careful follow-up is often lacking. Vena caval filters have also been endorsed for treatment of DVT and PE in patients with limited cardiopulmonary reserve, chronic obstructive pulmonary disease, postpulmonary embolectomy, after renal or cardiac transplantation, and during pregnancy. Yet, only a small number of unrandomized reports exist to support these recommendations. Because viable alternative regimens are available for all of these situations, careful assessment of the efficacy and safety of vena caval filters in these settings is necessary before firm recommendations can be made. Vena caval filters: Several different vena caval filters are available in Europe: the Greenfield filter, the titanium Greenfield filter, the bird's nest filter, the Simon nitinol filter, the Vena Tech filter, the Low Profile Vena Tech filter and the Trap Ease Filter. Greenfield filter The over-the-wire stainless steel alternating hook Greenfield filter has recently replaced the original stainless steel Greenfield filter, because its design modifications allow easier percutaneous placement. Both consist of a cone-shaped array of 6 stainless steel wires, which end in hooks that secure the device in the inferior vena cava (IVC). Titanium Greenfield filter To address concerns about venous trauma and thrombosis attributed to

the stainless steel filter's large insertion catheter (29F outer diameter), Greenfield and colleagues redesigned his filter using titanium to allow the use of a smaller introducer system (14F outer diameter). The initial model was troubled by a tendency to migrate, tilt, and perforate the inferior vena cava. Modifications of the filter limb and hook corrected these problems and this filter, the modified-hook titanium Greenfield filter, was introduced in 1991. Bird's nest filter In 1984, Roehm et al published the initial clinical investigation of the bird's nest filter. Unlike the conical design of the Greenfield filters, the bird's nest filter consists of 4 stainless steel wires, 0.18 mm in diameter and 25 cm in length, which are affixed to the vena caval wall by Vshaped struts bearing hook-like anchors. Its unique design allows it to be loaded into an 14F insertion catheter and gives it the flexibility to be placed into vena cavae up to 40 mm in diameter, the largest of any filter on the market (other filters are limited to cavae of 30 mm or less). Unfortunately, the original design of the bird's nest filter had a tendency to migrate, which resulted in several deaths. Thereafter, stiffer, wider struts were used, which greatly diminished this tendency. Simon nitinol filter The Simon nitinol vena caval filter is constructed of a nickel-titanium alloy that has thermal memory properties. At 4° to 10°C, the wires, which compose the filter, are folded into a straight and compact form fitting into a 9F catheter carrier. At body temperature, the wires are programmed to unfold into an umbrella filter composed of 7 "petals" and 6 hooked legs that anchor the device in the vena cava. Vena Tech filter The Vena Tech vena caval filter is constructed of Phynox, a unique nonparamagnetic alloy. Six struts are fused into a cone-shaped filter reminiscent of the Greenfield filter. Side rails attached to the filter cone anchor it to the vena caval wall. Low Profile, Vena Tech Filter This conical filter has thin side-rails placed along the sides of the legs. The filter is made with Phynox and placed via a 9F sheath and can be used in vessels less than 30 mm in diameter. The filter is 43 mm long. The Trapease™ filter, is a double basket symmetrical Nitinol vena cava filter, made from a single Nitinol tube. The filter has a non-expanded maximum length of 65mm and a length of 50mm when expanded to its maximum diameter of 35mm in vitro. The proximal and distal baskets, each comprised of six petal-shaped openings, are connected by 6 straight struts. A proximal and distal hook is located on each connecting strut for fixation of the filter to the vena cava wall. The filter is implanted through a 6F, 55cm long straight introducer sheath. Temporary or retrievable vena caval filters Because the long-term safety of permanent vena caval filters remains unknown and many patients have only temporary contraindications to anticoagulation, there has been considerable interest in developing effective temporary filtration devices. Several under investigation include the Tempo filter, the Gunther Tulip filter and the Optease™ filter. The cone-shaped Tempo filter is attached to a catheter, which is anchored in the subcutaneous tissue at the insertion site. It can be left in place for up to 6 weeks before removal. The Gunther Tulip filter and the Optease™ filter ( a new version of Trapease™ filter ) are permanent vena caval filters, which can be retrieved if desired in the first 10 days after implantation. If proven safe and effective, these devices could be valuable tools in the treatment and prevention of venous thromboembolic disease. Results: As a whole, the mean duration of follow-up for each filter types varies from 6 to 18 months. All are about equally effective in the prevention of pulmonary embolism (2.6%-3.8%). Deep venous thrombosis (6%-32%) and inferior vena cava thrombosis (3.6%-11.2%) after filter placement vary widely among different filter types primarily because of differences in outcome assessment. Thrombosis at the insertion site is a common complication of filter placement (23%-36%).

Until 1998, the only clinical data available on vena caval filters were derived almost exclusively from retrospective unrandomized case series. In 1998, Decousus et al published the only randomized study of vena caval filters in the prevention of PE. They randomized 400 patients using a 2 × 2 factorial design to a vena caval filter or no filter and enoxaparin or unfractionated heparin. Four different types of vena caval filters (titanium Greenfield, bird's nest, Vena Tech, and Cardial filters) were used. Vena caval filters were associated with a significant decrease in the incidence of PE compared with anticoagulation alone (1.1% versus 4.8%, P = .03) at 8 to 12 days of follow-up. After 2 years, however, this difference was no longer statistically significant although the trend still favored vena caval filters (3.4% versus 6.3%, P = .16). Symptomatic PE occurred at a similar frequency in both groups after 3 months. Fatal emboli were more common among patients treated solely with anticoagulation (0.5% versus 2.5%). In contrast, vena caval filters were associated with significantly more recurrent DVT than anticoagulation alone (20.8% versus 11.6%, P = .02). No difference in bleeding or overall mortality was documented. In light of these data, one can conclude that vena caval filters in combination with standard anticoagulation do appear to offer significantly more protection from PE than anticoagulation alone. This additional protection, however, appears to be short-lived and does not decrease overall mortality. In addition, vena caval filters are associated with a higher incidence of recurrent DVT over 2 years of follow-up. Conclusion : Undoubtedly, vena caval filters represent an important weapon in every clinician's armamentarium for the treatment of venous thromboembolic disease, but the optimal application of this technology remains to be defined. Because long-term safety and efficacy of filters remain uncertain, use of filters should be restricted to situations in which anticoagulation is clearly contraindicated. More liberal application of vena caval filters should be restricted to prospective randomized clinical trials. Temporary filtration devices in clinical testing may provide many of the advantages of permanent filters without the potential long-term side effects. References: 1. Greenfield LJ, Proctor MC. Twenty-year clinical experience with the Greenfield filter. Cardiovasc Surg. 1995;3:199-205 2. Greenfield LJ, Rutherford RB, and participants in the Vena Caval Filter Consensus Conference. Recommended reporting standards for vena caval filter placement and patient follow-up. J Vasc Interv Radiol. 1999;10:1013-1019 3. Cho KJ, Greenfield LJ, Proctor MC, et al. Evaluation of a new percutaneous stainless steel Greenfield filter. J Vasc Interv Radiol. 1997;8:181187 4. Crochet DP, Stora O, Ferry D, et al. Vena Tech-LGM filter: long-term results of a prospective study. Radiology. 1993;188:857-860 5. Katsamouris AA, Waltman AC, Delichatsios MA, Athanasoulis CA. Inferior vena cava filters: in vitro comparison of clot trapping and flow dynamics. Radiology. 1988;166:361-366 6. Teitelbaum GP, Jones DL, van Breda A, et al. Vena caval filter splaying: potential complication of use of the titanium Greenfield filter. Radiology. 1989;173:809-814 7. Greenfield LJ, Cho KJ, Proctor MC, et al. Results of a multicenter study of the modified-hook titanium Greenfield filter. J Vasc Surg. 1991;14:253-257 8. Mohan CR, Hoballah JJ, Sharp WJ, Kresowik TF, Lu CT, Corson JD. Comparative efficacy and complications of vena caval filters. J Vasc Surg. 1995;21:235-245 9. Ferris EJ, McCowan TC, Carver DK, McFarland D. Percutaneous inferior vena caval filters: follow-up of seven designs in 320 patients. Radiology. 1993;188:851-856 10. Wittenberg G, Kueppers V, Tschammler A, et al. Long-term results of vena cava filters: experiences with the LGM and the titanium Greenfield devices. Cardiovasc Interv Radiol. 1998;21:225-229. 11. Simon M, Rabkin DJ, Kleshinski S, Kim D, Ransil BJ. Comparative evaluation of clinically available inferior vena cava filters with an in

vitro physiologic simulation of the vena cava. Radiology. 1993;189:769774 12. Roehm JOJ. The bird's nest filter: a new percutaneous transcatheter inferior vena cava filter. J Vasc Surg. 1984;1:498-501 13. Dorfman GS. Percutaneous inferior vena caval filters. Radiology. 1990;174:987-992. 14. Wojtowycz MM, Stoehr T, Crummy AB, McDermott JC, Sproat IA. The bird's nest inferior vena caval filter: review of a single-center experience. J Vasc Interv Radiol. 1997;8:171-179 15. Poletti PA, Becker CD, Prina L, et al. Long-term results of the Simon nitinol inferior vena cava filter. Eur Radiol. 1998;8:289-294 16. Ricco JB, Crochet D, Sebilotte P, et al. Percutaneous transvenous caval interruption with the LGM filter: early results of a multicenter trial. Ann Vasc Surg. 1988;3:242-247. 17. Crochet DP, Brunel P, Trogrlic S, Grossetete R, Auget JL, Dary C. Long-term follow-up of Vena Tech-LGM filter: predictors and frequency of caval occlusion. J Vasc Interv Radiol. 1999;10:137-142 18. Decousus H, Leizorovicz A, Parent F, et al. A clinical trial of vena caval filters in the prevention of pulmonary embolism in patients with proximal deep-vein thrombosis. Prevention du Risque d'Embolie Pulmonaire par Interruption Cave Study Group. N Engl J Med. 1998;338:409415 19. Millward SF. Temporary and retrievable inferior vena cava filters: current status. J Vasc Interv Radiol. 1998;9:381-387 20. Bovyn G, Gory P, Reynaud P, Ricco JB. The Tempofilter: a multicenter study of a new temporary caval filter implantable for up to six weeks. Ann Vasc Surg. 1997;11:520-528 21. Rousseau H, Perreault P, Otal P, Stockx L, Golzarian J, Oliva V, Reynaud P, Raat F, Szatmari F, Santoro G, Emanuelli G, Nonent M, Hoogeveen Y. The 6-F nitinol TrapEase inferior vena cava filter: results of a prospective multicenter trial. J Vasc Interv Radiol 2001 Mar;12(3):299304

1.2 Interventions in central venous occlusion C. Zollikofer; Institut für Radiologie, Kantonsspital Winterthur, Winterthur, Switzerland. Introduction Central venous obstruction is caused by malignancies due to primary or secondary tumors of the mediastinum and retroperitoneum in 95 % of the cases. Only a minority of about 5 % are of benign nature and most commonly the sequellae of iatrogenic origin such as central venous lines, hemodialysis access fistulas, hematoma formation after surgery or percutaneous interventions and finally secondary to thrombotic occlusion other than central venous lines as in acute deep veinthrombosis, post irradiation fibrosis, trauma, mediastinitis and benign retroperitoneal fibrosis. Clinically central venous obstruction manifests itself by the so called superior vena cava syndrome (SVCS) if the superior vena cava or the brachiocephalic veins are involved. The syndrome comprises severe swelling and edema of the head, neck and the upper extremities, dyspnea, orthopnea and cyanosis, headache conjunctival edema and blurred vision. In up to 80% the SVCS is caused by bronchogenic carcinoma with mediastinal lymphnode involvement or by post irradiation fibrosis. The incidence of SVCS in bronchogenic carcinoma is reported to be from 3 to 15 %. An acute onset usually signifies a malignant nature due to rapid tumor progression or additional acute thrombosis of the veins peripheral to the obstruction. Unilateral obstruction of benign nature is usually due to brachiocephalic vein stenosis or occlusion secondary to central venous lines or hemodialysis access fistulas. The counterpart for the lower extremities is the inferior cava syndrome (IVCS), which in the majority of cases is caused by retroperitoneal lymphnode involvement in malignant disease with compression of the pelvic veins or IVC. The symptoms of IVCS are characterized by massive leg swelling and may be accompanied by venous claudication and edema of the genitals. In some cases the intrahepatic segment of the IVC may be compressed by metastatic enlargement of the liver, particularly the caudate lobe. IVC obstruction secondary to a hypertrophied caudate lobe is also seen in Budd-Chiari syndrome. Finally IVCS may be caused by other benign disease such as post irradiation fibrosis and retroperitoneal fibrosis. Not infrequently acute or chronic venous obstruction in the pelvic veins (often related to a pelvic venous spur, MayThurner syndrome) is a cause for significant swelling of the lower extremities.

Indications and contraindications for venous stenting Generally speaking, all lesions which cannot be satisfactorily dilated by balloon angioplasty alone may be treated by implantation of a stent. Thus obstruction by continuous extrinsic compression (i.e. malignant tumours, fibrous-inflammatory processes, scaring and shrinkage after surgery, percutaneous interventions or irradiation, and intraluminal blockage (e.g. old thrombus, intimal reaction, venous spurs and webs) are all lesions which have been treated successfully by stenting. There are practically no absolute contraindications to the method other than those also applicable to balloon angioplasty. Lesions of the SVC and IVC close to the heart where stents would reach the right atrium should be evaluated very carefully and may represent a contraindication to stenting because of an increased stent risk of migration or/and arrhythmia caused by the stents. Relative contraindications are coagulation disorders with hypercoagulability, and low flow situations related to a stented venous segment, both of which carry an increased risk of thrombotic occlusion. Cross-stenting of large venous branches does not represent an absolute contraindication since in the majority of cases such side branches remain patent. In our experience cross-stenting of the innominate and azygos veins in SVC obstruction or of the internal ilicac vein in pelvic vein stenting has had no adverse effect. All the same, it seems advisable to avoid cross-stenting of large branching collateral veins and particularly the renal veins if it is technically feasible to do so. Characteristics and selection of stents Only general guidelines can be given as to which stent is best suited for use in venous obstruction. Some important criteria are flexibility, maximal stent diameter, radial force and foreshortening of the stent. Felxibility and self-expansion have the advantage that they allow the stent to adapt to the natural course of the vessel (brachiocephalic veins) and compensate to some degree for differences in caliber along its course. In areas of movement (e.g. across a joint such as the hip or shoulder) it is clearly preferable to use a flexible stent. To date, the most widely used stents in the venous system have been the Gianturco or GianturcoRoesch stents and the Wallstent. But also balloon expandable stents such as the Palmaz stent and others have been successfully used for venous stenting. So far we have preferred the Wallstent for most venous lesions because it combines the advantageous features of a small diameter introducing instrument, a large stent lumen, high flexibility and adequate expansile force. A potential drawback of the Wallstent is its foreshortening during release and later expansion, but with some experience and choosing long enough stents, this is usually not a major problem. Apart from these objective criteria and the need for a flexible stent when crossing a large joint, the choice of the stent is largely a matter of personal experience or preference. So far, no significant difference in the clinical results obtained with different stent designs have been published. Adjuntive medication During all stent implantations 5’000 i.u. of heparin are given intravenously followed by heparinization or low molecular heparin for 3 to 7 days. If there is no contraindication, overlapping anticoagulation with coumarin is used for 6 months usually in benign disease (with the exception of hemodialysis acess fistulas) and for life term in patients with malignancies to prevent thrombosis secondary to paraneoplastic syndromes. Analgo-sedation may be necessary as the dilatation of both tumour stenosis and hemodialysis fistula stenosis may be painful. Stenting in malignant venous obstructin of the cava, the brachiocephalic veins and the pelvic veins In malignant disease the classical treatment for vena cava syndrome has been radiation therapy and/or chemotherapy. However, often the patients present with advanced metastatic disease where the conventional therapy is insufficient or shows only a slow regression. Though chemotherapy in small cell lung cancer has a high response of up to 80 %, relapse is frequent and non small cell lung cancer (NSCLC) has a comparatively low response rate of 15 % to 30 %. Also radiation therapy used in NSCLC has only a 15 % to 25 % complete response rate. Since many patients will already have had maximum tolerance radiation therapy or chemotherapy, the only available treatment then is medically with steroids and diuretics which, however, often show an insufficient effect. Surgical intervention with a venous Bypass-graft may be performed but this involves major surgery and therapeutic results seem disproportional to the burden for the patient who may be in a preterminal condition where only a minimally invasive intervention is justifiable. For this reason a low risk intervention using percutaneously implanted endovascular stents offers a valuable alternative therapy to rapidly treat

this severely distressed patients. The first two successful cases of palliative stenting of the SVC and IVC were reported in 1986 by Charnsangavey et al. In the last 15 years several hundred cases of palliative stenting in malignant vena cava syndrome have been reported. In most reports self-expandable stents of the Gianturco-Z and Gianturco-Roesch type or Wallstents were used. Indications, technique and results The indication for stenting of malignant vena cava syndrome is the presence of siginificant obstruction in the innominate veins, the SVC, the IVC or the pelvic veins. Obstruction by intraluminal tumour may be a contraindication but more so for the Gianturco-Z-stent, than for the Wallstent or other stents with a relatively tight mesh. If tumour infiltration into the vessel is demonstrated the creation of a tighter mesh with 2 or 3 overlaying stents may be considered to prevent tumour ingrowth. Because the latter phenomenon is relatively rare (approx. 5 %) the routine use of covered stents seems not warranted. Before stenting adequate imaging for tumour extension using CT and a phlebographic study of the obstructed vein should be obtained to exactly evaluate the inand outflow, the collaterals and the length and diameter of the stent to be used. In order not to disturb peripheral inflow we routinely use a femoral approach under local anesthesia with an introducing sheath for the IVC as well as for stenting of the SVC and brachiocephalic veins. Occasionally, however, and particularly for recanalisation of chronic occlusions of the brachiocephalic or pelvic veins an additional approach via an arm vein or the internal jugular vein using the so called pullthrough-technique is required. If acute thrombosis proximal to the occlusion is present we perform local thrombolysis before stent implantation to guarantee good inflow and to minimize the risk of embolism or stent occlusion. As soon as the obstruction is passed using steerable guidewires and angiographic catheters, we perform a predilatation with a balloon catheter. This allows easier passage of the introducing instrument and also an estimation of how easily the stent will open the obstruction. After stent implantation additional balloon dilatation is performed if we find that the stent is not adequately expanding spontaneously when using self-expandable stents. For the SVC stent diameters of 14 to 16mm and 10 to 14mm for the brachiocephalic veins are sufficient. Larger stents of 20 to 30mm diameter are required for the IVC and usually 12 to 16mm diameter stents are adequate for the common femoral and pelvic veins. A normal inflow and outflow of the stented segment has always to be ensured after stenting to minimize the risk of acute thrombotic occlusion. Significant relief after stenting can be expected in 70 to 100% of the patients. Particularly shortness of breath, cyanosis and the feeling of pressure in the head and neck is relived within hours, whereas edema of the head, neck and extremities usually takes one to two days ( in the lower extremities several days).The question, whether bilateral stenting is necessary if both innominate veins are obstructed depends on the extend of obstruction and functional collaterals. If both internal jugular veins, which serve as the major collaterals from one side to the other are patent, then usually unilateral stenting – as a rule the right side or SVC alone– is sufficient. Long-term palliation including secondary procedures usually lies in the range of 85 % or more, however, due to tumour progression additional stents may have to be implanted. Severe procedure related complications such as stent breackage, cardiac arrhythmias or stent migration are rare ( 0 to 10 %) and early thrombotic stent occlusion (5 to 14 %) may be treated with thrombolysis as long as an adequate in- and outflow is guaranteed. We believe it is important to use anticoagulation after stent insertion in malignancies. This is not only because of the increased risk of thrombosis in tumour patients who may develop paraneoplastic syndromes, but because the neo-endothelialisation process that normally operates over the stent surface seems to be delayed or absent in patients having had radiation and/or chemotherapy as evidence by the findings in animal experiments as well as in two of our own patients at autopsy. Stents in benign obstruction of the venae cavae and large veins unrelated hemodialysis access fistula The most frequent causes for benign obstruction of the superior vena cava and brachiocephalic veins are mediastinal fibrosis and post-thombotic sequallae of central venous catheters. In IVCS retroperitoneal fibrosis and stenosis of the intrahepatic segment of the IVC in BuddChiari syndrome are to be mentioned. Obstruction of the large femoral veins and the pelvic veins are usually sencondary to deep vein thrombosis, surgery and trauma. The May-Thurner syndrome or the so called venous spur is a further not infrequent cause for pelvic vein stenosis, usually on the left and preferably involving the female sex. Indications, techniques and results

In principle venous obstruction secondary to all of the above mentioned etiologies may be treated by stent placement because surgery, such as venovenous bypass operations or xeno-grafts usually carry a significant risk or have unsatisfactory long-term results. Post-thrombotic stenosis in pelvic veins including those caused by a venous spur in May-Thurner syndrome as well as post-surgical and post-interventional obstruction in the brachiocephalic, large peripheral and pelvic veins have been treated with great success in the past. Also benign stenosis of the SVC and IVC including congenital anomalies (webs) or obstructions of the vena cava and hepatic veins related to the Budd-Chiari syndrome may be successfully stented. As in malignant stenosis a preprocedural evaluation with CT and phlebography is mandatory for the differential diagnosis and choosing the best approach. Again a femoral approach is usually employed for stenting but an internal jugular approach for SVC or pelvic lesions is sometimes necessary for very tight or chronic occlusions. For the brachiocephalic vessels or the pelvic veins a double approach from the arm and the femoral veins with the so called pullthrough-method to stabilize the guidewire from both ends may be necessary. For chronic thombotic occlusions the use of thrombolysis with urokinase has not been of any benefit in our experience, however, recanalisation with a steerable hydrophilic guidewire may be possible even in longstanding occlusions. If lesions across joints, particularly the hip joint have to be treated a self-expandable stent with good longitudinal flexibility has to be used. In the event of additional local thrombosis close to the obstruction thrombolysis has to be initated before stenting. In extensive thrombosis or peripheral post-thrombotic syndrome however, where no adequate inflow can be established stenting is not indicated. A 90 to 100 % overall initial clinical success can be expected with a long-term primary stent patency ranging from 67 to 90 % and a secondary patency between 86 and 100 %. Clinical long-term success varies from 80 to 100 %. Particularly in patients with deep venous thrombosis caused by May-Thurner syndrome stenting after surgical or endovascular recanalisation has a high clinical success and a long-term patency of the 85 to 100 %. In our own patient material comprising 18 patients with May-Thurner syndrome in 10 and iliofemoral obstruction due to iatrogenic interventions in the other 8, there is a 100 % primary patency after a mean of follow-up of 7 years. It seems particularly noteworthy that treatment of common femoral vein stenosis with a flexible stent crossing the hip joint – a critical region for surgical intervention – seems to have excellent long-term results. As long as benign lesions can be passed with a guidewire and a sufficient lumen after stenting (which may be tested by probatory balloon dilatation) is established stent placement in benign disease is highly successful with an extremely low complication rate. Hemodialysis access fistulae related stenosis of the brachiocephalic veins Hemodialysis access fistula is known to cause stenosis or occlusion of the brachiocephalic veins in approximately 21 %. These central lesions tend typically to occur at sites of physiological narrowing and turbulence in the axilliary, subclavian and innominate veins. Previous vascular damage due to catheterization, flow turbulence through valves and abnormal function of platelets damaged by hemodialysis are all factors that have been considered as playing a possible role in the etiology of these lesions. As simple balloon dilatation usually is insufficient or hampered by early recurrence, there is a general agreement that all these lesions should be treated by primary stenting. Again because of the movement of the shoulder and possible compression between the clavicle and the first rib flexible, self-expandable stents should be used in this region. Depending on the location and severity of the obstruction the lesions may be approached from a femoral and/or peripheral approach through the venous limb of the shunt. However, we prefer again a femoral approach if possible in order not to disturb the central flow of the shunt by manual compression after the procedure. Results In contrast to peripheral lesions recurrencies due to intimal hyperplasia usually occur later i.e. in the 8 to 12 month’s period versus 3 to 6 months in the periphery. Initial success with primary stent placement is definitely improved compared to simple balloon dilatation where recurrent obstruction is seen in almost 75 % within 6 months (Glanz et al.1988, Beathard et al. 1992). Furthermore recurrencies are delayed. Close surveillance of the stented veins however is mandatory to early detect and treat reobstruction. Any increase in venous pressure at hemodilaysis should lead to prompt evaluation by Doppler ultrasound and/or a shuntogram. It is also important to check the entire venous outflow tract since new lesions may develop at some distance from the

original stent. In most cases once a stent is implanted, repeat dilatation is facilitated by the fact that an implant is in place and can be easily traversed and dilated even in the presence of a complete obstruction. Secondary patencies with prompt revision helps to achieve excellent long-term stent function with patencies reported of 78 and 86 % at one year (Vorwerk et al. 1995, Quinn et al. 1995) and of 77 % and 86 % at two years (Vorwerk et al. 1995, author’s results). Conclusions Stenting of malignant venous obstruction is a low risk procedure and highly efficient with rapid relief of the symptoms in these severely ill patients. In symptomatic patients stent placement should be considered as the method of choice and performed without undue delay to prevent potential additional thrombotic occlusion of the inflowing veins, a complication that makes stent placement much more difficult. Stent procedures for benign disease have a very high success rate and excellent long-term clinical results. Morbidity and complication rates are extremely low. Particularly for treatment of the May-Thurner syndrome and iatrogenic lesions of the large veins stents are the method of choice. Central venous lesions related to hemodialysis access fistulas are the third indication for primary stenting to improve the rather dismal results of simple balloon dilatation. A flexible, self-expanding stent should be used for the brachiocephalic veins. Because of the high rate of intimal hyperplasia usually occurring after 8 to 12 months a close shunt surveillance is mandatory with immediate evaluation if venous pressures are rising. Adequate results with re-dilatation are the rule. References: 1. Bell DR, Woods RL, Levi JA. Superior vena caval obstruction: a 10year experience. Med J Aust 1986; 145: 566-568. 2. Nieto AF, Doty DB. Superior vena cava obstruction: clinical syndrome, etiology and treatment. Curr Probl Cancer 1986; 10: 441-484. 3. Irving ID, Dondelinger RF, Reidy JF et al. Gianturco self-expanding stents: clinical experience in the vena cava and large veins. Cardiovasc Intervent Radiol 1992; 15: 328-333. 4. Rösch J, Uchida BT, Hall LD et al. Gianturco-Rösch expandable Zstents in the treatment of superior vena cava syndrome. Cardiovasc Intervent Radiol 1992; 15: 319-327. 5. Zollikofer CH, Schoch E, Stuckmann G, Espinosa N, Stiefel M. Perkutane transluminale Behandlung von Stenosen und Verschlüssen im venösen System mittels Gefässendoprothesen (Stents). Schweiz Med Wochenschr 1994; 124: 995-1009. 6. Elson JD, Becker GJ, Wholey MH et al. Vena caval and central venous stenoses: management with Palmaz balloon-expandable intraluminal stents. JVIR 1991; 2: 215-223. 7. Lokich JJ, Goodman R. Superior vena caval syndrome: clinical management. JAMA 1975; 231: 58-61. 8. Davenport D, Ferrec C, Blake D. radiation therapy in treatment of superior vena caval obstruction. Cancer 1978; 42: 2600-2603. 9. Levitt SH, Jona TK, Kilpatrick SJ et al. Treatment of malignant superior vena caval obstruction. Cancer 1969; 24: 447-452. 10. Perez CA, Presant CA, Van Amburg AL. Management of superior vena cava syndrome. Semin Oncol 1978; 5: 123-124. 11. Doty DB. Bypass of superior vena cava: six years experience with spiral vein graft for obstruction of superior vena cava due to benign or malignant disease. J Thorac Cardiovasc Surg 1982; 83: 326-338. 12. Bernstein EF, Knowles J, Saeed M. Should superior vena caval syndrome be treated by surgery any more ? Cardiovascular surgery 1994; Vol2 No5: 605-606. 13. Charnsangavej CH, Carrasco CH, Wallace S et al. Stenosis of the vena cava: preliminary assessment of treatment with expandable metallic stents. Radiology 1986; 161: 295-298. 14. Nicholson A, Duncan F, Ettles F, Arnold A, Greenstone M et al. Treatment of malignant superior vena cava obstruction: metal stents or radiation therapy. SCBIR 1997; 8: 781-788. 15. Kee ST, Kinoshita L, Mahmood K et al. Superior vena cava syndrome: treatment with catheter-directed thrombolysis and endovascular stent placement. Radiology 1998; 206: 187-193. 16. Thony F, Moro D, Witmeyer S et al. Endovascular treatment of superior vena cava obstruction in patients with malignancies. Eur Radiol 1999; 9:965-971. 17. Petersen BD, Uchida BT. Long-term results of treatment of benign central venous obstructions unrelated to dialysis with expandable Z stents. JVIR 1999; 10: 757-766. 18. Lopez RR, Benner KG, Hall L et al. Expandable venous stents for treatment of the Budd-Chiari syndrome. Gastroenterology 1991; 100: 1435-1441.

19. Binkert CA, Schoch E, Stuckmann G et al. Treatment of pelvic venous spur with self-expanding metallic endoprostheses. Cardiovasc Intervent Radiol 1998; 21: 22-26. 20. O’Sullivan GJ, Semba CP, Bittner CA et al. Endovascular management of iliac vein compression (May-Thurner) syndrome. JVIR 2000; 11: 823-836. 21. Barett N, Spencer S, McIvor J et al. Subclavian stenosis: a major complication of subclavian dialysis catheters. Nephrol Dial Transplant 1988; 3:423-425. 22. Glanz S, Gordon DH, Lipkowiz GS et al. Axillary and subclavian vein stenosis: percutaneous angioplasty. Radiology 1988; 168: 371-373. 23. Glanz S, Gordon DH, Butt KMH et al. Stenotic lesions in dialysisaccess fistulas: treatment by transluminal angioplasty using high-pressure balloons. Radiology 1985; 156: 236. 24. Quinn S, Schuman E, Demlow T et al. Percutaneous transluminal angioplasty versus endovascular stent placement in the treatment of venous stenoses in patients undergoing hemodialysis: intermediate results. JVIR 1995; 6: 851-855. 25. Vorwerk D, Guenther RW, Mann H et al. Venous stenosis and occlusion in hemodialysis shunts: follow-up results of stent placement in 65 patients. Radiology 1995; 195: 140-146. 26. Turmel-Rodrigues LA, Blanchard D, Pengloan J et al. Wallstents and craggstents in hemodialysis grafts and fistulas: results for selective indications. JVIR 1997; 8: 975-982.

1.3 Interventions in Pulmonary Embolism T. Schmitz-Rode; Diagnostic Radiology, University Hospital, Experimental Diagnostic and Interventional Radiology, Aachen, Germany. Patients with massive pulmonary embolism are at serious risk of death due to right ventricular failure. Survival depends on rapid recanalisation of the pulmonary arterial occlusion and reduction of the right ventricular afterload. In severe cases, even high-dose thrombolytic therapy may fail to prevent a fatal outcome. Percutaneous catheter treatment represents an therapeutic option for these patients. Catheter devices include standard diagnostic catheters, balloon catheters, dedicated suction catheters (Greenfield vacuum cup catheter), rotating pigtail catheter, hydrodynamic thrombectomy catheter, and high-speed rotating tip catheters. The principles employed are extraction and/or fragmentation of the embolic material. Fragmentation of central emboli and dislocation of the fragments to the periphery result in a relative gain of non-obstructed cross-sectional area. The increased total surface area of the fragments accelerates the efficacy of an accompanying thrombolysis or of spontaneous intrinsic lytic activity. Pulmonary placement and navigation of interventional catheter devices depend on different designs as well as modes of mechanical action for recanalisation. The indications, safety, efficacy, the influence on hemodynamic parameters, possible complications, and pitfalls will be discussed. Catheter fragmentation provides for an average recanalisation of about one third of the pulmonary embolic occlusion and is especially useful in high-risk patients to accelerate thrombolysis and as an alternative to surgical embolectomy. References: 1. Kasper W, Konstantinides S, Geibel A et al. Management strategies and determinants of outcome in acute major pulmonary embolism: results of a multicenter registry. J Am Coll Cardiol 1997;30:1165-71 2. Gulba DC, Schmid C, Borst HG, et al. Medical compared with surgical treatment for massive pulmonary embolism. Lancet 1994;343:576-77 3. Brady AJB, Crake T, Oakley CM. Percutaneous catheter fragmentation and distal dispersion of proximal pulmonary embolus. Lancet 1991;338:1186-89 4. Essop MR, Middlemost S, Skoularigis J, et al. Simul-taneous mechanical clot fragmentation and pharmacologic thrombolysis in acute massive pulmonary embolism. Am J Cardiol 1992;69:427-30 5. Fava M, Loyola S, Flores P, et al. Mechanical fragmentation and pharmacologic thrombolysis in massive pulmonary embolism. J Vasc Interv Rad 1997;8:261-66 6. Murphy JM, Mulvihill N, Mulcahy D, et al. Percutaneous catheter and guidewire fragmentation with local administration of recombinant tissue plasminogen activator as a treatment for massive pulmonary embolism. Eur Radiol 1999;9:959-64

7. Schmitz-Rode T, Janssens U, Duda SH, Erley CM, Günther RW. Massive pulmonary embolism: percutaneous emergency treatment by pigtail rotation catheter. J Am Coll Cardiol 2000;36:375-380 8. Koning R, Cribier A, Gerber L, Eltchaninoff H, Tron C, Gupta V, Soyer R, Letac B. A new treatment for severe pulmonary embolism: percutaneous rheolytic thrombectomy. Circulation 1997;96:2498-2500 9. Uflacker R, Strange C, Vujic I. Massive pulmonary embolism: preliminary results of treatment with the Amplatz thrombectomy device. J Vasc Interv Radiol 1996;7:519-528 10. Rocek M, Peregrin J, Velimsky T. Mechanical thrombectomy of massive pulmonary embolism using an Arrow-Trerotola percutaneous thrombolytic device. Eur Radiol 1998;8:1683-1685 11. Schmitz-Rode T, Günther RW, Pfeffer JG, et al. Acute massive pulmonary embolism: use of a rotatable pigtail catheter for diagnosis and fragmentation therapy. Radiology 1995;197:157-62 12. Schmitz-Rode T, Janssens U, Schild HH, et al. Fragmentation of massive pulmonary embolism using a pigtail rotation catheter. Chest 1998;114:1427-1436 13. Singhal S, Henderson R, Horsfield K, et al. Morphometry of the human pulmonary arterial tree. Circulation Research 1973;33:190-97 14. Goldhaber SZ. Integration of catheter thrombectomy into our armamentarium to treat acute pulmonary embolism. Chest 1998;114:123738 15. Goldhaber SZ, Kessler CM, Heit JA, et al. Recombinant tissue-type plasminogen activator versus a novel dosing regimen of urokinase in acute pulmo-nary embolism: a randomized controlled multicenter trial. J Am Coll Cardiol 1992;20:24-30 16. Meyer G, Sors H, Charbonnier B, et al. Effects of intravenous urokinase versus alteplase on total pulmonary resistance in acute massive pulmonary embolism: a European multicenter double-blind trial. J Am Coll Cardiol 1992;19:239-45 17. Meneveau N, Schiele F, Vuillemenot A, et al. Streptokinase vs alteplase in massive pulmonary embolism. Eur Heart J 1997;18:1141-48 18. Gray HH, Miller GAH, Paneth M. Pulmonary embolectomy: its place in the management of pulmonary embolism. Lancet 1988;1:1441-45 19. Uflacker R. Interventional therapy for pulmonary embolism. J Vasc Interv Radiol. 2001 Feb;12:147-64.

Special Session Femoropopliteal PTA / Stenting 2.3 Mechanical Recanalization Devices S. Müller-Hülsbeck1, H. J. Steinkamp2, T. Jahnke1; 1Radiology, Universityhospital, Kiel, Germany, 2Radiology, Universityhospital - Campus Virchow, Berlin, Germany. Various mechanical recanalization devices have been introduced for the treatment of either acute or chronic arterial occlusions. The therapeutic options and approaches to short- or long-distant occlusions as mentioned above are usually completely different; however one well known first step of interventional treatment is the same for both: crossing the lesion with a guide wire to place the distal end of the wire beyond the occluded segment. The decision for different treatment modalities after successful guide wire passage is dependent from the time range between occurrence of occlusion often obvious by clinical symptoms and interventional treatment. By means the device has to be more aggressive and powerful with increased time of occlusion. Success for the treamtment of acute and subacute thromboembolic infrainguinal occlusions is usually given when using mechanical thrombectomy devices in time, that means the duration of occlusion should not exceed 14 days (1). But what kind of modalities are available if much more older and chronic occlusions have to be treated. The lecture tries to give an answer to possible approaches for Mechanical Recanalization Devices to treat acute and chronic arterial infraaortal occlusions. · Acute occlusions (estimated time of occlusion less than 14 days) Various devices have been introduced for the purpose of percutaneous mechanical thrombectomy (PMT). These devices should concur with the gold-standard surgical method, the Fogarty balloon embolectomy and with local fibrinolysis therapy. The latter therapeutic option is usually deemed too time-consuming (2), whereas the surgical intervention might cause severe intimal vessel wall damage (3). Currently available PMT devices achieve a mechanical recanalization due to hydrodynamic principles of Venturi and Bernoulli [AngioJet™ rheolytic thrombectomy

catheter (4, 5), Hydrolyser™ catheter (6), Oasis™ catheter (7)], highspeed rotation of an impeller [Amplatz Thrombectomy device™ and Rotarex™, 8-12] or ultrasound energy. The major problem of all currently available PMT devices is the fact, that in at least in a quarter of applications additional techniques for the removal of residual thrombus are needed. Mechanical thrombectomy offers the following advantages over chemical thrombolysis: (a) the procedure requires only a few minutes; (b) typically, only a single catheterization session is required to treat the thrombus, compared in multiple examinations performed during chemical thrombolysis; (c) monitoring of the patients in an intensive care unit for several days is not necessary; (d) expensive, repeated laboratory tests to control the fibrinolytic effect are not required; (e) complications of chemical thrombolysis such as external and internal bleeding, pericatheter thrombosis and allergic reaction are uncommon with mechanical thrombectomy (4). Knowing all these facts, Fibrinolysis is still the standard percutaneous procedure for management of acute arterial occlusions. Treatment time ranges from several hours to 2 days (13). Local thrombolysis is successful in 80%-92% of arterial occlusions, for treatment of native arteries, this may be as high as 91% (14). Even the TOPAS data have not really changed the sale success of mechanical devices on the huge market of interventional radiology. The TOPAS trial compared the effectiveness of local thrombolysis with urokinase and vascular surgery as the initial treatment for acute arterial occlusions of lower limbs in 272 patients per group (15). For a mean follow-up period of 12 months a mortality rate / amputation free survival rate of 20% / 70% for urokinase and 17% / 65% for surgery was found at 1 year. Our data on PMT with the RTC device revealed a lower mortality rate of 10% and a higher amputation free survival rate of 80%. It should be mentioned, however, that comparisons of the TOPAS trial with other devices for mechanical thrombectomy or thrombolysis must be tempered by the differences in study design and patients characteristics. Ultimately, the only way to answer these questions is to conduct a randomized trial for different devices. · Chronic occlusions Additional techniques are of outmost importance when a chronic i.e. femoro-politeal occlusion was successfully recanalized with a guide wire. The techniques are similar and often identical to these used after successful PMT to handle the underlying lesion, the latter can be either treated with PTA alone or with direct stent implantation. But before switching to PTA and stenting which are widespread accepted, only two mechanical recanalization devices are currently available: the highspeed rotation Rotarex™ device, known for the successful treatment of acute and subacute occlusions and laser devices (10). Fact is that both techniques are not widespread in the huge group of interventional radiologists, but they more often used by angiologists and cardiologists. Radiologists prefer direct stenting techniques, either balloon-expandable or self-expanding stents to treat chronic occlusions (16-19), mechanical techniques using rotational ablation devices have been abandoned (20). The principle of the high-speed rotation Rotarex™ device will be explained more detailed as follows because an application for either acute and even chronic occlusion seems theoretically feasible: The Straub Rotarex catheter (Straub Medical, Wangs, Switzerland) consists of a coated stainless steel spiral which glides over an .020inch guide wire and runs inside along the whole length of the 8F outer catheter. The spiral is fixed distally to the catheter head and proximally to a motor drive. The catheter head consists of two cylinders fitting over each other, the outer rotating cylinder is attached to the spiral, the stationary inner one to the catheter shaft. At the tip of the 8F catheter, the spiral communicates with the vessel lumen through two oval slits. When the catheter and motor drive are connected by a magnetic clutch, the motor rotates the spiral at 40,000r/min, resulting in 80,000cuts/min at the catheter head (negative pressure exerted: 5.8kPa = 43.5mmHg). The occlusion material is sucked through the cutting slits and the fragments are transported by the spiral to the proximal sideport and discharged into a plastic bag. The size of fragmented particles is supposed to range between 100µm and 500µm. A control unit regulates the speed of the motor and increases or decreases the power steered to the motor to obtain an optimum rotation speed of the catheter. The momentary rotational speed of the catheter head is indicated optically and acoustically to avoid overload. Berczi et al. used the rotarex device in 18 patients with acute or subacute occlusions of the superficial femoral and popliteal artery. Primary procedural success including additional procedures such as angioplasty and/or stent-graft placement was 94%. The complication rate was

31.5% with two cases of perforations, one arteriovenous fistula and three distal embolizations, however, complications were treated interventionally and no surgical measures were necessary. The authors concluded that the Rotarex thrombectomy catheter is effective in treating acute and subacute occlusions of the superficial femoral and popliteal arteries but should not be used in arteries with heavily calcified plaques because of the risk of perforation (11). In a study by Zeller and coworkers the device was evaluated in a series of 28 patients, target lesions were mostly located in superficial femoral or popliteal arteries, but also one aortic-femoral bypass (in crossover technique), one common iliac artery, one external iliac artery and one common femoral artery were treated. Primary success rate was 90%. Again the complication rate was fairly high with 32% including 5 perforations, 3 cases of embolization (after PTA), one wire-induced dissection and one retroperitoneal bleeding. It was concluded that the device is a useful tool not only for the initial treatment of acute or subacute occlusions but also for neointima-related in-stent occlusions, which was the target lesion type in 15 cases of this study. The main complications were embolizations and vessel perforations, however, these were treated successfully by interventional techniques (12). While the clinical application of the Straub Rotarex™ device shows high initial technical and anatomical success, only short-term follow up is available. The first multicenter study which was presented by Jäger et al. involved 60 patients and yielded 6-months reobstruction rates of 57% (13). When laser angioplasty is used, an eximer laser seems to be the tool of choice. In a prospective trial in 94 cases with occlusion of the SFA and formerly unsuccessful conventional percutaneous transluminal angioplasty, laser angioplasty for recanalization was applied and demonstrated a successful recanalization of the SFA in 76/94 patients (80.9%). After a follow-up period of 12 months primary, primary-assisted and secondary patency rates were 50.0%, 65.8% and 73.7%, respectively. However, excimer laser recanalization is followed by balloon angioplasty with optional stenting in the superficial femoral, popliteal, and/or tibial arteries. Complex endovascular treatment combining laser debulking and angioplasty/stenting offers an alternative for patients with critical limb ischemia who lack good surgical options (21-24). It was shown that is was a successful therapeutic option for patients with femoral artery occlusion or high-grade stenosis after Angio-Seal application (25). What is proven so far for treatment of acute and chronic infra-aortic arterial occlusions? 1. Mechanical recananlization devices are a competitive and alternative minimal invasive treatment modality for restoring flow in acute infraaortal arterial thromboembolic occlusions. 2. Mechanical recananlization devices can concur either with the surgical technique using the Forgaty-Balloon and the local admission of fibrinolytic drugs. 3. Mechanical recananlization devices even may be suitable for treatment of chronic occlusions, however the treatment of choice seems still balloon angioplasty and direct stenting. Unfortunately, mechanical recananlization devices are expensive. The price and the effectiveness increases significantly with technical design of the device used. The physicians need some technical, device specific know-how when using mechanical recananlization devices, therefore often local fibrinolysis therapy is performed during the night or at the weekend for treatment of acute occlusions; for the treatment of chronic occlusions the treatment of choice seems still balloon angioplasty and direct stenting. The power of mechanical recananlization devices for a fast and effective resolution of acute and chronic arterial occlusions would be more and widespread accepted solving the problem of reimbursement for expensive mechanical recananlization devices. Solving the latter problem, the use of mechanical recananlization devices could be one of the TOP procedures in interventional radiology; especially in Europe for most interventional centers and physicians the topic of reimbursement of expensive devices is unsolved, therefore the use of expensive mechanical recanalization devices remains still limited to a few centers. References: 1. Müller-Hülsbeck S, Grimm J, Suwelack D, Heller M (1999) Mechanical assistance for enhancement of In vitro effectiveness for hydrodynamic thrombectomy. Invest Radiol 34: 669-677 2. Fogarty TJ, Cranley JJ, Krause RJ, Strasser ES, Hafner CD. A method of extraction of arterial emboli and thrombi. Surg Gynecol Obstet 1963; 116: 241-244 3. Van Ommen V, Van der Veen FH, Daemen M, Habets J, Wellens H. In

vivo evaluation of the hydrolyser hydrodynamic thrombectomy catheter. J Vasc Intervent Radiol 1994; 5: 823-826 4. Wagner HJ, Müller-Hülsbeck S, Pitton M, Weiss W, Weß M. Rapid thrombectomy with a hydrodynamic catheter: results from a prospective, multicenter trial. Radiology 1997; 205: 675-681 5. Sharafuddin MJ, Hicks ME, Jenson ML, Morris JE, Drasler WJ, Wilson GJ. Rheolytic thrombectomy with the use of the AngioJet-F105 catheter: preclinical evaluation of safety. J Vasc Intervent Radiol 1997; 8: 939-945 6. Reekers JA, Kromhout JG, van der Waal K. Catheter for percutaneous thrombectomy: first clinical experience. Radiology 1993; 188: 871-874 7. Vicol C, Dalichau H. Recanalization of aged venous thrombotic occlusions with the aid of a rheolytic system: an experimental study. Cardiovascular Intervent Radiol 1996; 19: 255-259 8. Gorich J, Rilinger N, Sokiranski R, Kramer S, Mickley V, Schutz A, Brambs HJ, Pamler R. Mechanical thrombolysis of acute occlusion of both the superficial and the deep femoral arteries using a thrombectomy device. AJR Am J Roentgenol 1998;170:1177-1180 9. Zana K, Otal P, Fornet B, Forrai G, Chabbert V, Smayra T, Joffre F, Rousseau H. In Vitro Evaluation of a New Rotational Thrombectomy Device: The Straub Rotarex Catheter. Cardiovasc Intervent Radiol 2001; 24:319-323 10. Berczi V, Deutschmann HA, Schedlbauer P, Tauss J, Hausegger KA Early Experience and Midterm Follow-up Results with a New, Rotational Thrombectomy Catheter. Cardiovasc Intervent Radiol 2002;25:275-281 11. Zeller T, Muller C, Frank U, Burgelin KH, Horn B, Roskamm H. The Straub-Rotarex thrombectomy system: initial experiences Rofo Fortschr Geb Rontgenstr Neuen Bildgeb Verfahr 2001;173:626-631 12. Jäger KA, Schmitt EM, Schmitt HE, Labs KH. Peripheral Thrombectomy With the new Straub-Rotarex® Catheter: A Multicenter Study. Intern Angiology 1999;Suppl.1: 17A 13. Kandarpa K, Copra PS, Aruny JE, Polak JF, Donaldson MC, Whittemore AD, Mannick JA, Goldhaber SZ, Meyerovitz MF. Intraarterial thrombolysis of the lower extremity occlusions: prospective, randomized comparison of forced periodic infusion and conventional slow continuous infusion. Radiology 1993; 188: 861-867 14. McNamara TO, Gardner KR, Bomberger RA, Greaser LE. Clinical and angiographic selection factors for thrombolysis as initial therapy for acute lower limb ischemia. J Vasc Interv Radiol 1995; 6: 36s-47s 15. Ouriel K, Veith FJ, Sasahara AA. A comparison of recombinant urokinsae with vascular surgery as initial treatment for acute arterial occlusions of the legs. N Engl J Med 1998; 338: 1105-1111 16. Jahnke T, Andresen R, Muller-Hulsbeck S, Schafer FK, Voshage G, Heller M, Brossmann J.Femoropopliteal Occlusions: Experience with Peripheral Hemobahn stent-grafts for treatment of femoropopliteal arterial obstructions: midterm results of a prospective trial. J Vasc Interv Radiol 2003 Jan;14(1):41-51 17. Jahnke T, Voshage G, Muller-Hulsbeck S, Grimm J, Heller M, Brossmann J. Endovascular placement of self-expanding nitinol coil stents for the treatment of femoropopliteal obstructive disease. J Vasc Interv Radiol 2002 Mar;13(3):257-66 18. Müller-Hülsbeck S, Grimm J, Jahnke T, Brossmann J, Hilbert C , Heller M (2000) Erste Ergebnisse nach Inplantation des neuen ballonexpandierbaren Bridge™-Stents in die Beckenarterien. Rofo- Fortschr Rontg 172: 836-841 19. Grimm J, Müller-Hülsbeck S, Jahnke T, Hilbert C, Brossmann J, Heller M (2001) Randomized study to compare PTA alone versus PTA with Palmaz stenting for femoropopliteal lesions. J Vasc Interv Radiol 12: 935942 20. Jahnke T, Link J, Muller-Hulsbeck S, Grimm J, Heller M, Brossman J. Treatment of infrapopliteal occlusive disease by high-speed rotational atherectomy: initial and mid-term results. J Vasc Interv Radiol 2001 Feb;12(2):221-6 21. Steinkamp HJ, Werk M, Haufe M, Felix R. Laser angioplasty of peripheral arteries after unsuccessful recanalization of the superficial femoral artery. Int J Cardiovasc Intervent 2000 Sep;3(3):153-160 22. Steinkamp HJ, Wissgott C, Rademaker J, Scheinert D, Werk M, Settmacher U, Felix R. Short (1-10 cm) superficial femoral artery occlusions: results of treatment with excimer laser angioplasty. Cardiovasc Intervent Radiol 2002 Sep-Oct;25(5):388-96 23. Steinkamp H, Werk M, Wissgott C, Settmacher U, Haufe M, Hierholzer C, Felix R. Stent placement in short unilateral iliac occlusion. Technique and 24-month results. Acta Radiol 2001 Sep;42(5):508-14 24. Gray BH, Laird JR, Ansel GM, Shuck JW.Complex endovascular treatment for critical limb ischemia in poor surgical candidates: a pilot study. J Endovasc Ther 2002 Oct;9(5):599-604

25. Steinkamp HJ, Werk M, Beck A, Teichgraber U, Haufe M, Felix R. Excimer laser-assisted recanalisation of femoral arterial stenosis or occlusion caused by the use of Angio-Seal. Eur Radiol 2001;11(8):136470

2.4 Stenting the SFA: Should we do it ? T. Jahnke, S. Müller-Hülsbeck; Department of Radiology, University Hospital Schleswig-Holstein, Campus Kiel, Kiel, Germany. Why is stenting of the SFA a problem ? In patients with peripheral vascular disease (PVD) the infrainguinal arteries are almost always affected and especially in this vascular territory we are often encountered with a diffuse type of disease. A relatively small vessel lumen with high plaque burden and a high frequency of primary occlusions makes the SFA very difficult to treat effectively, e.g. to maintain long-term patency. If initial balloon angioplasty fails due to hemodynamically relevant dissection or residual luminal narrowing, stents might offer a potential solution [1-5]. However, a significant problem is that long-term patency is compromised by neo-intimal hyperplasia and/or progression of the underlying disease [1, 6-14]. In addition, the process of activation, proliferation and migration of smooth muscle cells seems to be triggered by the implanted device itself [15, 16]. Modern strategies for prophylaxis of recurrent stenoses like drug-eluting stents or local catheter- based application of drugs or gene- carrying vectors and brachytherapy are under intensive scientific evaluation, but none of these modalities have yet reached widespread clinical practise. In a publication of the Transatlantic InterSociety Consensus on the Management of Peripheral Arterial Disease (TASC) which was worked out by representatives of vascular medical, surgical, and radiologic societies from Europe, Canada and the US, the consensus expert recommendation for femoropopliteal stenting was that this procedure is not indicated as a primary approach to the interventional treatment of intermittent claudication [17], nonetheless percutaneous stent-placement into femoro-popliteal arteries has become a widely accepted therapeutic strategy if transluminal angioplasty fails. This lectures gives a short overview about historical aspects and current developments of femoropoplitealeal stenting in patients with PVD. Are the results of stenting superior to PTA alone ? In the largest randomised study comparing PTA versus stenting with the Palmaz stent in the femoropopliteal artery, 154 obstructions (141 patients) were randomised to PTA (n=77) or PTA plus stenting (n=77) [18]. In the PTA group initial technical success was achieved in 65 of 77 limbs (84%) versus 76 of 77 limbs (99%) in the stent group. Cumulative primary patency rates at 12 and 24 months were 63% and 53%, respectively, for both groups. Thus initial success actually appears to be better with stenting, but long-term results are not convincing when compared to PTA. We also performed a randomised study to compare PTA alone versus PTA with Palmaz stenting for femoropopliteal lesions with a long-term follow up of more than 3 years. 21 patients underwent PTA in combination with stenting and 19 patients received PTA alone. Mean follow up for the PTA group was 33.8 (± 8.7) months and for the Palmaz group 29.1 (±6.2) months with a maximum follow-up period of 39 months for both groups. No significant differences in the primary or secondary patency rates could be observed, neither at 12 nor at 39 months: After 39 months the primary patency rate for PTA alone was 68.4%, secondary patency rate was 89.5%; primary patency rate for PTA with stenting was 65%, secondary 90%. Trends were observable favoring patients with shorter and non-occlusive lesions. However, the results of this study again showed that - when looking at a long-term follow up of more than 3 years - stenting of the femoropopliteal artery does not produce better results than PTA alone although providing a favourable initial luminal gain following the procedure [19]. Do et al. performed another comparative study in the femoropopliteal level using the Wall-Stent. Primary 1-year patency rates based on clinical status and ABI were not significantly different between PTA or stenting using this particular self-expandable stent design (65% vs 59% respectively) [12]. Which stents are being used and how do they perform ? During the last 20 or so years nearly all available stent designs have been used in the femoropopliteal level for salvage of PTA or even “offlabel” primary treatment. The stent designs first employed were the Palmaz-, the Strecker- and the Wallstent and results of these clinical studies were published in the late 1980s. Henry et al. reported the highest

patency rates for the Palmaz-stent with a primary 1-year patency of 81% [6]. Other stent models did not produce comparably good results. Martin et al. reported a primary patency of 61% at 1 year and 49% at two years for Wallstents in the SFA (secondary patency 84% and 71%, respectively) [7]. Strecker stents were used by Liermann et al. for peripheral infrainguinal lesions showing a primary patency of 71% at one year [8]. However, in a later study Strecker reported primary patency rates of 76% one year following treatment [14]. In a review article on femoropopliteal interventions Lammer [20] assessed the weighted average of published long-term patency rates after stenting of femoropopliteal artery lesions in 585 patients (600 limbs, 80% claudicants) to be 67% and 58% at 1 year and 3 years, respectively. These data underline the fact that although stents may dramatically improve the immediate post-angioplasty appearance, they are actually not able to prevent restenosis. Moreover, there is no proven difference in stentrelated restenosis or occlusion within the range of currently available stent designs and materials. What´s new with femoropopliteal stents ? The Intracoil™ self-expanding nitinol coil stent (Intratherapeutics, USA) This device is the first stent to be FDA-approved for the femoropopliteal artery and it is also the first non-coronary vascular stent to be approved for use as a primary treatment. The Intracoil consists of a spring-shaped nitinol wire mounted on a delivery catheter which - when released “jumps” into its nominal helical coil configuration. Actually this stent design is not new: In 1969 it was Dotter who introduced a tubular coilshaped stainless steel wire which he had successfully implanted into rabbit and canine arteries. 14 years later he described an expandable nitinol coil stent design for “nonoperative restoration and maintenance of patency in blood vessels” [21]. 1993 the Intracoil™ self-expanding peripheral stent (formerly called “VascuCoil” by InStent, Inc) became commercially available. The design of this stent makes it suitable especially for tortuous arteries and lesions of the femoral or popliteal artery situated at flexion points or at locations that may be subjected to external compression, but due to the helical configuration of the wires with spacing between the struts of up to 3,5 mm the device is not ideal for heavily calcified and eccentric lesions, and conventional balloonexpandable stents should be preferred in these situations.In a Multicenter study a total of 357 patients were prospectively randomised to treatment with the Intracoil-stent or PTA. Results of the first 93 “roll-in patients” - which constituted the learning phase for the 22 clinical sites - were published by Ansel et al. in 2002. Acute angiographic success was obtained in 98.9% of patients. Major complications occurred in 3.2% and no patient experienced abrupt or subacute closure. At 9month follow-up of successfully treated patients 77.9% remained free of major adverse clinical events; freedom from target lesion revascularization was 81.8%. The authors concluded that the use of the Intracoil stent showed excellent clinical and hemodynamic patency at 9 months [22]. We also had the opportunity to evaluate the Intracoil in a nonrandomised two-center trial for stenting of the SFA after PTA failure [23]. In our series with 37 patients implantation of the device showed promising initial success rates with primary patency of 97,1% (SE =2,9) and 86,2 % (SE=6,5) at 6 and 12 months. 24 months follow-up data with primary and secondary patency rates of 77,8 (SE=8,1) and 92,6 (SE=5,1) respectively became recently available and were presented at this years “HighTech Peripherique” Meeting in Paris. However, despite our excellent initial and mid-term results assessment of long-term patency by larger studies is still needed for evaluation of superiority over other stent designs or PTA alone. The Hemobahn™ self-expanding stent-graft (GORE, USA) To improve long-term results of femoro-popliteal stenting new stentdesigns are constantly being developed by the industry not only to improve flexibility, profile and mechanical resistance of the current devices, but also to overcome the problem of late restenosis due to negative remodelling and neointima-formation. In contrast to mere variations of the two basic types of vascular stents available today – the classic balloon-expandable systems and newer generations of selfexpanding stents like the Intracoil device – stent-grafts incorporate the theoretical advantage to limit neointimal hyperplasia. This is based on the assumption that the polymer cover might reduce tissue in-growth at the treatment site and thereby improve patency. Recently, we reported immediate and mid-term results of the Hemobahn selfexpanding stent-graft for percutaneous interventional treatment in 52 patients with long-segment high-grade stenoses and occlusions of the femoropopliteal artery [24]. Overall, in our series re-obstructions were observed in 25% (13/52) of the patients (reocclusions: 19,2% [10/52],

restenoses: 5,8% [3/52]). For both, stenoses and occlusions the primary patency rates were 78,4% and 74,1% at 12 and 24 months, respectively. Primary assisted patency was 82,4% at 12 months and 80,3% at 24 months. Secondary patency rates for 12 and 24 months were 88,3% and 83,2% respectively. These results - which were comparable to the first prospective Multicenter study published by Lammer et al. in 2000 [25] - led to the conclusion that endovascular placement of the Hemobahn stent-graft for treatment of high-grade stenoses and occlusions of the femoro-popliteal artery is a safe procedure with excellent initial success rates and promising mid-term results. Follow-up imaging studies also indicated that the graft cover of the device actually has the potential to reduce neointimal in-growth through the supporting stent, and thus to enhance long-term patency. This assumption can be based on the fact that our cases of reobstructions were associated with plaque progression and neointima formation in the genuine artery near the stent-ostia while neointimal hyperplasia inside the stent-graft was not detected at any time. Drug eluting stents Following the global excitement of Cardiologists when early trial results indicated that Sirolimus-eluting coronary artery stents may offer a definitive solution for restenosis, a pilot study of Sirolimus-eluting stents versus uncoated devices in superficial femoral artery obstructions was performed at one Canadian and five European hospitals (SIROCCO Trial). This prospective, double-blinded study tracked the clinical experience of 36 severely claudicated patients with femoral artery occlusions or stenoses. 18 patients were treated with the Cordis self-expanding nitinol SMART stent coated with a polymer impregnated with Sirolimus, the remaining 18 patients received bare SMART stents. At 6 months the in-stent mean percent diameter stenosis was 22.6% in the sirolimuseluting stent group versus 30.9% in the uncoated stent group (P=0.294). The in-stent mean lumen diameter was significantly larger in the sirolimus-eluting stent group (4.95 mm versus 4.31 mm in the uncoated stent group; P=0.047). The authors concluded that sirolimuseluting SMART stents for superficial femoral artery obstructions show a trend toward reducing late lumen loss when compared to uncoated stents [26]. Should we do it ? Due to the often complex lesion morphology in the SFA of patients with PVD this territory is regarded as a region so notorious for in-stent restenosis that some interventional radiologists consider stenting a waste of time. However, stent placement has an undeniably important role for salvage of failing PTA especially in patients with critical limb ischemia. Promising new stents – also designed for primary treatment of this vascular region - have been introduced to the market and seem to have advantages over conventional stents. Furthermore the development of drug-eluting stents has raised considerable hope to reach a definite solution for the problem of restenosis in future. However, despite excellent initial and mid-term result of new stent technology, long-term success still needs to be proven by larger randomized multicenter trials in order to assess a potential superiority over PTA alone or conventional stent systems. Another important issue when dealing with modern devices is the problem of reimbursement. Depending on the health system of the country in which the stents are to be used it will be more or less difficult to introduce sophisticated new techniques into clinical practice regardless of favourable results from large international randomised studies. So at this time the answer to the question “should we do it ?” is a definite: “yes ,…perhaps, …some day…”. References: 1. Jeans WD, Amstrong S, Cole SEA, et al. Fate of patients undergoing transluminal angioplasty for lower-limb ischemia. Radiology 1990; 177:559-564 2. Capek P, McLean GK, Berkowitz HD. Femoropopliteal angioplasty: factors influencing long-term success. Circulation 1991; 83 (Suppl. 2):I70I80 3. Johnston KW. Femoral and popliteal arteries: reanalysis of results of balloon angioplasty. Radiology 1992; 183:767-771 4. Matsi PJ, Manninen HI, Vanninen RL, et al.: Femoropopliteal angioplasty in patients with claudication: primary and secondary patency in 140 limbs with 1–3-year follow-up. Radiology 1994; 191:727-733 5. Murray JG, Apthorp LA, Wilkins RA. Long-segment (>10 cm) femoropopliteal angioplasty: improved technical success and long-term patency. Radiology 1995; 195:158-162 6. Henry M, Amor M, Ethevenot G, et al. Palmaz stent placement in iliac and femoropopliteal arteries: primary and secondary patency in 310

patients with 2–4-year follow-up. Radiology 1995; 197:167-174 7. Martin EC, Katzen BT, Benanti JF, et al. Multicenter trial of the Wallstent in iliac and femoral arteries. J Vasc Interv Radiol 1995; 6:843-849 8. Strecker EP, Hagen P, Liermann D, et al. Iliac and femoropopliteal vascular occlusive disease treated with flexible tantalum stents. Cardiovasc Intervent Radiol 1993; 16:158-162 9. Vorwerk D, Gunther RW, Schurmann K, Wendt G. Aortic and iliac stenoses: follow-up results of stent placement after insufficient balloon angioplasty in 118 cases. Radiology 1996; 198:45-48 10. Rousseau HP, Raillat CR, Joffre FG, et al. Treatment of femoropopliteal stenoses by means of self-expandable endoprostheses: midterm results. Radiology 1989; 172:961-964 11. Zollikofer CL, Antonucci F, Pfyffer M, et al. Arterial stent placement with use of the Wallstent: midterm results of clinical experience. Radiology 1991; 179:449-456 12. Do DD, Triller J, Walpoth BH, Stirnemann P, Mahler F. A comparison study of self-expandable stents vs balloon angioplasty alone in femoropopliteal artery occlusions. Cardiovasc Intervent Radiol 1992; 15:306312 13. Liermann D, Strecker EP, Peters J. The Strecker stent: indications and results in iliac and femoropopliteal arteries.Cardiovasc Intervent Radiol 1992; 15:298-305. 14. Strecker EPK, Boss IBL, Gotzmann D. Femoropopliteal artery stent placement: evaluation of long-term success. Radiology 1997; 205:375383 15. Kearney M, Pieczek A, Haley L, Losordo DW, Andres V, Schainfeld R, Rosenfield K, Isner JM. Histopathology of in-stent restenosis in patients with peripheral artery disease. Circulation. 1997 Apr 15;95(8):19982002 16. Barth KH, Virmani R, Froelich J, Takeda T, Lossef SV, Newsome J, Jones R, Lindisch D.Paired comparison of vascular wall reactions to Palmaz stents, Strecker tantalum stents, and Wallstents in canine iliac and femoral arteries. Circulation. 1996 Jun 15;93(12):2161-9 17. Management of peripheral arterial disease (PAD). TASC Working Group. TransAtlantic Inter-Society Concensus (TASC). J Vasc Surg. 2000 Jan;31(1 Pt 2):1-296 18. Cejna M, Thurnher S, Illiasch H, Horvath W, Waldenberger P, Hornick K, Lamer J.: Percutaneous transluminal angioplasty (PTA) versus Palmaz stent placement in femoropopliteal artery obstruction: A Multicenter prospective randomised study. J Vasc Interv Radiol 2001; 12: 23-31 19. Grimm J, Müller-Hülsbeck S, Jahnke T, Hilbert C, Brossmann J, Heller M: Randomized study to compare PTA alone versus PTA with Palmaz stenting for femoropopliteal lesions. J Vasc Interv Radiol 2001; 12: 935942 20. Lammer J: Femoropopliteal Artery Obstruction: From the Balloon to the Stent Graft. Cardiovasc Intervent Radiol (2001) 24:73-83 21. Dotter CT, Buschmann RW, McKinney MK, Rösch J. Transluminal expandable nitinol coil stent grafting: preliminary report. Radiology 1983, 2: 979-987 22. Ansel GM, Botti CF Jr, George BS, Kazienko BT; IntraCoil Femoralpopliteal Stent Trial Investigators. Clinical results for the trainingphase roll-in patients in the Intracoil femoralpopliteal stent trial. Catheter Cardiovasc Interv. 2002 Aug;56(4):450). 23. Jahnke T, Voshage G, Müller-Hülsbeck S, Grimm J, Heller M, Brossmann J: Endovascular placement of self-expanding nitinol coil stents for the treatment of femoro-popliteal obstructive disease. J Vasc Interv Radiol. 2002 Mar;13(3):257-66 24. Jahnke T, Andresen R, Müller-Hülsbeck S, Schäfer FKW, Voshage G, Heller M, Brossmann J: Hemobahn stent-grafts for treatment of femoropopliteal arterial obstructions: Mid-term results of a prospective trial. J Vasc Interv Radiol. 2003; 14: 41-51 25. Lammer J, Dake MD, Bleyn J, Katzen BT, Cejna M, Piquet P, Becker GJ, Settlage RA. Peripheral arterial obstruction: prospective study of treatment with a transluminally placed self-expanding stent-graft. International Trial Study Group. Radiology. 2000 Oct;217(1):95-104 26. Duda SH, Pusich B, Richter G, Landwehr P, Oliva L, Tielbeek A, Wiesinger B, Hak JB, Tielemans H, Ziemer G, Cristea E, Lansky A, Beregi JP: Sirolimus-eluting stents for the treatment of obstructive superficial femoral artery disease: six-month results. Circulation 2002 Sep 17;106(12):1505-9

Special Session Biliary Interventions I 3.1 Preprocedural imaging of bile ducts E. A. Eracleous; Radiology, Diagnostic Center "Ayios Therissos", Nicosia, Cyprus. The preprocedural image of the biliary system is very essential for therapeutic decisions. Anatomic demonstration of the bile ducts can be obtained by ultrasonography (US), computer tomography (CT), magnetic resonance cholangiopancreatography (MRCP) and endoscopy retrograde cholangiopancreatography (ERCP). Biliary function can be evaluated employing cholescintigraphy (CS). Helical or multislice CT cholangiography (HCTC), contrast enhanced MRCP (CE MRCP) and secretin and cholecystokinin stimulation can provide both anatomic and functional / kinetic information. Percutaneous transhepatic cholangiography may be used for biliary duct imaging in selected cases. Patients with bile duct dilatation, with or without jaundice, choledocholithiasis, bile duct neoplasms and congenital malformations of the biliary tree should be initially imaged by US. US is highly accurate in diagnosing surgical jaundice, demonstrating dilatation of the biliary tree in up to 95% of the patients with obstruction. It is also highly accurate in determining the level of the obstruction, although the actual cause may be diagnosed in only about one third of the patients. A negative US must not preclude the option of proceeding to other imaging modalities, since obstruction may occur without duct dilatation and US sensitivity for the detection of cholelithiasis is about 82%(1). Detection and staging of malignant hilar obstructions are improved by the use of Levovist in the postvascular phase of sonography compared with routine sonography(2). Endoscopic ultrasonography(EUS) (2-4), intraductal ultrasonography (5,6) , percutaneous transhepatic cholangioscopy (5) and intraoperative US(7), although not widely used, can provide high resolution images and can disclose small common bile duct stones and small strictures. Recent studies have shown that EUS is the most sensitive method for diagnosing choledocholithiasis with 95% sensitivity (3). In patients’ population with low disease prevalence, EUS is superior to MRCP for choledocholithiasis. EUS is most useful for confirming a normal biliary tree and should be considered a low-risk alternative to ERCP. Although MRCP had the lowest procedural reimbursement, the initial EUS strategy had the greatest cost effectiveness by avoiding unnecessary ERCP examinations (4). Recently developed ultrasonographic miniprobes (diameters about 2 mm; frequencies 12-20 MHz) can be passed through the working channel of standard endoscopes to provide high-frequency ultrasound images (6). Cross-sectional CT is less sensitive than US in demonstrating bile duct dilatation or non-calcified biliary stones but may demonstrate extrabiliary causes of obstruction and calcified calculi with sensitivity up to 86% (8). Multiplanar biliary tree minimum intensity projections of the biliary tree based on helical CT data sets can improve the conspicuity of the bile ducts (9). The combination of helical CT imaging and intravenous administration of biliary excreted iodinated contrast media enables volumetric anatomical imaging and evaluation of kinetics of the biliary tree (10-12). The technique is very sensitive in detecting small hypodense stones, in discriminating stones from air bubbles, in determining communication of the biliary tree with a cystic lesion, in diagnosing minimal or segmental biliary dilatation and in demonstrating pancreaticobiliary maljunction or pancreatic reflux into the common bile duct (10-12). Evaluation of anatomic variations prior to laparoscopic cholecystectomy, of bile kinetics in post cholecystectomy cases, of biliary-enteric anastomoses, or of bile leakage and assessment of patency of stents can be performed employing HCTC (12). However, the method is not indicated in patients with known allergy, hepatic or renal failure or elevated serum bilirubin levels. MRCP (13-29) is a safe non-ionizing, non-invasive method for visualizing the biliary tree, both proximal and distal to an obstructing lesion and it has replaced diagnostic ERCP in many centers. Excellent MRCP images can be obtained by high performance gradients allowing rapid acquisition times and dedicated phase array coils increasing the signal to noise ration. Projectional single shot turbo spin echo heavily T2weigheted images and multislice HASTE images, relatively insensitive to susceptibility artifacts, with maximum intensity projection reconstructions and in use to demonstrate the bile with high signal in the low signal intensity background of soft tissues. Congenital abnormalities,

choledocholithiasis, intrahepatic bile ducts lesion, sclerosing cholangitis (18) and obstruction are among the indications of the method. Differential diagnosis of an obstructing lesion requires evaluation of both MRCP and cross sectional MR images. MRCP is subjective to different types of artifacts and may simulate or mask pathological conditions (19, 20). MRCP is a more effective diagnostic method than ERCP for the evaluation of intrahepatic stones (21). Although MRCP provides the same imaging information as direct cholangiography, it has limited specificity for the diagnosis of malignant strictures (22). The use of MRCP to guide biliary stent placement in a patient with inoperable hilar obstruction reduces the overall cost of treatment (23).Initial MRCP in patients referred with abdominal pain would potentially have avoided ERCP in 44 % of cases, reduced ERCP workload by 13 % and significantly reduced patient morbidity and mortality. The relatively small reduction in ERCP workload among these patients reflects the fact that over half of them had probable sphincter dysfunction, a significant proportion of whom might have benefited from biliary manometry and/or endoscopic intervention despite a normal MRCP. Furthermore, a small number of patients with calculi and subtle biliary and pancreatic strictures would be missed by this approach (24). Contrast enhanced MR cholangiography is an alternative method for imaging of the biliary tree (25). The biliary excreted contrast allows for excellent demonstration of bile ducts on T1-weighted 3D FLASH sequence providing both anatomic and functional information (25). Endoscopy like views from the biliary tree can be obtained using either HASTE or contrast enhanced 3D FLASH images (26). Secretin-stimulated MRCP for non-invasive quantification of pancreatic exocrine function is also used (27, 28). Nonvisualization of sphincteric relaxation on kinematic MRCP indicates ampullary or periampullary lesions (29). Kinematic MRCP can be used to determine the necessity of biliary intervention in patients with biliary dilatation (29). Hepatobiliary function is traditionally evaluated by cholescintigraphy, but the method is not widely used nowadays (30). The current role for quantitative cholescintigraphy is the assessment of patients with biliary colic and a negative US or MRCP. US remain the initial imaging modality for the biliary tree. MRCP will replace ERCP for diagnostic purposes. Helical CT cholangiography, contrast enhanced MRCP, virtual endoscopy, Endoscopic or intraductal US, transhepatic cholangiography and MRCP after cholecystokinin stimulation may used in selected cases. References: 1. Pickuth D, Spielmann RP. Detection of choledocholithiasis: comparison of unenhanced spiral CT, US, and ERCP. Hepatogastroenterology 2000 Nov-Dec;47(36):1514-7 2. Khalili K, Metser U, Wilson SR. Hilar biliary obstruction: preliminary results with Levovist-enhanced sonography. AJR Am J Roentgenol 2003 Mar; 180(3):687-93 3. Deprez P. Approach of suspected common bile duct stones: endoscopic ultrasonography.Acta Gastroenterol Belg 2000 JulSep;63(3):295-8. 4. Scheiman JM, Carlos RC, Barnett JL, Elta GH, Nostrant TT, Chey WD, Francis IR, Nandi PS Can endoscopic ultrasound or magnetic resonance cholangiopancreatography replace ERCP in patients with suspected biliary disease? A prospective trial and cost analysis.. Am J Gastroenterol 2001 Oct;96(10):2900-4 5. Tamada K, Ohashi A, Tomiyama T, Wada S, Satoh Y, Higashizawa T, Ido K, Sugano K Comparison of intraductal ultrasonography with percutaneous transhepatic cholangioscopy for the identification of residual bile duct stones during lithotripsy.. J Gastroenterol Hepatol 2001 Jan;16(1):100-3 6. Menzel J, Domschke W. Intraductal ultrasonography (IDUS) of the pancreato-biliary duct system. Personal experience and review of literature.Eur J Ultrasound 1999 Nov;10(2-3):105-15 7. Kruskal JB, Kane RA. Intraoperative sonography of the biliary system.AJR Am J Roentgenol 2001 Aug;177(2):395-403 8. Pickuth D, Spielmann RP. Detection of choledocholithiasis: comparison of unenhanced spiral CT, US, and ERCP. Hepatogastroenterology 2000 Nov-Dec;47(36):1514-7 9. Raptopoulos V, Prassopoulos P, Chuttani R, McNicholas MM, McKee JD, Kressel HY Multiplanar CT pancreatography and distal cholangiography with minimum intensity projections. Radiology 1998 May;207(2):317-24 10. Zandrino F, Benzi L, Ferretti ML, Ferrando R, Reggiani G, Musante F. Multislice CT cholangiography without biliary contrast agent: technique

and initial clinical results in the assessment of patients with biliary obstruction.Eur Radiol 2002 May;12(5):1155-61 11. Hoglung M, Muren C, Boijen MW et al Computed Tomography with intravenous cholangiography contrast: a method for visualization choledochal cysts. Eur J Radiol. 1990; 10:159-161. 12. Stabile Ianora AA, Memeo M, Scardapane A, Rotondo A, Angelelli G. Oral contrast-enhanced three-dimensional helical-CT cholangiography: clinical applications. Eur Radiol 2003 Apr;13(4):867-73 13. Papanikolaou N, Karantanas AH, Eracleous E, Costa JC, Gourtsoyiannis N. Magnetic resonance cholangiopancreatography: comparison between respiratory-triggered turbo spin echo and breath hold singleshot turbo spin echo sequences.Magn Reson Imaging. 1999 Nov;17(9):1255-60. 14. Miyazaki T, Yamashita Y, Tsuchigame T, et. al. MR Cholangiopancreatography using HASTE (half-Fourier acquisition single-shot turbo spinecho) sequences. AJR 1996; 166:1297-1303. 15. Iriei Honda H Tajima T et al. Optimal MR cholangiopancreatographic sequence and its clinical application Radiology 1998; 206: 379387. 16. Fulcher A S, Turner M A, Cappes G E. MR Cholangiography: Technical advances and clinical application Radiographics 1999; 19: 25-41. 17. Neri E, Poraschi P Caramella D et al Common bile duct stone: Detections with MR virtual endoscopy SE 331 RSNA 1999. 18.Textor HJ, Flacke S, Pauleit D, Keller E, Neubrand M, Terjung B, Gieseke J, Scheurlen C, Sauerbruch T, Schild HH. Three-dimensional magnetic resonance cholangiopancreatography with respiratory triggering in the diagnosis of primary sclerosing cholangitis: comparison with endoscopic retrograde cholangiography.Endoscopy 2002 Dec;34(12):984-90 19.Sugita R, Sugimura E, Itoh M, Ohisa T, Takahashi S, Fujita N. Pseudolesion of the bile duct caused by flow effect: a diagnostic pitfall of MR cholangiopancreatography.AJR Am J Roentgenol 2003 Feb;180(2):467-71 20. Hakansson K, Christoffersson JO, Leander P, Ekberg O, Hakansson HO. On the appearance of bile in clinical MR cholangiopancreatography.Acta Radiol 2002 Jul;43(4):401-10 21. Kim TK, Kim BS, Kim JH, Ha HK, Kim PN, Kim AY, Lee MG. Diagnosis of intrahepatic stones: superiority of MR cholangiopancreatography over endoscopic retrograde cholangiopancreatography.AJR Am J Roentgenol 2002 Aug;179(2):429 22. Rosch T, Meining A, Fruhmorgen S, Zillinger C, Schusdziarra V, Hellerhoff K, Classen M, Helmberger H. A prospective comparison of the diagnostic accuracy of ERCP, MRCP, CT, and EUS in biliary strictures.Gastrointest Endosc 2002 Jun;55(7):870-6 23. Harewood GC, Baron TH Cost analysis of magnetic resonance cholangiography in the management of inoperable hilar biliary obstruction. Am J Gastroenterol 2002 May;97(5):1152-8 24. Farrell RJ, Noonan N, Mahmud N, Morrin MM, Kelleher D, Keeling PW. Potential impact of magnetic resonance cholangiopancreatography on endoscopic retrograde cholangiopancreatography workload and complication rate in patients referred because of abdominal pain.Endoscopy. 2001 Aug;33(8):705-8. 25. Papanikolaou N, Prassopoulos P, Eracleous E, Maris T, Gogas C, Gourtsoyiannis N. Contrast-enhanced magnetic resonance cholangiography versus heavily T2-weighted magnetic resonance cholangiography. Invest Radiol. 2001Nov;36(11):682-6. 26. Neri E, Caramella D, Boraschi P, Braccini G, Lehmann ED, Perri G, Bartolozzi C. Magnetic resonance virtual endoscopy of the common bile duct stones.Surg Endosc 1999 Jun;13(6):632-3 27. Punwani S, Gillams R, Lees R. Non-invasive quantification of pancreatic exocrine function using secretin-stimulated MRCP.Eur Radiol 2003 Feb;13(2):273-6 28. Matos C, Metens T, Deviere J et al. Pancreatic duct: Morphologic and functional evaluation with Dynamic MR Pancreatography after secretin stimulation. Radiology 1997; 203: 435-441. 29. Kim JH, Kim MJ, Park SI, Chung JJ, Song SY, Yoo HS, Lee JT. Using kinematic MR cholangiopancreatography to evaluate biliary dilatation.AJR Am J Roentgenol 2002 Apr;178(4):909-14

3.2 Puncture technique. Stent implantation technique and strategy A. A. Hatzidakis; Department of Radiology, University of Crete, Heraklion Crete, Greece. Table of Contents

o o o

PTC PTBD Biliary Metallic Stenting

PTC Percutaneous Transhepatic Cholangiography (PTC) is the basic procedure for opacification of dilated biliary tree before trying to decompress it by means of drainage placement. PTC is performed under sterile conditions, with the patient in supine position. After percutaneous local anesthesia, we puncture the liver with a fine 21-22 G Chiba needle, under fluoroscopic guidance. We enter the liver in the right middle axillary line between the 9th and 11th intercostal space targeting the xiphoid, keeping the needle on a horizontal level until we reach the right lateral spine margin. If we prefer to puncture the left biliary duct system, we then puncture the anterior abdominal wall just under the xiphoid, aiming posterior and right lateral. In order not to obscure the region of interest, bile aspiration can be used instead of contrast injection before we are sure the needle tip is inside a bile duct. Upper abdominal imaging available at procedure day, provide important information before starting PTC. Ultrasound, CT or MR Imaging can reveal liver segment anatomy, degree of obstruction in different liver parts, localization of the obstructing cause, presence of ascites or bowel interposition. US-real time imaging can be sometimes useful during puncturing. PTBD After opacification of the obstructed biliary system, a decision must be made about the site of puncture for placement of the drainage catheter. Percutaneous Transhepatic Biliary Drainage (PTBD) follows PTC procedure if malignant obstruction is revealed. There are two major techniques by which PTBD can be performed. First, the fine 22 G needle and second, the 18 G needle technique. Using the fine needle technique, the same fine PTC-needle can be used for puncturing of an appropriate dilated duct. If the duct used for initial cholangiography is appropriate, there is no need for a new puncture. Through the 22 G needle, a 0.018’’ inch fine stiff wire with a floppy tip can be initially inserted into a peripheral duct. Over this wire, a 5 Fr catheter-dilator system can be carefully advanced. Then, wire and dilator can be retrieved, leaving the outer catheter in the duct. Via that catheter, a 0.035’’ inch stiff type hydrophilic wire can be manipulated through the obstruction. Subsequently, a super stiff wire can exchange the hydrophilic, and 6 and 8 Fr dilators can be used to enlarge the transhepatic tract, before a larger 8 Fr biliary drainage catheter is placed. For the 18 G needle technique, a larger 18 G needle directly punctures the liver and after hitting the appropriate duct, the stylet is withdrawn and the plastic straight 5 Fr outer catheter remains in the duct. Through that a 0.035’’ inch stiff hydrophilic wire can be manipulated into the biliary tree. The major advantage of the fine needle technique is the fact that even multiple needle passages can cause less complication than the 18 G needle, so that the procedure can be considered as safer. The disadvantage is that fine needle technique can be more difficult in more inexperienced hands. Which duct is chosen to be punctured, depends on the site of obstruction and should be as peripheral as possible, in order to avoid injury of larger vascular structures. For obstruction of the common bile (CBD), a right-sided approach is preferred, except in cases of ascites or colon interposition. This route provides a more straight way for the wire and catheter down to the CBD, and keeps the operator’s hands away from the x-ray beam. In case of hilar strictures, puncture site should be decided after studying the tumor location, the tumor extent and the liver lobe infiltration or atrophy. In case of opacification of multiple obstructed bile ducts during PTC, the operator should try to drain as many opacified ducts as possible in order to avoid bacterial contamination and post procedural cholangitis. Special biliary manipulation catheters can assist the passage of the hydrophilic wire through the obstructing mass. Some times, if wire pas-

sage through the obstruction is not possible, only external drainage is initially placed. New attempt can be done after several days, which is usually successful, because of the smaller diameter of the decompressed CBD. Biliary Metallic Stenting Metallic Stenting can be performed after internal-external biliary drainage. Primary or secondary stenting technique is possible. Primary stenting means that the metallic stent is expanded immediately after the initial drainage, with or without balloon dilatation of the malignant stricture. Post-stenting balloon dilatation can be decided if the stenting end result is not satisfactory. Secondary stenting means that the physician waits several days before stent placement until the dilated system decompresses and a transhepatic tract starts to form. Especially, if hemobilia occurs during PTBD, one should wait before stenting. Hemobilia occurs less frequently by using the fine needle technique, and if no stricture dilatation is performed. Primary stenting technique has the major advantage of immediate internal bile drainage through a wide lumen of 5-10 mm, depending on how strong the metallic mesh radial force is. Post-stenting balloon dilatation can increase the inner stent diameter, if needed. Primary stenting can be followed by insertion of a 4 Fr catheter, which can be retrieved one day later after check-cholangiography. Some interventional radiologists (IR) retrieve the catheter immediately after primary stenting, placing a hemostatic plug in the transhepatic tract. The secondary stenting technique has the advantage of tamponating the transhepatic tract for several days, so that the risk of subcapsular hematoma, intraperitoneal hemorrhage or bile peritonitis decreases. After stent implantation, the draining catheter can stay in place for a total of 14 days, after initial placement, until the tract matures, before removal. This technique increases patient discomfort but decreases complication rate. Nevertheless, pain or infection risk increases. An arterial branch injury does not heal even after 2 weeks, so that major hemorrhage is possible after catheter removal in case of pseudoaneurysms formation. Emergency arterial selective embolization is then indicated. It is obvious that there is no gold-standard technique available, and every IR should find the appropriate technique, which suits him, according to his experience, the institutional conditions, the level of nursing and medical care, the length of patient’s hospital stay, the available material, etc.). The number of stents used, depends on the length of stricture, the intention to stent the papilla of Vater and the number of ducts and lobes one has to stent. If both lobes have to be stented, perhaps both stents may be introduced through one puncture site, for example through the right liver lobe, avoiding left liver puncture. Finally, the use of covered stents is still controversial and perhaps indicated in cases of secondary intervention after occlusion of the initially placed uncovered stent. Higher cost or stent availability, as well as the fact that they are not found to significantly improve patient’s survival or live quality has to be taken in account. Covered stents are less likely to be indicated in high hilar lesions. Uncovered metallic stents with a narrow woven mesh, seem to prevent tumor ingrowth more than stents with wide woven metallic construction. References: 1. Mueller PR, Harbin WP, Ferrucci JT Jr, Wittenberg J, vanSonnenberg E. Fine needle transhepatic cholangiography: reflections after 450 cases. AJR 1981; 136: 85-90. 2. Butch RJ, Mueller PR. Fine-needle transhepatic cholangiography. Semin Intervent Radiol 1985; 2: 1-20. 3. Nilsson U, Evander A, Ihse I, Lunderquist A, Mocibob A. Percutaneous transhepatic cholangiography and drainage. Acta Radiologica 1983; 24: 433-439. 4. Ferrucci JT Jr, Mueller PR, Harbin WP. Percutaneous transhepatic biliary drainage: technique, results, and complications. Radiology 1980; 135: 1-13. 5. Mueller PR, vanSonnenberg E, Ferrucci JT Jr. Percutaneous biliary drainage: technical and catheter-related problems in 200 procedures. AJR 1982; 138: 17-23. 6. Lammer J, Neumayer K. Biliary drainage endoprostheses: experience with 201 placements. Radiology 1986; 159: 625-629. 7. Dick BW, Gordon RL, LaBerge JM, Doherty MM, Ring EJ. Percutaneous transhepatic placement of biliary endoprosthesis: results in 100 consecutive patients. JVIR 1990; 1: 97-100. 8. Gordon RL, Ring EJ, LaBerge JM, Doherty MM. Malignant biliary obstruction: treatment with expandable metallic stents - follow-up

of 50 consecutive patients. Radiology 1992; 182: 697-701. Lameris JS, Stoker J, Nijs HGT, Zonderland HM, Terpstra OT, van 9. Blankenstein M, Schutte HE. Malignant biliary obstruction: percutaneous use of self-expandable stents. Radiology 1991; 179: 703-707. 10. Becker CD, Giatti A, Malbach R, Bauer HU. Percutaneous palliation of malignant obstructive jaundice with the Wallstent endoprosthesis: follow-up and re-intervention in patients with hilar and nonhilar obstruction.JVIR 1993;4:597-604. 11. Rossi P, Bezzi M, Rossi M, Adam A, Chetty N, Roddie ME, Iacari V, Cwikiel W, Zollikofer CL, Antonucci F, Boguth L. Metallic stents in malignant biliary obstruction: results of a multicenter European study of 240 patients. JVIR 1994; 5: 279-285.

Special Session Aortoiliac PTA/Stenting 7.1 Aortoiliac PTA/ Stenting : Preprocedural Imaging, Indications, Technique D. K. Tsetis; Radiology, University Hospital Heraklion, Heraklion, Greece. Preprocedural Imaging Intraarterial digital subtraction angiography (DSA) is still accepted as the gold standard imaging modality in aortoiliac occlusive disease. Its advantages include high spatial resolution, excellent imaging of diseased arteries and collateral pathways and high accuracy in the diagnosis of the etiology of the arterial obstruction. However it causes discomfort to the patient and carries a 0.7% risk of severe complications, a 0.1% risk of severe reaction to contrast medium and a 0.16% mortality risk (1,2). In patients with normal renal function, the abdominal aorta is studied in both anterioposterior (AP) and lateral views; the aortic bifurcation and iliac arteries are examined in AP and bilateral oblique views with the catheter at the aortic bifurcation, since disease preferably is located posteriorly in these segments. When treating distal common iliac (CIA) and proximal external iliac (EIA) arterial lesions, the contralateral oblique view also helps in separating the origin of the internal and external iliac arteries. In cases of impalpable femoral pulses due to aortobiliac occlusion, intravenous (i.v) DSA runs with the catheter tip in the right atrium are usually sufficient to outline the proximal end of the occlusion and the proximal femoral run-off arteries. In patients with increased risk of contrast medium (CM) induced nephropathy, the concentration of CM should be kept as low as 140-200mg I/ ml, the number of runs should be reduced by selection of the best primary views to avoid superimposition of vessels and intraarterial manometry should be used more frequently to evaluate questionable lesions. In such cases CO2 can replace iodinated CM but only at the expense of poor filling of dependent vessels and both over and underestimation of the stenoses. Three-dimensional rotational digital subtraction angiography (3D-DSA), an advanced angiographic technique in which sequential imaging is obtained with rotation of the X-ray tube, permits imaging of vascular structures in three dimensions and visualizes the course and structure of blood vessels more precisely than conventional angiography. This method enables detection of ulcerations of the iliac arteries as well as the dimensional structure of a dissection, which cannot be visualized using conventional single plane angiography (3). In addition 3D-DSA studies can potentially be used to create intraluminal images simulating angioscopy (4). Colour duplex scanning is usually the first technique which is applied to a patient with a clinical suspicion of aortoiliac occlusive disease. It is an easily repeatable method, free of any risk complications, causes minimal discomfort to the patient and is relatively cost effective. Currently the examination of the iliac arteries with this method is at best 84-92% sensitive (5). Despite its high negative predictive value in the diagnosis of lower-extremity arterial disease (6) and its potential to replace angiography in the selection of short aortoiliac lesions suitable for percutaneous endovascular treatment (7), very often the presence of excess bowel gas, obesity, and tortuous and calcified arteries that disallow optimal visualization of the aortoiliac segment mandates the arteriographic evaluation prior to the decision-making for treatment of the aortoiliac segment (8). The examination is more technically demanding in the iliac region compared to other segments of the peripheral arterial tree. The patient should be examined both supine and lying slightly on the opposite site. By moderate compression of the probe towards the

psoas muscle and directing it over the iliac arteries with slight medial retraction of the abdomen, satisfactory visualization of the CIA and EIA can be achieved. In 0-49% stenoses the peak systolic velocity (PSV) ratio (the stenotic PSV divided by the pre-or post stenotic PSV) is less than 2.5, 50-74% stenoses have a PSV ratio of 2.5 or more, 75-99% stenoses have an additional end-diastolic velocity (EDV) of 40 cm/s or more, and in occluded segments there is no detectable flow (9). In cases with very slow flow, especially in multiple lesions in series and in cases with a long distance between skin surface and the vessel ultrasound contrast media may be of value (10). Contrast-enhanced (CE) 3D MRA, which combines fast 3D spoiled gradient echo sequences with the i.v bolus administration of paramagnetic contrast has the advantage over conventional MR imaging strategies in that it is not flow-dependent. The 3D volume obtained from a single acquisition can be reformatted at a workstation and the aortoiliac anatomy viewed in any projection and with a variety of section thicknesses, making this MR imaging technique particularly useful in the evaluation of stenoses, aneurysms and dissection flaps. The use of multiplanar reformat and source image allows differentiation between high grade stenoses and occlusions (11). The availability of high-performance gradient systems permits the acquisition of complex 3D data sets within the confines of a comfortable breath-hold interval of under 30 seconds. CE 3D MRA is a technique with a potential for further improvement and shows superior sensitivity, specificity and accuracy compared to 2D TOF MRA for the detection of significant (>50%) stenosis and occlusion of aortoiliac arteries (12). Also with this technique the length of the iliac occlusion is accurately determined while it is overestimated on TOF MRA and is often uncertain on US (13). The excellent safety profile of gadolinium chelate contrast media makes MRA invaluable in patients with known allergy to iodinated contrast media and in patients with renal failure in which X-ray contrast media are contraindicated . The recent introduction of multislice CT (MSCT) scanners, offers the advantages of substantially shorter acquisition times, retrospective creation of thinner or thicker sections from the same raw data, and improved three-dimensional rendering with diminished helical artifacts. As a result of the shorter acquisition time, the contrast dose can be significantly reduced using a compact, fast, bolus of a high-concentration contrast material (e.g 400mg/ml iodine). Aortoiliac enhancement is improved with higher initial iodine and lowering continuing flow administration rates which lead to a more favorable plateau-like enhancement; therefore biphasic injections are preferable compared to uniphasic injections (14). Volume rendering (VR) is the first-line postprocessing technique in younger patients and in those with embolic disease, providing the lumen is not obscured by vessel wall calcifications. Maximum intensity projection (MIP), provides arteriography likeimages, but is limited by obscuring vessel calcifications as well ; in addition image editing (e.g removing bony elements) is needed before display. Curved planar reformations (sagittal/coronal) overcome the problem of atherosclerotic plaque calcification, but are time-consuming and tortuous vessels may lead to confusing artifacts. Indications Probably the most important clinical application of aortoiliac PTA/ stenting is in critical limb ischemia (CLI). Many, if not most, patients with CLI have multilevel disease, but proximal revascularization will suffice in approximately 75-85% of them, especially if one or more stenosis of high hemodynamic significance or complete occlusions are treated. Regarding claudicants, the decision for intervention varies widely, depending on the impact of the symptoms on the life style and professional requirements of an individual patient. In general, a painfree walking distance of less than 200 m corresponding to a stage IIb according to the Fontaine classification is accepted as a borderline to indicate invasive treatment. According to the transatlantic consencus paper on treatment of peripheral arterial disease (T.A.S.C), single iliac stenoses up to 10 cm long or a total of two stenoses <5 cm long [in either case not extending into the common femoral artery (CFA)], as well as unilateral CIA occlusions, should be treated either definitely or preferably with an endovascular procedure. Diffuse disease involving aorta and both iliac arteries, unilateral occlusions involving both the CIA and EIA, bilateral EIA occlusions, and iliac stenoses in a patient with an abdominal aortic aneurysm, should be treated surgically. Bilateral 5-10 cm long stenoses of the CIA and/or EIA not extending into the CFA, unilateral EIA occlusions not extending into the CFA, unilateral EIA stenosis extending into the CFA, and bilateral CIA occlusions, should be preferably treated by surgery, although CIRSE believes that in clinical practise these lesions are more commonly treated by endovascular techniques with a reasonable tech-

nical success rate (15). The application of stents has improved the immediate hemodynamic and long-term clinical results of aortoiliac PTA (16). The indications for their use, which are generally accepted in clinical practise are as follows: a) insufficient hemodynamic result of PTA due to elastic recoil based on residual pressure gradient b) massive, lumen-obstructing dissection c) treatment of chronic occlusions d) iliac artery ulceration associated with symptoms e) restenoses after previously performed PTA f) complex lesions for which primary stenting may give more satisfactory results.

Technique The most commonly used vascular access for aortoiliac revascularization is retrograde CFA access. The contralateral femoral or brachial access may be necessary when the very distal portion of the EIA / CFA are involved or when dealing with a total occlusion. Very often a combination of different accesses is needed in order to negotiate complex lesions. A stenosis is crossed with a combination of a soft-tip 0.035-in guidewire and a 5F straight or curved (Multipurpose or Angled Taper) catheter. A Bentson type wire which is extremely floppy, is often helpful. A hydrophilic guidewire (Terumo, Tokyo, Japan) by its extremely slippery nature when wet has proved invaluable in difficult lesions. In occlusions it is our policy to place initially 5F sheaths in both groins. This facilitates initial angiography from the contralateral side with a specific attention to the accurate imaging of aortic bifurcation in cases of CIA stenting, enables crossing of the occlusion from the contralateral side if the ipsilateral approach fails, and allows simultaneous pressure measurements from the aorta and the ipsilateral groin. We prefer to access EIA occlusions on an antegrade and contralateral approach using a 5F curved catheter (eg Cobra 1 and 2, Multipurpose or Angled Taper) in combination with a 0.035-in hydrophilic guidewire, usually of the straight type especially when the occlusion shows a small nipple just distal to the origin of the internal iliac artery. By contrast with EIA occlusions we principally start from an ipsilateral approach in case of a CIA occlusion. The first choice guidewire would be a conventional 0.035-in floppy tip straight guidewire directed using a curved catheter. A conventional wire provides more tactile response than a hydrophilic guidewire and is less likely to follow a subintimal route. If repeat gentle probing fails then we try the antegrade and contralateral approach; a 5F or 7F Sidewinder I catheter is introduced and the tip fixed within the ipsilateral occluded CIA. The occlusion is negotiated with a 0.035-in hydrophilic guidewire, preferably of the straight type especially if the tip of the catheter is wedged into to occluding material. Once it has transversed the occlusion, the distal tip of the wire is snared through the ipsilateral access sheath and pulled through. If contralateral approach fails, we return to the ipsilateral approach and we try a 0.035-in hydrophilic wire; in case this follows a subintimal route, we prefer not to try reentering the free lumen from ipsilateral approach as it is very likely for reentry to occur far away from the level of the occlusion, an event which may be disastrous for the procedure; at this stage we return to the antegrade contralateral approach using a 7F Sidewinder I for extra support in order to enter the subintimal space that has been formed from ipsilateral 5F catheter manipulation. Once the space has been entered with a guidewire, the wire is brought down close to the sheath, snared and pulled out. Occasionally the guidewire stops in the subintimal space, but it is still possible to capture it using a goosneck snare. Murphy et al have described an alternative technique in cases that reentry into the true lumen cannot be achieved with use of standard guidewires and catheters when the subintimal space is entered from ipsilateral approach ; an 18-gauge sharp needle is used to cross short distances and indent an occlusion balloon inserted from contralateral approach and indicating proper orientation toward the true lumen of the aorta (17). Should it prove impossible to cross the lesion from the femoral arteries, then the brachial approach should be used. The left brachial artery is to be prefered and a long sheath is inserted and its tip brought just proximal to the lesion. A 7F curved catheter may be required to provide the sufficient support to enter an occlusion followed by the same steps described above from the contralateral femoral artery. According to the traditional view in iliac artery interventions, the balloon diameter should be chosen equal to the distal reference vessel diameter of the target segment. When dilating aortic lesions one should

be very cautious because of the greater risk of rupture when larger vessels are dilated. A single balloon technique with “underdilation” of the aorta may not provide an excellent cosmetic result on the postangioplasty aortogram but is usually associated with a satisfactory clinical result. In heavily calcified lesions, the balloon diameter should be 1 to 2 mm smaller than the reference vessel diameter. The inflation has to be performed cautiously using a pressure not higher than 8 atm and an inflation time no longer than 60 seconds. In case of insufficient postangioplasty result stent placement is indicated. In general balloon expandable stents are prefered when a precise stent placement is required (for instance an ostial lesion), and selfexpanding stents when precision is not a critical factor and the vessel tapers in size. For balloon expandable stents an arterial sheath long enough to cross the lesion is recommended in order to avoid having the undeployed stent catch on the lesion and risk embolization or dislodgement of the stent. If a self-expanding sent is intended for use, then one with an intended diameter of approximatel 1mm greater than the measured vessel diameter should be used and if in doubt between two sizes -for example 8mm or 9mm stent-, then consider using the larger diameter as these stents cannot be expanded to a larger size if there has been a marginal error in measurements.

Summary Modern noninvasive imaging modalities have a vast potential as a replacement for diagnostic catheter angiography in the aortoiliac arterial region. Due to technical developments and estsablished expertise by the interventional radiologists even bilateral up to 10 cm long obstructing lesions can be treated with a high technical success by endovascular means. The lesions can be accessed from ipsilateral femoral, contralateral femoral and brachial approach and both the intraluminal and subintimal space can be used for successful recanalization. A variety of endovascular devices can accomplish this task. The application of stents has improved the immediate hemodynamic and long-term clinical results of aortoiliac PTA. References: 1. Waugh JR, Sacharias N. Arteriographic complications in the DSA era. Radiology 1992; 182: 243-246. 2. Bettmann MA, et al for the SCVIR contrast agent registry investigators. Radiology 1997; 203: 611-620 3. Yamamoto K, Maeda S, Kameoka N, et al. Rotational digital angiography for the evaluation of iliac artery disease. Int J Angiol 1999; 11: 4. Unno N, Mitsuoka H, Takei Y, et al. Virtual angioscopy using 3dimensional rotational digital subtraction angiography for endovascular assessment. J Endovasc Ther 2002; 9 (4): 529-34 5. Whiteley MS, Harris RA, Horrocks M. Aortoiliac segment examination with colour flow Duplex- a pilot study using Klean Prep to improve the image quality. Eur J Vasc Endovasc Surg 1995; 10(2): 192-7 6. Lai DT, Huber D, Glasson R et al. Colour duplex ultrasound versus angiography in the diagnosis of lower-extremity arterial disease. Cardiovasc Surg 1996; 4: 384-388 7. de Smet AAEA, Visser K, Kitslaar PJEHM. Duplex scanning for grading aortoiliac obstructive disease and guiding treatment. Eur J Vasc Surg 1994; 8: 711-715 8. Katsamouris AN, Giannoukas AD, Tsetis D, et al. Can ultrasound replace arteriography in the management of chronic arterial occlusive disease of the lower limb? Eur J Vasc Endovasc Surg 2001;21:155-159. 9. Legemate DA, Teeuven C, Hoeneveld H, et al. Value of duplex scanning compared with angiography and pressure measurements in the assessment of aortoiliac arterial lesions. Br J Surg 1991; 78: 1003-1008. 10. Fobbe F, Ohnesorge I, Reichel M, et al. Color-coded duplex sonography and ultrasound contrast medium in the study of periheral arteriesinitial clinical experiences. Ultraschall Med 1992; 13:93 11. Wikstrom J. On contrast-enhanced magnetic resonance angiography of the aortoiliac arteries. Acta Radiol Suppl 2001; 42 (426):1-30. 12. Bourlet P, De Fraissinnette B, Garcier JM, et al.Comparative assessment of helical CT-angiography, 2D TOF MR-angiography and 3D gadolinium enhaced MRA in aorto-iliac occlusive disease. J Radiol 2000 ; 81(11) : 1619-25 13. Lundin P, Svensson A, Henriksen E, et l. Imaging of aortoiliac arterial disease. Duplex ultrasound and MR angiography versus digital subtraction angiography. Acta Radiol 2000; 41(2): 125-32 14. Fleischmann D. Present and future trends in multiple detector-row CT applications: CT angiography. Eur. Radiology 2002; 12(suppl.2) S11-

S16. 15. TASC. Management of Peripheral Arterial Disease. Internat. Angiol. 2000;19: suppl.1 16. Bosch JL, Hunink MGM. Metaanalysis of the results of percutaneous transluminal angioplasty and stent placement in aortoiliac occlusive disease. Radiology 1997; 204: 87-96 17. Murphy TP, Marks MJ, Webb MS. Technical note: Use of a curved needle for true lumen re-entry during subintimal iliac artery revascularization. JVIR 1997; 8: 633-636.

7.2 Aortoiliac PTA/Stenting Results and Complications. E. N. Brountzos, G. Papaioannou, D. A. Kelekis; 2nd Radiology, Athens University, Athens, Greece. PTA and stenting of iliac arteries have since long been proven to result in successful treatment of aortoiliac obstructive vascular disease. Published series of iliac artery PTA date back to 1980, while the preliminary results of the clinical use of the first clinically used stent, the Palmaz stent, were published in 1988 [1,2]. Indications for PTA of an iliac artery occlusive lesion are identical to surgical indications and include a) life limiting claudication, b) limbthreatening ischemia, c) vasculogenic impotence. In addition iliac artery PTA are indicated prior to a surgical bypass procedure to increase blood inflow [3]. Indications for iliac artery stenting include a) suboptimal results of PTA, b) total occlusions, c) restenoses, and d) dissections [4]. According to the Society of Interventional Radiology (SIR) [5], iliac artery lesions are classified in categories 1 to 4 with increasing degree of severity according to their anatomic characteristics: a) Category 1- concentric, non-calcified stenoses < 3 cm in length. b) Category 2- stenoses 3-5 cm in length, or calcified, eccentric stenoses < 3 cm in length. c) Category 3 – stenoses 5-10 cm, or occlusions 10 cm, occlusions > 5 cm, extensive bilateral aortoiliac occlusive disease, or a concomitant abdominal aortic aneurysm. Category 1 lesions respond well to PTA, while PTA of lesions category-2 and -3 yields less favorable results. Surgery is indicated for category-4 lesions [3,5]. With the use of stents the results of percutaneous procedures improved even with more difficult lesions. The indications for stenting are different among different institutions; while some centers use stents in all iliac occlusive lesions, most authors agree that stenting should be reserved for the failures or complications of PTA, or when dealing with total occlusions [6]. Technical success of PTA and stenting is very high (>90%) according to multiple literature reports. A meta- analysis published in 1997, including data from six PTA studies (1,300 patients), and eight stent studies (816 patients), showed that stenting was more technically successful than PTA (96% versus 91%, p<0.05) [2]. Complications occur in 9-10% of the patients, but only 2-2,5% require treatment. Systemic complications include sepsis (0.2%), contrast induced acute renal failure (0.3-1%), cardiac and pulmonary problems [7-9]. Major local complications include: distal embolization, aorto-iliac artery rupture, hematoma, pseudoaneurysm, and arteriovenous fistula. Procedure related mortality is approximately 0.4%, but 30-day mortality is higher (approximately 1.4%) especially in elderly patients with critical ischemia [2,7]. Nevertheless complication and mortality rates were not different between PTA and stenting procedures [2]. Patency of iliac arteries treated with PTA and stenting is generally favorable: primary patency of PTA of iliac stenoses for the management of claudication at 1, 3 and 4 years of is 79%, 68% and 65% respectively; primary patency of stenting for stenoses is 91%, 80%, and 77% respectively. Primary patency of PTA of iliac occlusions at 1, 3, and 4 years is 66%, 57%, and 54% respectively. Primary patency of stenting for iliac occlusion is 72%, 64%, and 61% respectively. These rates are worse when treating patients with critical ischemia: 4-year patency was 53% for stenoses and 44% for occlusions after PTA to treat critical ischemia; 4-year patency was 67% for stenoses and 53% for occlusions after stenting to treat critical ischemia [2]. When compared with PTA alone stent treatment of iliac artery lesions resulted in a higher technical success rate, with similar mortality and morbidity. The risk of long-term failure was reduced by 39% with stent treatment as compared with PTA alone [2]. These favorable long-term results are corroborated by a recent longterm retrospective study including 110 patients [10]: Primary stent

patency at 5 and 10 years was 66% and 46% respectively. Secondary patency was 79% and 55% respectively. When compared to surgery, PTA and stenting of iliac artery have worse patency rates. A meta-analysis including 23 studies of aorto-iliac or aorto-femoral bypass grafts reports patency rates at 5 and 10 years of 91% and 86.8% respectively, for patients with claudication, and of 87.5% and 81.8% for patients with ischemia [11]. In a randomized controlled trial comparing PTA to surgery in 292 patients, 5-year primary patency of PTA was 62%, whereas that of surgery was 81% [12] Nevertheless surgery is usually avoided in this patient group because of the significantly higher mortality (4%) and major complication rate (21%) [13]. Because of the better immediate and long-term results of stenting primary stent placement would be the treatment of choice for patients with aorto-iliac occlusive disease. Nevertheless there are serious concerns with the increased costs of this procedure, that might have deleterious impact on health care budgets [14]. A randomized clinical trial and cost-effectiveness analysis was performed in the Netherlands to compare primary stenting with PTA with selective stenting –when PTA failed: Initial technical results and complications rates were similar between the 2 treatment strategies. 2-year patency rates were also similar-70% versus 71% respectively. The cost of stenting was higher. The investigators concluded that PTA with selective stenting is the treatment of choice [15]. Additional publications of the same group showed that quality of life improves equally after primary stent placement and PTA with selective stent placement, but the latter costs less ($957) [14,16]. Although selective stenting is advocated after failure of PTA, criteria for PTA failure vary widely. Using intraarterial residual transtenotic pressure gradient as an indication for stenting Kamphuis et al found a complete lack of uniformity in the literature: The use of the various thresholds would have led to stent placement in 4-87% of the Dutch Iliac Stent Trial Study population [17]. The actual stenting rate in this group using a 10mm Hg residual pressure threshold was 43%. This discrepancy and the difficulty to clearly define lesions as stents candidates encourages many interventionalists to adopt a primary stenting approach. There is a strong feeling that the development of newer flexible stents with low profile broadens the applicability of percutaneous treatments with much better than previously results. In short stents may treat lesions previously managed exclusively by surgery [18]. Between 1996 and 2001 we have treated 200 limbs in 159 patients with peripheral vascular occlusive disease with stents. Pre-procedural clinical stage (Fontaine) was as follows: I: 1.2%, IIa: 12.8%, IIb: 69.8%, III and IV: 16.2%. Lesions' category was as follows: 1: 10.7%, 2: 55%, 3: 21.9%, 4: 12.4%. 10.5% of the lesions were occlusions, the remaining were stenoses (>75%). We have used the Palmaz stent in 58.5%, the Wallstent in 28.1%, and others (Symphony, Expander, Jostent, Strecker) in 13.4%. In 36 patients aortic bifurcation reconstruction was performed using the kissing stent technique [18]. Follow-up was performed prospectively with physical examination, ABI measurement, and Color duplex sonography. Intra-arterial DSA was performed when restenosis was diagnosed with the non-invasive tests 19.42%). Technical success was 93.6%. There was substantial clinical inprovement; clinical Fontaine stage after stenting was as follows: I: 71,2%, IIa: 19.4%, IIb: 7.1%, III-IV: 2.4%. We encountered only 2 pseudoaneurysms, and 5 hematomas –requiring only observation. There was zero 30-day mortality. Follow-up time ranged from 1 to 66 months (mean: 22.10±16.43 months). 10 patients died during the follow-up: 5 from malignant neoplasms, 4 from cardiovascular causes, one from chronic obstructive pulmorary dissease. Follow-up was available in 173 limbs (139). Primary patency at 12, 24, 36, and 48 months was 89.7%, 81%, 76%, and 66% respectively. In our experience iliac artery stenting was a safe procedure with good short-term results. Long-term patency was moderate. References: 1. Palmaz JC, Richter GM, Noeldge G, et al. Intraluminal stents in atherosclerotic iliac artery stenosis: preliminary report of a multicenter study.Radiology 1988; 68: 727-731. 2. Bosch JL, Hunink MGM. Meta-analysis of the results of percutaneous transluminal angioplasty and stent placement for aortoiliac occlusive disease. Radiology 1997; 204: 87-96. 3. Pentecost MJ, Criqui MH, Dorros G, et al. AHA Medical/Scientific Statement. Guideline for percutaneous transluminal angioplasty of the

abdominal aorta and lower extremity vessels. Circulation 1996; 94: 511531. 4. Johnston KW. Iliac arteries: Reanalysis of results of balloon angioplasty. Radiology 1993; 186: 207-212. 5. Standards of Practice Committee of the Society of Cardiovascular and Interventional Radiology. Guidelines for percutaneous transluminal angioplasty. Radiology 1990; 177: 619-626. 6. Sheeran SR and Murphy TP. Aortoiliac angioplasty and stenting. In: Kandarpa K and Aruny JE, eds. Handbook of Interventional Radiologic procedures: A Lippincott Williams & Wilkins, 3rd edition 2001: 173-182. 7. Becker GJ, Katzen BT, Dake MD. Noncoronary angioplasty. Radiology 1989; 170: 921-940. 8. McDermott JC, Crummy AB. Complications of angioplasty. In: Kadir S, ed. Current practice of interventional radiology: BC Decker, 1991:311319. 9. Brountzos E N, Athanasoulis C A. Contrast material induced acute renal failure: clinical course and outcome. Radiology 1992 (suppl); 185: 317. 10. Schurmann K, Mahnken A, Meyer J, et al. Long-term results 10 years after iliac arterial stent placement. Radiology 2002; 224: 731-738. 11. De Vries SO, Hunink MG. Results of aortic bifurcation grafts for aortoiliac occlusive disease: a meta-analysis. J Vasc Surg 1997; 26: 558569. 12. Wilson SE, Wolf GL, Cross AP. Percutaneous transluminal angioplasty versus operation for peripheral arteriosclerosis: report of a prospective randomized trial in a selected group of patients. J Vasc Surg 1989; 9: 19. 13. Murphy TP, Khwaja AA, Webb MS. Aortoiliac stent placement in patients treated for intermittent claudication. J Vasc Interv Radiol 1998; 9: 421-428. 14. Bosch JL, Haaring C, Meyerovitz MF, et al. Cost-effectiveness of percutaneous treatment of iliac artery occlusive disease in the United States. Am J Radiol 2000; 175: 517-521. 15. Tetteroo E, Van der Graaf Y, Bosch JL, et al. Randomized comparison of primary stent placement versus primary angioplasty followed by selective stent placement in patients with iliac –artery occlusive disease. Dutch Iliac Stent Trial Study Group. Lancet 1998; 351: 1153-1159. 16. Bosch JL, Tetteroo E, Mali WP, Hunink MG. Iliac artery occlusive disease: cost-effectiveness analysis of stent placement versus percutaneous transluminal angioplasty. Dutch Iliac Stent Trial Study Group. Radiology 1998; 208: 641-648. 17. Kamphuis AGA, Van Engelen AD, Tetteroo E, Hunink MG, Mali WPT. Impact of different hemodynamic criteria for stent placement after suboptimal iliac angioplasty. J Vasc Interv Radiol 1999; 10: 741-746. 18. Stainken B. Primary stenting vs primary angioplasty and selective stenting for PVD. J Vasc Interv Radiol (suppl) 2002; 13: P159-P161. 19. Scheinert D, Schroeder M, Balzer JO, Steinkamp H, Biamino G. Stent-supported reconstruction of the aortoiliac bifurcation with the kissing balloon technique. Circulation 1999; 100: II-295-II-300.

Special Session Percutaneous Endurology 8.2 Nephrostomy: complications and their management M. N. Özmen; Radiology, Hacettepe University, School of Medicine, Ankara, Turkey. Percutaneous nephrostomy (PCN) is a well established method for drainage of an obstructed urinary collecting system. Although PCN represents some minor and few major complications, these complications can easily be managed by various methods. Secondary urinary manipulations like percutaneous nephrolithotomy or antegrade ureteral stenting, increases the risk of complication. However life threatening complications are rare. Major complications occur in 3-7 % of cases (1-5). These complications are defined as complications requiring therapy, prolonged hospitalization, increase level of care, or that have permanent adverse sequel or result in death (6). Some authors limited major complications as death or complication requiring surgical management. In such studies major complications was reported to be as low as 0.29% (7). Minor complications occur in 7-10 % of cases (1,2). These cases require no or minimal therapy, sometimes causing overnight admission for observation only (6). In emergency PCN, overall complication rate is up to 34% (8). PCN has a low mortality of 0.2 % compared with a surgical mortality of 6 %

(2,5,7-9). Standards Committee of Society of Interventional Radiology published Standards of Practice of PCN. In this article thresholds and reported rates for major complications of percutaneous nephrostomy was reported (table 1). It is advocated that when complication rates exceed a maximum threshold, a review should be performed to determine causes and to implement changes, if necessary (6). Pain after PCN might be due to too close puncture to a rib. Such cases are best managed by intercostal nerve block. Catheter tip pushing against renal pelvic wall may be the cause of pain. It is managed by repositioning the catheter. Catheter related problems may occur during drainage. These might be as a result of occlusion of drainage holes due to debris or blood clots. Therefore, particularly in pyonephrosis cases, large bore drainage catheters with multiple side holes are preferred. Blood clots can be managed by small amount of saline flushes. Sometimes catheter kink might be the cause of catheter blockade. In such cases catheter must be repositioned under fluoroscopic guidance. If total obstruction occurs, catheter should be replaced either by managing a guide wire through the catheter or by placing a peel-away sheath over the catheter. Catheter dislodgment is a relatively frequent problem encountered during PCN drainage. It has been reported to be as high as 15% to 25% (3,4). Although several skin fixation devices have been available, dislodgment of nephrostomy catheters is still a problem for interventional radiologist. Self-retaining catheters are less prone to catheter dislodgment. In a series where this type of catheter was used, catheter dislocation rate was 0.9% (5). When dislodgment occurs, the sooner the patient is managed, the better. Usually tract is patent for 6 to 8 hours after dislodgment. Blind tract negotiation and excessive amount of local anesthetic infusion should be avoided. Tract should be delineated with contrast medium injection under fluoroscopy. After visualization of the tract, tract negotiation can be carried out with a 5 Fr 45-degree catheter along with a soft, straight, flexible tip guide wire or hydrophilic guide wire. If the patient is referred to hospital after more than 72 hours following dislodgment, the tract is usually closed. In that case, creating a new PCN is the best method. Urine leakage which can be seen as urinoma, is usually a result of ineffective drainage. Sometimes, particularly in percutaneous nephrolithotomy cases, renal pelvis or ureter lacerations may occur. Unless there is not an effective drainage, these tears heal spontaneously. The reason for ineffective drainage must be evaluated by nephrostogram. If there is an obstruction, kink, or malposition of the catheter, appropriate measures must be taken. Sometimes large bore catheters are needed for this purpose. Rarely it is necessary to drain urinoma with a separate drainage catheter. In cases where infected urinoma turns out to be a perinephritic abscess, prompt percutaneous abscess drainage should be performed combined with appropriate antibiotics regimen. In percutaneous nephrolithotomy cases, perforations usually occur in cases of small renal pelvis, bifid collecting systems or renal pelvis filled with calculi. Also perforations are more common when nephrostomy tract is dilated with metal or semirigid dilators instead of balloon dilation. Infection or urosepsis is a serious complication of PCN. The risk of septic complications is reported to be 1-3% (2,6,9). Temperature rise more than 2 Fo (1.1 Co) and lasting more than 6 hours is accepted as a complication. In a series of emergency PCN, although all patients presented transient increase in body temperature, only in 14 % of them temperature rise was considered as a complication (with a rise of more than 2 Fo and lasting more than 6 hours) (8). Periprocedural antibiotic coverage is effective for preventing urinary infection or urosepsis (5,10). Thus most centers either give single dose of intravenous antibiotics or administer a 36 to 72 hours antibiotic coverage. In pyenophrosis or urosepsis cases, antibiotic regimen is continued for several days after PCN. Cochran has shown that in the high risk patients (patients with positive urine analysis and/or urine culture, struvite stone, or urinary ostomy), risk of sepsis was 50% when antibiotics were not used, however the risk was decreased to 9% when antibiotics were used (11). One of the most important reasons of urosepsis after PCN is irrigation with excessive amount of saline with improper use of aseptic techniques or obtaining antegrade pyelography with use of excessive amount of contrast media. In that case, due to high pressure in collecting system, pyelotubular or pyelovenous back-flow can easily occur resulting in bacteriemia. Over distention of collecting system should be avoided (1,2). Therefore radiologists must be careful while performing antegrade pyelography after PCN. Particularly in pyonephrosis cases, this can easily lead to urosepsis, even death. If signs of infection occurs after PCN in a patient without prior pyonephrosis, besides prompt administration of appropriate anti-

biotics regimen, drainage system must be examined for partial obstruction. If needed, catheter should be exchanged. It is well known that, in spite of infected urine, adequately drained patients remain asymptomatic. Like all the secondary urinary manipulations, percutaneous nephrolithotomy procedure should not be performed on patients with obstructed infected collecting system unless infection is brought under control with parenteral antibiotics conjoined with PCN. In such cases manipulations must be minimized. In most cases of PCN, transient hematuria occur which clears within 24-48 hours. This is not considered as a complication and no treatment is needed (3,10). However bleeding can be an immediate or a delayed problem. If urine does not clear in 48 hours due to minor bleeding and there is no falling of hematocrit, it is usually due to side holes to be positioned in the parenchyma. It can easily be managed by catheter repositioning. Sometimes renal pelvis may be filled with blood clot and side holes of the catheter might be blocked with this clot. In that case catheter must be flushed with saline, however excessive flushing must be avoided as the clot in the renal pelvis tamponades further bleeding into collecting system. The aim of the flushing should be keeping catheter open, not to resolve the clot in the renal pelvis. The clot in the renal pelvis will resolve spontaneously. Slow oozing of blood with a drop in the hematocrit is conservatively treated with bed rest and transfusion. The threshold for transfusion is reported as a hematocrit of 30% or less or, in the preexisting anemia, a drop of 4% in the hematocrit (8). Although transfusion requirement is about 1-4% after PCN, this complication occurs about 10 % following stone extraction. It has been reported that in cases of percutaneous staghorn stone removal, transfusion requirement is as high as 57% (4, 12). Farrel reported that while only 2% of non-coagulopathic patients required transfusion in their series of PCN, 4% of coagulopathic patients, all of whom treated prior to the procedure to correct coagulopathy, required transfusion. However, while in patients with a baseline platelet count more than 100.000/mm3 only 5% transfusion required, in patients with a baseline platelet count less than 100.000/mm3 50% transfusion requirement occurred despite routine preprocedural administration of platelets (5). Major vascular injury may cause severe bleeding. Usually the reason is arterial laceration, pseudoaneurysm or arteriovenous fistula formation. Larger caliber catheter insertion usually solve the bleeding problem due to vascular laceration. Sometimes balloon catheters are preferred for this purpose (9,13). If these methods are not successful, then selective catheterization must be performed followed by embolization. Very rarely, if embolization fails, surgical intervention might lead to partial or total nephrectomy (9,13). It has been stated according to the results of 2.200 cases of percutaneous nephrolithotomy that angiography and embolization was needed only in 0.8 % of cases (13). Angiographic intervention was advocated under 3 conditions: 1) in the immediate postoperative period when the clamping of the nephrostomy tube and a tamponade balloon catheter fail to control hemorrhage, 2) in the early postoperative period (2 to 7 days) when patient requires 3 or 4 units of blood after replacement of the initial blood loss and 3) for sudden hemorrhage more than 7 days postoperatively (13). The bleeding usually arises from injury of the anterior or posterior segmental arteries rather than the smaller peripheral arteries. Atherosclerosis decreases the ability of a lacerated vessel to close because of the loss of its normal muscle and elastin layers. Therefore, patients with atherosclerosis and hypertension are at higher risk for hemorrhage (13,14). Arteriovenous fistula or pseudoaneurysm formation more common in patients who undergo renal biopsies rather than PCN patients. Severe bleeding or hemorrhage due to arteriovenous fistula or pseudoaneurysm can be resolved with selective catheterization and embolization. Although thoracic complications are rare in PCN (0.1-0.2%), it is not infrequent in patients who undergo percutaneous nephrolithotomy (515%) (5,6,9,15). Particularly in cases where stone removal is performed via supracostal approach, there is a high risk of injuring the pleura and lung. It has been reported that puncture done at the level of 10th intercostal space has a high risk of thoracic complications up to 66% (16). The posterior costodiaphragmatic recess usually extends to the level of the 12th rib in prone position. Therefore, especially where the upper pole intervention is required through 11th intercostal space, careful fluoroscopic observation is needed to avoid puncturing the lung at the time of initial nephrostomy placement. During percutaneous nephrolithotomy, routine use of 30 Fr working sheath is believed to lessen the risk of leakage of air and fluid into pleural space in these patients (16). Urine leakage resulting in hydrothorax can be managed by inserting a large

bore nephrostomy catheter. If needed, chest tube can be placed under fluoroscopic guidance in cases of pneumothorax or hydrothorax (15). Colonic perforations are very rare, less than 0.5 % in percutaneous nephrostomy cases (6). This risk is higher in cases of percutaneous nephrolithotomy (2.7%) (15). It is usually encountered in patients with very little retroperitoneal fat or renal anomaly such as horseshoe kidney as colon tends to have more posterior location. High risk patients for colon injuries during percutaneous nephrolithotomy are young, lean males with minimal retroperitoneal fat (17). In addition, patients in whom the nephrostomy puncture is performed too laterally have an increased risk of colon injury. The presenting symptoms are extremely variable. The clinical presentations range from subtle (low grade fever, paralytic ileus, and leukocytosis) to dramatic (fecaluria, pneumaturia, rectal bleeding and obvious peritonits). Sometimes it is incidentaly diagnosed during routine nephrostogram. Retroperitoneal colon injuries can be successfully managed conservatively with early recognition and appropriate drainage of the urinary and intestinal tracts separately besides administration of broad spectrum antibiotics. However intraperitoneal colon injuries after percutaneous renal stone extraction demand open extraction and repair. It has been shown that the experience of the operators play little role in the initial success rate of procedure after they have performed PCN 10 or more per year. However, the level of experience influence the number of procedures that needed to be repeated, the incidence of minor complications and the mean fluoroscopy and procedure times (18). On account of low complication rate, it has been recommended to perform PCN on an outpatient basis in selected cases (10,11,14). However those with coagulation disorders, signs of urosepsis, staghorn calculi (especially struvit) and hypertension should be treated as inpatients.

Table 1 Thresholds (%) for Major Complications of Percutaneous Nephrostomy* Complication

Reported Rate

Threshold

• Septic shock (fever, chills with hypotension, requiring

1-3

4

• Septic shock (in setting of pyonephrosis)

7-9

10

-PCN alone

1-4

4

-With PCNL

12-14

15

0.1-1

1

0.2

<1

-PCN alone

0.1-0.2

<1

-With PCNL or endopyelotomy (intercostal procedures)

8.7-12

15

4-7

5

• Hemorrhage (requiring transfusion)

• Vascular injury (requiring embolization or nephrectomy) • Bowel transgression • Pleural Complications (pneumothorax, empyema, hydrothorax, hemothorax)

• Individual Threshold Complications that result in unexpected transfer to an intensive care unit, emergency surgery or delayed discharge from the hospital

Note.-PCN = percutaneous nephrostomy; PCNL = percutaneous nephrolithotomy. *From: Ramchandani et al. J Vasc Intervent Radiol 2001; 12:1247-1251 tube insertion for hydrothorax after percutaneous nephrolithotomy. Urology 2002; 60:1098-1099 References: 16. Narasimham DL, Jacobsson B, Vijayan P et al. Percutaneous neph1. Vehmas T, Kivisaari L, Mankinen P et al. Results and complications of rolithotomy through an intercostal approach. Acta Radiol 1991; percutaneous nephrostomy. Ann Clin Res 1988; 20:423-427 32:162-165 2. Stables DP. Percutaneous nephrostomy: Techniques, indications and 17. Gerspach JM, Bellman GC, Stoller ML, Fugelso P. Conservative manresults. Urol Clin North Am 1982; 9:15-29 agement of colon injury following percutaneous renal surgery. Urology 3. Von der Recke P, Nielsen MB, Pedersen JF. Complications of ultra1997; 49:831-836 sound-guided nephrostomy. A 5-year experience. Acta Radiol 1994; 18. Lee WL, Patel U, Patel S, Pilliari GP. Emergency percutaneous neph35:452-454 rostomy: results and complications. JVIR 1994; 5.135-139 4. Gupta S, Gulat1 M, Shankar U, et al. Percutaneous nephrostomy with real time sonographic guidance. Acta Radiol 1997; 38:454-457 Special Session 5. Farrel TA, Hicks ME. A review of radiologically guided percutaneous nephrostomies in 303 patients. J Vasc Intervent Radiol 1997; 8:769-774 Peripheral Venous Imaging and Interventions 6. Ramchandani P, Cardella JF, Grassi CJ, et al. Quality improvement guidelines for percutaneous nephrostomy. J Vasc Intervent Radiol 2001; 9.2 12:1247-1251 Peripheral Venous Interventions 7. Kaskarelis IS, Papadaki MG, Malliaraki NE, et al. Complications of perS. Trerotola; cutaneous nephrostomy, percutaneous insertion of ureteral endoprosHospital of the University of Pennsylvania, Philadelphia, United States. thesis, and replacement procedures. Cardiovasc Intervent Radiol 2001; 24:224-228. Peripheral Venous Interventions 8. Lee WL, Mond DJ, Patel M, Pilliari GP. Emergency percutaneous nephBroadly defined, peripheral venous interventions make up an increasrostomy: Technical success based on level of operator experience. J Vasc ingly large proportion of the practice of Interventional Radiology. Intervent Radiol 1994; 5.327-330 Hemodialysis access interventions, peripherally inserted central cathe9. Dyer RB, Regan JD, Kavanagh PV, et al. Percutaneous nephrostomy ters, and the treatment of venous vascular malformations are areas in with extensions of the technique: Step by step. Radiographics 2002; which growth has leveled off, while the field of venous thrombosis and 22:503-525 its sequelae of venous insufficiency is beginning to see substantial 10. Hogan MJ, Coley BD, Jayanthi VR, et al. Percutaneous nephrostomy growth. To cover all of these subjects is well beyond the scope of this in children and adolescents: outpatient management. Radiology 2001; presentation and I would like to confine my subject matter to DVT and 218:207-210 venous insufficiency. However, one must remember that even with this 11. Cochran ST, Barbaric ZL, Lee JJ, Kashfian P. Percutaneous nephrosdistinction, there is overlap with the other above-mentioned subjects: tomy tube placement: An outpatient procedure? Radiology 1991; for example, PICC lines may be complicated by venous thrombosis. 179:843-847 Deep vein thrombosis is estimated to affect more than 2 million people 12. Lee WJ, Snyder JA, Smith AD. Staghorn calculi: endourologic manannually (1) and the sequelae of venous insufficiency develop in up to agement in 120 patients. Radiology 1987; 165;85. 2/3 of these patients (2). It appears that only a subset of patients with 13. Kessaris DN, Bellman GC, Pardalidis NP, Smith AG. Management of DVT will be candidates for percutaneous management, including those hemorrhage after percutaneous renal surgery. J Urol 1995; 153:604with phlegmasia cerulea dolens, iliofemoral thrombosis, isolated iliac 608 disease, young age, and thrombosed IVC filters; this subset may be in 14. Gray RR, So CB, McLoughlin RF, et al. Outpatient percutaneous the vicinity of 20% of all DVT but represents those at highest risk for nephrostomy. Radiology 1996; 198:85-88 post phlebitic syndrome (3). Even confining treatment to this subset, 15. Ogan K, Pearle MS. Oops we got in the chest: fluoroscopic chest

the potential for growth of this area is staggering. Yet DVT thrombolysis has been available, either systemically or as catheter directed therapy, for decades (4-9), with ample published evidence including prospective randomized trials (for systemic therapy) to support its use, and yet only a tiny minority of patients receive these therapies. The answer to this paradox is found, in my opinion, in two distinct realities. The first is a relative lack of outcomes data regarding DVT thrombolysis, in particular catheter-directed thrombolysis. Specifically, we have failed to prove that in doing catheter directed lysis, we are preventing or reducing the post phlebitic syndrome, or for that matter reducing lost work days, hospitalization, etc. Indeed, the second reality is intimately tied to the hospitalization issue. While previously, management of acute DVT involved days of hospitalization for heparin therapy and conversion to oral anticoagulation, present management involves exclusively outpatient therapy with low molecular weight heparin or similar agents. Since catheter directed therapy for DVT, as described in the Venous Registry, averaged 53 hours of hospitalization (6), it is highly unlikely that we will be able to show reduced hospitalization unless DVT lysis can be accomplished entirely as an outpatient. It is my belief that this is possible, but only with combination therapy marrying mechanical devices and thrombolytic agents. Further, since the burden of proof is on the IR community, we need to construct prospective randomized trials of DVT thrombolysis that challenge conventional therapy. On the surface, this should be easy. Anticoagulation therapy only prevents further clot formation and does little or nothing to spare valves and prevent post phlebitic syndrome. Early recanalization in patients treated with anticoagulation is associated with better outcomes (2). The theoretical advantages of DVT lysis are just that, but other potential outcomes variables to explore include more rapid relief of symptoms and return to work. It is unlikely that absolute cost reductions can be shown, thus more careful cost effectiveness analyses taking productivity into consideration will have to be performed. Until recently, such trials had been hampered by a number of factors, including the lack of FDA approval for any thrombolytic agent for DVT lysis. However, several investigators have now obtained Investigational New Drug exemptions (IND) and single- and multicenter US clinical trials are beginning to get underway. These trials consist chiefly of infusion thrombolysis, and it may prove difficult to enroll patients in them for reasons cited above. It is my belief that true buy-in from other practitioners treating DVT, such as Emergency Room and Primary Care physicians, will only come in the form of a randomized trial. I am aware of only one such trial in the US, comparing low molecular weight heparin to a combination of mechanical device and lytic agent. No results from this trial, which has only recently begun, are available. I am aware of several other trials of mechanical devices alone or in combination with lytic agents, but these are not randomized against standard medical therapy. In the US, such trials will require IND and probably also IDE (Investigational Device Exemption) approval from the FDA and will almost certainly be locally investigator driven, rather than industrydriven. At present I am aware of only one such trial in the making, but there may well be others. Looking briefly at the results of existing trials, DVT lysis appears reasonably safe and effective. The Venous Registry described treatment of 303 limbs in 287 patients with one year follow-up. Complete thrombolysis was achieved in 31% and partial (ie >50%) lysis in 52%.12 month primary patency was 80%. Valvular reflux was evaluated and showed better results in those patients with complete lysis (28% reflux) compared with 58% for the entire group (6). The average dose in the DVT registry was 7.8 million units of urokinase. Bjarnason et al treated 87 limbs in 77 patients with a technical success rate of 70% and 63% 1 year primary patency (78% secondary) for iliac veins and 40% primary (51% secondary) patency at one year for femoral veins (9). Valvular function was not evaluated. Comerota et al described better quality of life scores for patients undergoing thrombolysis compared with those receiving anticoagulation therapy at 16 month mean follow-up (8). Several smaller series have reported similar results. To my knowledge there has been only one prospective randomized trial comparing catheter directed lysis to anticoagulation (10). Preliminary results from this series (n=35) showed 72% patency at 6 months for lysis compared to only 12% for anticoagulation (p<0.001). Venous reflux was also higher in the anticoagulation group (41%) compared to the lysis group (11%, p=0.04). The consensus regarding catheter-directed lysis appears to be that if offers better results with fewer complications that systemic therapy (2,7). Despite this, it is not in widespread use as the primary therapy for iliofemoral DVT. Clinical experience with mechanical devices is quite limited at present,

though probably more widespread than the published literature would suggest. Yamauchi et al reported partially successful use of the salinejet aspiration catheter on two occasions in a single patient (11). Uflacker reported use of the Amplatz device in three patients with iliac and/or caval thrombosis. Adjunctive urokinase and stents were required in two of three patients. All three patients had improvement in their symptoms (12). Vorwerk et al described a single case in which a combination of stents and the Hydrolyser were used to treat iliocaval thrombosis (13). The largest series to date reported 28 limbs in 20 patients using a variety of devices combined with lytic agents. 82% procedural success was achieved, importantly with lower does of lytic agents than previous reports of infusion alone (eg 2.7 million units of urokinase) and shorter procedure times (mean 23 hours infusion time)(14). Late outcomes and valvular reflux were not reported. The reports share the common thread of partial success in clot removal using a mechanical device supplemented by thrombolytic agents (or vice versa-some start with a device and convert to lytics, while others start with lytics then use a device) and/or stents. To summarize, widespread application of catheter-directed DVT lysis is likely to occur only after it has been objectively shown in a large prospective, randomized fashion, to be superior to modern outpatient anticoagulation therapy using a variety of measures. Such a trial would be extremely newsworthy, and if published in a high visibility journal such as NEJM, Lancet or JAMA could dramatically change the volume of this potentially valve-sparing and post-phlebitic syndrome preventing therapy worldwide. On the other hand, continued performance of nonrandomized “feasibility” and registry studies will almost certainly keep this procedure in the stagnation it has experienced for the last decade. The small randomized study described above is a step in the right direction (10). Intimately related to DVT lysis, but even more experimental at present, is the percutaneous placement of valve prostheses for the treatment of venous insufficiency (VI). It has been estimated that there are 24 million Americans with venous insufficiency (2). Although a small percentage have primary disease, the vast majority have lost valves as a result of prior DVT. Given the likelihood that non-valve sparing treatment (ie anticoagulation) is likely to remain the gold standard for the foreseeable future, the potential applications for a valve prosthesis are enormous. To my knowledge, there are three percutaneous prostheses in clinical trials in Europe and North America, but virtually no data at present other than animal data. The underlying premise of valve replacement is obvious; rather than treating varicosities, ulcers and other symptoms of the post phlebitic syndrome, the underlying cause can be treated. This premise is deceptively simple, however, since the majority of patients with postphlebitic syndrome have both venous obstruction (from incomplete recanalization) and reflux; valve prostheses may or may not improve outcomes in this segment of the VI population. This has hampered enrollment in trials of venous prostheses, whether percutaneous or surgically placed. Further, developing a device that functions short term in an animal has proven relatively easy compared to maintaining patency and function in humans. Indeed, no device to date has been shown to do so in more than a handful of patients. To briefly review the animal work in this area, Pavcnik et al have described a prosthetic valve made of porcine small intestinal submucosa (SIS) based an a “square stent” platform. These authors have reported the longest follow-up for a percutaneously placed valve prosthesis, with 22/26 prostheses retaining function at 6 months in a sheep model (15). Gomez-Jorge et al described a glutaraldehyde-fixed bovine valve mounted on a stent. This device was used in 11 animals with limited follow-up (maximum 2 weeks), but the feasibility of percutaneous delivery was well-documented (16). In summary, while valve prostheses may well become a clinical reality in this decade, they will likely only benefit a subset of VI patients and is not clear at present how large this subset will be. Patients with primary valvular venous insufficiency, while forming the small minority of this population, will benefit the most. Whether Interventional Radiologists will be actively involved in the placement of these devices will depend largely on whether they have developed clinical practices devoted to treatment of patients with DVT and VI, a segment traditionally treated by vascular surgeons. Bridging these two concepts is the practice of venous ablation, presently seeing dramatic growth in the US. It is possible that practitioners who develop a large venous ablation practice, treating the effects of venous insufficiency, will be the best positioned to treat the underlying cause of VI if and when valve prostheses become viable options. In the

US, much of the venous ablation practice is cosmetic, and it is not clear how many patients undergoing venous ablation would ultimately be candidates for valve prostheses. Nonetheless, this highly consumer-oriented peripheral venous intervention has appealed to significant numbers of IR physicians. There is a great deal of literature documenting the success rates for various surgical approaches to venous insufficiency, including stripping, ligation, and sclerotherapy. However, percutaneous approaches are so new that relatively little has been published to date to support their use. Min et al described 90 greater saphenous vein ablations in 84 patients using a laser-based system. They reported a 97% success rate (100% after repeat treatment) with 99% persistent occlusion at 9 months and documented greater saphenous vein diameter reductions at 6 and 9 months of 73% and 81% respectively (17). Results from several multicenter registries evaluating radiofrequency closure have been reported (18,19). Manfrini et al reported results in 210 patients for two different procedures, RF ablation and a device designed to restore function to incompetent valves (18). RF ablation was successful in 141/151 limbs (93%), with 6 month follow-up showing CEAP class 0 or 1 in 87%, with 75% symptom free. Results with the device designed to restore valve function were poor (18). Most recent results describe 319 limbs: 194/232 (84%) were shown to be persistently occluded at 12 months and 121/142 (85%) at 2 years. Clinically, 94% reported complete relief from pain and 87% from varicosities (19). 24 month data show 90% to be reflux free. Patient satisfaction was reportedly very high. These results compare very well with surgical techniques; but like DVT, the burden of proof is on the endovascular community to show in a prospective randomized fashion that these approaches are as effective as (or more effective than) the surgical alternatives. In conclusion, peripheral venous interventions are the mainstay of many IR practices. Venous thromboembolic disease and its sequelae may represent the single largest growth opportunity for IR at present. However, unless the appropriate evidence-based medicine approach is applied to existing and emerging peripheral venous interventions, such growth potential will not be realized. Organizations such as FIRE and CIRREF should place a strong emphasis on funding projects that support what we do as Interventional Radiologists in an evidence based fashion that is understandable and acceptable to our medical colleagues. With such an approach, the field of peripheral venous interventions can contribute heavily to the bright future of Interventional Radiology. References: 1. Hirsch J, Hoak J. Management of deep vein thrombosis and pulmonary embolism- A statement for healthcare professionals. From the Council on Thrombosis (in consultation with the Council on Cardiovascular Radiology), American Heart Association. Circulation 1996; 93:2212-2245. 2. Sharafuddin MJ et al, Endovascular management of venous thrombotic and occlusive diseases of the lower extremities. JVIR 2003; 14:405-423. 3. Strandness DE, Langlois YE, Cramer M, et al. Long-term sequelae of acute venous thrombosis. JAMA 1983; 250:1289-1292. 4. Goldhaber SZ, Buring JE, Lipnick RJ, Hennekens CH. Pooled analyses of randomized trials of streptokinase and heparin in phlebographically documented acute deep venous thrombosis. Am J Med 1984; 76:393-397. 5. Semba CP, Dake MD: Ileofemoral deep venous thrombosis: Aggressive therapy with catheter-directed thrombolysis. Radiology 1994; 191: 487-494. 6. Mewissen MW, Seabrook GR, Meissner MH, Cynamon J, Labropoulos N, Haughton SH, Catheter-directed thrombolysis for lower extremity deep venous thrombosis: Report of a national multicenter registry. Radiology 1999; 211:39-49. 7. Comerota A, Aldridge SC. Thrombolytic therapy for deep venous thrombosis: a clinical review. Can J. Surg 1993: 36:359-364. 8. Comerota AJ, Throm RC, Mathias SD, Haughton S, Mewissen M. Catheter-directed thrombolysis for iliofemoral deep vein thrombosis improves health-related quality of life. J Vasc Surg 2000; 32:130-137. 9. Bjarnason H, Kruse JR, Asinger DA et al, Iliofemoral deep vein thrombosis: Safety and efficacy outcome during 5 years of catheterdirected thrombolytic therapy. JVIR 1997; 8:405-418. 10. Elsharawy M, Elzayat E Early results of thrombolysis vs anticoagulation in iliofemoral venous thrombosis. A randomised clinical trial Eur J Vasc Endovasc Surg 2002 Sep;24(3):209-14 11. Yamauchi T, Furui S, Katoh R et al, Acute thrombosis of the inferior vena cava: treatment with saline-jet aspiration thrombectomy catheter.

AJR 1993; 161:405-407. 12. Uflacker R, Mechanical thrombectomy in acute and subacute thrombosis with use of the Amplatz device: Arterial and venous applications. JVIR 1997; 8:923-932.12 13. Vorwerk D, Guenther RW, Wendt G, Neuerburg J, Schurmann K, Iliocaval stenosis and iliac venous thrombosis in retroperitoneal fibrosis: Percutaneous treatment by use of hydrodynamic thrombectomy and stenting. Cardiovasc Intervent Radiol 1996; 19:40-42. 14. Vedantham S, Vesely TM, Parti N, Darcy M, Hovsepian DM, Picus D Lower extremity venous thrombolysis with adjunctive mechanical thrombectomy JVIR 2002 Oct;13(10):1001-8 15. Pavcnik D, Uchida BT, Timmermans HA, et al, Percutaneous bioprosthetic venous valve: a long term study in sheep. J Vasc Surg 2002; 35:598-602. 16. Gomez-Jorge J, Venbrux AC, Magee C. Percutaneous development of a valved bovine jugular vein in the swine venous system: A potential treatemtn for venous insufficiency. JVIR 2000; 11:931-936. 17. Min RJ, Zimmet SE, Isaacs MN, Forrestal MD. Endovenous laser treatment of the incompetent greater saphenous vein. JVIR 2001; 12:1167-1171. 18. Manfrini S, Gasbarro V, Danielsson G, Norgren L, Chandler JG, Lennox AF, Zarka ZA, Nicolaides AN. Endovenous management of saphenous vein reflux. Endovenous Reflux Management Study Group J Vasc Surg 2000 Aug;32(2):330-42 19.Merchant RF, DePalma RG, Kabnick LS. Endovascular obliteration of saphenous reflux: a multicenter study. J Vasc Surg 2002 Jun;35(6):1190-6

Special Session Percutaneous Treatment of Portal Hypertension 10.2 TIPS: Stents Versus Grafts, Technique, Results, Complications A. Krajina, M. Lojik, V. Chovanec, J. Raupach, P. Hulek; Radiology, University Hospital, Hradec Kralove, Czech Republic. While the TIPS procedure has become safer and more efficacious than alternative treatments during 10 years of its clinical use, it was limited by the need for frequent periodic revisions. Depending upon the definition of the shunt stenosis and surveillance protocols, stenoses of greater than 50 % were reported in 25-50 % of cases within 6-12 months of shunt creation (1-4). Fortunately, these patients were rarely symptomatic and recurrent variceal bleeding was seen in just 25 % of patients with significant TIPS stenosis (5, 6). The high rate of shunt stenoses was the major reason for the relatively poor cost-effectiveness when compared TIPS with endoscopic or surgical treatment. 1.Pathophysiology of TIPS stenosis: bare stents. 1.1. Acute shunt stenosis. A complex process called shunt maturation begins after intrahepatic tract dilatation and metallic stent implantation. This includes at least two stages. The formation of a trombus which is followed by a pseudointimal lining. Ideally, thrombus formation can be detected only as a thin layer along the stent wires as a result of mechanical forces of the blood flow through the shunt, luminal thrombogenicity, and coagulability of the patient´s blood. The dilatation of the parenchymal tract of the liver causes laceration of the biliary ducts. The injured ducts may be of various sizes and thus its role in thrombus formation can be of various range. Hemobilia, which might result from these injuries, is in TIPS procedures extremely rare (7). On the other hand biliary – TIPS fistulae have been repeatedly reported to be observed in patients with acute and/or recurrent shunt thromboses (8-12). The exact pathophysiological mechanism of the bile ducts and bile influence on shunt stenosis and intraluminal thrombus formation is not clear. In vitro studies revealed that bile itself paradoxically inhibits smooth muscle activity and perhaps more important for the excessive thrombus formation, that bile may inhibit endothelialization process of the shunt lumen (13, 14). There is another theory that the mucinous secretion of the biliary endothelium can facilitate the excessive thrombus formation in some shunts (10). 1.2. Chronic shunt stenosis. The more voluminous intraluminal thrombus might provide a physically larger matrix for subsequent pseudointimal hyperplasia and the parenchymal tract stenosis development. This factor and most probably also the other triggering mechanisms may cause chronic shunt stenosis. The similar hyperplasia of smooth muscle cells, as was described in the

parenchymal tract, can be observed in the outflow hepatic vein. Bile leak has evidently no apparent cause in the hepatic vein stenosis development in TIPS. Development of this stenosis is probably triggered by chronic shear stress which results from elevated blood flow and can be observed at many surgical graft anastomoses, including hemodialysis AV shunts and arterial bypasses. This type seems to be more benign in terms of progression and recurrence and may require restenting more frequently than the intraparenchymal stenosis (4). The TIPS stented to the junction of the hepatic vein with the inferior vena cava have been proved to have longer primary patency than shunts stented only within part of the hepatic veins (15). Among various parameters such as portal vein diameter, anatomical locations, angulations of the shunt, blood count and others, only age reached a statistically significant difference betwen group of selected excellent and problematic shunts (16). The patients with poor shunt patency were significantly younger. The patients with Budd-Chiari syndrome secondary to hypercoagulabity have been reported to be at higher risk of shunt thrombosis or stenosis than other population of patients treated with TIPS (17). In patients with more advanced liver cirrhosis, clotting may be diminished because of thrombocytopenia and hypoalbuminemia. The administration of heparin and coumadin derivatives can prevent early shunt thrombosis (18). The preventive application of intraluminal irradiation is technically feasible, however, data on long-term outcomes is not yet available. Local delivery of paclitaxel at the time of TIPS creation was revealed to be effective in reducing pseudointimal hyperplasia in experiment (19). 2. Stent-grafts in TIPS. Experimental (20, 21) and clinical (22, 23) studies have shown that implantation of a stent-graft instead of a bare stent improves the longterm patency, presumably by physically separating blood flow within the shunt from the liver parenchyma and flow overloaded hepatic vein. The most widely used graft material, either for homemade or commercially available stent-grafts, is expanded polytetrafluorethylene (ePTFE). Other graft materials have not shown improved patency versus bare stents, due mainly to their porosity, increased thrombogenicity and inflammatory reaction (24). 2.1. Technique of implantation. The only commercially available stent-graft dedicated for TIPS is the Viatorr TIPS Endoprostheses (W. L. Gore&Assoicates; Inc., AZ, USA) (25, 26). It consists of a flexible nitinol selfexpandable stent. This stent provides an exoskeleton for the ePTFE prosthesis with reduced permeability. The unlined part is 2 cm long and serves for anchoring the stent-graft in the portal vein branch, the PTFE lined part is from 4 to 8 cm long and is designed to extend the stent-graft to the orifice of the hepatic vein at the inferior vena cava. 2.2. Results. The latest reported series included from 65 to 89 patients with primary and secondary patency from 80 % to 90 % in 6 months and 12 months, which is significantly higher than in bare stents. Outflow vein stenoses were observed in cases without full coverage of the hepatic veins (2527). 2.3. Complications. The rate of clinical important portosystemic encephalopathy is increased. This can be due to improved patency and the 8 or 10 mm stent-graft should be preferred over 12 mm one. By its design, the dedicated stent-graft occludes normal drainage through the hepatic vein that receives the stent-graft. The clinical significance of this occlusion has not yet been defined (26). However, since the largest hepatic vein is usually used for TIPS, impairment of liver perfusion might be expected as it was revealed by experimental study (28). Direct intrahepatic portocaval shunts (DIPS) (29) or start to puncture the hepatic vein as close as possible to its ostium into the inferior vena cava (26) might potentially avoid this problem. 2.4. Other indications for stent-grafts in TIPS. Except improving primary and secondary patency of TIPS stent-grafts play important role in sealing extrahepatic portal vein leak during TIPS (30), and using special design of stent-grafts to create shunt stenosis in patients with debilitating encephalopathy without liver failure, which was proved to be uncontrollable despite optimal medical therapy (31). Another potential role of stent-grafts might be in prevention of malignant cell dissemination and potential tumor ingrowth in patients with the centrally located or undiagnosed hepatocellular carcinomas. 2.4.1. Treatment of portal vein tear with intraperitoneal hemorrhage during TIPS. The major procedure-related complication is intraperitoneal hemorrhage from portal vein laceration, which ocurred in 1 % of our series of

more than 500 TIPS procedures. In this situation, an open emergency surgery is associated with significant perioperative mortality. The closer puncture site is to the portal vein bifurcation, the higher is the risk of an extrahepatic entry with subsequent extrahepatic laceration of the portal vein.Though free intraperitoneal bleeding occurs rarely, this observation is contrary to anatomical studies showing an extrahepatic location of the portal vein bifurcation in 40 – 48 % of patients. Temporary occlusion of the portal vein tear with an angioplasty balloon can stop intraperitoneal bleeding or serve as a bridge for appropriate fluid replacement and stent-graft preparing. Implantation of stent-graft seals the blood leak and decreases risk of continual bleeding by decompression of the portal vein. Consideration should be given to maintaining the stent-graft available for emergency use in laboratories where TIPS procedures are performed. 3. Conclusion. Although the results of randomized comparative trials between ePTFE stent-grafts and bare stents are not available yet, it is obvious from current literature that ePTFE stent-grafts have significantly improved acute and long-term primary and secondary patency of TIPS. The endoprostheses have to be placed from the portal vein to the ostium of the hepatic vein to prevent not only acute shunt thrombosis and the parenchymal tract stenosis but also the outflow tract stenosis. Stent-graft implantation plays important role in sealing extrahepatic portal vein leak during TIPS and treats this potentially fatal complication of the TIPS procedure. References: 1. Haskal ZJ, Pentecost MJ, Soulen MC et al. Transjugular intrahepatic portosystemic shunt stenosis and revision: early and midterm results. AJR 1994; 163: 439-444. 2. Hausegger KA, Sternthal HM, Klein GE et al. Transjugular intrahepatic portosystemic shunt: angiographic follow up and secondary interventions. Radiology 1994; 191: 177-181. 3. Nazarian GK, Ferral H, Castaneda-Zuniga WR et al. Development of stenoses in transjugular intrahepatic portosystemic shunts. Radiology 1994; 192: 231-234. 4. Sterling KM, Darcy MD. Stenosis of transjugular intrahepatic portosystemic shunts: presentation and management. AJR 1997; 168: 239244. 5. Barton RE, Rösch J, Saxon RR, Lakin PC, Petersen BD, Keller FS. TIPS: short- and long-term results: a survey of 1750 patients. Semin Intervent Radiol 1995; 12: 364-367. 6. Saxon RR, Ross PL, Mendel-Hartvig J et al. Transjugular intrahepatic portosystemic shunt patency and the importance of stenosis location in the delopment of recurrent symptoms. Radiology 1998; 207: 683-693. 7. Freedman AM, Sanyal AJ. Complications of transjugular intrahepatic portosystemic shunts. Sem Intervent radiol 1994; 2: 161-177. 8. LaBerge JM, Ferrell LD, Ring EJ, Gordon RL. Histopathologic study of stenotic and occluded transjugular intrahepatic portosystemic shunts. JVIR 1993; 4: 779-786. 9. Stout LC, Lyon RE, Murray GB, Barth MH. Pseuodintimal biliary epithelial proliferation and Zahn´s infarct associated with a 6½ - month – old transjugular intrahepatic portosystemic shunt. Am J Gastroenterol 1995; 90: No. 1, 126-130. 10. Saxon RR, Mendel-Hartvig J, Corless CL et al. Bile duct injury as a major cause of stenosis and occlusion in transjugular intrahepatic portosystemic shunts: comparative histopathologic analysis in humans and swine. JVIR 1996; 7: 487-497. 11. Jalan R, Harrison DJ, Redhead DN, Hayes PC. Transjugular intrahepatic portosystemic stent-shunt (TIPSS) occlusion and the role of biliary venous fistulae. J Hepatol 1996; 24: 169-176. 12. Sanyal AJ, Contos MJ, Yager D et al. Development of pseudointima and stenosis after transjugular intrahepatic portasystemic shunts: characterization of cell phenotype and function. Hepatology 1998; 28: 2232. 13. Teng GJ, Bettmann MA, Hoopes PJ et al. Transjugular intrahepatic portosystemic shunt: effect of bile leak on smooth muscle cell proliferation. Radiology 1998; 208: 799-805. 14. Teng G, Lu Q. Bile leakage in transjugular intrahepatic portosystemic shunt: effect on human umbilical vein endothelium. JVIR 2003; 14: S 74. 15. Agarwal R, Stavropoulos SW, Clark TWI. The effect of venous outflow stent position on TIPS patency. JVIR 2003; 14: S 74. 16. Chovanec V, Krajina A, Skrabkova Z, Fejfar T. Potential factors influencing TIPS patency. Cardiovasc Intervent Radiol 2001; 24, suppl. 1: 137. 17. Cejna M, Peck-Radosavljevic M, Schoder et al. Repeat interventions for maintenance of transjugular intrahepatic portosystemic shunt func-

tion in patients with Budd-Chiari syndrome. JVIR 2002; 13: 193-199. 18. Sauer RR, Theilmann L, Richter G et al. Phenprocoumon for prevention of early shunt occlusion after transjugular intrahepatic portosystemic shunt (TIPS) results of a randomized trial (abstract). Gastroenterol 1995; 108: A 1164. 19. Park SV, Cha IH, Lee SH, Kim KA, Park CM, Kim CH. Inhibition of pseudointimal hyperplasia in swine TIPS model: the efficacy of local delivery of paclitaxel. JVIR 2003; 14, S 113. 20. Nishimine K, Saxon RR, Kichikawa K et al. Improved transjugular intrahepatic portosystemic shunt patency with PTFE – covered stentgrafts: experimental results in swine. Radiology 1995; 196: 341-347. 21. Haskal ZJ, Davis A, McAllister A, Furth EE. PTFE – encapsulated endovascular stent-graft for transjugular intrahepatic portosystemic shunts: experimental evaluation. Radiology 1997; 205: 682-688. 22. Cohen GS, Young HY, Ball DS. Stent-graft as treatment for TIPS-biliary fistula. JVIR 1996; 7: 665-668. 23. Saxon RR, Timmermans HA, Uchida BT. Stent-grafts for revision of TIPS stenoses and occlusions: a clinical pilot study. JVIR 1997; 8: 539548. 24. Haskal ZJ. TIPS stenosis and reinterventions. In Hulek P, Krajina A. Current practice of TIPS, Olga Stambergova, Hradec Kralove 2001, 172182. 25. Cejna M, Peck-RadosavljevicM, Thurnher SA et al. Creation of transjugular intrahepatic portosystemic shunts with stent-grafts: initial experiences with a polytetrafluoroethylene-covered nitinol endoprosthesis. Radiology 2001; 221: 437-446. 26. Otal P, Smayra T, Bureau C et al. Preliminary results of a new expanded-polytetraufluoroethylene-covered stent-graft for transjugular intrahepatic portosystemic shunt procedures. AJR 2002; 178: 141147. 27. Hausegger KA, Karnel F, Tauss J, Portugaller H, Georgieva B, Deutschmann. TIPS creation with the Viatorr stent-graft: experience with 71 patients. JVIR 2003; 14: S 34. 28. Keussen I, Song HY, Bajc M, Cwikiel W. Changes in the distribution of hepatic arterial blood flow following TIPS with uncovered stent and stent-graft: an experimental study. Cardiovasc Intervent Radiol 2002; 25: 314-317. 29. Petersen B. Intravascular ultrasound-guided direct intrahepatic portacaval shunt: description of technique and technical refinements. JVIR 2003; 14: 21-32. 30. Krajina A, Hulek P, Ferko A, Nozicka J. Extrahepatic portal venous laceration in TIPS treated with stent graft placement. Hepatogastroenterol 1997; 44: 667-670. 31. Madoff DC, Perez-Young IV, Wallace MJ, Skolkin MD, Toombs BD. Management of TIPS-related refractory hepatic encephalopathy with reduced wallgraft endoprostheses. JVIR 2003; 14: 369-374.

Special Session Interventions in Stroke and SAH 11.2 Thrombolysis in Stroke C. Islak; Radiology Dept. Nororadiology Division, Cerrahpasa Medical School, Istanbul, Turkey. Thrombolysis In Stroke Although in use over the last two decades, thrombolysis for ischemic stroke still stirs a lot of discussion among interventionalist and neuroscientist alike. This is mostly due to the heterogeneity of results published on the subject. But nowadays Class A evidence supports that under certain circumstances for selective cases thrombolysis offers an opportunity for better clinical outcome. Technically speaking it can be achieved either through intra arterial or intravenous routes. Even if Intravenous thrombolysis (IVT) appears easier to apply and offers shorter time interval to install the treatment it lacks selectivity. Therefore higher drug doses are needed and that may actually leads more complication. Since IVT does not need any vascular intervention the one that is to be discussed here will be limited to intraarterial one. Before beginning to overview the Thrombolysis the disease per se and certain facts about the Stroke needs to be remembered. It is well known that not only is it 3rd leading cause of dead in developed countries, Stroke takes a very high toll as far as the morbidity is concerned. When we look the natural history of it we immediately see that mortality rates

can go up as high as 70%. In spite of those figures it is known a very few cases could be selected for probable treatment. That fact might be observed when numbers in such studies are taken into consideration. For example in one large randomized study only less than 2% of cases presenting ischemic stroke (IS) could be treated. That limitation occurs because the treatment for IS has to be initiated in a given time interval called “Therapeutic Window”. Since any treatment beyond that interval will most probably harm the patient It has to be strictly obeyed with. It becomes therefore immediately apparent that the primary task in Thrombolysis for IS is patient selection not the technical point of treatment. The aim of treatment in IS is primarily to reperfuse the ischemic but viable tissue with nutrient and oxygen. Therefore the obstacle for blood flow should be taken away only if still electrically silent but living neurons exist. A lot of efforts have been put forward to find out to outline the so-called ischemic penumbra in which the neurons receive blood supply below threshold level. Since “Time is Brain” is 100% reality all those efforts should not consume a lot of time. To day in most centers MR diffusion and variety of perfusion studies are in use to show ischemic but salvageable neurons. For IAT two types of intervention can be done. The first one is mechanical lysis and the second one is drug assisted lysis. . In mechanical lysis the clot is disrupted with micro wires while thrombolytic drugs are used for drug assisted lysis. In general both types are combined in order to achieve recanalization as fast as possible. For certain hard clots resisting to lysis can be retrieved with retrieval devises such as micro snares. The most widely used drugs now are rTPA and pro UK. It should be remembered that any recanalization effort might lead to reperfusion injury and hemorrhage although this at least theorically is higher with drugs. The risk of hemorrhage leading to neurologic deterioration is around 25% in patient receiving thrombolytic therapy. Although the figure is nearly same for untreated patient it occurs later. The predictive factors for hemorrhage are location of occlusion, age, previous hypertension, higher NIHSS scores and longer time to recanalization. Despite an increased frequency of early symptomatic intracranial hemorrhage, thrombolytic therapy appears to be the only effective treatment for IS provided that initiated during theTherapeutic window. References: 1. Ueda T, Sakaki S, Kumon Y, Ohta S. Multivariable analysis of predictive factors related to outcome at 6 months after intra-arterial thrombolysis for acute ischemic stroke. Stroke 1999;30:2360-2365 2. Becker J, Monsein LH, Ulatowski J, et al. Intraarterial thrombolysis in vertebrobasilar occlusion. AJNR 1996:17:255-262 3. Suarez JL, Sunshine JL, Tarr R, et al. Predictors of clinical improvement, angiographic recanalization, and intracranial hemorrhage after intra-arterial trombolysis for acute ischemic stroke. Stroke. 1999;30:2094-2100. 4. Cross DT, Derdeyn CP, Moran JM. Bleeding complications after basilar artery fibrinolysis with tissue plasminogen activator. AJNR 2001;22:521-525. 5. Jahan R, Duckvilker GR, Kidwell CS, et al. Intraarterial thrombolysis for treatment of acute stroke: Experience in 26 patients with long term follow-up. AJNR. 1999;20:1291-1299. 6. Wechsler LR, MTech RR, Furlan AJ, et al. Factors Influencing outcome and treatment effect in PROACT II. Stroke. 2003;34:1224-1229. 7. Kaste M, Thomassen L, Grond M, et al. Thrombolysis for acute ischemic stroke. A consensus statement of the 3rd Karolinska Stroke Update, October 30-31, 2000. Stroke 2001;3;2:2717-2718. 8. Kidwell CS, Saver JL, Carneado J, et al. Predictors of hemorrhagic transformation in patients receiving intra-arterial thrombolysis. Stroke. 2002:33; 717-724. 9. Lansberg MG, Tong DC, Norbash AM, et al. Intra-arterial rtPA treatment of stroke assessed by diffusion- and perfusion –weighted MRI. Stroke 1999;30:678-680. 10. Lev MH, Segal AZ, Farkas J, et al. Utility of Perfusion weighted CT imaging in acute middle cerebral artery stroke treated with intra-arterial thrombolysis. Prediction of final infarct volume and clinical outcome. Stroke 2001;32:2021-2028. 11. Shrier DA, Tanaka H, Numaguchi Y, et al. CT angiography in the evaluation of acute stroke. AJNR 1997;18:1011-1020. 12. Wildermuth S, Knauth M, Brandt T, et al. Role of CT angiography in patient selection for thrombolytic therapy in acute hemispheric stroke. Stroke 1998;29:935-938.

13. Jansen O, von Kummer R, Forsting M, Hacke W and Sartor K. Thrombolytic therapy in acute occlusion of the intracranial internal carotid artery bifurcation. AJNR 1995;16:1977-1986. 14. Endo S, Kuwayama N, Hirashima Y, et al. Results of urgent thrombolysis in patients with major stroke and atherothrombotic occlusion of the cervical internal carotid artery. AJNR 1998;19:1169-1175. 15. del Zoppo GJ, Higashida RT, Furlan AJ, et al. PROACT: A phase II randomized trial of recombinant pro-urokinase by direct arterial delivery in acute middle cerebral artery stroke. Stroke 1998;29:4-11. 16. Furlan A, Higashida R, Wechsler L, et al. &Idot_upper;ntra-arterial prourokinase for acute ischemicstroke the PROACT II study : A randomized controlled trial. JAMA 1999; 282:2003-2011.

Special Session Complications in Non-Vascular Interventions 12.1 Ultrasound guided percutaneous subclavian vein port implantation F. Orsi, G. Bonomo; Radiology, European Institute of Oncology, Milan, Italy. The need for central venous access has found wide application in intravenous administration of medicaments, hemoderivates, in parenteral nutrition and in blood sampling. In spite of this growing wide spread, there is not a standardized implantation procedure: there is no agreement about the vein access site, the technique of vein access (surgical vs percutaneous, with anatomical landmarks vs ultrasound guided) and about device type (open ended vs valved tip, sylicon vs polyurethane catheter) and size. Many of these problems are justified by the lack of evidenced base data in litterature (small series or not randomized studies) that may lead daily activity; some of these problems are instead based on economical point of view or on the limitations of physician different attitude (surgeon, anaesthesiologist, interventional radiologist). Access sites most commonly used are internal jugular vein (JV) and subclavian vein (SV); other sites rarely used are femoral vein, saphenous vein, inferior vena cava (with retroperitoneal direct puncture) and so on. Surgeons instead have often friendliness with surgical cut-down of cephalic vein. Moreover the choice of the access site and the technique is based on the potentially related complications. Due to this fact JV is considered the most safer access site in order to avoid pneumothorax, but a limitation not enough analized in the litterature of this way is patient discomfort due to the subcutaneous catheter loop on the neck side. However most transjugular and trans-subclavian catheters are only positioned by referring to the anatomical landmarks (so called “blind technique”). This method might have nevertheless some disadvantages, such as complications (i.e. repeated vein puncture, pneumothorax and accidental puncture of the artery) and prolonged procedural times. Moreover the learning curve of new operators is long and full of unaspected complications. The use of ultrasound (US) guidance for jugular and subclavian vein puncture provides a preliminary study of vessels anatomy by detecting possible abnormal pathways and allowing the diagnosis of asymptomatic venous thrombosis patterns. US guidance enables to reduce the number of puncture attempts, diminishes the incidence of complications, thus eliminating a lot of patient discomfort. The learning curve of new operators seems to be very shorter and safer, mainly due to the possibility to “see” what they are doing under the skin surface. The aim of this presentation is to describe the US-guided subclavian vein Port placement technique. Patient selection Indications for the implantation of port vein has to be defined by the clinician who refers the patient. In this step is very important the professional relationship with the clinician in order to show him possibilities and limitations of this device. Inside the limitations it is to be mentioned patient origin: if the patient will be refferred to a different hospital for the treatment is important that they know how to use the device that we implanted; depending on different local reality, there are many situation in which physicians are not friendliness with totally implanted access vein device. Coagulation values have to be accurately checked due to the need of a surgical subcutaneous pocket for the port reservoir more than for vein

puncture. Raccomanded baseline values are mainly Prothrombin Time (PT = 10-13 seconds), Partial Thromboplastin time (PTT= 22-35 seconds) and platelet count (150.000 – 450.000 /mm3). Is important to exclude important anatomical variation or surgical and/ or pathological modifications on the subclavian area: an extended mastectomy or a previous local radiotherapy may result in a very difficult subcutaneous pocket preparation. The side of implantation depends moreover by lungs conditions: due to the fact that one of the most serious complications is pneumothorax, especially at the beginning of the experience, is very important to choose the side where lung is in the worst conditions (i.e. the side of pneumonectomy). Procedure preparation Due to the fact that infection is a serious midterm complication, is very important that all the procedure will be performed under surgical sterile conditions; for this reason the angio-room has to be particularly cleaned and free from unnecessary items. The brilliance intensifier (BI) must to be covered with a sterile plastic cuff because some steps of implantation will be perform with fluoroscopic guidance, with the patient exactly under the BI. All the surgical tools for the procedure and the device have to be prepared on a cart easy to use. A 7.5 MHz linear probe for ultrasound guidance has to be covered with a dedicated or an adapted sterile cuff. In order to make the procedure safer and quicker we suggest to use a common electrocautery to control the bleeding during surgical pocket preparation. Procedures steps 1) patient disinfection: this step is to be accurately followed; the implanted area must to be limited with sterile cloth; 2) local subcutaneous anaesthesia on the puncture site; 3) skin incision and ultrasound guided vein puncture; 4) guide-wire insertion under fluoro guide, with J-tip on superior vena cava/right atrium (SVC/RA) junction; 5) needle extraction and external part of guide-wire protection (with sterile cloth); 6) skin incision and subcutaneous pocket preparation; it will be about five cm down the vein access; 7) dilation of venous access with dedicated peel-away sheath and catheter introduction (under fluoro-guidance) over the sheath tip; 8) removal of peel-away sheath and withdrawing of the catheter on the right position (SVC/RA junction); 9) catheter subcutaneous tunnellization between vein access site and the pocket; 10) catheter connection and port insertion inside the subcutaneous pocket; 11) port anchorage with suture on the muscle fascia and cutaneous suture; 12) port check and chest X-ray. References: 1.Caridi Jg. AJR 1998; 171:1259-63 2.Orsi F. J Vasc Access 2000; 1:73-77 3.Ruesch S. Crit Care Med 2002 Feb; 30(2):454-60

Special Session Infrapopliteal PTA 16.2 Infra-popliteal PTA : results, follow-up, complications M. Sapoval; Cardio Vascular Radiology, HEGP, Paris, France. Infra-popliteal PTA is currently reserved to patients suffering of critical limb ischemia This patient population is also a special subset of our ordinary case load, as they are often older and more fragile than current patients referred for interventionnal treatment. Spontaneous prognosis of these patients is poor regarding both limb salvage and life expectancy. As stated in the TASC consensus : « in contrast to patients with intermittent claudication, patients with critical limb ischemia are on a rapidly downward course in terms of both life and limb , and current non interventionnal treatment or pharmacotherapy can help only a minority » (1). In fact it is estimated that 25 % will undergo an amputation during their first year of follow-up and 90 % will undergo angioplasty or bypass at the

same term (2) When analysing the data of the literature and the results the Fire Registry it appears that the mean age of these patients is around 73 years old and that 53% have myocardial ischemia , 63% have hypertension and 28% have renal failure (3). The first line intervention should be endovascular treatment that can be performed using standard angioplasty, subintimal recanalization, cutting balloon and stent placement in selected cases. This approach was recommended in the Second European consensus document published in 1991(4). It was reconfirmed by the experience of Bolia and Bell in a letter to the editor in 1999 stating that when looking at consecutive patients referred for treatment of critical limb ischemia including 222 limbs in 188 patients treated in 1994, 42 % were treated using angioplasty and 24 % using bypass surgery (5). They underline that in 1999, these figures had moved to two thirds treated with angioplasty with a primary success rate of more than 90 %. Soder reports even more dramatic figures: in a consecutive experience using the “PTA first” approach 77 PTA were performed in this setting compared to only 9 femoro-distal bypass over a one year period (6) Other authors suggest also the interest to prefer a primary endovascular approach in these patients, because it offers the same results than bypass surgery without precluding the possibility of surgical approach in case of failure (7) Thus it is current practice in most experienced teams that these patients should be assessed on a multidisciplinary approach with the aim to look for a possible endovascular approach. Using this appraoch, angiography followed by angioplasty in the same session can be performed in most patients. The role of the medical and anaesthesiologist team is of utmost importance in the pre, per and post operative procedure. In fact, management of anticoagulation, puncture site complications possible myocardial ischemia and renal failure is one of the key aspect of clinical success and reduction of peri-operative complication. Primary angiographic success Immediate primary angiographic success is obtained in 70- 91 % of the cases, the more difficult cases are those with extensive occlusive lesions. Modern materials can enhance the initial success rate as well as sub-intimal recanalization which can solve very difficult angiographic cases (6-9). Peri-operative complications: Despite the use of 5 F sheath which is sufficient in the vast majority of cases, puncture site complications can be life threatening especially because of multiple comorbidities in these patients. They are more frequent in case of antegrade puncture and in older women with diabetes and hypertension (10). It can be recommended as soon as the patient is not able to maintain strict decubitus during 24 hours to use closing devices. The immediate complication rate is reported between 0 and 16 % (11). In the ongoing experience of Fire registry comprising 283 consecutive patients treated in 23 centres from 11 countries, a 8.7 % puncture site rate complications was reported (3). Another complication that should be feared is the risk for infection and systematic use of prophylactic antibiotherapy should be recommended. They can be favoured if the procedure is prolonged also because of more or less infected distal ulceration. In case of infected gangrene, debridment surgery is to be performed sometimes on an emergent basis. These patients are more or less immuno-depressed because of extensive wounds, frequent renal failure and chronic pain. Peri-operative mortality: Related to puncture site bleeding complicated by myocardial ischemia but also by possible septic choc, it is reported anywhere between 0 to 5 % (3.8 % in the Fire registry)(3,6,11) Immediate clinical success: Depending on the criteria used (the definition of TASC, which is used in the Fire registry is :”clinical success is obtained in case of reopening of a pulse distal to the lesion treated or increase in ankle brachial index of more than .015”) (1), is reported between 63-90 % (6, 9). In the Fire registry we recorded a 88 % clinical success rate at 48 hours and 80% at hospital discharge (3); Mid-term and long-term results: The purpose of revascularization in these patients is to save limbs and lives. Without aggressive treatment , most of them will die in the short to mid term. The clinical success is more important than angiographic success in the long term, because once healing has been obtained, restenosis at the level treated can be of little or no clinical consequence. Results can be assessed using cumulated patency curves and the limb salvage rate. The later is probably more important for the reason explained higher. Primary patency is comprised between 40 % and 81 % at one year (6,7,11) and between 48 and 78 % at 2 years (9). Limb sal-

vage rate is higher, between 77 % and 89 % at one year (5,6,9,11). In the Fire registry, 73 % of limb salvage rate at 6 months is currently obtained. Factors predicting clinical and angiographic success: From a clinical stand point of view, the presence of diabetes and renal failure are related to lower limb salvage rate (6, 13). It is also recognised following distal surgical revascularization that healing of ischemic foot ulcer is higher in non diabetics as well as survival at 1 year (14). From a technical point of view, the ability to restore a strain line flow to the foot as well as enhancement of flow to the ischemic region are predictors of limb salvage (6, 15, 16) Conclusion: Using modern techniques including lower profile balloons through 5F sheath, sub-intimal recanlization and maybe stents in the near future, most of the patients suffering of critical limb ischemia can be safely treated. However, the interventionnal radiologist should always remember that these subset of our patients are especially fragile and at high risk of life threatening complications. The fire registry is an ongoing collective experience of data collection on this topic and all team in Europe should be encouraged to enter there cases (3). References: 1- TASC consensus. J Vasc Surg 2000, Part 2 ;31 number 1 2- Bower BL Valentine RJ, Myers Si et al. The natural history of patients with claudication with toe pressure of 40 mm Hg or less. J Vasc Surg 1993; 18:506-11 3- Fire registry on infra-popliteal angioplasty for critical limb ischemia: .org 4- European Working group on critical leg ischemia second European consensus document on chronic critical leg ischemia. Circulation 1991; 84 (suppl IV): IV-1- IV-26. 5- Varty, K, Nydahi S, Butterworth P et al. Changes in the management of critical limb ischemia. Br j Surg 1996; 83: 953-7 6- Soder HK, Manninen HI, Jaakkola P et al. Prospective trial of infra popliteal artery balloon angioplasty for critical limb ischemia: angiographic and clinical results . J Vasc Intervent Radiol 2000; 11:1021-31 7- Lofberg AM, Karacagil S, Ljungman C et al. Percutaneous transluminal angioplasty of the femoropopliteal arteries in limbs with chronic critical lower limb ischemia. J Vasc Surg 2001; 34:114-21 8- Matsi PJ, Manninen HI, Suhonen MT et al. Chronic critical lower-limb ischemia Prospective trial of angioplasty with 1-36 months follow-up. Radiology 1993; 188:381-87 9- London NJ, Varty K, Sayers RD et al. Percutaneous transluminal angioplasty for lower-limb critical ischemia. Br J Surg 1996; 83:135-6 10- Heintzen MP, Strauer BE. Peripheral arterial complications after heart catheterization. Herz 1998; 23:4-20. 11- Boyer L, Therre T, Garcier JM et al. Infra popliteal percutaneous transluminal angioplasty limb salvage. Acta Radiol 2000; 41:73-7. 12- Brillu C, Piquet J, Villapardiena F et al. Percutaneous transluminal angioplasty for management of critical ischemia in arteries below the knee. Ann Vasc Surg 2001: 15: 175-81. 13- Vainio E, Salenius JP, Lepantalo M et al. Endovascualr surgery for chronic lower limb ischemia. Factors predicting immediate outcome on the basis of a natiowide vascular registry. Ann Chir Gynaecol 2001; 90: 86-91 14- Wolfle KD, Bruijnen H, Loeprecht H et al. Graft patency and clinical outcome of femorodistal arterial reconstruction in diabetic and non diabetic patients: results of a multicenter comparative analysis. Eur J Vasc Endovasc Surg 2003; 25:229-34. 15- Matsi PJ, Manninen HI, Suhonen MT Chronic critical lower limb ischemia: prospective trial of angioplasty with 1-36 monthes follow-up. Radiology 1993; 188: 381-87 16- Bakal CW, Sprayregen S, Scheinbaum K et al. Percutaneous transluminal angioplasty o f the infra-popliteal arteries: results in 53 patients. Am J Roentgenol 1990; 154:171-4

16.3 Interventions in femoro-popliteal and femoro-crural bypasses L. Boyer, J. Garcier, A. Ravel, E. Rosset; Radiologie, CHU G Montpied, Clermont Ferrand Cedex 1, France. Bypass occlusion within the first 2 months after surgery has to be considered as a surgical failure : a surgical revision (anastomoses, graft kinking or torsion, remaining valves) is a priori mandatory in such cases. In opposition, the long term patency of femoro-popliteal and femorocural bypasses not only depends on the initial surgical technique, but

also on the quality of the medical follow up, to detect stenosis as soon as possible in order to avoid occlusion. Most of the obstructive lesions begin during the year following surgery : endovascular interventions are then of interest, to treat proximal or distal anastomosis stenosis, stenosis of the bypass itself (vein graft), or inflow and distal outflow arteries stenoses. Thrombo aspiration, fibrinolysis, balloon PTA and stents authorize alternative therapeutic options to conventional surgical reinterventions. Surgical revision in our mind have to be preferred to treat anastomotic false aneurysms, long and diffuse stenoses, and valve remnants. When a reoperation is planned, an imaging staging to define the length and quality of remaining superficial veins is of interest. PTA is valuable to treat short segmental stenosis, via crossover technique access (proximal lesions) or via antegrade direct puncture (distal lesions). Primary assisted patency rates of 68-80% after 5 years are reported. To treat bypasses occlusions, heparin alone is often not sufficient. Urokinase and recombinant tissue plasmine activator (rtPA), associated with heparin infusion, are usually used to perform intra arterial in situ fibrinolysis infusion of occluded grafts (via crossover techniques or ipsilateral antegrade puncture, depending on location of proximal anastomosis and thrombosis). The global rate of complete clot lysis in case of fresh thrombosis is 80 - 90%, higher with prosthetic grafts than with vein grafts. To avoid reocclusion, thrombolysis has to be considered as a non exclusive therapy, but as the first step before the treatment of the prime cause of thrombosis. Large variations of long term patency rates are reported (18 - 70%), depending on multiple factors (type and age of the bypass, causal mechanism of occlusion, delay to begin and fibrinolysis protocol…). Fogarty surgical thrombectomy often stay the first choice in case of graft occlusion causing acute ischemia. More complete and not so aggressive, thrombo aspiration is also now and alternative to surgical Fogarty desobstruction. It may be performed (immediately followed if necessary by PTA correction of an unmasked stenosis) alone or in association with fibrinolysis, authorizing lower doses protocols. Long term results of thrombolysis, thromboaspiration or surgical thrombectomy remains poor, leading to discuss in each case the opportunity of a new bypass ; but the results of redux operations are inferior than those of such primary surgery. No References

Special Session GI Interventions 17.1 Esophageal stenting B. Acunas; Radiology, Istanbul Medical Faculty, Istanbul, Turkey. Carcinoma of the esophagus is notoriously difficult to manage, and approximately 60% of all patients have extensive tumors that are suitable only for palliative treatment. Even after curative surgery, almost 20% of the patients will have further dysphagia owing to recurrence or stricture at the anastomotic sites. Treatment of dysphagia with the insertion of a metal stent was proposed by Frimberger in 1983 and since then, many reports considering experience with covered or uncovered metallic stents have been published. The indications for metallic stent placement in esophageal diseases can be listed as follows: 1. Palliative treatment of malignant dysphagia, 2. Palliation of malignant esophageal fistulas and perforations due to local tumor invasion, surgery or esophageal stent placement and, 3. Benign esophageal strictures in patients who become resistant to balloon dilation, and who are poor operative candidates. The stents most commonly used in the esophagus are covered and uncovered Ultraflex stents ( Boston Scientific, Medi-tech, Natick, MA ), cone shaped, covered Flamingo stent ( Boston Scientific ), and covered Gianturco-Rösch Z stents ( William Cook Europe, Bjaeverskov, Denmark ). First an initial esophagogram is performed to delineate the site and the length of the stricture. The patient is then placed in the right lateral decubitus position. A topical anesthetic is applied to the pharynx. The stricture is crossed with Terumo guidewire and angled tip catheter

manipulation. A stiff Amplatz guidewire is advanced into the stomach, and the catheter is withdrawn. Predilation is performed with an esophageal balloon catheter 6-8 cm long and 12-15 mm in diameter. A stent of appropriate size is selected; the stent length is chosen so that at least 2 cm of normal esophagus above and below the stricture is covered by the stent. After stent deployment, contrast examination is performed to confirm correct stent location and to exclude complications such as perforation. Esophageal stent placement is technically successful in 97-100% and all stent types are effective in relieving dysphagia in 83-100% of patients. Covered and uncovered stents are prone to different complications. The most common compliaction seen in covered stents is migration while tumor ingrowth is the most common complication seen in uncovered stents. Other complications include tumor overgrowth, ulceration, fistula and perforation, hemorrhage and severe pain. In conclusion, metal stents are safe and effective for palliation of dysphagia. But however, despite multiple recent advances in stent design, the quest for the ideal stent continues. References: 1. R. Earlam and J.R. Cunha-Melo , Malignant esophageal strictures: a review of techniques for palliative intubation. Br. J. Surg. 69 (1982), pp. 61–68. 2. E. Frimberger , Expanding spiral: a new type of prosthesis for the palliative treatment of malignant esophageal stenosis. Endoscopy 15 (1983), pp. 213–214. 3. B. Acunaş, &Idot_upper;. Rozanes, . Say et al., Treatment of malignant dysphagia with nitinol stents. Eur. Radiol. 5 (1995), pp. 599–602. 4. B. Acunaş, &Idot_upper;. Rozanes, S. Akpinar, A. Tunaci, M. Tunaci and G. Acunaş , Palliation of malignant esophageal strictures with self-expanding nitinol stents: drawbacks and complications. Radiology 199 (1996), pp. 648–652. 5. H.Y. Song, Y.S. Do, Y.M. Han et al., Covered, expandable esophageal metallic stent tubes: experience in 119 patients. Radiology 193 (1994), pp. 689–695. 6. W. Cwikiel, K.-G. Tranberg, M. Cwikiel and R. Lillo-Gil , Malignant dysphagia: palliation with esophageal stents––long term results in 100 patients. Radiology 207 (1998), pp. 513–518. 7. A. Poyanli, S. Sencer, &Idot_upper;. Rozanes and B. Acunaş, Palliative treatment of inoperable malignant esophageal strictures with conically shaped covered self expanding stents. Acta Radiol. 42 (2001), pp. 166–171. 8. H.U. Laasch, D.A. Nicholson, C.L. Kay et al., The clinical effectiveness of the Gianturco esophageal stent in malignant esophageal obstruction. Clin. Radiol. 53 (1998), pp. 666–672. 9. R.A. Morgan, E.R.E. Denton, A.F. Watkinson et al., Metallic stenting of malignant esophageal strictures. Cardiovasc. Intervent. Radiol. 69 Suppl (1996) abstract .

17.3 Colon stenting: A review M. A. De Gregorio1, A. Mainar1, M. J. Gimeno1, J. Medrano1, R. Tobio2, M. Herrera1, H. D'Agostino3; 1 Interventional Radiology, Universidad de Zaragoza, Zaragoza, Spain, 2Interventional Radiology, Clinica La Zarzuela, Madrid, Spain, 3 Interventional Radiology, Lousiana State University, Shreveport, United States. Objetive: Objective: To know and to understand the main indications, results, complications and clinical and economic implications of the stenting for treatment of the colonic stenosis and obstruction. Abstract Up to 85 % of the colonic obstructions are due to malignant lesion. Between the 7 and 29% of these patients presents total or partial obstruction of colon. So alone 20% of the patients that debut with acute colonic obstruction survive 5 years. The intervention of emergency in the colonic obstruction is associated with discharge morbidity and mortality rates. To only 40% of the patients with obstructive carcinoma of left colon can be treated with surgical resection without any need for a stoma. The temporary or permanent colostomy supposes an important impact on quality of life. The colonic decompression by means of stent supposes a treatment, durable, simple and effective as palliation in the patients with metastatic disease. Stent deployment

provides an effective solution to the acute colonic obstruction and allows surgery in elective conditions. In both situations the cost is cheaper and avoids the need of a stoma. Key words: Colonic stenosis or obstruction, stents, neoplasm Introduction Colorectal cancer is an important health problem: there are nearly one million new cases of colorectal cancer world- wide diagnosed each year and half a million deaths. (1). Between 10 and 30% of these patients present complete or partial obstruction of the colon (2). Traditionally the treatment of the large-bowel obstruction to colon cancer has been the surgery (3). Nevertheless, acute obstruction of the left side colon pose a important medical problem (4). Emergency surgery in a patient with a unprepared colon is associated with high morbility and mortality (5). Patient who present with obstruction of the colon have five year survival rate of less than 20 % a far poorer prognosis than patients who present without obstruction (6). Mortality decrease from 15-20 per cent to 0.9-6 per cent when patients undergo elective surgery (7). Only 40 % of left –sided carcinoma colonic obstructions can be treated with intraoperative lavage and subtotal colectomy (8). The rest of patients need a temporary or permanent colostomy, which has an impact on quality of life (9). Colonic stents were first used by Dohmoto in 1990 (10). Metallic endoprosthesis have been used in recent years, as a palliative treatment, to manage the acute phase of large-bowel obstruction in patients with colonic malignancy (11-13). Tejero et al. in 1994 described the use of stents as a “ bridge to surgery”. Other non-surgical treatments were used with limited effectiveness such as balloon dilatation, endoscopic laser ablation and decompression tubes (14). Metallic stent placement is a reasonable alternative that can provide an effective, non-surgical decompression of the obstructed left-sided colon to avoid and emergency colostomy (13). Indications and contraindications Main indications for endoluminal metallic stents placement in the colon and rectum are: 1. for temporary colonic decompression in patients with acute malignancy obstruction as a “bridge to elective surgery”, 2. for long-term colonic decompression in patients with large-bowel obstruction due an unresectable colonic carcinoma, 3. for long-term colonic decompression in patients with large-bowel benign strictures due a fibrosis post surgery or post radiotherapy 4. for temporary colonic decompression in patients with diverticulitis and inflammation and obstruction to permit elective colon cleansing prior to surgery resection. 5. as palliative treatment, for close coloenteric or colovesical fistulae. The main contraindication to the procedure are: clinical and radiological evidence of perforation of the colon. Other contraindication o limitations are: tumors that are too long, tumors are too proximal or too distal in the colon and too much colon kinked (15-16). Stents design. Many metallic stents have been used for the management of colonic obstruction. Despite their commercial availability the current generation of stents continues to be modified and improved. The ideal device should include the following features: 1. High expansion ratio 2. High flexibility, 3. Big diameter (> 25 mm), 4. Mechanical stability ( sandwatch shape ) adequate radial expandable force. 5. Prevention of reestenosis due to tumor ingrowth or hyperplasia. 6. Small delivery system. 7. Biogradable or readily removable stent for benign strictures. The different current types of metallic stents are: 1. Enteral stent Wallstent (Boston Scientific La Garenne Colombes . Cedex. France ). Gianturco Z Stent (William Cook Europe Bjaeversok Denmark) 3. Memotherm colorectal stent ( C R Bard Inc Billerica MA USA). 4.Ultraflex esophageal stent (Boston Scientific La Garenne Colombes. Cedex France ). 5 Stent Choostent. Life Europe Bagnolet. France. Technical considerations Stent deployment in colon can be achieved under fluoroscopy guidance alone, under colonoscopy guidance alone and combined (fluorosscopy and colonoscopy). In distal colonic and rectal regions can be achieved under fluoroscopic and colonoscopy guidance alone. Due to redundant access path to the more proximal portion of the colon, especially with the tortuosity in the sigmoid segment, stent placement in the descending colon same authors (14-21) preferred guided with combined fluoroscopy and endoscopy. Endoscopy is frequently used to cross colonic stricture and place stent, although it is not strictly necessary. However fluoroscopy is an absolute requirement. (16). We preferred use fluoroscopic guide alone approach. Bowel cleansing may be desirable to facilitate endoscopic stenting, although it is not always possible to perform a complete cleansing. Moreover, oral bowel preparation is contraindicated due to the risk of

worsening the obstruction. The procedure doesn't require general anaesthesia, but if the patient, doesn't collaborate, he is nervous or he requires a procedure combined with colonoscopy, sedation with midazolan and analgesia is administered. A water-soluble contrast enema should be performed prior to stent placement in order to identify the location, length, and calibre of the obstructive lesion. With either a ventral decubitus or a lateral decubitus position, a high-torque angiographic catheter or guiding catheter is advanced over a 0.035-inch angled hydrophilic guide super stiff wire (Radiofocus Terumo Europe NV Leuven Belgium) to traverse the obstructive segment under fluoroscopic guidance. If an excessively tortuous or redundant recto sigmoid region is encountered, the use of a 0.035-icnh Lunderquist extra stiff wire guide (William Cook Europe Bjaeversok Denmark) is recommended to facilitate advancement of the angiographic catheter to the level of the obstruction. As soon as the catheter is positioned proximal to the obstruction, water-soluble contrast medium is injected through the catheter to document the characteristics and length of the obstruction and to rule out the possibility of perforation. The procedure is necessary to achieve under fluoroscopy guidance The radiopaque markers can be placed to identify the two ends of the obstruction. An appropriately sized stent and its delivery system are chosen based on the information obtained from the enema examination. A 0.038-inch Amplatz super stiff wire is introduced through the catheter into the proximal segment well beyond the obstruction. After the catheter is withdrawn, a delivery system with a loaded stent is advanced over the stiff wire and is positioned in the obstructive segment. The deployed stent should be long enough not only to cover the entire obstructive segment but also to extend beyond both the proximal and distal margins of the lesion by at least 1-2 cm. In case of inadequate coverage, an additional stent can be deployed to completely cover the lesion. Several self-expanding metallic stents of appropriate calibre can be used. The covered stents is not indicated since although they prevent the intra-stent tumor growth, they can migrate .Choo I et al. have used two stent types recovered in 20 patients. Of the stents completely covered stents migrated 50% (17). The endocopy can help to advance the guide in the proximal colonic lesions (high descending colon and transverse colon), and in those cases that sharp and marked loops of the colon exist. We have used for these cases, with success, 10-12 F and 40 mm length introducer sheath (18) with double wire guides and catheter: hydrophilic wire guide with catheter and Lunderquist superstar wire guide. The introducer sheath allows to advance with easiness in these cases of tortuous colon and also, it has facilitated the biopsy with forceps of the lesion and it has allowed us to value the length of the lesion with the introduction of the contrast medium. Finally through the own introducer sheath we have deployed the stent . In palliative occasions when it is not possible to access for via anal we antegrade placement of colonic stent through cecostomy or colostomy (19). In two patients, non candidates for surgery, with obstruction in transverse colon and great dolicocolon, we carried out the deployement of the stent respectively through a percutaneous colostomy and cecostomy (20). Additional balloon dilation is not recommended because it is associated with a high risk of perforation because of excessive manipulation in these friable tumoral areas. Due to the nature of the self-expansion, stents are allowed to slowly expand over time. The peristaltic movement of the bowel after decompression may facilitate full expansion. Immediately after stent placement, the enema examination should be repeated to document post-stenting patency and correct positioning. The ends of the stent should not be covered by colonic folds. Otherwise, the position of the stent should be adjusted subsequently. Some authors suggest a low-residue diet and mineral oil given to the patients to prevent early intra-stent blockage. Twenty-four hours after stent placement, a plain radiograph of the abdomen is obtained to evaluate the position of the stent and to assess changes in the radiographic appearance of the obstruction. As the obstruction is decompressed, it is possible to stage the malignancy to determine whether the patient is a good surgical candidate. Follow-up radiographs should be taken whenever the patient becomes symptomatic to rule out stent migration, restenosis and perforation. Concluded the procedure the patient's constants are monitored and it recovers the electrolytic balance. In next days the oncologic stage can begin by means of computed tomography or ultrasound . If the patient is candidate for surgery, will receive preparation intestinal laxative. Simple abdominal x-ray at the 24 hours is achieved to check the situation and expansion of the stent. If the patient has appropriate general

state, the study to oncologic stage is carried out in an ambulatory way. Results Technical success in deploying the stents with fluoroscopic, endoscopic or combined guidance has ranged from 88% to 100% (15). Clinical or medical success (improvement of obstructive symptoms has been reported in 80%-92% (15). In a systematic review of the published data (1990-2000) with 598 cases of colonic stents, the technical success was achieved in 92% and clinical success in 88%. Palliation was achieved in 302 of 336 cases (90%) and in 223 of 262 cases (85%) the stent placement as a “ bridge to surgery” (21). In the same study the technical failure was reported in 47 of 598 patients (8%). The main causes were: 1. inability to across the stenosis with the wire guide in 36 cases malposition in 4 cases and perforation in 2. In remaining five patients there was inadequate information. Clinical failure was in 26 of 551 patients (5%). The main causes were: perforation in 8 cases, persistence of obstructive symptoms in 10 and adhesions of the colonic wall to the stent in 3. No data remaining five cases. J Beltran (22) compares two groups of patient with carcinoma of left colon. To the group A with carcinoma of left colon previously to the surgery a colonic stent was implanted, while the group B with neoplasia of left colon without obstruction was intervened in an elective way. In the study significant differences were not observed groups between both as much as morbility (50. 5 vs 45%), mortality (3.5%vs 2.8%), recurrence (3.5 vs 8.4) and survival at three years (76% vs 66%). Complications Lesser complications of stent placement such as minor rectal bleeding, transient anorectal pain, temporary incontinence and fecal impactation are common to many reports. More severe life threatening complications are described however, inclusively deaths. In the review of Khot et al (21): three patients (1%) of 598 died from a perforation and not successfully decompression . Perforation occurred in 22 of 598 patients (3.6%). Was a higher incidence of perforation in the dilatation group (10%). Stent migration occurred after 54 of 551 patients (9.8%). Usually migration occurred after 3 days (7.2%). Overall reobstruction occurred in 52 of 525 patients (9.9%). The causes of reobstruction were: tumor ingrowth 32 cases, stent migration 7 cases and fecal impactation in 13 cases. Lesser bleeding occurred in 24 patients and major bleeding in three patients [total 27 (4.5%) of 598 patients]. And finally, minor abdominal o rectal pain was described in 31 of 598 patients (5.1%). Limitations: Irradiation, cost and others The radiation exposure is a very important problem for the patient and for the operator when the procedure its carried out with fluoroscopy guide. We measured the duration of the procedure (in minutes) and the fluoroscopy time and studied the total radiation dose (dGycm2) in 12 patients The average procedural time transanal stent placement was 57.3 minutes and fluoroscopy time was 20.9 minutes. Mean total dose was 3593 dGycm2 (23). Transanal placement of metallic stents performed in a combined fashion with colonoscopic assistance may decrease fluoroscopy time (23). Osman H et al. (25) estimated that the cost of palliative treatment was less than half of that of a surgically decompressed case (1445£ vs 3205£). The cost of the “ bridge to surgery “ treatment was reduced by 12 % compared with a two-stage operative procedure (5035£ vs 5720 £). Binker CA et al. carried out a retrospective study comparing the total costs of two groups of patients: preoperative stent (12 patients) versus surgically treatment (11 patients) (26). They reported saving of up to 20% for presurgery stent placement versus surgical management. The decreases in the cost were attributed mainly to the shorter hospital stay. In relation with special location stents should not be inserted across distal rectal tumors because they can cause severe tenesmus or fecal incontinence (27). Is not there a completed randomised, prospective studies that compared preoperative stents with standard surgery in a group of patients with potentially resectable primary colorectal carcinoma and obstruction (28). Radiologic guidance vs endoscopy guidance The placement of colorectal stent has been carried out with endoscopic guidance, with radiological control and in a combined way. For both procedures advantages and disadvantages exist. Stent may be placed endoscopic in the right colon whereas with radiologic guidance alone, usually, stent placement limited to left colon (26). In both the effectiveness and the security would be identical. While in the endoscopy there would be smaller irradiation, this would be more expensive and more bothersome for the patient. Nevertheless the combination of both it could be the election technique.

Own clinical experience (16) The study included 136 consecutive patients with large bowel obstruction from colon carcinoma treated from January 1994 to September 1998. Barium demonstrated the lesion location in the sigmoid colon (76 patients), descending colon (33 patients), rectosigmoid junction (11 patients) and transverse colon (6 patients). Patients selection criteria for transanal stent insertion included malignant colonic obstruction distal to the left of the spine on barium examination. Wallstent 22 mm diameter with length ranging 66-100 mm were used. Criteria for successful stent placement were relief of colonic obstruction to allow for preoperative bowel preparation and as a palliative procedure for those patients not eligible for surgery. Results: Colonic obstruction was relieved in 115 (91%) patients. Bowel movements occurred within the first 24 hours in 98 (85%) patients and after 24 hours in 17 (15%) patients. Bowel preparation was followed by colonic resection and primary anastomosis in 81(70%) patients. Palliative decompression was achieved in 34 (30%) patients who attained a mean survival of 6.8 months (range 1-12 months) after stent insertion. Colostomy was performed in 11 (9%) patients in whom colonic obstruction persisted because of inability to place the stent (9 patients) and in two patients despite stent placement. Complications occurred in 15 (12%) patients: self limited rectal bleeding (10 patients), stent migration (3 patients) and symptomatic colonic perforation (2 patients). In our experience, transanal metallic stent placement is an effective method for relief of malignant colonic obstruction. The procedure allows preoperative bowel preparation and a single stage operation. Palliation of large obstruction improves the quality of life of patients in whom resection not indicated. Future direction Many authors agree that the management of the colon obstructed by means of stents is a sure and effective technique as previous treatment to the elective surgery and as palliative treatment in the obstructive carcinoma of left colon. Nevertheless they are necessary studies clinical randomised that evidence the obtained results. On the other hand it is necessary to design new devices of more diameter and with better subjection systems that allow to carry out the procedure in a easy and safety way.(29) References: 1.Boyle P Leon M. Epidemiology of colorectal cancer Br Med Bull 2002; 64:1-25 2. Dean GT, Krukowski Z Irwin S. Malignant obstruction of the left colon. Br J Surg 1994; 81: 1270-1276 3. Duxbury M, Brodribb A, Oppong F, Hosie K Management of colorectal cancer: variations in practice in one hospital Eur J Surg Oncol. 2003; 29: 400-402 4. Griffith R Prospective medical obstacles to surgery Cancer 1992; 70:1333-1341 5. Smothers L, Hynan L, Fleming J, Turnage R, Simmang C, Anthony T. Emergency surgery for colon carcinoma. Dis Colon Rectum 2003; 46: 24-30 6. Ohman U. Prognosis in patients with obstructing colorectal carcinoma. Am J Surg 1982; 143: 742-747 7. Witzig J, Morel P, Erne M, Egeli R, Borst F, Rhoner A. Chirugie des cancers digestifs des patients de plus de 80 ans Hel Chir Acta 1993; 59: 767-769 8. Brillari P, Aurello P, de Angelis R, Valabrega S, Ramacciato G, D’Angello F. Management and survival of patients affected with obstructive colorectal cancer . Int Surg 1992; 7: 251-255 9. Neugent K, Daniels P, Stewart B,Patankar R, Johnson Cd. Quality of life in stoma patients. Dis Colon Rectum 1999; 42: 1569-1574 10. Dohmoto M, Rupp K, Hohlbach G. Endoscopically-implanted prosthesis in rectal carcinoma Dtsch Med Wochenschr. 1990; 115: 915 11. Cwsikiel W, Andren-Sandberg A. Malignant stricture with colovesical fistula: stent insertion in the colon. Radiology 1993; 186: 363-364 12. Keen R, Orsay CP. Rectosigmoid stent for obstructing colonic neoplasms Dis Colon Rectum 1992; 35:912-913 13. Tejero E, Mainar A, Fernandez L, Tobio R, De Gregorio MA. New procedure for the treatment of colorectal neoplastic obstructions. Dis Colon Rectum 1994; 37: 1158-1159 14. Zollikofer C, Jost R, Schonh E, Decurtins M Gastrointestinal stenting Eur Radiol 2000; 10: 1158-1159 15. Mauro M, Koehler R, Baron T. Advances in gastrointestinal intervention: The treatment of gastroduodenal and colorectal obstructions with metallic stents. Radiology 2000; 215:659-669 16. De Gregorio MA, Mainar A, D’Agostino H, Herrera M, Tejero E, Medrano J. Trasanal metallic stents for malignant colonic obstructions

in 126 patients. JVIR 1999; 10 Suplement: 219. 17. Choo I, Do Y, Suh S, Chun H, Choo S, Park H et al. malignant colorectal obstruction: treatment with a flexible covered stent . Radiology 1998; 206. 415-421 18. De Gregorio MA, Mainar A, Tejero E, Alfonso ER, Gimeno MJ, Herrera M. Use of an introducer sheath for colonic stent placement Eur Radiol 2002; 12: 2250-2252 19.Vellin TE, Hall L, Brenan F. Colonic stent placement facilitated by percutaneous cecostomy and antegrade enema. AJR 2000; 175: 119-120 20.Gimeno MJ, Alfonso ER, Herrera M, Tobio R, Medrano J, De Gregorio MA. Palliative treatment of malignant stenoses of transverse colon by autoexpandable metallic stent through percutaneous cecostomy or colostomy. Cardiovas Intervent Radiol 2002; 25 (Supl): 208 21. Khot U, Wenk Lang A, Murali K, Parker M. Systematic review of the efficacy and safety of colorectal stents. Br J Surg 2002; 89:1096-1102 22. Beltran J. Left obstructive colonic carcinoma. Comparative study of short and middle-term results after a new therapeutic procedure based in self-expanding metallic stents placement. Doctoral Tesis.Universidad de Zaragoza Junio 2003 23. De Gregorio MA, Mainar A, Gimeno MJ, Tejero E, Medrano J, Herrera M et al Radiation exposure during metallic stent placement for the treatment of malignant colonic obstruction: experience with 12 patients. Cardiovas Intervent Radiol 2001; 24: S 189 24. Gimeno MJ, Medrano J, Alfonso ER, De Gregorio MA. Utility of flouroscopy combined with endoscopy in the treatment of malignat colorectal stenosis with metallic stents Cardiovas Intervent Radiol 2002; 25: S 203 25.Osman H, Rashid H, Sathananthan N, Parker M.The cost effectiveness of self- expanding metal stents in the management of malignant left sided bowel obstruction. Colorectal Dis 2000; 2: 233-237 26. Binkert C, Lederman H, Jost R, Saurenmann P, Decurtins M, Zollikofer C. Acute colonic obstruction: clinical aspects and cost efectiveness of preoperative and palliative treatment with self-expanding metallic stents. A preliminary report. Radiology 1998; 206: 199-204 27. Turegano F, Echenagusia A, Simo A,Camuñez F, Muñoz F, Del Valle E et al. Transanal self expanding metal stents as an alternative to palliative colostomy in selected patients with maligant obstruction of the left colon. Br J Surg 1998; 85: 232-235 28. Baron T Expandable metal stents for the treatment of cancerous obstruction of the gastrointestinal tract. N Engl J Med 2001, 344. 16811687 29. Maynar M, Quian Z. Current status of the expandable metallic stent for the treatment of colorectal obstruction. Cardiovasc Intervent Radiol 200; 21 Supl: 114-116

Special Session New Trends in IR I 18.1 Sirolimus Coated Stents in the SFA S. H. Duda, B. Wiesinger, G. H. Tepe; Dept. of Diagnostic Radiology, University Hospital, Tuebingen, Germany. Restenosis Restenosis after PTA procedures or peripheral vascular stent implantation is known to be the crucial point determining the long term patency rates in vascular intervention. Late lumen loss is basically the result of three distinct processes: vessel remodeling, elastic recoil and neointimal hyperplasia. Stents are able to abolish recoil and to prevent remodeling in the coronary and superficial femoral arteries.1,2 Restenosis rates for the superficial femoral arteries after PTA and/or stent placement is estimated between 22 and 65% after 12 months.3,4 Recently, many different approaches were attempted to prevent the problem of neointimal hyperplasia. Among the most promising is the approach of drug-eluting stents, where pharmacologically different agents are delivered via the stent surface in the adjacent vessel wall and the blood. Among these agents Sirolimus (Rapamycin) was shown to virtually abolish neointimal hyperplasia in many different coronary trials, such as RAVEL and SIRIUS.5,6 The SIROCCO trial was intended as a feasibility trial with 36 patients enrolled. In a second phase another 54 patients were enrolled . New approaches to fight restenosis - Drug Eluting Stents Recently many different coating techniques were evaluated. Corrosion prevention with materials such as titanium or gold, ceramics (e.g. diamond-like carbon), polymers, and radioactive coatings were tested for

the prevention of restenosis. The most encouraging results were obtained with pharmacologically active coatings, such as sirolimus (Rapamycin). In impressive coronary trials such as RAVEL (238 patients) no restenosis and no late loss (-0.01mm) was seen for the sirolimuseluting group at 6 months, whereas the late loss was 0.80mm in the uncoated group.5 SIRIUS, an 1101 patient coronary trial with more challenging lesions, revealed 0.14mm of late lumen loss for the sirolimus group versus 0.92mm in the bare stent group during the interim analysis of the first 400 patients. This translated into significantly better clinical event free survival rates.6 Other agents, like paclitaxel are currently under investigation in the TAXUS, ASPECT and ELUTES trials. Here again, results are better than those obtained with bare stents. In the TAXUS1 trial for example, the 6month late loss was 0.35mm compared to 0.71mm in the bare stent group. However, restenosis prevention with paclitaxel seems to be dose dependent.7 With dexamethasone (STRIDE trial, 70 patients) on a BiodivYsio stent platform, results at 6 months showed a restenosis rate of 13.3% and a late loss of 0.45mm.8 Other agents used are batimastat, a metalloprotease inhibitor (BRILLIANT1 trial), and actinomycin, a cytotoxic agent investigated in the ACTION trial. Both trials were suspended for the occurrence of infarctions and deaths in the early follow-up. With 17beta estradiol (EASTER) as well as tacrolimus (PRESENT trial) encouraging 30- day follow-up data were reported.9,10 The PRESENT trial, however, revealed 2 cases of in-stent restenosis at 30 days so that until now, the only really solid coronary database exists for Sirolimus. Sirolimus Sirolimus (Rapamycin) is a natural, macrocyclic, lipophilic lactone with immunosuppressive, antibiotic activity originally isolated in the mid 70s from Streptomyces hygroscopicus found in Easter Island soil samples (C51H79NO13). Its antimicrobial and antifungal activity led finally to its approval as the immunosuppressive agent Rapamune by the FDA in 1999 for the prophylaxis of renal transplant rejection.11,12 It belongs to the same class of immunosuppressive agents as cyclosporin A and tacrolimus binding to specific cytosolic proteins (immunophilins). However, binding to FK binding protein 12, it inhibits the mammalian target of rapamycin mTOR leading to elevated p27KIP1 levels reducing the CDK4/ Cyclin D and CDK2/Cyclin E complex activities and in the reduction of RB S6 kinase protein phosphorylation. This leads finally to an early blockade in the G1-S cell cycle and to an inhibition of cell proliferation.12 In vitro and in vivo studies demonstrated that sirolimus is capable of virtually abolishing both T cell proliferation and SMC (small muscle cell) proliferation and migration, which is a cardinal pathophysiologic target for the prevention of restenosis.13,14 SIROCCO Trial design 36 patients with symptomatic peripheral atherosclerotic disease (Rutherford stages 2-4) were treated in a multicenter, doubleblind, prospective, 2-arm trial (n=18 with bare stents and n=18 with coated stents). All had obstructive, native, de novo or restenotic lesions with a diameter stenosis of greater 70% over a length ranging between 7 to 20cm or occlusions ranging between 4 and 20cm in the superficial femoral artery. Exclusion criteria included poor aortoiliac or common femoral inflow, previously stented regions, tandem lesions and ischemic tissue loss. The study endpoints were the in-stent diameter stenosis via quantitative angiography at 6 months as assessed by an independent core lab. Restenosis was evaluated defining binary restenosis as a lumen reduction of 50%. The secondary endpoints included duplex ultrasound and ankle brachial index at discharge and at 1, 6, 9 and 18 months, hemodynamic failure, X-rays at 18 months, pharmacokinetic sampling and the incidence of serious adverse events. SIROCCO Stents The stent used was a self-expandable SMARTTM CORDIS, crush recoverable nitinol stent with 8cm length. The unconstrained stent diameters were chosen 1-2mm larger than the vessel diameter (6 or 7 millimeters). The translucent coating comprised of a drug/copolymer matrix delivers an equivalent amount of drug/vessel as the CYPHER™ Coronary Stents. A maximum of three stents per patient were implanted. SIROCCO Results The mean number of stents placed was 2.2. Baseline demographics were similar in the 2 treatment groups. Only the percentage of patients who smoked or had diabetes was slightly higher in the sirolimus-eluting stent group than in the bare stent group (32.9% to 16.7%, P =0.305, and 50.0% to 28.8%, P =0.264, respectively). The differences between

the treatment groups otherwise were not statistically significant. The level of ischemia (higher Rutherford stages, stages 3 and 4) was also more pronounced, but not significantly different between the sirolimus coated stent group and the uncoated stent group (61.1% and 33.3%, P =0.181). The mean lesion length was 85 mm. Total occlusions were present in 57% of patients. One noteworthy difference between the treatment groups was that 100% of the lesions in the sirolimus-eluting stent patients had moderate-to-severe calcification (defined as readily apparent densities noted within the vascular wall) as compared to 46.7% of lesions in the uncoated stent patients (P =0.002). The patients who received sirolimus-eluting stents in this small cohort were generally at greater risk of restenosis or complications than the control group. The in-stent mean percent diameter stenosis evaluated via quantitative angiography immediately after the procedure was 21.2±18% in the sirolimus-eluting stent group and 23.3%±9.5% in the uncoated stent group (P =0.554). At the 6-month follow-up, the mean percent stenosis of the diameter was 22.6±16.5% in the sirolimus-eluting stent groupand 30.9±27.2% in the uncoated stent group (P =0.294). Also at 6 months, the in-stent mean lumen diameter was 4.95±0.59 mm in the sirolimus-eluting stent group and 4.31±1.39 mm in the uncoated stent group (P =0.047). The in-stent mean lumen diameter of 4.95 mm in the sirolimus-eluting stent group was larger than the reference diameter (4.81 mm). This step-up/step-down effect is often seen with the use of self-expanding stents that are 1 to 2 mm larger than the vessel diameter. As a result, with suppression of intimal hyperplasia, it would be possible for the in-stent mean diameter to be larger than those of the native vessel. There was no occlusion observed in the sirolimus group versus one in the bare stent group at 6 months. Binary in-lesion restenosis rate was 0% in the coated versus 23.5% in the uncoated group, binary restenosis was defined as stenosis>50%. In both treatment groups no revascularization procedure was required. Figure 1 shows an angiographic followup after 19 months because the patient underwent an iliac artery stenting and Figure 2 an ultrasound image showing neointimal hyperplasia in a bare stent patient. The six stent fractures (plus one questionable) that occurred were equally distributed in both stent groups. For one patient a covered stent was placed to treat an asymptomatic ulceration of the adjacent vessel wall in the fractured area. Serious adverse events were two inflammations of the treated leg that were observed in the sirolimus group at 6 months and two unrelated deaths the follow up period in two other sirolimus patients for stroke, acute myocardial infarction, and pulmonary artery embolism. Conclusion The use of sirolimus-eluting stents in the SFA is safe and feasible. The release kinetic of the drug seems to be crucial.15 The available data on sirolimus-eluting stents is consistent among all the studies and presently demonstrate the best results among the efforts in the prevention of restenosis, such as intraluminal radiation therapy16 or novel approaches like sonotherapy and photoangioplasty.17,18,19 Sirolimus-eluting stents will have a role in the treatment of peripheral SFA disease,especially in high risk patients (diabetics, heavily calcified lesions, long occlusions). Furthermore, the bare stent group performed very well and much better than expected in the light of previous reports with Wallstents and PALMAZ® Stents. An extension including 54 patients is currently underway with the slow sirolimus-eluting SMART Stents to corroborate the previous findings. References: 1. Fischman DL, Leon MB, Baim DS, et al. A randomized comparison of coronary stent placement and balloon angioplasty in the treatment of coronary artery disease. Stent Restenosis Study Investigators. N Engl J Med 1994; 331: 496–501 2. Grimm J, Muller-Hulsbeck S, Jahnke T, et al. Randomized study to compare PTA alone versus PTA with Palmaz stent placement for femoropopliteal lesions. J Vasc Interv Radiol 2001; 12: 935–942 3. Gray BH, Olin JW. Limitations of percutaneous transluminal angioplasty with stenting for femoropopliteal arterial occlusive disease. Semin Vasc Surg 1997; 10: 8–16 4. Cheng SW, Ting AC, Wong J. Endovascular stenting of superficial femoral artery stenosis and occlusions: results and risk factor analysis. Cardiovasc Surg 2001; 9: 133–140 5. Morice MC, Serruys PW, Sousa JE. A randomized comparison of a Sirolimus-eluting stent with a standard stent for coronary revascularization. N Engl J Med 2002; 346: 1773–1780

6. Leon MB, Moses JW, Fitzgerald PJ, Popma JJ. SIRIUS 400: A US multicenter, randomized, double blind study of the sirolimus eluting stent in de novo coronary lesions. Sirolimus/Expert presentations. Available at http://www.tctmd.com/expert-presentations/multislide.html?product_id=2237. Accessed May 5, 2003 7. Grube E. Taxol/Taxane coated stents (polymer based). Paclitaxel/ Expert presentations. Available at http://www . tctmd.com/expert-presentations/multi- slide.html?product_id=2204. Accessed May 5, 2003 8. De Scheerder IK. Study of anti- restenosis with the Biodivysio Dexamethasone-eluting stent (STRIDE). Dexamethasone/Clinical trials and scientific slides. Available at http://www.tctmd.com/expert-presentations/ multi-slide.html?product_id=1502. Accessed May 5, 2003 9. New G. 17 beta Estradiol and the EASTER trial. Estrogen/Expert presentations. Available at http://www.tctmd.com/expert-presentation/ multi-slide.html?product_id =3086. Accessed May 5,2003 10. Diestel E. Promising results with Tacrolimus-the PRESENT and EVIDENT trials. Associated meeting ACC 2002, news from major meetings. Available at http://www.tctmd.com/meeting-news/ one.html?news_item_id =2529. Accessed May 5, 2003 11. Marx SO, Marks AR. The development of rapamycin and its application to stent restenosis. Circulation 2001; 104: 852–855 12. Zohlnhofer D, Klein CA, Richter T, et al. Gene expression profiling of human stent-induced neointima by cDNA array analysis of microscopic specimens retrieved by helix cutter atherectomy: detection of FK506binding protein upregulation. Circulation 2001; 103: 1396–1402 13. Suzuki T, Kopia G, Hayashi S, et al. Stent based delivery of sirolimus reduces neointimal formation in a porcine coronary model. Circulation 2001; 104: 1188–1193 14. Sousa JE, Costa MA, Abizaid A, et al. Lack of neointimal proliferation after implantation of sirolimus-eluting stents in human coronary arteries: a quantitative coronary angiography and three-dimensional intravascular ultrasound study. Circulation 2001; 103: 192–195 15.Duda S. H., Pusich B., Richter G., Landwehr P., Oliva V. L., Tielbeek A., Wiesinger B., Hak J. B., Tielemans H., Ziemer G., Cristea E., Lansky A., Beregi J. P. Sirolimus-Eluting Stents for the Treatment of Obstructive Superficial Femoral Artery Disease: Six-Month Results. Circulation 2002; 106(12): 1505-1509 16. Leon MB, Teirstein PS, Moses JW, et al. Localized intracoronary gamma-radiation therapy to inhibit the recurrence of restenosis after stenting. N Engl J Med 2001; 344: 250–256 17. Kuntz RE, Moses JW, Abizaid AAC, et al. Intravascular sonotherapy in human coronary arteries: first results of a feasibility trial (abstr). J Am Coll Cardiol 2001; 37: 45A 18. Rockson SG, Lorenz DP, Cheong WE, et al. Photoangioplasty: an emerging clinical cardiovascular role for photodynamic therapy. Circulation 2000; 102: 591–596 19. Duda, Stephan H. MD; Poerner, Tudor C. MD; Wiesinger, Benjamin MD; Rundback, John H. MD; Tepe, Gunnar MD; Wiskirchen, Jakub MD, and; Haase, Karl K. MD, FACC. Drug-eluting Stents: Potential Applications for Peripheral Arterial Occlusive Disease. JVIR 2003; 14(3): 291301

Special Session Imaging of the Heart and Coronary Arteries 20.1 Imaging of the heart and coronary arteries S. M. Tepe; Radiology, TEST Cardiovascular Imaging, Istanbul, Turkey. OBJECTIVES:Coronary artery calcium is a marker for arteriosclerosis. As symptoms of cardiac disease may be silent, ultrafast computed tomography has been shown to be a valuable screening method to assess the coronary event risk. It predicts the coronary event risk by demonstrating the amount of calcifications in the coronary arteries. Plaque volume and mean atenuation values are very important factors for coronary artery disease diagnosis, prognosis and follow up. Electron Beam Computed Tomography Angiography is a non invasive test for evaluation of anatomy, stenoses of the native coronary arteries and the patency of native arteries and bypass grafts. MATERIALS AND METHODS: Between february 2000 and april 2003 we performed more than 1000 (thousand) coronary angiography and over 5000 coronary calcium screening test with Ulltrafast Electron Beam CT (C-150 scanner GE-Imatron) scanner which is without a mov-

ing part much faster (50 msec exposure time/per slice) than a modern spiral CT. Rapid acquisition of a large number of images with EBCT allows accurate imaging of the coronary arteries and their walls. Imaging time for a cardiovascular EBCT examination is completed in 20 to 40 seconds during a breath hold. A scan run consisted of the acquisition of 40 contiguous transverse two-dimentional images of 3- mm-thick sections at the level above the coronary artery origins to the cardiac apex. Parameters used were 130 kVp, 630 mA. Images were acquired with ECG triggering at 80 % of the R-R interval during diastole, obtained by using a 26 cm2 field of view and a 512x 512 reconstruction matrix was used. Image quality and scoring accuracy was assessed by the radiologist who carefully inspected each image and made vessel-by-vessel and calcific focus-by-calcific focus. Care was taken that no calcific focus was overlooked and and that no noncoronary calcific focus was included. Images could be negatively affected by cardiac, respiratory or patient motion, arrythmias, noise, artifact, triggering problems, periarterial calcifications. RESULTS:Calcium scores show the severity and dimension of atherosclerosis. The ideal calcium score is zero for all of age groups. Atenuation values gives information about plaques age. The plaques with a low density values are young, recently calcified, high density value lesions are old plaques. Calcium scores are evaluated as Agatston scores and have ranges between 0-400 and higher. With EBTA, non or critical stenoses, coronary anomalies, patency of bypass grafts has been accurately demonstrated. CONCLUSION:Increased coronary event risk was associated with even mild calcium scores according to patients`s age. Noninvasive ultrafast coronary angiography allows coronary imaging with a robust technique and cine cardiac studies make possible to calculate ejection fraction, visualize ventricules, wall thickness and motion. References: 1.Araoz AP, Mulvagh SL, Tazelaar HD, Julsrud PR, Breen JF. CT and MR imaging of Benign Primary Cardiac Neoplasms with Echocardiographic Correlation. Radiographics 2000; 20:1303-9. 2.Burke A, Virmani R. Tumors of the heart and great vessels In: Atlas of tumor pathology. Fasc 16, ser 3. Washington DC: Armed Forces Institute of Pathology, 1996; 1-98. 3.Araoz AP, Eklund HE, Welch TJ, Breen JF. CT and MR imaging of Primary Cardiac Malignancies. Radiographics 1999;19: 1421-34. 4.Chiles C, Woodard PK, Gutierrez FR et al. Metastatic involvement of the Heart and Pericardium: CT and MR imaging. Radiographics 2001; 21: 439-449. 5.Thurber DL, Edwards JE, Achor RWP. Secondary malignant tumors of the pericardium. Circulation 1962; 26:228-241. 6.Waller BF, Gering L, Branyas NA et al. Anatomy, histology and pathology of the cardiac conduction system: part IV. Clin Cardiol 1993; 16:507-511. 7.Reyen K. Cardiac myxomas. N Engl J Med 1995;333:1610-7. 8.Kuhlman JE, Singha NK. Complex disease of the pleural space: Radiographic and CT evaluation. Radiographics 1997; 17:63-79. 9.Knollman FD, Hummel M, Hetzer R, Felix R. CT of heart transplant Recipients: Spectrum of disease. Radiographics. 2000; 20:1637-48. 10.Grebenc ML, Rosado-de- Christenson ML, Green CE, Burke AP, Galvin JR. Cardiac myxoma: imaging features in 83 patients. Radiographics 2002;22:673-89. 11.Gomes AS, Lois JF, Child JS, Brown K, Batra P. Cardiac tumors and thrombus: evaluation with MR imaging. American Journal of Roentgenelogy 1987; 149:895-99. 12.Funari M, Fujita N, Peck WW, Higgins CB. Cardiac tumors: assesment with Gd- DTPA enhanced MR imaging. J Comput Assist Tomogr 1991; 15:953-8. 13.Higgins CB. Acquired heart disease. In: Higgins CB, Hricak H, Helms CA, eds. Magnetic resonance imaging of the body. Philadelphia , Pa: Lippincott –Raven, 1997;409-60. 14.Stanford W, Rumberger J. Ultrafast CT in Cardiac Imaging: Principles and Practice. Stanford W, Rooholamini SA, Galvin JR. Ultrafast CT in the detection of intracardiac masses and pulmonary artery thromboembolism. 1992, Chapter 13 : 235-41 15.Achenbach S, Giesler T, Ropers D, et al. Detection of coronary artery stenosis by contrast preenhanced, retrospectively ECG –gated, multislice computed tomography. Circulation 2001; 103: 2535-38. 16.Funabashi N, Kobayashi Y, Perloth M, Rubin GD. Coronary artery: Quantitative evaluation of normal diameter determined with Electron Beam CT compared with Cine Coronary Angiography-Initial experience.

Radiology 2003; 226: 263-71. 17.Hunold P, Vogt FM, Schmermund A, Debatin JF, Kerkhoff G, Budde T, Erbel R, Ewen K, Barkhausen J. Radiation exposure during CT: Effective doses at Multidetector row CT and Electron beam CT. Radiology 2003; 226:145-52. 18.Achenbach S, Moshage M W, Ropers D, Nossen J, Daniel W G. Value of Electron beam computed tomography for the noninvasive detection of high pregrade coronary artery stenosis and occlusions. N Engl J Med 1998; 339: 1964-71. 19.Achenbach S, Ulzheimer S, Baum U, et al. Noninvasice coronary angiography by retrospectively ECG gated multislice spiral CT. Circulation 2000; 102: 2823-28. 20.Warren R. Janowitz. Coronary Artery Calcium. Radiology 1999;211:288-290. 21.Agatston AS, Janowitz WR, Hildner FJ, Zusmer NR, Viamonte M, Jr, Detrano R. Quantification of coronary artery calcium using ultrafast computed tomography. J Am Coll Cardiol 1990; 15:827-832. 22.Shemesh J, Apter S, Stroh CI, Itzchak Y, Motro M. Cardiac Imaging. Tracking coronary calcification by using Dual -section spiral CT:A 3 year follow up. Radiology 2000:217:461-465. 23.SL Mautner, GC Mautner, J Froehlich, IM Feuerstein, MA Proschan, WC Roberts, JL Doppman. Coronary artery disease:prediction with in vitro electron beam CT. Radiology 1994:192:625-630.

20.3 Imaging of the heart and coronary arteries: multislice CT imaging A. F. Kopp; Department of Diagnostic Radiology, University of Tuebingen, Tuebingen, Germany. Despite a multitude of different medical and interventional strategies to treat coronary artery disease (CAD), the natural course of CAD is a relentless progression. The current gold standard to assess the degree of stenosis is coronary angiography. In Germany alone, the total number of angiographic procedures rose by 45 % from 1995 to 2000, while the fraction of interventional procedures remained almost constantly low at about 30%. Although coronary angiography has become a safe procedure with only a small risk associated, the inconvenience for the patient as well as the economic burden have fueled the quest to find an alternative, non-invasive method to visualize an assess coronary arteries. In the last few years mechanical Multidetector-row CT (MDCT) systems with simultaneous acquisition of up to 16 slices and less than half second scanner rotation have become available (1,2). Multi-row acquisition with these scanners allow for considerably improved visualization of the coronary arteries. Initial experiences have shown that coronary lesions can be detected with good sensitivity and specificity (3). CTA coronary angiography The imaging protocol for multislice CT angiography of the coronary arteries is relatively straightforward. To establish the scan delay time a test bolus of 15 ml CM and 20 ml saline chaser bolus is used. The circulation time is determined by measurements of CT density values in the ascending aorta. Imaging commences at the circulation time plus 3 s. A bolus of 120 ml nonionic contrast (400 mg I/ml) is injected trough an 18-gauge catheter into an antecubital vein at 4 ml/sec followed by a 50 ml saline chaser bolus. The angiographic series is a continuous spiral scan with the calculated pitch. Even with 4-row-technology the z-resolution is significantly improved compared to electron beam CTA, where only sequential ECG-triggered 3-mm slices can be obtained. With 4slice technology 1 mm slices and sub-millimeter image increment provide nearly isotropic volume data within a single breath-hold for high resolution CT volume imaging. Non-invasive MSCTA with 4-slice technology showed a high diagnostic accuracy in the detection and quantification of coronary lesions (4). The results of MSCT coronary angiography published so far from different centers are very encouraging. CTA of the coronary arteries in 103 patients yielded a sensitivity of 88%, a specificity of 93%, a positive predictive value of 0.91, and a negative predictive value of 0.90 for detection of hemodynamically significant stenoses in the major segments of the coronary arteries (4). The results revealed a high diagnostic accuracy of MSCT, in the detection and quantification of coronary lesions since not only severe, but also intermediate and mild lesions could be visualized. One of the most important findings, however, of the ongoing studies is that MSCTA allowed detection and assessment of noncalcified lipid-rich plaques (5). To confirm these findings and to investigate non-invasive detection of coronary plaques and plaque composition by multislice-CT

intracoronary ultrasound (ICUS) was used as a gold-standard. Multislice-CT and ICUS yielded identical results in regard of plaque composition and quantification of lesions (6). Thus, this new technology holds promise to allow for the non-invasive detection of rupture-prone soft coronary lesions and may have the option to lead to early onset of therapy (7). Limitations A number of factors are known to decrease image quality and make image interpretation difficult. The two factors mostly held responsible are higher heart rates and severe calcifications. Becker was the first to describe the negative effect of higher heart rates on image quality. These data have been confirmed by others: excellent diagnostic image quality can only be obtained at heart rates < 65 bpm (8). The reason for this heart rate limitation lies in the temporal resolution of the CT image acquisition and reconstruction system. To obtain heart rates below 65 bpm for optimal image quality either 80mg Esmolol i.v. or 50-100 mg metoprololtartrate orally is administered prior to the scan. Assessment of luminal diameter in the presence of severe calcifications yields unsatisfactory results. Especially if non-high grade coronary lesions are known, it can be difficult to determine the progress of that specific lesion. However there is only limited published data available that quantifies the amount of calcification critical for image interpretation. Evaluation of Cardiac Function Beside the diagnosis of cardiac and coronary morphology, evaluation and quantification of cardiac function provides important information for the assessment of cardiac and coronary diseases. In addition to image reconstruction in the diastolic phase of the cardiac cycle for assessment of morphology ECG-gated multi-slice spiral CT can provide additional 3D reconstructions in different heart phases based on the same scan data. Thus, the diagnosis of cardiac and coronary morphology and also of basic cardiac function parameters such as left and right ventricular ejection fraction and end-diastolic and end-systolic volume can be derived from a single contrast enhanced ECG-gated spiral examination with thin-section acquisition. First study results show that basic cardiac function parameters derived with 4-slice CT correlate well with the gold-standard techniques MRI and coronary angiography based on a standardized heart phase selection for end-diastolic and end-systolic CT reconstruction and semiautomated evaluation tools. Use of multi-planar reformations in short and long heart axis with thickness of 5-8mm enables appropriate delineation of the ventricle wall in both end-diastolic and end-systolic reconstruction. Automated direct 3D reconstruction of oblique planes in pre-defined views such as short and long heart axis and in multiple phases of the cardiac cycle will enable a more efficient workflow for cardiac function analysis in the future. Rapid cardiac motion during the systolic phase of the cardiac cycle can cause motion artifacts in the end-systolic reconstructions. However, correct delineation of the ventricle walls is usually still possible with sufficient accuracy. Latest 16-slice CT scanners have the potential to further improve the accuracy of cardiac function measurement as compared to 4- and 8-slice CT scanners based on increased gantry rotation speed from 0.5 s to 0.42 s with a best possible temporal resolution of 105 ms. More advanced analysis of cardiac function such as quantitative wall motion analysis does require temporal resolution consistently better than 100 ms. At this point even faster gantry rotation speed will be needed in combination with new segmented reconstruction algorithms for retrospective improvement of temporal resolution at the expense of reduced spatial resolution in the longitudinal direction. From 4 to 16 rows… True isotropic resolution has not yet been reached with 4-slice CT systems. Consequently an increased number of simultaneously acquired slices and sub-millimeter collimation for cardiac applications was the next step on the way towards true isotropic scanning with multislice CT. The recently introduced multislice CT scanner Siemens SOMATOM Sensation 16, offering simultaneous acquisition of 16 slices with 0.75 mm or 1.5mm collimated slice width each, is the first scanner of this new generation. Similar to the 4-slice CT scanner SOMATOM Volume Zoom the SOMATOM Sensation 16 has an Adaptive Array Detector. It consists of 24 detector rows, the 16 central ones being 0.75 mm wide in the center of rotation, the 4 outer ones on both sides being 1.5 mm wide. The total z-coverage in the iso-center is 24 mm. For CTA of the coronary arteries a collimation of 0.75mm can (13.2mm/s feed) with a gantry rotation time of 420ms is used. Spiral scanning with 16 sub-millimeter slices represents a breakthrough on the way towards true isotropic resolution for routine clinical applications. As a consequence, the distinc-

tion between longitudinal and in-plane resolution will gradually become a historical remnant, and the traditional axial slice will loose its clinical predominance. The simultaneous acquisition of 12 0.75-mm cross-sections yields higher spatial resolution, and at the same time, the overall scan time is substantially shorter than with previous scanners. Improved spatial resolution, through reduction of partial volume effects, improves diagnostic accuracy and potentially reduces the problems caused by calcification, whereas a shorter scan time may improve scan quality through the shorter breathhold duration and, in addition, requires less contrast agent. Finally, the faster rotation (420 ms) compared with previous scanners provides higher temporal resolution. Ropers et al. analyzed coronary stenoses in 16-row coronary MDCTA with a reference diameter down to 1.5 mm, thus covering all lesions that may be potential targets for revascularization (9). Both motion and calcification rendered fewer arteries (12%) unevaluable than in most previous studies and a high sensitivity (92%) and specificity (93%) for the detection of coronary stenoses was achieved, confirming results previously obtained with this technology (9). In a recent study to evaluate feasibility, image quality and clinical accuracy in detecting coronary artery lesions using a CT scanner with 16 detector-slices and faster gantry rotation speed MDCTA yielded a sensitivity of 98 %, specificity of 98%, a positive predictive value of 0.8 and a negative predictive value of 1.0% if all coronary segments, regardless of their size were included (3). To answer the question whether a patient would benefit from an interventional procedure 58 out of 60 (97%) patients studies were correctly diagnosed using MDCT. Nieman et al. (10) demonstrated similar results in a comparable patient population using the same 16 slice CT technology. However, their analysis was restricted to the major branches with a vessel size 2 mm. Outlook The temporal resolution of current multi-slice CT scanners needs to be improved to provide motion-free and robust coronary imaging also for moderate and high heart rates. Increased temporal resolution can be achieved by segmented reconstruction techniques using 2 or more segments from consecutive heart cycles for reconstruction. Latest 16-slice scanners achieve 0.42s rotation time and a best temporal resolution of 105ms by using 2-segment reconstruction. However, segmented reconstruction can improve temporal resolution only in a limited range of heart rates and at the expense of blurring artifacts in the presence of heart rate changes. Further increased rotation speed is the most favorable approach to increase temporal resolution. E.g. 0.3s rotation time can produce a consistent temporal resolution of 150ms for all heart rates that may be able to provide motion-free data in patients with low and moderate heart rate (presumably up to 80min-1), thereby reducing the number of patients that require heart rate controlling medication. Obviously, significant development efforts will be needed to handle the increase of mechanical forces and the increased data transmission rates. The ultimate CT-scanner should cover the entire coronary anatomy in a single heart beat without movement of the table. This can be achieved with area detectors that cover about 120mm scan range with at least 0.5mm spatial resolution. Area detector technology and related new cone beam reconstruction techniques are in research, that can provide in-plane and through-plane spatial resolution of 0.2mm. With these CT scanners imaging of high-resolution morphology as well as dynamic and functional information via repeated scanning of the same scan range may be possible. The application potential of such technology is being evaluated with first experimental systems using phantom models and post mortem hearts. Initial experience shows that today's flat panel detector technology is yet too limited in low contrast resolution and the high radiation dose that is needed to provide adequate signal-to-noise ratio even for high contrast studies is unacceptable for use in human subjects. Due to the intrinsic slow decay times of flat panel detectors available today only slow rotation times 10s are possible at the present time. References: (1) Kopp AF, Küttner A, Schröder S, Heuschmid M, Claussen CD. New developments in cardiac imaging: the role of MDCT. J Clin Basic Cardiol 2001; 4:253-60. (2) Kopp AF, Schröder S, Küttner A, Heuschmid M, Georg C, Ohnesorge B, Kuzo R, Claussen CD. Coronary arteries: retrospectively ECG-gated multi-detector row CT angiography with selective optimization of the image reconstruction window. Radiology 2001; 221:683-88. (3) Heuschmid M, Küttner A, Flohr T, Wildberger JE, Lell M, Kopp AF,

Schröder S, Baum U, Schaller S, Hartung A, Ohnesorge B, Claussen CD. Visualization of coronary arteries in CT as assessed by a new 16 slice technology and reduced gantry rotation time: first experiences. Fortschr Röntgenstr 2002; 174:721-24. (4) Kopp AF, Schröder S, Küttner A, Baumbach A, Heuschmid M, Georg C, Kuzo R, Ohnesorge B, Karsch KR, Claussen CD. Non-invasive coronary angiography with high resolution multidetector-row computed tomography: results in 102 patients. Eur Heart J 2002; 23:73-84. (5) Schröder S, Kopp AF, Baumbach A, Küttner A, Georg C, Ohnesorge B, Herdeg C, Claussen CD, Karsch KR. Non-invasive detection and evaluation of atherosclerotic plaques with multi-slice computed tomography. JACC 2001; 37:1430-1435. (6) Kopp AF, Schröder S, Baumbach A, Küttner A, Georg C, Ohnesorge B, Heuschmid M, Kuzo R, Claussen CD. Non-invasive characterisation of coronary lesion morphology and composition by multislice computed tomography: first results in comparison with intracoronary ultrasound. Eur Radiol 2000; 11:1607-11. (7) Schröder S, Flohr T, Kopp AF, Küttner A, Herdeg C, Baumbach A, Ohnesorge B. Accuracy of density measurements within plaques located in artificial coronary arteries by multi-slice computed tomography: results of a phantom study. J Comput Assist Tomogr 2001; 25:900-906. (8) Giesler T, Baum U, Ropers D, Ulzheimer S, Wenkel E, Mennicke M, Bautz W, Kalender WA, Daniel WG, Achenbach S. Noninvasive visualization of coronary arteries using contrast-enhanced multidetector CT: influence of heart rate on image quality and stenosis detection. AJR Am J Roentgenol 2002; 179:911-16. (9) Ropers D, Baum U, Pohle K, Anders K, Ulzheimer S, Ohnesorge B, Schlundt C, Bautz W, Daniel WG, Achenbach S. Detection of coronary artery stenoses with thin-slice multi-detector row spiral computed tomography and multiplanar reconstruction. Circulation 2003; 107:664-66. (10) Nieman K, Cademartiri F, Lemos PA, Raaijmakers R, Pattynama PM, de Feyter PJ. Reliable noninvasive coronary angiography with fast submillimeter multislice spiral computed tomography. Circulation 2002; 106:2051-54.

Honorary Lecture Andreas Grüntzig Lecture 24.1 Thrombosis and ultrasound: a marriage of art and innovation K. R. Thomson; Radiology, The Alfred Hospital, Prahran, Australia. Since the advent of balloon catheter angioplasty by Andreas Gruentzig in 1978 1, the pace of technological innovation in revascularization techniques has been very rapid. This has resulted in tremendous improvements and tangible patient benefits. But there are still major opportunities to reduce restenosis rates, gain access to smaller and more tortuous vessels and demonstrate better results in removing obstructive material. The presence of recent or partly organised thrombus presents a unique challenge to the angioplasty procedure. The precise regulation of interactions between the vessel wall, blood constituents and the blood flow limits thrombosis under normal conditions but haemorrhage, thrombosis and embolism are common manifestations of many diseases. Angioplasty and its accompanying injury to the vessel wall causes such a disturbance to the three components of Virchow’s triad it is a wonder that it works at all. The risk of acute reclosure of the vessel or embolisation of the obstructing material has caused the development of numerous devices and methods for the bulk removal of thrombus from the circulation. The innovative genius of man is amply demonstrated in these devices but the very number of them is probably a negative indicator of their effectiveness. Recent developments in the understanding of mechanism of haemostasis has lead to an ability to control almost every aspect of the coagulation pathway and to reverse coagulation once it begins to develop. These pharmacologic methods however are unable to fragment crosslinked Fibrin polymer or organised thrombus. In the arteries, transluminal angioplasty has displaced surgical removal or bypass of chronic obstructions and thrombolytic drugs have revolutionised the treatment of acute arterial thrombosis. However most of the devices for mechanical removal of thrombus and arteriosclerotic

plaque are still just adjuncts to the angioplasty process as the lumen they provide is insufficient to restore normal flow on their own. In the venous system surgical thrombectomy has been shown to be superior to interventional techniques by meta-analysis in prosthetic dialysis grafts 2 but surgery has proved less effective for acute thrombosis of the ilio-femoral system. The degree of morbidity after deep venous thrombosis is related to the use of conservative treatment with drugs given not to remove thrombus but to inhibit further thrombus formation. A randomised controlled study showed a clear difference in the rate of thrombolysis in acute deep venous thrombosis with t-PA compared to patients treated only with Heparin 3.In this study of 64 patients it was notable that almost half of the patients receiving t-PA did not show more than 50% clearance of the thrombus burden. Angioplasty in venous thrombosis is only effective when an underlying stenosis is unmasked after thrombus clearance by other means and the presence of valves makes retrograde catheterisation difficult. Yet venous thrombosis is a common medical problem with a high community cost in terms of individual patient morbidity from post thrombotic syndrome and pulmonary embolism. As a person who lives in a remote continent, I have a keen interest in the risk of deep venous thrombosis inducing during long distance air travel. Post thrombotic syndrome represents a continuum of disease ranging from minor leg swelling and heaviness to haemosiderosis, non-healing ulceration and severe scarring. This condition is probably under-recognised but has a profound impact on the patient’s enjoyment of life. It is my observation there is a huge gulf between the doctor’s and the patient’s perception of the disability the patient suffers. If we are to make an impact on these arterial and venous disease processes we need a rapid and easy method of bulk removal of the offending material. Early mechanical devices used alone or in combination with thrombolytic drugs show promise but there are no randomised trials to prove this. In such circumstances, health care payors are reluctant to support increased expenditure for treatment. The ideal method of clearing an occluded vein would ensure that not only was the thrombus completely removed but also that the venous valves were not damaged in the process. There is evidence that if this can be achieved that the post-thrombotic syndrome will not occur 4.The major attraction of mechanical devices for thrombectomy is that they are not subject to the restrictive contraindications for the use of thrombolytic drugs. The current mechanical devices have a mode of action more akin to those required by a chef than an interventional radiologist. These devices use whisks, beaters, brushes, suction or a fluid jet to disrupt the thrombus. Most require very soft fresh thrombus to be effective and they only affect a small diameter of the vessel. While some have a risk of fluid overload many still do not protect the patient from significant emboli. They are generally all used in an adjunctive way with either thrombolytic drugs or subsequent angioplasty. As a method of preventing pulmonary thrombo-embolism, inferior vena cava filters have been increasingly used, particularly after major trauma. In younger patients, or those with a very limited period of thrombotic potential, a temporary or removable inferior vena cava filter has merit. After the reports of the Dublin experience 5 the Gunther retrievable filter has rapidly become the filter of first choice, at least in Australia, and we have safely removed these filters at intervals up to several months without difficulty or complication. The concern with longstanding vena cava filters is significant rate of vena cava occlusion ranging from 0-28% and what to do if there is a large amount of partly organised thrombus adherent to a filter. Some of this wide variation in occlusion rates is due to small case series and inconsistent post placement screening 6. Ultrasonic angioplasty was developed in the late 1960’s and two US patents were issued for clot lysis and for removing cholesterol and other deposits from blood vessels 7,8. Rosenchein 9 investigated the use of ultrasound to create a disruptive force for removal of occlusive material and demonstrated that high-energy (20+/- 10 W at 20kHz) ultrasound would disrupt atherosclerotic plaque and thrombus in vitro. The debris produced by passage of the probe wire through an occluded segment varied in size but was uniformly smaller when the probe wire was energised with ultrasound than when it was not. However the limitations of the device included the need to have a solid metal wire with a small amount of flexibility to prevent loss of energy. Another problem to be overcome was heat production and disruption of the red blood cells. The design of the Rosenchein device produced ultrasound energy only at the tip and a significant amount of energy was lost in transmission. The theoretical advantage of ultrasound is that the energy produced is dis-

sipated within flexible material, which provides a degree of protection for normal elastic tissues while disrupting rigid plaque or thrombus. Ultrasound energy applied above the inertial cavitation threshold produces potential spaces in the blood. These spaces then collapse and produce shear forces in their vicinity that exceed the tensile strength of nearby structures such as the strands of fibrin that compose a blood clot. As a result, depending on the power and the frequency it is possible to have a profound effect on thrombus several millimetres from the probe without affecting the normal vessel wall. More recently others 10 have used low power (0.15 – 1.2W at 20 kHz) ultrasonic energy to enhance the efficiency of thrombolytic agents and shown that the rate of clot dissolution is faster and the process more complete than with thrombolytics alone. At the power levels used, ultrasound does not affect the drug itself but produces more surface area of thrombus and thus a higher interaction of thrombus and lytic drug. A similar concept is used in the EKOS MicroLysUS infusion catheter 11 but this operates below the cavitation threshold and is designed to augment acute thrombolytic infusions. In October 2002 we used a new type of ultrasonic wire (Omnisonics Resolution 360) under an approved human research protocol for recanalisation of a chronic arterial occlusion. This wire, unlike its predecessors, releases energy along a 20cm length of the probe, instead of just at the tip or from an embedded piezo-electric crystal. The field of ultrasonic energy produced extends up to a 6mm circle from an 0.015inch wire. Unlike earlier ultrasound devices the acoustic energy is transmitted to the wire in a transverse format. This format provides higher efficiencies with reduced losses and cavitation is achieved at 1.8 W at 20kHz 12. The lower power causes insignificant heating and almost no haemolysis even after operation for several minutes. In vivo particle size was less than 10 microns. The first patient had a 45 cm long femoro-popliteal artery occlusion of several years standing and was recanalised with the Resolution wire and subsequent angioplasty. After this successful patient 15 others were treated with the Resolution wire. Twenty-five occluded arterial segments were treated. The average length of occlusion was > 25cm and average treatment time was 6 minutes. 19 occluded arterial segments were successfully crossed with the Resolution wire resulting in a lumen of 1-6mm. Adjunctive angioplasty was required in all patients and in two patients arterial stents were required to maintain immediate vessel patency. There was no evidence of haemolysis related to the use of the Resolution wire. In some patients with pre-existing stents there was no interaction between the stent and the Resolution wire. In chronic occlusions the recanalisation was rapid with less force than is necessary with alternative methods. The safety profile was excellent with minimal rates of complications such as perforation and emboli. The overall procedural outcomes even in this difficult group of long and chronically occluded native arteries were equivalent to angioplasty in less complex and shorter lesions. At the time of preparation of this abstract we are about to begin a further cohort with an improved ultrasound generator and some changes to the configuration of the probe itself. These have been designed to improve performance. Studies in chronic total occlusions in human coronary arteries are also about to commence at our institution. If the promise of this ultrasound technology is kept we shall have a method of bulk removal of not only fresh but old, organised thrombus. The need for adjunctive angioplasty will be removed and the degree of wall injury might be reduced to the extent that restenosis becomes less of a problem. Although this technology is being applied to chronic and acute arterial occlusions, there is an equally large body of patients with venous thrombosis. A review by Sharafuddin et al 13 showed that chronic venous insufficiency is a major cause of disability in Western countries afflicting as many as two-thirds of all patients in whom deep venous thrombosis develops. Preliminary studies at our institution show that the Resolution wire is likely to be just as effective in chronic venous occlusions as chronic arterial occlusion. The small diameter of the wire allows for treatment of multiple veins at the same sitting. As the Resolution wire is very flexible it does not damage the venous valves in animal studies even at higher energy levels. Once the valves have been destroyed in a chronic occlusion, provided the lumen can be re-established by a device such as the Resolution wire, then venous valvular function could be restored by percutaneous insertion of a prosthetic valve 14. We are also investigating the possibility of augmenting venous flow during the healing period after recanali-

sation using an implantable but removable device. There is much work to do and thanks to our industry partners we have the tools to do it. The legacy that Andreas Gruentzig and others such as Charles Dotter and have left us is a rich and varied one. We can honour them by our continued endeavours to improve our patient’s outcomes in chronic and acute vascular occlusion. References: 1. Gruentzig AR. Transluminal dilatation of coronary artery stenosis. Lancet 1978;1:263 2. Green LD, Lee DS, Kucey DS. A metaanalysis comparing surgical thrombectomy, mechanical thrombectomy and Pharmacomechanical thrombolysis for thrombosed dialysis grafts. J Vasc Surg 2002; 36:93945 3. Goldhaber SZ, Meyerovitz MF, Green D, Vogelzang RL, Citrin P, Heit J, Sobel M, Wheeler HB, Plante D, Kim H, et al. Randomized controlled trial of tissue plasminogen activator in proximal deep venous thrombosis. AM J Med 1990; 88:235-240 4. Eichlisberger R, Frauchiger B, Widmer MT, Widmer LK, Jager K. Late sequelae of deep venous thrombosis; a 13 year follow up of 223 patients. Vasa 1994; 23:234-243 5. Lees MJ. Personal Communication 6. Kinney TB. Update on inferior vena cava filters. J vas Interv Radiol 2003; 14:425-440 7. Delaney LJ: Method for clot lysis. US Patent No. 3,352,303 July 28 1968 8. Kuris A: Ultrasonic method and apparatus for removing cholesterol and other deposits from blood vessels and the like. US Patent No. 3.352,303 June 3 1965 9. Rosenchein U, Bernstein JJ, DiSegni E, Kaplinsky E, Bernheim, J, Rozenszajn LA. Experimental ultrasonic angioplasty: Disruption of atherosclerotic plaques and thrombi in vitro and arterial recanalisation in vivo. J Am Coll Cardiol 1990; 15:711-717 10. Nedelmann M, Eicke BM, Lierke EG, Heimann A, Kempski O, Hopf HC. Low frequency ultrasound induces nonenzymatic thrombolysis in vivo. J Ultrasound Med 2002; 21:649-656 11. Mahon BR, Nesbit GM, Barnwell SL, Clark W, Marotta TR, Weill A, Teal PA, Qureshi AI. North American Clinical Experience with the EKOS MicroLysUS Infusion Catheter for the Treatment of Embolic Stroke. AJNR Am J Neuroradiol 2003 Mar;24:534-538 12. Polak JF, Bloch S, Rabiner R, Marciante R, Hare B. Cavitation effect as a mechanism of thrombus dissolution by transverse application of ultrasonic energy to a wire. CVIR 2001; 24 (Suppl.1):S154 13.Sharafuddin MJ, Sun S, Hoballah JJ, Youness FM, Sharp WJ, Roh BS. Endovascular management of venous thrombotic and occlusive diseases of the lower extremities. J Vasc Interv Radiol 2003; 14:405-423 14. Pavcnik D, Uchida BT, Timmermans HA, Corless CL, O'Hara M, Toyota N, Moneta GL, Keller FS, Rosch J. Percutaneous bioprosthetic venous valve: a long-term study in sheep. J Vasc Surg 2002 ;3:598-602

Special Session Uterine Fibroid Embolization 25.1 UFE Imaging and Embolization Technique I. Pinto Pabón; Hospital de Madrid-Montepríncipe, Madrid, Spain. Uterine artery embolization was introduced in the seventies to treat postpartum hemorrhage. In the 1990s, this technique was used before myomectomy to reduce bleeding during the surgical phase. In 1995, Ravina proposed embolization of uterine arteries as an alternative to surgical treatment of uterine fibroid. Several papers on the subject have appeared since then. In the management of the patients, before the procedure, all patients require a gynaecologic history and physical examination to exclude other pathologies, and the gynecologist can provide an overview of all therapies for fibroids. We always obtain, a laboratory test, a biopsy of the endometrium associated with a hysteroscopy in case of doubt and cross-sectional images with MRI or ultrasound, to evaluate uterus and fibroid size, type and number. Magnetic Resonance is more sensitive and accurate than US in evaluating the endometrium, exclude adenomyosis or any other pathologies; and we can obtain also an angioresonance sequence to visualize the size of the uterine arteries and chose the biggest to begin.

PROCEDURE: How to do UAE? Inmediatly before the procedure, an intravenous line and Foley catheter are placed in all patients. Not all authors use prophylactic antibiotics, we always use cefazoline 2 gr. IV. During the procedure adequate sedation and analgesia is achieved with midazolan and fentanyl, in our practice. Concerning the approach the unilateral right common femoral artery is preferred for access but some authors make a bilateral puncture to reduce the time of intervention and radiation. We begin with a Sidewinder catheter 4-5 F to catheterize the internal iliac artery and further angiogram obtained to see the uterine artery. After that, selective uterine artery catheterization is performed, with catheter tips placed within the transverse portion of the uterine artery. The intramural branches (also called arcuate arteries) are visible, and the embolization can begin. We can recognize the uterine arteries, because they are highly twisted; and the characteristic course is a parietal segment, an arch part and a marginal or ascending segment. The embolization is performed with particles bigger than 300 µ (PVA particles or microspheres). Which endpoint is the best?: the end point for embolization with PVA described in the literature is stasis or near-stasis of uterine flow. The mean PVA dose is quite variable between 100 to 500 mg. It is more difficult to determine the embolization endpoint when using microspheres and the identifiable markers to indicate that the embolization is sufficient are: the curvilinear enlarged vessels that surround the fibroid need to reach stasis or near-stasis, and this usually coincides with sluggish forward flow remaining in the main uterine artery. Secondary signs of completion include easy reflux into the ovarian arteries when it had not been present earlier and filling of cross-uterine branches without filling of intervening fibroid tissue. The key points are complete disappearance of fibroid vascularity and slow flow in the ascending uterine artery. False endpoint may occur due to clumping of the embolic material or due to spasm. The technical success rate for bilateral embolization of both uterine arteries is between 84 - 98%. The need to employ microcatheter to complete the procedure varies from 5% to 90% in the literature, and it´s used when the uterine arteries are very small to avoid the spasm. The procedure time range between 45´ and more than two hours. In the patient management, after the procedure, the most important consideration is to control the pain. The degree of pain is quite variable. Some may experience only mild cramping and other patients have pain that is comparable to that of childbirth. Most practitioners offer PCA pumps to deliver intravenous morphine for the first 12 to 24 hours. We don’t use it because the problem with vomiting and nauseas, that we observed in the first patients. Other physicians have advocated the use of epidural analgesia, but it is not clear at this time, that these convey any advantage. Currently, we use nonsteroidal anti-inflammatory agents such as ketorolac 30 mg IV alternating every 6 hours with propacetamol 2 g IV during 24 hours. After discharge (24h usually) the patient is switched to oral analgesics and anti-inflammatories. References: 1. Ravina JH, Merland JJ, Herbreteau D, Houdart E, Bouret JM, Madelenat P. Embolization pré-opératoire des fibromes utérins. Presse Med 1994;23:1540. 2. Ravina JH, Herbreteau D, Ciaru-Vigneron N, et al. Arterial embolization to treat uterine myomata. Lancet 1995;346:671-672. 3. Goodwin SC, Vedantham S, McLucas B, Forno AE, Perrella R. Preliminary experience with uterine artery embolization for uterine fibroids. JVIR 1997;8:517-526. 4. Ellis PK, Kidney DD, Deutsch LS. Uterine arterial embolization in the management of symptomatic leiomyomata. J Intervent Radiol 1998;13:16 –19. 5. Ravina JH, Aymard A, Ciraru-Vigneron N, Bouret JM, Houdart E, Le Dref O. Embolisation arterielle particulaire: un nouveau traitement des hemorragies des leiomyomes uterins. Press Méd 1998;27:299-303. 6. Worthington-Kirsch RL, Popky GL, Hutchins Jr FL. Uterine arterial embolization for the management of leiomyomas: quality -of- life assessment and clinical response. Radiology 1998;208:25-29. 7. Bradley EA, Reidy JF, Forman RG, et al. Transcatheter uterine artery embolization to treat large uterine fibroids. Br J Obstet Gynaecol 1998;105:235-240. 8. Spies JB, Scialli AR, Jha RC, et. Initial results from uterine fibroid embolization for symptomatic leiomyomata. JVIR 1999;10:1149-1157. 9. Siskin GP, Eaton LA Jr, Stainken BF, Dowling K, Herr A, Schwartz J. Pathologic findings in a uterine leiomyoma after bilateral uterine artery

embolization. JVIR 1999;10: 891-894. 10. Goodwin SC, McLucas B, Lee M, et al. Uterine artery embolization for the treatment of uterine leiomyomata: midterm results. JVIR 1999;10:1159-1165. 11. Burn P, Mccall J, Chinn R, Healy J. Embolization of uterine fibroids. Br J Radiol 1999;72:159-161. 12. Hutchins FL Jr, Worthington-Kirsch R, Berkowitz RP. Selective uterine artery embolization as primary treatment for symptomatic leiomyomata uteri. J Am Assoc Gynecol Laparosc 1999;6:279- 284. 13. Brunereau L, Herbreteau D, Gallas S, et al. Uterine artery embolization in the primary treatment of uterine leiomyomas: technical features and prospective follow-up with clinical and sonographic examinations in 58 patients. AJR 2000;175:1267-1272. 14. Siskin GP, Stainken BF, Dowling K, Meo P, Ahn J, Dolen EG. Outpatient uterine artery embolization for symptomatic uterine fibroids: experience in 49 patients. JVIR 2000;11:305-311. 15. Pelage JP, Le Dref O, Soyer P, et al. Fibroid-related menorrhagia: treatment with superselective embolization of the uterine arteries and midterm follow-up. Radiology 2000; 215:428-431. 16. Pelage JP, Le Dref O, Mateo J, et al. Life-threatening primary postpartum hemorrhage: treatment with emergency selective arterial embolization. Radiology 1998;208:359-362. 17. Braude P, Reidy J, Nott V, Taylor A, Forman R. Embolization of uterine leiomyomata: current conceps in management . Hum Reprod Update 2000;6: 603-608. 18. Pelage JP, Le Dref O, Soyer P, Jacob D, Kardache M, Dahan H, Lassau JP, Rymer R. Arterial anatomy of the female genital tract: variations and relevance to transcatheter embolization of the uterus. Am J Roentgenol 1999; 172 (4): 989-94. 19. Spies JB, Roth AR, Jha RC, Gomez-Jorge J, Levy EB, Chang TC, Ascher SA. Leiomyomata treated with uterine artery embolization: factors associated with successful symptom and imaging outcome. Radiology 2002; 222 (1) :45-52. 20. Roth AR, Spies JB, Walsh SM, Wood BJ, Gomez-Jorge J, Levy EB. Pain after uterine artery embolization for leiomyomata: Can its severity be predicted and does severity predict outcome? JVIR. 2000; 11 (8):1047-52. 21. Pelage JP, Laurent A, Wassef M, Bonneau M, Germain D, Rymer R, Flaud P, Martal J, Merland JJ. Uterine artery embolization in sheep: comparison of acute effects with polyvinyl alcohol particles and calibrated microspheres. Radiology 2002; 224 (2):436-45. 22. Spies JB, Warren EH, Mathias SD, Walsh SM, Roth AR, Pentecost MJ. Uterine fibroid embolization: measurement of health-related quality of life before and after therapy. JVIR 1999;10 (10):1293-303. 23. Pinto I, Chimeno P, Romo A, Paul L, Haya J, de la Cal MA, Bajo J. Uterine fibroids: uterine artery embolization versus abdominal hysterectomy for treatment--a prospective, randomized, and controlled clinical trial. Radiology. 2003; 226 (2):425-31. 24. Spies JB. Uterine Artery Embolization for Fibroids: Understanding the Technical Causes of Failure. JVIR, 2003;14:11-14. 25. Pelage JP, Le Dref O, Beregi JP et al. Limited Uterine Artery Embolization with Tris-acryl Gelatin Microspheres for Uterine Fibroids. JVIR, 2003; 14:15-20.

25.2 Uterine Fibroid Embolization: Indications and Results J. Pelage; Hôpital Ambroise Paré, University Paris-Ouest,, Boulogne-Billancourt Cedex, France. Introduction Leiomyomata, commonly known as fibroids, are benign neoplasms derived from the smooth muscle of the uterus. They are among the commonest lesions of the female genital tract and occur in 20-25% of women of reproductive age and are more common in women of AfroCaribbean origin than in white women [1]. The most common presenting symptom is menorrhagia often leading to anemia. In other cases fibroids may grow to a large size and compress adjacent organs causing pelvic pain or pressure, constipation, bloating, backache and urinary frequency [1]. Because of poor results associated with hormone suppressant therapy, conventional treatment of fibroids has been surgical i.e. hysterectomy or myomectomy [1,2]. In recent years the surgical armamentarium has been expanded with further procedures such as laparoscopic myomectomy and hysteroscopic resection of sub-mucous myomas [3,4]. The major drawback associated with myomectomy is the risk of recurrence

requiring further surgery which has been estimated as between 20 and 50% at 5 years [5]. It is important that any radiologist embarking on fibroid embolization has a good working knowledge of surgical procedures for fibroids, their efficacy and their complications. Although embolization has been increasingly employed by radiologists since the 1970 for many differing conditions, fibroid embolization is comparatively new, despite fibroids being a logical target for embolization [6]. The first report of uterine fibroid embolization (UFE) was only published in the Lancet in 1995 consisting of a group of patients treated at Lariboisiere Hospital in Paris, France [7]. Indications

ischemia [8-11]. The overall incidence of ovarian failure is approximately 2-15% but is age dependant [9,11,25]. Patients under the age of 45 have a low risk of ovarian failure (<2%) [11,25]. Premature menopause has also been reported after hysterectomy with conservation of the ovaries [26]. Ovarian ischaemia caused by uterine artery ligation or embolization is the supposed mechanism and is more likely to be observed in women over the age of 45 with already compromised ovarian reserve. In a randomized trial comparing embolization and hysterectomy, the rate of major postprocedural complications was higher in the hysterectomy group (35 vs 2%) [20]. Conclusion and perspectives

Before UFE is carried out it is particularly important that the radiologist and gynaecologist are confident in the diagnosis and that UFE is indicated. Typical inclusion criteria for patients undergoing UFE include abnormal menstrual bleeding, pelvic pain and bulk-related symptoms attributable to fibroids [8-12]. To determine if heavy menstrual bleeding is present, a standard menstrual history is obtained. In case of interperiod bleeding or if the bleeding is more frequent than 21 days, it is recommanded to exclude other cause of bleeding by performing endometrial biopsy. Preprocedure imaging should provide precise information on the number, type and size of fibroids with special attention paid to subserous and submucous fibroids [13,14]. Magnetic Resonance Imaging (MRI) is particularly useful as fibroids can be more accurately imaged than with ultrasound and adenomyosis more consistently excluded [13]. It is important to correlate the fibroid’s location with the presenting symptoms of the patient. During the follow-up, MRI is also more accurate in the assessment of fibroid shrinkage [14]. UFE should be offered as an alternative to hysterectomy and multiple myomectomies. Currently, UFE is not offered as an alternative to a single myomectomy by most groups. The published evidence on fertility after embolization is still scanty whereas the literature on pregnancy following myomectomy is extensive. However, there are some clear indications for UFE in patients with multiple fibroids, those who have undergone failed myomectomy or in whom the only treatment on offer is hysterectomy. Results and complications

Although much has already been published on UFE there are still some unanswered questions. The precise indications for UFE and the impact on fertility are still controversial. Nevertheless a sufficient number of embolizations have been carried out and the results published to date strongly indicate that this is a viable alternative to hysterectomy and multiple myomectomy for symptomatic women. References: 1. Buttram VC, Reiter RC. Uterine leiomyomata: etiology, symptomatology and management. Fertil Steril 1981; 36: 433-445. 2. Healy DL, Vollenhoven BJ. The role of GnRH agonists in the treatment of uterine 2. fibroids. Br J Obstet Gynaecol 1992; 99: 23-26. 3. Mais V, Ajossa S, Guerriero S, Mascia M, Solla E, Mellis GB. Laparoscopic versus abdominal myomectomy: a prospective randomized trial to evaluate benefits in early outcome. Obstet Gynecol 1996; 174: 654658. 4. Vercellini P, Zaina B, Yaylayan L, Pisacreta A, De Giorgi O, Crosignani PG. Hysteroscopic myomectomy: long-term effects on menstrual pattern and fertility. Obstet Gynecol 1999; 94: 341-347. 5. Fedele L, Parazzini F, Luchini L, Mezzopane R, Tozzi L, Villa L. Recurrence of fibroids after myomectomy: a transvaginal ultrasonographic study. Hum Reprod 1995: 10: 1795-1796. 6. Greenwood LH, Glickman MG, Schwartz PE, Morse S, Denny DF. Obstetric and nonmalignant gynecologic bleeding: treatment with angiographic embolization. Radiology 1987; 164: 155-159. 7. Ravina JH, Herbreteau D, Ciraru-Vigneron N, et al. Arterial embolization to treat uterine myomata. Lancet 1995; 346: 671-672. 8. Goodwin SC, Mc Lucas B, Lee M, et al. Uterine artery embolization for the treatment of uterine leiomyomata: mid-term results. JVIR 1999; 10:1159-1165. 9. Pelage JP, Le Dref O, Soyer P, et al. Fibroid-related menorrhagia: treatment with superselective embolization of the uterine arteries and mid-term follow-up. Radiology 2000; 215: 428-431. 10. Spies JB, Ascher SA, Roth AR, Kim J, Levy EB, Gomez-Jorge J. Uterine artery embolization for leiomyomata. Obstet Gynecol 2001; 98: 29-34. 11.Walker WJ, Pelage JP. Uterine artery embolisation for symptomatic fibroids: clinical results in 400 women with imaging follow up. BJOG 2002; 109: 1262-72. 12. Pron G, Bennett J, Common A, et al. The Ontario uterine fibroid embolization trial: Uterine fibroid reduction and symptom relief after uterine artery embolization for fibroids. Fertil Steril 2003; 79: 120-127. 13. Togashi K, Ozasa H, Konishi I et al. Enlarged uterus: differentiation between adenomyosis and leiomyoma with MR imaging. Radiology 1991; 180: 81-83. 14. Jha RC, Ascher SM, Imaoka I, Spies JB. Symptomatic fibroleiomyoma: MR imaging of the uterus before and after uterine arterial embolization. Radiology 2000; 217: 22-235. 15. Spies JB, Benenati JE, Worthington-Kirsch RL, Pelage JP. Initial US experience using trisacryl gelatin microspheres for uterine artery embolization for leiomyomata. JVIR 2001; 12: 1059-1063. 16. Spies JB, Warren EH, Mathias SD, Walsh SM, Roth AR, Pentecost MJ. Uterine fibroid embolization: measurement of health-related qualityof-life before and after therapy. JVIR 1999; 10: 1293-1303. 17. Smith SJ, Sewall LE, Handelsman A. A clinical failure of uterine fibroid embolization due to denomyosis. JVIR 1999; 10: 1171-1174. 18. Siskin GP, Tublin ME, Stainken BF, Dowling K, Dolen EG. Uterine artery embolization for the treatment of adenomyosis: clinical response and evaluation with MR imaging. AJR 2001; 177: 297-302. 19. Subramanian S, Spies JB. Uterine artery embolization for leiomyomata: resource use and cost estimation. JVIR 2001; 12: 571-574. 20. Pinto I, Chimeno P, Romo A, et al. Uterine fibroids: uterine artery embolization versus abdominal hysterectomy for treatment. Radiology 226: 425-431.

Criteria for success are not universally accepted but patient satisfaction and resolution of symptoms seem to be essential. Results of UFE are generally similar with a success rate of over 85% for control of menorrhagia or bulk-related symptoms [7-12]. Validated bleeding scores have recently been used to quantify precisely menstrual bleeding before and after embolization [15]. Quality-of-life improvement has also been reported after embolization [16]. Overall, approximately, 90% of patients are symptomatically improved and 10% are not. Among the causes of failures, there are obvious technical failures (i.e. unilateral emboilization or spasm) or anatomical failures (secondary source of blood supply from the ovarian arteries, absence of one uterine artery). Finally, associated adenomyosis is often considered as a cause of failure of UFE despite initial symptom relief [17,18]. The majority of shrinkage of the fibroids occurs within a six month period but some further reduction in size occurs between six and twelve months [11]. The volume reduction of the dominant fibroid is larger than uterine volume reduction and varies between 43 to 50% [8-12] for the fibroid and 29 to 48% [7-12] for the whole uterus. From an economical point of view, embolization is a valuable alternative to hysterectomy and myomectomy mainly because of a shorter hospital stay and faster recovery [19,20]. Even if UFE is a safe and well-tolerated procedure, serious complications such as pelvic infections have led to hysterectomy [8,9,11,21]. In a survey of 10,501 UFE procedures carried out world-wide, 40 major complications within 30 days of embolization requiring gynaecological surgery and one death were reported in 1999 [22]. It has been stated that infective complications are more likely with sub-mucus fibroids where exposure to intra-cavitary pathogens may occur or with pedunculated subserosal fibroids [9,11]. In some cases fibroids may be passed vaginally and it is particularly important that if infection occur the need for hysteroscopic resection is recognized where appropriate [9-11]. Four deaths have occurred following UFE, two from pulmonary emboli and two due to infection in approximately 40,000 cases (0.10:1,000) [21,23]. It should be remembered that the mortality rate for hysterectomy for benign disease excluding complications of pregnancy is 0.62:1,000 [24]. A further documented complication of fibroid embolization is ovarian failure. This may be related to temporary ovarian or endometrial

21. Vashisht A, Studd J, Carey A, et al. Fatal septicaemia after fibroid embolization. Lancet 1999; 354: 307-308. 22. Anonymous. Uterine Artery Embolization survey results: 10,500 procedures performed world-wide. Society of CardioVascular Interventional radiology, October 2000 23. Lanocita R, Frigerio LF, Patelli G, Di Tolla G, Spreafico C. A fatal complication of percutaneous transcatheter embolization for treatment of uterine fibroids. SMIT/CIMIT. 11th Annual Scientific Meeting. Boston, Mas, USA, 16-18 Sept 1999. 24.Wingo PA, Huezo CM, Rubin GL, et al. The mortality risk associated with hysterectomy. Am J Obstet Gynecol 1986; 152: 803-808. 25.Chrisman HB, Saker MB, Ryu RK, et al. The impact of uterine fibroid embolization on resumption of menses and ovarian function. JVIR 2000; 11:699-703. 26. Siddle N, Sarrel P, Whitehead M. The effect of hysterectomy on the age at ovarian failure: identification of a subgroup of women with premature loss of ovarian function and literature review. Fertil Steril 1987; 47: 94-100.

Special Session Venous Access 26.2 Venous access: Technique of insertion B. E. Çil; Department of Radiology, Hacettepe University School of Medicine, Ankara, Turkey. Central venous catheters include 4 major groups: temporary catheters, peripherally inserted central catheters (PICC), tunneled catheters and implantable port devices. Routine preplacement screening blood work includes prothrombine time and international normalized ratio (INR), platelet count, blood urea nitrogen and creatinine. INR of 1.3 or less and platelet count of 50,000/ dL or higher are preferrable for tunneled catheter and port insertion to avoid bleeding complications and hematomas, which increase the risk for infection. With strict adherence to imaging-guided techniques, temporary catheters and PICCs can be placed in patients with uncorrectable coagulopathy or thrombocytopenia. Conscious sedation is used for the placement of tunneled catheters and ports, which requires that the patient not eat for 6 hours prior to the procedure. The operator should wear a cap and mask and follow meticulous sterile technique. A 5-minute full surgical scrub is required prior to tunneled catheter and port insertion. Skin at the insertion site is widely prepared from mandible to the nipple craniocaudally, and from the sternum to the mid-axillary line laterally for tunneled catheter and port placement. Traditionally, central venous catheters were placed either by cut-down or percutaneous access using landmark technique. In addition to the risks of arterial puncture and pneumothorax inherent with this blind insertion, in subclavian vein (SCV) approach, catheter entry between the first rib and clavicle can result in the pinch-off syndrome, which is due to catheter entrapment between the costoclavicular ligament and the subclavius muscle and subsequently may result in catheter fracture and embolization (1). Complication rates associated with blind percutaneous puncture range from 1 to 7%, and include pneumothorax, arterial puncture, nerve injury, arteriovenous fistula and catheter malposition (2,3). More importantly, the higher incidence of subclavian vein stenosis and thrombosis compared with the internal jugular vein (IJV) access has been shown in dialysis and cancer patients (4,5). Allowing for these considerations, the most commonly used venous access route today is the IJV. Imaging guidance has been shown to be superior to the blind technique and is characterized by a 0 to 2% overall complication rate (2,3,6,7). The right IJV approach is preferred over the left for practical reasons. The right IJV provides a straighter course to the right atrium; the left IJV route requires traversing the relatively acute angles at the left IJV-left innominate and left innominate-SVC confluences. The resulting anatomic configuration on the left is more prone to catheter malposition. Some studies have demonstrated that catheter patency rates are lower using the left IJV approach, but this might be due in part to the greater propensity for malposition on the left side(8,9). Therefore, we resort to the left IJV access if the right is occluded. Alternatives to the jugular and subclavian approaches often need to be considered when these routes have become thrombosed, as in patients

requiring chronic central venous access. Among the alternatives are the femoral, translumbar, transhepatic, transrenal, and azygous routes. For the IJV access, the patient is placed on the table with his or her head near the top of the table. This allows the operater to be comfortably positioned at the head of the table with the ultrasonography unit placed facing the operator; this positioning makes it easy for the operator to easily perform the puncture parallel to the course of the IJV. The operator should perform ultrasonographic examination of the neck veins prior to draping the patient. Once the IJV patency is confirmed and the neck and chest are prepared and draped, the venipucture site is anesthetized with 1% lidocaine. For the initial venous entry, 21-gauge micropuncture needle is preferred to minimize the puncture site complications. A mid-neck access is preferred for temporary catheters and a low IJV pucture is preferred for tunneled catheters and ports. The needle tip should be followed by the ultrasound probe, from the cutaneous incision to the venous entry. Once the needle has entered the vein, a 0.018-inch guidewire is advanced through the needle and its tip is positioned in the right atrium. Then, 4F coaxial dilator of the micropuncture set is inserted, and the tip of the wire is positioned in the proximal right atrium, and its trailing segment is bent at the dilator hub. The 0.018-inch guidewire and the inner dilator are removed and replaced with a 0.035-inch J-shaped guidewire from the catheter kit. A flow switch is advanced over the wire, Luerlocked to the 4F dilator and locked on the wire. Then the subcutaneous tunnel is created for tunneled catheters or alternatively subcutaneous pocket is created for the ports. Tunneled catheters require subcutaneous tunnel which can be created either laterally or medially. Medial parasternal tunneling minimizes upward displacement of the catheter tip in obese patients or in women with large breasts. Lateral skin tunnel provides a more gently curved course that may be less prone to catheter kinking at the venotomy site. The decision should be made on a patient-by-patient basis. The micropuncture needle can be used to anesthetize the anticipated subcutaneous tunnel tract. A skin incision large enough to accommodate the catheter and its Dacron cuff is made at the skin exit site. The tunneling device from the catheter kit is then advanced from the skin exit site to the cervical venotomy site. With the tunneler in place, the catheter is attached to its trailing end, and both are advanced through the tunnel. If it is a infusion catheter or a port catheter, the catheter is then trimmed to length using the previously bent 0.018-inch guidewire. Hemodialysis and apheresis catheters cannot be cut to length, so the tunnel length is adjusted to the catheter length. Then, under fluoroscopic guidance, the peel-away sheath is inserted over the previously placed guidewire. After the peel-away sheath is placed, the sheath is held with thumb and index finger of one hand while the dilator and the guidewire are removed with the other hand and the sheath is pinched to prevent air embolism. Then the catheter is advanced with the other hand as the pinch is released. After insertion, the catheter tip position and catheter curve at the venotomy site are evaluated under fluoroscopy. The catheter tip should be in the mid right atrium for hemodialysis or apheresis catheters and in the proximal right atrium for Hickman catheters and ports. The venotomy site is closed with a single stitch and the catheter is secured to the skin exit site with a nonabsorbable suture material. Each catheter port is flushed with heparinized saline solution. Implantable port devices require subcutaneous pocket creation. The port is placed on the skin of the chest to assess incision size and location. (For arm ports, basilic or cephalic veins are accessed under sonographic guidance and the pocket is created in the arm or forearm.) The skin and the subcutaneous tissues at the pocket site are infiltrated with lidocaine containing 1% epinephrine. An incision just large enough to accommodate the port reservoir is made with a no.15 blade. Once the incision is made, blunt dissection is used to create a small pocket under the skin. Care should be taken to keep at least 0.5-1 cm thickness of tissue between the port and the skin to prevent skin necrosis. Once the pocket is created, the port should be fitted inside to check for positioning and size. After creating the port pocket, the access catheter is tunneled from the port pocket to the venotomy site with a tunneling device. The catheter is then connected to the port reservoir and the reservoir is placed into the pocket. Two resorbable 3-0 vicryl stay sutures are placed into the medial and lateral aspects of the port base to prevent rotation of the port in the pocket. The catheter is then trimmed to length using the previously bent 0.018-inch guidewire, as in the case of Hickman catheters. The tip of the port catheter must be in the high right atrium.Then the catheter is inserted through the peel-away sheath as described above. Using a Huber needle, the port is accessed and function is confirmed with aspiration of blood. The reservoir is flushed

with 5-10 mL of 100U/mL heparin solution while carefully observing for any leakage at the connection site. The incision is closed in layers with resorbable 3-0 vicryl interrupted inverted mattress sutures subcutaneously, and the skin is closed with a running subcuticular stitch. Steristrips are also applied. The venotomy incision is also closed as described for tunneled catheters. PICCs are catheters inserted via a peripheral vein whose tip terminates in the cavo-atrial junction, like other central venous catheters. The preferred location for placement of a PICC is the upper arm. Generally, PICCs are inserted either at the bedside by a nurse or in the interventional radiology suite with imaging-guided techniques.Usually no ansthesia, analgesics, or sedation is required. Strict attention to sterile technique is imperative when placing PICCs. Although, basilic, cephalic, and brachial veins can be used for access, basilic vein is the preferred access vein. Venous access can be obtained by direct visualization/palpation, venographic or ultrasonographic guidance. Ultrasonographic guidance is the preferred imaging guided technique. A 7-10 mHz linear array transducer usually provides adequate vascular visualization. Ultrasonographic evaluation is a useful tool in the initial assessment of the patency of upper arm veins and in demonstrating the location of the artery. After the patients arm is prepared and draped, a tourniquet is placed around the upper arm and an appropriate vein is chosed for puncture.The intended PICC insertion site is anesthetized with lidocaine. Under ultrasound guidance, a 21G needle is introduced into the vein and a 0.018-inch guidewire is then advanced into the vein using fluoroscopic guidance. After a small cutaneous incision, a peel-away introducer sheath is advanced over the guidewire. The guidewire is advanced to the junction of superior vena cava/right atrium and the length of the guidewire is used to measure the appropriate length of the PICC. Once the catheter is trimmed to length, it is advanced through the peel-away sheath to the cavo-atrial junction and the peel-away sheath is removed. Then, the catheter is sutured or taped to the skin. References: 1. Hinke DH, Zandt-Stastny DA, Goodman LR, Quebbeman EJ, Krzywda EA, Andris DA. Pinch-off syndrome: a complication of implantable subclavian venous access devices. Radiology 1990; 177:353-356. 2. Lameris JS, Post PJM, Zonderland HM, Gerritsen PG, Kappers-Klunne MC, Schutte HE. Percutaneous placement of Hickman catheters: comparison of sonographically guided and blind techniques. Am J Roentgenol 1990; 155:1097-1099. 3. Mallory DL, McGee WT, Shawker TH, et al. Ultrasound guidance improves the success rate of internal jugular vein cannulation. Chest 1990; 98:157-160. 4. Macdonald S, Watt AJB, Edwards RD, Moss JG. Comparison of the internal jugular and subclavian venous routes in radiologically placed tunneled venous access lines. [Abstract] Cardiovasc Intervent Radiol 1998; 21 (suupl 1) :S81. 5. Fulton JK, Trerotola SO, Johnson MS, Shah H, Namyslowski J. Symptomatic venous thrombosis from tunneled infusion catheters: internal jugular versus subclavian approach. [Abstract] J Vasc Intervent Radiol 1990; 10 Pt 2 :234. 6. Denys BG, Uretsky BF, Reddy PS. Ultrasound-assisted cannulation of the internal jugular vein: a prospective comparison with the external landmark –guided technique. Circulation 1993; 87:1557-1562. 7. Farrell J, Gellens M. Ultrasound-guided cannulation versusthe landmark-guided technique for acute hemodialysis access. Nephrol Dial Transplant 1997; 12:1234-1237. 8. De Meester J, Vanholder R, Ringoir S. Factors affecting catheter and technique survival in permanent silicone single-lumen dialysis catheters [Abstract]. Nephrol Dial Transplant 1994; 9:678-683. 9. Puel V, Caudry M, Le Metayer P, et al. Superior vena cava thrombosis relative to catheter malposition in cancer chemotherapy given through implanted ports. Cancer 1993; 72:2248-2252.

Special Session Percutaneous Treatment of Liver Tumors II 27.1 New Trends in Chemoembolization of Liver Tumors. J. Llovlet1, M. Real2, X. Montana2, J. Bruit1; 1Liver Unit, Hospital Clinico, Barcelona, Spain, 2Radiology, Hospital Clinico, Barcelona, Spain. There is no standard treatment for unresectable hepatocellular carcinoma (HCC). Survival benefits derived from medical interventions are controversial, particularly regarding the impact of arterial embolization and chemoembolization in survival. We performed a 3-arm randomized controlled trial (RCT) in 112 patients with unresectable HCC which received embolization (37 patients), chemoembolization (gelfoam and lipiodol+ doxorubicin, 40 patients) or conservative treatment (35 patients). Chemoembolization improved survival in comparison to controls (2-year survival 63% vs 27%, p= 0.009), induced objective responses sustained for six months in 35% of cases, and significantly decreased the incidence of portal vein invasion. Treatment allocation was the sole variable independently related to survival (odds ratio:0.45, 95%CI: 0.25 to 0.81, p=0.02). Considering these positive results, we performed a systematic review of RCT published between 1978-2002 to assess the impact of arterial embolization/chemoembolization in survival in HCC patients. We identified 61 randomized trials, and 7 met the criteria to perform a metaanalysis assessing embolization (545 patients). Arterial embolization improved 2-year survival compared to control [OR, 0.53; 95% CI: 0.320.89, p=0.017]. Sensitivity analysis showed a significant benefit of chemoembolization with cisplatin or doxorubicin [OR, 0.42 (95%CI: 0.20-0.88)], but none with embolization alone [OR, 0.59 (95% CI: 0.291.20)]. Overall, treatment induced objective responses in 35% of patients (range 16-61%). In conclusion, chemoembolization improves survival of stringently selected patients with unresectable HCC, and may become the standard treatment. No References

27.2 Percutaneous ablation techniques in hepatocellular carcinoma R. Lencioni; Department of Radiology, University of Pisa, Pisa, Italy. INTRODUCTION There is no universal algorithm implemented worldwide for the treatment of hepatocellular carcinoma (HCC) in cirrhosis. If diagnosed at an early stage, patients should be considered for any of the available options that may provide a high rate of complete response. These include surgical resection, liver transplantation, and percutaneous techniques of tumor ablation (1). Indication to surgical resection is currently restricted to patients with single asymptomatic HCC and extremely well-preserved liver function, who have neither clinically significant portal hypertension nor abnormal bilirubin. Cadaveric liver transplantation is limited by the shortage of donors and living donor liver transplantation is still at an early stage of clinical application. As a result, percutaneous ablation plays a key role in the therapeutic management of HCC. While ethanol injection is a well-established technique for percutaneous ablation, several newer methods of tumor destruction have been developed and clinically tested over the past few years. Among these methods, radiofrequency (RF) thermal ablation constitutes the most extensively studied alternative to ethanol injection (2). ELIGIBILITY CRITERIA A careful clinical, laboratory, and imaging assessment has to be performed in each individual patient by a multidisciplinary team to evaluate eligibility for percutaneous ablation. Patients classified as stage A according to the Barcelona Clinic Liver Cancer staging classification, that are not candidates for surgery, qualify for percutaneous ablation (3). The tumor to treat by RF must be a focal, nodular-type lesion. The presence of a clear and easy-to-detect target for needle placement is crucial for the outcome of treatment. Tumor size should be preferentially smaller than 3-5 cm in greatest dimension. When using thermal

methods of tissue destruction, some additional points have to be considered. Treatment of lesions adjacent to the gallbladder or to the hepatic hilum is at risk of thermal injury of the biliary tract. Lesions located along the surface of the liver can be considered for thermal ablation, although their treatment requires experienced hands and may be associated with a higher risk of complications. A careful assessment of the coagulation status is mandatory before percutaneous ablation. A prothrombin time ratio (normal time / patient's time) greater than 50% as well as a platelet count higher than 50,000/µl are required to keep the risk of bleeding at an acceptable low level. ETHANOL INJECTION Percutaneous ethanol injection (PEI) is a well-established technique for tumor ablation (4) Several studies have shown that PEI is an effective treatment for small (3 cm or less), nodular-type HCC. Although there have not been any prospective randomized trials comparing PEI and surgical resection, several series have shown that the long-term outcome of selected PEI-treated patients was similar to that of patients who had undergone resection, with 5-year survival rates of 32-59% (59). The major limitation of PEI, besides the uncertainty of tumor ablation and the long treatment times, is the high local recurrence rate, that may reach 33% in lesions smaller than 3 cm and 43% in lesions exceeding 3 cm (10, 11). The injected ethanol does not always accomplish complete tumor necrosis because of its inhomogeneous distribution within the lesion - especially in presence of intratumoral septa - and the limited effect on extracapsular cancerous spread. Also, PEI is unable to create a safety margin of ablation in the liver parenchyma surrounding the nodule, where satellite nodules are most frequently located. RF THERMAL ABLATION The goal of RF ablation is to induce thermal injury to the tissue through electromagnetic energy deposition. In RF ablation, the patient is part of a closed-loop circuit, that includes a RF generator, an electrode needle, and a large dispersive electrode (ground pads). An alternating electric field is created within the tissue of the patient. Because of the relatively high electrical resistance of tissue in comparison with the metal electrodes, there is marked agitation of the ions present in the target tissue that surrounds the electrode, since the tissue ions attempt to follow the changes in direction of alternating electric current. The agitation results in frictional heat around the electrode. The discrepancy between the small surface area of the needle electrode and the large area of the ground pads causes the generated heat to be focused and concentrated around the needle electrode (12). The thermal damage caused by RF heating is dependent on both the tissue temperature achieved and the duration of heating. Heating of tissue at 55°C for 4–6 minutes produces irreversible cellular damage. At temperatures between 60°C and 100°C near immediate coagulation of tissue is induced, with irreversible damage to mitochondrial and cytosolic enzymes of the cells. At more than 100°–110°C, tissue vaporizes and carbonizes. For adequate destruction of tumor tissue, the entire target volume must be subjected to cytotoxic temperatures. Thus, an essential objective of ablative therapy is achievement and maintenance of a 55°– 100°C temperature throughout the entire target volume for at least 4–6 minutes. However, the relatively slow thermal conduction from the electrode surface through the tissues may increase the duration of application up to 30 minutes. On the other hand, the tissue temperature should not be increased over these values to avoid carbonization around the tip of the electrode due to excessive heating (12). In the early experiences with RF ablation, a major limitation the technique was the small volume of necrosis created by conventional monopolar electrodes. These devices were capable of producing cylindrical lesions not greater than 1.6 cm in diameter. Therefore, multiple electrode insertions were necessary to treat of all but the smallest lesions. Subsequently, several strategies for increasing the area of thermal necrosis achieved with RF treatment have been tested, including the use of multiprobe arrays, bipolar arrays, and saline injections during RF application. These devices were shown to increase the volume of coagulation necrosis that can be obtained in a single treatment session. However, such techniques were either technically challenging and timeconsuming or produced irregularly-shaped thermal lesions, thereby substantially limiting their clinical usefulness. A major progress in the RF technology was achieved with the introduction of modified electrodes, including cooled-tip electrode needles and expandable electrode needles with multiple retractable lateral-exit

prongs on the tip. Cooled-tip electrodes consist of dual-lumen needles with uninsulated active tips, in which internal cooling is obtained by continuous perfusion with chilled saline. Needle cooling is aimed at preventing overheating of tissues nearest to the electrode, which may cause charring, thereby limiting the propagation of RF waves. They are available either as a single needle or as a cluster array with three needles spaced 0.5 cm apart. Expandable needles have an active surface which can be substantially expanded by hooks deployed laterally from the tip. The number of hooks and the lenght of the hooks deployment may vary according to the desired volume of necrosis. These techniques enabled a substantial and reproducible enlargement of the volume of thermal necrosis produced with a single needle insertion, and prompted the start of clinical application of RF ablation. Technique and Methodology At our institution, we currently use 150- or 200-W RF generators and 14-gauge expandable electrode needles (StarBurst XL, RITA Medical Systems). The needle electrode consist of an insulated outer cannula that houses nine curved electrodes of various lengths, that deploy out from the trocar tip. This design decreases the distance between the tissue and the electrodes, thereby ensuring uniform heating that relies less on heat conduction over a large distance. Five of the electrodes are hollow and contain thermocouples in their tips that are used to measure the temperature of the adjacent tissue. Probe-tips temperatures, tissue impedance, and wattage are displayed on the RF generator and graphically recorded by a dedicated software, installed on a personal computer. Maximum power output of the RF generator, amount of electrode array deployment from the trocar, and duration of the effective time of the ablation (time at target temperature) depend on the desired volume of ablation. This is established at the beginning of the procedure with the goal to destroy ideally the visible tumor mass plus a 1-cm safety margin of ablation all around. To perform a typical ablation, two grounding pads are placed on the patient's thighs. The tip of the needle (with retracted electrodes) is advanced under ultrasound guidance to the proximal edge of the lesion, and the electrodes are deployed to 2 cm. The generator is turned on and runs by an automated program. The temperature at the tips of the electrodes are controlled and the peak power is maintined until the temperature exceeds the preselected target temperature (typically between 90° and 100°C). After the target temperature is achieved, the curved electrodes can be advanced stepby-step to full deployment. When the electrodes are fully deployed, the program maintains the target temperature by regulating the wattage. At the end of the procedure, when the generator runs off, a “cool down cycle” is automatically performed. After retracting the hooks, the coagulation of the needle track can be done (track ablation) mantaining the temperature above 75°C with the aim to prevent any tumor cell dissemination. In our center, percutaneous RF is usually performed under conscious sedation. The association of an hypnotic drug with an ultrashort halflife analgesic drug allows a mild sedation and the patient, who can cooperate with the operator and bear the pain induced by treatment. Our standard protocol consists in admistering a bolus of ketorolac (0.5-0.8 mg/kg) followed by infusion of propofol (1-2 mg/kg/h) and remifentanil (0.1 mg/kg/min). However, drug posology has to be modulated in relation to the individual patient compliance and to the different phases of the procedure. The infusion of the hypnotic drug can be varied between 0.5 and 2 mg/kg/h to achieve a patient sedation that preserves the ability to do easy actions. The infusion of remifentanil can be varied between 0.05 and 0.15/kg/min to obtain an optimal analgesia. Attention has to be made in order to avoid bolus administration of remifentanil, as this may cause respiratory depression. The procedure is performed under standard cardiac, pressure, and oxigen monitoring with continuous oxigen administration. A careful post-treatment protocol is to be recommended following RF ablation. The patient is kept under close medical observation and rescanned with ultrasound 1-2 hours after the procedure. An overnight hospital stay is scheduled. Contrast-enhanced ultrasound performed shortly after the procedure may allow an initial evaluation of tumor response, by showing disappearance of intratumoral signals (13). Spiral CT obtained 1-3 days after the ablation shows a core of hypoattenuation surrounded by an enhancing rim. The peripheral enhancing rim which is due to the inflammatory reaction surrounding the area of necrosis - should not be misinterpreted as tumor persistence. Since the enhancing rim tends to disappear over time, spiral CT at 1 month is con-

sidered the most reliable method to evaluate the outcome of treatment (1). If there is imaging evidence of residual tumor, the patient can be considered for repeated RF ablation, provided that requirements for treatment are still met. Follow-up ultrasound and spiral CT studies are usually scheduled at 3- or 6-month intervals. Clinical Results In most clinical experiences with RF thermal ablation, patients were treated in the framework of feasibility studies, aimed at analysing safety, tolerability, and local therapeutic effect of the treatment.. In the series of Shirato et al (14), 28 of 30 HCC nodules smaller than 3 cm did not show residual tumor on imaging studies performed after RF ablation, and did not show any local recurrence after a follow-up period of 3-15 months (mean, 8.4 months). Livraghi et al (18) also reported that complete response was seen on post-treatment CT in 47 of 52 HCC lesions smaller than or equal to 3 cm in diameter after treatment with cooled-tip RF ablation. The rate of complete responses was higher than that achieved by the same authors in a group of comparable patients who underwent PEI (90% vs. 80%). In addition, the average number of treatment sessions needed to achieve ablation was substantially lower for RF than for PEI. However, one major complication and four minor complications occurred in patients treated with cooled-tip RF ablation, whereas no complications occurred in patients treated with PEI. One of the current recommendations for research of the European Associaton for the Study of the Liver is to compare newer methods of tumor destruction, such as RF, with the well-established and accepted PEI through randomized trials assessing not only initial tumor response, but also long-term survival outcomes (1). We therefore undertook a prospective randomized study aimed at comparing the efficacy of RF ablation with that of PEI for the first-line treatment of small HCC in cirrhosis (16). Primary end-point of the study was overall survival. Secondary end-points were local recurrence-free survival and event-free survival (i.e., survival free from local recurrence, new HCC tumors, and extrahepatic metastases). Fifty-two patients with 69 HCC tumors were treated with RF ablation (“RF group”), while 50 patients with 73 HCC tumors received PEI (“PEI group”). No statistically significant differences between RF and PEI groups were observed with respect to baseline characteristics, except for patients age and albumin concentration. Primarly due to relatively short observation period and small number of deaths in both treatment groups, we were not able to demonstrate a statistically significant difference between RF ablation and PEI with respect to overall survival. Nevertheless, a trend towards increased survival in the RF thermal ablation group was obvious. The overall survival rates after 1 and 2 years were 96% and 88%, respectively, in the PEI group, and 100% and 98%, respectively, in the RF group (p=0.138). However, 1- and 2-year local recurrence-free survival rates were significantly higher in the RF group (98% and 96%, respectively) than in the PEI group (83% and 62%, respectively, p=0.002). One- and 2-year event-free survival rates were also higher in RF-treated patients (86% and 64%, respectively) than in PEI-treated patients (77% and 43%, respectively, p=0.012, Fig. 2). RF treatment was confirmed as independent prognostic factor for local recurrence-free survival by multivariate analysis (adjusted RR=0.20, p=0.015). Despite the fact that the RF generator used in this study has been currently replaced by newer devices, RF ablation was shown to be more effective than PEI in the treatment of small HCC in cirrhosis. However, treatment of large tumors is still problematic. Livraghi et al (17) treated 114 patients with 126 HCC lesions greater than 3 cm in diameter. Complete necrosis (on imaging) was attained in only 60 lesions (47.6%), nearly complete (90%-99%) necrosis in 40 lesions (31.7%), and partial (50%-89%) necrosis in the remaining 26 lesions (20.6%). Medium and/ or noninfiltrating tumors were treated successfully significantly more often than large and/or infiltrating tumors. Therefore, there is currently a focus on a multimodality strategy in attempts to ensure a more effective treatment of large tumors. Recent studies have proved the influence of perfusion-mediated tissue cooling on the area of thermal necrosis achievable with RF treatment. Goldberg et al (18) applied RF in vivo to normal porcine liver without and with balloon occlusion of the portal vein, celiac artery, or hepatic artery, and to ex vivo calf liver: RF application during vascular occlusion produced larger areas of coagulation necrosis than RF with unaltered blood flow. The same authors demonstrated that intraoperative RF application produced greater coagulation diameter for human hepatic metastases treated during portal inflow occlusion than for tumors treated with normal blood flow.

Assuming that the volume of thermal necrosis produced by RF treatment is strongly dependend on blood flow, and considering that in HCC blood flow is mainly sustained by the hepatic artery, we designed a multicenter clinical trial aimed at investigated whether interruption of the tumor arterial blood supply by means of occlusion of either the hepatic artery with a balloon catheter or the feeding arteries with gelatin sponge particles could increase the extent of RF-induced coagulation necrosis (19). A series of 62 consecutive patients with a single, large HCC ranging from 3.5 to 8.5 cm in diameter (mean, 4.7 cm) accompanying cirrhosis underwent RF ablation after occlusion of the tumor arterial supply. The RF energy was delivered by using an expandable electrode needle at the time of balloon catheter occlusion of the hepatic artery (n = 40), at the time of occlusion of the HCC feeding arteries with gelatin sponge particles (n = 13), or 2–5 days thereafter (n = 9). Two patients underwent liver resection after the thermal ablation; the remaining 60 patients were followed up for a mean of 12.1 months (range, 3–26 months). During the follow-up, 49 (82%) of the 60 treated HCC nodules showed stable complete response, while the remaining 11 (18%) nodules showed local progression. Histopathologic analysis of one autopsy and of the two surgical specimens revealed more than 90% necrosis in one specimen and 100% necrosis in two. No fatal or major complications related to the treatment occurred, despite the more aggressive RF treatment protocol. Results of this study provide evidence that areas of coagulative necrosis that are much larger than those previously reported can be created if RF thermal ablation is performed in HCC nodules after occlusion of their arterial supply. The results achieved with this technique were confirmed by two recent studies. Yamasaki et al (20) compared the coagulation diameters obtained with balloon-occluded RF and standard RF in 31 patients with 42 HCC lesions measuring less than 4 cm in the greatest dimension. There were no significant differences in the ablation conditions such as the frequency of a fully expanded electrode, the number of needle insertions, application cycles, or treatment times between the two groups. However, the greatest dimension of the area coagulated by balloon-occluded RFA was significantly larger than that coagulated by standard RFA. Yamakado et al (21) evaluated the local therapeutic efficacy of RF ablation after transarterial chemoembolization in 64 patients with 108 lesions. Sixty-five lesions were small (3 cm or less), 32 were intermediate in size (3.1-5 cm), and 11 were large (5.1-12 cm). Complete necrosis was achieved in all lesions, and there were no local recurrences in small and intermediate-sized lesions during a mean follow-up of 12.5 months. The safety profile of RF ablation has been assessed by a recent multicenter survey in which 2,320 patients with 3,554 lesions were included (22). The majority of patients (1,610) had HCC with chronic liver disease. Six deaths (0.3%) were noted, including two caused by multiorgan failure following intestinal perforation; one case each of septic shock following Staphylococcus aureus-caused peritonitis, massive hemorrhage following tumor rupture, liver failure following stenosis of right bile duct; and one case of sudden death of unknown cause 3 days after the procedure. Fifty (2.2%) patients had additional major complications. Results of this study confirm that RF ablation is a relatively low-risk procedure for the treatment of focal liver tumors. Nevertheless, subcapsular HCC lesions should be treated with caution as they appear to be associated with a higher risk of major complications, especially tumor seeding. OTHER METHODS OF PERCUTANEOUS ABLATION Other methods of percutaneous ablation have been clinically tested over the past few years. Ohnishi et al (23) compared percutaneous acetic acid injection and PEI. Sixty patients with one to four HCCs smaller than 3 cm were entered onto a randomized controlled trial. Thirty-one and 29 patients, respectively, were treated by percutaneous acetic acid injection using 50% acetic acid or by PEI using absolute ethanol. All original tumors were treated successfully by either therapy. However, 8% of 38 tumors treated with percutaneous acetic acid injection and 37% of 35 tumors treated with PEI developed a local recurrence (P <.001) during the follow-up periods of 29 +/- 8 months and 23 +/- 10 months, respectively. The 1- and 2-year survival rates were 100% and 92% in percutaneous acetic acid injection and 83% and 63% in percutaneous ethanol injection (P = .0017). A multivariate analysis of prognostic factors revealed that treatment was an independent predictor of survival. The authors concluded that percutaneous acetic acid injection was superior to PEI in the treatment of small HCC. However, the results of acetic acid injection were not established in large series of patients.

Other groups investigated the usefulness of percutaneous microwave coagulation therapy (MCT). Seki et al (24) performed a retrospective study in which 90 patients with small HCC were included. The overall 5year survival rates for patients with well-differentiated HCC treated with MCT (70%) and PEI (78%) were not significantly different. Among the patients with moderately or poorly differentiated HCC, overall survival with MCT was significantly better than with PEI (5-year survival rate, 35%) (P = 0.03), and 9 of 22 patients with moderately or poorly differentiated HCC treated with PEI experienced recurrence in the original target subsegment compared to 2 of 25 patients treated with MCT. The authors concluded that MCT may be superior to PEI for the local control of moderately or poorly differentiated small HCC. Shibata et al (25) compared the effectiveness of MCT with that of RF ablation in 72 patients with 94 HCC nodules who were randomly assigned to PMC and RF ablation groups. The number of treatment sessions per nodule was significantly lower in the RF ablation group than in the MCT group (1.1 vs 2.4; P <.001). Complete therapeutic effect was achieved in 41 (89%) of 46 nodules treated with MCT and in 46 (96%) of 48 nodules treated with RF ablation (P =.26). Major complications occurred in four patients treated with MCT and in one patient treated with RF ablation (P =.36). During follow-up (range, 6-27 months), residual foci of untreated disease were seen in eight of 46 nodules treated with MCT and in four of 48 nodules treated with RF ablation. They concluded that MCT and RF ablation had equivalent therapeutic effects, but that RF achieves tumor ablation with fewer sessions. Recently, Adam et al (26) compared the complication and success rates in patients treated with percutaneous cryosurgery (PCS) or percutaneous RF ablation for unresectable hepatic malignancies. Sixty-four patients were treated with either PCS (n = 31) or PRF (n = 33). Patient treatment was based on the random availability of the probes. The distribution of tumor types was similar in the two groups (P =.76). Initial treatment success was comparable in the two treatment groups. However, local recurrences occurred more frequently after PCS than after RF ablation (16 [53%] of 30 vs 6 [18%] of 34; P =.003). Multivariate analysis demonstrated that the use of PCS (P =.003) was an independent risk factors for local tumor recurrence. The authors concluded that, while similar initial treatment success and complication rates are observed following either PCS or RF ablation, local recurrences occur more frequently following PCS. CONCLUSION From the early studies with PEI to the latest reports on RF ablation, percutaneous techniques have been refined and their clinical efficacy better defined. While PEI is a valuable and accepted treatment for small HCC lesions, RF ablation seems to achieve more effective local control of disease with fewer treatment sessions and appears to be superior to PEI as well as to other chemical or thermal methods of tissue destruction. RF ablation could therefore be currently considered as the firstline treatment of choice for patients with small HCC who are not suitable candidates for resection or transplantation. Nevertheless, further studies investigating the long-term oucomes of RF-treated patients are warranted to fully establish the clinical efficacy of this technique. PEI as well as segmental transarterial chemoembolization - will continue to have a valuable complementary role in patients with early-stage HCC. Appropriate use of each treatment technique can only be done when the therapeutic strategy is decided by a multidisciplinary team and is tailored to the individual patient and to the features of the disease. References: 1. Bruix J, Sherman M, Llovet JM, Beaugrand M, Lencioni R. Clinical management of hepatocellular carcinoma. Conclusions of the Barcelona-2000 EASL conference. European Association for the Study of the Liver. J Hepatol 2001; 35: 421-430 2. Lencioni R, Cioni D, Bartolozzi C. Percutaneous radiofrequency thermal ablation of liver malignancies: techniques, indications, imaging findings, and clinical results. Abdom Imaging 2001; 26: 345-360 3. Llovet JM, Bru C, Bruix J. Prognosis of hepatocellular carcinoma: the BCLC staging classification. Semin Liver Dis 1999; 19: 329-338 4. Bartolozzi C, Lencioni R. Ethanol injection for the treatment of hepatic tumours. Eur Radiol 1996; 6: 682-696 5. Castells A, Bruix J, Bru C, et al. Treatment of small hepatocellular carcinoma in cirrhotic patients: a cohort study comparing surgical resection and percutaneous ethanol injection. Hepatology 1993; 18: 11211126 6. Lencioni R, Bartolozzi C, Caramella D, et al. Treatment of small hepa-

tocellular carcinoma with percutaneous ethanol injection. Analysis of prognostic factors in 105 Western patients. Cancer 1995; 76: 17371746 7. Livraghi T, Giorgio A, Marin G, et al. Hepatocellular carcinoma and cirrhosis in 746 patients: long-term results of percutaneous ethanol injection. Radiology 1995; 197: 101-108 8. Yamamoto J, Okada S, Shimada K, et al. Treatment strategy for small hepatocellular carcinoma: comparison of long-term results after percutaneous ethanol injection therapy and surgical resection. Hepatology 2001; 34: 707-713 9. Shiina S, Teratani T, Obi S, Hamamura K, Koike Y, Omata M. Nonsurgical treatment of hepatocellular carcinoma: from percutaneous ethanol injection therapy and percutaneous microwave coagulation therapy to radiofrequency ablation. Oncology 2002; 62 Suppl 1: 64-68 10. Khan KN, Yatsuhashi H, Yamasaki K, et al. Prospective analysis of risk factors for early intrahepatic recurrence of hepatocellular carcinoma following ethanol injection. J Hepatol 2000; 32: 269-278 11. Koda M, Murawaki Y, Mitsuda A, et al. Predictive factors for intrahepatic recurrence after percutaneous ethanol injection therapy for small hepatocellular carcinoma. Cancer 2000; 88: 529-537 12. Goldberg SN, Gazelle GS, Mueller PR. Thermal ablation therapy for focal malignancies: a unfied approach to underlyng principles, techniques, and diagnostic imaging guidance. AJR Am J Roentgenol 2000; 174: 323-331 13. Cioni D, Lencioni R, Rossi S, et al. Radiofrequency thermal ablation of hepatocellular carcinoma: using contrast-enhanced harmonic power doppler sonography to assess treatment outcome. AJR Am J Roentgenol 2001; 177: 783-788 14. Shirato K, Morimoto M, Tomita N, et al. Small hepatocellular carcinoma: therapeutic effectiveness of percutaneous radiofrequency ablation therapy with a LeVeen needle electrode. J Ultrasound Med 2002; 21: 67-76 15. Livraghi T, Goldberg SN, Lazzaroni S, Meloni F, Solbiati L, Gazelle GS. Small hepatocellular carcinoma: treatment with radio-frequency ablation versus ethanol injection. Radiology 1999; 210: 655-661 16. Lencioni R, Allgaier HP, Cioni D, et al. Small hepatocellular carcinoma in cirrhosis: randomized comparison of radiofrequency thermal ablation versus percutaneous ethanol injection. Radiology 2003; 228 (in press) 17. Livraghi T, Goldberg SN, Lazzaroni S, et al. Hepatocellular carcinoma: radio-frequency ablation of medium and large lesions. Radiology 2000; 214: 761-768 18. Goldberg SN, Hahn PF, Tanabe KK, et al. Percutaneous radiofrequency tissue ablation: does perfusion-mediated tissue cooling limit coagulation necrosis? J Vasc Interv Radiol 1998; 9: 101-115 19. Rossi S, Garbagnati F, Lencioni R, et al. Percutaneous radio-frequency thermal ablation of nonresectable hepatocellular carcinoma after occlusion of tumor blood supply. Radiology 2000; 217: 119-126 20. Yamasaki T, Kurokawa F, Shirahashi H, Kusano N, Hironaka K, Okita K. Percutaneous radiofrequency ablation therapy for patients with hepatocellular carcinoma during occlusion of hepatic blood flow. Comparison with standard percutaneous radiofrequency ablation therapy. Cancer 2002; 95: 2353-2360 21. Yamakado K, Nakatsuka A, Ohmori S, et al. Radiofrequency ablation combined with chemoembolization in hepatocellular carcinoma: treatment response based on tumor size and morphology. J Vasc Interv Radiol 2002; 13: 1225-1232 22. Livraghi T, Solbiati L, Meloni MF, Gazelle GS, Halpern EF, Goldberg SN. Treatment of focal liver tumors with percutaneous radio-frequency ablation: complications encountered in a multicenter study. Radiology 2003; 226: 441-451 23. Ohnishi K, Yoshioka H, Ito S, Fujiwara K. Prospective randomized controlled trial comparing percutaneous acetic acid injection and percutaneous ethanol injection for small hepatocellular carcinoma. Hepatology 1998; 27: 67-72 24. Seki T, Wakabayashi M, Nakagawa T, et al. Percutaneous microwave coagulation therapy for patients with small hepatocellular carcinoma: comparison with percutaneous ethanol injection therapy. Cancer 1999; 85: 1694-1702 25. Shibata T, Iimuro Y, Yamamoto Y, et al. Small hepatocellular carcinoma: comparison of radio-frequency ablation and percutaneous microwave coagulation therapy. Radiology 2002; 223: 331-337 26. Adam R, Hagopian EJ, Linhares M, et al. A comparison of percutaneous cryosurgery and percutaneous radiofrequency for unresectable hepatic malignancies. Arch Surg 2002; 137: 1332-1339

Special Session Aortic Stent Grafts 28.3 Endoleak Management K. Ivancev1, N. Dias1, B. Sonesson2, M. Malina2; 1Dept of Radiology, Malmo University Hospital, Malmö, Sweden, 2Dept of Vascular Surgery, Malmo University Hospital, Malmö, Sweden. Endovascular abdominal aortic aneurysm repair (EVAR) has dramatically changed the treatment of abdominal aortic aneurysms (AAA) during the last decade. A constantly increasing number of patients are undergoing EVAR. Although EVAR has been repeatedly shown to have obvious advantages over open repair (OR), there is a continuous concern about the long-term durability. One of the principal reasons for failure after the first year following EVAR is the occurrence of endoleak (EL). The definition of EL is: the presence of blood flow outside the lumen of the endoluminal stent-graft (SG) but within the aneurysm sac as determined by an imaging study. An EL can be classified according to the time of occurrence and site of origin1. When observed during the first post-operative month, it is defined as primary EL, and when detected later on - a secondary EL. More important is the origin of the EL. A type I EL indicates a persistent blood flow along the SG attachment sites and is divided further into subgroups according to the precise position of the attachment site. Type II EL relates to flow from branch vessels without attachment site connection. The branch vessels may include lumbar arteries, the inferior mesenteric artery (IMA), or accessory renal arteries. The number of branch vessels involved determine whether it is A. simple, from one vessel to another, or B., complex, with multiple vessels involved. Type III EL is associated with a SG defect, use of subscript A denotes a junctional leak or disconnection of module elements, and B fabric rupture. Type IV EL denotes a blood flow through an intact, but porous, graft fabric, and is observed only during the first 30 days after SG implantation. The final, and most controversial, EL is type V, or endotension. The strict definition includes an increased intrasac pressure after EVAR, without a visualization of EL on the imaging study, i.e., although there is no demonstrable blood flow in the aneurysm sac, endotension is the ultimate evidence of failed EVAR2. In this presentation the prevalence of different types of EL and their management will be discussed. Diagnostic imaging The most commonly used imaging modality for EVAR surveillance is contrast-enhanced spiral CT. Usually it involves scanning prior to, and thereafter following intravenous injection of iodinated contrast medium. In addition delayed scanning approximately 1 minute after the end of the intravenous contrast-medium injection enhances the ability of the spiral CT to detect endoleaks3. Colour Duplex ultrasonography (CDU) has been claimed to be more sensitive than the so called triphasic spiral CT scanning, especially when ultrasound contrast medium is used4. However CDU is strongly operator dependent and in addition has been shown to be afflicted with a high percentage of falsely positive results. Finally gadolinium-enhanced MR-angiography (MRA) may have the potential to be a valuable alternative to spiral CT provided that the SG is compatible with MR which is not always the case, especially when the SG is made out of stainless steel. Digital subtraction angiography with selective catheterization if indicated into the internal iliac artery, iliolumbar arteries or the superior mesenteric artery and middle colic artery is regarded as the golden standard for identifying the source of different endoleaks. Type I and Type III endoleaks These two types of EL are always associated with either persistant or recurrent pressurization of the aneurysm sac. Their frequency has been reported to vary from less than 1 % up to 4-5%. The latter reflects experience from early generations of SG. By now there is strong evidence to suggest that these two types of EL are strong predictors of aneurysm rupture and therefore they need to be treated aggressively5. This may be accomplished endoluminally by the placement of additional stent-grafts as proximal or distal extensions or, if necessary, conversion to OR. Type II endoleaks These branch endoleaks vary in frequency from 8 to 24%5-9,18,24,28-30. This variation may be explained by the fact that type II EL have a tendency to resolve spontaneously6-9. Though there are no exact numbers

of how often this happens, a spontaneous thrombosis of type II EL has been reported to occur in approximately one third of the patients. In the majority of type II EL the feeding arteries are the lumbar arteries, whereas in approximately one third of the patients the IMA and possibly an accessory renal artery may also be involved. There is no correlation between the number of existing lumbar arteries and/or patent inferior mesentric artery and the risk for occurrence of type II EL10-12. Neither is there an increased risk for type II endoleaks in the presence of coumadin treatment which may instead contribute to the persistence of type II EL. Aneurysms exposed to type II EL usually remain unchanged with time13-17 whereas upon resolution, regardless of whether this is spontaneous or the result of successful treatment, there is a decrease in diameter and volume of the aneurysm sac18-19. Enlargement of the aneurysm sac or aneurysm rupture caused by type II EL is a very rare event20-22. Pre-operative embolization of lumbar arteries and the inferior mesentric artery does not decrease the risk for development of type II EL14. Transcatheter embolization of feeding arteries is frequently unsuccessful, owing to recruitment of new collaterals to the perfused aneurysm sac7,23-24. Direct translumbar puncture of the EL has been claimed to be the most effective treatment modality available25. Laparoscopic clipping of lumbar arteries is another more seldom used but also effective treatment26. Both modalities may come in handy in the rare situations of enlargement of the aneurysm sac caused by type II EL. Type IV endoleak This type of endoleak occurs due to porosity of thin wall graft material. It is claimed to thrombose spontaneously during the first 30-day postoperative follow up. It is unclear what the effect of thrombosed type IV EL may have on the degree of exclusion of the AAA. Type V endoleak – endotension Persistent or recurrent pressurization of the aneurysm sac without demonstrable endoleak is often demonstrated by enlarging aneurysm size. This can happen at any time after EVAR. The most common cause is SG migration, progressive aneurysm neck dilatation or separation of modular components. Though there is strictly speaking no contrast accumulation outside the SG, the systemic pressure is still obviously transmitted across the thrombus surrounding the SG27. A channel for transmission of systemic pressure that is more difficult to identify is defects in the graft fabric, especially in early generations of SG. Sometimes, spontaneously sealed type I EL may also be the source of endotension. In situations of growing aneurysm size without an obvious explanation for the pressurization of the aneurysm sac, conversion to OR may be the last option. Conclusion Endoleaks along the implantation site of the SG (type I) or through SG defects, either in the fabric or due to separation of modular elements (type III), are associated with a high risk for aneurysm rupture and should therefore be treated aggressively. Endoleak through persistent flow of branch vessels (type II), either lumbar arteries or IMA often result in unchanged aneurysm size and in that case can be safely followed up by repeated spiral CT examinations. Upon increase of aneurysm size, these type II EL should be either embolized by direct translumbar puncture or using laparoscopic technique and clipping of the lumbar arteries. In case of increasing aneurysm size without EL (endotension) an explanation should be looked for, such as SG migration or separation of modular components, or SG defects, in order to choose between conversion to open surgery or further endoluminal solutions to prevent aneurysm rupture. In situations of unchanged aneurysm size, direct intrasac pressure measurement may be helpful in defining follow-up policy and possible treatment. References: 1. Chaikof, E. L. et al. Reporting standards for endovascular aortic aneurysm repair. J Vasc Surg 35, 1048-60. (2002). 2. Veith, F. J. et al. Nature and significance of endoleaks and endotension: summary of opinions expressed at an international conference. J Vasc Surg 35, 1029-35. (2002). 3. Schurink, G. W. et al. Endoleakage after stent-graft treatment of abdominal aneurysm: implications on pressure and imaging--an in vitro study. J Vasc Surg 28, 234-41 (1998). 4. Bendick, P. J. et al. Efficacy of ultrasound scan contrast agents in the noninvasive follow-up of aortic stent grafts. J Vasc Surg 37, 381-5 (2003). 5. van Marrewijk, C. et al. Significance of endoleaks after endovascular repair of abdominal aortic aneurysms: The EUROSTAR experience.

J Vasc Surg 35, 461-73. (2002). 6. Becker, G. J. et al. Risk stratification and outcomes of transluminal endografting for abdominal aortic aneurysm: 7-year experience and long-term follow-up. J Vasc Interv Radiol 12, 1033-46. (2001). 7. Karch, L. A. et al. Algorithm for the diagnosis and treatment of endoleaks. Am J Surg 178, 225-31 (1999). 8. Abraham, C. Z. et al. Abdominal aortic aneurysm repair with the Zenith stent graft: Short to midterm results. J Vasc Surg 36, 217-25. (2002). 9. Faries, P. L. et al. A multicenter experience with the Talent endovascular graft for the treatment of abdominal aortic aneurysms. J Vasc Surg 35, 1123-8. (2002). 10. Walker, S. R. et al. A study on the patency of the inferior mesenteric and lumbar arteries in the incidence of endoleak following endovascular repair of infra- renal aortic aneurysms. Clin Radiol 53, 593-5 (1998). 11. Gould, D. A. et al. Aortic Side Branch Embolization before Endovascular Aneurysm Repair: Incidence of Type II Endoleak. J Vasc Interv Radiol 12, 337-341. (2001). 12. Petrik, P. V. & Moore, W. S. Endoleaks following endovascular repair of abdominal aortic aneurysm: the predictive value of preoperative anatomic factors--a review of 100 cases. J Vasc Surg 33, 739-44. (2001). 13. Arko, F. R. et al. Type-II endoleaks following endovascular AAA repair: preoperative predictors and long-term effects. J Endovasc Ther 8, 503-10. (2001). 14. Parry, D. J. et al. Type II endoleaks: predictable, preventable, and sometimes treatable? J Vasc Surg 36, 105-10. (2002). 15. Rhee, R. Y., Eskandari, M. K., Zajko, A. B. & Makaroun, M. S. Longterm fate of the aneurysmal sac after endoluminal exclusion of abdominal aortic aneurysms. J Vasc Surg 32, 689-696 (2000). 16. Czermak, B. V. et al. Serial CT volume measurements after endovascular aortic aneurysm repair. J Endovasc Ther 8, 380-9. (2001). 17. Resch, T. et al. Persistent collateral perfusion of abdominal aortic aneurysm after endovascular repair does not lead to progressive change in aneurysm diameter. J Vasc Surg 28, 242-9. (1998). 18. Liewald, F. et al. Influence of treatment of type II leaks on the aneurysm surface area. Eur J Vasc Endovasc Surg 21, 339-43. (2001). 19. Ermis, C. et al. Does successful embolization of endoleaks lead to aneurysm sac shrinkage? J Endovasc Ther 7, 441-5. (2000). 20. Dattilo, J. B. et al. Clinical failures of endovascular abdominal aortic aneurysm repair: incidence, causes, and management. J Vasc Surg 35, 1137-44. (2002). 21. Hinchliffe, R. J., Singh-Ranger, R., Davidson, I. R. & Hopkinson, B. R. Rupture of an Abdominal Aortic Aneurysm Secondary to Type II Endoleak. Eur J Vasc Endovasc Surg 22, 563-5. (2001). 22. White, R. A., Donayre, C., Walot, I. & Stewart, M. Abdominal aortic aneurysm rupture following endoluminal graft deployment: report of a predictable event. J Endovasc Ther 7, 257-62 (2000). 23. Solis, M. M. et al. Mechanism of failure in the treatment of type II endoleak with percutaneous coil embolization. J Vasc Surg 36, 485-91. (2002). 24. Chuter, T. A. et al. Endoleak after endovascular repair of abdominal aortic aneurysm. J Vasc Surg 34, 98-105. (2001). 25. Baum, R. A. et al. Treatment of type 2 endoleaks after endovascular repair of abdominal aortic aneurysms: comparison of transarterial and translumbar techniques. J Vasc Surg 35, 23-9. (2002). 26. Wisselink, et al. Retroperitoneal endoscopic ligation of lumbar and inferior mesenteric arteries as a treatment of persistent endoleak after endoluminal aortic aneurysm repair. J Vasc Surg 31, 1240-4 (2000). 27. White, G. H. et al. Endotension: An explanation for continued AAA growth after successful endoluminal repair. Journal of Endovascular Surgery 6, 308-315 (1999). 28. Makaroun, M. S.et al.. Efficacy of a bifurcated endograft versus open repair of abdominal aortic aneurysms: a reappraisal. J Vasc Surg 35, 203-10. (2002). 29. Hölzenbein, T. J. et al. Midterm durability of abdominal aortic aneurysm endograft repair: a word of caution. J Vasc Surg 33, S46-54. (2001). 30. Fairman, R. M. et al. Potential impact of therapeutic warfarin treatment on type II endoleaks and sac shrinkage rates on midterm follow-up examination. J Vasc Surg 35, 679-85. (2002).

Special Session MR Guided Interventions 29.1 MRI-guided interventions - Where are we now? R. W. Guenther; Diagnostic Radiology, Aachen University Hospital, Aachen, Germany. MRI has become a very important imaging modality today and offers a great potential in future. Consequently, MRI-guided interventions deserve our attention and interest. Interventional procedures can be performed with different MRI units: Open and closed systems, highfield and low field systems, and hybrid systems, i.e. combination of MRI and x-ray fluoroscopy . The latter system compensates for the lack of free access, improves the safety of the interventions by providing a possibility to continue or to extend the intervention if restricted by MRI. It also allows the feasibility of vascular intervention to be studied. The properties of MRI make it a very suitable technique also for interventions. MRI provides: High intrinsic tissue contrast, vascular imaging is possible without administration of contrast medium, multiplanar imaging, temperature sensitivity and lack of irradiation. Potential applications of MR-guided interventions include: Percutaneous biopsy, interstitial therapy, focussed ultrasound, vascular and cardiac interventions, and sterotactic neurosurgical interventions Percutaneous biopsy and interstitial therapy seem to be realistic applications of MRI. To what extent, remains rather speculative for the time being: at least those lesions will be biopsied which are either not visible by other modalities or are in a difficult anatomic location. Interstitial therapy using MRI-guidance may be used for the treatment of liver metastases and palliation of recurrent head and neck tumors. Limitations of interstitial therapy are obvious: It seems to provide only a palliative treatment and tumor necrosis achieved depends on tumor size among other factors. Focussed ultrasound is in a very early stage of development and evaluation regarding its potential in tumor therapy. It may lead to coagulation necrosis due to a thermal effect, but application of ultrasound is limited by overlying bones and air. Vascular and cardiac interventions may be very complex requiring realtime and rapid imaging with high spatial resolution. Even though the feasibitlity of interventions such as coronary angioplasty, coronary stent placement and percutaneous atrial septal occlusion has been demonstrated experimentally, cardiovascular interventions in MRI continue to be a great challenge in clinical application. Freedom of access and real-time imaging with high temporal and spatial resolution will be of the utmost importance in the further development of MRI-guided interventions. The perspectives are very promising and interesting enough for this technique to be pursued very intensively in the next few years. Refrences: Buecker A, Spuentrup E, Grabitz R, Freudenthal F, Muehler EG, Schaefter T, van Vaals JJ, Günther RW: Magnetic resonance-guided placement of atrial septal closure device in animal model of patent foramen ovale. Circulation 2002; 106: 511-515 Buecker A, Adam G, Neuerburg JM, Kinzel S, Glowinski A, Schaeffter T, Rasche V, vanVaals JJ, Guenther RW: Simultaneous real-time visualization of the catheter tip and vascular anatomy for MR-guided PTA of iliac arteries in an animal model. J Magn. Res. Imaging 2002; 16: 201208 Günther, R.W., A. Bücker, G. Adam: Interventional magnetic resonance: Realistic prospect or wishful thinking? Cardiovasc. Intervent. Radiol. 1999; 22 187-195 Spuentrup E, Ruebben A, Schaeffter T et al.: MR-guided coronary stent placement in a pig model. Circulation 2002; 105: 874-879

29.3 Interventional MR in Cardiovascular Disease C. B. Higgins; Radiology, University of California, S.F., San Francisco, United States. The development of fast MR imaging techniques in recent years has rendered MR imaging capable of guiding interventional procedures. The rationale to use MR instead of X-ray fluoroscopy for guiding endovascular therapy is the imperative to reduce radiation exposure to the patient and the interventionalist. In order to develop MR for endovascular therapy, a dual X-ray/MR facility has been devised; this consists of

an X-ray catheterization laboratory and 1.5 Tesla MR imager in adjacent rooms separated by a sliding door which is impervious to both X-rays and radiofrequency waves. The patient can be moved in a direct line between the X-ray equipment and the MR imager in less than one minute. This dual-imaging interventional facility is called XMR (Philips Medical Systems, the Netherlands). In experimental animals, the XMR has been used for the following cardiovascular interventional procedures: 1. Transcatheter delivery of valved stents to the pulmonic area. After implantation of the stent, velocity encoded cine MR is employed to assess and quantify flow dynamics. 2. Delivery of closure devices for atrial septal defects. 3. Delivery of stents to the thoracic aortic, pulmonary arteries, and coronary arteries. 4. Delivery of therapeutic solutions such as an angiogenesis growth factor to the myocardium. MR contrast media (dilute Gd DTPA) is used to document the site and volume of injectate in the myocardium. Thus, the guidance of endovascular interventional procedures is feasible. A major limitation is the need for fabrication of catheters and devices which are readily visible on MR imaging but free of distorting image artifacts. No References

Special Session Renal Artery Interventions 31.1 Renal Artery Imaging D. A. Kelekis, S. I. Argentos; 2nd Radiology, Athens University, Athens, Greece. Introduction Renal artery interventions (either angioplasty or stenting) are performed to treat renal artery stenosis. Thus, among other indications for renal artery imaging, focusing on renal artery stenosis (RAS) is of special interest for interventional radiologists. Follow-up after intervention is also an issue of utmost importance. The aim of diagnosing RAS is to suggest a causal relationship to hypertension or renal failure and consider therapeutic intervention. Among the population of hypertensive patients approximately 1-5% have Renovascular Hypertension (RVH). Clinical features suggestive of renovascular hypertension as listed by the Sixth Report of the Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure (Arch Intern Med 1997, 157:2413-2446), include: · Onset of hypertension before 30 years of age, especially without a family history, or recent onset of significant hypertension after 55 years of age. · Abdominal bruit, particularly if it continues into diastole and is lateralized. · Accelerated or resistant hypertension. · Recurrent (flash) pulmonary edema. · Renal failure of uncertain cause, especially with a normal urinary sediment. · Coexisting, diffuse atherosclerotic vascular disease, especially in heavy smokers. · Acute renal failure precipitated by antihypertensive therapy, particularly Angiotensin-converting enzyme (ACE) inhibitors or angiotensin II receptor blockers. In the presence of these clinical criteria the prevalence of RVH is 2030%. Consequently, in this selected population the investigation should be extended to noninvasive detection of RAS. Such noninvasive modalities are ACE Inhibitor Scintigraphy, Color Doppler Ultrasound, Computed Tomography Angiography (CTA) and Magnetic Resonance Angiography (MRA). Digital subtraction angiography (DSA), the “gold-standard” method for detection and quantification of renal artery stenosis, cannot be used as a screening tool. A renal artery imaging study must: a. identify the number and origin of the renal arteries, b. detect, localize and characterize renal artery stenosis, c. exhibit pathologic involvement of adjacent vascular structures such as aorta, d. provide additional information about kidney size, cortical thickness or incidental lesions of the kidneys or adrenals.

The most common causes of RAS are atherosclerosis and, less commonly, fibromuscular dysplasia (FMD). Atherosclerosis is more common in patients over the age of 50 yrs and involves the proximal renal artery, whereas FMD is more common in women under the age of 40 yrs and involves the distal main renal artery and segmental renal arteries. Morphologically, RAS is categorized as ostial (at its origin from the aorta, extending within its proximal 5-10 mm) or truncal (more than 10mm distal to the aortic lumen, proximal to renal artery branching). The degree of stenosis is graded as mild (25% -50 %), moderate (50% 75 %), severe (> 75% -99%), occlusion (100%). Scintigraphy The diagnosis of RAS is based on exhibition of ipsilateral impaired renal function (as indicated in the scintigraphic nephrogram) during ACE inhibition.This is believed to be caused by disruption of the autoregulation system of the glomerular filtration rate (GFR), which becomes dependent on angiotensin II under conditions of low perfusion. In the case of unilateral RAS, a unilateral change in renal function can be revealed with Scintigraphy. Such changes are not observed inpatients with nonsignificant RAS or normal renal arteries. ACE inhibitor therapy should be stopped at least 5 days before the examination and adequate hydration is required. Baseline and ACE Inhibitor (1 hour after administration of 25-50 mg of Captopril) Scintigraphy are performed after intravenous injection of DTPA-99mTc or MAG3-99mTc. DTPA-99mTc is excreted by glomerular filtration, whereas MAG3-99mTc by tubular secretion. The latter is preferred in case of renal failure. Diagnostic criteria include worsening of the scintigraphic curve, reduction in the relative uptake, prolongation of the transit time and delayed maximum activity. ACE Inhibitor Scintigrams are interpreted as low, intermediate or high probability of Renovascular disease. The sensitivity of the method is 90-95%. Bilateral RAS, impaired renal function, urinary obstruction are factors that lower its sensitivity. Color Doppler Ultrasound Two approaches are used to detect RAS: Direct visualization of the renal arteries (proximal criteria at the level of stenosis) Exploration of the main renal arteries with anterior or anterolateral approach using color or power Doppler US, followed by estimation of renal artery velocities. Criteria to diagnose significant stenosis or occlusion are: · Increase of peak systolic velocity (PSV) in the renal artery (> 150-200 cm/sec), · Renal/aortic ratio of PSV greater than 3-3,5, · Turbulent flow in the poststenotic segment of the renal artery, · No detectable Doppler signal (finding indicating occlusion) Contrast agents enhancing echo signal can be administered. These facilitate renal artery visualization and decrease the number of inconclusive examinations. Their clinical impact and potential has to be determined. This approach is relatively time-consuming, difficult to reproduce and limited by individual variability in the course of the main renal artery. Factors such as gas interposition or obesity result in non-feasible or technically inadequate examinations. Accessory renal arteries are also a limiting factor. According to literature, the sensitivity of the method ranges from 70% to 93%. Study of Intrarenal Waveforms(distal criteria- flow changes in the renal vasculature) Translumbar approach is used and segmental renal arteries are evaluated. Qualitative criteria (waveform pattern recognition): · Tardus parvus intrarenal waveform indicates stenosis. Quantitative criteria proposed for detection of significant RAS: · Acceleration of less than 370-470 cm/sec, · Acceleration time greater than 50-80 msec, · Change of resistive index of greater than 5% between the right and left kidney. This approach is easier, but limited in elderly and atherosclerotic patients. Combining morphologic and quantitative criteria, sensitivity for detection of RAS varies from 64% to 89% and specificity from 82% to 99%. Improved detection of RAS can be achieved by administration of Captopril. Technical advances such as matrix-array transducers, harmonic imaging techniques, 3D US Angiography are rapidly developing and can play an

increasing role in renal vascular imaging DISADVANTAGES · Up to 25% of the examinations may yield inadequate results, · The method is highly operator-dependent, · Inadequate visualization of accessory renal arteries CT Angiography CT Angiography is a fast, noninvasive method. It is performed during a single breath hold at peak vascular enhancement after I.V. injection of iodinated contrast medium via a power injector. Single-slice spiral CT scanners allow evaluation of vessels with diameter 2mm or more, although visualization of vessels smaller than 1 mm is possible. To improve spatial resolution and obtain the best possible longitudinal resolution the collimation, table speed (pitch) and reconstruction interval (spacing) should be minimized. In single-slice systems, is better to use thinner collimation and increase the pitch to cover the required region. 2 mm collimation with table speed of 3-4 mm/rotation (pitch 1,5-2) is considered optimal. 3 mm collimation with table speed of 3 mm/rotation (pitch 1) is also a protocol with good results. In both cases, images are reconstructed at 1 mm intervals. Rotation time (scan time) ranges from 0,75-1 sec. With the advent of multislice technology, near isotropic submillimeter resolution has become feasible. In multislice systems, thinner collimation (1-1,25 mm) is used, table speed of 6mm/rotation and rotation time 0,5 sec. Even thinner sections (0,5 mm) are technically feasible but, however, they are not routinely used because of noise and dose issues. (Multislice scanning suffers from low signal-to-noise ratios at thin sections). Images are reconstructed at a spacing of 0,5 mm. The total scan time is 20-30 sec and breath-holding is required. High degree vascular opacification is essential. Bolus I.V. injection of 100-150 ml of high-concentration low osmolar iodinated contrast agent at a flow rate of 3-5 ml/sec is used. Bolus timing (scan delay) can be determined either empirically or by means of test-bolus and bolustriggering techniques. Usually, scan delay time range from 25-40 sec. CT image data postprocessing techniques include curved planar reformations (CPR), shaded surface display (SSD), maximum intensity projections (MIP) and volume rendering (VR). For assessment of RAS we rely mainly on CPR and MIP-VR displays (MIP yields better results for small intrarenal branches). Curved thin-slab MIP views provide superior results, better than CPR. Generally, due to the possibility of an eccentric stenosis, reformations in the axial (caudocranial) and coronal (anteroposterior) planes should be viewed. The optimum approach is to generate reformations in 10º steps in longitudinal and horizontal axis. It is important to review cross-sectional (source) images, especially in the case of calcified plaques. Breathing or pulsation artifacts, partial volume effects (thick collimation), poor vascular enhancement (slow rate or wrong timing) may result in limited diagnostic accuracy of a renal CTA study. CTA is a sensitive and accurate method for the detection of significant RAS. Both, its sensitivity and specificity rise above 90%. It has also very high negative predictive value, which ranges above 95%. Lower sensitivity has been reported for FMD (75-85%-single-slice CTA). The main disadvantage of CTA is the need for intravenous nephrotoxic contrast agents and ionizing radiation. Further technical advances of multislice scanners can achieve reduction in contrast volumes and patient radiation dose. The advantages over angiography (DSA) include: · Noninvasive method, less cost, less radiation, · Simultaneous visualization of the vessel wall and lumen (distinguish between soft and calcified plaques), · Ability to study vessels at arbitrary angles using one acquisition (better quantification of eccentric stenoses and differentiation between ostial and truncal stenoses). MR Angiography 3D contrast enhanced MRA is currently used during a 20 to 30 sec breath hold. “High performance” gradient systems allow such a reduction in imaging times. A high-resolution 3D T1-W gradient-echo pulse sequence in conjunction with I.V. injection of high-dose Gadolinium is performed. Phased array coils are preferred for better image quality. Other protocol parameters include: slice thickness 2-3 mm, FOV 30-35 cm, minimum repetition time (TR 3-5 msec) and echo time (TE < 3 msec), flip angle of 30º-60º Contrast material timing and dose are critical for image quality. Typically, injection of 0,2-0,3 mmol/kgr gadolinium at a rate of 2-4 ml/sec is adequate (flushing with 20-ml normal saline solution is necessary). Accurate bolus timing can be achieved either by test-bolus or bolus-

triggering techniques. Postprocessing on an independent workstation is important. Shaded surface display (SSD), Maximum intensity projection (MIP) or volume rendering (VR) algorithm is used. Subvolume MIP or VR images in coronal oblique and axial oblique planes are mandatory to identify eccentric plaques. Although the most widely used postprocessing technique is MIP, recent studies refer more accurate detection and quantification of RAS with the VR technique. Curved planar reconstructions (CPRs) are very helpful. Correlation with source images is very important to avoid pitfalls and inaccurate evaluation. Poor quality images can result from failure to suspend breathing, inadequate dose of contrast material or poor bolus timing. To avoid pitfalls from poor image reconstruction, multiple views of renal arteries should be evaluated, so eccentric disease is revealed. The spatial resolution of 3D-Gd-MRA is limited compared with that of conventional Angiography. Even with state-of-the-art systems, the maximum spatial resolution is in the range of 1mm³. For this reason, MRA is prone to overestimation of stenoses. Acquisition of complementary MR sequences to assess the hemodynamic significance of stenoses is proposed, particularly for borderline (50-60%) stenoses. These sequences include mainly 3D Phase Contrast (PC) MRA and cardiacgated cine PC flow measurements. PC imaging is limited in-patients with slow flow or renal failure. Inadequate results are also encountered when there is inappropriate selection of velocity encoding. In addition, a common problem is artifactual spin dephasing at the renal artery origins. Several pulse sequences for the assessment of renal size and parenchymal disease should be employed, including fat suppressed pre- and post-contrast T1-W sequences. This comprehensive approach allows hemodynamic evaluation of renal artery stenosis during a single MRI examination that can be performed in about 1 hr. Most series since 1995 report sensitivity for the detection of RAS ranging from 88-100% and specificity ranging from 79-100%. Subtle irregularities of the distal main renal arteries, as is the case in fibromuscular dysplasia, may be difficult to demonstrate because in this case maximum spatial resolution is required. The accuracy of MRA in diagnosis of FMD has not been established. Major advantages of MRA are: · Noninvasive method, · No need for radiation or nephrotoxic contrast agents, · Can be performed in patients with renal failure or known allergy to iodinated contrast media, · Good contrast resolution, · Easier 3D postprocessing techniques, · Flow quantification (important for therapeutic decision). Main disadvantage is its low spatial resolution. Angiography It is the reference method (gold-standard). Its spatial resolution is very high (0,2-0,3 mm). However, it requires ionizing radiation, nephrotoxic contrast agents and catheterization. Consequently, it cannot be used as screening method. Several limitations exist, mainly correlated to inappropriate vessel projection and eccentric stenosis or anteroposterior calcifications. Technical advances such as 3D rotational angiography can further improve its diagnostic accuracy. Follow-up after intervention Apart from blood pressure and serum creatinine, radiological assessment of stent patency is required, usually at yearly intervals. Radiological follow-up is necessary because clinical follow-up may not enable detection of restenosis. Little consensus exists on what is the ideal follow-up modality after renal artery intervention. DSA cannot be used routinely for follow-up. MRA is the first-line examination for the follow-up of patients treated with angioplasty, but its role in the evaluation of stent patency is controversial. In the latter case, Color Doppler Ultrasonography, and CT Angiography are used to evaluate renal stent patency. As already mentioned, Color Doppler US is highly operator-dependent and sometimes yields inadequate results. Recent studies suggest an important potential role for US in noninvasive assessment of the immediate hemodynamic outcome and long-term follow-up after renal artery stenting. Comparing pre- and postprocedural US findings is valuable. CTA can achieve accurate assessment, particularly with thin-slice multislice scanning. In-stent stenosis is best evaluated with curved planar reconstructions or VR images. Subvolume VR images are particularly helpful, allowing direct visualization of the stent lumen without the

need of time-consuming segmentation. In addition, color- coding of different attenuating materials enables delineation of the stent lumen. Compared to angiography, the intrastent luminal diameter is underestimated in most patients who undergo CTA. Wide window settings improve evaluation of the stent lumen. When a significant restenosis is suspected on the basis of clinical findings, color doppler US or CTA, further investigation with DSA is performed. Reduction of 50% or more in stent lumen diameter is considered significant restenosis. On DSA, manometric gradient = 20 mm Hg indicates significant restenosis. References: 1. Soulez G, Oliva VL, Turpin S, Lambert R, Nicolet V, Therasse E. Imaging of Renovascular Hypertension: Respective values of Renal Scintigraphy, Renal Doppler US, and MR Angiography. Radiographics 2000; 20:13551368. 2. Martin LG, Runback JH, Sacks D, Cardella JF, et al. Quality Improvement Guidelines for Angiography, Angioplasty, and Stent Placement in the Diagnosis and Treatment of Renal Artery Stenosis in Adults. JVIR 2002; 13:1069-1083. 3. Taylor AT, Fletcher JW, Nally JV, et al. Procedure guidelines for diagnosis of renovascular hypertension. J Nucl Med 1998; 39:1297-1302. 4. Prigent A. The diagnosis of renovascular hypertension: the role of captopril renal scintigraphy and related issues. Eur J Nucl Med 1993; 20:625-644. 5. Helenon O, Rody F, Correas JM, Melki P et al. Color Doppler US of renovascular disease in native kidneys. Radiographics 1995; 15:833-854. 6. Ripolles T, Aliaga R, Morote V, Lonjedo E, Delgado F, Martinez M, Vilar J. Utility of intrarenal Doppler ultrasound in the diagnosis of renal artery stenosis. Eur J Radiol 2001; 40(1): 54-63. 7. Gottlieb RH, Snitzer EL, Hartley DF, et al. Interobserver and intraobserver variation in determining intrarenal parameters by Doppler sonography. AJR 1997; 168:627-631. 8. Oliva VL, Soulez G, Lesage D et al. Detection of renal artery stenosis with Doppler sonography before and after administration of captopril: value of early systolic rise. AJR 1998; 170:169-175. 9. Claudon M, Plouin PF, Baxter GM et al. Renal Arteries in Patients at Risk of Renal Arterial Stenosis: Multicenter Evaluation of the Echoenhancer SH U508A at Color and Specteral Doppler US. Radiology 2000; 214:739-746 10. Brink GD, Lim JT, Wang G et al. Technical optimization of Spiral CT for depiction of renal artery stenosis: in vitro analysis. Radiology 1995; 194:157-163 11. Rubin GD, Dake MD, Napel S, et al. Spiral CT of Renal Artery Stenosis: Comparison of Three-dimensional Rendering Techniques. Radiology 1994; 190:181-189. 12. Rubin GD. Spiral (Helical) CT of the renal vasculature. Seminars in Ultrasound, CT, and MRI 1996; 17:374-397 13. Kaatee R, Beek FJA, de Lange EE et al. Renal Artery Stenosis: Detection and Quantification with Spiral CT Angiography versus Optimized Digital Subtraction Angiography. Radiology 1997; 205:121-127 14. Kim TS, Chung JW, Park JH et al. Renal Artery Evaluation: Comparison of Spiral CT Angiography to Intra-arterial DSA. JVIR 1998;9:553559. 15. Wittenberg G, Kenn W, Tschammler A, Sandstede J, Hahn D. Spiral CT angiography of renal arteries: comparison with angiography. Eur. Radiol. 1999; 9:546-551. 16. Beregi J-P, Mauroy B, Willoteaux S, et al. Anatomic variation in the origin of the main renal arteries: spiral CT evaluation. Eur. Radiol. 1999; 9:1330-1334. 17. Beregi J-P, Louvegny S, Gautier C, et al. Fibromuscular Dysplasia of the Renal Arteries: Comparison of Helical CT Angiography and Arteriography. AJR 1999; 172:27-34. 18. Johnson PT , Halpern EJ, Kuszyk BS, et al. Renal Artery Stenosis: CT Angiography-Comparison of Real-time Volume-rendering and Maximum Intensity Projection Algorithms. Radiology 1999; 211:337-343. 19. Urban BA, Ratner LE, Fishman EK. Three-dimensional Volume-rendered CT Angiography of the Renal Arteries and Veins: Normal Anatomy, Variants, and Clinical Applications. Radiographics 2001; 21:373-386. 20. Dong Q, Shoenberg S, Carlos RC, et al. Diagnosis of Renal Vascular Disease with MR Angiography. Radiographics 1999; 19:1535-1554. 21. Leung DA, Hagspiel KD, Angle F, Spinosa DJ, Matsumoto AH, Butty S. MR Angiography of the renal arteries. Radiol Clin N Am 2002; 40(4): 847-865. 22. Fain SB, King BF, Breen JF et al. High-Spatial-Resolution Contrast-

enhanced MR Angiography of the Renal Arteries: A Prospective Comparison with Digital Subtraction Angiography. Radiology 2001; 218:481-490. 23. Shetty AN, Bis KG, Kirsch M, et al. Contrast-enhanced breath-hold three-dimensional magnetic resonance angiography in the evaluation of renal arteries: optimization of technique and pitfalls. JMRI 2000; 12:912-923. 24. De Cobelli F, Venturini M, Vanzulli A, et al. Renal Arterial Stenosis: Prospective Comparison of Color Doppler US and Breath-hold, Threedimensional Dynamic, Gadolinium-enhanced MR Angiography. Radiology 2000; 214:373-380. 25. Mallouhi A, Schocke M, Judmaier W, et al. 3D MR Angiography of Renal Arteries: Comparison of Volume Rendering and Maximum Intensity Projection Algorithms. Radiology 2002; 223:509-516. 26. Boudewijn G, Vasbinder C, Nelemans PJ, et al. Diagnostic Tests for Renal Artery Stenosis in Patients Suspected of Having Renovascular Hypertension: A Meta-Analysis. Ann Intern Med 2001; 135:401-411. 27. Blum U, Krumme B, FlÜge P, et al. Treatment of Ostial Renal-Artery Stenoses with Vascular Endoprostheses after Unsuccessful Balloon Angioplasty. NEJM 1997; 336:459-465. 28. Hilfiker PR, Quick HH, Debatin JF. Plain and Covered Stent-Grafts. In Vitro Evaluation of Characteristics at Three-dimensional MR Angiography. Radiology 1999; 211:693-697. 29. Sharafuddin MJA, Raboi CA, Abu-Yousef M, et al. Renal Artery Stenosis: Duplex US after Angioplasty and Stent Placement. Radiology 2001; 220:168-173 30. Beharm JV, Nelson RC, Zidar JP, DeLong DM, Smith TP. Thin-Section Multidetector CT Angiography of Renal Artery Stents. AJR 2002; 178:1155-1159.

31.3 Complications of renal artery PTA/stenting J. H. Peregrin; ZRIR -Radiology, IKEM, Prague, Prague 4, Czech Republic. Reports on complications of endovascular renal artery revascularization in the literature are not comparable and the rate of complications varies from 0 % - 83 %. The rate of complications reported by different authors depends on definition of the complications, selection of the patients and on how meticulously the records of patient’s follow-up are kept. To make the complication reports more comparable a definition of complications suggested by Beek has recently been used (1): a) Complications with considerable clinical significance (SCS) prolonging hospitalisation for more than 2 days or increasing the number of outpatient’s visits These include: renal artery occlusion/injury requiring nephrectomy, cholesterol spray (to renal or to peripheral arteries), deterioration of renal function with temporary or permanent haemodialysis, severe allegro-toxic reactions, severe bleeding at access site requiring transfusion and/or surgical repair, renal artery perforation/rupture, and others. The incidence of SCS reported in literature is shown in table. Reported one month mortality rate is from 0 % (2) to 5 % (3), the metaanalysis of 24 articles performed by Leertower et all (4) indicates a one-month mortality rate below 1 %. Secondary nephrectomy either surgical or by embolization was necessary in less than 1 %. Surgical salvage operation was performed in 2.5 % according to Martin (5), but in later reported metaanalysis it was less than 1 % (4) (the lower incidence of rescue surgery is most probably due to treatment of occlusive renal artery dissection by stent implantation). b) Complications with minor clinical significance (MCS) prolonging hospitalisation for 2 days or less (usually minor groin bleeding/haematoma, minor to mild allergo-toxic reaction). The incidence of MCSs varies from 0 % - 21 % (table). c) Technical complications (RTC): events, that occurred during procedure and had no clinical consequence, but led to an increase of procedural time and/or cost (stent misplacement, balloon rupture, dissection requiring additional stent implantation). They could sometimes (but rarely) have resulted in complications of clinical significance (e.g. peripheral stent embolization causing lower limb artery thrombosis). The incidence of RTCs is reported from 0 % to 83 % (6), but the true average incidence is not known, as some authors do not report these complications at all.

First author Year

N (patients/arteries)

SCS (%)

Rees 1991 (8)

28/28

10

Joffre 1992 (2)

16/16

0

0

Weibull 1993 (6)

29/29

17

11

Erdoes 1996 (3)

16/18

33

Iannone 1996 (9)

63/83

35

Boisclar 1997 (10)

33/35

21

Beek 1997 (1)

50/61

10

Taylor 1997 (11)

29/32

3

21

White 1997 (12)

100/133

1

9

Rundback 1998 (13)

45/54

4.4

Dorros 1998 (14)

163/202

2

10

Fiala 1998 (15)

21/25

14

10

Van de Ven 1999 (16)

85/85

41

21

Bakker 1999 (17)

106/120

6

8

Paulsen 1999 (18)

135/233

Martin et al. (5,7) demonstrated a trend towards reduced complications rate with increasing experience of the operator and improved technology and devices. To summarize the literature review, the mean incidence of complications after endovascular renal revascularization procedures is 5 % – 20 %. Majority of these complications are not life threatening and do not result in loss of renal function, but they are dependant on patients selection, operator’s experience and device selection. Even if the incidence of SCSs is low and the loss of life directly related to renal artery intervention is rare, each case should be evaluated individually and the anticipated procedure must outweigh the risks involved. References: 1. Beek FJA, Kaatee R, Beutler JJ, van der Ven PJ, Mali WPTM: Complications during renal artery stent placement for atherosclerotic ostial stenosis. Cardiovasc Intervent Radiol (1997) 20: 184-190. 2. Joffre F, Rousseau H, Bernadet P, Nomblot Ch, Montoy JC, Chemali R, Knight Ch: Midterm results of renal artery stenting. Cardiovasc Intervent Radiol (1992) 15: 313-318. 3. Erdoes LS, Berman SS, Hunter GC, Mills JL: Comparative analysis of percutaneous transluminal angioplasty and operation for renal revascularization. American Journal of Kidney Diseases (1996) 4: 496-503. 4. Leertouwer TC, Gussenhoven EJ, Bosch JL, van Jaarsveld BC, van Dijk LC, Deinum J, Man in ´t Veld AJ: Stent placement for renal arterial stenosis: where do we stand? A meta-analysis. Radiology (2000) 216: 78-85. 5. Martin LG: Renal revascularization using percutaneous balloon angioplasty for fibromuscular dysplasia and atherosclerotic disease. In: Calligaro KD, Doughtery MJ, eds. Modern Management of Renovascular Hypertension and Renal Salvage. 1st ed. Baltimore, MD: Williams & Wilkins, 1996; 125-144. 6. Weibull H, Bergqvist D, Bergentz SE, Jonsson K, Hulthén L, Manhem P: Percutaneous transluminal renal angioplasty versus surgical reconstruction of atherosclerotic renal artery stenosis: A prospective randomized study. J Vasc Surg (1993) 18: 841-52. 7. Martin LG, Casarella WJ, Alspaugh JP, Chuang VP: Renal artery angioplasty: increased technical success and decreased complications in the second 100 patients. Radiology (1986) 159: 631-644.

MCS (%)

13 8. Rees CHR, Palmaz JC, Becker GJ, Ehrman KO, Richter GM, Noeldge G, Katzen BT, Dake MD, Schwarten DE: Palmaz stent in atherosclerotic stenoses involving the ostia of the renal arteries: Preliminary report of a multicenter study. Radiology (1991) 181: 507-514. 9. Iannone LA, Underwood PL, Nath A, Tannenbaum MA, Ghali MGH, Clevenger LD: Effect of primary balloon expandable renal artery stents on long-term patency, renal function, and blood pressure in hypertensive and renal insufficient patients with renal artery stenosis. Cathet Cardiovasc Diagn (1996) 37: 243-250. 10. Boisclair Ch, Therasse E, Oliva VL, Soulez G, Bui BT, Quérin S, Robillard P: Treatment of renal angioplasty failure by percutaneous renal artery stenting with Palmaz stents: Midterm technical and clinical results. AJR (1997) 167: 245-251. 11. Taylor A, Sheppard D, Macleod MJ, Harden P, Baxter GM, Edwards RD, Moos JG: Renal artery stent placement in renal artery stenosis: technical and early clinical results. Clin Radiol (1997) 52: 451-7. 12. White CJ, Ramee SR, Collins TJ, Jenkins JS, Escobar A, Shaw D: Renal artery stent placement: utility in lesions difficult to treat with balloon angioplasty. J Am Coll Cardiol (1997) 30: 1445-50. 13. Rundback JH, Gray RJ, Rozenblit G, Poplausky MR, Babu S, Shah P, Butt K, Tomasula J, Garrick R, Goodman A, Dolmatch B, Horton K: Renal artery stent placement for the management of ischemic nephropathy. JVIR (1998) 9: 413-420. 14. Dorros G, Jaff M, Mathiak L, Dorros II, Lowe A, Murphy K, He T: Fouryear follow-up of Palmaz-Schatz stent revascularization as treatment for atherosclerotic renal artery stenosis. Circulation (1998) 98: 642-7. 15. Fiala LA, Jackson MR, Gillespie DL, O'Donnell SD, Lukens M, Gorman P: Primary stenting of atherosclerotic renal artery ostial stenosis. Ann Vasc Surg (1998) 12: 128-33. 16. Van de Ven PJG, Kaatee R, Beutler JJ, Beek FJA, Woittiez AJJ, Buskens E, Koomans HA, Mali WPTh: Arterial stenting and balloon angioplasty in ostial atherosclerotic renovascular disease: a randomised trial. Lancet (1999) 353: 282-86. 17. Bakker J, Goffette PP, Henry M, Mali WPTM, Melki JP, Moss JG, Rabbia C, Therasse E, Thomson KR, Thurnher S, Vignali C: The Erasme study: a multicenter study on the safety and technical results of the Palmaz stent used for the treatment of atherosclerotic ostial renal artery stenosis. Cardiovasc Intervent Radiol (1999) 22: 468-474. 18. Paulsen D, Klow NE, Rogstad B, Leivestad T, Lien B, Vatne K, Fauchald

P: Preservation of renal function by percutaneous transluminal angioplasty in ischaemic renal disease. Nephrol Dial Transplant (1999) 14: 1454-1461.

Special Session AV Shunts 32.2 Interventional radiology in functional and failing autogenous fistulae L. A. Turmel-Rodrigues; Radiology, Clinique St-Gatien, Tours, France. All the published series concerning percutaneous treatment of thrombosed autogenous fistulae report that an underlying tight stenosis is unmasked in close to 100% of cases, which means that preventive treatment of such stenoses could have avoided acute thrombosis. However, we also know that autogenous fistulae are much less prone to thrombosis than prosthetic grafts and that not all stenoses threaten access patency. The challenge is to detect stenoses that will worsen and to tolerate stenoses at rest. Flow rate monitoring appears to be one the best, if not the best, tool to detect active stenoses whose reduced luminal diameter decreases fistula flow proportionally. Some series have already demonstrated a clear reduction in the thrombosis rate of monitored autogenous fistulae. Suggested thresholds are a decrease of 20% at 2 monthly examinations in fistulas with a flow rate below 1L/min [1]. This means that a fistula with a flow rate of 400mL/min that does not change with time probably suffers from a stenosis somewhere but that this stenosis is steady, does not threaten patency, and must not be treated as long as adequate dialysis is possible. Stenosis detection programmes with the aim of prevention of thrombosis of autogenous fistulae should therefore be initiated in all dialysis centres. According to the study of Mc Carley they are cost-effective [2]. Recent American multidisciplinary guidelines suggest classifying stenosed autogenous fistulas into 2 groups: functional and failing. Functional fistulas are fistulas allowing for normal dialysis sessions at the usual 300 to 400 mL/min flow rate, and failing fistulae are those whose flow-rates, below 350 mL/min, do not allow normal extra-renal epuration [3]. INDICATIONS FOR IMAGING Apart from significant decrease in routinely monitored flow-rates, any clinical abnormality requires imaging with a view to its correction: difficulties in cannulation, vacuum phenomenon during dialysis, pain in the hand, increased venous pressure, development of collaterals, increased final compression times, delayed healing of previous needle holes with formation of a black scar, arm oedema… The place of non-invasive ultrasound examinations prior to invasive angiography is still not clear. Their goal is to avoid unnecessary angiograms but they must not delay treatment. There are only 2 clear indications: immature fistulae and insufficient inflow in forearm fistulae. In cases of delayed maturation, echography rules out the specific problem of deep location of the vein, which is treated by surgical transposition into a more superficial location. In cases of insufficient inflow in forearm fistulae, echography can confirm an isolated stenosis located in the anastomotic area, which is better treated by surgical creation of a new anastomosis. Angiography is of little value in these 2 clinical situations which indicate surgery. In all other cases, only angiography evaluating the whole limb from the subclavian artery to the hand and from the hand to the superior vena cava is able to diagnose all the problems and to offer concomitant treatment in the vast majority of cases. Prior ultrasound examinations are not harmful but they can be a waste of time and money. Our experience indicates that ultrasound examinations are used particularly (overused?) in institutions where interventional radiology is not available. DIAGNOSTIC ANGIOGRAPHY Technique Diagnostic angiography is performed within a framework of a concomitant interventional procedure and normal angiograms should be rare. Gadolinium can be used as the contrast medium in cases of absolute contraindication for iodine injection. In patients in whom dialysis has not yet been initiated iodine diluted to 90% can be used with little risk of nephrotoxicity for a much lower cost than gadolinium-based con-

trast media [4-5]. The angiography technique depends on the clinical situation. In cases of poor inflow (the most frequent situation in forearm fistulae), retrograde puncture of the brachial artery at the elbow with an 18-20 G needle is the most physiologically appropriate approach to visualize the arterio-venous anastomosis and to avoid misinterpretation of the quality of arterial feed to the fistula. In addition, reflux to the axilla is easily achieved and provides evidence of high bifurcations of the brachial artery (20% of patients). This retrograde approach also makes it possible to push a 4F catheter up to the aortic arch to rule out stenosis of the subclavian or axillary artery when no peripheral stenosis is unmasked to explain low fistula flow or hand ischaemia. This elbow artery route is the most common approach for the exploration of delayed maturation in both forearm and upper arm fistulae, a specific case where inflow and venous cannulation problems predominate [6]. One hour’s bedrest and arm rest are required after brachial artery needle removal before discharging outpatients. In cases of outflow obstruction, antegrade puncture of the vein is performed just after the anastomosis. This is usually easy since this venous segment is under tension. In cases of high outflow or hand ischaemia, complete visualization of the arteries of the limb is required to evaluate the possibility of fistula flow reduction or peripheral flow improvement. Retrograde puncture of the brachial artery is usually sufficient and the femoral route is rarely necessary. In cases of distal ischemia, acquisitions focusing on the forearm and hand must be performed before and during compression of the fistula in order to evaluate the respective roles of steal and intrinsic arterial lesions in the process. DILATATION Indications Stenosis should be treated only if it explains clinical impairment or if it threatens access patency because of decreasing flow (DOQI guideline 3) [7]. For example, apparent stenoses of the anastomosis of autogenous fistulae must be dilated only in cases of insufficient flow, because dilation of a well-functioning fistula could lead to secondary high flow or steal phenomenon. Similarly, subclavian and brachio-cephalic stenoses are rarely responsible for fistula thrombosis and must be dilated only in cases of clinically impairing upper limb oedema. Regular dilation technique Dilation is an outpatient procedure [8]. Mature hemodialysis fistulae are made to be punctured and must be used for introduction of dilation catheters. Clinical examination is essential to choose the best cannulation site. Puncture of the vein for dilation must be performed under local anesthesia in the direction of but far enough from (3 cm) the presumed stenosis. An antegrade approach should be used for stenoses located far enough from the anastomosis, and a retrograde approach should be used for stenoses close to the anastomosis. A tourniquet often helps this initial catheterization stage. A 16 to 18G needle and a 0.035 inch guide wire offer good support for placement of a 6 to 9 F introducer-sheath. A small subcutaneous tunnel between the skin entry-point and the vein entry-point will facilitate the final compression and decrease the risk of pseudoaneurysm. A soft-tip "Bentson-type" guide wire is preferred to traverse stenoses. Angled catheters are helpful for selective catheterization of branches. Hydrophilic guide-wires are very effective but must be used carefully because of their propensity to dissect vessel walls in stenosed areas. Heparin may be injected (2,000 to 3,000 units) once a guide wire has passed through the stenosis but our experience indicates that it is only necessary in small diameter or low flow fistulae. In typical cases (but there are exceptions, see below), the diameter of the dilatation balloon should be equal to or 1 millimeter greater than the diameter of the immediately upstream or downstream normal vessel. Angiography of the stenosed area should therefore be performed with a graduated wire through stenosis or with the superimposition of a lead millimeter ruler. A fistula of less than 3 months must be dilated at least to 5 mm, a fistula of more than 1 year must be dilated at least to 7 mm. The dilatation balloon is inflated with a manometer filled with diluted contrast medium. Pressure is slowly increased to abolish the waist of the stenosis on the balloon, the edges of which must be completely parallel. No residual notch should be accepted. The inflated balloon is left in place for 1 to 3 minutes. Dilation is often painful locally. Local anesthesia can be administered when stenoses are just under the skin. Venous stenoses are often very hard and high pressure balloons with

bursting pressures over 25 atmospheres must be used ( "Blue-Max", Medi-Tech®, Natick, MA / "Centurion", Bard®, Covington, GA / “Extreme”, Cordis®, Miami, FL). Immediate post-dilation angiography, with the guide wire left in place through the dilated area, may show several possibilities: 1) No residual stenosis and no parietal damage: the procedure is then complete. 2) Minor parietal damage: 3 to 5 minutes' low pressure ballooning is performed to try to smooth the vessel wall. 3) Rupture with clear extravasation of contrast medium and creation of a hematoma: the dilation balloon must be rapidly reinflated to 2 atmospheres for repeated periods of 10 minutes. 4) Residual stenosis. If there is no parietal damage, a new dilation is performed with a balloon which is 1 mm greater in diameter. Less than 30% residual stenosis is acceptable at this stage only if it results from the first dilation ever performed for this stenosis. If there is greater than 30% residual stenosis, new dilatation should be performed with a larger balloon but more than 2 mm overdilatation is not recommended at the time of the first intervention. Once the dilatation has been performed, the catheters and introducers are removed. The puncture site must be compressed as gently as possible to stop bleeding without stopping flow through the fistula. We tend to use systematically the technique of suture with a minitourniquet [9]. The outpatient can then get up and the vascular access is immediately usable for hemodialysis. Specific typical cases Delayed maturation is always explained by an underlying stenosis which is diagnosed in 100% of cases if the fistula is appropriately evaluated by puncture of the brachial artery. Our experience indicates therefore that the best treatment is surgical or radiological correction of the stenosis, and not ligation or embolization of collaterals [6]. Even long stenoses on the artery or on the vein must be dilated, since this is the only way to save the fistula and to preserve the patient’s venous reserve. Stenoses resistant to 25 bar pressures are infrequent. In our experience, they occur in 1% of forearm fistulae and 5% of upper arm fistulae (final arch of the cephalic vein) [10]. They can be treated with expensive cutting balloons, although the slightly rudimentary but cheap technique of multiple direct percutaneous puncture of the area of fibrotic stenosis frequently works. However, it seems that this problem should soon be a problem of the past since a new very high pressure balloon inflatable up to 40 atm is now available on the American market and should be launched on the European market in the near future ("Conquest", Bard®, Covington, GA). Stenoses of the feeding artery and of the arterio-venous anastomosis may be impossible to traverse using a regular retrograde fistula approach. Antegrade puncture of the brachial artery and catheterization of the feeding artery is feasible but with a greater risk of local complications at the puncture site. This is the reason why, unlike Maninnen, we recommend avoiding this brachial artery route whenever possible [11-12]. Dilatation of anastomotic venous stenosis in the forearm can necessitate placing the dilation balloon across the anastomosis. The balloon diameter must therefore be limited to 1mm greater than the diameter of the artery, which means that residual stenosis will be left on the vein. Due to the difference in compliance between arterial and venous walls, a vein must be larger than an artery to allow the same flow-rate (for example, compare the diameter of the ascending aorta with the diameter of the 2 venae cavae on CT scan). The diameter of the vein of a fistula should always therefore be larger than its proximal feeding artery and even more if the distal artery has retrograde filling and also contributes to feeding the fistula. This goal cannot be achieved when the dilation balloon is placed across the anastomosis. This is one of the arguments for surgical revision of such stenoses. Manninen has reported that interventional radiology in this area provided mid-term disappointing results (20% primary patency rate at 1 year versus 52% in other locations)[11. In contrast Oakes reported a 79% 1-year primary patency rate after surgical revision of such stenoses [13].The problem is rare with upper arm fistulae since the most proximal feeding arteries have a greater diameter. Distal ischemia Dilatation has no place if the arteries are normal and if distal ischemia is simply due to the steal of the fistula. In contrast, dilatation of an arterial stenosis above or below the arterio-venous anastomosis can significantly improve distal flow and cure the patient [14]. Non typical cases Underdilatation and subsequent residual stenosis after dilatation should be the rule in 2 non-typical situations:

Precarious arterial supply of the hand is particularly frequent in diabetic and elderly patients. If the arteries feeding the hand are partially destroyed or stenosed, full dilatation of a stenosis located on the arterialized vein might increase fistula flow so much that blood can become stolen from the distal arteries, which can sever a previous fragile balance between fistula and hand arterial supplies. The risk is much greater with elbow fistulae. In such cases it may be advisable to dilate venous stenoses cautiously only to 5 mm, especially when they are located in the anastomotic area. The ideal is to be able to measure fistula flow rate during procedures and to stop dilatation when fistula flow reaches 1L/min [15]. Concomitant asymptomatic central vein stenoses can become symptomatic if full dilatation of a peripheral stenosis increases fistula flow so much that collaterals are no longer able to divert the excess flow, resulting in handicapping and painful arm oedema. As advised above, the ideal is to undersize the first dilatation balloon, to monitor fistula flow during dilatation and to stop when fistula flow reaches 1L/min. It might of course be suggested that the central vein stenosis should have beeen dilated concomitantly. Unfortunately, we know that the immediate results of central vein dilatation are frequently disappointing, necessitating stent placement and often needing 3-monthly redilatation thereafter. The less we treat central vein stenosis, the better it is for the long term functional patency of fistulae. Complications of dilatation Rupture The patient often reports a sudden acute pain at the dilated site as soon as the balloon is deflated. If rupture is confirmed, the balloon must be reinflated quickly to 2 atmospheres and left in place for 10 minutes. If previously injected, heparin must be neutralized by injection of protamin (except in patients under insulin) and the hematoma can be compressed manually if clinically accessible. If extravasation continues in spite of 3 periods of 10 minute balloon-tamponade, a stent must be placed [16-17]. In our experience, rupture during dilation occurred in 8% of forearm fistulae and 15% of upper arm fistulae [10]. However, the majority were controlled by prolonged balloon inflation and very few required a stent. The final arch of the cephalic vein, the end of the transposition of brachio-basilic fistulae and the post-anastomotic area of forearm fistulae are the most prone to rupture. Pseudoaneurysms can be the consequence of a dilation-induced rupture which was initially controlled but which reopened and was limited in its expansion. Such aneurysms can be treated by placement of a stent through the base except in cannulation areas where surgery is preferable. Bleeding As after a dialysis session, the cannulation site can subsequently reopen, and control of such bleeding by manual compression is a part of the basic education of hemodialyzed patients. Infection is a potentially fatal complication and it has been reported in all series because hemodialyzed patients have immunological disorders. Systematic treatment with antistaphylococcal antibiotics is required when a difficult dilation procedure has been prolonged, when several cannulations have been necessary, clots are present, the fistula has recently been created and above all in cases of breach in sterile technique,. Contraindications to dilation Absolute contraindications Local infection, which is much rarer than in grafts, contraindicates any endovascular procedure. Concomitant steal syndrome is an absolute contraindication for treatment of any stenosis located on the vein (downstream from the anastomosis) since it would increase flow in the fistula and subsequent steal. Relative contraindications High flow is mainly a complication of upper arm fistulae and is a contraindication for dilatation because treatment of the stenosis would increase the already high flow. When no flow-rate measurement has been provided by the referring nephrologists, high flow can be suspected when the vein is tortuous and aneurismal. These high flow elbow fistulae are often associated with mild central vein vein stenosis that are responsible for increased venous pressure and longer compression times after dialysis. These mild stenoses must not be dilated when flow rate is above 1.5 L/min since dilatation might increased flow above 2L and cause either hand ischaemia or mid-term cardiac failure. Anastomotic stenoses in fistulae of less than 6 weeks must be dilated with extreme caution because of the high risk of rupture. Isolated stenosis within 10 cm of the wrist in forearm fistulae can be dilated but surgical creation of a new anastomosis above the stenosis is simple, minimally invasive and proven to be more durable. However, dilation remains the best treatment when the radial artery is occluded

above the anastomosis, the fistula being fed by the ulnar artery via the palmar arch and retrograde flow from the distal radial artery. STENTS Indications for stent placement are extremely unusual in forearm fistulae where rupture is the only valuable indication. Elastic recoil in forearm veins always indicates insufficient balloon sizing. Indications are much more frequent in the upper arm and in central veins. Choice of the stent The anatomic and mechanical constraints of hemodialysis fistulae mean that flexible self-expandable stents (Wallstent type) are preferable to balloon expandable and rigid stents. When a stent has to be placed in a cannulation area, which is a rare indication, a wide mesh puncturable stent such as the "Symphony*" (Medi-tech ®) or the “ZA” (Cook®) must be used. The disappointing results to date for covered stents lead to recommending them only for rupture control or elimination of aneurysms in non-cannulation areas. Think of the future Stents must not obviate further surgery and must protect the ease of making new anastomoses. For example, stent placement is contraindicated at the elbow if the stent would overlap the basilic vein and obviate future creation of a transposed brachio-basilic fistula, and a stent placed in the final arch of the cephalic vein must not protrude into the subclavian vein where it could lead to stenosis, which would preclude future use of the basilic and axillary veins for a direct fistula or for drainage of an upper arm graft [18]. Indications Acute rupture is an emergency for which a stent can save the fistula. Covered stents are effective in 100% of cases but bare Wallstents frequently work [16-17]. Stent placement could also prevent the development of a secondary pseudoaneurysm. Aneurysms can be eliminated by placement of a (preferably covered ) stent through the neck or base. The indication is valuable for aneurysms located in non-cannulation areas where surgery may be difficult. Surgery is preferable in cannulation areas. Soft post-PTA recoil . Recoil needs a stent when the residual stenosis is greater than 30%. It is important to have dilated enough before placing the stent, i.e. for at least 5 minutes with a slightly oversized balloon. Significant recoil is extremely unusual in forearm veins and is due in the vast majority of cases to an undersized balloon. There are stenoses whose waisting on the dilation balloon is not abolished in spite of the highest pressure applicable with the strongest balloons (25 atm). Stents are not a logical indication in these cases, in which are indications for more mechanically aggressive tools, i.e. multiple direct percutaneous punctures, atherectomy device or cutting balloon, or surgery. Early recurring stenoses (<4 months). There is no consensus regarding the indications for and value of stents to delay restenosis. Follow-up There is no need for anticoagulants or antiplatelet drugs before, during or after stenting of a fistula. There is a high risk of restenosis, although delayed, either in-stent or at its extremities. RESULTS There are only 2 studies in the international literature about dilation of autogenous fistulae with a sufficiently high number of patients to reach some statistical significance [10-11]. Primary patency rates after dilation of forearm fistulae are very close in both studies, with 44% and 51% at 1 year. We have emphasized in our study that forearm fistulae fare significantly better than grafts since in our experience they have provided a 51% rate versus 29% for grafts at 1 year and 37% versus 13% at 2 years. On the other hand, upper arm fistulae have provided intermediate results, with 35% at 1 year and 24% at 2 years [10]. Compared to the forearm, these poorer results in upper arm fistulae are due to difficulties in the management of stenoses of the final arch of the cephalic vein (rupture or resistance to dilation, early recurrence). In all groups, significant morbidity was low (2-4%) and the mortality of 2% was not attributable to the radiological procedures. Clark reported a low 26% primary patency rate at 1 year in a cohort of fistulae located in the forearm (n=37) and in the upper arm (n=28) and emphasized the detrimental role of lesion length and of comorbidities (diabetes, coronary or peripheral vascular disease)[19]. A Dutch study reports a low 50% primary patency rate at 6 months but the location of the 33 fistulas is not provided [20]. Our opinion is that stenoses are underdilated in these 2 series, which might explain these poorer primary patency rates. Primary patency rates after stent placement

Primary patency rates after stent placement in our experience were 36% and 20% at 1 year in the forearm and in the upper arm respectively, which is low. The beneficial effect of stents is clearer when intervals between maintenance reinterventions are compared before and after stent placement. In our study the mean interval approximately doubled from 4.1 to 9.7 months in forearm fistulae and from 2.26 to 4.22 months in the upper arm (p<0.05 in both groups) [10]. Manninen reports similar findings, with a mean interval increasing from 3.4 to 6.3 months after stent placement in the forearm [11]. Secondary patency rates after dilation and stent placement were exactly the same 85% for forearm fistulae at 1 year in Manninen’s and in our study. It was necessary in our experience to reintervene only every 18 months after the initial dilation to achieve this rate of 85% at 1 year and then 77% at 4 years. During the same period, the rates for grafts were 92% at 1 year and 60% at 4 years, with reinterventions every 10 months. Rates for upper arm fistulae were 82% at 1 year and 51% at 4 years, with reinterventions every 11 months. Upper arm fistulae require significantly more frequent reinterventions than forearm fistulae and do not fare better than grafts after dilation. Finally, we recently confirmed the influence of the age of the fistula at the time of the first dilation on its outcome after dilation. The interval between maintenance reinterventions is significantly shorter in forearm fistulae of less than 1 year compared with those of more than 1 year (16 vs 30 months in our experience). The younger the access, the faster the restenosis. CONCLUSION Treatment of stenosis in autogenous fistulae by interventional radiology achieves excellent long-term secondary patency rates, which does not mean that all stenoses are good indications for dilatation. Multidisciplinary cooperation with nephrologists and surgeons is essential to determine the best therapeutic option according to stenosis location, patient history and life expectancy [21]. It can be said in 2003 that there is an unacceptable loss of opportunity for patients who are dialyzed in centers not working in close cooperation with an interventional radiologist. References: 1-Schwab S, Oliver M, Suhocki P, McCann R. Hemodialysis arteriovenous access: detection of stenosis and response to treatment by vascular access blood flow. Kidney Int 2001; 5:358-362 2-Mc Carley P, Wingard R, Shir Y, Pettus, Hakim R, Ikizler T. Vascular access blood flow monitoring reduces access morbidity and costs. Kidney Int 2001;60:1164-1172 3-Sidawy A, Gray R, Besarab A, Henry M, Ascher E, Silva M, Miller A, Scher L, Trerotola S, Gregory R, Rutherford R, Kent C, Recommended standards for reports dealing with arteriovenous hemodialysis accesses. J Vasc Surg 2002;35:603-610 4- Spinosa D, Angle F, Hagspiel K, Schenk W, Matsumoto A. CO2 and gadopentetate. Kidney Int 1998; 54:945-950 5-Nyman U, Elmstahl B, Leander P, Nilsson M, Golman K, Almen T. Are gadolinium-based contrast media really safer than iodinated media for digital subtraction angiography in patients with azotemia? Radiology 2002;223:311-318 6-Turmel-Rodrigues, Mouton A, Birmelé B et al. Salvage of immature forearm fistulas for haemodialysis by interventional radiology. Nephrol Dial Transplant 2001;16:2365-2371 7-Schwab S, Besarab A, Beathard G, Brouwer D, Etheredge E, Hartigan M, Levine M, McCann R, Sherman R, Trerotola S. NKF-DOQI Clinical Practice Guidelines for Vascular Access. Am J Kidney Dis 1997; 30(Suppl 4):S150-189. 8- Turmel-Rodrigues L, Pengloan J, Blanchier et al. Insufficient dialysis shunts: improved long-term patency rates with close hemodynamic monitoring, repeated percutaneous balloon angioplasty, and stent placement Radiology 1993; 187:273-278 9- Zaleski G, Funaki B, Gentile L, Garofalo R Purse-string sutures and miniature tourniquet to achieve immediate hemostasis of percutaneous grafts and fistulas. Am J Radiol 2000;175:1643-1645 10- Turmel-Rodrigues L, Pengloan J, Baudin S et al. Treatment of stenosis and thrombosis in haemodialysis fistulas and grafts by interventional radiology. Nephrol Dial Transplant 2000; 15:2029-2036 11- Manninen H, Kaukanen E, Ikaheimo R, et al, Endovascular treatment of failing Brescia-Cimino hemodialysis fistulae by brachial artery

access: Initial success and long-term results. Radiology 2001;218:711-718. 12- Trerotola S, Turmel-Rodrigues L. Off the beaten path: transbrachial approach for native fistula interventions. Radiology 2001;218:617-619 13-Oakes D, Sherck J, Cobb L. Surgical salvage of failed radiocephalic arteriovenous fistulae: techniques and results in 29 patients. Kidney Int 1998; 53:480-487 14-Valji K, Hye R, Roberts A, Oglevie S, Ziegler T, Bookstein J. Hand ischemia in patients with hemodialysis access grafts: angiographic diagnosis and treatment. Radiology 1995; 196:697-701 15- Ahya S, Windus D, Vesely T. Flow in hemodialysis grafts after angioplasty: do radiologic criteria predict success? Kidney Int 2001;59:19741978 16- Sapoval M, Turmel-Rodrigues L, Raynaud A, Bourquelot P, Rodrigue H, Gaux JC. Cragg covered stents in hemodialysis access: initial and mid-term results. J Vasc Interv Radiol 1996; 7:335-342 17- Raynaud A, Angel C, Sapoval M, Beyssen B, Pagny J, Auguste M. Treatment of hemodialysis access rupture during PTA with Wallstent implantation. J Vasc Interv Radiol 1998; 9:437-442 18-Turmel-Rodrigues L, Bourquelot P, Raynaud A, Sapoval M. Primary stent placement in hemodialysis-related central venous stenoses: the dangers of a potential "Radiologic dictatorship". Radiology 2000; 217:600-602 19-Clark T, Hirsch D, Jindal K, Veugelers P, Le Blanc J. Outcome and prognostic factors of restenosis after percutaneous treatment of native hemodialysis fistulas. J Vasc Interv Radiol 2002; 13:51-59 20- Van der Linden J, Smits J, Assink J, Wolterbeek D, Zijlstra J, de Jong G, Van der Dorpel M, Blankestijn P. Short- and long-term functional effects of percutaneous transluminal angioplasty in hemodialysis vascular access. J Am Soc Nephrol 2002; 13:715-720 21-Turmel-Rodrigues L, Pengloan J, Bourquelot P. Interventional radiology in hemodialysis fistulae and grafts: a multidisciplinary approach. Cardiovasc Intervent Radiol 2002; 25:3-16

32.4 Mechanical Devices For Hemodialysis Graft and Fistula Thrombolysis S. Trerotola; Hospital of the University of Pennsylvania, Philadelphia, United States. Mechanical thrombolysis of hemodialysis grafts has been developed to address a number of significant problems associated with chemical thrombolysis. These problems include bleeding complications due to the multiple puncture sites in these grafts, prolonged procedure times, the cost and risk of the thrombolytic agent, and residual clot in the graft. The basic tenets of mechanical thrombolysis are: 1) crossed catheter access, 2) use of some mechanical device or technique to macerate or pulverize the clot, 3) embolization and/or aspiration of the resulting particulate and 4) treatment of the underlying cause of graft thrombosis, usually venous outflow stenosis. Mechanical thrombolysis can be performed in native fistulae as well as grafts, with immediate results that are slightly worse than grafts but with much better long-term results. The most significant impediments to restoration of flow in a dialysis graft or fistula are the underlying stenosis, usually at the venous outflow, and the arterial plug (discussed below). In order to understand mechanical thrombolysis and its importance relative to chemical techniques, one must understand the composition of clot in a thrombosed graft. The solid clot volume in an average thrombosed forearm loop dialysis graft was approximately 3 cc, and never more than one teaspoon (5 cc) in our series (1). This clot is predominantly soft red thrombus which fragments and/or lyses relatively easily. A small portion of the clot, which invariably rests at the arterial anastomosis, consists of firm, compacted, mature white thrombus characterized histologically by a dense onion-skin appearance consisting of alternating layers of red cells and platelets/fibrin. This “arterial plug” has a characteristic appearance when removed surgically, consisting of a cylindrical piece of white clot with a concave surface facing the artery, leading many surgeons to describe it as the “bullet” . Removal of the arterial plug is considered by surgeons the sine qua non of successful thrombectomy and it is documented that failure to remove this plug results in decreased patency after thrombectomy (2). The arterial plug is well documented to be extremely resistant to chemical thrombolysis, which was previously thought to be due to platelet-richness of the clot

(3,4), rather than dense compaction. Since the thrombolytics used in chemical lysis do not affect the arterial plug nor the venous outflow stenosis, and these are the two most important determinants of success of the procedure, it is not surprising that mechanical techniques have been found to be equivalent: this concept forms the basis for mechanical thrombolysis. Mechanical devices There are now nine devices with US FDA approval for lysis of hemodialysis grafts, and two for lysis of fistulae. While most appear to have similar acute and long-term results, evidence now suggests that at least some devices may not offer acceptable long-term results in dialysis grafts. This information is in stark contrast to the long-held hypothesis that the results of thrombolysis were technique-independent. It is the author’s belief that the failure of some devices to achieve acceptable long-term results is their inability to completely remove the clot from the graft wall. There are data in the surgical literature to support this concept: Schmitz-Rode et al performed an in vitro study of four different devices designed for mechanical thrombolysis in hemodialysis access (Trac, ultrasound-oscillating probe, electric oscillating probe, rotating spiral aspiration thrombectomy) and showed that residual adherent clot was a significant problem for all devices tested (5). Holzenbein et al performed angioscopy during surgical revision and found that detection and treatment of residual abnormalities including wall adherent thrombus and pseudointima resulted in improved 30 day patency (66% vs 33%) (6). Puckett et al used an endometrial curette during surgical revision to remove pseudointima and showed a 30 day patency of 97% (vs 71% when not used) and 6 month patency 65% (vs 36%) (7), again showing that removal of wall-adherent material (clot and pseudointima) is not only desirable but essential fro improved patency. Those devices for which lower patency results have been reported can all be characterized by being non-wall contact devices, and the above literature may help explain the poorer results achieved with these devices. It should be borne in mind that the DOQI recommendations (8) for patency after percutaneous declotting are 85% immediate success and 40% 3 month patency. In the order of FDA approval, the available devices are: The Amplatz mechanical thrombectomy device (Microvena, White Bear Lake, MN). It is approved by for use in hemodialysis access, including native fistulae, but it should be noted the US clinical trial leading to the device’s approval did not include any native fistulae. The version currently available for sale is the 7F “Helix” device, consists of a gasdriven high-speed (>100,000 rpm) cam which creates a vortex which pulverizes the clot. In a randomized series (9) comparing surgical thrombectomy (n=18) to the 8F device (n=19), 89% success was achieved in the device group and 83% in the surgery group. 30 day patency was lower with the device (47% primary) than surgery (77%). Adjunctive urokinase was required in some cases. Hemolysis occurred in 63% of procedures. While not yet published, the complete results of the ATD clinical trial (8F device) were presented by Horn et al at the 1999 RSNA meeting. They reported 36% 90 day primary patency in the surgical arm (n=65) and 32% primary patency in the ATD arm (n=109). The immediate ability to dialyze was 77% in the surgical arm and 92% in the device arm. A study comparing pulse-spray lysis to the 8F device (n=57) showed 36% 90 day patency, notably there were higher complications in the device group (10). Residual thrombus may occur with this device, and it may not work well to treat the arterial plug, this could help explain why the final results of these trials fall short of the DOQI recommendations. Results with the 6F device have not been published, nor have any results from the currently available 7F Helix device. The ATD is not available in an over the wire design. This device costs $550 for the short version and $950 for a long version. The AngioJet (Possis Medical, Minneapolis, MN), uses high velocity saline jets to pulverize clot as well as aspirate the particulate using the Venturi effect. The device currently marketed for dialysis grafts (Xpeedior) consists of a 6F catheter used over a 0.035 inch wire. It is connected to an external drive unit which costs $25,000. The catheter and other disposables together cost $600-1450 depending on catheter length. The results of the US clinical trial of the AngioJet device (5F version), randomized to surgical thrombectomy and revision, showed 90 day patencies achieved with the device far below the DOQI recommendations at 15% (11). Based on the published evidence, this device cannot be recommended for treatment of clotted dialysis grafts at the present time. No published randomized data exist concerning the presently marketed device. Like the Amplatz device, some hemolysis occurs with the AngioJet. The Arrow-Trerotola percutaneous thrombolytic device (Arrow Inter-

national, Reading, PA) consists of a low-speed (3000 rpm) rotating stainless steel basket-like fragmentation cage which pulverizes the clot and effectively strips it from the walls of the graft. It is used through crossing 5 or 6F sheaths and requires only a minute or two to restore flow in the graft. The resulting particulate is 3mm in diameter or less, with the vast majority of particles smaller than 1mm. Much of this particulate can be aspirated via the sheaths. A comparative study of the PTD versus PSPMT in an animal model showed 100% effectiveness in graft thrombolysis with only 2/11 animals showing small pulmonary emboli on perfusion scans versus 10/11 in the PSPMT group (p<0.001) (12). In a clinical trial comparing the device to PSPMT in 122 patients, the technical success rate was identical (95%), procedure time shorter with the PTD (median 75 vs 85 min, p<0.04), and 90-day patency identical (13). It has been demonstrated that the PTD can be used to treat the arterial plug as well as intragraft thrombus, and is the only device capable of (and FDA approved for) doing so. Results using that approach yielded a 3 month patency of 41% (14). The PTD is one of two devices which can be described as wall-contact devices. This device is FDA approved for use in dialysis grafts and native fistulae and costs $600, and a 120 cm version costs $775. It is available in over the wire (0.025 inch, 7F) design as well. The Castaneda thrombolytic brush (MicroTherapeutics, Inc., San Clemente, CA), combines mechanical thrombolysis with thrombolytics to shorten procedure time and reduce lytic dose. While this is not a purely mechanical thrombolytic approach, it takes advantage of many of the principles of mechanical thrombolysis. This 6F device consists a polyester (Castaneda) brush which rotates at low speed (1800 rpm). The Castaneda goes over a 0.035 inch wire. Results from the clinical trial of the Cragg brush (15), an earlier version of the Castaneda, meet DOQI recommendations (see table). The brush is approved for dialysis grafts only in conjunction with urokinase, although some use it without lytics (offlabel). No data, randomized or otherwise, have been published for the Castaneda brush, either with or without lytics. The device costs $495 each, plus a vial of urokinase ($380). The Gelbfish EndoVac (NeoVascular Technologies, New York, NY), another FDA approved device, uses a “clot spoon” with a combination of suction and maceration with as well as irrigation to remove clot. The spoon/sheath combination is 6F and costs $350. Once flow is restored, the device can no longer be used and balloon maceration of any residual clot is performed. No data are available as to the clinical efficacy of this device either in print or from the manufacturer. This device is no longer available for sale in the US. The Oasis (formerly the SET, Boston Scientific, Natick, MA), is a 6F device is driven by an angiographic injector, and uses the Venturi effect to achieve thrombolysis, similar to the AngioJet. It can be used over a 0.018 inch wire. In a clinical trial randomized to PSPMT with urokinase, the technical success was 95% (90% control), procedure time 89 min (77 min control), ability to dialyze 89% (81% control) and 90 day patency 40% (41% control) (16). It costs $600. The Hydrolyser (Cordis, Miami, FL) is a 7F device which also uses the Venturi effect to achieve mechanical thrombolysis. The catheter is driven using a conventional angiographic injector. Vorwerk et al initially described use of the Hydrolyser in native (n=8) and synthetic (n=7) access (17). Although thrombectomy was successful in 15 of 16 instances, 5 rethrombosed within the first 24 hours. Secondary patency was 41% at six months. Overbosch et al reported 65 procedures in a heterogeneous graft population with 89% success and 55% 3 month patency. One pulmonary embolus occurred as did 2 arterial emboli and 2 major bleeding complications (18). The higher 3 month patency is likely due to the inclusion of native fistulae, which have better longterm results than grafts; the US trial (see table), not yet published, reported results similar to many other non-wall contact devices and just short of the DOQI recommendations. The Hydrolyser costs $500-700 depending on device length. One concern with this device and other Venturi-type devices is the amount of blood aspirated during the procedure (50-150 cc) which could be problematic for these chronically anemic patients; this has yet to be determined clinically. The Thrombex PMT (Baxter, Irvine CA) has received FDA approval for use in hemodialysis grafts. This 6F, 0.018 inch over-the-wire, non wall contact device uses an Archimedean screw approach to mechanical thrombectomy. A suction bottle is attached. No results from the clinical trial have been published or made available by the manufacturer. Cost is $650. The Solera (Bacchus Vascular, Santa Clara, CA) has received FDA approval for use in hemodialysis grafts. This 7F, 0.018 inch over-thewire device consists of a rotating spiral inside a fixed outer basket. Suc-

tion is applied with a syringe device. This is essentially a non-wall contact device, although the manufacturer indicates the outer basket can be rotated mechanically for limited wall contact. This device received FDA approval without a formal clinical trial, thus no outcomes data are available. Cost is $995 for the 55 cm device and $1095 for the 110 cm device. Native Fistula Lysis As the number of native fistulae increase in the US in accordance with the DOQI guidelines, more problems with native fistulae will be observed. Mechanical lysis is effective in native fistulae as well, as evidenced by several European reports with various devices including the Hydrolyser (17,18,19), ATD (19) and PTD (20). In fact, the long-term results appear to be better than those seen in grafts, although the procedure time is generally longer and the acute success rate somewhat less. Undoubtedly more reports on use of mechanical devices in native fistulae will emerge with time. Conclusion Mechanical thrombolytic devices have become the mainstay of percutaneous treatment of clotted dialysis grafts in many areas. The choice of percutaneous lytic technique and/or device depends on the operator's experience and availability of mechanical thrombolytic devices. Mechanical thrombolysis, properly used, is faster, safer and equally or more effective than chemical lysis. In an era of increasing reliance on evidence-based medicine, it is reassuring to know that for at least some devices are able to achieve DOQI recommendations, and that there are prospective, randomized published data to support this. Finally, it should be noted that mechanical devices are currently the only FDA approved way to percutaneously treat a clotted dialysis graft in the US. References: 1. Winkler TA, Trerotola SO,Davidson DD,Milgrom ML.Study of thrombus from thrombosed hemodialysis access grafts.Radiology 1995; 197:461465. 2. Etheredge E, Haid S, Maeser M, Sicard G, Anderson C. Salvage operations for malfunctioning polytetrafluoroethylene hemodialysis access grafts. Surgery 1983; 94(3):464-470. 3. Valji K, Bookstein JJ, Roberts AC, Davis GB. Pharmacomechanical thrombolysis and angioplasty in the management of clotted hemodialysis grafts: early and late clinical results. Radiology 1991; 178:243-247. 4. Valji K, Bookstein JJ,Roberts AC,Oglevie SB,Pittman C,O’Neill MP.Pulse-spray pharmacomechanical thrombolysis of thrombosed hemodialysis access grafts: long-term experience and comparison of original and current techniques.AJR 1995; 164:1495-1500. 5. Schmitz-Rode T, Pfeffer J, Bohndorf K, Gunther R. Percutaneous thrombectomy of the acutely thrombosed dialysis graft: in vitro evaluation of four devices. Cardiovasc Intervent Radiol 1993; 16:72-75. 6. Holzenbein TJ, Miller A, Gottlieb MN, Gupta SK. The role of routine angioscopy in vascular access surgery. J Endovasc Surg 1995; 2:10-25. 7. Puckett J, Lindsay S. Midgraft curettage as a routine adjunct to salvage operations for thrombosed polytetrafluoroethylene hemodialysis access grafts. Am J Surg 1988; 156:139-143. 8. Schwab S, Besarab A, Beathard G, Brouwer D, Etheredge E, Hartigan M, Levine M, McCann R, Sherman R, Trerotola S. NKF-DOQI Clinical Practice Guidelines for Vascular Access, Am J Kid Dis 1997; 30(4,Supp 3), S150-S191 9. Uflacker, R, Rajagopalan, PR, Vujic, I, Stutley, JE, Treatment of thrombosed dialysis access grafts: Randomized trial of surgical thrombectomy versus mechanical thrombectomy with the Amplatz device. JVIR 1996; 7:185-192. 10. Sofocleous CT, Cooper SG, Schur I, Patel RI, Iqbal A, Walker S, Retrospective comparison of the Amplatz thrombectomy device with modified pulse-spray pharmacomechanical thrombolysis in the treatment of thrombosed hemodialysis access grafts. Radiology. 1999 Nov;213(2):561-7 11. Vesely TM, Williams D, Weiss M, et al, Comparison of the AngioJet rheolytic catheter to surgical thrombectomy for the treatment of thrombosed hemodialysis grafts. JVIR 1999;10:1195-1205. 12. Trerotola, SO, Johnson, MS, Schauwecker, DS et al, Pulmonary emboli from pulse-spray versus mechanical thrombolysis in an animal dialysis graft model. Radiology 1996, 200: 169-176. 13. Trerotola SO, Vesely TM, Lund GB, Soulen MC, Ehrman KO, Cardella JF. Treatment of thrombosed hemodialysis access grafts: Arrow-Trerotola Percutaneous Thrombolytic Device versus pulse-spray thrombolysis. Radiology 1998; 206:403-414. 14. Lazzaro CR, Trerotola SO, Shah H, Namyslowski J, Moresco KP, Patel NH. Modified use of the Arrow-Trerotola percutaneous thrombolytic device for the treatment of thrombosed dialysis access grafts. JVIR

1999; 10:1025-1031. 15. Dolmatch BL, Casteneda F, McNamara TO, Zemel G, Lieber M, Cragg AH. Synthetic dialysis shunts: thrombolysis with the Cragg thrombolytic brush catheter. Radiology 1999; 213:180-184. 16. Barth KH, Gosnell MR, Palestrant AM, et al, Hydrodynamic thrombectomy system versus pulse-spray thrombolysis for thrombosed hemodialysis grafts: a multicenter prospective randomized comparison Radiology 2000 Dec;217(3):678-84. 17. Vorwerk D, Sohn M, Schurmann K, Hoogeveen Y, Gladziwa U, Guenther R. Hydrodynamic thrombectomy of hemodialysis fistulas: first clinical results. JVIR 1994; 5:813-821. 18. Overbosch EH, Pattynama PMT, Aarts HJCNM, Kool LJS, Hermans J, Reekers JA.Occluded hemodialysis shunts: Dutch multicenter experience with the Hydrolyser catheter. Radiology 1996; 201:485-488. 19. Haage P, Vorwerk D, Wildberger JE, Piroth W, Schurmann K, Gunther RW. Percutaneous treatment of thrombosed primary arteriovenous hemodialysis access fistulae. Kidney International 2000; 57:1169-1175. 20. Rocek M, Peregrin JH, Lasovickova J, Krajickova D, Slaviokova M Mechanical thrombolysis of thrombosed hemodialysis native fistulas with use of the Arrow-Trerotola percutaneous thrombolytic device: our preliminary experience J Vasc Interv Radiol 2000 Oct;11(9):1153-8 17. ATD package insert. 18. Data courtesy Cordis

Special Session Non-Invasive Vascular Imaging 33.3 Is an arterial screening programme workable? S. G. Ruehm; Diagnostic and Interventional Radiology, University Hospital Essen, Essen, Germany. As the western population continues to age, atherosclerotic disease is becoming more prevalent. With 90% of atherosclerotic lesions identified below the aortic bifurcation, the lower limbs are the most frequently affected vascular territory. Presently, peripheral vascular disease (PVD) accounts for 50.000-60.000 percutaneous transluminal angioplasties, implantation of 110.000 vascular prostheses and 100.000 amputations annually in the U.S. alone [1-2]. The management of a patient adversely affected by PVD has to be seen in the context of the disease's epidemiology and, in particular, apparent risk factors or markers predicting spontaneous deterioration [3]. Thus, reflecting the systemic nature of atherosclerotic disease, PVD is frequently associated with coronary, renal, and carotid arterial disease. Proper management of arterial disease requires a comprehensive assessment of the underlying vascular morphology. Localizing and grading the severity of arterial lesions is crucial for therapeutic decisionmaking. For this purpose, several imaging modalities, including conventional catheter angiography, duplex ultrasound, as well as CT- and MRangiography (MRA) are in clinical use. Lack of ionizing radiation and contrast agents void of any nephrotoxicity [4-5], in conjunction with high diagnostic accuracy have driven the rapid implementation of MRA as the modality of choice for assessing arterial disease in many centers throughout the world [6-9]. Since atherosclerotic disease affects the entire arterial system, extended coverage allowing the concomitant assessment of the arterial system from supraaortic arteries to the distal runoff vessels appears desirable. Subsequent parenchymal enhancement and contrast dose limitations had initially curtailed contrast-enhanced 3D MRA to the display of the arterial territory contained within a single field-of-view extending over 40-48 cm. The implementation of “bolus chase” techniques extended coverage to encompass the entire run-off vasculature, including the pelvic, femoral, popliteal and trifurcation arteries [10-12]. The implementation of faster gradient systems has laid the foundation for a further extension of the bolus chase technique: whole body coverage extending from the carotid arteries to the trifurcation vessels with 3D MRA has become possible in only 72 seconds [13]. Correlation with a limited number of regional DSA exams revealed the diagnostic performance of whole-body MRA to be sufficient to warrant its consideration as a noninvasive alternative to DSA. The performance of whole-body MRA was further improved with the introduction of AngioSURF, which integrates the torso-surface coil for signal reception. Use of the surface coil results in higher SNR and CNR values translating into sensitivity and specificity values of 95.3% and 95.2%, respectively, for the detec-

tion of significant stenoses (luminal narrowing > 50%) in lower extremity peripheral vascular disease [14]. Noninvasiveness, three-dimensionality, extended coverage and high contrast conspicuity are the characteristics of the applied whole body 3D MRA-approach allowing for a quick, risk-free, and comprehensive evaluation of the arterial system beyond the peripheral vasculature in patients referred to display clinically relevant atherosclerotic lesions. References: 1. Martin EC: Transcatheter therapies in peripheral and nonvascular disease. Circulation 1991; 83: 1-5 2. Rutkow IM, Ernst CB: An analysis of vascular surgical manpower requirements and vascular surgical rates in the United States. J Vasc Surg 1986; 3:74-83 3. Management of peripheral arterial disease (PAD). TransAtlantic InterSociety Consensus (TASC). 4. J Vasc Surg 2000; 31, part 2 (Supplement): S5 5. Shehadi WH: Contrast media adverse reactions: occurrence, recurrence, and distribution patterns. Radiology 1982;143: 11 -17 6. Shellock FG, Kanal E: Safety of magnetic resonance imaging contrast agents. J Magn Reson Imaging 1999;10:477-84 7. Prince MR: Gadolinium-enhanced MR aortography. Radiology 1994; 191:155-164. 8. Prince MR, Narasimham DL, Stanley JC, Chenevert TL, Williams DM, Marx MV, Cho KJ: Breath-hold gadolinium-enhanced MR angiography of the abdominal aorta and its major branches. Radiology 1995; 197: 785-792 9. Meaney JF, Weg JG, Chenevert TL, Stafford-Johnson D, Hamilton BH, Prince MR.: Diagnosis of pulmonary embolism with magnetic resonance angiography. N Engl J Med 1997; 336: 1422-1427 10. Goyen M, Debatin JF, Ruehm SG: Peripheral MR-Angiography. Top Magn Res Imaging 2001; 12: 327-335 11. Meaney JF, Ridgway JP, Chakraverty S, et al. : Stepping-table gadolinium-enhanced digital subtraction MR angiography of the aorta and lower extremity arteries: preliminary experience. Radiology 1999; 211:59-67 12. Ho KY, Leiner T, de Haan MW, et al.: Peripheral vascular tree stenoses: evaluation with moving-bed infusion-tracking MR angiography. Radiology 1998; 206: 683-692 13. Ruehm SG, Hany TF, Pfammatter T, et al.: Pelvic and lower extremity arterial imaging: diagnostic performance of three-dimensional contrast-enhanced MR angiography. Am J Roentgenol 2000; 174:11271135 14. Ruehm SG, Goyen M, Barkhausen J, et al.. Rapid magnetic resonance angiography for detection of atherosclerosis. Lancet 2001; 357: 10861091 15. Goyen M, Quick HH, Debatin JF, Ladd ME, Barkhausen J, Herborn CU, Bosk S, Kuehl H, Schlepütz M, Ruehm SG: Whole Body 3D MR Angiography Using AngioSURF: Initial Clinical Experience, Radiology (in press)

Special Session Thrombolysis 36.1 Thrombolytic (& Antiplatelet) Drugs and their use K. Kandarpa; UMass Memorial Medical Center, Massachuesetts, United States. Introduction This talk will focus primarily on thrombolytic agents currently in use for treating peripheral arterial occlusions (PAO). The pharmacology and outcomes for this use have been previously described extensively. Since there has been a recent interest in combination therapy with antiplatelet agents, the abstract reviews recent studies in this area as added material to complement the talk. Clot Formation Platelet adhesion, activation, and aggregation (1-4) Circulating platelets are inactive. When arterial injury occurs, either spontaneously or as result of a percutaneous intervention, sub-endot-

helial adhesive glycoproteins are exposed at the luminal surface. These glycoproteins (which include collagen, fibrinogen, vWF, fibronectin, and vitronectin) bind with high affinity to surface receptors on platelets within the boundary layer of the bloodstream. This results in the adhesion of a monolayer of platelets onto the injured surface. A hemostatic or occlusive thrombus forms when this layer of platelets is activated and platelet aggregation is initiated. Adhesion alone, as well as the local secretion or release of pro-aggregatory agents, lead to the expression of platelet receptor sites that play a key role in platelet aggregation. The more important of these agonists are adenosine diphosphate (ADP), epinephrine, and thrombin. ADP is contained in intracellular granules that are released when the platelets are activated. ADP, like the other agonists, binds to specific receptors on circulating platelets and induces intracellular pathways, such as the release of arachidonic acid (AA) and formation of thromboxane A2 (TxA2), that activate the GP IIb/IIIa receptor. The GP IIb/IIIa receptors, which are specific to platelets, represent the final common pathway for aggregation that occurs when these receptors bind to fibrinogen and cross-link adjacent platelets. There are approximately 80,000 GP IIb/IIIa receptors per platelet. These receptors recognize the Arg-Gly-Asp (RGD) amino acid sequence, which is the adhesive motif of circulating fibrinogen and other adhesive proteins. Intravenous Antiplatelet Agents (2-7) Currently, three intravenous agents are approved by the FDA: abciximab (ReoPro®), eptifibatide (Integrilin®), and tirofiban hydrochloride (Aggrastat®). All three are direct GP IIb/IIIa receptor antagonists that prevent fibrinogen-mediated platelet aggregation. Aciximab is the Fab fragment of the chimeric human-murine monoclonal antibody 7E3 that binds to the GP IIb/IIIA receptor. It binds with similar affinity to the vitronectin receptor (αvβ 3). Integrilin is a synthetic heptapeptide that reversibly blocks the GP IIb/IIIa receptor. Aggrastat is a reversible nonpeptide blocker of the GP IIb/IIIa receptor. All three drugs are approved as adjunctive therapy for the prevention of cardiac ischemic events during percutaneous coronary interventions (PCI), with concomitant PO ASA and IV heparin administration. The latter two agents have shorter biological half-lives than abciximab. Since excess bleeding is a concern with all of these agents manufacturer guidelines and contraindications, which are similar to those for thrombolytic agents, should be followed. The recommended dose for ReoPro is a bolus of 0.25 mg/kg given within an hour prior to PCI, followed by an infusion of 0.125 µg/kg/hr (for a maximum of 10µg/min) for 12 hours. This has been the dose used for peripheral interventions as well. Low dose heparin and PO ASA are recommended. Following this bolus and infusion dose, abciximab blocks 80% of the GP IIb/IIIa receptors. Upon discontinuation, platelet function gradually returns over about 48 hours. Although plasma levels fall rapidly upon termination of infusion, abciximab remains platelet-bound in the circulation for about 15 days. When used with low dose heparin, there is less than a 1% incidence of severe thrombocytopenia, which generally occurs within the first 24 hours of treatment. Thus, monitoring should include platelet counts at baseline, 2-4 hours after initiation, and at 24 hours. If there is a need to restore platelet function, abciximab should be discontinued, and platelets may be transfused. With transfusion, abciximab restores platelet function partially by redistributing itself over the entire population of platelets, leading to a lower level of receptor blockade. Generally, prior to the initiation of the agent, baseline ACT, PT, and aPTT levels are also recommended. Early sheath removal is encouraged, but an aPTT < 50 seconds and an ACT < 175 seconds must be first achieved. Although, neither excess anaphylactic nor hypersensitivity reactions have been reported with primary administration or re-administration of this agent, the potential for this does exist. The dosing of Integrilin has been modified from the original coronary trials to include an initial double-bolus (180mcg/kg x 2 - 10 minutes apart) followed by a continuous infusion (2mcg/kg/min) for 24 hours. Doses must be adjusted for severely compromised renal function. Low dose heparin and PO ASA are recommended. Upon dosing, inhibition of platelet aggregation (> 80%) is rapid and bleeding time increases 2- to 4- fold, but platelet function and bleeding time return to normal within 4 - 8 hours of termination. Aggrastat is given for PCI at an initial rate of 0.4 µg/kg/min over 30 minutes, followed by a continuous infusion of 0.1 µg/kg/min for 1224 hours. This dose must be adjusted for severe renal insufficiency. Greater than 90% inhibition of platelet aggregation is attained within 30 minutes of dosing, and this is promptly reversed by stopping the infusion. For PCI, concomitant heparin is recommended because of the risk of excessive mortality with Aggrastat alone. For peripheral inter-

ventions, low dose heparin is probably appropriate. Thrombolytic & Antiplatelet Agents for PAO (7-21) The primary goal of using GP IIb/IIIa inhibitors in PVD would be to prevent rethrombosis during thrombolytic therapy and to passivate the vessel wall after angioplasty or stenting with view towards preventing acute thrombosis and restenosis. At the time of this writing there is little documented evidence of such benefit from using these agents in the peripheral arteries. Nevertheless, the combined results of the coronary intervention trials discussed above and some preliminary reports of applications in the peripheral vessels lend some support for extending the use of these drugs for peripheral vascular interventions. Early anecdotal reports of spontaneous resolution of coronary thrombi with either IV or local IC abciximab infusions were published in abstract form. Platelet disaggregation has been shown to occur in vitro. The observation of spontaneous disruption of acute IC thrombi by abciximab therapy is probably due to the combined effects of platelet disaggregation and endogenous lysis. The TAMI-8 trial (Thrombolysis and Angioplasty in Acute Myocardial Infarction-8) studied a small group of patients to assess the activity and safety of combining rtPA and abciximab (n=60) compared to controls (rtPA only, n=10). Ninety percent of the treated patients had angiographically patent infarct-related arteries, versus only 5 of the 9 controls. Eighty-seven percent of the treated patients were free of recurrent ischemia, and there were no excess bleeding complications due to abciximab. Similarly, the multi-center PARADIGM (Platelet Aggregation Receptor Antagonist Dose Investigation for Reperfusion Gain in Myocardial Infarction) trial studied various doses of lamifiban in combination with either rtPA or streptokinase for patients with acute onset (<12 hours) MI. Of the patients treated with lamifiban, about seventy five percent (vs 63% control) had angiographically patent infarctrelated vessels at 90 minutes. Lamifiban also reduced the time to steady-state reperfusion by 28%. However, 30-day clinical outcomes were not different between the study and control groups, and the incidence of major bleeding was higher in the lamifiban patients (3% vs 1.7%). Another study combined eptifibatide (with ASA and heparin) with rtPA to treat AMI patients. The results showed that the incidence and speed of reperfusion was significantly enhanced by such a strategy, but there we no significant differences between the study group and controls with respect to clinical outcomes or severe bleeding. Thus, the combination of platelet inhibitors with fibrinolytic agents appears to accelerate thrombolysis. The results of experimental and early clinical studies using combination platelet inhibitor and thrombolytic therapy in peripheral arteries are encouraging. A recent experimental study showed that treatment with reteplase (Retavase™) and abciximab of peripheral arterial thrombi in primates significantly reduces the time to sustained reperfusion of the limb. Haas et al reported that the addition of abciximab therapy restored patency in 8 of 8 patients in whom lytic therapy with urokinase alone had failed. There were three minor bleeds. Tepe et al. reported a 100% technical success rate during their early clinical experience with combining abciximab and urokinase in fourteen patients peripheral arterial occlusions (< 6 months duration). Treatment times ranged from 50 minutes to 8 hours. There were only two minor early bleeds and one intermediate-term reocclusion (at 8 months; follow-up range 1-15 months). They have recently reported treatment times below 4 hours, high technical success, and 2 major and 2 minor bleeds in a total of 38 patients. Yoon et al reported their preliminary experience with treating a small group (n=7 in each arm) of PAO patients with a combination of eptifibatide, rtPA, and heparin or rtPA and heparin only. Although treatment times were dramatically longer than reported by Tepe et al, the time to treatment was nearly double in the control group in comparison with the treated group (40 vs 22 hours). However, more recently, in a larger trial, the same group failed to show a reduction in treatment time, but did show a significant reduction in the rt-PA dose necessary for achieving thrombolysis, without loss of efficacy or safety, when Eptifibatide is used concomitantly. Recently, Schweizer et al. reported on their experience with 84 patients who were randomized to two equal groups receiving either IV ASA or abciximab during thrombolytic therapy with rtPA. Interestingly, they used an aggressive short-term thrombolysis protocol in which they combined suction thrombectomy with rapid 5 mg infusions of rtPA until the thrombus was cleared. This was followed by angioplasty as needed. Sub-therapeutic doses of IV heparin were used during treatment and all patients were anticoagulated for six months. There were no major complications and the few minor complications occurred with equal frequency in both groups. Lysis times were significantly shorter in the abciximab group and a composite clinical

outcome at six months significantly favored the use of this agent. Most interestingly, claudication distance was not only significantly higher in the abciximab group, but it also showed a gradual improvement over the six-month period of observation, whereas the ASA group showed some deterioration. Duda et al have recently published the results of the PROMPT study. In a randomized comparison of patients with acute lower extremity PAO treated either with urokinase alone or urokinase plus abciximab, they showed that the combination group had faster dissolution of thrombus and a significantly improved amputation-free survival rate at 90 days. Many patients who receive GPIIb/IIIa inhibitor therapy during peripheral vascular interventions will need elective, and at times emergent, surgery. Abciximab, for instance, is associated with higher incidence mortality and bleeding following major surgery (e.g., CABG). Thus, the problem of rapid reversal of the antiaggregation effects of these drugs becomes important. Eptifibatide and tirofiban are rapidly reversed when treatment is terminated and platelets are transfused. Abciximab is less easily reversed, but the degree of blockade can be reduced with platelet transfusion. These should be considerations when choosing these agents for specific situations. References: 1) Wilentz JR, Sanborn TA, Haudenschild CC, et al. Platelet accumulation in experimental angioplasty: time course and relation to vascular injury. Circulation 1987; 75: 636-642. 2) Mousa SA, Bennett JS. Platelets in health and disease: platelet GPIIb/IIIa structure and function – recent advances in antiplatelet therapy. Drugs of the future 1996; 21: 1141-1154. 3) Coller BS. Platelet GPIIb/IIIa antagonists: The first anti-Integrinreceptor therapeutics. J. Clin. Invest 1997; 99: 1467-1471 4) Gonzalez ER. Antiplatelet therapy in atherosclerotic cardiovascular disease. Clinical Therapeutics 1998; 20: B18-B41. 5) Dyke CM. Safety of GPIIb/IIIa inhibitors: a heart surgeon’s perspective. Am Heart J 1999; 38:S307-S316. 6) Topol EJ, Byzova TV, Plow EF. Platelet GPIIb/IIIa blockers. Lancet 1999; 353:227-231. 7) Adgey AAJ. An overview of the results of clinical trials with GPIIb/ IIIa inhibitors. Eur Heart J 1998; 19(Suppl D): D10-D21. 8) Tepe G, Schoff U, Erley CM, et al. Platelet GPIIb/IIIa receptor antagonists used in conjunction with thrombolysis for peripheral arterial thrombosis. AJR 1999; 172:1343-1345. 9) Verheugt FWA, Meijer A, Lagrand WK, Eenige MJ. Reocclusion: the flip side of coronary thrombolysis. JACC 1996; 27: 766-773. 10) Coller B. Platelets and thrombolytic therapy. N Engl J Med 1990; 322: 33-42. 11) Coller BS, Anderson KM, Weisman HF. The anti-GPII/IIIa agents: fundamental and clinical aspects. Haemostasis 1996; 26: 285-293. 12) Kleiman NS, Ohman EM, Callif RM, et al. Profound inhibition of platelet aggregation with monoclonal antibody 7E3 Fab after thrombolytic therapy: results of the TAMI-8 pilot study. JACC 1993; 22: 381189. 13) The IMPACT-AMI Investigators. Combined accelerated tissueplasminogen activator and platelet GPIIb/IIIa Integrin receptor blockade with Intergrillin in acute myocardial infraction. Circulation 1997; 95: 846-854. 14) Harrington RA. PARADIGM. AHA 69th Annual Scientific Sessions. New Orleans, LA. 1996. 15) Ohman EM, Kleiman NS, Gacioch G, et al. Combined accelerated tPA and platelet GPIIb/IIIa Integrin receptor blockade with Integrilin in AMI. Circulation 1997; 95:846-854. 16) Nakada MT, Nedelman MA, Barnathan EB, Cohen SA, Jordan RE. Catheter-directed Reteplase alone and in combination with systemic Abciximab promote sustained reperfusion in a primate model of peripheral vascular disease. Abstract No. 104, SCVIR Annual Scientific Meeting, 2000, San Diego. 17) Haas PC, Yaryura R, Krajcer NE, et al. Use of GPIIb/IIIA antagonist (abciximab) in peripheral interventions. Am J Cardiol 1997; 80: 46S. 18) Yoon H, Lessie T, Miller FJ. The use of peptide inhibitors of GPIIb/ IIIa: preliminary results in patients with acute peripheral arterial occlusions. Abstract No. 101, SCVIR Annual Scientific Meeting, 2000, San Diego. 19) Yoon H, Miller FJ. The use of a peptide inhibitor of the GPIIb/IIIa platelet receptor: initial experience in patients with acute peripheral arterial occlusions. (AJR 2001, in press) 20) Schweizer J, Kirch W, Koch R, et al. Short- and long-term results of abciximab versus aspirin in conjunction with thrombolysis for

patients with peripheral occlusive arterial disease and arterial thrombosis. Angiology 2000; 51:913-923. 21) Duda SH, Tepe G, Oliver O, et al. Peripheral artery occlusions: treatment with abciximab plus urokinase versus with urokinase alone – a randomized pilot trial (PROMPT Study). Radiology 2001; 221: 689696.

36.2 Thrombolytic Therapy: update M. Maynar1, S. Casey2, Z. Qian2; 1 Endoluminal Therapy, Hospiten Rambla, Santa Cruz de Tenerife, Spain, 2Radiology, Louisiana State University Health Sciences Center, New Orleans, United States. Thrombolytic therapy refers to the use of a pharmaceutical agent to restore blood flow and vessel patency for tissue reperfusion and vascular healing. The rationale behind thrombolysis is the generation of plasmin triggered by plasminogen activator leading to the enzymatic breakdown of cross-linked fibrin strands within the thrombus. Catheter-directed thrombolysis (CDT), local administration of thrombolytic agent, is more efficient at drug delivery and minimizes systemic complications. Thrombolysis causes rapid clot dissolution and potentially unmasks underlying stenosis. Results of successful clinical applications during the last decade demonstrate that this treatment modality is improving efficacy for the management of vascular occlusive disease. Brief history and back ground The first clinical investigation of a thrombolytic agent was reported in 1955 followed by reports on the use of thrombolytic agents in the treatment of occluding vascular thrombi. In 1974 Dotter reported local use of streptokinase (SK) for selective clot lysis [1]. Since then researchers and clinicians around the world have been devoted to the development of higher purified thrombolytic agents and improved approaches in the management of intravascular thrombosis that have significantly changed the outcome of thrombotic events. However the precise role of thrombolytic therapy in the treatment of thrombotic disorders was not clear for a long time because there were no large-scale randomized prospective studies comparing thrombolytic therapy with the standard surgical treatment in terms of safety and efficacy. In 1994 Ouriel et al. reported the randomized comparison of urokinase (UK) versus surgical intervention in 114 patients with severe acute limb ischemia [2]. Subsequently the STILE, TOPAS, and POSE trials were instrumental in the acquisition of knowledge for thrombolysis that led to improved standardization of the definitions and objectives of thrombolysis [3-6]. Commonly used agents SK (Astra Pharmaceuticals, Wayne, PA) is a first generation thrombolytic drug isolated from ß-hemolitc streptococcal cultures. Although it is inexpensive and has a prolonged thrombolytic effect, SK has not gained popularity because of problems related to hypersensitivity reactions and decreased efficacy after repeated use [7]. SK is not fibrin-specific and causes severe plasminogen and fibrinogen depletion resulting in a systemic lytic (anticoagulated) state. SK has proved to be inferior to urokinase (UK) and rt-PA in the capability to dissolve peripheral arterial thrombus. UK (Abbott Laboratory, Abbott Park, IL) previously produced by human tissue culture was the most widely used thrombolytic agent. In 1999 distribution of UK in the United States was halted due to a theoretical risk of viral contamination. UK manufactured by genetic engineering resumed marketing in September 2002. Because of its remarkable advantages over SK, including consistency in performance, predictability, lack of immunogenicity, and reduced hemorrhagic complications [8], the use of UK is broadly accepted for CDT. The recombinantly derived t-PA, also known as rt-PA (Alteplase; Genetech, South San Francisco, CA), is produced by recombinant DNA technology. rt-PA is isolated, purified and then packaged as a sterile powder in vials [9]. In 1987 the FDA approved the use of rt-PA for the treatment of acute MI, acute PE, and acute ischemic stroke. It gained much interest due to the shortage of UK in 1999. The primary advantages of rt-PA are the high fibrin and clot selectivity and a short halflife that allows focused enzymatic activity on the clot with minimal systemic fibrinolysis [10]. The main disadvantage is that the clinical doses for CDT vary widely from 0.25 to 10 mg/hr. Other emerging agents include reteplase (r-PA; Centocor, Malvern , PA) and anistreplase (Eminase; Roberts Pharmarceutical, Eatontown, NJ). There is only limited published information on CDT with these agents

thus far. Clinical applications CDT is a complementary and not by any means a competing technology with surgical or percutaneous revascularization. In appropriately selected patients it can reduce or obviate subsequent surgery for acute limb ischemia (<14 days) [9]. It also appears to be effective in the management of DVT, thrombosed vascular grafts, thrombosis–related catheter occlusion, acute ischemic stroke, and acute myocardial infarction [9, 11-15]. Peripheral arterial occlusion The best candidates for CDT are patients with acute occlusion. In the case of severe ischemia progressing to sensory and motor deficits, surgical intervention is the first line of treatment. A delay due to a CDT treatment under those circumstances could result in the loss of a limb. A prospective randomized trial comparing CDT with UK and rt-PA to surgical recanalization for limb ischemia demonstrated that surgical revascularization was more effective and safer than CDT for patients with chronic ischemia. Better amputation-free survival and a shorter hospital stay were observed in patients with acute ischemia treated with CDT. CDT also resulted in a significant reduction in planned surgery. UK seemed to be equal to rt-PA in efficacy and safety [3]. Several methods of CDT have been described including IA infusion, bolus injection, and pulse-spray technique. Intraarterial (IA) continuous infusion remains the most commonly used method in the treatment of peripheral arterial occlusion. Intrathrombus bolus injection (lacing) has been proposed to shorten the duration of CDT and is particularly important for acute limb ischemia. Results from a British randomized trial reported an 80% reduction in median duration of infusion time from 20 hours to 4 hours with bolus injection technique [16]. However it may potentially increase the risk of bleeding complications. Pulse-spray technique (PST) has also been used in CDT for peripheral arterial occlusion and purportedly has an advantage of rapid clot lysis compared to continuous infusion. PST refers to the technique of forcefully injecting thrombolytic agent into the thrombus. There is not wide acceptance of this technique because of the potential risk of downstream emboli, the need to use a special injecting device and uncertain benefits [17]. Whether or not concomitant anticoagulation with heparin during CDT is beneficial to patients remains controversial. A SCVIR Multidisciplinary Advisory Panel recommended the conservative use of heparin until further studies are performed. Meanwhile it endorsed a subtherapeutic regimen during CDT. After a successful CDT, anticoagulation with heparin has become a common practice to prevent acute rethrombosis. DVT The principal goal of thrombolytic therapy for DVT is to rapidly remove acute thrombus, prevent pulmonary embolism, and preserve venous valve function. Patient population suitable for this form of treatment modality is primarily for those who have acute DVT. A number of clinical trials of CDT with UK for DVT have documented its safety and efficacy. Initial Stanford experience with CDT demonstrated that a significant lysis (72% complete and 20% partial) had been achieved in 27 affected limbs of 21 patients with iliofemoral venous thrombosis. Additionally disappearance of concomitant symptoms was noted [18]. More recently, data from a multicenter registry of 473 patients with DVT found that either complete lysis or more than 50% lysis was achieved in more than 80% of the patients [19]. Although some chronic DVT (more than 10 days) could be significantly lysed with CDT, the best results with CDT were obtained in patients with acute symptoms without a prior history of DVT [10]. A recent NIH trial using once-daily t-PA through pulsespray catheter over 30 minutes resulted in extensive or complete lysis in 90% cases. Their protocol used t-PA dose of 50mg/d per leg and a maximum of 4 sessions [20]. Any patient presenting with phlegmasia cerulean dolens should be a candidate for CDT unless an absolute contraindication prevents such treatment. Thrombosed hemodialysis access grafts Percutaneous treatment options for a thrombosed access graft include thrombolysis, pharmacomechanical thrombolysis (PMT), and mechanical thrombectomy. Although thrombolytic therapy seems to be somewhat effective, it usually requires 2 to 20 hours infusion time [12] and is not sufficient to remove a clot without the need for adjunctive mechanical means. To address these problems with thrombolysis PMT has been used to accelerate the thrombolytic process. A pulse-spray technique described by UCSD group with forceful injection of concentrated thrombolytic agent into the clot has achieved more consistent results with 90-99% initial success rates [21]. With this technique the mean infusion duration has been considerably reduced to 49 minutes without significant hemorrhagic complications [22]. A modified pulse-spray pro-

cedure added balloon clot fragmentation, heparin and mechanical thrombectomy that further shortened the infusion time to 23 minutes. However the initial success rate has remained unchanged. Recently Cynamon proposed a “lyse-and-wait” technique to simplify the procedure while keeping a high success rate [23]. With this technique a catheter is inserted into the treated graft near arterial anastomosis aiming at the venous limb without imaging guidance. While the graft is simultaneously compressed at both arterial and venous anastomosis, UK (250,000 IU) mixed with heparin (5,000 U) is infused into the closed graft over 1 minute. After 2 hours if the platelet plugs remain in the graft, mechanical measures can be used. Primary patency was attained in 94% of cases with this technique. Other authors suggest that a low dose of 5,000 IU is also effective for “lyse-and-wait” declotting. Thrombosis-related catheter occlusion Central venous catheter (CVC) dysfunction occurs in up to 25% of all catheters with 60% of them attributable to thrombosis related occlusion [24]. Both UK and rt-PA have been extensively used in the management of thrombosed catheters. A double-blind randomized clinical trial involving 50 occluded CVC compared UK (10,000 IU) with rt-PA (2mg) in clearance rates for a two-hour treatment duration. The study found more efficiency in restoring catheter patency with rt-PA (89%) than with UK (59%) (P<0.05) [25]. In 2000, a consensus protocol for the use of rt-PA in the management of thrombosed CVC was drafted by the nine health care organizations including SCVIR and Oncology Nursing Society. The protocol is as follows: use an adequate volume of the 1-mg/ml (usually a 2-ml bolus) rt-PA solution to fill the CVC with a dwell time of 30-120 minutes. The dose can be repeated if catheter function is not restored following the initial instillation. If the catheter does not respond to two instillations or if dysfunction recurs further evaluation of the occlusion may be required [26]. Stroke and MI The NINDS trial opened a new chapter for acute stroke treatment by intravenous administration of rt-PA to reduce stroke mortality [27]. The rationale behind thrombolytic therapy in acute ischemic stroke is based on the concept that rapid recanalization provides improved cerebral blood flow and rescues neurons at an early reversible stage. Although the outcomes of the treatment greatly depend on the time when thrombolysis is initiated, the exact “therapeutic window” remains unknown. Animal studies indicate that irreversible ischemic brain injury may occur when vessel occlusion lasts for 2-3 hours. However a recent clinical trial using pro-UK showed a positive outcome in patients with severer middle cerebral artery (MCA) occlusion within a 6-hour window [28]. Clinical experience has revealed that prognosis for proximal MCA (M1/M2) occlusion is generally poor, while distal MCA occlusions have a better prognosis. There are two treatment options. First, IV rt-PA is an easy and rapid approach being particularly effective for small infarcts. Second, IA administration has a local direct effect on thrombi and may be more effective in patients with greater deficits. Improved perfusion and survival benefits with thrombolysis have been reported in acute MI, especially with mutant PA reteplase. Immediate angioplasty currently remains a reasonable and feasible approach to early coronary revascularization because shorter delay between the onset of a MI and therapeutic intervention results in a better survival outcome. Complications, prevention, and management The most common and serious complication from thrombolytic therapy is procedure-related hemorrhage. The best way to prevent complications from occurring is to comply with proper guidelines and minimize the use of anticoagulant. Relative contraindications include diabetic retinopathy, remote stroke or TIA, recent percutaneous intervention, hepatic insufficiency, renal insufficiency, oral anticoagulation therapy with an international normal ratio of more than 2.0, major GI bleeding within the past 3 months, and contrast dye allergy. The most important absolute contraindications to thrombolysis are a history of stroke or TIA within the past 6 months, recent major surgery or trauma, active internal hemorrhage, intracranial malignancy or vascular malformation, uncontrolled hypertension, pregnant women, and those in the immediate postpartum period [9,29]. Although all of the available agents usually have a short half-life, a hypocoagulable state can persistent for a long time due to the fibrin degradation products resulting from fibrinolysis. If a bleeding complication becomes severe it can be corrected by fresh frozen plasma or cryoprecipitate. In addition the fibrinolytic process can also be reversed with aminocaproic acid (Amicar, Immunex, Seattle, WA) [29]. Future directions Besides the continuing development of more effective and safer agents,

the future directions may lie on new strategies to improve thrombolysis outcomes. Because of the crucial role of platelet deposition in thrombus formation, prevention of on-going fibrin deposition and blockade of platelet mediated process are not only logic for a comprehensive thrombolysis treatment but also important for avoidance of rethrombosis. Glycoprotein IIb/IIa (GP IIb/IIIa) inhibitor has been successfully used to block GP IIb/IIIa receptors on the platelet surface, that bind fibrin(ogen) to form the framework of the platelet thrombus. Several pilot studies have clearly demonstrated clinical benefits resulting from combination treatment with the thrombolytic agent and GP IIb/IIIa inhibitor. A large-scale prospective randomized trial will further define the role of the combination therapy in thrombolysis. Other potential strategies include 1) CDT with a low dose of thrombolytic agent in combination with the use of a thrombolytic enhancer, 2) CDT at a standard dose coupled with the intravenous injection of an anti-thrombolytic agent, or 3) combination of a low dose CDT, a local thrombolytic enhancer, and a systemic anti-thrombolytic agent to protect remote sites from hemorrhagic complications [30]. References: 1. Dotter CT, Rosch J, Seaman AJ. Selective clot lysis with low dose streptokinase. Radiology 1974;111:31-37. 2. Ouriel K, Shortell CK, DeWeese JA et al. A comparison of thrombolytic therapy with operative revascularization in the initial treatment of acute peripheral arterial ischemia. J Vasc Surg 1994;19:1021-1030. 3. Anonymous. Results of a prospective randomized trial evaluating surgery versus thrombolysis for ischemia of the lower extremity: the STILE trial. Ann Surg 1994;220:251-266. 4. Ouriel K, Veith FJ. Sasahara AA. Thrombolysis or peripheral arterial surgery: phase I results. TOPAS investigators. J Vasc Surg 1996;23:6473. 5. Ouriel K, Veith FJ. Sasahara AA.A comparison of recombinant urokinase with vascular surgery as initial treatment for acute arterial occlusion of the legs. N Engl J Med 1998;338:1105-1111. 6. Ouriel K, Kandarpa K, Schuerr DM, Hultquist M, Hodkinson G, Wallin B. Prourokinase vs. urokinase for recanalization of peripheral occlusions, safety and efficacy: the PURPOSE trials. JVIR 199;10:1083-1091. 7. Dube M, Soulez G, Therasse E et al. Comparison of stretokinase and urokinase in local thrombolysis or peripheral arterial occlusions for lower limb salvage. JVIR 1998;7:587-593. 8. Van Breda A, Katzen BT, Deutsch AS. Urokinase versus streptokinase in local thrombolysis. Radiology 1987;165:109-111. 9. Semba CP, Murphy TP, Bakal CW, Calis KA, Matalon TAS and the Advisory Panel. Thrombolytic therapy with use of Alteplase (rt-PA) in peripheral arterial occlusive disease: review of the clinical literature. JVIR 2000;11:149-161. 10. Sobel BE. The language of lysis: a proposal to standardize the nomenclature. Am Coll Cardiol 1999;34:12226-1227. 11. Mewissen MW. Catheter-directed thrombolysis for lower extremity deep vein thrombosis. Techniques in Vasc & Intervent Radiol 2001;4:111-114. 12. Gibbens DT, Triolo J, Yu T, Depalma J, Iglasias J, Castner D. Comtemporary treatment of thrombosed hemodialysis grafts. Techniques in Vasc & Intervent Radiol 2001;4:122-126. 13. Haire WD. Techniques in dosing for thrombolysis of occluded central venous catheter. Techniques in Vasc & Intervent Radiol 2001;4:127-130. 14. Perl J-II, Samples SD. Thrombolytic therapy for acute ischemic stroke. Techniques in Vasc & Intervent Radiol 2001;4:115-121. 15. Spinler SA. Inverso SM. Update on strategies to improve thrombolysis for acute myocardial infarction. Pharmacotherapy. 2001;21:691-716. 16. Braithwaite BD, Buckenham TM, Galland RB et al. Prospective randominzed trial of high-dose versus low-dose tissue plasminogen activator infusion in the management of acute limb ischemia. Br J Surg 1997;84:646-650. 17. Kandarpa K, Goldhaber SZ, Meyerovitz MF. Pulse-spray thrombosis: the ¨carful analysis¨. Radiology 1994;320:324. 18. Semba CP, Dake MD. Catheter-directed thrombolysis for iliofemoral venous thrombosis. Radiology 1994;191:487-494. 19. Mewissen MW, Seabrook GR Mewissen MH, et al. Catheter-directed thrombolysis for lower extremity deep venous thrombosis: report of a national multicenter reristry. Radiology 211:39-49,1999. 20. Chang R, Shawker TH, Chen CD, et al. Technique and outcomes of single daily intraclot injections of recombinant tissue plasminogen activtivator in treatment of acute deep venous thrombosis of the lower extremity. JVIR 2000;11(part 2):305. 21. Grubbs GE. Hemodialysis vascular access: a role for thrombolytic therapy? Clinical Courier 2000;18:7-9.

22. Bookstein JJFellmath B, Roberts A , et al. Pulsed-spray pharmacomechanical thrombolysis: preliminary clinical results. AJR 1989;152:10971100. 23. Cynamon J, Lakritz PS, Wahl SI, et al. Hemodialysis graft declotting: description of the “Lyse and Wait” technique. JVIR 1997;8:825-829. 24. Stevens LC, Haire WD, Kotulak GD. Are clinical signs accurate indicators of the cause of central venous catheter occlusion? J Parent Ent Nutr. 1995;19:75-79. 25. Haire WD, Atkinson JB, Stephens LC. Urokinase versus recombinant tissue plasminogen activator in thrombosed central venous catheters: a double-blinded, randomized trial. Thromb Haemost 1994;72:543-547. 26. Haire WD. Identification and management of thrombotic catheter occlusion. Clinical Courier 2000;18:9-11. 27. Anonymous: Tissue plasminogen activator for acute ischemic stroke. The national Institute of Neurological Disorders and Stroke rt-PA Stroke Study Group. N.Engl J Med 1995;3333:1581-1587. 28. Furlan A, Higashida R, Wechsler L, et al. Intra-arterial prourokinase for acute ischemic stroke. The PROACT II study: a randomized controlled trial. JAMA 1999;282:2003-2011. 29. Samlling RW, Bode C, Kalbfleisch J, et al. More rapid, complete and stable coronary thrombolysis with bolus administration of reteplase compared with alteplase infusion in acute myocardial infarction. Circulation 1995;91:2725-2732. 30. Murphy TP. Thrombolysis on the horizon: new thrombolytic agents and strategies. JVIR 2003; 14:(suppl)P245-249.

Special Session Less Common Non-Vascular Interventions 38.1 Percutaneous Treatment of Hydatid Cysts O. Akhan; Radiology, Hacettepe University, Ankara, Turkey. Hydatic disease caused by Echinococcus granulosus is still an endemic disease in Mediterranean, Middle East and South American countries. The clinical findings and the serologic tests are not specific for the diagnosis and activity of this disease (1,2). Among the imaging modalities, US has a special importance for the diagnosis of liver hydatid disease and is considered the method of choice to classify them (3). The conventional treatment of hydatid liver disease is surgery. This is associated with considerable mortality (up to 6.3%), morbidity (12.5 80%), high recurrence rates and a long hospital stay (uncomplicated cases 14 , complicated cases 30 days). Postoperative recurrence rates are 2.2% , 8.5% and 11.3% respectively in three different series published before 1980. In other series where US and CT were used for follow-up, recurrent disease was found in as many as 10 - 30% of cases (4-11). Medical therapy with mebendazole or albendazole is still controversial and low success rates are observed in the series with long-term follow up. However, these drugs have been used successfully for prophylaxis against spillage before and after surgery or percutaneous treatment. Follow-up criteria of the medical treatment are also controversial (12,13). The percutaneous aspiration and treatment of liver hydatid cysts were considered to be contraindicated due to two main potential risks: anaphylactic shock and abdominal dissemination of the disease. However, in 1980’s, it was understood that some cystic lesions punctured for diagnostic purposes were hydatid cysts and important complications did not occur in these patients. Since the first case percutaneously treated was published, several series of successful percutaneous treatment of the liver thorax, kidney, peritoneum, parotid gland and orbital cavity hydatid cysts have appeared in the literature (14-28). According to Gharbi’s classification, the liver hydatid cysts were classified in five groups and these types are as follows; type I (pure fluid collection), type II (fluid collection with a split wall), type III(fluid collection with septa), type IV (hydatid cysts with heterogeneous echo patterns),and type V (hydatid cysts with reflecting thick walls) (3). The size, number or localization of the hydatid cysts in the liver are not contraindications for percutaneous treatment. The main indications of percutaneous treatment of liver hydatid cysts include type I and type II liver hydatid cysts. The others are some subtypes of type III that do not involve non-drainable solid material, some subtypes of type IV that have considerable amount of fluid component, suspected postoperative fluid collections, the infected hydatid cysts, in the groups such as the

patients rejecting operation, inoperable patients, pregnant patients, patients with multiple or disseminated cysts. The relative contraindications of the treatment of hydatid cysts include some subgroups of Type III that contains of non-drainable solid materials and ruptured liver cysts into the biliary system or peritoneum. These cases should be treated by either surgery or modified catheterization techniques. Any treatment is unnecessary in patients with totally calcified liver hydatid cysts (type V) and some subtypes of type IV liver hydatid cysts which do not contain fluid component (11). Two traditional techniques are used in percutaneous treatment of liver hydatid cysts. The first one is; the PAIR technique described by Ben- Amour et al. in 1986 (16). PAIR is the abbreviation of Puncture, Aspiration of cyst content, Injection of hypertonic saline solution, and Reaspiration of all fluid. After the first puncture is applied by a 18G Seldinger needle (or a smaller one), approximately 25-35% of the estimated volume of the punctured cyst is aspirated. Hypertonic saline (%15 or 20 NaCl) less than 5-10% of the aspirated volume is injected into the cavity. The concentration of the hypertonic saline is increased in the cavity after that. It is known that the hypertonic saline kills all the viable protoscolices in five minutes. It is waited till endocyst is seperated from pericyst. This seperation generally takes place within 10 minutes. Hypertonic saline concentration is increased in the cavity, if needed, in order to provide separation of endocyst from pericyst. After the whole seperation is demonstrated by US, all the fluid is reaspirated from the cavity. The second one is the Catheterization technique with hypertonic saline and alcohol described by Akhan et al. in an experimental study in sheep in 1993 (15). Initially, as it is in the PAIR technique, the cyst is punctured by Seldinger needle and the crystal-clear fluid is aspirated. The hypertonic saline is applied and the endocyst is separated from pericyst wholly, then a 6F pig-tail catheter is put in the cavity by using modified Seldinger technique. This application is realized under the guidance of fluoroscopy. The cavity is washed by hypertonic saline and all the cavity contents are aspirated via this catheter. The catheter is fixed to the skin and left to free drainage for 24 hours. If the amount of free drainage is less than 10cc within 24 hrs, it is indicated that there is no communication between the biliary system and the cavity. This is confirmed by obtaining a cystogram under the fluoroscopic guidance. 95% absolute alcohol is given to the cavity (approximately 25-35% of the volume) and wait for 20 minutes. Later, all of the fluid is reaspirated, and the catheter is withdrawn. The reason why the alcohol is given is to sclerose the cavity and prevent or minimize the possible collection which may exist for a long period. If the amount of drainage in 24 hrs is more than 10 cc and/or containing bile, the catheter is kept in place until the daily amount of drainage is decreased under 10 cc. After that, the cystogram and sclerosis applications are realized as already mentioned. The alcohol application in cysts having biliary communications is contraindicated. Some researchers suggest to treat all the liver hydatid cysts by PAIR technique if indicated. Whereas some others suggest that the cysts smaller than 6cm in diameter (volume less than approximately 100cc) should be treated by PAIR technique, and cysts larger than 6 cm in diameter (volume more than approximately 100cc) should be treated by the catheterization technique. The aim of the sclerosis of cysts that have volumes more than 100 cc by alcohol is to prevent the occurrence of a large cystic collection after the procedure. The large cystic collections are subject to the risks of infection or fistula into the biliary system in the long term. It is accepted that sclerosis by alcohol minimizes these risks. The sclerosis effect of alcohol has already been delineated in the percutaneous treatment of simple renal cysts and lymphocysts (29,30). The absolute alcohol fixes the epithelial lining cells in the cyst wall in 1-3 minutes, but penetrates into the cyst capsule only after 412 hours. For that reason, if the alcohol remains in the cavity longer, some undesired sclerotic effects might occur in the surrounding tissues. The absolute alcohol also has scolecidal effect as well as the effect of destroying daughter cysts besides sclerosis. Therefore, its usage when the hypertonic saline is not sufficient to destroy daughter cysts is also appropriate. Modified catheterization techniques have recently been defined besides two traditional techniques. The aim of this technique is to evacuate not only the fluid component of the cavity but also degenerated membranes of the cyst. Ultrasonography is performed once in every three months during the first year after the procedure and then twice a year at the second year and once for the following years. CT examination is carried out once in one or two years period during the follow-up (11). It is important to extend the follow-up period to be between 5-10 years in order to com-

prehend the long term effects of the procedure. The healing criteria described in an experimental study in sheep include considerable reduction in the size and volume of the cyst, irregularity and thickening of the cyst wall, decrease in the fluid component of the cavity, and solidification of the cyst in time by disappearance of fluid component. This solid remnant is defined as pseudotumor appearance on US. These changes in the treated hydatid cysts occur gradually after the procedure (11,15,19). Major complications include the abdominal dissemination after percutaneous treatment which was not reported in any series published and death due toanaphalactic shock which were reported in two cases (the mortality rate 0.1-0.2%). Minor complications include some allergic reactions such as urticaria, itching and hypotension that may be treated by antihistaminics and fever not more than 38.5 C which does not require any medication. More serious minor complications (10%) include infection in the cavity, fistula between cavity and the biliary system and more serious anaphylactic reaction. The recurrence rates in the reported series is between 0% and 4. Recurrent cysts may also be treated percutaneously. The mean hospitalization period after percutaneous treatment of liver hydatid cysts ranges between 2.5-4.2 days (11,19,23,24) Today, the percutaneous approach has an important role in treatment of liver hydatid cysts with the results of macroscopic and microscopic changes demonstrated after percutaneous treatment and long term follow-up results published in the literature. Therefore, if indicated, percutaneous treatment of liver hydatid cysts is an effective and reliable interventional radiological procedure as an alternative to surgery. References: 1. Dawson JL, Stamatakis JD, Stringer MD, Williams R. Surgical treatment of hepatic hydatid disease. Br J Surg 1988; 75:946-950. 2. Morris DL, Richards KS. Hydatid disease: Current medical and surgical management. London, Butterworth-Heinemann. 1992 3. Gharbi HA, Hassine W, Brauner MW, Dupuch K. Ultrasound examination of the hydatic liver. Radiology 1981; 139:459-463 4. Sayek &Idot_upper;, Yal1n R, Sanaç Y. Surgical treatment of hydatid disease of the liver. Arch Surg 1980; 115:847-850 5. Menteş A. Hydatid liver disease: a perspective in treatment. Dig Dis 1994; 12:150-160 6. Yal1n R, Oğuz M, Y1ld1r1r C, Dülger M. Surgical treatment of hepatic hydatid cyst. Med Principles Pract 1989; 1:154-159 7. Magistrelli P, Masetti R, Coppola R, Messia A, Nuzzo G, Picciocchi A. Surgical treatment of hydatid disease of the liver. Arch Surg 1991; 126:518-523 8. Langer JC, Rose DB, Keystone JS. Diagnosis and management of the hydatid disease of the liver: a 15 year North American experience. Ann Surg 1984; 199:412-417 9. Demirci S, Eraslan Ş, Anadol E, Bozatl1 L. Comparison of the results of the different surgical techniques in the management of the hydatid cysts of the liver. World J Surg 1989; 13:88-91 10. Little JM, Hollands MT, Eckberg H. Recurrence of hydatid disease. World J Surg 1988; 12:700-704 11. Akhan O, Özmen MN. Percutaneous treatment of liver hydatid cysts. European Journal of Radiology 1999; 32:76-85 12. Davis A, Pawlowski ZS, Dixon H. Multicentre clinical trials of benzimidazole carbamates in human echinococcosis. Bull World Health Org 1986; 64:383-388 13. Todorov T, Vutova K, Mechkov G, Petkov D, Nedelkov G, Tonchev Z. Evaluation of response to chemotherapy of human cystic echinococcosis. Br J Radiol 1990; 63:523-531 14. McCorkell SJ. Unintended percutaneous aspiration of pulmonary echinococcal cysts. AJR 1984; 143:123-126 15. Akhan O, Dinçer A, Gököz A, et al. Percutaneous treatment of abdominal hydatid cysts with hypertonic saline and alcohol: an experimental study in sheep. Invest Radiol 1993; 28:121-127 16. Ben Amor N, Gargouri M, Gharbi HA, et al. Traitment du kyste hydatique du foie du mouton par ponction sous échographie. La Tunisie Medicale 1986; 64:325-331 17. Gargouri M, Amor NB, Chehida FB, et al. Percutaneous treatment of hydatid cysts (Echinococcus granulosus) Cardiovasc Intervent Radiol 1990; 13:169-173 18. Khuroo MS, Zargar SA, Mahajan R. Echinococcus granulosus cysts in the liver: management with percutaneous drainage. Radiology 1991; 180:141-145 19. Akhan O, Özmen MN, Dinçer A, Sayek &Idot_upper;, Göçmen A. Liver hydatid disease: long-term results of percutaneous treatment.

Radiology 1996; 198:259-264 20. Khuroo MS, Dar MY, Yattoo GN, Zargar SA, Javaid G, Khan BA, Boda MI. Percutaneous drainage versus albendazole therapy in hepatic hydatidosis: a prospective, randomized study. Gastroenterology 1993; 104:1452-1459 21. Xiaozhi W et al. Clinacal treatment of hepatic and abdominal hydatidosis with percutaneous puncture drainage and curettage (Report of 869 cases). Chin J Parasitol and Parasitic Dis . 1994; 12:285-287 22. Khuroo MS, Wani NA, Javid G, Khan BA, Yattoo GN, Shah AH, Jeelani SG. Percutaneous drainage compared with surgery for hepatic hydatid cysts. N Engl J Med 1997; 337:881-887 23. Üstünsöz B, Akhan O, Kamiloğlu MA, Somuncu &Idot_upper;, Uğurel MŞ, Çetiner S. Percutaneous treatment of hydatid cyst of the liver: long-term results. AJR. 1999; 172:91-96 24. Men S, Hekimoğlu B, Yücesoy C, Arda &Idot_upper;S, Baran &Idot_upper;. Percutaneous treatment of hepatic hydatic cysts: a justified alternative to surgery. AJR. 1999; 172:83-89 25. Akhan O, Özmen MN, Dinçer A, Göçmen A, Kalyoncu F. Percutaneous treatment of pulmonary hydatid cysts. Cardiovasc Intervent Radiol 1994; 17:271-275 26. Akhan O, Üstünsöz B, Somuncu &Idot_upper;, Özmen M, Öner A, Alemdaroğlu A, Besim A. Percutaneous renal hydatid cyst treatment: long-term results. Abdom Imaging 1998; 23:209-213 27. Akhan O, Bilgiç S, Akata D, K1ratl1 H, Özmen MN. Percutaneous treatment of an orbital hydatid cyst: a new therapeautic approach. Am J Ophthalmol 1998;125:877-879 28. Akhan O, Ensari S, Özmen M. Percutaneous treatment of a parotid gland hydatid cyst: a possible alternative to surgery European Radiology 2002; 12:597-599 29. Bean WJ, Rodan BA. Hepatic cysts: treatment with alcohol. AJR 1985; 144:237-241 30. Akhan O, Çekirge S, Özmen MN, Besim A. Percutaneous transcatheter sclerotherapy of postoperative pelvic lymphoceles. Cardiovasc Intervent Radiol 1992; 15:224-227

38.2 Dacryocystoplasty H. Song; Radiology, Asan Medical Center, University of Ulsan College of Medicine, Seoul, Republic of Korea. Introduction The lacrimal drainage system includes the superior and inferior punctum, superior and inferior canaliculi, common canaliculus, lacrimal sac, and nasolacrimal duct (1). When there is an obstruction in the lacrimal drainage system, drainage of tears into the inferior meatus of the nasal cavity is not sufficient. As a result, the tears fall over the lid margin onto the cheek, which is termed epiphora (2). While there are many causes of lacrimal outflow obstructions such as congenital cause or infection, the majority of cases are due to idiopathic inflammation and scarring of the nasolacrimal duct (2). Congenital obstruction of the lacrimal drainage system is a relatively common clinical problem, affecting as many as 20% of all infants (3). The most common form of congenital obstruction is caused by a persistent layer of lacrimal and nasal epithelial cells at the level of the valve of Hasner (4). Congenital obstruction responds well to conservative treatment or to probing, if done in the first year of life (4). After 24 months of age, probing fails in up to 67% of cases (5). Historically, external dacryocystorhinostomy (DCR) has been the standard treatment for acquired lacrimal drainage system obstructions below the common canaliculus, yielding success rates of 89%-95% in primary repairs (6-8). DCR is edge-to-edge anastomosis of the lacrimal sac mucosa to nasal mucosa over the margins of a hole made through the lacrimal bone. The invasive surgical procedure usually requires general anesthesia and leaves a facial scar that is permanent, even though minimal. Although complications of DCR are uncommon, they can include a hyperthropic facial scar and regrowth of mucous membrane over the nasolacrimal opening (8). Recently, several studies (9,10) have investigated adjunctive mitomycin-C for DCR or probing to confirm improved success rate. Mitomycin-C, an antineoplastic antibiotic isolated from Streptomyces caespitosus, acts as a potent fibroblastic inhibitor and can selectively inhibit the synthesis of DNA, cellular RNA, and protein. The conventional methods of treatment for acquired canalicular stenosis can be divided into two basic types: 1) invasive surgical procedures such as conjuntival dacryocystorhinostomy (C-DCR) with permanent

bypass tube placement (11-13), and 2) nonsurgical procedures such as temporary silicone tube placement (14). In C-DCR, a conventional DCR is performed and a tract is made from the conjunctival surface to the internal anastomosis, creating a continuous tract from the conjunctiva to the nasal cavity. A bypass tube is permanently inserted into the tract. The reported success rate has been variable, from 57% to 98% (11, 12). C-DCR usually requires general anesthesia and leaves a permanent facial scar. Rose and Welham (12) reported that, during follow-up of 23 years after C-DCR with the Lester Jones bypass tube in 326 eyes of 310 patients with canalicular obstruction, replacement of the bypass tube was needed in 44% of the cases due to spontaneous loss of the tubes. Steinsapir et al (13) reviewed 79 C-DCR cases performed over a 16-year period in 75 patients. They reported complication rates of extrusion in 51%, obstruction in 23%, medial migration in 9%, malposition in 9%, hypermobility of the tube in 4%, granuloma formation in 7%, and infection in 3%. Although probing is effective for treatment of congenital stenosis of the lacrimal system, it is usually ineffective after the age of 2 years (8). Temporary silicone tube placement without DCR has not met with widespread acceptance or universal success, probably because the manipulation of probes down the narrow bony canal of the nasolacrimal duct leads to further fibrosis and aggravates the obstruction (11). In addition, silicone tube placement is disagreeable because the tubes must be left in place for long periods, which may cause complications such as recurrent dacryocystitis, granuloma formation, and canaliculi erosion (15). As a means of overcoming the problems of DCR, C-DCR, and silicone intubation, interventional procedures such as balloon dacryocystoplasty and stent placement have been developed for the treatment of epiphora secondary to obstruction of the lacrimal drainage system. Since Becker and Berry (6) reported dilation of the lacrimal system using an angioplasty balloon catheter in 1989, there have been many reports (15-28) stating that the procedure is safe and easy for the treatment of the lacrimal system. Technique Balloon dilation can be performed as an outpatient procedure under local anesthesia except for children who are not cooperative with the procedure. The site and severity of the obstruction are usually evaluated before balloon dilation by means of dacryocystography. Under fluoroscopic guidance, a guide wire is introduced through the punctum across the obstruction into the inferior meatus of the nasal cavity. As the ampulla portions of the lacrimal systems are vertically oriented for about 2 mm, the guide wire is first introduced vertically into the punctum and then rotated 90º horizontally to conform to the bend in the first portion of the canaliculus. In patients with incomplete obstructions of the lacrimal drainage system, one can use a 0.014-inch flexible guide wire to negotiate the obstruction (16,17). In patients with complete obstructions, an 0.018-inch ball-tipped guide wire is useful for opening the complete obstruction (18,19). A hook or a hemostat is placed in the nasal cavity and aimed laterally toward the inferior meatus to pull the tip of the guide wire out of the external naris. A deflated balloon catheter is then advanced over the guide wire and positioned across the obstruction. Dilatation is performed by inflating the balloon with water-soluble contrast media. The balloon catheter and the guide wire are removed from the lacrimal drainage system. Dacryocystography can be performed immediately after the procedure to assess the patency of the lacrimal system. Becker and Berry (6) introduced a 3-4 mm coronary angioplasty balloon catheter through the canaliculus in an antegrade approach, whereas Munk et al (15) introduced a 3-4 mm tibial angioplasty balloon catheter through the inferior opening of the nasolacrimal duct in a retrograde approach. The retrograde approach seems to be safer than the antegrade approach, not only because the antegrade approach imposes limits on the size of the balloon that can be introduced into the lacrimal drainage system, but also because the canaliculus can be lacerated during insertion or removal of the balloon catheter (18,20). The duration of inflation of the balloon catheter ranges from 20 seconds to 5 minutes (15-28). Janssen et al (16) emphasized that the balloon catheter should be inflated for a short period, such as 30 seconds, to prevent severe damage to the lacrimal drainage system. Results and Discussion 1. Congenital obstruction Becker et al (26) reported a 95% initial improvement rate of epiphora in 61 eyes with congenital lacrimal drainage system obstruction. We performed balloon dilation in 20 eyes of 16 patients with an age range of 12-78 months (mean, 33 months) for congenital lacrimal drainage system obstruction (20). The technical success rate and initial improve-

ment rate were both 95%. No patients showed recurrence during the mean follow-up of two years. A variety of treatments have been described for congenital lacrimal drainage system obstruction, including local massage, probing, silicone intubation, and DCR (5, 20,26, 28-30). After 24 months of age, probing fails in up to 67% of cases (5). Silicone intubation is an effective treatment in children in whom probing fails. However, complications such as slitting of the canaliculi and early removal of the tube by the child may occur (29,30). The No.0 Bowman probe is 0.71 mm and the silicone tube 0.64 mm to 1.19 mm in diameter. In contrast, the inflated balloon catheter is 30 mm in diameter. Therefore, it is not surprising that the initial and long-term improvement rates of balloon dilation are greater than those of probing or silicone intubation. 2. Acquired obstruction Technical success rates of 89 - 95% have been reported (6,15-28). The technical failure rate was higher in patients with traumatic obstruction than in those with idiopathic obstruction. No major complications have been reported. A majority of the patients express mild pain during the inflation of the balloon and complain of slightly blood-tinged nasal discharge for 1 to 72 hours after the procedure. Ilgit et al (25) reported a 69% initial improvement rate of epiphora in 80 eyes, while Janssen et al (28) reported a 96% initial improvement rate and an 84% 1-year improvement rate in 100 eyes. According to our experience with 430 eyes of 350 patients (22), the technical success rate and the overall initial improvement rates were 95% and 57%, respectively. The 2-month, 1-year, and 5-year improvement rates were 48%, 39%, and 37%, respectively. A problem in comparing these studies is that a variety of treatment techniques and methods of analysis have been utilized. There may be different factors contributing to results, including factors such as patient selection, negotiation technique, balloon diameter, balloon inflation time, and length of the follow-up period. As many authors have already described, initial and long-term improvement rates were greatest in the cases of lesions with obstructions of the nasolacrimal duct. We found that the long-term improvement rate of the canalicular obstructions was also high (22) and that the initial improvement rate was influenced by the severity (p=.014) and the site (p<.001) of the obstruction, and the diameter of the balloon (p=.047). The cause of obstruction has some impact on the results, although the cause of the epiphora in most reported cases was idiopathic. Recently, we started irrigation of the lacrimal system with mitomycin-C at a low concentration (0.2 mg/ml) after dacryocystoplasty to investigate its effectiveness and safety in the treatment of lacrimal system obstruction. Conclusion The technique of dacryocystoplasty is simple, safe, and cost-effective. It causes no facial scars and fewer problems with bleeding, and patients need less postoperative care. Dacryocystoplasty seems to be of as an initial therapy for children with congenital obstruction in whom local massage fails, regardless of their age. The initial success rate of balloon dilation is relatively low in cases with acquired obstruction of the lacrimal drainage system. In cases that have initial success, however, a relatively high long-term success rate can be expected. References: 1. McCord CD. The lacrimal drainage system. IN: Duane TD, Jager EA, eds. Clinical ophthalmology. Vol 4. New York, NY: Haper & Row,1985;124. 2. Linberg JV, McCormick SA. Primary acquired nasolacrimal duct obstruction: a clinicopathologic report and biopsy technique. Ophthalmology 1986;93:1055-1063. 3. MacEwen CJ, Young JDH. Epiphora during the first year of life. Eye 1991;5:596-600. 4. Kushner BJ. Congenital nasolacrimal system obstruction. Arch Ophthalmol 1982;100:597-600. 5. Katowitz JA, Welsh MG. Timing of initial probing and irrigation in congenital nasolacrimal duct obstruction. Ophthalmology 1987;94:698705. 6. Becker BB, Berry FD. Balloon catheter dilatation in lacrimal surgery. Ophthalmic Surg 1989;20:193-198. 7. Welham RAN, Henderson P. Failed dacryocystorhinostomy. Trans Am Acad Ophthalmol Otolaryngol 1974;78:824-828. 8. Thornton SP. Nasolacrimal duct reconstruction with the nasolacrimal duct prosthesis: an alternative to standard dacryocystorhinostomy. Ann Ophthalmol 1977;9:1575-1582. 9. Yeatts RP, Neves RB. Use of mitomycin C in repeat dacryocystorhinostomy. Ophthalmic Plast Reconstr Surg 1999;15:19-22. 10. Tsai CC, Kau HC, Kao SC, Hsu WM, Liu JH. Efficacy of probing the nasolacrimal duct with adjunctive Mitomycin-C for epiphora in adults.

Ophthalmology 2002;109:172-174. 11. Katowitz JA. Silicone tubing in canalicular obstructions. A preliminary report. Arch Ophthalmol 1974;91:459-462. 12. Rose GE, Welham RA. Jones' lacrimal canalicular bypass tubes: twenty-five years' experience. Eye 1991;5:13-19. 13. Steinsapir KD, Glatt HJ, Putterman AM. A 16-year study of conjunctival dacryocystorhinostomy. Am J Ophthalmol 1990;109:387-393. 14. Lauring L. Silicone intubation of the lacrimal system : pitfalls, problems and complications. Ann Ophthamol 1976;8: 489-498. 15. Munk PL, Lin DTC, Morris DC. Epiphora: treatment by means of dacryocystoplasty with balloon dilation of the nasolacrimal drainage apparatus. Radiology 1990;177:687-690. 16. Janssen AG, Mansour K, Krabbe GJ, et al. Dacryocystoplasty: treatment of epiphora by means of balloon dilation of the obstructed nasolacrimal duct system. Radiology 1994;193:453-456. 17. Wilhelm KE, Hofer U, Textor HJ, Bőker T, Strunk HM, Schild HH. Dacryoliths: nonsurgical fluoroscopically guided treatment during dacryocystoplasty. Radiology 1999;212:365-370. 18. Song HY, Ahn HS, Park CK, et al. Complete obstruction of the nasolacrimal system. Part I. Treatment with balloon dilatation. Radiology 1993;186:367-371. 19. Lee JM, Song HY, Han YM, Chung GH, Sohn MH, Kim CS, Choi KC. Balloon dacryocystoplasty: results in the treatment of complete and partial obstructions of the nasolacrimal system. Radiology 1994;192: 503-508. 20. Cho YS, Song HY, Ko GY, Yoon CH, Ahn HS, Yoon HK, Sung KB. Congenital lacrimal system obstruction: treatment with balloon dilation. J Vasc Interv Radiol 2000;11:1319-1324. 21. Ko GY, Lee DH, Ahn HS, Yoon HK, Sung KB, Song HY. Balloon catheter dilation in common canalicular obstruction of the lacrimal system: safety and long-term effectiveness. Radiology 2000;214:781-786. 22. Lee DH, Song HY, Ahn HS, et al. Balloon dacryocystoplasty: results and factors influencing outcome in 350 patients. J Vasc Interv Radiol 2001;12:500-506. 23. Song HY, Lee CH, Park SS, et al. Lacrimal canaliculus obstruction: safety and effectiveness of balloon dilation. J Vasc Interv Radiol 1996;7:929-934. 24. Becker BB, Berry FD. Balloon catheter dilatation in pediatric patients. Ophthalmic Surg 1991;22:750-752. 25. Ilgit ET, Yuksel D, Unal M, et al. Transluminal balloon dilatation of the lacrimal drainage system for the treatment of epiphora. AJR Am J Roentgenol 1995;165:1517-1524. 26. Becker BB, Berry FD, Koller H. Balloon catheter dilatation for treatment of congenital nasolacrimal duct obstruction. Am J Ophthalmol 1996;121:304-309. 27. Berkefeld J, Kirchner J, Müller HM, Fries U, Kollath J. Balloon dacryocystoplasty: indications and contraindications. Radiology 1997;205:785-790. 28. Janssen AG, Mansour K, Bos JJ. Obstructed nasolacrimal duct system in epiphora: long-term results of dacryocystoplasty by means of balloon dilation. Radiology 1997;205:791-796. 29. Migliori ME, Putterman AM. Silicone intubation for treatment of congenital lacrimal duct obstruction: successful results removing the tubes after six weeks. Ophthalmology 1988;95:792-795. 30. Leone CR, Van Gemert JV. The success rate of silicone intubation in congenital nasolacrimal duct obstruction. Ophthalmic Surg 1990;21:90-92.

Special Session Biliary Interventions II 41.1 The interventional radiological management of cholangiocarcinoma A. Adam; Interventional Radiology, Guy's and St. Thomas' Hospital, London, United Kingdom. (This abstract is entirely based on a Review Article by A. Hatzidakis and A. Adam published in Clinical Radiology in 2003 , Volume 58, p 91) Introduction Cholangiocarcinomas are relatively rare tumours with an autopsy incidence of 0.5 - 2 % and 4500 new cases annually in the USA. The incidence increases with age, with most tumours occurring in between 5070 years, commonly in men. Histologically, the vast majority of cholangiocarcinomas are adenocar-

cinomas and they are topographically differentiated as intra- and extrahepatic. Intrahepatic lesions are classified according to the macroscopic types, as mass forming, periductal-infiltrating and intraductal, whereas extrahepatic tumours are divided as hilar and distal. Intrahepatic tumours usually do not cause extensive bile duct dilatation, so that jaundice, as a primary symptom is rare. They can present as a large peripheral mass, a cystic lesion or an intraductal tumour with segmental bile duct dilatation. Extrahepatic tumours usually present with obstructive jaundice due to obstruction at the hilum at common bile duct (CBD) level. Up to two thirds of all cholangiocarcinomas are located at or near the hilum. Diagnostic Imaging Current imaging techniques effectively evaluate the location and extent of the tumour, permitting accurate staging. Ultrasonography, Computed Tomography and Magnetic Resonance Imaging can provide valuable information about the extent tumour, the level of the obstruction, metastatic disease and infiltration or encasement of vascular structures. The decision whether the tumour is resectable is based on the imaging information and the result of any biopsies taken. Surgical treatment The best chance of cure is offered by complete surgical resection with negative margins and restoration of biliary-enteric continuity. Resectability criteria are based on tumour extension, vascular involvement, distant metastases, the presence of liver cirrhosis and dysfunction, portal hypertension, and the general condition of the patient. Cholangiocarcinomas are resectable if vascular and biliary ductal involvement is limited to one lobe of the liver and if there is no extrahepatic disease in patients who are fit for surgery. Extended right hepatic resection is possible in some patients with involvement of the quadrate lobe. In patients in whom the proportion of the liver to be resected is substantial it may be useful to carry out embolization of the appropriate branch of the portal vein, in order to achieve atrophy of the lobe to be resected and at the same time compensatory hypertrophy of the remaining parenchyma, minimizing the risk of post-operative liver failure. If pre-operative investigations do not definitively demonstrate an irresectable tumour, surgical exploration can be performed. Some surgeons advocate the use of staging laparoscopy in order to avoid exploratory laparotomy. Hilar tumours present a challenge to surgeons because of their proximity to the portal vein, hepatic arteries and liver parenchyma. Nevertheless, large Japanese series report resectability rates as high as 80% for patients treated by combined bile duct and liver resection with caudate lobectomy. Other centres report much lower rates of 10 - 56% . In many cases, tumours thought to be resectable, are shown to be irresectable at the time of surgery, so that only a palliative surgical bypass is performed. If curative resection is not possible, surgical, endoscopic or percutaneous methods of palliation can be used. Palliative surgical bypass has a procedure-related mortality of 13.6%, an overall 30-day mortality of 7% for all types of lesions and, in patients with hilar tumours, a postoperative complication rate of approximately 20%. Curative resection of hilar tumours is associated with an operative mortality rate of 5 – 7% and a postoperative complication rate of 22%. The median survival after resection of hilar cholangiocarcinoma is 27-59 months, with 5- and 10-year survival rates ranging between 0 - 56% and 0 - 18%, respectively. Other authors report a lower overall median survival rate of 14 to 19 months. Hilar tumours of Bismuth types I and II (Table 1) have a relatively better 5-year survival rate (14 - 33%) in comparison to type III and IV tumours (0 - 25%). Survival is lower if there is local lymph node involvement, but it still remains longer than for the unresectable patients. The management of patients with hilar tumours is decided according to the local expertise and is based on a number of factors, including age, performance status, other co morbid conditions, location of the tumour, and resectability criteria. The 5-year survival rate after curative resection of distal extrahepatic tumours, varies between 0-39% (mean 25%). Peripheral cholangiocarcinomas show a post resection median survival of about 18 months. Adjuvant treatment The benefit of external-beam radiotherapy or intra-arterial local chemotherapy is uncertain. Preoperative treatment using external-beam irradiation in combination with bolus 5-FU chemotherapy, then followed by brachytherapy plus protracted venous infusion of 5-FU, may increase the survival rate. Also Ir192 brachytherapy may restrict the tumour spread. Other authors believe that intraluminal brachytherapy does not provide any advantage compared to external-beam irradiation. Photodynamic therapy is a safe, minimally invasive palliative therapy, which can be effective in reducing malignant stenosis. It also seems to

lead to a significant decrease in serum bilirubin, longer survival and improved quality of life in patients with nonresectable Bismuth type III and IV tumours. Endoscopic treatment Endoscopic biliary drainage with stent placement is a widely used method of palliation. Endoscopic retrograde cholangio-pancreatography (ERCP) has several advantages over percutaneous transhepatic cholangiography (PTC) in distal common bile duct lesions. ERCP provides direct inspection of the duodenum and the papilla, can help to differentiate malignant from benign causes, and allows cannulation of the stricture. Cytologic samples can be easily obtained, with subsequent placement of a plastic stent to relieve the patient’s jaundice. If endoscopic stent placement fails, percutaneous transhepatic biliary drainage (PTBD) or a combined endoscopic-radiologic procedure can be performed. Failure of endoscopic drainage may be due to anatomic alterations, a periampullary diverticulum, duodenal obstruction by adjacent tumour, failure to cannulate the biliary tree, or inability to advance the guide wire or push a stent through the stricture. Nevertheless, plastic stents are inserted successfully is approximately 90-95% of patients. The procedure is more often successful in patients with distal rather than proximal lesions. Endoscopic management for proximal bile duct strictures has a lower overall success rate than percutaneous treatment (19% vs. 97%) with comparable complication rates (26% vs. 25%). There is only a single study that reported a much higher endoscopic success rate (81 - 89%) compared with that for the percutaneous approach (57-83%). Percutaneous treatment Many authors favour the percutaneous approach over the endoscopic approach for the palliation of patients with malignant hilar biliary obstruction because the percutaneous technique provides better demonstration of the proximal extent of the tumour and allows easier placement of drainage catheters. Endoscopic stent insertion is often successful in type I hilar tumours, but the percutaneous route is preferable in type II-IV lesions. Endoscopic stent placement is particularly problematic in type III and IV strictures in which successful drainage is achieved in only 15% of the patients. Nevertheless, there is one report of successful endoscopic plastic stent placement in 100% of patients with type III and IV tumours, followed by stent exchange every four months. The palliative management of hilar lesions is strongly dependent on individual expertise. Percutaneous transhepatic transluminal forceps biopsy is a safe technique, which is easy to perform through an existing biliary track, producing a diagnostic result in 71% of cases. Percutaneous cholangioscopic endoluminal forceps biopsy has a sensitivity of 89%, and is appropriate even for very small intraductal lesions. Placement of metallic stents does not preclude subsequent biopsy. Therefore, the requirement for histology is not in itself a contraindication to the placement of metallic endoprostheses. Plastic stents Most endoscopists use plastic stents to achieve biliary drainage. Plastic devices should be used in preference to metallic endoprostheses in patients being drained percutaneously when a definitive diagnosis of malignancy has not been made or when there is extensive malignant infiltration of the duodenum. However, metallic stents are generally preferable to plastic devices for percutaneous drainage, especially if the tumour is quite advanced and considered unresectable, and can be placed as a one stage procedure. In patients in whom metallic stents protruding into the duodenum have become occluded, patency can often be restored by endoscopic insertion of a plastic endoprosthesis. Plastic stent types A large number of different types of plastic endoprostheses, such as Carey-Coons and Miller stents, are available. Plastic stents usually become occluded because of bile encrustation. The frequency of occlusion is inversely proportional to the size of the stent lumen. However, large-diameter tracks are necessary for transhepatic placement of large-calibre plastic stents, increasing patient discomfort and the risk of haemobilia. Such devices are usually inserted in two stages, increasing the patient’s stay in hospital and the accompanying costs. Metallic stents Metallic stents have revolutionized the interventional radiological management of patients with malignant obstructive jaundice. They can be inserted in a single-stage procedure in most patients, minimizing hospital stay and reducing the associated costs, even though the cost of metallic stents are significantly higher than plastic stents. Patients with life expectancy of more than three months are particularly good candi-

dates for an expandable metallic stent but it is difficult to predict which patients are likely to survive to stent exchange. The main disadvantage of metallic endoprostheses is that they cannot be removed. Therefore, staging and resectability should be determined before stent placement, as this could compromise subsequent surgery. Metallic Stent types The majority of the stents used are self-expandable devices, such as the Wallstent (Boston Scientific, Watertown, MA, USA), and the Zilver stent (Cook, Bloomington, IN, USA). Dilatation of the stricture prior to stent deployment may be helpful. Balloon-expandable stents are infrequently used in the biliary system, because their relatively high rigidity is a significant disadvantage, especially when they have to be deployed along a curve, which is very frequently necessary in patients with hilar lesions. Stenting procedure After percutaneous cannulation of the bile ducts and passing a hydrophilic guidewire across the obstructing lesion, the stricture can be carefully dilated with an 8 - 10 mm balloon. This allows easier passage of the stent delivery catheter and accelerates full expansion of the metallic stent. The proximal end of the stent should be placed in a peripheral intrahepatic duct, in order to minimize the risk of occlusion by tumour overgrowth. The distal part of the stent should protrude about one cm into the duodenum to allow easy endoscopic access, in case it is needed for the restoration of patency of occluded stents. In patients with hilar tumours obstructing both the left and the right hepatic ducts, bilateral stent insertion is usually indicated. If the contralateral duct system is not opacified during percutaneous cholangiography it may be sufficient to drain a single lobe, particularly if staging CT/MRI demonstrates atrophy of the non opacified lobe of the liver. However, if the patient’s jaundice fails to resolve, or if there is infection of the contralateral lobe, bilateral drainage should be carried out. Deployment of two stents can be achieved either through a single transhepatic tract or by using two separate punctures to gain access to each lobe. The latter arrangement is preferable as parallel deployment of two metallic stents allows easier recanalization in cases of obstruction. Non-surgical complications Complications of percutaneous treatment include cholangitis (5 6.5%), haemorrhage (2%), bile leakage (< 2%), abscess, and catheter dislodgment. Minor complications are seen in 10 - 23% of the cases, while the incidence of major ones ranges from 2.3 - 20.8%. Most complications are related to the transhepatic puncture than the stent placement. Endoscopic complications include all the above-mentioned ones, plus acute pancreatitis and bowel perforation. Aggressive guide wire and catheter manipulations, as well as vigorous filling of multiple undrained segments with contrast during cholangiography, can cause cholangitis and septicaemia, especially if stent insertion is unsuccessful. Incomplete biliary drainage after ERCP leads to cholangitis in up to 87% of the patients. Septicaemia may be caused by bacteria, already present in obstructed ducts, which enter the blood stream during the procedure because of the already high biliary pressure, which rises even further following injection of contrast medium. Management of cholangitis When imaging studies suggest obstruction proximal to the distal third of the extrahepatic bile duct percutaneous drainage is the preferred method of palliation because of it is much more likely to be completed successfully than endoscopic drainage thus avoiding the risk of septicaemia due to a failed attempt at endoscopic stent insertion. If PTC demonstrates biliary obstruction it is desirable that drainage is carried out, in order to prevent cholangitis and sepsis. However PTC does not contaminate the bile ducts with enteric flora, as ERCP does, and is less likely to lead to sepsis. Late complications due to non-obstructive cholangitis have been reported in 3% of the patients. Ensuring that the stent extends through the sphincter of Oddi minimizes the risk of late non-obstructive cholangitis and reduces post-procedural morbidity. Stent-related complications include malpositioning, migration, inadequate expansion, failure of release of the stent, duodenal erosion or perforation when stent protrudes too much into the bowel, and stent occlusion. The 30-day mortality is 6 - 39%, depending on the patient’s general condition, the tumour stage and the therapeutic method used, whereas the procedure related death rate ranges between 0.8 and 3.4%. Preoperative biliary drainage Endoscopic stents inserted pre-operatively may increase the incidence of bile contamination and lead to a higher postoperative infection rate. Except for patients with acute cholangitis, in whom emergency drainage is mandatory, preoperative biliary drainage is not indicated in lower bile duct obstruction. Some authors have reported that preoperative

PTBD reduces the incidence of postoperative sepsis, bleeding and renal failure. However, several prospective, randomized trials have shown that preoperative PTBD has no effect on mortality, morbidity or duration of hospitalization. Stenting results and Stent patency Metallic stents can be inserted successfully in up to 100% of patients. Survival after metallic stent placement has been reported as ranging between 93 and 420 days, depending on patient population, tumour location and stage. In bifurcation tumours, the longest survival rate has been observed in patients in whom both lobes have been drained and the worst in those with cholangiographic opacification of both lobes but drainage of only one lobe. Stent occlusion is usually caused by tumour overgrowth (2.4–16%) and less frequently by tumour ingrowth (2.4-7%) or bile encrustation. This condition may cause jaundice and cholangitis and should be managed by insertion of a new endoprosthesis to relieve the obstruction. Plastic stent become occluded after 3-6 months with an average patency of 126 days in comparison with a patency of 273 days for metallic selfexpandable stents. The use of metallic stents reduces hospital stay and minimizes costs. The patency of metallic endoprostheses at 25 and 50 weeks in patients with CBD obstruction was 74% and 59% respectively, and 62 % and 47% respectively in patients with hilar obstruction. The twelve-month patency was 46% for hilar and 89% for non-hilar obstructions, with an overall 12-month survival of 35%. The reintervention rate was 18 19.2% after a mean period of 5.9 months. In a prospective, randomized trial comparing percutaneously placed metallic stents with plastic stents in patients with malignant tumours at the hilum of the liver, metallic stents were found to have higher technical success and patency rates to be more cost-effective than plastic stents. Covered metallic stents offer no significant improvement in patency in comparison to uncovered stents. Stent occlusion due to encrustation of bile is uncommon, because of the large stent diameter. Covered stents prevent tumour ingrowth, but overgrowth remains a problem. If the covering membrane is damaged, ingrowth is possible through the tear. The stent cover can be damaged during deployment or through physical and mechanical factors, such as increasing pressure from the tumour on the covering membrane, corrosion due to reflex of duodenal and gastric juices. The frequency of occlusion due to tumour overgrowth can be minimized by peripheral stent placement but, this may not be possible with covered stents, as it may lead to the inadvertent occlusion of intrahepatic side branches. Nevertheless the possible future role of impregnated stents has yet to be determined. Cystic or pancreatic duct obstruction by covered stents usually does not result in cholecystitis or pancreatitis. Summary The management of patients with cholangiocarcinoma presents great challenges for several medical subspecialty disciplines. It requires a high degree of expertise in diagnostic imaging techniques, as well as endoscopic, interventional radiological and surgical skills. A dedicated team approach is essential. Early diagnosis and accurate staging is the key to reaching the correct decision regarding the resectability of the tumour. If resection is not possible, an endoscopic approach should be used for palliation of distal lesions, with percutaneous intervention reserved for endoscopic failures. However, in hilar lesions interventional radiological palliation is the method of choice, as endoscopic attempts at drainage frequently fail, leading to cholangitis. Tables Table 1 The Bismuth classification of biliary strictures - Type I Stricture of the common bile duct, at least 2 cm distal to the bifurcation - Type II Stricture involving the proximal common hepatic duct - Type III Stricture involving the confluence of the left and right main bile ducts but with remaining communication between the two lobes of the liver. - Type IV Stricture involving the confluence of the left and right main bile ducts with lack of communication between the two lobes of the liver. References: 1. Zimmermann A. Tumours of the bile duct – Pathologic aspects, In: Blumgart LH, Fong Y, ed. Surgery of the liver and Biliary Tract. 3rd ed. London, WB Saunders 2000; 953-975. 2. Blumgart LH, Benjamin IS. Cancer of the bile ducts. In: Blumgart LH,

ed. Surgery of the liver and Biliary Tract. 2nd ed. Edinburgh, Churchill Livingstone 1994; 967-995. 3. Chamberlain RS, Blumgart LH. Hilar cholangiocarcinoma: a review and commentary. Ann Surg Oncol 2000; 7: 55-66 4. Adam A, Benjamin IS. The staging of cholangiocarcinoma. Clin Radiol 1992; 46: 299-303 5. Nimura Y, Kamiya J, Kondo S, et al. Aggressive preoperative management and extended surgery for hilar cholangiocarcinoma: Nagoya experience. J Hepatobiliary Pancreat Surg 2000; 7: 155-162 6. Bismuth H, Castaing D, Traynor O. Resection or palliation: priority of surgery in treatment of hilar cancer. World J Surg 1988; 12: 39-47 7. Lillemoe KD, Cameron JL. Surgery for hilar cholangiocarcinoma: the Johns Hopkins approach. J Hepatobiliary Pancreat Surgery 2000; 7: 115121 8. Smith AC, Dowsett JF, Russell RCG, Hatfield ARW, Cotton PB. Randomised trial of endoscopic stenting versus surgical bypass in malignant low bile duct obstruction. Lancet 1994; 344: 1655-1660 9. Byrnes V, Afdhal N. Cholangiocarcinoma of the hepatic hilum (Klatskin tumor). Curr Treat Options Gastroenterol 2002; 5: 87-94 10. Hausegger KA, Thurnher S, Bodendorfer G, et al. Treatment of malignant biliary obstruction with polyurethane-covered Wallstents. AJR 1998; 170: 403-408 11. Bathe OF, Pacheco JT, Ossi PB, et al. Management of hilar bile duct carcinoma. Hepatogastroenterology 2001; 48: 1289-1294 12. Figueras J, Llado L, Valls C, et al. Changing strategies in diagnosis and management of hilar cholangiocarcinoma. Liver Transpl 2000; 6: 786-794 13. Rumalla A, Baron TH. Evaluation and endoscopic palliation of cholangio-carcinoma. Management of cholangiocarcinoma. Dig Dis 1999; 17: 194-200 14. Torok N, Gores GJ. Cholangiocarcinoma. Sem Gastrointest Dis 2001; 12: 125-132 15. Zoepf T, Jakobs R, Arnold JC, Apel D, Rosenbaum A, Riemann JF. Photodynamic therapy for palliation of nonresectable bile duct cancer – preliminary results with a new diode laser system. Am J Gastroenterology 2001; 96: 2093-2097 16. Sherman S. Current status of endoscopic pancreaticobiliary interventions. J Vasc Intervent Radiol 2001; 12: 140-155 (Suppl) 17. England RE, Martin DF. Endoscopic and percutaneous intervention in malignant obstructive jaundice. Cardiovasc Intervent Radiol 1996; 19: 381-387 18. Sutton R, Slavin JP, Neoptolemos JP. Invited commentary in: Cowling MG, Adam A. Internal stenting in malignant biliary obstruction. World J Surg 2001; 25: 355-361 19. Nelsen KN, Kastan DJ, Shetty PC, Burke MW, Sharma RP, Venugopal C. Utilization pattern and efficacy of nonsurgical techniques to establish drainage for high biliary obstruction. J Vasc Intervent Radiol 1996; 7: 751-756 20. Cowling MG, Adam A. Internal stenting in malignant biliary obstruction. World J Surg 2001; 25: 355-361 21. Hwang MH, Tsai CC, Chou CY, et al. Percutaneous cholangiofiberoscopic endoluminal forceps biopsy of intrabile duct diseases. Hepatogastrenterology 1998; 45: 2073-2078 22. Adam A, Chetty N, Roddie M, Yeung E, Benjamin IS. Self-expandable stainless steel endoprostheses for treatment of malignant bile duct obstruction. American Journal of Roentgenology 1991; 156:321-325 23. Lammer J, Hausegger KA, Fluckiger F, et al. Common bile duct obstruction due to malignancy: treatment with plastic versus metal stents. Radiology 1996; 201: 167-72 24. Schima W, Prokesch R, Oesterreicher C, et al. Biliary Wallstent endoprosthesis in malignant hilar obstruction: long-term results with regard to the type of obstruction. Clin Radiol 1997; 52: 213-219 25. Rossi P, Bezzi M, Rossi M, et al. Metallic stents in malignant biliary obstruction: results of a multicenter European study of 240 patients. J Vasc Intervent Radiol 1994; 5: 279-285 26. Hatzidakis AA, Tsetis D, Chrysou E, Sanidas E, Petrakis J, Gourtsoyiannis NC. Nitinol stents for palliative treatment of malignant obstructive jaundice. Should we stent the sphincter of Oddi in every case? Cardiovasc Intervent Radiol 2001; 24: 245-248 27. Nicholson AA, Royston CMS. Palliation of inoperable biliary obstruction with self-expanding metallic endoprostheses: a review of 77 patients. Clin Radiol 1993; 47: 245-250 28. Davids PHP, Groen AK, Rauws EA, Tytgat GNJ, Huibregtse K. Randomized trial of self-expanding metal stents versus polyethylene stents for distal malignant biliary obstruction. Lancet 1992; 340: 1488-1492

29. Wagner HJ, Knyrim K, Vakil N, Klose KJ. Plastic endoprostheses versus metal stents in the palliative treatment of malignant hilar biliary obstruction: A prospective randomized trial. Endoscopy 1993; 25: 213218

6.1.1

6.1.3

Carotid artery stenting: three years' experience F. Fanelli, F. M. Salvatori, M. Bezzi, S. Vagnarelli, G. Marcelli, D. Pepino, P. Rossi, R. Passariello; Department of radiological sciences, University of Rome, Rome, Italy.

Carotid stenting: personal experience with filter neuroprotection T. Jargiello, M. Szajner, A. Drelich-Zbroja, M. Szczerbo-Trojanowska; Department of Interventional Radiology, University School of Medicine, Lublin, Poland.

Purpose: To evaluate the safety and efficacy of carotid artery stenting (CAS) performed with cerebral protection devices. Materials & Methods: Since February 2000, CAS with cerebral protection was performed in 73 patients with extracranial carotid artery stenoses (>70 %). The plaque was ulcerated in 13 cases, soft in 28 cases, and calcified in 32 cases. Forty-nine stenoses were primary, while 24 were secondary to TEA. Seventy-nine stents (69 Carotid Wallstent and 10 Acculynk) were implanted in 76 carotid arteries. All procedures were performed using a cerebral protection device: Parodi system (8), EPI filter wire (51), Angioguard (5), NeuroShield (9), or Spider embolic protection (3). All patients underwent independent neurologic examination before and after the procedure. Results: Technical success (residual stenosis <30%) was achieved in all cases. No death or restenoses occurred throughtout follow-up, which ranged from 1 to 36 months. Six complications (8.2 %) were observed: 1 major stroke (1.3 %) and 5 minor strokes (6.8 %). In 7 cases (9.5 %) embolic material was found inside the cerebral protection. Conclusion: CAS represent a feasible procedure that can be performed in high-risk patients and is associated with a low restenosis rate.

Purpose: To evaluate procedural complications using filter neuroprotection for carotid stenting. Materials & Methods: Carotid stenting was performed in 75 patients with ICA stenosis >70% (primary–62, restenosis–13). All patients were symptomatic, studied by duplex Doppler, TCD and neurological examinations. For neuroprotection we used Angioguard™ –59 and FilterWire™ –16. Results: In 67 patients the filter device passed the stenosis easily; in eight patients (10.6%) with severe stenoses it could not be advanced through the stenosis. In these cases predilatation was performed before placing the filter device over the stenosis. In 22 patients (29.3%) there was macroscopic debris in the filter when it was removed after stenting; in 12 of them angiography after stent postdilatation showed ICA occlusion due to debris in the filter. Occlusion was always temporary, resolving when the filter device was removed. In 15 patients arterial spasm occurred during control angiography, resolving spontaneously after the filter device was removed. We encountered no complications while removing filter devices after the completed procedure. There were no major neurological complications. Five patients (6.6%) had a TIA, probably because of temporary filter occlusion. Conclusion: Filter neuroprotection makes carotid stenting safe from distal embolization. Minor complications are occasional and temporary, resolving without additional treatment.

6.1.2

6.1.4

Endovascular treatment of carotid atherosclerotic disease: the use of self-expandable carotid Wallstents. Experience in a single center R. Manetti, F. Castriota, E. Ricci, A. Liso, K. Oshoala, A. Cremonesi; Unità Operativa di Cardio-Angiologia Diagnostica e Interventistica, Villa Maria Cecilia Hospital, Cotignola (RA), Italy.

A comparison of the high-intensity signal rate using transcranial Doppler during unprotected and protected carotid stenting (NeuroShield™ filter) within a randomised controlled trial S. Macdonald1, T. J. Cleveland1, D. Evans2, P. D. Griffiths3, P. A. Gaines1; 1Sheffield Vascular Institute, Sheffield, United Kingdom, 2Department of Medical Physics, Leicester Royal Infirmary, Leicester, United Kingdom, 3Academic Department of Radiology, Royal Hallamshire Hospital, Sheffield, United Kingdom.

Purpose: Primary endpoints of this study were to evaluate procedural success rate and neurological complications in 373 patients treated from October 1999 to January 2003 by endovascular therapy of carotid critical stenosis by using self-expandable low-profile carotid Wallstents. Materials & Methods: A total of 373 patients underwent percutaneous angioplasty and stenting of the extracranial carotid artery by using a carotid Wallstent. In 42 (11.26%) patients there was a history of major or minor stroke; 87 (23.32%) patients presented a previous TIA. In 28 (7.51%) cases there was a restenosis after carotid endarterectomy (CEA). In 202 (54.16%) patients a direct stenting and in 171 (45.84%) patients a pre-dilatation were performed. Results: The percutaneous procedure was effective in 373/373 patients (100%). No deaths occurred. The overall symptomatic complications rate was 3.75%: one major stroke (0.27%); seven minor strokes (1.88%); five intracranial hemorrhages (1.34%); one TIA (0.27%). Embolic complications were nine (2.41%): one (0.27%) major stroke and four (1.07%) minor strokes were late events, while three (0.80%) minor strokes and one (0.27%) TIA were early events. Conclusion: Our data suggest that percutaneous dilatation and stenting of the carotid artery by using the carotid Wallstent is feasible and effective.

Purpose: To compare the total number of recorded events, particulate and gaseous emboli using the sample volume length parameter (SVL). Materials & Methods: A subset of nine consecutive patients (five unprotected and four protected) recruited into this randomised trial were evaluated. All had symptomatic carotid artery stenosis =70% (NASCET) and underwent procedural insonation of the middle cerebral artery upstream of the target lesion. The digital audiotape recording was analysed off-line using a high-temporal-resolution spectral analyser to measure embolic duration and velocity and provide the SVL parameter. An SVL =1.28cm was considered to represent a particulate embolus. Results: Some results are outstanding but will be presented. The mean number of total events (SVL parameter) was 466 (SD 160.8) in the protected group and 165.2 (SD 100.1) in the unprotected group and numbers of particulate emboli were 299.75 (SD 104.9) protected and 92 (SD 40.497) unprotected. There was a smaller difference for the number of discrete gaseous emboli. Statistical significance will be presented. There were no adverse clinical sequelae. Conclusion: Cerebral protection using this filter-type device (NeuroShield™) may be associated with a greater procedural high-intensity signal rate on transcranial Doppler compared with unprotected carotid stenting. The clinical relevance of this is unclear.

6.1.5

6.1.7

Jugular venous neurone specific enolase and S100B protein following unprotected and protected carotid stenting (NeuroShield™ filter) within a randomised controlled trial S. Macdonald1, T. J. Cleveland1, D. Thompson2, P. D. Griffiths3, P. A. Gaines1; 1 Sheffield Vascular Institute, Sheffield, United Kingdom, 2Department of Biochemistry, Leeds General Infirmary, Leeds, United Kingdom, 3 Academic Department of Radiology, Royal Hallamshire Hospital, Sheffield, United Kingdom.

Percutaneous transluminal angioplasty and stenting for symptomatic intracranial stenoses H. G. Bertoni1, R. D. Garcia Monaco1, D. Cassina1, M. Fava1, C. G. Videla2, S. M. Valiensi2, P. L. Bauso Toselli2; 1 Interventional Radiology, Hospital Italiano, Buenos Aires, Argentina, 2 Neurology, Hospital Italiano, Buenos Aires, Argentina.

Purpose: To utilise biochemical markers of neurone specific enolase (NSE) and glial (S100B) injury as surrogates of adverse neurological events to compare unprotected and protected carotid stenting. Materials & Methods: Thirty patients with symptomatic carotid artery stenosis (=70%, NASCET) were randomised between protected and unprotected treatment limbs. Cystatin C, a sensitive marker of glomerular function, was assayed as measurement of S100B as sensitive to renal impairment. Jugular venous samples were taken for NSE, S100B and cystatin C at baseline, two, six and 24 hours post-stenting. Groups were compared using a summary statistic for serial measurements (area under the curve). Data were rendered normal (logarithmic and squareroot transformations) and rise-to-peak values compared using independent-samples t-tests. The influence of cystatin C on the rise in S100B was addressed using ANCOVA. Results: There were no significant differences between groups. There was a significant rise in S100B and cystatin C (reflecting subclinical contrast nephrotoxicity) in both groups. There was no significant rise in NSE. The rise in S100B cannot be explained by the rise in cystatin C alone (p=0.067, eta squared=0.18, correlation= -0.42). Conclusion: Carotid stenting causes a release of S100B implying transient glial injury regardless of whether cerebral protection (NeuroShield™) is utilised.

Purpose: Although percutaneous transluminal angioplasty (PTA) is an effective treatment modality in coronary and peripheral artery diseases, its efficacy for intracranial atherosclerotic stenoses has not been verified. We report our experience in the treatment of symptomatic intracranial stenoses with endovascular techniques. Materials & Methods: PTA and stenting was performed in six patients with symptomatic high-grade stenoses (>70%) who had either recurrent transient ischemic attacks (TIAs) refractory to the medical therapy or perfusion problems. Short-term clinical and angiographic data were obtained. The lesions involved the middle cerebral artery in one case, the distal end of the internal carotid artery in one case, the basilar artery in two cases, and the intracranial section of the vertebral artery in two cases. Results: All the procedures were technically successful (100%). There was one immediate complication. All the patients underwent continuous neurological follow-ups with a mean of 8.2 months (range: 6-14) and, presently, they are asymptomatic. Procedure-related mortalities were not recorded in this series. Conclusion: Although the treatment of intracranial artery stenoses with angioplasty and stenting is promising, long-term angiographic and clinical follow-up monitoring of a larger patient population is needed.

6.1.6

6.1.8

Why implement a carotid angioplasty program in a center with experience in carotid endarterectomy? M. Bosiers1, P. Peeters2, K. Deloose1, J. Verbist2; 1Vascular Surgery, AZ St-Blasius, Dendermonde, Belgium, 2Cardiovascular and Thoracic Surgery, Imelda Hospital, Bonheiden, Belgium.

PTA/stenting of the intrathoracic portion of the common carotid artery in 68 patients V. Bérczi, A. Á. Molnár, B. Nemes, Á. Simonffy, K. Hüttl; Radiology, Clinics of Cardiovascular Surgery, Semmelweis University Medical School, Budapest, Hungary.

Purpose: To argue the complementarity of carotid endarterectomy (CEA) and carotid artery stenting (CAS), based on peri- and post-procedural investigational results, patients' preferences, and financial arguments. Materials & Methods: Retrospective and prospective analysis of the major adverse neurological event (MANE) rate and the CEA/CAS ratio of all carotid procedures performed in our vascular centers. Results: Our vascular centers perform about 250 carotid procedures yearly, out of which about 70% consist in endovascular repair. This high percentage is directly related to an equal MANE rate for both treatment modalities (CEA 2.3% and CAS 1.5%). Apart from the 20% with contraindications for either CEA or CAS, all patients are offered a choice between classical and endovascular treatment. Despite the higher cost of the endovascular procedure, body integrity and local anesthesia are stated as the main arguments in its favor. Of patients with the choice, 75% opt for CAS. Conclusion: Experience with both CEA and CAS is mandatory in one center, as contra-indications exist for both treatment modalities. Offering patients the freedom of choice between these modalities is only ethical when the MANE rates are equal and acceptable for both.

Purpose: To assess primary success, efficacy, and safety of PTA/stenting of the origin (intrathoracic portion) of the common carotid artery (CCA). Materials & Methods: Between 1995 and 2002, 68 patients (70 stenoses) (38 men; 56 left, 14 right) with >80% stenoses of the CCA were treated. Co-morbidity factors were: hypertension (36), high serum lipid levels (27), and diabetes mellitus (7). All the procedures were done through a transfemoral approach. A stent was implanted in 21 cases. Embolic protection devices were not used. Follow-ups included neurological examinations and carotid Duplex scans. Results: Primary technical success was 98.6%. In one case, the advancement of the guidewire through the stenotic right CCA was not possible. Periprocedural complications included two strokes, six TIAs, one cardiac asthma. No periprocedural deaths occurred (periprocedural death/stroke rate 2.9%). Follow-ups (33 patients) showed 16 neurological improvements, three unchanged, and one impairment; the remaining 13 asymptomatic patients showed an unchanged neurological condition. All arteries (33) remained patent at a mean follow-up period of 33.1 months, two rePTAs (after 28 and 31 months) were however done due to significant restenosis. Conclusion: PTA/stenting of the intrathoracic portion of the CCA is a successful and effective treatment with an acceptable safety level.

6.1.9

6.2.2

Brachiocephalic arterial occlusive disease: one center's experience of endovascular treatment with stenting E. N. Brountzos1, B. T. Petersen2, C. A. Binkert2, I. E. Panagiotou1, J. A. Kaufman2; 1 2nd Radiology, Athens University, Athens, Greece, 2Dotter Interventional Institute, Oregon Health and Sciences University, Portland, United States.

Vibrational angioplasty in the treatment of chronic infrainguinal arterial occlusions: mid-term results D. Tsetis1, L. Michalis2, M. Rees3, A. Katsamouris4, M. Matsagas5, Z. Androulakakis6, A. Hatzidakis1, D. Sideris2, N. Gourtsoyiannis1; 1 Radiology, University Hospital Heraklion, Heraklion, Greece, 2 Cardiology, University Hospital Ioannina, Ioannina, Greece, 3Radiology, Bristol Royal Infirmary, Bristol, United Kingdom, 4Division of Vascular Surgery, University Hospital Heraklion, Heraklion, Greece, 5Surgery, University Hospital Ioannina, Ioannina, Greece, 6Surgery, Venizelion Hospital Heraklion, Heraklion, Greece.

Purpose: Outcome evaluation of stent-treatment of innominate and subclavian artery occlusive lesions. Materials & Methods: Retrospective data collection of 48 consecutive symptomatic patients (27 men, median age: 64) with 49 brachiocephalic artery lesions treated with stenting. Eighty-two percent of patients had cardiovascular comorbidity. Indication for treatment was vertebrobasilar insufficiency (VBI) in 16.6%; upper limb ischemia (ULI) in 31.3%; VBI and ULI in 12.5%; failure of other vascular conduits in 22.9%. Balloon expandable stents were used in 44 lesions, self-expandable stents in 5 lesions; in total 53 stents were placed in the 48 patients. Results: Technical success was 96%, clinical success 94%. We encountered four complications (two puncture site hematomas, one distal – hand-embolization, one transient cerebral ischemia). Mean follow-up time was 16.7 months (range: 0-68). Five patients had recurrent lesions, treated by surgical (n=2) or endovascular (n=3) means. Cumulative primary and secondary patency rates at 12 months were 91.7% and 96.5% respectively. Conclusion: Stenting of brachiocephalic artery lesions resulted in immediate resolution of our patients' symptoms with durable effect and few complications in this large patient group with serious comorbid conditions.

Purpose: To present our mid-term results of vibrational angioplasty in chronic infrainguinal arterial occlusions. Materials & Methods: Thirty-seven patients (39 legs) were treated for 18 chronic femoropopliteal and 26 tibioperoneal occlusions not amenable to surgical revascularization. The occlusion length ranged from 2 to 15 cm. Pre-procedural presentations (SVS/ISCVS criteria) were: category 3 (4 patients), category 4 (6 patients) and category 5 (29 legs). Vibrational angioplasty using coronary equipment was applied to assist the passage of a coronary guidewire, followed by a conventional angioplasty. Clinical follow-up ranged from 1 to 24 months (mean, 11 months). Results: Recanalization was successful in 40 of the 44 occlusions (91%). One calf hematoma was treated conservatively. The time for crossing the occlusions with the wire ranged from 3 to 30 minutes. During follow-up, two patients died, ten underwent below-the-knee amputation, four showed no improvement, seven remained free from claudication or rest pain and in 16 legs with ulceration or gangrene a good healing progress was observed. The amputation-free survival rate is 94% at one, 84% at three, 81% at six, and 73% at 12 months. Conclusion: Vibrational angioplasty is a feasible way for safely recanalizing chronic infrainguinal arterial occlusions with satisfactory midterm results.

6.2.1

6.2.3

Auth-Rotablator rotational aterectomy in infrapopliteal diabetic critical limb ischemia. Long-term results A. R. Cotroneo1, G. Marano1, A. Cina2, C. Di Stasi2; 1Istituto di Radiologia, A.O. "Maggiore della Carità", Novara, Italy, 2Istituto di Radiologia, Università Cattolica del S. Cuore, Roma, Italy.

A new atherectomy catheter for the treatment of short occlusions of the superficial femoral artery: first clinical experience H. Krankenberg, T. Tübler; Angiology, Center for Cardiology and Vascular Intervention, Hamburg, Germany.

Purpose: To evaluate indications and long-term results of Auth-Rotablator aterectomy of tibial arteries in diabetic critical limb ischemia (CLI). Materials & Methods: From January 1992 to December 2002, 227 diabetic patients (275 limbs) with CLI (LF III-IV) underwent percutaneous recanalization of infrapopliteal stenotic/occlusive lesions with AuthRotablator alone or associated with PTA/stenting. Patients' selection included clinical evaluations and US-Doppler examinations, while angiography was both diagnostic and therapeutic. Ostial, multiple, or plurivascular lesions and segmentary occlusions were treated. Followup was based on clinical and sonographic surveillance. We report here our long-term results in145 patients and 175 limbs, with a follow-up >36 months. Results: Fifty-eight FIII and 87 FIV patients were treated with aterectomy. In 43 of the 175 limbs PTA/stenting were required. We recorded two inguinal hematomas, one distal embolization, one ATA perforation, and one A-V fistula. During follow-up, 58 FIII patients were classified as F0 (49) or F2a (9), while 82/87 FIV patients passed to the F2b stage. Five major and nine minor amputations became necessary. Seventy-three complete remissions were observed. Conclusion: Our long-term results, low complication rate, and feasibility in selected cases of diabetic CLI proved the effectiveness and safety of rotational aterectomy.

Purpose: PTA is associated, especially in the superficial femoral artery (SFA) region, with high restenosis/reintervention rates. We present a new endovascular technique for SFA lesions <5 cm. Materials & Methods: The Fox Hollow Technologies (FHT) debulking catheter is an atherectomy system with a spinning steel blade allowing debulking of atherosclerotic plaques. The cut material is captured within a storage chamber at the catheter tip. In 20 patients (16 men, age 67±7y) with intermittent claudication, 20 single, distal SFA/ popliteal artery lesions were treated. At baseline, relative/absolute claudication distances (RCD/ACD) were 156±55/224±44 m, the ankle-brachial index (ABI) was 0.61±0.3 before and 0.53±0.2 after exercise. Results: A median of six cuts per lesion were carried out. A primary angiographic success rate was achieved in all patients. Except for two patients in whom additional PTA was required, no further adjunctive therapies were needed. ACD improved to 281±19 m and the ABI at rest and after exercise were 0.89±0.09 and 0.81±0.11, respectively. No complications were observed during or after intervention. Thirty-day followup revealed a clinically significant improvement in claudication distance and ABI. Conclusion: FHT debulking catheter allows a safe and effective treatment of SFA lesions. Long-term evaluation must prove the value of this new technology.

6.2.4

6.2.6

Distal flow protection devices during percutaneous treatment of acute and subacute lesions of lower extremity arteries D. Siablis1, D. Karnabatidis1, K. Katsanos1, P. Zabakis1, T. Kordonouri1, I. Tsota1, C. Kotsarini1, I. Tsolakis2; 1 Department of Radiology, University of Patras, Patras, Greece, 2 Department of Vascular Surgery, University of Patras, Patras, Greece.

Subintimal angioplasty in critical ischemia of lower limbs C. Setacci, I. Baldi, G. de Donato, A. Cappelli, M. Pieraccini; Vascular surgery, University of Siena, Siena, Italy.

Purpose: Outflow protection devices are well-established in the PTA of carotid lesions. Yet there is no sufficient experience regarding their role in the performance of other peripheral interventions. We present our experience using these devices during percutaneous interventions of acute and subacute occlusions of the lower extremity arteries. Materials & Methods: In 21 consecutive patients with acute and subacute arterial occlusion (estimated age of thrombuses: one day-three months) distal protection devices were applied during percutaneous interventional treatment. The treated lesions were located in: external iliac (1), femoral (6), popliteal (9), infrapopliteal arteries (5). A catheter occlusion device was used in nine cases and a guidewire filter in 12. Results: All devices were successfully applied. In 14 cases >1-mm particles were extracted. In two cases both device dislocation and vasospasm occurred in both types of devices. In both cases, vasospasm was treated with local spasmolytic therapy. In all cases the interventional procedure was successful without angiographic and clinical signs of distal embolization. All patients showed clinical improvements at 24hour and 30-day follow-ups. Conclusion: Distal flow protection devices seem to be feasible and safe when applied during percutaneous interventions of lower extremity arteries to prevent distal embolization.

Purpose: Though advances in endovascular technology hold promise for less invasive approaches to arterial diseases, the mainstay of therapy for patients with peripheral occlusions remains surgical bypass grafting. Materials & Methods: We reviewed our experience with 33 patients with attempted subintimal angioplasty, 24 men and nine women, 56-83 years in age. Eighteen patients had gangrene, 13 rest pain and two disabling claudication. There were 15 superficial femoral and popliteal lesions, six popliteal, three popliteal-tibial, and nine tibial lesions. Results: Subintimal angioplasty was technically successful in 27/33 patients. The patency rates at 12 and 24 months showed complete resolution in 15/15 patients and healing of areas of gangrene in 12/18. In the 27 technically successful subintimal angioplasties rest pain and disabling claudication resolved in 15/15 patients and areas of gangrene healed in 12/18. Among the 27 technically successful subintimal angioplasties patency rates at 12 and 24 months were 71% and 59% respectively. Conclusion: We believe subintimal angioplasty will continue to have a role in treating disabling claudication and critical limb ischemia. It offers many advantages including reduced anaesthesia requirements, a minimally invasive approach, and potential reduction of length of stay and cost. When applied judiciously, bypass options can be preserved.

6.2.5

6.2.7

Subintimal angioplasty: technical success and long-term patency J. A. Brookes1, M. J. Guiney1, S. Aly2, M. Adiseshiah2; 1Radiology, The Middlesex Hospital, UCLH, London, United Kingdom, 2Vascular surgery, The Middlesex Hospital, UCLH, London, United Kingdom.

Use of transcutaneous oxygen tension for the assessment of successful percutaneous transluminal recanalization in diabetic patients with ischemic foot ulcers S. Fabiano1, R. Gandini1, M. Stefanini1, V. Latini2, A. Caselli2, L. Uccioli2, G. Menzinger2, G. Simonetti1; 1Diagnostic Imaging and Interventional Radiology, University of "Tor Vergata", Roma, Italy, 2Endocrinology, University of "Tor Vergata", Roma, Italy.

Purpose: To assess the technical feasibility, clinical outcome and patency rate of femoro-popliteal segment subintimal angioplasty in patients with disabling claudication and critical limb ischaemia. Materials & Methods: Patients with long femoro-popliteal segment occlusion (>15cm) including failed conventional angioplasty and failed femoro-popliteal/distal bypass were entered into the study. All of them were symptomatic. Some of them were not suitable candidates for bypass surgery. Subintimal angioplasty recanalisation from both antegrade and/or retrograde approach was carried out. Patients had Duplex ultrasound scans immediately before and after the procedure to assess patency. Long-term patency was assessed by Duplex at clinic visits. Results: Thirty-three patients were recruited (19 men, age range 43-89 years [mean 70 years]. We attained primary technical success in 27/33 (83%) cases. There were 3/33 early failures (<48hrs) and a further 3/33 late failures (48hrs-6 months). The “intention-to-treat” patency rate at eight months was 21/33 (65%) or 21/24 (88%) of those patent at 48 hrs. Conclusion: Subintimal angioplasty is a feasible management option for femoro-popliteal occlusive disease. It should be considered as an alternative line of management for those patients not suitable for bypass surgery.

Purpose: To assess the efficacy of transcutaneous oxygen tension (TcPO2) for the assessment of successful percutaneous transluminal recanalization in diabetic patients with ischemic foot ulcers. Materials & Methods: Twelve diabetic patients with ischemic foot ulcers (Fontaine stage IV) underwent peripheral revascularization in our unit. In all cases, angiography was performed prior to revascularization. TcPO2 was recorded before (T0) and 1, 2, 3 and 4 weeks after revascularization. Results: As shown in the Table progressive improvement of cutaneous oxygen tension (pO2) was observed after revascularization, peaking at 3 weeks after surgery. At the same time, a decrease of cutaneous carbon dioxide tension (pCO2) was observed 1 week after revascularization and reached a plateau during the following weeks. In agreement with TcPO2 findings, none of our patients had complications after revascularization (100% limb salvage rate) or needed to undergo a second revascularization procedure. Table: T0 1week 2week 3week 4week pO2 (mmHg) (8.97+/-9) (26.83+/-17) (34.12+/-22) (41.6+/-22) (40.93+/-27) pCO2 (mmHg) (71.2+/-27) (43.75+/-10) (39.7+/-14) (34.18+/-8) (33.28+/-7) Conclusion: TcPO2 measurement is a valid tool for assessing successful revascularization. It might also be used to establish the optimal time to perform more aggressive surgical debridement in diabetic patients with ischemic foot ulcers.

6.2.8

6.3.1

Endoluminal femoro-popliteal bypass: our experience D. Laganà1, R. Caronno2, G. Piffaretti2, M. Carnini2, M. Tozzi2, M. Canziani2, P. Castelli2; 1 Institute of Radiology, University of Insubria, Varese, Italy, 2Vascular Surgery, University of Insubria, Varese, Italy.

Percutaneous radiofrequency ablation of malignant hepatic tumors S. Terraz1, P. Majno2, C. Becker1; 1Diagnostic Imaging and Interventional Radiology, CHUGE, Geneva, Switzerland, 2Abdominal surgery, CHUGE, Geneva, Switzerland.

Purpose: To determine the feasibility and effectivness of endoluminal bypass. Materials & Methods: In the last 12 months we treated 17 patients, mean age 62 years, with critical limb ischemia. Preoperative angiography showed complete occlusion of the superficial femoral artery (SFA) in all patients, with patency of the popliteal artery and at least one of the leg arteries. Under spinal anesthesia we implanted 17 stents (Wallgraft 13, Hemobahn 4) through a femoral surgical approach; we also combined endarterectomy at the origin of the SFA (in seven cases) and PTA of the popliteal artery (in nine cases). Results: The immediate success rate was 88% and the limb salvage rate was 100%. In two cases we were forced to perform femoro-popliteal bypass. We observed one early occlusion successfully treated by thrombolysis. We performed clinical and echocolordoppler follow-up at 1, 3, 6, and 12 months after the procedure; mean follow-up was 12 months (range 3-24), and in all patients we documented the patency of the endoluminal bypass. Conclusion: In our preliminary experience endoluminal bypass has been demonstrated to be a feasible technique; it showed early and mediumterm results as good as those of conventional surgery.

Purpose: To assess the efficacy and safety of radiofrequency (RF) ablation of nonresectable primary and secondary liver tumors. Materials & Methods: 69 tumor nodules (38 hepatocellular carcinomas [HCC], 30 metastases, and 1 cholangiocarcinoma) were treated in 38 patients. Under US, CT, or MRI guidance, RF ablation was accomplished within 15 min for each lesion. Contrast-enhanced CT and/or MRI were systematically performed immediately, after one month and then every 3 months for a median follow-up of 17 months. Results: Complete necrosis was achieved initially in 65 lesions (94%) by imaging criteria. No local recurrence was observed in 32 HCC (84%), 19 metastases (63%) and the cholangiocarcinoma. There was a statistically significant difference in the rate of complete ablation between patients with HCC and those with metastases (p = 0.019). We observed two complications: a biliary fistula requiring percutaneous drainage and an arterio-portal fistula treated by embolization. Conclusion: Our preliminary results suggest that RF ablation is a safe, effective treatment for local control of HCC. The approach in patients with metastases still needs to be improved.

6.2.9

6.3.2

Prospective randomized trial evaluating the efficacy and safety of a newly developed staple-based vascular closure device, the HemoStape device D. M. Duke1, C. Martin2, J. Coleman2, P. Brennan1, J. Thornton1; 1Dept. of Neuroradiology, Beaumont hospital, Dublin, Ireland, 2Dept. of Biomedical Research, Nova Science Ltd., Dublin, Ireland.

Percutaneous coil placement prior to liver radiofrequency ablation under fluoroscopic guidance A. A. Hatzidakis1, T. Sabharwal2, M. Hamady2, A. Gangi3, D. Tsetis1, A. Adam2; 1Department of Radiology, University of Crete, Heraklion Crete, Greece, 2Department of Interventional Radiology, Guy's and St. Thomas' Hospital, London, United Kingdom, 3Department of Radiology B, University of Strasbourg, Strasbourg, France.

Purpose: To establish the efficacy and safety of a newly designed vascular closure device, the HemoStape, in a randomized controlled trial. Materials & Methods: Forty-one patients (18 male) aged 25-72 years (mean 48.6) undergoing diagnostic cerebral angiography were randomized to receive either manual pressure (MP) or the HemoStape device(HD) for vascular hemostasis. All patients were monitored postprocedure and at 4 weeks, with ultrasound scans performed on those receiving HD. Results: In 5 patients the device was only positioned within the vessel, with no staple fired. Eighteen patients underwent MP and 18 received HD. Mean time to hemostasis was HD: MP=156:614 sec; mean time to ambulation was HD:MP=161:465mins. Post-procedure complications included bruising in 3 (HD:MP=1:2), hematoma in 1 (HD), oozing in 3 (HD); none were clinically significant. Three MP patients had significant complications (2 cerebrovascular accidents and 1 deep venous thrombosis), all unrelated to the procedure. Five patients in each group underwent therapeutic angiography (1-64 days later, average HD: MP=5.6:14.4 days), when a suture-based closure device was used. Two of these, both in the HD group, developed pseudoaneurysms. Conclusion: These early clinical trial results indicate that the HemoStape device is an efficient and safe closure device. Further evaluation is ongoing.

Purpose: Some hepatic neoplasms are invisible on ultrasound, appearing only transiently on computed tomography following enhancement with intravenous contrast. The aim of this work was to evaluate a new technique for localization of such tumors before radiofrequency ablation (RFA) treatment. Materials & Methods: The technique was used in four cases. A 21G needle was inserted before IV contrast injection in the approximate position of the lesions using anatomical landmarks demonstrated near the mass on previous contrast-enhanced CT images. After contrast injection, the needle location was placed more precisely with its tip in the center of the mass. A metallic coil (40 mm x 3 mm) was inserted via the needle and its position confirmed with CT imaging. Then fluoroscopy was used to guide the RF electrode into the tumor using the coil as target. Results: Placement of the coil was satisfactory in all cases. RFA was performed successfully and adequate coagulation of tumors was seen on follow-up CT. No complications were observed. The coil did not interfere with delivery of RF energy. Conclusion: The technique described, safe and effective, is a very good method of guiding RF electrode placement in patients with tumors poorly visualized on US and unenhanced CT.

6.3.3

6.3.5

Adverse events recorded during the radiofrequency treatment of 582 liver tumors T. de Baere1, V. Kuoch1, C. Dromain1, C. Sengel2, C. Le Hong1, O. Risse2, C. Letoublon2, D. Elias1; 1 Institut Gustave Roussy, Villejuif, France, 2CHU de Grenoble, Grenoble, France.

Characterization of radiofrequency-treated areas with MRI: impact of native T2-weighted imaging B. B. Frericks1, T. D. Kirchhoff1, S. Merkesdal2, H. Rosenthal1, T. Becker3, M. Galanski1; 1 Diagnostische Radiologie, Medizinische Hochschule Hannover, Hannover, Germany, 2Medizinische Hochschule Hannover, Hannover, Germany, 3Viszeral- und Transplantationschirurgie, Medizinische Hochschule Hannover, Hannover, Germany.

Purpose: To describe rates and potential complications risk factors of liver tumors radiofrequency (RF) ablation. Materials & Methods: Over a period of five years, 312 patients underwent 350 (124 pre-operative and 226 percutaneous) sessions of RF for the treatment of 582 liver tumors (115 HCC, 467 metastases). Results: Thirty-seven (10.6%) adverse events and five (1.4%) deaths occurred. Deceases were due to liver insufficiency (1), colon perforation (1), and portal vein thrombosis (3). Portal vein thrombosis was significantly more frequent after preoperative RF performed during a Pringle maneuver in cirrhotic livers (2/5) that in non-cirrhotic livers (0/54). Liver abscess (7) was the most common complication with bilio-enteric anastomosis as a major risk factor. We recorded five pleural effusions, five skin burns, four hypoxemias, three pneumothoraces, two small subcapsular hematomas, one regressive acute renal insufficiency, one hemoperitoneal effusion, and one needle tract seeding. Complications which did not require specific treatments or prolonged hospital stays were 6.3%, while 3.7% of complications required less than five days of hospitalization for treatment or surveillance, and 2% needed a treatment requiring more than five days of hospitalization. Conclusion: RF is well-tolerated. Patients with bilio-enteric anastomosis, and RF during vascular occlusion in cirrhotic livers are at risk.

Purpose: The treatment results after radiofrequency ablation cannot be assessed by lesion size alone. Precise characterization of the treated areas is mandatory. Therefore, signal intensities of treated areas were evaluated and the impact of the different sequences for evaluation of treatment efficacy was assessed. Materials & Methods: 30 patients with 82 malignant liver tumors treated with RF-ablation were examined using MRI at day one, day 30 and every third month after RF-ablation. Signal intensities of the ablated areas in pre- and post-contrast sequences were assessed especially to evaluate the efficacy of treatment. Results: After a mean follow-up of 14 (6-27) months six patients developed local recurrence of nine hepatic tumours. Nine patients developed new metastases. The ablated areas showed characteristic time-dependent appearances in the different sequences. Due to the high contrast between low signal intensity of completely ablated tumours and intermediate signal intensities in viable tumor areas, native T2-weighted sequences were best suited for early detection of local recurrence. Conclusion: Characterization of RF-treated areas is possible in MRI, but knowledge of the changes in signal intensities over time is essential. Native T2-weighted images are of major importance for the early detection of viable tumor.

6.3.4

6.3.6

Computed tomography-guided percutaneous radiofrequency ablation of liver tumors using a LeVeen 4-cm array probe and a 200watt generator M. B. Pitton1, S. Herber1, J. Schneider1, G. Otto2, M. Thelen1; 1Radiology, University Hospital Mainz, Mainz, Germany, 2Transplantation Surgery, University Hospital Mainz, Mainz, Germany.

PET scanning in patient selection and follow-up for radiofrequency thermal ablation of liver metastases: work in progress L. Monfardini1, F. Orsi1, G. Bonomo1, P. Della Vigna1, S. Penco2, M. Bellomi2; 1Radiology, European Institute of Oncology, Milan, Italy, 2Radiology, University of Milan, Milan, Italy.

Purpose: To evaluate safety and efficacy of a radiofrequency (RF) ablation system. Materials & Methods: Liver tumors were treated using a LeVeen 4-cm array probe and a RF3000-generator (Radiotherapeutics). Primary tumors were colorectal (n=22), mammary (n=4), pancreatic (n=2), hepatic (n=3) gastric (1), duodenal (1), gastrointestinal stroma tumor (1), and lung tumor (1). Follow-up included multislice-CT at four weeks, and every three months. Results: In 35 patients (age 63.9±12.6y) 65 tumors were percutaneously treated. Lesion size was 2.3±1.2cm, corresponding necrosis was 4.6±1.4cm. Primary success was obtained in 60/65 lesions, four cases required second treatment. One procedure was aborted (close relation to colon). Sixty-three tumors were treated in local anesthesia, two required general anesthesia. Morbidity was 9.2%; bleedings (n=3) were treated with coil-embolization. One subcapsular hematoma required blood transfusion. One hematothorax was treated by drainage. One cholecystitis required cholecystectomy. Mortality was 0, follow-up 5.6±3.3 months (range, 0-13). Tumor recurrence occurred in 13 patients (two local recurrences, five local+new recurrences, six with new metastases). Three recurrences were treated by RF-ablation, ten patients received chemotherapy. Conclusion: RF-ablation can be performed in analgo-sedation and local anesthesia with low peri-interventional morbidity and mortality. Using LeVeen probes and 200-watt generators, appropriate necrosis sizes can be achieved.

Purpose: PET and CT scan were performed before and after liver radiofrequency (RF) ablation in patients with hepatic metastases to select patients candidates for local treatment and to follow them up. Materials & Methods: Eleven liver metastatic patients (6 F; age 49-77, average 61) were selected for liver RF as a part of a multimodal therapeutic strategy. They were followed up with CT after 1 and 3 months and with PET at least 2 months after liver ablation. Results: PET and CT scans showed similar results in 9/11 patients: in 5 both PET and CT scans were negative and in 4 both were positive for local recurrence; in two patients PET was also positive for systemic disease (lung and mediastinum). In 2 pts PET and CT disagreed because of PET emission at RF-treated sites, whereas CT was negative in one patient and not clearly positive in the other. Conclusion: In our preliminary experience PET seems to be a very sensitive instrumental examination for local and systemic follow-up in patients subjected to liver RF. PET could also be useful in patient selection, in order to avoid treatment in case of extra-hepatic disease.

6.3.7

6.4.1

How many colorectal liver metastases can we treat? Analysis of the 6-10 group A. R. Gillams, W. R. Lees; Imaging Dept, The Middlesex Hospital, London, United Kingdom.

Radiologically-controlled balloon dilatation of colorectal strictures T. Betschart1, S. Dincler2, W. Schöpke1, T. Jung1, R. Naef2, P. Buchmann2, B. Czermak1; 1 Department of Radiology, Waid Hospital, Zuerich, Switzerland, 2 Department of Visceral Surgery, Waid Hospital, Zuerich, Switzerland.

Purpose: Most centres offer ablation to patients with five or fewer metastases. Reported median survival from the diagnosis of liver metastases is 32-34 months. We have performed radiofrequency (RF) ablation in a cohort of patients with 6-9 colorectal metastases. Materials & Methods: Twenty-six patients (14 men) were treated. The maximum tumour diameter was 4.5 cm. Six patients had six metastases, ten had seven, eight had eight and two had nine. RF was performed under general anaesthesia using US and CT guidance. Single or triple water-cooled cluster electrodes were used depending on tumour size. Maximum power (<200 W) was applied for 10-15 minutes. If residual tumour persisted after the first ablation, then a second ablation was performed four weeks later. Results: The median number of treatments was two (range, 1-8). Median survival from the diagnosis of liver metastases was 25 months (range 12-32). Median survival from the first thermal ablation was 14 months (range 1-29). Conclusion: Although this cohort fares less well than patients with more limited disease, RF may still be beneficial. These results are of particular importance in light of the on-going European trial that has acceptance criteria of <10 metastases.

Purpose: To evaluate feasibility and safety of fluoroscopically guided balloon dilatation of postoperative anastomotic colorectal strictures. Materials & Methods: From 1999 to 2002 a total of 49 dilatations of postoperative rectal or colonic strictures were performed in our department in 14 patients using wire-guided Gruentzig-type balloon catheters under fluoroscopic control. Before dilatation the strictures were documented by partially distending the balloons with contrast solution. Degree of hourglass effect (indentation at the point of maximum stricture) was evaluated and the diameter of the stricture was measured in mm. For dilatation, balloons with diameters from 5 to 25 mm were used. Results: Before intervention stricture diameter measured 4-12 mm (mean 8.4 mm). Diameters measured at the final session were 14-25 mm (mean 20.4 mm). The postoperative diameter was wider by a factor 1.75-5 (mean 2.66) than the preoperative diameter. Complications encountered were a circumscribed intramural hematoma (n=1) and a small mucosal injury without perforation (n=1). Conclusion: In our experience radiologically controlled wire-guided balloon dilatation of colorectal strictures appears to be a safe and effective method and is well tolerated by patients. Wider use can therefore be recommended.

6.3.8

6.4.2

Safety of percutaneous portal vein embolization using polyvinyl alcohol particles R. T. Gandhi, K. T. Brown, A. M. Covey, L. Brody, C. T. Sofocleous, G. I. Getrajdman, E. J. Ring; Department of Radiology, Memorial Sloan-Kettering Cancer Center, New York, United States.

Percutaneous gastrostomy: ten years' experience F. Perona1, A. Zerbi1, M. Cugnasca2, A. Barile3, A. Morbidelli2, A. De Troia4; 1Diagnostic & Interventional Radiology, Galeazzi Institute, Milan, Italy, 2Vascular Surgery, Galeazzi Institute, Milan, Italy, 3Radiology, ASL 2, Savona, Italy, 4Vascular Surgery, Galeazzi Institute, Milan, Italy.

Purpose: Percutaneous portal vein embolization (PVE) is a technique utilized to induce hepatic hypertrophy and make more extensive hepatectomy possible. The purpose of this study is to determine the technical success and complications of pre-operative PVE with polyvinyl alcohol (PVA) particles. Materials & Methods: From 9/99 to 2/03, 40 patients with hepatic malignancies without underlying cirrhosis underwent 41 PVE procedures with PVA. Patient demographics and technical data were gathered prospectively. After PVE, patients were evaluated for complications and post-embolization syndrome (PES). Results: Technical success was achieved in 40/41 procedures (97.6%). In one patient, the right portal vein could not be accessed via a right-sided approach; hence a left-sided approach was utilized. Puncture site was right anterior (14), right posterior (21), and left (5). The only complication encountered was intractable hypertension in a single patient following PVE requiring cardiology consultation. Classic PES was not encountered in any patient. Length of hospitalization averaged 1.8 days. Conclusion: PVE with PVA is a safe and effective pre-operative procedure for inducing contralateral hepatic hypertrophy. Site of puncture did not influence technical success or incidence of complication. The complication rate in this series is significantly lower then that previously reported using other methods.

Purpose: Interventional radiologists have witnessed the technical evolution of gastrostomy from the first surgical/endoscopical approaches to percutaneous interventional procedures. The aim of this study is to evaluate the results of a ten-year series. Materials & Methods: During the period of December 1992 – December 2002, 187 new consecutive gastrostomies and 236 replacement procedures were performed. All cases used T- fastener gastropexy and tube placement. For all procedures, the indications, patient selection, procedure duration, type of sedation, morbidity, and mortality were examined. Results: All procedures were carried out successfully. One patient with severe neurologic disorders (0.53%) died of a stroke after tube placement without signs of complications related to the procedure. Six (3.2%) major complications occurred (three duodenal lesions with peritoneal leakage, two gastric bleeding, and one gastric lesion). Minor complications including pain at the site of the T-fastener, skin excoriation, and tube occlusion were easily managed. Conclusion: This series of consecutive patients confirms that a team of interventional radiologists can provide excellent service for gastrostomy placement and long term tube management. Percutaneous gastrostomy is a reliable technique which matches therapeutic needs even in high risk patients .

6.4.3

6.4.6

Percutaneous drainage of 300 intraperitoneal abscesses with longterm follow-up D. Akýncý1, N. Karabulut2, A. Öztürk1, M. Karçaaltýncaba1, B. Çil1, S. Geyik1, M. Özmen1, O. Akhan1; 1 Department of Radiology, Hacettepe University School of Medicine, Ankara, Turkey, 2Department of Radiology, Pamukkale University School of Medicine, Denizli, Turkey.

Correct needle alignment during percutaneous pelvic osteoplasty: projections, patient position, and results of treatment A. D. Kelekis, H. Yilmaz, K. Lovblad, A. Mehdizade, D. A. Ruefenacht, J. Martin; Radiology, University Hospital of Geneva, Geneva, Switzerland.

Purpose: To evaluate the efficacy of percutaneous drainage of intraperitoneal abscesses, with attention to recurrence and failure rates. Materials & Methods: A retrospective analysis was performed of percutaneous treatment of 300 intraperitoneal abscesses in 255 patients (147 male; average age: 38 years, range: 40 days-90 years) for whom at least one-year follow-up data were available. Abscesses were drained with fluoroscopic, sonographic, or computed tomographic guidance. Nine abscesses were drained by simple aspiration, whereas catheter drainage either by Seldinger or trocar technique with 6-14 Fr catheters was used in the remaining 291 abscesses. Results: Initial cure and failure rates were 68% (203/300) and 12% (36/ 300) respectively. Sixty-one abscesses (20%) were either palliated or temporized. The recurrence rate was 4% (12/300) and nine of thes 12 were cured by recatheterization, while the other three were treated by medication or surgery. The overall success and failure rates after the catheterization of recurrences were 91% (273/300) and 9% (27/300) respectively. Overall mortality rate was 3.1% (8/255). Mean drainage time was 12 days. Conclusion: Intraperitoneal abscesses with safe access routes should be drained percutaneously because of the high success rate and the low morbidity, mortality and recurrence rates.

Purpose: To report clinical experience with radiological projections, needle placement and results of percutaneous pelvic osteoplasty. Materials & Methods: During the last three years percutaneous pelvic osteoplasty was performed, usually under general anesthesia, in a consecutive series of 25 patients with intractable pain from bony metastases. Positioning was supine or prone, depending on lesion site, with multiplanar fluoroscopic guidance using specific anatomic landmarks. Among 44 treated lesions, 14 were sacral, eight ilial, 14 iliopubic, and eight acetabular. CT was used to assess implant distribution. Clinical evaluation included immediate and delayed follow-up studies of general condition and neurological status. Results: Six different projections were established and standardized, achieving safe needle placement. CT established the adequacy of cement distribution. Follow-up was available in 26/28 patients. Complete pain relief was obtained in 21/26 (81 %), partial in 3/26 (11.5 %), and no pain relief 2/26 (7.5 %). Pain medication could be withdrawn in 24/26. Overall mobility improved in 24/26. One complication was observed, due to leakage. One case of superior iliac wing lytic metastasis remained painful. Conclusion: Safe standardized needle placement is achievable under fluoroscopy for percutaneous pelvic osteoplasty. Treatment failure was related to PMMA injection.

6.4.5

6.4.7

Kyphoplasty: a promising method? The Heidelberg experience G. Noeldge1, J. Hillmeier2, C. Kasperk3, H. Meeder2, G. W. Kauffmann1; 1Diagnostic Radiology, University Hospital, Heidelberg, Germany, 2Traumatology, University Hospital, Heidelberg, Germany, 3Endocrinology, University Hospital, Heidelberg, Germany.

MRI monitoring and characterization of post-vertebropasty lesions A. D. Kelekis, K. Lovblad, H. Yilmaz, T. Somon, J. Delavelle, D. A. Ruefenacht, J. Martin; Radiology, University Hospital of Geneva, Geneva, Switzerland.

Purpose: The effects on pain relief and vertebral stabilization of kyphoplasty with PMMA and biocement injections were evaluated. Materials & Methods: CT-/fluoroscopic-guided kyphoplasty was performed under general anesthesia in 92 patients (19 men) aged 18-88 (mean: 72y). Results: PMMA was used in 68 patients and biocement in 24; 168 vertebrae in total were treated with injections of 3.5-9.9 ml (mean: 4.5 ml) between the 8th thoracic and the 4th lumbar vertebral body. In 38 patients one vertebral body was injected, in 40 cases two, and in 14 three. Minor cement leakage occurred in 15.2% of cases without neurologic symptoms. Two major complications occurred: one patient with misplacement of the puncture set and one patient with bleeding 12 hours later (unknown clotting disturbance). An optimal pain relief was achieved immediately in 92% of cases, in 75% of them this persisted for three months and in 62% for six months. Recurrence of pain in adjacent untreated bodies occurred in 25% of cases six months later. Follow-up was done clinically and by CT, MR and plain films up to six months. Conclusion: Our preliminary results in the treatment of pain due to osteoporotic fractures are encouraging. An associated endocrinologic medical treatment is mandatory.

Purpose: MR imaging, by showing bone edemas and epidural involvements, is a well-established pre-operative technique. In this study, aimed at evaluating MR as a follow-up tool, post-procedural MR images were analyzed and implant signal was characterized. Materials & Methods: Twelve patients candidates for vertebroplasty were examined by MR and CT before and after the treatment. MR sequences included FSE T1- and FSE T2-weighted images in sagittal and axial planes. Additionally, three cadavers were scanned before and after multilevel vertebroplasty. Images were correlated with anatomopathology and histopathology reports. Results: In post-vertebroplasty MR images, PMMA was visualized as hypointense, corresponding exactly to the cement on post-procedural axial CT scans. Pre-treatment bone edemas, visualized as hypointense in T1-weighted images, disappeared at follow-ups. The presence of a hyperintense rim was noted on both T1- and T2-weighted images, in patients and in cadavers. Histopathology reports confirmed the absence of fibrosis in cadaveric material. Conclusion: Post-vertebroplasty, MR showed cement as hypointense in T1- and T2-weighted images correlating well with CT. The presence of a hyperintense rim did not correspond to post-procedural fibrosis or edema, but to a susceptibility artifact. MR imaging can be used not only to evaluate implant distribution, but treatment efficacy also.

6.4.8

6.5.1

Ultrasound-guided percutaneous 14-G core skeletal muscle biopsy in neuromuscular disease P. J. O'Sullivan1, G. Gorman1, M. Farrell2, M. P. Logan1; 1Radiology, Beaumont Hospital, Dublin, Ireland, 2Pathology, Beaumont Hospital, Dublin, Ireland.

Transcatheter embolization of male varicocele with Fibro-Vein® mousse infusion: experience with 103 patients C. A. Reale, R. Gandini, S. Fabiano, A. Spinelli, V. Pipitone, M. Stefanini, G. Simonetti; Diagnostica per Immagini, Policlinico Tor Vergata, Rome, Italy.

Purpose: To evaluate the sensitivity of ultrasound-guided percutaneous muscle biopsy in investigation of neuromuscular disorders. Materials & Methods: Ultrasound-guided needle biopsy of skeletal muscle was performed using a 14-G Core Biopty System in 40 patients over 24 months. All patients were referred from the department of neurology under investigation for neuromuscular disorders. Ultrasound was used to find suitable tissue and avoid major vascular structures. Twothree milliliters 1% lignocaine anaesthetic were applied subcutaneously; a 3-4 mm incision was made. Three samples were obtained using a 14-G Biopty gun from vastus lateralis, deltoid, biceps and/or gastronemius. Samples were placed on saline, dampened gauze and sent to neuropathology. Results: One-hundred twenty samples were obtained in total; 99 samples (72.5%) were of diagnostic quality. Overall, 38 of 40 patients (95%) had a diagnosis made on the basis of needle biopsy. No complications, early or late, were recorded in any patient. Conclusion: Ultrasound-guided percutaneous 14-G core skeletal muscle biopsy is an easy and well tolerated procedure that produces reliable results and avoids open surgical biopsy. Three biopsy samples should be taken at each site to ensure diagnostic tissue is obtained.

Purpose: The aim of this study was to evaluate efficacy and safety of the Fibro-Vein® mousse in the treatment of male varicocele. Materials & Methods: From January 2000 to January 2002, 103 patients (mean age: 28.2 years; range: 15-44) were treated by sclerotization of the pampiniform plexus venous dilatation. Ten patients only (9.7%) had bilateral varicocele. All the patients had infertility and/or pain; the diagnosis was routinely obtained by physical examination and Doppler ultrasound. Phlebography was done under local anesthesia with access through the basilic or the femoral vein using a 4-F Simmons 2 or multipurpose catheter; in all the patients embolization was performed with Fibro-Vein® mousse at 3% and air (ratio 1:4). Clinical follow-up was routinely performed by telephone calls and Doppler ultrasound controls. Results: A technical success was obtained in all but seven patients (6.79%); of the 51 patients (49.51%) with infertility, 32 (62.7%) achieved pregnancy after the treatment; of the 42 patients with pain (40.77%) 40 had a relief after the treatment (95%). Conclusion: Male varicoceles treatment by Fibro-Vein® mousse and air is a safe and very effective technique that allows an easy and quick pampiniform plexus sclerotization.

6.4.9

6.5.2

Treatment of secondary hypertension and Cushing's syndrome by CT-guided percutaneous radiofrequency thermal ablation (RFTA) of primary adrenal tumours P. A. Almeida1, M. Pêgo2, I. Paiva3, F. Carrilho3, L. Providência2, M. Carvalheira3, M. Ferreira1; 1Radiology, University Hospital Coimbra, Coimbra, Portugal, 2Cardiology, University Hospital Coimbra, Coimbra, Portugal, 3Endocrinology, University Hospital Coimbra, Coimbra, Portugal.

The Günther-Tulip retrievable filter: prolonged temporary filtration by frequent percutaneous repositioning within the inferior vena cava M. Gimeno, B. Madariaga, J. Medrano, P. Gamboa, M. Herrera, M. De Gregorio; Interventional radiology, Clinico Universitario Lozano Blesa, Zaragoza, Spain.

Purpose: Radiofrequency thermal ablation (RFTA) in the percutaneous treatment of functioning adrenal adenomas in patients with secondary high blood pressure (HBP) due to primary hyperaldosteronism and in patients with Cushing’s syndrome. Materials & Methods: We used RTFA to treat adrenal tumors in 12 Caucasian patients (four men). In three cases, a CT-guided biopsy prior to RTFA was performed. RFTA was carried out for primary hyperaldosteronism in ten patients (four men) and for glucocorticoid-secreting tumors in two women. Lab values of plasma renin activation, plasma concentration of aldosterone, glycemia, cortisol, urinary-free cortisol, and adrenocorticotropic hormone were determined. Complications, hospital stay and costs were analyzed. Results: Results were assessed by clinical examinations, lab values, and HBP measurements six months after RFTA. A normal BP was achieved in nine patients (75%). In the remaining patients, the diary intake of antihypertensive drugs was reduced. In the group treated with RFTA for secondary HBP due to hyperaldosteronism, seven patients (70%) were healed; in both patients treated for glucocorticoid-secreting tumors plasma levels of cortisol normalized within 24h. Conclusion: The results obtained by RFTA are comparable with those achieved by percutaneous ethanol injection or surgery.

Purpose: To report our experience using the Günther-Tulip retrievable Filter (GTF) as a means of temporary caval filtration utilizing a technique that prolongs filter dwell time beyond 14 days. Materials & Methods: We implanted 90 GTF in 90 patients (39 female, 51 male) between December 1998 and October 2002. Indications were massive PE/DVT (58); DVT or PE with anticoagulation contraindicated (18); prophylaxis in high risk surgical patients (8); major trauma (2); reTEP in anticoagulated patients (3); and DVT-PE during pregnancy (1). GTFs were placed with the intention of retrieval in 14 days. In 27 there was a need to prolong filtration. This was successfully accomplished by repositioning the filter to a different location within the inferior vena cava (IVC) prior to definitive removal. Results: 76 GTF were successfully retrieved, and 14 were left permanently. 49 were removed after initial implantation (mean time 13.7 days, range 7 – 19). 27 patients required repositioning to prolong temporary caval filtration, with a mean dwell time of 33.4 days (range 21– 62); mean repositioning procedures were 1.51 (range 1–4). Conclusion: The GTF approaches the ideal caval filter, usable either as a temporary or a permanent filtering device. It can be repositioned to prolong filtration time.

6.5.3

6.5.5

Superior vena cava obstruction: treatment of 18 patients using endoluminal stents H. Bjarnason, W. Wang, S. Misra, M. A. McKusick, J. C. Andrews, M. C. Johnson, E. A. Sabater; Radiology, Mayo Clinic Rochester, Rochester, United States.

Percutaneous venous thrombectomy using the Arrow-Trerotola percutaneous thrombolytic device with temporary caval filtration: invitro investigations J. E. Wildberger1, P. Haage1, J. Bovelander2, J. G. Pfeffer1, C. Weiß3, D. Vorwerk4, T. Schmitz-Rode1, R. W. Günther1; 1 Dept. of Diagnostic Radiology, University Hospital, Aachen, Germany, 2 Department of Clinical Research, Arrow Int., Delft, Netherlands, 3Dept. of Medical Statistics, University Hospital, Aachen, Germany, 4Dept. of Diagnostic Radiology, Klinikum Ingolstadt, Ingolstadt, Germany.

Purpose: To present the outcomes of endovascular stent placement for Superior vena cava (SVC) obstruction Materials & Methods: Procedural reports were reviewed for 18 patients (11 male) who underwent endovascular stent placement for SVC obstruction between January 1996 and May 2002. The average age was 49.4 years(range from 22 to 78). All patients underwent SVC +/- innominate vein stent as well as balloon angioplasty, and all had SCV syndrome. The median follow up was 18.5 months. Results: Six patients had benign SVC obstruction (5 fibrosing mediastinitis, one central line) and 12 had malignant obstruction (breast cancer 1, lung cancer 11). The 6 patients with benign causes are still alive after a median follow up of 30 months. Nine of 12 patients with malignant causes have died, with median survival 11 months. Primary stent patency was 18.1 months for benign causes and 10.8 months for malignant causes. Only 2 of 9 deceased patients underwent PTA due to stent stenosis prior to death. Immediate symptom relief was documented in 14 patients. Conclusion: Stenting for SVC obstruction is effective and can achieve immediate symptom relief. Survival in patients with malignant obstruction is brief, and most did not require re-intervention.

Purpose: To evaluate the size/quantity of downstream emboli using the Arrow-Trerotola percutaneous thrombolytic device (PTD) with/without temporary filtration for mechanical thrombectomy (MT). Materials & Methods: Iliocaval thrombi were simulated in a flowmodel. Each experiment was performed 10 times, measuring thrombus particles and distribution in the effluent. 1) After retrograde insertion, MT was performed using the PTD. 2) A tulip-shaped vena cava stent-filter was inserted additionally and removed after the intervention, without further manipulation. A thrombus-stuffed filter was carefully closed within the flow circuit: 3) without additional MT and 4) after running the PTD within the filter lumen. 5) MT was performed within the thrombus-filled tubing and in the filter lumen. The latter was closed at the end of the procedure. Both devices were removed from the flow circuit. Results: Experiment1 led to fragmentation of 75.5% of the clot into 500µ particles. Controlled closing of the thrombus-filled filter within the flow circuit macerated 75.2%, while additional MT (running the PTD within the occluded filter) led to dissolution of 90.4%. In the final setup, an overall fragmentation rate of 99.6% was achieved. Conclusion: The combined use of the Arrow-Trerotola PTD with temporary filtration proved to be effective for removing even large clots.

6.5.4

6.5.6

Pitfalls in arterial stimulation with venous sampling for localization of insulin-secreting tumors T. Pfammatter1, P. Wiesli2, C. Schmid2, B. Marincek1; 1Institut für Diagnostische Radiologie, Universitätsspital Zürich, Switzerland, 2Abteilung Endokrinologie, Universitätsspital Zürich, Switzerland.

Experience with more than 3000 central venous access devices in interventional radiology R. Uflacker, L. Williams, J. B. Selby, C. Hannegan; Interventional Radiology, Medical University of South Carolina, Charleston, United States.

Purpose: To report the reasons for false-negative arterial stimulation with venous sampling (ASVS-tests). Materials & Methods: ASVS was performed in 24 patients evaluated for hypoglycemic disorders. Final diagnoses were insulinoma (n=21), islet cell hyperplasia (n=2) and hepatic insulin secreting neuroendocrine carcinoma (n=1). Results: In five cases samples were initially negative after calcium stimulation. Repeated testing with glucose instead of calcium stimulation (n=1), after discontinuing medication with diazoxide or calcium antagonists (n=2), or widening the time interval between the stimulations in a patient with high basal insulin led to correct localization in four of these patients. Conclusion: If these pitfalls had been avoided, ASVS would have yielded true positive results in 23 of 24 patients.

Purpose: To report the outcomes of patients with central venous access (CVA) systems. Materials & Methods: Three-thousand twenty patients, age range three weeks to 91 years, were treated by interventional radiology. Multiple systems were placed, picclines, tunneled catheters, temporary catheters, subcutaneous ports, and others. The patients were managed by the CVA nurse coordinator, which kept the information. Outcomes included infection rate, revisions and function of the devices. Results: In 1999, 591 devices were placed, including 191 picclines, 82 ports, 261 tunneled and 28 temporary catheters. Infection rate was (+)cultures in 8.5%. In 2000, 783 devices were placed, including 304 picclines, 92 ports, 311 tunneled and 55 temporary catheters. Infection rate was (+)cultures in 9.4%. In 2001, 1038 devices were placed, including 529 picclines, 84 ports, 390 tunneled and temporary catheters. Infection rate was (+) cultures in 7.8%. In 2002, 608 devices were placed, including 256 picclines, 63 ports, 214 tunneled and temporary catheters. Infection rate was (+) cultures in 5.7%. Conclusion: The volume of CVA placement is increasing at 20%/year. The infection rate is dropping and the longevity of the accesses is increasing. The venous access coordinator played important role in the management of the patients.

6.5.7

6.6.2

Efficacy of percutaneous interventions in thrombosed hemodialysis access: our six-year experience M. Rocek, J. H. Peregrin; Dpt. Diagnostic and Interventional Radiology, IKEM, Prague 4, Czech Republic.

Pelvic abscess drainage in children: the trans-gluteal approach A. M. Cahill1, C. R. Fitz2, R. D. Kaye1, R. B. Towbin1; 1Radiology, Children's Hospital of Philadelphia, Philadelphia, United States, 2Radiology, Children's Hospital of Pittsburgh, Pittsburgh, United States.

Purpose: To present a review of 6 years' experience with percutaneous interventions in thrombosed hemodialysis access with high-volume blood clots. Materials & Methods: Between 1/1/1997 and 31/12/2002 215 patients with thrombosed hemodialysis access were treated percutaneously using PTA, Amplatz thrombectomy device, Arrow-Trerotola PTD, or Hydrolyser and aspiration. The study included 157 patients with occlusions longer than 10 cm and at least 6 mm in diameter. Results: The clinical success rate was 90% (142 of 157), 88% (69 of 78) for synthetic grafts vs. 92% (58 of 63) for native fistulas vs. 94% (15 of 16) for allografts. Mean procedure time was 101 minutes (104 vs. 130 vs. 77 minutes, respectively). We observed nine complications: access rupture in three cases, peripheral arterial embolism successfully handled by percutaneous aspiration in two cases, pseudoaneurysm in two cases, and huge hematoma in two cases. One-year primary patency and assisted primary patency were 18% and 32% for synthetic grafts vs. 73% and 82% for native fistulas vs. 23% and 41% for allografts, respectively. Conclusion: Mechanical interventions are effective procedures for percutaneous treatment of thrombosed hemodialysis access. Our experience shows the technique to be safe and rapid for both native fistula and graft occlusions.

Purpose: To describe our experience with percutaneous pelvic abscess drainage using the trans-gluteal approach. Materials & Methods: From September 1995 to April 2002, 41 children (16M) had pelvic abscess drainage performed via the trans-gluteal approach; mean age was 14.4 years, range 7.8-19. The indications included appendicitis with perforation (35), Crohn's disease (2), perineal injury (1), spina bifida related infection (2), and sigmoid perforation (1). Percutaneous guidance was performed with CT in 33 patients (80%) and ultrasound and fluoroscopy in 8 patients (20%). All procedures were performed under sedation. Results: All percutaneous drainage procedures were successful. The mean volume of aspirate was 98 cc (range 20-250cc). A positive culture was obtained in 21/41 patients. The mean duration of catheter placement was 4 days (range 2-10days). Two major complications occurred: one patient sustained a significant pelvic hematoma after drain removal, the second patient developed a bowel fistula 4 days post catheter placement. There were no minor complications. Conclusion: The trans-gluteal approach to pelvic abscess drainage is a safe and effective method of pelvic abscess drainage, providing a safe alternative to trans-rectal pelvic abscess drainage.

6.6.1

6.6.3

Air-online: feasibility of a multi-center global database for interventional radiology procedures M. Sapoval, A. Air-online user group; Cardio Vascular radiology, HEGP, Paris, France.

Percutaneous CT-guided radiofrequency thermal ablation of small, unresectable non-small cell lung cancer R. Lencioni1, L. Crocetti1, R. Cioni1, C. Franchini1, C. A. Angeletti2, C. Bartolozzi1; 1Department of Radiology, University of Pisa, Pisa, Italy, 2Department of Thoracic Surgery, University of Pisa, Pisa, Italy.

Purpose: To assess the feasibility of an online database allowing each center to create their own database of vascular interventional radiology (IR) procedures. Materials & Methods: Eight centers entered the phase 1 of the "Aironline" project which started in January 2001. Current web browsers were used to access www.air-online.org and confidential login/passwords for all the radiologists of these centers were attributed. All centers were asked about the user-friendliness of the application and main limitations of use. Results: All centers answered that Air-online is user-friendly and allowed to create an effective database. The time needed to enter a case ranged from three to 15 minutes, according to the level of training and the quality of the local network. In some centers, the main limitation was the lack of a PC in the angio-suit. More than 1400 files are now in the database. Conclusion: According to this preliminary experience, Air-online allowed to enter rapidly in all the files and it was quoted as an userfriendly tool by all users. The widespread employment of Air-online could be of help in gaining more data on the actual activity and results of IR procedures over Europe and could assist in the quality control.

Purpose: To assess feasibility, safety, and effectiveness of percutaneous CT-guided radiofrequency (RF) thermal ablation of small, unresectable, non-small cell lung cancer (NSCLC). Materials & Methods: Ten patients with unresectable, biopsy-proven, NSCLC, 3 cm or less in diameter (range, 1.1-3 cm; mean, 2.1 ± 0.6 cm), were enrolled in a pilot clinical trial. Patients underwent CT-guided RF ablation under consciuos sedation by using a 150-W RF generator and 15-G, 9-hook expandable electrode needles (RITA Medical Systems, Mountain View, CA). Follow-up period ranged from three to 11 months (mean, 6.1 ± 2.1 months). Results: Single-session ablations were technically successful in all cases. Pneumothorax was observed after the procedure in four of the ten patients, two of whom required drainage. No other complication occurred. At one-month CT studies, an area of thermal injury encompassing the native tumor was observed in all cases. Treated areas showed a reduction in size in 10/10 patients assessed three months after the procedure and in 6/6 patients assessed six months after the procedure. No signs of recurrence were detected on CT scans in any case. Conclusion: Preliminary results of this pilot clinical study suggest that RF ablation could enable successful percutaneous treatment of small NSCLC with acceptable morbidity.

6.6.4

6.6.6

Microsphere localization in the vasculature of embolized nasopharyngeal angiofibromas and paragangliomas R. Chapot1, M. Wassef2, Y. Wang3, P. Tran ba huy4, J. Mikol2, E. Houdart1, L. Feng3, A. Laurent1; 1 Neuroradiology, Lariboisière hospital, Paris 75010, France, 2Pathology, Lariboisière hospital, Paris 75010, France, 3Neurosurgery & Neuroradiology, Xuan Wu Hospital, Beijing, China, 4ENT Surgery, Lariboisière hospital, Paris 75010, France.

Treatment of low-flow angiomas by local injection of Polidocanol S. Baus1, T. Weiss2, M. Pichlmaier3, A. Chavan1, M. Galanski1; 1Diagnostische Radiologie, Medizinische Hochschule Hannover, Hannover, Germany, 2Kardiologie und Angiologie, Medizinische Hochschule Hannover, Hannover, Germany, 3Thorax-, Herz- und Gefäßchirurgie, Medizinische Hochschule Hannover, Hannover, Germany.

Purpose: To determine histologically where calibrated microspheres (MS) block the vasculature of paragangliomas and nasopharyngeal angiofibromas. Materials & Methods: Embolization with 100, 200, 400, 600, 800 and 1000 mum MS (Embosphere, Biosphere Medical); surgery in 25 PGG and 25 NPA patients. Measurements of: size of occluded vessels and MS, and MS intra- or extra- tumoral (INT and EXT) localization. Results: Among 1061 occluded vessels, the median number of MS per vessel was one (mean SD : 2.2 ± 4.2). The diameter of occluded vessels correlated with MS size (r=0.537, p< .0001) and increased significantly with MS size (p<.0001). INT/EXT percentages were 90%/10%. INT Vessel diameter was significantly smaller than EXT (mean±SD: 298±196, median : 240, 443±198 median: 398, respectively, p < .0001). No difference was found between the two tumor types in terms of size of occluded vessels or INT/EXT percentages Conclusion: The size of vessels occluded by MS and their intra- or extra-tumoral location depend directly on the size of MS injected. Penetration of calibrated MS seem to be identical in paraganglioma and nasopharyngeal angiofibroma vessels, suggesting that the vasculature of both tumors have similarities.

Purpose: To evaluate the effectiveness of sclerotherapy with Polidocanol injections as a treatment option in vascular malformations. Materials & Methods: A group of 33 patients was evaluated by an interdisciplinary panel. After the exclusion of surgical and medical options, 14 lesions in 12 patients were treated with local injections of Polidocanol. A local pain was the main symptom in all the patients, aggravated by an occasional swelling in some cases and resulting in a significant reduction of daily activities in all of them. Diagnostic workup included clinical examination and palpation, ultrasound including color-Doppler studies, phlebography of the affected limb, magnetic resonance imaging and digital subtraction angiography (DSA). All interventions were carried out under local anaesthesia. Peri-intervention medications included anti-inflammatory drugs and pain killers on demand. Results: The treatment was technically successful in 100% of cases with relief of symptoms in all the patients. No major complications occurred. Minor complications included local swelling, hematoma, pain and temporary loss of function due to swelling. Circumscribed skin necrosis occurred in one patient. Conclusion: Local injections of Polidocanol in the treatment of lowflow malformations of the extremities is a promising option requiring no general anesthesia.

6.6.5

6.6.7

Management of gastrointestinal bleeding by means of Embosphere microspheres V. Pipitone, R. Gandini, C. A. Reale, M. Stefanini, S. Fabiano, G. Pendenza, G. Simonetti; Diagnostic Imaging, University of Tor Vergata, Rome, Italy.

Five-year experience using three-dimensional ultrasound (3-D US) to assist interventional procedures: when it helps and when it does not S. C. Rose, T. B. Kinney, A. C. Roberts, K. Valji, F. J. Miller, G. M. Rivera SanFeliz, T. R. Nelson, D. H. Pretorius; Radiology 8756, University of California, San Diego, San Diego, United States.

Purpose: To assess efficacy and complications in the management of superselective embolization by the use of Embosphere microspheres (BioSphere). Materials & Methods: Eleven patients (age range 25-77 years; mean 59.05) with acute gastrointestinal bleeding underwent transfemoral angiography with subsequent superselective embolization by Embosphere. Four patients had upper and seven had lower gastrointestinal bleeding. Embosphere (500-700 m) microspheres were employed after superselective catheterization of the bleeding feeding artery with the use of coaxial catheter systems. Results: In all the patients, hemostasis was achieved and bleeding was successfully stopped after the first intervention. Conclusion: The use of Embosphere as an embolizing material is an efficient and safe alternative to other embolizing materials utilized in the treatment of gastrointestinal bleedings.

Purpose: To review our institutional experience using 3-D US to assist interventional procedures in order to define the most appropriate use of this technology. Materials & Methods: Between 11/1997 and 11/2002, 3-D US was used in a variety of interventional procedures. Information provided by 3-D US, technique modifications, and limitations were tabulated. Retrospective analysis was performed to determine clinical scenarios best suited for 3-D US, specific information provided, and how interventional techniques were modified. Results: Three-dimensional US was used in 149 procedures (123 patients). Procedures included TIPS (51), transjugular liver biopsy (21), focal ablation of liver malignancies (29), percutaneous drainage of complex abdominal fluid collections (28), endocavitary drainage of pelvic fluid collections (11), aspiration of liver cysts (4), drainage of musculoskeletal abscesses (2), biliary drainage (2), and empyema drainage (1). Three-dimensional US was unable to visualize relevant structures in 15 (10%) procedures; 3-D US added information in 128 (86%) procedures, and modified interventional technique in 105 (70%). It was most useful in TIPS procedures, transjugular liver biopsies, and drainage of complex abdominal and pelvic fluid collections. Conclusion: In selected applications, 3-D US can provide significant unique structural and relationship information that frequently leads to modifications of interventional technique.

6.6.8

15.1.1

Percutaneous CT-guided lung biopsy in children: a ten-year review A. M. Cahill1, C. R. Fitz2, R. D. Kaye1, R. B. Towbin1; 1Radiology, Children's Hospital of Philadelphia, Philadelphia, United States, 2Radiology, Children's Hospital of Pittsburgh, Pittsburgh, United States.

Balloon occlusion techniques enhance detection of endoleaks in patients who have had endovascular repair of abdominal aortic aneurysms T. B. Kinney1, E. Owens2, N. H. Kumins1, G. M. Rivera SanFeliz1, S. C. Rose1, K. Valji1, A. C. Roberts1; 1 Radiology, UCSD Medical Center, San Diego, United States, 2Surgery, Veterans Administration Hospital, San Diego, United States.

Purpose: To describe our techniques and results of CT-guided percutaneous lung biopsy in children. Materials & Methods: From April 1992 to June 2002, 71 CT guided lung biopsies were performed on 64 children (35M), mean age 10.2 yrs, range 0.65-22 yrs. The indications included unresolving chest x-ray abnormality (34), metastatic disease (23), post-transplant PTLD/fungal infection (10), and chronic granulomatous disease (4). Fifty-nine (83%) biopsies were performed using a coaxial core biopsy technique, 10 (14%) using an aspiration technique and 2 (3%) with both. The mean diameter was 2.5 cm (range 1-10cm) in the core biopsy group and 1.0 cm (range 0.51.7cm) in the aspiration biopsy group. Results: 60/71 biopsies (84%) were diagnostic; 11 were non-diagnostic, and of those 6 underwent diagnostic surgical biopsy, 3 repeat diagnostic percutaneous biopsy and 2 were treated conservatively. Complications included asymptomatic focal pneumothorax (6), asymptomatic peri-lesional hemorrhage (3), and tension pneumothorax (1). One delayed complication occurred: a pneumothorax 7 days post lung biopsy Conclusion: 1. Percutaneous CT guided lung biopsy is a safe and accurate method in children. 2. Percutaneous techniques obviate the need for open surgical biopsy in most patients

Purpose: To evaluate the added benefit in detecting endoleaks AAA stent-graft cases using balloon occlusion technique in addition to conventional angiography. Materials & Methods: Forty-nine patients had 43 Ancure and six AneuRx endografts placed for AAAs (1/00 and 12/02). Ten patients underwent 14 procedures for problems including growth of endosac (4), lack of shrinkage of endosac (4), type I/II leaks on CT (4/2) and loss of pulses or claudication (2). In eight of these cases, patients were studied with conventional angiography and retrograde bilateral iliac occlusion balloon angiography to identify leaks. Results: Only two subtle type-I leaks were detected in the eight patients. Retrograde balloon occlusion angiograms showed five type-I (distal limbs), three type-II, and one gate leak (AneuRx). All patients with type-I leaks were treated: four reduced their sac sizes and one has stabilized. The gate leak patient was treated, but lost to follow-up. Both type-II leaks have reduced sac sizes, one without treatment and one with embolization and sac puncture. Conclusion: While conventional angiography may show various endoleaks, the application of retrograde occlusion balloon angiography in the common or external iliac arteries may unmask more leaks, which respond to therapy.

6.6.9

15.1.2

Embolization in the management of diffuse pulmonary arteriovenous malformations C. Lagrange, M. El Hajjam, J. P. Pelage, A. Demir, A. Foudali, S. Chagnon, P. Lacombe; Radiology, Hopital Ambroise Paré, Boulogne, France.

Aortic aneurysm sac filling to prevent endoleaks after stent-graft treatment: experience with an animal model R. Uflacker1, T. Brothers2, C. Hannegan1, J. B. Selby1; 1Interventional Radiology, Medical University of South Carolina, Charleston, United States, 2Vascular Surgery, Medical University of South Carolina, Charleston, United States.

Purpose: To evaluate the outcome of embolization to treat diffuse pulmonary arteriovenous malformations (PAVM) and assess improvement of blood oxygenation after embolization. Materials & Methods: Embolization was performed in 14 patients (12 women, mean age 40.3 years) with diffuse PAVM, all secondary to hereditary hemorrhagic telangiectasia. Coils of a size depending on the diameter of the feeding artery were deployed in all angiographically visible PAVMs. Post-procedural follow-ups included clinical assessment, arterial oxygen saturation (PaO2) measurements, and radiological evaluation. Results: A total of 230 PAVMs (mean 16.4/patient) were treated by 359 coils (mean 25.6) requiring one to seven procedures (mean: two). No technical failure or paradoxical coil migration was observed. Post-procedural complications included a transient ischemic attack in two patients and a pulmonary infarction in other two cases. After a mean follow-up of 48.3 months, an improvement was observed in all the patients. A significant improvement in the PaO2 (75.5 versus 56.5 mm/ Hg) was initially observed. In the long-term, however, hypoxemia (<85 mm) was still present in 10/14 patients. Conclusion: Embolization is a safe and effective treatment for diffuse PAVM. Clinical, physiological, and radiological follow-up is of crucial importance to confirm the successful and permanent occlusion of the treated lesions.

Purpose: To present the results of a new technique of abdominal aortic aneurysm (AAA) sac filling to prevent endoleaks in an animal model. Materials & Methods: Eight Yucatan pigs, six months of age, weighing 40-60 kg, were used to surgically create an AAA with a Dacron pouch. Lumbar arteries were kept patent. Two weeks after creation, AAA was treated with a straight tube stent-graft from a common carotid approach. The residual aneurysmal sac was filled with a variable amount of a hemostatic polymer, DEAC-Poly-N-Acetyl Glucosamine (DPNAG), (5-15 ml) according to the AAA size. There was filling of a lumbar artery in two animals. Follow-up was two to 12 weeks, with angiographic and pathologic examination. Results: AAA creation, stent-graft insertion, and DPNAG injection were successful in all animals, with elimination of initially present endoleaks in five animals and long-term sac obliteration. The AAA was permanently obliterated by an uniform filling polymer mass, and no endoleaks were observed. There was AAA shrinkage into a firm mass around the aorta, with no procedure-related complications. Conclusions: AAA sac filling with DPNAG seems to be adequate to avoid endoleaks and to promote shrinkage of the AAA in an animal model.

15.1.3

15.1.5

Seven-year results of endovascular abdominal aortic aneurysm treatment: embolization of collateral vessels improves clinical outcome and aneurysm shrinkage M. B. Pitton1, S. Herber1, J. Schneider1, H. Schweitzer1, C. Düber2, A. Neufang3, W. Schmiedt3, M. Thelen1; 1 Radiology, University Hospital Mainz, Mainz, Germany, 2Radiology, University Hospital Mannheim, Mannheim, Germany, 3Cardiovascular Surgery, University Hospital Mainz, Mainz, Germany.

Endovascular repair of ruptured abdominal aortic aneurysms: our preliminary experience G. Piffaretti1, R. Caronno1, D. Laganà2, G. Carrafiello2, M. Carnini1, M. Canziani1, P. Castelli1, M. Tozzi1; 1 Vascular Surgery, University of Insubria, Varese, Italy, 2Institute of Radiology, University of Insubria, Varese, Italy.

Purpose: Seven-year follow-up of endovascular aneurysm treatment. Does embolization of collateral vessels influence abdominal aortic aneurysm shrinkage? Materials & Methods: Endovascular treatment was performed in 121 patients using Vanguard stent-grafts in 53 cases and Talent grafts in 68. In group I patent collateral vessels remained untreated. In group II patent lumbar arteries and inferior mesenteric arteries were completely occluded before stent-grafting. Results: Technical success was 100% and thirty-day mortality 0.8% (myocardial infarction). Twenty patients died during follow-up due to cardiac insufficiency (n=11), cancer (n=6), others (n=3). Mean followup is 38 months for Vanguard (0-84 months) and 20 months for Talent stent-grafts (3-48 months). In 26/121 cases (21.5%,18 Vanguard, 8 Talent) a total of 45 radiological and/or operative reinterventions were performed including 11 graft explants. In contrast to group I, cases with completely occluded collateral vessels (spontaneously or by coil embolization, group II) showed reduced number and size of endoleaks. Group II showed significantly improved aneurysm shrinkage at 3-year followup (39.8±9.6mm (II) vs. 47.8±9.6mm (I), p<0.05) compared to postintervention diameter (59.7±13.6mm (II) vs 55.8±10.2mm (I), n.s.). Conclusion: Endovascular treatment is safe and effective in selected patients. Embolization of collateral vessels prior to stent-grafting improves clinical outcome and aneurysm shrinkage.

Purpose: To describe safety and efficacy of primary endovascular repair of ruptured abdominal aortic aneurysms (RAAA). Materials & Methods: In the last year, nine patients with RAAA were treated. Their mean age was 78y (range, 70-88y), RAAA mean diameter was 68 mm (range, 37-100). Each hemodinamically stable patient underwent preoperative helical-CT with 3-D reconstructions; in two patients hypovolemic shock occurred and an immediate repair was carried out. Bifurcated devices were always implanted and spiral-CT follow-ups were performed at one, three, six, and 12 months after the procedure. Results: Primary technical success rate was 100%; mortality and morbidity rates were 0. Average surgical time was 144 minutes (range, 60330) and mean blood loss was 244 cc (range, 100-950). Mean length of hospitalization was three days (range, 2-6). Mean follow-up was eight months (range, 3-12). One iliac branch occlusion--successfully treated with fem-fem bypass--and two type-II endoleaks--one from the patent inferior mesenteric artery and one from the lumbar artery spontaneously thrombosed four months after diagnosis--were recorded. Conclusion: Endovascular repair of RAAA is feasible and effective; it will probably become the first treatment option, particularly when a wide endograft choice will be available.

15.1.4

15.1.6

Endovascular treatment of abdominal aortic aneurysm with conscious sedation: a single center's experience E. N. Brountzos1, K. Malagari1, N. L. Kelekis1, A. Gougoulakis2, S. Argentos1, D. A. Kelekis1; 12nd Radiology, Athens University, Athens, Greece, 23rd Department of Surgery, Athens University, Athens, Greece.

Aneurysms treated with stentgrafts: MRI monitoring of their histologic organization J. Schneider, B. Blecher, S. Herber, P. Schmenger, G. Klos, M. Thelen, M. B. Pitton; Dept. for Radiology, University of Mainz, Mainz, Germany.

Purpose: Prospective evaluation of our experience with the endovascular treatment of abdominal aortic aneurysms (AAA) under conscious sedation. Endpoints were technical success, safety, and durability of the procedure. Materials & Methods: Fifty-two patients were treated with stentgrafting. The procedures were performed in the angiography suite using conscious sedation. Stent-graft selection was based on availability and anatomic criteria (Vanguard in 12, Excluder in 14, Zenith in 26). Followup duration was 930±446 days. Results: Primary technical success with accurate stent-graft placement was achieved in 51 of 52 patients (98%). Three type-I endoleaks occurred immediately after implantation, managed with cuff placement in two patients. In the remaining patient the endoleak healed spontaneously. There were two immediate type-II endoleaks. One patient required conversion to conventional surgery. Major complications occurred in two patients (3.8%) including one death. Two late type-I endoleaks occurred within six months and were managed percutaneously. The type-II endoleak rate was 9.6% at six months, 1.9% at 12 months, and 0 % at 24 months. Aneurysm diameter decreased from 59.6mm±9.9 to 52.3mm±8.9 at 24 months (p<0.001). Conclusion: In our experience the endovascular treatment of AAA with conscious sedation results in successful exclusion of the aneurysm with low risk and long-lasting effect.

Purpose: MRI monitoring for the histological organization of excluded aneurysms. Materioals & Methods: Twenty-eight patients with post-endovascular aneuryms treatment underwent MRI monitoring: T1-w TSE (TR173ms; TE4,1ms), T2-w (TR5572ms; TE99ms) and contrast-enhanced T1-w. Follow-up was at one week, three and six months, and annually. After aneurysm segmentation (complete aneurysm, preexisting mural thrombus (PMT), areas without organization) signal intensities (SI) were categorized and referred to histological organization classes previously experimentally defined: detritus, tissue with increasing fibre proliferation, and tight connective tissue. The distribution of distinct SIs in anatomical subregions were quantitatively analyzed. Results: Follow-ups showed various diameter changes. Aneurysm shrinkages were associated with decreasing SIs in T2-w and contrastenhanced (T1-w) images, thus indicating a histological organization. Considerable signal decreases (70%) (SI T2-w) 104±78 vs. 177±102) were associated with a diameter decrease of 42% after six months. A missing signal decrease indicated detritus or poor fibre proliferation and was associated with reduced aneurysm shrinkages. PMT showed low SIs (T2-w) without changes during follow-up. There was no correlation between PMT and the diameter decrease. Conclusion: T2-w sequences are superior for discriminating different organization classes. Decreasing SIs (T1-w, T2-w) and contrast enhancement (T1-w) indicate an ongoing organization. PMT was not correlated with aneurysm shrinkage.

15.1.7

15.2.1

The use of Wallstents to prevent iliac leg kinking of AAA stentgraft in patients with tortuous iliac arteries T. Jargiello1, T. Zubilewicz2, F. Brakowiecki3, J. Kesik2, M. SzczerboTrojanowska1; 1 Department of Interventional Radiology, University School of Medicine, Lublin, Poland, 2Department of Vascular Surgery, University School of Medicine, Lublin, Poland, 3Department of Vascular Surgery, District Hospital, Lublin, Poland.

Transjugular intrahepatic portosystemic shunt versus sclerotherapy for variceal rebleeding P. Popoviè, M. Šurlan, T. Kljuèevšek, P. Berden, J. Klanèar, D. Kuhelj, V. Salapura; Clinical Institute of Radiology, UMC, Ljubljana, Slovenia.

Purpose: We present our experience in the prevention of Zenith AAA stentgraft iliac leg kinking with the use of Wallstents. Zenith AAA stentgraft with its suprarenal fixation has a very good stability of the main body, but its modular construction with several Z-stents covered and connected by polyester material makes iliac leg kinking possible in tortuous iliacs. Materials & Methods: Among 132 patients with AAA treated with stentgrafts, in 92 cases aorto-biiliac Zenith devices were used. Twelve of them had tortuous iliac arteries with angles <90°. In the two first patients we did not intend to use any supporting stent during stentgraft planning. They both developed occlusion due to leg kinking corrected, in the first case, with a Wallstent during the procedure. The second patient developed ischemic symptoms three days after implantation. Reintervention with a Wallstent restored iliac leg patency. In next patients, Wallstent implantation into the iliac leg of the Zenith stentgraft was elective, planned together with stentgraft implantation. Results: None of these patients, followed-up for 9-36 months (average 19), had iliac leg kinking and symptoms of lower extremity ischemia. Conclusion: Wallstent implantation into the iliac leg of a Zenith stentgraft prevents kinking in patients with tortuous iliac arteries.

Purpose: The aim of our study was to compare elective TIPS with endoscopic sclerotherapy (ES) for the secondary prophylaxis of variceal bleeding in cirrhotic patients. Materials & Methods: From April 1994 to December 2000, 96 patients with recurrent variceal bleeding were randomly assigned to either a transjugular shunt or to an endoscopic treatment. Results: Median observation time was 35.5 months in the TIPS group and 19.6 months in the ES group. Cumulative rebleeding rates were 6.25 % in the TIPS group and 45.7% in the ES group (p=0.001; χ2 test). Incidence of mortality secondary to rebleeding was higher in the ES group (57% versus 15%). Mortality was 45.6% in the ES group and 27.1% in the TIPS group. Cumulative survival after TIPS was 83% at one, 79.9% at two, and 73.5% at four years; after ES it was 69.8% at one, 53.3 % at two, and 39.8% at four years (p = 0.013 log-rang). Incidence of new or worsening hepatic encephalopathy was less common in the ES group (2.2% versus 8.2 %; p = 0.316; χ2 test). Conclusion: Elective TIPS is significantly better then ES in preventing rebleeding, it improves survival and is associated with a similar rate of encephalopathy.

15.1.8

15.2.2

Endovascular treatment of abdominal aortic aneurysms: five years' experience F. Fanelli, F. M. Salvatori, M. Bezzi, G. Marcelli, D. Pepino, S. Vagnarelli, P. Rossi, R. Passariello; Department of radiological sciences, University of Rome, Rome, Italy.

Outcome after transjugular intrahepatic portosystemic shunt (TIPS) for refractory hepatic hydrothorax: a long-term study P. P. Goffette1, J. Y. Wilputte2, J. P. Lerut3, A. Godoy-Gepert2, F. Zech4, A. P. Geubel2; 1Interventional Radiology, St-luc University Hospital, Brussels, Belgium, 2Gastroenterology, St-luc University Hospital, Brussels, Belgium, 3Digestive surgery, St-luc University Hospital, Brussels, Belgium, 4internal Medicine, St-luc University Hospital, Brussels, Belgium.

Purpose: To report five years' experience in the treatment of abdominal aortic aneurysms (AAA). Materials & Methods: From September 1998, 94 patients with AAA (mean diameter 47.5 mm) were treated using various stent-grafts: 27 Vanguard-II (Boston-Scientific), 59 Excluder (WL-Gore), 9 Talent (World-Medical), 3 Anaconda (Sulzer-Medical), 4 Quantum (Cordis), and 2 Endologix (Endologix) Procedures were performed in the angio-suite (28) or operating-room (66). Results: Technical success with complete exclusion of the aneurysmatic sac was achieved in 93/94 patients, while in 1 case surgical conversion was necessary. After follow-up ranging from 1 to 59 months, 79 patients are alive. Sac diameter decreased in 31 patients (mean 2.3 mm), remained stable in 51 patients, and increased in 11 patients. Endoleaks were detected in 13 patients (13.9%): 7 type II, 2 type III, 4 type IV. Two late conversions were performed after 22 and 31 months for multiple fractures of the stent-graft and for a persistent type-II endoleak. In one case after 4 months occlusion of the right renal artery was observed due to stentgraft migration. Limb obstruction occurred in 11 patients (11.8%), resolved by femoro-femoral by-pass. Conclusion: Percutaneous treatment of AAA seems to be a feasible technique only in selected patients.

Purpose: To assess long-term results of TIPS in the treatment of hepatic hydrothorax and to evaluate predictors of early mortality. Materials & Methods: Between 1992 and 2001, 28 consecutive patients (mean age 54, Child class B/C: 43/57%) with hepatic hydrothorax refractory to the medical therapy underwent TIPS. The average follow-up was 358 (± 121) days. Mann Whitney's and Wilcoxon's rank tests were used for the comparisons of variables. Survival was expressed by Kaplan-Meier analysis. Results: Reduction in the pleural effusion volume with clinical improvement was observed in 68% of cases (Child B: 75%, C: 15%); a complete radiological disappearance of hydrothorax was documented in 57% after a mean period of four months. Thirty-day mortality was 14%, 90and 150-day mortalities were 25% and 53%, respectively. All patients surviving longer than seven months had a successful procedure. Oneyear survival without liver transplantation (LT) was 41.2%. Six patients underwent LT after a mean period of 222 days. A Child-Pugh score <10 was a strong predictor of non-response and early mortality (odds ratio at 7.62). Conclusion: TIPS is an effective treatment for hepatic hydrothorax in selected patients. A poor liver function is a strong predictor of bad outcome and early mortality.

15.2.3

15.2.5

TIPS and local thrombolysis for the treatment of early portal vein thrombosis after liver transplantation P. P. Goffette1, O. Cicarelli2, P. F. Laterre3, X. Wittebolle3, J. P. Lerut2; 1Interventional Radiology, St-luc University Hospital, Brussels, Belgium, 2 Digestive surgery, St-luc University Hospital, Brussels, Belgium, 3 Intensive Care, St-luc University Hospital, Brussels, Belgium.

TIPS creation with the VIATORR stent-graft. Experience with 71 patients K. A. Hausegger1, F. Karnel2, B. Georgieva3, J. Tauss3, H. R. Portugaller3; 1Deparment of Radiology, Federalhospital Klagenfurt, Klagenfurt, Austria, 2Department of Radiology, Kaiser Franz-Josef Spital, Vienna, Austria, 3Department of Radiology, Univ. Hospital Graz, Graz, Austria.

Purpose: Evaluating the efficacy of a combination: TIPS, thrombolysis and PTA to treat early portal thrombosis after liver transplantation (LT). Materials & Methods: Following orthotopic (1) and split (2) LT, an early extended portal thrombosis (delay: four, ten and 18 days) was managed non-operatively in three patients. Symptoms were variceal bleeding (1), ascitis (2) and encephalopathy (n=2). Liver tests were markedly increased in all patients. Small-caliber TIPSs were created to increase the portal output and allow the safe placement of an infusion catheter into the portal vein; r-TPA was infused (2.5 mg/h). Two cases of huge splenorenal shunts were embolized. Results: TIPS was successful in all. At 7-, 12- and 15-hour lysis, the complete declotting of extra-/intra-hepatic portal veins allowed to disclose one anastomotic stenosis and one compression, treated by angioplasty or stenting. One subcapsular hematoma requiring percutaneous drainage and one peritoneal oozing requiring surgery occurred. A full clinical recovery was recorded in two cases (four- and 24-month follow-up). Despite a successful lysis, persistent liver test abnormalities required an early redo-LT in the remaining patient. Conclusion: The combination of TIPS, local lysis and angioplasty may be effective to manage early portal thrombosis, obviating the need for a new graft.

Purpose: To evaluate patency of transjugular intrahepatic portosystemic shunt (TIPS) created with the VIATORR stent-graft. Materials & Methods: This two-center study comprises 71 patients. Ten patients were Child-Pugh class-A liver cirrhosis, 43 were class-B, and 18 were class-C. Indication for TIPS was recurrent bleeding in 26 patients, refractory ascites in 44, and Budd-Chiari syndrome in one. Follow-up included color-coded ultrasound 3-5 days after TIPS, after one, three, six months, and biannually thereafter. Shunt venograms were performed every six months. Patency was evaluated with the KaplanMeyer analysis after six and 12 months. Results: TIPS creation was successful in all patients with an effective portal decompression. One shunt thrombosis was observed four days after TIPS and was easily reopened. Two further occlusions were observed after one and two months, both due a kink in the hepatic vein. Hepatic vein stenosis with incomplete hepatic vein coverage was observed in four patients. Seven shunt revisions were performed without complications. Primary and secondary shunt patencies after six and 12 months were 80 and 78%, respectively. De-novo encephalopathy or a deterioration was observed in 20 patients (28%). Conclusion: TIPS with the VIATORR stent-graft had a high primary patency; rate of hepatic encephalopathy had however markedly increased.

15.2.4

15.2.6

Endovascular treatment of patients with obstructed portal and/or mesenteric veins W. Cwikiel; Department of Radiology, University of Michigan Hospital, Ann Arbor, United States.

Expanded-PTFE covered stent graft vs. conventional bare stent in transjugural intrahepatic portosystemic shunt: prospective evaluation of shunt efficacy and patency M. Bezzi, F. Fanelli, F. M. Salvatori, D. Pepino, G. Marcelli, S. Vagnarelli, P. Rossi, R. Passariello; Department of radiological sciences, University of Rome, Rome, Italy.

Purpose: To evaluate interventional treatment of patients with obstruction of splanchnic veins. Materials & Methods: Eleven patients, three women, 8-79 years old (mean: 35 years), with symptomatic portal hypertension secondary to obstruction of portal and/or mesenteric veins were treated. Symptoms were pain at the gastrointestinal tract (n=5), enlarged spleen and thrombocytopenia (n=5), refractory ascites (n=1), encephalopaty (n=1) and one or more bleedings (n=9). A partial spleen embolization was performed in four patients. Attempt to recanalize occluded veins was done in all patients using a transjugular (n=6), transhepatic (n=8) or transsplenic (n=4) access. Results: All procedures but one were carried out without serious complications and followed by a decrease of symptoms. Intraperitoneal bleeding occurred in two patients and required blood transfusion in one of them. Thrombolysis was performed in two patients. Stent or stentgraft recanalization of occluded veins was possible in four patients. Creation of transjugular intrahepatic portosystemic shunt was necessary in four patients. During the mean observation time of 22 months (3-52 months) additional balloon dilatation and placement of new stents or stent-grafts was necessary in four patients. Conclusion: Interventional procedures are valuable in the majority of patients with symptoms secondary to obstruction of veins in the portal circulation.

Purpose: To assess patency of transjugular intrahepatic portosystemic shunt (TIPS) performed with a stent-graft versus a control group treated with bare-stents. Materials & Methods: Sixty-two patients were treated with e-PTFE covered stent-grafts (VIATORR, W.L. Gore) and 87 with bare stents (Wallstent). Patients were treated and followed according to a standard diagnostic work-up and a prospective surveillance strategy. The groups were matched for sex, age, etiology of liver disease, Child's class, previous variceal bleeding, presence of ascites. Patency was assessed by Doppler-ultrasound and venography. Results: Stent revision was necessary in 11/62 patients (17.7%) with stent-grafts (mean follow-up, 15 months) and in 45/87 patients (51.7%) with Wallstents (mean follow-up, 38.5 months). Secondary reinterventions in the stent-graft group were performed in four cases and in the bare-stent group in 19 cases. Stenoses inside the stent were observed only in the bare-stent group (88%) while stenoses of the hepatic vein were more frequent in the stent-graft group (13.2% vs. 9%). Portal vein stenoses were seen in four patients (6.4 %) treated with the VIATORR. Primary patency at 12 months was 79.3% for the VIATORR and 48.3% for the Wallstent. Conclusion: Expanded-PTFE covered stent-grafts, compared with bare stents, reduce reintervention rate prolonging shunt patency.

15.2.7

15.2.9

TIPS: final results of a multicenter study using the VIATORR stentgraft M. Grosso1, G. Passalacqua2, F. Maglione3, I. Bargellini4, C. Vignali5; 1Radiology, S. Croce Hospital, Cuneo, Italy, 2Radiology, S. Filippo e Nicola Hospital, Avezzano, Italy, 3Radiology, Caldarelli Hospital, Napoli, Italy, 4Diagnostic and Interventional Radiology, University of Pisa, Pisa, Italy, 5Radiology, University of Pisa, Pisa, Italy.

Percutaneous portal hepatic islet cell transplantation for Type I diabetes: technique and clinical follow-up L. Hechelhammer1, T. Pfammatter1, B. Marincek1, R. Lehmann2; 1Diagnostic Radiology, University Hospital, Zürich, Switzerland, 2 Endocrinology, University Hospital, Zürich, Switzerland.

Purpose: To present the final results (12-month follow-up) of an Italian multicentric study of TIPS using the Viatorr stent-graft, a PTFE covered stent-graft with a 2-cm uncovered portion on the portal side. Materials & Methods: The study included 66 patients (23 female; mean age 59). Five were Child’s class A, 41 Child B, and 20 Child C. Indications for TIPS were variceal bleeding in 39 patients, refractory ascites in 25, Budd-Chiari syndrome in two. Fifty-eight de-novo TIPS and eight occluded TIPS revisions were performed. On follow-up, patients underwent color-Doppler US at one week, one, three, and six months, and phlebography at six and 12 months. Results: The procedure was successful in 64/66 cases (97%); in 2 cases with portal vein thrombosis, vein puncture was not feasible. Mortality rate was 13.6% (9/66), with 6.1% (4/66) 30-day mortality. Eleven patients (16.6%) developed encephalopathy. Stenosis occurred in one patient within 30 days and in two cases later suring follow-up, requiring further revision. Conclusion: The use of the Viatorr stent-graft seems to improve primary patency of TIPS; to prevent hepatic vein stenosis, the entire tract from the portal vein to the hepatic ostium of the inferior vena cava must be covered.

Purpose: To report radiologic-interventional aspects and clinical results of portal hepatic islet cell transplantation (PICT). Materials & Methods: A single-center prospective series included 6 patients with insulin-dependent diabetes mellitus (IDDM) underwent transplantation between September 2000 and January 2003 (mean age=50.3 yrs, range 40-65; mean IDDM duration=39.7 yrs, all with normal coagulation parameters). Four patients had prior intraoperative portal islet cell injection during renal transplantation. Fluoroscopyguided percutaneous right transhepatic access (Neff®-Set, 6F, Cook, Bjaeverskov, DK) was followed by islet cell suspension infusion into the right branch. Access coil embolization was performed in 5/11 interventions, liver ultrasound the day after, and immunosuppression according to the Edmonton protocol. Results: All interventions were technically successful with a mean fluoroscopy time of 14.5 min and mean dose area product of 157 Gycm2. Baseline portal pressure increased by 8.1 mmHg (range 1-25) immediately after PICT. Number of PICT per patient: 1, n=3; 2, n=2, 4, n=1. One patient, in whom the tract was not embolized, had a subcapsular hematoma which was treated conservatively. At a mean follow-up of 18 months (range 3–31) 5 patients are no longer insulin dependent. Conclusion: PICT is a low-complication radiological intervention with promising clinical medium-term results.

15.2.8

15.3.1

Percutaneous intrahepatic transplantation of pancreatic human islet cells in type 1 diabetes: technique, complications and therapeutic efficacy M. Venturini1, E. Angeli1, P. Maffi2, P. Fiorina2, A. Secchi2, A. Del Maschio1; 1Radiology, San Raffaele Hospital, Milan, Italy, 2Internal Medicine, San Raffaele Hospital, Milan, Italy.

Antibodies of platelet receptors and reteplase for thrombolysis in peripheral arteries (APART). Phase-II trial G. Tepe, J. Schmehl, T. Dietrich, B. Wiesinger, J. Wiskirchen, M. Bitzer, C. C. Claussen, S. H. Duda; Diagnostic Radiology, University of Tuebingen, Tuebingen, Germany.

Purpose: To evaluate technique, complications, and therapeutic efficacy of percutaneous intrahepatic transplantation of pancreatic human islet cells (PHIC) in patients with long-term type 1 diabetes. Materials & Methods: A suspension of approximately 150 ml containing 300,000-800,000 purified PHIC was percutaneously injected (4F catheter) in the portal vein of 45 diabetic uremic patients (1989-2001), after kidney alone transplantation in 37 cases. The procedure was performed with local anaesthesia, under ultrasonographic and fluoroscopic guidance. PHIC therapeutic efficacy was evaluated based on insulin independence period, Exogenous Insulin Requirement (EIR) (reduction > or o < 1ng/ml was considered to indicate functioning or nonfunctioning PHIC, respectively. Results: The procedure was successful in all cases. Complications, conservatively treated, occurred in three patients (two cases of hemoperitoneum, one of thrombosis). In all cases CPsc > 1ng/dl was observed after PHIC transplantation. These values were maintained in 93%, 86%, and 63% of patients at 1 month, 1, and 5 years, respectively. EIR < 50% of pre-transplant dose were 64%, 67%, and 43% in the same periods. An insulin independence period was achievied in 19 patients (45%). Conclusion: In type 1 diabetes, PHIC transplantation can be considered a safe therapeutic approach, eventually repeatable, alternative to pancreas transplantation.

Purpose: Glycoprotein IIb/IIIa receptor inhibitors enhance thrombolysis in patients with acute peripheral artery occlusions. In a recent trial (PROMPT) the combined therapy abciximab+urokinase showed to be safer and more effective than urokinase alone. This prospective randomised trial (APART) evaluates the safety and efficacy of the combination of abciximab with a newer thrombolytic agent (reteplase) versus the combination with urokinase. Materials & Methods: Seventy patients with lower extremity thrombi were randomised to reteplase+abciximab or to urokinase+abciximab. The fibrinolytic agent was given in a pulsed spray fashion. Results: Currently, 68/70 patients are enrolled in the trial. The success rate did not differ between both groups. The duration of thrombolysis was 216±124 min in the reteplase and 181±61 min in the urokinase group. The total amount or fibrinolytic agent used was 3.42 ± 1.14 U (reteplase) and 644,164 ± 159,671 U (urokinase). Those patients who received the combination with reteplase showed a slightly higher tendency to bleeding complications. Conclusion: Our preliminary conclusion is that a strategy using abciximab as an adjunct to urokinase or reteplase in patients with subacute peripheral artery occlusions was safe in any case and the efficacy of urokinase and reteplase did not differ significantly.

15.3.2

15.3.4

Impact of IA reteplase on fibrinogen levels in the treatment of peripheral arterial disease (RELAX trial) G. Tepe1, K. Ouriel2, J. Smith3, T. McNamara4, F. Casteneda5, S. H. Duda1; 1 Diagnostic Radiology, University of Tuebingen, Tuebingen, Germany, 2 Cleveland Clinic, Cleveland, United States, 3Centorcor, Waaren, United States, 4UCLA Medical Center, Los Angeles, United States, 5St. Francis Hospital, Peoria, Ill, United States.

Ultrasound ablation in chronic lower extremity deep vein thrombosis A. Wang, D. Jing, L. Liu, J. Liu; Vascular Surgery, Hei Lonh Jiang Provincal Hospital, Harbin, China.

Purpose: The RELAX study was a multi-center pilot study to evaluate the safety and efficacy of escalating doses of reteplase with or without abciximab in patients with acute or subacute peripheral arterial occlusive disease. We examined fibrinogen (Fb) levels by dose, over time, from the RELAX trial. Materials & Methods: Patients (6/group) were assigned to intra-arterial reteplase (r-PA) 0.1, 0.2, 0.5 or 1.0 U/h with or without intravenous abciximab (abx) infusion. A dose confirmation phase added patients in selected dosing panels to a total enrollment of 74 patients. Fb levels were measured at baseline and at prespecified angiography (6 ± 2, 20 ± 4 hours). Results: Across all doses, Fb levels showed only minor reductions from baseline (see table). The addition of abciximab to reteplase monotherapy at each r-PA had no significant effect on Fb depletion. Observed differences were not statistically significant by Anova. Values expressed as mean ± SD (N). Six patients had any Fb level <100 mg/dl across all dosing groups. Conclusion: The infusion of intra-arterial reteplase alone or in combination with intravenous abx in acute peripheral arterial occlusive disease appears to minimally effect systemic levels of Fb in this dose finding safety trial.

Purpose: To evaluation the clinical effiency of ultrasound ablation in patients with chronic deep vein thrombosis of the lower extremity. Materials & Methods: In 38 patients with chronic deep vein thrombosis, with total occlusion of the left ilio-femoral vein, an ultrasonic ablation catheter was inserted into the occluded vein segment through a surgical approach . The ablation device consists of a flexible wire 2.2 mm in diameter attached to a piezoelectric crystal generating ultrasound at 24 KHz. Results: A channel was easily created within the occluded ilio-femoral vein and achieved successful recanalization in 31 of 38 patients. Conclusion: Ultrasound ablation is an effective therapeutic modality for patients with chronic ilio-femoral vein occlusion and no addtional endovascular treatment modalities are necessary to obtain a good clinical result

15.3.3

15.3.5

Acceleration of rt-PA thrombolysis in acute and subacute lower limb ischemia using the Trellis™ peripheral infusion system D. Tsetis1, A. Katsamouris2, Z. Androulakakis3, A. Hatzidakis1, E. Karagiannakidis1, T. Kostas2, C. Ioannou2, N. Gourtsoyiannis1; 1Radiology, University Hospital Heraklion, Heraklion, Greece, 2Division of Vascular Surgery, University Hospital Heraklion, Heraklion, Greece, 3Surgery, Venizelion Hospital Heraklion, Heraklion, Greece.

Management of limb ischemia with the Angiojet rheolytic thrombectomy device: four years of experience D. Siablis1, P. Zabakis1, D. Karnabatidis1, T. Kordonouri1, K. Katsanos1, P. Kraniotis1, I. Tsolakis2; 1Department of Radiology, University of Patras, Patras, Greece, 2Department of Vascular Surgery, University of Patras, Patras, Greece.

Purpose: To present our preliminary experience using the Trellis™ Peripheral Infusion System for the treatment of <30 days lower limb ischemia. Materials & Methods: Five men (age: 48-77) with iliac (n=1), femoropopliteal (n=3) and femoropopliteal bypass graft (n=1) arterial thrombosis were treated. The length of the occlusion ranged from 8 to 30 cm. The Trellis catheter was advanced over a 0.035-inch guidewire through an 8-F sheath. The thrombus was isolated between the inflated proximal and distal catheter balloons and saturated with 4-6 mg rt-PA boluses injected through the infusion catheter port. The guidewire was exchanged for the dispersion wire of the specific Oscillation Drive Unit which was activated for 5-15 minutes after each rt-PA bolus. Finally, the proximal balloon was deflated and the macerated clot was aspirated through the sheath. Results: There was 80-100% clot dissolution using a total of 9-14 mg rt-PA. The duration of thrombolysis ranged from 35 to 60 minutes. In two cases, the stents were placed due to underlying stenoses. No complications were observed. Conclusion: The Trellis Peripheral Infusion System isolates the thrombosed arterial segment, delivers and rapidly disperses small amounts of rt-PA into the thrombus and potentially eliminates the continuous infusion of the lytic agent.

Purpose: To evaluate the safety and efficacy of rheolytic thrombectomy with the AngioJet device for the treatment of lower extremity ischemia due to arterial/graft occlusion. Materials & Methods: Forty-six consecutive patients who underwent rheolytic thrombectomy for thrombotic occlusions in native arteries (n=36) or bypass grafts (n=10) were retrospectively studied. In-hospital and post-procedural events were evaluated. Amputations at 30 days were recorded. Results: Eighteen patients had acute lower limb ischemia (estimated age of thrombus up to two weeks) while the remaining 28 presented with subacute thrombosis (>2 weeks- 3 months). An additional postthrombectomy thrombolysis was performed in those patients who presented with subacute thrombosis. Underlying stenoses found in 44 patients were treated with PTA or/and stenting. In-hospital complications included four cases of peripheral emboli; a major amputation was deemed necessary in two of them. Four re-thromboses occurred due to a poor run-off, while the remaining patients had viable limbs at discharge. Amputation rate at 30 days was 4.0%. Conclusion: Rheolytic thrombectomy with the AngioJet device appears to be safe and effective for a rapid thrombectomy; it is a reliable and equivalent therapeutic alternative to surgery, especially for high-risk patients with limb-threatening ischemia.

15.3.6

15.4.2

Mechanical local thrombolysis for the treatment of acute and subacute occlusions of peripheral arteries: a novel approach H. Krankenberg, T. Tübler; Angiology, Center for Cardiology and Vascular Intervention, Hamburg, Germany.

Double aneurysm of the internal carotid artery: a new option for endovascular therapy F. Perona1, A. De Santis2, M. Fornari2, L. Valvassori3, E. Boccardi3; 1Interventional Radiology, Galeazzi Institute, Milan, Italy, 2 Neurosurgery, Galeazzi Institute, Milan, Italy, 3Neuroradiology, Niguarda, Milan, Italy.

Purpose: There is no consensus concerning the treatment of thromboembolic occlusions. We report on a new system combining local lysis with mechanical thrombectomy. Materials & Methods: Between October 2002 and January 2003, seven patients (six men, age 56±5.6) with acute/subacute occlusions of iliac/ femoral arteries were treated. After passage of the wire, a novel Catheter (Trellis™, Bacchus Vascular) was placed. One occlusion-balloon was positioned proximally/distally and a defined dose of r-tPA was injected in between. With an oscillating wire the thrombus was dissolved. Results: Interventions were performed in nine lesions (nine patients): A. iliaca externa (n=1), A. femoralis superficialis (n=3), A. poplitea (n=5). The age of the occlusion was a median of seven days (range 1-100) and the median length was 10 cm (4-35 cm). The occlusions were dissolved once or twice with 4-6 mg r-tPA. In seven patients the former occlusion was post-dilated and in five an additional stent was implanted. No acute complications were observed. At 30-day follow-up two re-occlusions occurred; one could be successfully treated with laser recanalization and stenting. Conclusion: The Trellis Embolectomy System is a safe and effective treatment option for acute and subacute occlusions. A larger cohort has to be enrolled to evaluate long-term efficacy.

Purpose: Endovascular therapy of intracranial aneurysms is a safe and effective method. Despite some experience there is a medical need to maximize this technique in multiple lesions that are frequently not suitable for endovascular occlusion. Neuroform ( Boston Scientific, USA ) offers an advantage in complex aneurysms as a support for difficult necks. Materials & Methods: A 29 year old female with recurrent headache was investigated using RMN angiography. A double aneurysm of the right internal intracranial carotid artery was detected. The indication for endovascular therapy was arguable due to the wide necks of both aneurysms. At the time of diagnosis Neuroform was about to be introduced into radiological practice, and the prompt use of this device was planned. Results: The patient was treated in November 2002 with a Neuroform implant in the intracanial tract of the right internal carotid artery. Both aneurysmal necks were optimized for endovascular therapy, and occlusion was achieved successfully in one session. Short term follow-up confirmed patency of the I.C.A. without signs of procedure-related complications. Conclusion: Treatment of complicated multiple intracranial aneuryms remains a significant challenge, but new devices such as Neuroform allow broadening of the indications with an improvement in therapeutic effectiveness.

15.4.1

15.4.3

Occlusion of large and giant intracranial aneurysms with mixed fibered and non-fibered coils: risk of thromboembolism H. K. Habboub1, a. Akaileh2, F. Radi2, m. Habahbeh3; 1Radiology, King Hussein Medical Center, Amman-Jordan, Jordan, 2Neurosurgery, King Hussein Medical Center, Amman-Jordan, Jordan, 3Neurology, King Hussein Medical Center, Amman-Jordan, Jordan.

Neuropathology of meningiomas post-Embosphere embolotherapy A. O'Connell1, M. Farrell2, J. Thornton1, P. Brennan1; 1Department of Neuroradiology, Beaumont Hospital, Dublin, Ireland, 2Department of Neuropathology, Beaumont Hospital, Dublin, Ireland.

Purpose: To identify the risk of thromboembolism after treatment of large intracranial aneurysms (ICA) with the mixed use of fibered and non fibered coils. Materials & Methods: Between May 2001 and December 2002, 15 large (>15mm) and giant (>25mm) ICA were treated using a mixture of non-fibered and fibered electrically-detachable coils (Dendron GmbH) to pack the aneurysm. An average of 2.75 fibered coils were used in large aneurysms and 3.5 in giant aneurysms. Post coiling, patients were maintained on heparin for 1 week. Results: Total occlusion was obtained in 13 aneurysms (87%). Mean follow-up time was 4.5 months. The incidence of significant thromboembolic manifestations was monitored by a neurologist for 2 weeks following aneurysm occlusion. In no cases did immediate motor or sensory deficit occur in the first 24 hours. In 3 cases transient unilateral weakness suggestive of ischemic changes was observed that resolved completely within 1 week. Conclusion: The use of fibered coils in large and giant ICA is helpful in certain cases to enhance thrombosis within these aneurysms. Thromboembolic phenomena may occur within the first week, but no permanent neurological deficit was encountered.

Purpose: Pre-surgical embolization of intracranial meningiomas reduces the risk of operative hemorrhage. However, embolotherapy leads to tumor necrosis which may cause difficult neuropathological interpretations, particularly in the diagnosis of atypical meningiomas. We evaluated embolized meningiomas for necrosis and features of atypia. Pre-operative imaging was reviewed for tumor necrosis. Materials & Methods: Ten patients with intracranial meningioma and angiographically demonstrated tumor hypervascularity underwent preoperative arterial embolization using tris-acryl gelatin microspheres [Embospheres]. Tumors were evaluated for presence of necrosis and level of mitotic activity using accepted criteria. Results: In 9/10 preoperative images no necrosis was present. Pathology, post-embolization, revealed: necrosis and mitoses (4); necrosis alone (3) and mitosis alone (1) (of eight meningiomas). Two meningiomas showed neither necrosis nor mitosis. In the patient with radiographic tumor necrosis, neuropathology confirmed necrosis and increased mitoses. None of the postoperative images demonstrated residual tumor. One patient had transient visual obscurations postembolization. No other clinical deterioration was evident post-embolotherapy. Conclusion: Although pathologic findings of meningiomas post-embolization must be interpreted with caution, correlation of pathology with preoperative imaging together with strict application of histologic diagnostic criteria should reduce risk of incorrect tumor-grading. Routine post-embolization scanning should be performed to document necrosis and confirm effective embolization.

15.4.4

15.4.6

Vertebral artery injury related to blunt and penetrating trauma: an update S. A. Naik, O. Diaz-Daza, C. J. Whigham, G. P. Hobbs; Radiology, Baylor College of Medicine, Houston, United States.

Vascular distribution of two types of microspheres (Embosphere and Contour SE): a comparative study J. Saint-Maurice1, M. Wassef2, J. Namur1, J. Merland3, A. Laurent3; 1Neuroradiology, Lariboisière hospital, Paris 75010, France, 2Pathology, Lariboisière hospital, Paris 75010, France, 3Neuroradiology, Lariboisière Hospital, Paris 75010, France.

Purpose: To review the incidence and nature of traumatic vertebral artery injuries at a large metropolitan trauma center. Materials & Methods: A retrospective review was performed of all trauma-related cerebral angiograms completed over a 36-month period at a Level I Trauma Center. Results: Two-hundred and seventeen cerebral angiograms were performed for trauma. Mechanisms of trauma were blunt (113) and penetrating (104). Twenty-four vertebral artery injuries were found in 20 patients. Injuries included: occlusion (5), subintimal dissection (8), pseudoaneurysm (2), intimal flap (6), active extravasation (1), and fistula (2). Segmental vertebral artery analysis indicated four injuries at V1 segment, nine injuries at V2 segment, 11 injuries at V3 segment, and no injuries at V4 segment. Nine of the 24 vertebral artery injuries (38%) were treated with endovascular therapy. Four patients suffered strokes prior to angiography. All patients had associated comorbid injuries. Conclusion: Vertebral artery injury is an uncommon sequela of trauma and may result in significant neurological impairment. The presentation of vertebral artery injury is variable with respect to both type and location of injury. As these injuries may be clinically silent initially, a high index of suspicion must be maintained in patients presenting with sufficient mechanism.

Purpose: To compare the distribution, in kidney vasculature, of two types of microspheres (MS): Embosphere (ES - Biosphere Medical) and Contour SE (CSE - Boston Scientific), in three size ranges: A=500-700, B=700-900 and C=900-1200 micrometers. Materials & Methods: Six sheep were anesthetized and randomization of MS type and size for the 12 kidneys was performed. Embolization at the renal artery ostium of 0.2 ml MS sediment was carried out. Dilutions and injection times differed with sizes: A: 1 ml/30 sec; B: 5 ml/90 sec; C: 5 ml/180 sec. Sheep were sacrificed 24 hours later and histology of kidneys was performed; 284 vessels were examined for: diameter of occluded vessels, number of MS per vessel, MS length and width, MS location in five classes from proximal arteries to distal capillaries. Results: MS deformation was +50.9% for CSE and +22.4% for ES(p<.0001). The size of occluded vessels increased with the size range for both MS types (p<.0001 and p<.0001). CSE penetrate deeper in the vasculature than E for all three sizes (p<.0001) as well as for each size range: A: p<.0001, B: p<.0001, and C: p=.0127. Conclusion: In sheep kidney vasculature, CSE block more distally than ES since their deformation is higher.

15.4.5

15.4.7

Endovascular treatment of dural type-D carotid cavernous sinus fistulae and long-term clinical outcome I. E. Ioannidis1, G. Orfanidis2, K. Papapetrou1, A. Mitsos2, M. Psomas3, A. Andreou1; 1Neurosurgery, Hygeia, Athens, Greece, 2Neurosurgery, G. Gennimatas, Athens, Greece, 3Anesthesiology, Hygeia, Athens, Greece.

Local intraarterial fibrinolysis for acute central retinal artery occlusions: initial experience D. M. Gor1, J. Farkas1, R. Turbin2, K. Shah1, S. Gujrathi1, V. Sharma1; 1Radiology, New Jersey Medical School, Newark, United States, 2Ophthalmology, New Jersey Medical School, Newark, United States.

Purpose: Dural carotid-cavernous fistulae are arteriovenous shunts between meningeal branches of external (ECA) or internal carotid arteries (ICA) and the cavernous sinus (CS). Chemosis, eyelid swelling, proptosis, diplopia and glaucoma with impaired vision may all result from these fistulas. We report the method and the results of treatment in 15 patients with type-D dural carotid cavernous sinus fistulae (CCF). Materials & Methods: All the patients had type-D CCFs. Two patients underwent transvenous embolization by introduction of coils into the CS via the inferior petrosal sinus. Thirteen patients underwent transarterial particle embolization of dural branches feeding the fistula from one or both ECAs. In five patients in whom a significant supply from ICA branches to the fistula was evident, particle embolization of ICA branches was also performed. In one patient a second particle embolization of ECA's feeding branches was performed six months later. Results: In one patient a diminished vision of the left eye was evident immediately after embolization. A clinical improvement was achieved in all patients. None of them developed recurrent symptoms and signs suggesting a recurrence of the fistula during the follow-up period (3-37 months). Conclusion: Transvenous or transarterial endovascular techniques can be used successfully to occlude dural CCFs.

Purpose: Traditional therapies for acute central retinal artery occlusion (CRAO) are largely unsuccessful. We emphasize local intra-arterial fibrinolysis (LIF) with the selected infusion of recombinant tissue plasminogen activator (r-TPA) directly into the ophthalmic artery as a safe and effective treatment by presenting two cases of CRAO treated with LIF. Materials & Methods: A retrospective review of three consecutive cases is presented. Superselective catheterization of the ophthalmic artery was performed with a 0.10-inch microcatheter (Prowler 10, Cordis). r-TPA was reconstituted to 0.5mg per milliliter and slowly infused into the ophthalmic artery. The procedure was halted when the patient symptoms improved or a maximum of 16mg were infused. Results: One treatment was aborted prior to the superselective catheterization, due to severe narrowing of the internal carotid. The CRAO was determined to be less than 18 hours in duration in each of the three cases, and the two successfully treated patients realized significant visual improvement after receiving 16 and 10mg of r-TPA, respectively (hand motions improving to 20/640 and light perception improving to 20/400). There was fundoscopic evidence of restored retinal arterial perfusion. Conclusion: LIF may be an effective method to treat CRAO if the diagnosis is made early and fibrinolysis performed promptly.

15.5.1

15.5.3

Superselective embolization in post-traumatic priapism with Glubran 2 acrylic glue C. A. Reale, R. Gandini, A. Spinelli, D. Konda, G. Pendenza, G. Simonetti; Diagnostica per Immagini, Policlinico Tor Vergata, Rome, Italy.

Transvaginal aspiration of ovarian cysts: long-term follow-up D. M. Duke1, J. Colville1, D. Broe2, T. Fotheringham1, M. J. Lee1; 1Dept. of Academic Radiology, Beaumont hospital, Dublin, Ireland, 2 Dept. of Surgery, Beaumont hospital, Dublin, Ireland.

Purpose: The aim of this study was to evaluate efficacy and feasibility of selective transcatheter embolization using the Glubran 2 acrylic glue in the treatment of post-traumatic priapism. Materials & Methods: Two patients (24- and 26-year-old) with posttraumatic priapism underwent transcatheter embolization with 4-mm straight-fibered microcoils (Target; Boston Scientific Co.), resulting in temporary penile detumescence and apparent resolution of the arterovenous fistula (AVF). In both cases, priapism recurred 24 hours after and was treated through selective transcatheter embolization of the nidus using the Glubran 2 acrylic glue (GEM Srl, Viareggio, Italy) mixed with Lipiodol in a 1:1 ratio. Results: Both patients developed a gradual penile detumescence with complete resolution of priapism within three days from the procedure. Ultrasonographic controls, performed post-procedurally at days one, five, 30 and after three months in one case and six months in the other, confirmed the exclusion of the AVF. The patients showed the complete recovery of their sexual activity within 45 days from procedure. Conclusion: Selective transcatheter embolization using acrylic glues, and especially Glubran 2, in the treatment of post-traumatic priapism is safe, feasible and efficient. This acrylic glue may be considered for the primary embolization of AVFs in the management of this disease.

Purpose: To establish the benefit of transvaginal drainage of ovarian cysts with particular reference to complications, patient tolerance, and recurrence rate. Materials & Methods: All transvaginal drainages of ovarian cysts between 1996 and 2002 were analysed by review of medical records. All patients, age range 21-85 (mean 44.9) had either previous pelvic surgery or obesity precluding laparoscopic intervention. Complications, recurrence rates, patient tolerance and need for subsequent surgical intervention were recorded. Results: Twenty-four patients underwent 34 transvaginal drainage procedures of ovarian cysts. The presenting symptoms included lower abdominal pain (16), abdominal swelling (1), unrecorded (5), incidental finding (2). Two drainage attempts failed due to poor visualisation. The amount of fluid drained was 10-400cc (mean 97cc). One patient developed vaginal spotting post-procedure while another complained that the procedure was embarrassing. Cytology was available in 25/32 with no malignant cells identified; 14/24 patients had recurrence (64%). Seven patients underwent repeat aspiration(s) and 3/7 had cyst sclerosis performed; 4/22 patients had subsequent gynaecological surgery. Conclusion: Transvaginal aspiration of ovarian cysts is simple, safe and useful and may be proposed as an alternative treatment when surgery is undesired or contraindicated. Our long-term follow-up indicates that recurrence rates are higher than previously reported.

15.5.2

15.5.4

Percutaneous nephrostomy: special indications and technique in newborns and infants W. Hruby1, M. Urban1, E. Frigo2, A. Rokitansky2; 1Radiology, Danube hospital, Vienna, Austria, 2Pediatric Surgery, Danube hospital, Vienna, Austria.

Transabdominal ultrasound-guided cystoscopic bladder tumour resection and retrograde ureteric stent insertion D. Fay1, C. Nice1, T. Hasan2, P. J. Haslam1; 1X-Ray Department, Freeman Hospital, Newcastle upon Tyne, United Kingdom, 2Department of Urology, Freeman Hospital, Newcastle upon Tyne, United Kingdom.

Purpose: Urinary tract obstruction caused by intrinsic or extrinsic ureteral ostruction or congenital disorders of the urinary tract in newborns or infants are rare indications for preoperative percutaneous nephrostomy (PcN). Patient selection, indications and technique of PcN with consideration of interdisciplinary management are evaluated. Materials & Methods: From January 1999 to January 2003 292 fetuses were diagnosed with congenital disorders in our prenatal screening program. Forty-seven (17%) had disorders of the urinary tract. Sixteen inborn disorders were diagnosed postnatally. Six infants underwent preoperative PcN. Technique: Our long standing experience in PcN has led to a standardized access and puncture technique and development of an "Infant-Newborn Puncture Set" which was used in all cases. USguided puncture was performed under dose-reduced fluoroscopy control with a one-stitch procedure. Results: Puncture and catheter placement were successful in all cases. There were no peri- or postprocedural complications. One drain dislocation was observed, but the drain was easily replaced. PcN was left in place until the childrens' weight was more than 9 kg. Conclusion: PcN in newborns and infants with obstructive congenital urinary tract disorders is a safe and reliable preoperative method to preserve renal function and parenchyma

Purpose: To describe the use of transabdominal ultrasound to guide resection of bladder tumour occluding ureteric orifice and to aid retrograde ureteric stent insertion. Materials & Methods: Three patients (two men, one woman, ages 6581) presented to our department with ureteric obstruction secondary to bladder carcinoma. In all three patients the ureteric orifice was obscured by tumour at cystoscopy. Retrograde ureteric access was gained using portable transabdominal ultrasound (Sonosite 180) in the operating theatre to localise the ureteric orifice and guide diathermy resection. Guidewire passage into the distal ureter was then performed under ultrasound visualisation and retrograde stents inserted, avoiding the need for percutaneous nephrostomy. Results: In all three patients ureteric access was successful and retrograde stenting was performed. Conclusion: Transabdominal ultrasound guidance is a useful technique to aid retrograde ureteric access when the ureteric orifice is obscured and can avoid the need for nephrostomy.

15.5.5

15.5.7

Percutaneous balloon dilatation for the treatment of early and late ureteral strictures after renal transplantation: long term follow-up A. Belenky1, G. Bartal2, E. Atar1, N. Goldberg1, E. Mor3, G. N. Bachar1; 1Radiology, Rabin Medical Center, Petah Tiqva, Israel, 2Radiology, HYMC, Hadera, Israel, 3Transplantation, Rabin Medical Center, Petah Tiqva, Israel.

The long-term results of single-session percutaneous drainage and ethanol sclerotherapy in simple renal cysts D. Akýncý, B. Gümüþ, M. Haliloðlu, B. Çil, M. Özmen, O. Akhan; Department of Radiology, Hacettepe University School of Medicine, Ankara, Turkey.

Purpose: To report our experience with percutaneous balloon dilatation (PBD) for the treatment of ureteral strictures in renal allografts. Materials & Methods: We reviewed the records of 422 consecutive renal transplant patients. In 21/422 patients (4.9 %) ureteral stricture was diagnosed using sonography and scintigraphy. This study sample consisted of 15 men and six women aged 16 to 67 years. Each of these patients underwent 2-4 PBD's at 7 to 10-day inter-session intervals. Clinical success was defined as resolution of the stenosis and of hydronephrosis, confirmed by sequential ultrasound and normalization of creatinine levels. Results: PBD was successful in 13 of 21 patients (62%). There was no statistically significant difference in the success rate of ureteral dilatation between the early and the late groups (seven of 12 early stenoses (58.4%) and six of nine late stenoses (66%)). No major complications occurred. Conclusion: PBD of the ureteral strictures is a safe and simple interventional procedure with low morbidity. We advocate PBD as an initial minimally invasive treatment in renal transplant patients with early and late ureteral strictures.

Purpose: To demonstrate the efficacy and long-term results of singlesession ethanol sclerotherapy in management of benign renal cysts. Materials & Methods: Eighty patients, with a mean age of 53.2 year, were studied. Indications were flank pain in 64 patients, hypertension in five patients, and hydronephrosis in 11 patients. The mean follow-up period was 23.3 months. The procedure was performed with the guidance of fluoroscopy and ultrasonography using a 5-6 Fr pigtail catheter. After the cystogram, 95% ethanol to a volume of 40% of the cyst volume was used. The cavity was then emptied and the catheter withdrawn. Results: No major complications were detected during the procedure or follow-up. Mean volume reduction was 95.46 % at the end of the first year, and the cyst disappeared completely in 11 (13.75 %) patients. Following the procedure 51 patients (79.6 %) were free of pain, normotension was obtained in four of the five (80 %) hypertensive patients, and hydronephrosis resolved in nine of 11 patients (81.8 %). A second intervention was required in two patients. Conclusion: Percutaneous treatment of renal cysts with single-session sclerotherapy is a safe, effective and minimally invasive method.

15.5.6

15.6.1

Percutaneous management of ureteral injuries after Cesarian sections B. Üstünsöz, &. Ugurel, M. Kocaoglu, F. Örs, &. Somuncu; Radiology, GATA, EtlÝk-Ankara, Turkey.

Radiofrequency ablation: tissue modulation in an experimental liver perfusion model A. Lubienski, L. Wirth-Jaworski, T. Hahn, R. Bitsch, K. Lubienski, C. Dechow, G. W. Kauffmann, M. Düx; Diagnostic Radiology, University Hospital Heidelberg, Heidelberg, Germany.

Purpose: To discuss the role of percutaneous nephrostomy with/without antegrade ureteral stenting in patients with ureteral injury after a Cesarian section. Materials & Methods: During the last six years, 15 women (average age 29 years), six with a complete block due to ureteral suturing, six with an incomplete ureteral rupture, and three with a complete ureteral rupture, were referred to the care of interventional radiologists. All the patients underwent percutaneous nephrostomy. In those six cases with an incomplete ureteral rupture antegrade ureteral stents with an average length of 24 cm were also implanted (Double J, Cook, USA). Ultrasonographic and fluoroscopic guidance was used in 14 cases. In one patient only, CT guidance was needed for the inital access while fluoroscopy guided the rest of the procedure. Results: The six patients with an incomplete ureteral rupture plus one case with a fistula between the ureter and the uterus were managed by interventional radiologists only. Percutaneous nephrostomies required an elective surgical correction. Conclusion: Percutaneous nephrostomies and antegrade ureteral stent placements for incomplete ureteral injuries may be resolved without surgical correction. Percutaneous nephrostomies for a complete block or for ureteral ruptures require elective surgery.

Purpose: Evaluation of different liquids for tissue modulation during radiofrequency ablation in order to improve necrosis size. Materials & Methods: Perfusion of fresh bovine livers was simulated in a liver perfusion tank developed for the present experiments. Swivels mounted on the liver perfusion tank allowed the simultaneous reproducible generation of several RFA lesions of standardized depths and angles with respect to the liver surface. A 2cm LeVeen needle and the RF 3000 generator (Boston Scientific) were used for RFA. Additional puncture needles allowed the independent injection of liquids. For tissue modulation saline (0.9% and 10%), ethanol (98%), acetic acid (50%) and Lipiodol were used. The size of the thermolesions was assessed. Results: Two hundred seventy thermolesions were generated. Mean size was 3.9cm2 (RFA only), 4.3cm2 (saline 0.9%), 4.7cm2 (saline 10%), 4.7cm2 (ethanol), 5.4cm2 (acetic acid) and 4.2cm2 (Lipiodol). There was no significant difference in thermolesion size between the different liquids. (Friedmann test p=0.526). Conclusion: Although tissue modulation during RFA with acetic acid seemed to create the largest thermolesions, there was no significant difference in lesion size between the evaluated liquids.

TH I W

N W A DR

15.6.2

15.6.4

A canine model of endoleak after AAA endograft with preservation of a large aneurysmal sac and patent lumbar vessels D. Pavcnik1, R. T. Andrews1, Q. Yin2, B. T. Uchida1, H. A. Timmermans1, N. Toyota1, M. Nakata1, J. A. Kaufman1, F. S. Keller1, J. Rösch1; 1 Dotter Interventional Institute, OHSU, Portland, OR, United States, 2 Section of Liver and PancreasTransplantation, OHSU, Portland, OR, United States.

Development and radiofrequency ablation of VX-II carcinomas in a rabbit uterus model: evaluation with histopathology B. Kang; Radiology, Mokdong Hospital, Seoul, Republic of Korea.

Purpose: To create an animal model of endoleak after AAA stentgraft placement. Materials & Methods: AAAs were created in nine dogs by anastomosis of an isolated segment of the IVC to the aorta. Animals underwent implantation of PTFE-covered endografts having 3-mm diameter holes through the fabric. Aortograms were obtained before and after surgery, following endograft placement, and at the time of animal sacrifice at one week or one, two, three or six months. Results: Following endograft placement, nine animals had artificial type-III endoleaks. One aneurysm ruptured at one week. At completion of the study, six endografts were patent and two were occluded. The aneurysm sac had enlarged by roughly 50% in seven animals. Type-I endoleaks were present in three animals, type II-in three, and an artificial type-III endoleak was present in six animals with patent endografts. The aneurysmal sac exhibited early thrombus formation at one week, which progressed to complete thrombosis from one to six months. Conclusion: The model is feasible but would only be useful in testing occlusive techniques for residual aneurysm sacs in the acute phase after endograft placement. It would be not reliable for chronic evaluation because of progressive thrombosis in most aneurysm sacs.

Purpose: To study experimental rabbit uterine tumor models with VX-II carcinomas and to evaluate our experience with uterine tumor ablation with percutaneous radiofrequency (RF). Materials & Methods: Nine rabbits with variable-sized uterine tumors underwent RF treatment. Immediately after having sacrificed them with CO2 inhalations, we opened their lower abdomen and RF ablation of the tumors was performed. After the procedure, uterine tumors were collected and preserved in a formalin solution. The effect of RF ablation on the tumors was evaluated by histopathology. Results: The tumors measured from 0.3x0.4 cm to 4x4.5 cm. An extensive necrosis was noted but intact viable nests of tumor cells were recorded in two rabbits. Minimal to extensive apoptosis was present; four of the nine tumors showed an extensive apoptosis after RF ablation. Conclusion: Histopathologic findings showed foci of intact tumor cells after RF ablation. Controlled studies with longer follow-ups without sacrificing the animals are necessary to better understand these findings.

15.6.3

15.6.5

A newly designed balloon-sheath for gastrointestinal interventions: preliminary experimental study in a phantom X. He1, W. Choi2, J. Lim2, T. Kim2, H. Park2, C. Yoon2, J. Shin2, H. Song2; 1Radiology, Nanjing First Hospital, Nanjing, China, 2Radiology, Asan Medical Center, Seoul, Republic of Korea.

Influence of paclitaxel-eluting covered stents on granulation tissue formation in canine urethra models J. Shin1, C. Yoon1, S. Yuk2, J. Kim3, T. Kim1, J. Seo1, H. Song1; 1Radiology, Asan Medical Center, Seoul, Republic of Korea, 2Polymer science and engineering, Hannam University, Daejeon, Republic of Korea, 3Pathology, Asan Medical Center, Seoul, Republic of Korea.

Purpose: To evaluate the performance of a newly designed balloonsheath for gastroduodenal interventions. Materials & Methods: This balloon-sheath consists of an introducer sheath and a supporting balloon. It was divided into three segments: distal bended, balloon, and proximal straight. To evaluate its supportability, we assessed the positions of the coil catheter tip while advancing in a silicon gastroduodenal phantom, with/without the balloonsheath, in different situations. Experiments were performed in two minutes and repeated ten times in each situation. Results: The tip of the coil catheter reached the first (n=3) or the second (n=7) part of the duodenum without the balloon-sheath. It reached the third (n=8) or the fourth (n=2) part of the duodenum through a deflated balloon-sheath. It reached the fourth part of the duodenum (n=8) or the proximal jejunum (n=2) through the sheath with an inflated balloon near the gastric fundus. It reached the fourth part of the duodenum (n=1) or the proximal jejunum (n=9) through the sheath with an inflated balloon near the pylorus. Statistical analysis showed a significant difference between every group (p < 0.001). Conclusion: This balloon-sheath seems to be a valid tool in gastroduodenal interventions, especially the sheath with an inflated balloon near the pylorus.

Purpose: To evaluate the influence of paclitaxel-eluting covered stents on granulation tissue proliferation in canine urethral models. Materials & Methods: Twenty paclitaxel-eluting stents (drug stent, DS) and 20 control stents (CS) were alternatively placed in the proximal and distal parts of each urethra in 20 male mongrel dogs. The amount of paclitaxel was 1800 µg for each DS. Dogs were sacrificed four (n=10) and eight weeks (n=10) after placement. Microscopic findings (number of epithelial layers, granulation tissue thickness, papillary projection thickness, and inflammatory cell infiltration degree) were obtained for each stent. Results: Follow-up urethrographies showed partial proximal migration of one DS. In the urethral proximal part of the eight-week group papillary projection thickness was less in DS than in CS (p<0.05). Inflammatory cell infiltration, granulation tissue thickness, and papillary projection thickness were less in the proximal than in the distal part of the urethra in the DS of the four-week group, in the CS of the fourweek group, and in the DS of the eight-week group, (p<0.05). Conclusion: The new paclitaxel-eluting covered stent does not seem to exhert much effect in preventing granulation tissue. Urethral location (proximal rather than distal) seems instead to induce less granulation tissue around the stent.

15.6.6

35.1.1

Extarterectomy for tumors invading the carotid artery: an animal model I. Oran1, C. Bilgen2, K. Oner3, T. Kirazli2, A. Memis1; 1Radiology, Ege University Medical School, Izmir, Turkey, 2 Otorhinolaryngology, Ege University Medical School, Izmir, Turkey, 3 Neurosurgery, Ege University Medical School, Izmir, Turkey.

Complications related to stent-graft placement in aortic dissection S. Suh, J. Y. Won, D. Y. Lee; Diagnostic Radiology, Research Institute of Radiological Science, Medical College, Yonsei University, Seoul, Republic of Korea.

Purpose: Tumors of the head and neck invading the carotid artery pose a management challenge to the surgeon.Carotid artery sacrifice, peeling the tumor, resection with carotid artery reconstruction are the proposed surgical treatment options. The aim of this study is 1. to propose carotid extarterectomy (following carotid stenting) as an alternative to current techniques, which allows tumor resection without distrupting the blood flow and 2. to validate this technique in an animal model. Materials & Methods: Eight bare (Wallstent) and two covered (Wallgraft) self-expanding stents were bilaterally implanted using the Seldinger technique in the carotid arteries of five male sheeps. Results: Six weeks later, the wall of the carotid artery surrounding the implanted stent was peeled away with a microsurgical technique without complications. The neoendothelium, which had formed within the stent, prevented arterial bleeding and maintained the blood flow. Angiographic controls performed after six months revealed a normal blood flow within the carotid arteries without any endothelial irregularity. Conclusion: With this technique aggressive tumor resections without interrupting the blood flow in the carotid artery may be possible. In addition, it prevents the risk of cerebral ischemia, which is always a problem with other techniques.

Purpose: To report about the complications related to stent-graft placement in patients with aortic dissection. Materials & Methods: Between July 1994 and June 2002, 47 patients with aortic dissection (five Stanford type-A and 42 type-B dissections) underwent stent-graft placement. Nine were acute and 38 subacute/ chronic dissections. Patients were followed for 6-102 months. Their clinical condition was monitored and a CT scan was taken within one month from the procedure and at 3- to 12-month intervals afterwards. We reviewed every complication during the procedure and the followup period. Results: During the procedure we recorded two stent-graft migrations, two stent-graft foldings due to oversize of the stent-graft, one stentgraft torsion, one persistent thoracic false lumen flow by the proximal new intimal tear which had not been identified preoperatively, and 11 cases of endoleaks immediatly after the procedure. On follow-up, we observed 28 patients (60%) with post-implantation syndrome, five persistent endoleaks, three dilated abdominal false lumina, one transient cerebral ischemia, and five pseudoaneurysms at the end of the stentgraft. Four patients (8.5%) were converted to surgery without any mortality. Conclusion: Various complications accompany stent-graft placement of aortic dissections. The stent-graft selection, a careful execution of the procedure and a close follow-up are mandatory.

15.6.7

35.1.2

MR-detectable embolization microspheres. An experimental study R. Chapot1, P. Reb2, A. Schwarz3, M. Wassef4, A. Laurent1; 1Neuroradiology, Lariboisière hospital, Paris 75010, France, 2R&d, Biosphere Medical, Louvres, France, 3R&d, Biosphere Medical, Rockland, United States, 4Pathology, Lariboisière hospital, Paris 75010, France.

Endovascular management of thoracic aortic disease M. Lojik1, J. Raupach1, A. Krajina1, J. Dominik2, V. Brzek2; 1Department of Radiology, University Hospital, Hradec Kralove, Czech Republic, 2Department of Cardiosurgery, University Hospital, Hradec Kralove, Czech Republic.

Purpose: To mark embolization microspheres (MS) to make them detectable in tissues with MR. Materials & Methods: Trisacryl MS were manufactured by a reverse emulsion process. A MR marker was incorporated to the emulsion at five different concentrations before polymerization. MS were washed, sieved to obtain the size ranges commonly used in embolization, and then sterilized. Different studies were performed with a 1.5-T MR scan with clinical sequences: 1. MR examination of MS suspended in saline and supernatant before and after ageing (70°C and 120°C) to test potential release, 2. MR signal change evaluation of MS dispersed in a gel, according to the marker concentrations and sizes, 3. MR examination of different embolized organs in sheep and histological analysis. Results: The general manufacturing process was applied successfully. MS obtained presented a signal decrease that was related to marker concentration. MR detectability of MS in vitro and in tissues was depending on three factors: marker concentration, MS size, and MS amount in the occluded vessels. Conclusion: Marking trisacryl MS for MR detection is possible and effective in vitro and in vivo. Clinical evaluation of its interest in the management of embolized patients can be proposed.

Purpose: To report medium-term results with endovascular stent-graft management of thoracic aorta disease. Materials & Methods: From April 1996 to September 2002 we treated twelve patients with thoracic stent-graft placement. Three patients had thoracic aortic aneurysms, eight patients had posttraumatic rupture of the aortic isthmus and one was treated due to unstable B-type dissection. A total of thirteen stent-grafts and one bare stent were deployed. Results: Two patients had died at 30 days (16.7 %). The first patient died due to collapse of a rigid stent-graft. The second patient died two hours after successful stent-graft placement during abdominal surgery. Paraparesis appeared in one patient after treatment of posttraumatic transection. The patient with unstable B-type dissection had ischemic chest pain the third day after treatment, and CT surprisingly revealed a retrograde dissection of the ascending aorta. In spite of conservative therapy spontaneous thrombosis occurred in the false lumen in the ascending and descending aorta. Regular CT follow-up revealed one endoleak, successfully managed with another stent-graft placement. Conclusion: Further improvement of stent-grafting equipment is mandatory for better treatment results. Non-surgical reconstruction of the thoracic aorta must be subjected to a randomized long-term study before it can be considered as routine practice.

35.1.3

35.1.5

Endovascular reconstruction of type-B dissections and aneurysms of the thoracic aorta using Talent endoprostheses H. G. Bertoni, R. D. Garcia Monaco, M. Fava, D. Bracco, V. Cesareo, D. Cassina; Interventional Radiology, Hospital Italiano, Buenos Aires, Argentina.

Medium term follow-up in patients treated for acute dissections of Stanford type B by means of endovascular repair B. V. Czermak1, A. Mallouhi1, R. Perkmann2, I. E. Steingruber1, T. Jung3, G. Fraedrich2, W. R. Jaschke1; 1 Radiology, University Hospital of Innsbruck, Innsbruck, Austria, 2 Vascular Surgery, University Hospital of Innsbruck, Innsbruck, Austria, 3 Radiology, Waid Hospital, Zürich, Switzerland.

Purpose: To evaluate immediate and short-term safety and effectiveness of Talent endovascular prosthesis placement in type B thoracic aortic dissection and aneurysms. Materials & Methods: Between September 1999 and December 2002, 36 patients with a mean age of 60.2 years ( range: 19-78 years) underwent treatment with a diagnosis of thoracic aneurysm (n=10), type B dissection (n=25), and penetrating ulcer (n=1). The Talent devices were placed via access from the femoral artery, except for one patient who recquired iliac access using a retroperitoneal approach, and were released via endovascular approach. Mean follow-up was 12 months and was performed in all patients using spiral CT. Results: Placement of the endoprostheses was technically satisfactory in all patients. Inmediate exclusion of the dissection and thrombosis of the aneurysm sac was demonstrated in 32 patients. There were no conversions and no cases of paraplegia. Complications included stroke (n=1), GI bleeding and multiorganic failure that led to death (n=2). Conclusion: Thoracic endovascular grafting with Talent endoprostheses is a safe and effective procedure. Early data suggests an endoluminal technique may be preferable to classical graft replacement.

Purpose: To evaluate medium-term outcome of transluminal stentgraft placement in Stanford type B dissections using postoperative lumen volume and thrombus length measurements. Material & Methods: In 13 consecutive patients who completed at least a 12 months follow-up volumetric analysis of total true lumen volume (TTLV), total false lumen volume (TFLV), true lumen volume and false lumen volume within (STLV and SFLV) and distal to (DTLV and DFLV) the stent-graft was performed. Clot formation in the false lumen was evaluated. Results: Over time a significant increase of TTLV (p<0.01) and STLV (p<0.01) was measured. SFVL showed a statistically significant decrease (p=0.02). However, DFLV distal to the stent-graft showed an insignificant volume increase over time (p=0.737). Aneurysmal degeneration of the aorta distal to the graft was observed in two patients. Extension of clot formation in the false lumen differed substantially when comparing the patients to each other. Conclusions: Endovascular Stanford type B dissection repair results in significant increase of STLV and significant decrease of SFLV along the stent-graft over time, which indicates an optimal outcome limited to this area. However, distal to the graft volume increase is observed over time. Clot formation in the false lumen is unpredictable.

35.1.4

35.1.6

Stanford type-B aortic dissections: conservative versus interventional/surgical management B. V. Czermak1, A. Mallouhi1, I. E. Steingruber1, R. Perkmann2, M. Rieger1, T. Jung3, G. Fraedrich2, W. R. Jaschke1; 1Radiology, University Hospital of Innsbruck, Innsbruck, Austria, 2Vascular Surgery, University Hospital of Innsbruck, Innsbruck, Austria, 3Radiology, Waid Hospital, Zürich, Switzerland.

Endovascular stent-graft exclusion for Stanford Type B aortic dissection Z. P. Jing; Vascular Surgery, Changhai Teaching Hospital, 2nd Medical University of Chinese Army, Shanghai, China.

Purpose: To evaluate the outcome of patients treated for Stanford typeB aortic dissections by different therapeutic strategies. Materials & Methods: From January 1996, 43 patients with Stanford type-B aortic dissection were treated in our hospital by endovascular management (n=22), major surgery (n=6), surgical fenestration (n=4), or conservatively (n=11). Results: In those patients treated surgically, the mortality during follow-up was 50% (perioperative mortality: 40%); in those treated by endovascular repair it was 15% (perioperative mortality: 10%), and in those treated conservatively the mortality was 25%. Whereas the false lumen volume (FLV) increased significantly among those cases treated conservatively (p<0.05), a significant decrease was observed in the endovascular group (p<0.01). The total aortic volume (TAV) increased significantly in the conservative group (p<0.05) and non significantly in the endovascular group (0.125). However, in the surgical and the endovascular groups, the volume increase was observed distally to the prosthesis. Severe clinical complications were recorded in all groups. Conclusion: TAV and FLV increases were more obvious in those patients treated conservatively or by fenestration. However, in those cases treated by major surgery or endovascular stent-graft insertion, TAV or FLV increases were observed distally to the prosthesis.

Purpose: To evaluate the feasibility of endovascular exclusion for Stanford type B dissections. Materials & Methods: One hundred and forty-two cases of Stanford type B aortic dissections underwent intravascular stent-graft exclusion. The stent-grafts were deployed to seal the tears and exclude the false lumen. Results: The deploying procedures succeeded in all but one case, who required conversion to open surgery. Complete sealing of the tears was achieved in 125 cases after deplying one stent-graft (88.03%). Immediately proximal endoleaks occurred in 16 cases (11.27%) and distal reflux in two cases (1.41%) from just below the stent-graft. In 14 cases (9.86% ), the reflux originated from distal tears. Additional stent-grafts were deployed in 14 cases with proximal leakage and in 12 of the distal tears. After the additional procedure, the sealing ratio rose to 95.77%. The peri-operative mortality ratio was 4.23%. Conclusion: Endovascular exclusion is a safe and long'lasting treatment for patients with thoracic aortic dissection. Endoleak may result in continued aneurysmal expansion and rupture. Additional endoprosthesis is highly effective for sealing leakage. For multiple tears, additional deployment of stent-grafts to seal all the tears may be the best management.

35.1.7

35.2.1

The CH endograft for one-step treatment of co-existent ascending aortic, aortic arch and descending aortic pathology A. Chavan1, M. Karck2, C. Hagl2, S. Baus1, M. Galanski1, A. Haverich2; 1Department of Diagnostic Radiology, Hannover Medical School, Hannover, Germany, 2Department of Thoracic and Cardiovascular Surgery, Hannover Medical School, Hannover, Germany.

SIR/CIRREF FIBROID Registry - procedural and short-term outcomes R. L. Worthington-Kirsch1, s. FIBROID Registry Steering Committee2; 1Image Guided Surgery Associates, Philadelphia, United States, 2Society of Interventional Radiology, Fairfax, VA, United States.

Purpose: A two-step approach is considered necessary for treating coexistent pathologies involving the proximal and the descending thoracic aorta. We report a prototype stent graft which enables such treatment during a single operation. Materials & Methods: The CH endograft consists of a vascular prosthesis with stainless steel stents affixed distally. Under total circulatory arrest, the CH endograft was deployed antegradely in the descending aorta in 12 patients. The proximal non-stented Dacron graft was used to reconstruct the aortic arch or was anastomosed to a proximal aortic graft (follow-up: 1-15 months). Results: Seven patients had chronic type A dissections, whereas five required ascending aortic surgery and had co-existent arch or descending aortic aneurysms/dissections. Endografting was successful in 11 of 12 patients. Transient neurological deficit and death occured in one patient each. Recovery was uneventful in the rest. All aneurysms were successfully excluded. Aortic dissections showed partial or complete false lumen thrombosis up to the level of the stents. Conclusion: The CH endograft enables treatment of pathologies of the aortic arch and descending aorta during surgery on the ascending aorta. It can reduce the multiple thoracotomies required, especially for type A aortic dissections, to a one-step procedure.

Purpose: The FIBROID Registry was developed to collect demographic, procedural, patient safety, and effectiveness data for uterine artery embolization (UAE) for fibroids. This presentation reviews initial patient demographic data, UAE procedural data, and short-term (30-day) outcomes for 2332 cases entered into the Registry between 13 December 2001 and 31 August 2002. Materials & Methods: The Registry is a prospective observational database. Participating IR practices (54 USA, four non-USA) enroll all consecutive patients into the registry. Initial data, and short-term followup are entered into the database by the IR practices. Follow-up information is obtained by a central data-coordinating center. The protocol and case report forms can be accessed for review at http://www.fibroidregistry.org Results: African-American women tend to have worse fibroid disease than white women. Procedures and patient management techniques will be described. Technical success rate was 98%. In-hospital adverse events were rare (3%). Adverse events within 30 days occurred in 26% of patients. Most adverse events after the procedure were minor. Reintervention within 30 days of UAE occurred in 1%. Conclusion: The Registry population is similar to other fibroid treatment populations. Procedure and fluoroscopy times were within acceptable ranges. Significant peri-procedural and short-term complications are rare.

35.1.8

35.2.2

3-D imaging after endovascular repair of thoracic aortic disease using archived CT datasets B. V. Czermak1, A. Mallouhi1, I. E. Steingruber1, T. Jung2, B. Neuhauser3, G. Fraedrich3, W. R. Jaschke1; 1Radiology, University Hospital of Innsbruck, Innsbruck, Austria, 2Radiology, Waid Hospital, Zürich, Switzerland, 3Vascular Surgery, University Hospital of Innsbruck, Innsbruck, Austria.

Mid-term results of uterine artery embolization for symptomatic adenomyosis J. P. Pelage, O. Le Dref, D. Jacob, A. Fazel, J. Namur, A. Laurent; Body and Vascular Imaging, Hopital Lariboisiere, Paris, France.

Purpose: To evaluate the value of digitalization and PACS storage of CT datasets in patients treated for aortic disease by means of endovascular repair. Material & Methods: From January 1996 a total of 130 patients underwent endovascular repair of various aortic diseases. Follow-up studies included routine enhanced CT study and an extra nonenhanced spiral CT of the stent-graft (section thickness 1.25mm, tablefeed/rotation 3.75, HQ mode). The datasets were transferred and stored in the PACS. During follow-up, 3-D VR and MIP reconstructions of the graft and the aorta were performed and rotated into the most appropriate plane. Images were compared with each other Results: On the base of 3-D reconstructions, evaluation of time-related alterations in shape, position, and structure of the stent-graft was possible in all patients. In addition severity and causes of complications could be analyzed in 20/25 patients. The reconstructions provided important information concerning medium-term outcome of different interventional procedures and stent-graft designs. Conclusions: The possibility of storing postinterventional CT datasets in the PACS enables problem-specific 3-D reconstructions of the graft and the aorta during follow-up. These results are important for both clinical application and research.

Purpose: To evaluate mid-term results of uterine artery embolization in the management of menorrhagia due to adenomyosis. Materials & Methods: Eighteen women (mean age 44.4y, range 36-52) with menorrhagia due to adenomyosis underwent bilateral uterine artery embolization. Diffuse adenomyosis was diagnosed in presence of an enlarged junctional zone and myometrial cysts. Focal adenomyosis was diagnosed in case of a circumscribed nodular lesion mimicking an intramural fibroid. Uterine artery embolization was performed using PVA particles, trisacryl calibrated microspheres and/or gelatin sponge pledgets. Clinical evaluation was made at regular intervals to assess outcome. Follow-up ultrasound or MRI was performed six months after embolization. Results: Fourteen women had diffuse and four had focal adenomyosis. Bilateral uterine artery embolization was performed in all but one woman. At six months, 16/18 women (89%) reported a significant improvement of symptoms. Follow-up imaging demonstrated a slight decrease of the uterine volume in 17/18 women (94%). At a mean follow-up of 20.5 months (7-72 months), only 5/18 women (39%) were still significantly improved. Five women had been treated by hysterectomy for recurrence. Conclusion: Bilateral uterine artery embolization provides an initial symptomatic relief but long-term evaluation suggests a rate of recurrence higher than that with uterine fibroid embolization.

35.2.3

35.2.5

Long-term behaviour of uterine fibroids after flow-guided spherical particle embolisation G. M. Richter1, B. Radeleff1, S. Rimbach2, G. W. Kauffmann1; 1Diagnostic Radiology, University Hospital Heidelberg, Heidelberg, Germany, 2Obstetrics and Gynecology, University Hospital Heidelberg, Heidelberg, Germany.

Uterine fibroid embolization with PVA: comparison of two different angiographic end-points J. Golzarian1, F. Tannouri1, M. Laureys1, N. Mathieu2, S. Murgo1; 1Radiology, Erasme hospital, Brussels, Belgium, 2Anestesiology, Erasme hospital, Brussels, Belgium.

Purpose: To report the course of uterine fibroids after spherical particle embolisation. Materials & Methods: From January 2001, in 47 women who had undergone uterine fibroid embolisation, a strict pre- and post-procedural MR imaging protocol was performed to determine the natural course of the embolised fibroids. Embospheres ranging in size between 500 and 900 µ were used. Results: All but one procedures were technically successful. One technical failure was due to a rare anatomic variation. One procedure failed clinically, requiring late hysterectomy. Otherwise no major complications were recorded. Presently, a follow-up longer than 12 months is available in 24 women. In 7/24 there are no more fibroids, while in the remaining 17/24 the residual size of the dominant fibroid measures 29% of the original one. According to standardized test questionnaires for pain and menstrual blood loss, clinical symptoms diminished in all the patients. Similarly, after the procedure the general health score increased from an average of 37% to 95% (100% represents the maximum). Conclusion: Size-adapted and flow-controlled particle embolisation of uterine fibroids provided an effective control of symptomatic myomatous disease with a high probability of significant fibroid size reduction or even passage on flow-up, as demonstrated by MRI.

Purpose: To compare limited uterine artery (UAs) embolization (LUE) with complete UAs embolization using PVA particles. Materials & Methods: LUE was performed in 11 consecutive patients with uterine fibroids (group A) and results compared with another group of 11 consecutive patients who had undergone complete UAs embolization (group B). PVA particles (500-700 µm) were used in all patients. Mean morphine doses, duration of intense pain, clinical success, uterine volume and complications were compared for each group. Results: Mean doses of morphine for groups A and B were 14 and 38 mg, respectively. Mean duration of intense pain was 115 min. (group A) and 476 min. (group B). Uterine volume reduction was 62 and 69.8 %, respectively. Mean follow-up was ten months in group A and 25 months in group B. Symptoms resolved completely or significantly improved in 90% of patients in both groups. Fibroid expulsion occurred in one group-A patient and in two group-B patients. A transient amenorrhea was observed in one patient in group A and in two patients in group B. Vaginal discharge was observed in two group-B patients. Conclusion: Preliminary results demonstrate that LUE embolization with PVA is effective and compares well with complete UAs embolization.

35.2.4

35.2.6

Targeted fibroid embolization with calibrated microspheres: shortterm clinical and MRI results P. N. Lohle1, L. E. Lampmann1, A. J. Smeets1, F. P. Boekkooi2, H. A. Vervest2, J. J. Pieters2; 1Radiology, St. Elisabeth Hospital, Tilburg, Netherlands, 2Gynecology, St. Elisabeth Hospital, Tilburg, Netherlands.

Uterine artery embolization for symptomatic uterine fibroids: the Belgian experience J. Niedercorn1, X. Kos2, O. d'Archambeau3, S. Murgo1, P. Magotteaux2, J. Golzarian1; 1Radiology, Erasme Hospital, Brussels, Belgium, 2Radiology, Saint Joseph, Liege, Belgium, 3Radiology, St Augustin, Antwerpen, Belgium.

Purpose: To evaluate the safety and efficacy of uterine fibroid embolization (UFE) with calibrated microspheres. Materials & Methods: Seventy women (mean age 43 years) with symptomatic uterine fibroids underwent embolization of the uterine arteries using calibrated microspheres (Embospheres and Embogold, 500900um). Indications for treatment were heavy menstrual bleeding in 93%, pain in 74%, and bulk related symptoms in 47% of patients. Clinical and MRI follow-up was obtained at three months in 12, at six months in 25, and at 12 months in 33 patients. Relief of symptoms and volume reduction of the uterus and fibroids were assessed. Results: UFE was accomplished in all 70 patients. Mean clinical folluwup was six months (range 3-12 months). At the last follow-up MRI the mean uterine volume reduction was 42%, and the mean dominant fibroid volume reduction was 54%. At the last clinical follow-up improvement of presenting symptoms was reported to be: 88% for bleeding, 87% for pain, and 82% for bulk-related symptoms. No emergency hysterectomies were needed. Patient satisfaction at last clinical follow-up was: 81% very satisfied, 13% satisfied, 6% not satisfied. Conclusion: UFE using calibrated microspheres is safe and effective in controlling menstrual bleeding, pain and bulk related symptoms.

Purpose: To evaluate the effectiveness of uterine artery embolization for symptomatic fibroids in a multicenter study. Materials & Methods: This is a retrospective study from a prospective database in three Belgian hospitals. Ninety-two patients (mean age: 42.6) with symptomatic fibroids were treated by bilateral uterine artery embolization. Pre- and post-embolization MRI or ultrasound were performed to assess fibroids and uterine volumes. Clinical success was monitored with office consultations and telephone follow-ups. Results: All the procedures were technically successful. At a mean of 14.5 months of follow-up, menstrual bleeding had significantly improved in 86% of cases. Pelvic pain and bulk-related symptoms had improved in 87% and 87% of patients, respectively. Mean dominant fibroid volume and uterine volume decreased to 58.5% and 52.6%, respectively. One patient had hysterectomy for infection. Fibroid expulsion was observed in 4.3% of patients. Transient and permanent amenorrhea occurred in 7.6% and 2.2% of cases, respectively. Conclusion: Uterine artery embolization provides an effective control of symptomatic fibroids with a low morbidity rate.

35.2.7

35.3.1

The management of uterine fibroids. A comparison of uterine fibroid embolization therapy versus myomectomy -A retrospective review S. M. Ford, C. J. Johnston, J. McCann, E. Laffan, N. McEniff; Diagnostic Imaging Department, St.James's Hospital, Dublin, Ireland.

Is there a role for the Silicon-Carbide coated stent below the groin? M. Bosiers1, P. Peeters2, K. Deloose1, J. Verbist2; 1Vascular Surgery, AZ St-Blasius, Dendermonde, Belgium, 2 Cardiovascular and Thoracic Surgery, Imelda Hospital, Bonheiden, Belgium.

Purpose: To compare uterine fibroid embolization with myomectomy as a treatment option for uterine fibroids, under the headings of symptom improvement, length of hospital stay and complication rates. Materials & Methods: We started performing uterine fibroid embolization in our institution three years ago and in that time 60 cases have been carried out; in the same time period, eight patients have undergone myomectomy. We reviewed all 60 cases of uterine fibroid embolization and compared our findings with those of the patients who had undergone myomectomy. Results: There were significantly lower major complication rates in the uterine fibroid embolization group than in the myomectomy group, with only one out of the 60 patients experiencing a major complication, whereas two myomectomy patients experienced major complications (1.6% vs. 25%). The length of hospital stay was also significantly shorter in the embolization group than in the myomectomy group. Symptomatic improvement was similar in both groups. Conclusion: Uterine fibroid embolization is a safe and effective therapy for uterine fibroids, with shorter hospital stay times and lower complication rates than myomectomy.

Purpose: To investigate the effectiveness of Silicon-Carbide coated stents in SFA (Peiron) and BTK-lesions (Lekton). Materials & Methods: Silicon-Carbide coatings prevent direct contact between blood or surrounding tissue and allergenic metal-ions, resulting in limited release of nickel and chromium. The passive diffusion barrier seals the bare metal surface, preventing corrosion and allergic reactions, and furthermore reduces fibrin, platelet, and leukocyte deposition, significantly lowering the risk of early and late restenosis. From June 2001 to April 2002, 91 patients with 100 SFA-lesions up to 50mm were included in the Peiron-group of the trial. Inclusion of patients with short below-the-knee lesions and CLI in the Lekton-group started in September 2002; a total of 50 will be included by March 2003. Results: To date, our series of Peiron-implants in short SFA-lesions shows a six-month primary patency rate of 92% (N=88). Twelve-month results of the Peiron-series and six-month patency rates of the Lektonseries will be available by September 2003. Conclusion: Preliminary results for Silicon-Carbide stents below the groin are promising. The passive coating significantly increases patency rates after stenting and offers a valuable cost-effective alternative for the more expensive active coated stents.

35.2.8

35.3.2

Serum levels of lactate dehydroganase and creatine kinase in women after uterine artery embolization as treatment for symptomatic leiomyomas R. Pietura1, M. Janczarek1, A. Szymańska1, W. Bednarek2, J. Kotarski2, M. Szczerbo-Trojanowska1; 1Interventional Radiology & Neuroradiology, University Medical School of lublin, Lublin, Poland, 2Gynecology, University Medical School of lublin, Lublin, Poland.

Sirolimus-eluting stents in SFA obstructions: long-term SIROCCO trial results S. H. Duda1, B. Wiesinger1, G. M. Richter2, P. Landwehr3, V. Oliva4, A. Tielbeek5, H. Tielemans6, J. Beregi7; 1Radiology, Univ.of Tuebingen, Tuebingen, Germany, 2Radiology, Univ.of Heidelberg, Heidelberg, Germany, 3Radiology, Henriettenstiftung, Hannover, Germany, 4Radiology, Univ.of Montreal, Montreal, Canada, 5Radiology, Catharina Hospital, Eindhoven, Netherlands, 6a Johnson & Johnson Company, Cordis, Waterloo, Belgium, 7Radiology, Univ.of Lille, Lille, France.

Purpose: To assess serum levels of lactate dehydroganase (LDH) and creatine kinase (CK) after uterine artery embolization (UAE). Materials & Methods: Between April and June 2002, 50 women (mean age 45.5±4.8) with symptomatic leiomyoma underwent UAE. The mean volume of the dominant leiomyoma was 223.9 ± 48.2 cm3. Mean volume of uterus was 525.6 ± 62.6 cm3. In all patients we evaluated total activity of LDH and CK, as markers of necrosis, in venous blood serum before the procedure and 8, 24, and 48 hours after UAE. In 20 patients we further evaluated these enzymes at 2, 4, 6, and 8 weeks after UAE. Results: In all patients we observed significant elevation of serum LDH and CK within 48 hours after procedure, without however exceeding normal limits. A strong correlation between increase in serum LDH within 48 hours and the number of leiomyomas was observed. A similar correlation was revealed between serum CK within 24 hours and the number of leiomyomas. Conclusion: The serum LDH and CK increases after UAE but never to higher than normal limits. This suggests slow, not massive necrotic changes in leiomyomas after UAE.

Purpose: Long-term evaluation of self-expanding Sirolimus-eluting nitinol SMART stents versus bare SMART stents in the superficial femoral artery (SFA). Materials & Methods: Thirty-six patients with chronic limb ischemia, femoral artery occlusions (57%) or stenoses (average lesion length, 85±57 mm) were treated with Sirolimus-eluting (n=18 with a fast or slow Sirolimus release kinetic) and uncoated stents (n=18) in a doubleblind, randomized, prospective trial. Study endpoints were the in-stent percent mean diameter stenosis quantified via angiography at six months and by duplex ultrasound at 18 months. Results: Angiographic six-month in-stent percent mean diameter stenosis was 11.7% for the slow release and 27.5% for the fast release group versus 30.9% in the control group. This translates into a sixmonth 0% binary restenosis rate for the coated versus 23.5% for the control group; 18-month binary in-stent restenosis quantified by duplex US was 0% and 33% in the slow and fast release groups, respectively, versus 30% in the control group. Six stent fractures and two nonrelated deaths (Sirolimus group) were observed. Conclusion: The use of Sirolimus-eluting nitinol SMART stents in the femoral artery is feasible and seems to be safe and effective with the slow release kinetic. Uncoated SMART stents performed unexpectedly well.

35.3.3

35.3.5

Peripheral stenting using the silicon-carbide coated Peiron stent: medium-term follow-up O. C. d'Archambeau1, O. G. François2, J. Hendriks3, J. Poniewierski3, A. M. De Schepper2; 1 Interventional radiology, University Hospital Antwerp, Edegem, Belgium, 2Radiology, University Hospital Antwerp, Edegem, Belgium, 3 Vascular surgery, University Hospital Antwerp, Edegem, Belgium.

Radial force and flexibility as parameters for vessel wall adaptation of self-expanding peripheral Nitinol stents W. Schmidt1, R. Andresen2, P. Behrens1, K. P. Schmitz1; 1Institute for Biomedical Engineering, University of Rostock, Rostock, Germany, 2Radiology Clinic, Hospital Güstrow, Academic Teaching Hospital of the University of Rostock, Güstrow, Germany.

Purpose: A prospective study was undertaken to evaluate the efficacy of the silicon-carbide coated balloon-expandable Peiron stent in the iliac- and femoropopliteal arteries Materials & Methods: From May 2001 to January 2003, 130 Peiron stents were placed in 97 patients (72 male) with PAOD (59 iliac stents, 65 femoral stents and 6 popliteal stents). Patients were treated with Fontaine stage IIa (5), IIb (41), III (19) and IV (32). Indications for femoropopliteal stenting were short (eccentric) calcified stenosis or occlusions. For long lesions or suboptimal angioplasty results, a selfexpanding nitinol stent was used. Follow-up was performed by clinical examination and color-duplex sonography before and after the procedure, at 3, 6, 12 and 18 months. The follow-up was 2-90 weeks (mean, 39). Results: The technical success rate was 100%. The initial clinical success rate was 96% (four patients required amputation for stage IV disease). Complications occurred in 10 patients (eight minor, two major). The six-month primary patency rate (n=67) was 91% (100% iliac, 84% femoropopliteal). The one-year primary patency rate (n=28) was 81%. Conclusion: The success at medium-term follow-up of Peiron stents in peripheral arteries is better than other balloon-expandable stents and compares favorably to bare nitinol stents.

Purpose: Self-expanding peripheral stents were investigated mechanically to assess their suitability to be used in specific vascular regions. Materials & Methods: In-vitro tests included: Guidant Acculink, Cook Zilver stent, Biotronik Philon, Cordis Precise, Bard Luminexx, Jomed Jostent SelfX, and Optimed Sinus Super-Flex. All stents were 8 mm in diameter, 38-44 mm long, and made of nitinol. Their radial force during expansion, their flexural stiffness, and their vessel wall adaptation in a "zigzag-shaped" vessel model as well as in a model with a diameter step from 5 to 7 mm, were measured. Results: At 7 mm, diameter radial forces of 1.09 (Sinus Super-Flex) up to 2.59 N (Philon) were measured. The stent stiffness ranged from 11.7 (Acculink) to 88.1 Nmm² (Luminexx). Vessel wall adaptation was best for Precise, SelfX, Philon, and Zilver stents. Rapid changes of the lumen cross-section were directly matched by stents with shorter segments (Precise, Acculink). Conclusion: The conformity of a stent to a curved vessel or a diameter step depends on its structure more than its radial force and stiffness. Flexible stents avoid the straightening of curved vessels. The differences noted in our study may support the clinician in the stent selection for a particular vascular lesion.

35.3.4

35.3.6

Complement inhibition reduces restenosis and inflammation after stent implantation in New Zealand White Rabbits: investigation with coated stents G. Tepe1, J. Trück1, H. J. Wendel2, U. Brehme1, C. C. Claussen1, S. H. Duda1; 1Diagnostic Radiology, University of Tuebingen, Tuebingen, Germany, 2University of Tuebingen, Tuebingen, Germany.

Stent versus percutaneous transluminal angioplasty in the treatment of infrapopliteal lesions T. Rand, C. Loewe, M. Schoder, J. Kettenbach, M. Cejna, J. Lammer, M. Funovic; Interventional Radiology, Vienna University, Vienna, Austria.

Purpose: To investigate the role of inflammation during stent placement and the value of complement inhibition by stent coating with C1/ C3 esterase inhibitor (CI). Materials & Methods: Twenty-one New Zealand White rabbits were fed a 0.5% cholesterol diet before balloon angioplasty and insertion of Jomed stainless steel stents in the infrarenal aorta. The animals were sacrificed 6 weeks after stent implantation. Bare stents (n=7) were compared with hydrogel (HI) coated (n=7) and hydrogel-CI (HCI) coated (n=7) stents. The inflammation index was assessed using hematoxylineosin staining. Results: No acute or subacute thrombosis was observed. After 6 weeks neointima formation was 1.8 ± 0.6 mm2 (bare stent) and 2.6 ± 1.3 mm2 (HI) in the control groups. In the treatment group (HCI) neointima reduction was detectable (1.5 ± 0.6 mm2). Inflammation was increased by HI coating. In the HIC stents the inflammation was significantly reduced compared to the HI stents. Conclusion: Inflammation is a key component of restenosis following stent administration. Inflammation can be reduced by C1/C3 esterase inhibitor coated stents. Because the hydrogel coating itself induced inflammation, this coating seems not to be the ideal for intravascular use.

Purpose: To assess the potential benefit of infrapopliteal stent insertion versus percutaneous transluminal angioplasty (PTA) in patients with high-grade stenoses of infrapopliteal arteries. Materials & Methods: In a prospective, randomized multicenter study patients with uni or multilocular high-grade stenoses (>80%) of the infrapopliteal arteries up to 3 cm in length were treated by PTA or by stent insertion (CarboStent, coronary stent system, Sorin). Patency rates and clinical outcomes were evaluated six to18 months after the procedure by conventional or CT-angiography. Until now, the procedures including the follow-up have been performed 54 times (32 patients). Results: Twenty-eight PTA procedures were performed in total. Followups showed patency in 12 cases and occlusion in 16. In 27 stent insertions, the follow-up showed patency in 23 cases and occlusion in four. The median patency rate for PTA was seven months, for stent implantation >21 months. The six-month patency rate for PTA was 51.2%, for stent insertion 81.2%, the12-month patency rate for the stented group was 81.2%. These results are statistically significant. Conclusion: Stent implantation proved to be a feasible treatment modality for infrapopliteal stenoses with a benefit for the patients as compared with angioplasty.

35.3.7

35.4.1

Assisting patency with stent placement in distal anastomotic stenosis of lower limb bypass grafts G. Marano1, A. R. Cotroneo1, E. Migazzo1, G. Guzzardi1, F. Nessi2, F. Bellini2, A. Carriero1; 1 Istituto di Radiologia, A.O. "Maggiore della Carità", Novara, Italy, 2 Divisione di Chirurgia Vascolare, A.O. "Maggiore della Carità", Novara, Italy.

Computational coronary flow dynamics based on 16-row multidetector CT data-sets T. Frauenfelder1, E. Boutsianis2, T. Schertler1, H. Dave3, J. Gruenenfelder3, B. Marincek1, Y. Ventikos2, S. Wildermuth1; 1 Inst. Diagostic Radiology, University Hospital Zuerich, Zuerich, Switzerland, 2Ltnt, Swiss Federal Institute of Technology, Zuerich, Switzerland, 3 Clinic for Cardiovascular Surgery, University Hospital Zuerich, Zuerich, Switzerland.

Purpose: To evaluate feasibility and early results of endovascular stent placement for treating distal anastomotic stenosis in below-knee bypass grafts. Materials & Methods: Between November 2001 and December 2002, 10 consecutive patients (seven male, ten female; age 55-89 years, average 71) were treated with endovascular stents for stenosis of distal anastomoses during follow-up for infragenicular bypass grafts (seven venous, two prosthetic-vein composite, one prosthetic). In 14 procedures (four patients were treated twice), 17 stents were positioned. In all cases femoral anterograde puncture was performed, using 4-6 F introducer sheaths. Only monorail stents were employed (14 balloonexpandable, three self-expandable). Patient selection and follow-up were performed with color duplex ultrasonographic surveillance. In no case was predilatation required before stent placement. Results: Technical success was achieved in 14/14 procedures (100%), without early complications. During follow-up (1-14 months, average 5.8), 13/14 by-passes are still patent. Four in-stent restenoses presenting respectively at 4, 7, 9 and 10 months were treated with repeat stenting. One bypass occlusion occurred three months after the procedure and required surgical revision. No major amputation was needed. Conclusion: In this short-term follow-up, endovascular treatment of infracrural anastomotic stenosis with stenting proved feasible, safe and repeatable, with improvement of patency rate.

Purpose: Modern paradigm of cardiovascular surgical planning involves the use of computational tools to construct and evaluate combined anatomic/physiologic models, thus improving the therapy accuracy and consistency. We simulated coronary flow using computational hemodynamic models based on cardiac multidetector-CT (MDCT). Materials & Methods: The coronaries of an explanted porcine heart and five patients after coronary stenting were included; ECG-gated 16-row MDCT-data were used to reconstruct the 3-D coronary geometry; which in turn was used to generate a volumetric tetrahedral mesh of approximately 600.000 elements. The patient-specific, anatomically accurate hemodynamics are simulated using validated computational fluid dynamic tools. The unsteady equations of momentum and mass conservation were numerically solved. Results: Computational simulations yielded a wealth of quantitative information, including pressure and wall shear stress distributions, bifurcation mass flow rates and detailed velocity profiles. The variability of these quantities within the heart cycle was investigated at a temporal resolution of 1/100 second. Conclusion: Such computational tools along with the possibility of MDCT-based noninvasive high temporal/spatial resolution coronary imaging could enable objective quantification of the effects of stenosis on the distal flow and pressure and thus aid in effective surgical/interventional planning. Additionally, it could enhance also our insight into coronary pathogenesis.

35.3.8

35.4.2

Temporary ZA-stent: a clinical study Z. A. Kavteladze, D. P. Doundoua, S. A. Drozdov, D. S. Kartashov, K. V. Bylov, A. M. Babunashvili; Cvir, CELT, Moscow, Russian Federation.

Comparison of MRI and echocardiography in the evaluation of cardiac masses S. Sharma1, G. Gulati1, S. S. Kothari2, R. Juneja2, A. Saxena2, P. Venugopal3; 1Cardiovascular Radiology, All India Institute of Medical Sciencse, New Delhi, India, 2Cardiology, All India Institute of Medical Sciencse, New Delhi, India, 3Cardiothoracic & Vascular Surgery, All India Institute of Medical Sciencse, New Delhi, India.

Purpose: To evaluate the new temporary ZA-stent in peripheral vascular. interventions. Materials & Methods: Temporary ZA-stents were implanted in 14 patients with stenotic iliac arterties lesions for a period of seven days and one patient with iatrogenic common iliac dissection for two days. Fifteen modified ZA-stents (W.Cook Europe A\S, Demnark) with anchoring elements at the distal ends were implanted in common and superficial iliac arteries. Stent diameters were 8 and 10 mm and the length was 40 or 60 mm. Stents were deployed and retrieved via an 8F delivery system. Clinical evaluation included ABI, duplex sonography, quantitative angiography. Results: All stents were implanted and retrieved without difficulty, stent thrombosis, late migration, or damage from distal arterial embolization. The minimal arterial lumen of 7.8+0.4 mm achieved after stent implantation did not changed significantly after stent retrieval. ABI increased from 0.5+ 0.2 up to 1.0+0.05 after implantation, and had not changed significantly 1-12 months later. Ultrasonic Doppler measurements calculating transstenotic gradient did not increase during the follow up. Six-month angiographic data is emerging. Conclusion: Temporary ZA-stent implantation and retrieval in patients with peripheral vascular disease is technically feasible and safe. Clinical follow-up revealed sustained clinical benefit.

Purpose: To compare MRI and echocardiography in evaluating intracardiac masses. Materials & Methods: 28 patients (21 males, mean age 25 years) with intracardiac masses on echocardiography were studied by MRI. Lesion location, relation to adjacent structures, and morphology were evaluated. Diagnosis was confirmed by histopathology in 18, surgical inspection in four, resolution on specific medical therapy in five, and typical extracardiac manifestations of disease in one. Results: There were 15 tumors ( benign 12, malignant three), five hematomas, and four each with infective and vascular origins. Thirtyfour masses were seen on MRI (13 in ventricle, 11 septal, seven atrial, two AV valve and one PA), of which echocardiography detected 28. TTE and MRI were of diagnostic quality in 82% and 100% respectively. Nine patients required additional TEE for echocardiographic diagnosis. History and echocardiography were diagnostic in 29%, and MRI in 75%. Intramural components, extension into the inflow or outflow of the involved chamber, outflow obstruction, and additional masses were better depicted on MRI. In seven cases, MRI changed management in favor of conservative treatment (four mural clots), medical therapy (one TB) and adjunct surgery (two hydatid cysts) Conclusion: MRI is superior to TTE and TEE in evaluating intracardiac masses.

35.4.3

35.4.5

Coronary artery atherosclerotic disease: what role for multidetector row CT-angiography? A. Romagnoli, M. Tomassini, R. Gandini, M. Sperandio, G. Simonetti; Diagnostic Imaging and Interventional Radiology, University of Tor Vergata, Rome, Italy.

Six-month angiographic and intravascular ultrasound follow-up after implantation of drug-eluting stents in the coronary arteries G. Gabrielli1, T. Piva1, A. Pangrazi1, F. Schicchi1, C. Costantini2, R. Piva1; 1Cardioradiology - Haemodynamics, "G.M. Lancisi" Cardiological Hospital, Ancona, Italy, 2Intensive Care Unit, "G.M. Lancisi" Cardiological Hospital, Ancona, Italy.

Purpose: To report our experience with multislice-CT (MSCT) in comparison with conventional angiography in the detection and grading of coronary diseases. Materials & Methods: From December 2002 to Febraury 2003, 65 patients (mean age 61, range 45-72) previously studied with conventional coronarography underwent MSCT in our departement (GE LightSpeed Plus, 1.25-mm collimation, 0.5-second rotation time). Inclusion criteria were: no previous history of coronary disease or suspicion of ischemia at stress ECG or myocardial scintigraphy. Results: Of 195 arteries studied, 29 (14.9%) were considered notassessable due to stair-step and motion artifacts. Eighty-six (44.1%) arteries were considered normal, 42 (21.5%) had minimal wall irregularities with <50% stenoses, 20 (10.2%) arteries presented a >50% stenosis, while in 18 (9.3%) vessels a total occlusion was demonstrated. Sensitivity and specificity of MSCT in comparison with DSA will be discussed for every group. Conclusion: The use of MSCT in the study of coronary arteries is a promising technique with a good potential for its use in the routine clinical practice. In selected patients, it can provide very interesting results and could be used as a mass-screening tool to select patients to be referred for coronary angiography.

Purpose: To evaluate the effectiveness of limiting restenosis of drugeluting coronary stents, we present our six-month follow-up results in terms of angiographic, intravascular ultrasound (IVUS) and clinical outcomes. Materials & Methods: Since May 2002, Sirolimus-eluting stents were implanted in 79 patients, all anatomical or clinical complex cases (typeC lesions, patients in dialytic treatment, in-stent restenosis, diabetic patients). Immediate results were evaluated angiographically and with IVUS. Patients underwent angiographic and IVUS studies six and 12 months after percutaneous coronary intervention (PCI). Results: We present the results of six-month follow-up for the first 24 patients. One patient presented subacute thrombosis, two were readmitted for clinical symptoms: one had a focal restenosis, the other had an extensive restenosis in a long segment of a bypass graft. No deaths occurred. All the other patients had angiographically evaluated <10% diameter stenoses. In-stent neointimal hyperplasia remained minimal, as evaluated by IVUS. More complete series for one-year follow-up will be available in September. Conclusion: Sirolimus-eluting stents seem effective in limiting restenosis after PCI. A longer term follow-up is needed. IVUS is a useful tool in guiding PCI and evaluating the characteristics and sites of restenosis.

35.4.4

35.4.6

Stent patency evaluation by multislice CT: our preliminary experience M. Tomassini, A. Romagnoli, R. Gandini, A. Nisini, G. Simonetti; Diagnostic Imaging and Interventional Radiology, University of Tor Vergata, Rome, Italy.

Sixteen-row multislice computed tomography versus conventional pulmonary venography: assessment of pulmonary veins before and after cryothermal energy ablation R. Maksimovic1, M. Scholten2, F. Cademartiri1, v. Jordaen2, P. M. Pattynama1; 1Department of Radiology, Erasmus MC, Rotterdam, Netherlands, 2Department of Cardiology, Erasmus MC, Rotterdam, Netherlands.

Purpose: To report our preliminary experience in the follow-up of coronary stenting. Can multislice computed tomography (MSCT) evaluate stent patency? Materials & Methods: Twenty-one patients (13 men, mean age: 64 years) with a history of ischemic cardiopathy treated by coronary stent implantation were followed using MSCT coronary angiography with retrospective ECG gating (GE LightSpeed Plus). Post-stent coronarographies were available for all the patients. Since seven patients had undergone a double stent implantation, an overall of 28 stents were studied. MSCT results were compared with conventional coronarography, considered as the gold standard in the diagnostic of coronary diseases. Results: Four stents were considered as non-assessable because of movement artifacts due to a high cardiac frequency. The follow-up of the remaining 24 stents yielded: three complete occlusions; four mild restenoses; 17 regularly patent stents. Although the visualization inside the mesh of five stents placed distally was not possible, they were considered patent because of the blood flow present within the distal portion of the artery. Conclusion: MSCT coronary angiography may be considered as a potential non-invasive alternative in the follow-up of coronary stenting. Furthermore, the technique enables the direct evaluation of in-stent restenoses without the need of conventional coronarography.

Purpose: In the treatment of atrial fibrillation, cryothermal energy (CE) ablation is a new method to obtain the electrical isolation of pulmonary veins (PVs). PVs are visualized by 16-row multislice computed tomography (MSCT). The aim of this study was to compare PVs diameters on MSCT with conventional pulmonary venography (CPV) for the presence of stenoses and for a non-invasive patients' follow-up. Materials & Methods: Twenty patients (80 PVs), (mean age 52.6±7.6 years, 15 men) with atrial fibrillation underwent CE ablation. CPV for the quantitative vessel analysis (Pie Medical) and MSCT images were obtained before the procedure, to evaluate the ostium and the proximal 10 mm of all PVs. Three months after CE, MSCT only was repeated, for the assessment of PVs. Results: There was a good correlation between MSCT and CPV measurements (r=0.82, p<0.01). Thirty-nine (48.7%) PVs were treated with CE ablation. Diameters of the treated PVs before and after the procedure did not differ significantly on MSCT (19.6±2.9 mm vs. 18.3±2.3 mm, p>0.05, respectively), as well as those of non-treated PVs (17.2±1.8 mm vs. 17.2±2.1 mm, respectively, p>0.05). Conclusion: MSCT is a reliable and reproducible method to visualize PVs' diameters prior to isolation, and a non-invasive follow-up after ablation.

35.4.7

35.5.2

Impact of 16-slice CT on visualization of subsegmental pulmonary arteries P. Herzog, K. Herrmann, T. K. Helmberger, M. F. Reiser; Dep. of Clinical Radiology, University of Munich, Munich, Germany.

Nitric oxide and oxidative stress in atherosclerotic renovascular hypertension: effect of endovascular treatment M. Parildar1, Z. Parildar2, I. Oran1, C. Kabaroglu2, A. Memis1, O. Bayindir2; 1 Radiology, Ege University, Izmir, Turkey, 2Clinical Biochemistry, Ege University, Izmir, Turkey.

Purpose: According to the published literature only 20-60% of all subsegmental arteries can be evaluated using 4-slice CT, depending on the section thickness used. Our goal was to evaluate the diagnostic benefit of 16-slice CT for the evaluation of segmental and subsegmental pulmonary arteries. Materials & Methods: One hundred four consecutive patients with suspected acute pulmonary embolism were examined using a 16-slice CT scanner (Somatom Sensation16, Siemens Medical Solutions). For contrast enhancement, a bolus of 120cc was injected at a flow rate of 5 cc/ sec followed by a saline flush of 50cc. A thoracic radiologist with six years experience in interpreting thoracic CT read the axial slices off a workstation. In the 104 patients, each segmental (n=20) and subsegmental (n=40) pulmonary artery was individually rated as analyzable or indeterminate. The contrast enhancement in the main pulmonary artery and in each segmental artery was measured in patent vessels. Results: 97.9% of subsegmental pulmonary arteries and 100% of segmental pulmonary arteries were rated as analyzable. Conclusion:16-slice CT with 0.75-mm thick sections improves visualization of segmental and subsegmental pulmonary arteries compared to results reported for 4-slice CT. With this technique, the percentage of non-diagnostic studies should be further reduced.

Purpose: We studied nitric oxide (NO), superoxide dismutase (SOD), catalase, and malondialdehyde (MDA) levels and the effect of endovascular treatment on these parameters in atherosclerotic renovascular hypertensive patients and their relationship with blood pressure and renal function indices. Materials & Methods: In this prospective cohort study, serum creatinine, NO , SOD, catalase, plasma MDA, urinary microalbumin, NO and blood pressure were determined in 21 hypertensive patients with unilateral renal artery stenosis due to atherosclerosis at entry and at 24 hours, 2 and 6 weeks after endovascular treatment. Results: MDA concentrations decreased 24 hours after the intervention and remained low thereafter. Serum SOD, NO and urine NO levels increased significantly 24 hours after endovascular treatment and decreased thereafter. However, serum catalase levels did not change after the intervention. There were no significant differences in urinary microalbumin or estimated glomerular filtration rate (EGFR). Conclusion: Endovascular treatment decreases oxidative stress and may offer new benefits in the treatment of patients with hypertension associated with renal-artery stenosis. The decrease in oxidative stress and/or the upregulation of SOD may increase the bioavailability of NO, which in turn may lead to the rapid hypotensive response.

35.5.1

35.5.3

Clinical evaluation and restenosis rate after percutaneous transluminal renal artery angioplasty and stenting (PTRAS) in diabetic and non-diabetic patients with hypertension and renal artery stenosis (RAS) M. Venturini1, R. Quartagno2, F. De Cobelli1, S. Tedoldi2, E. Angeli1, A. Del Maschio1; 1Radiology, San Raffaele Hospital, Milan, Italy, 2Nephrology, San Raffaele Hospital, Milan, Italy.

Renal artery stenosis: overvalued as a cause of dialysis T. C. Gill-Leertouwer1, M. van Vliet1, B. Zietse2, P. M. Pattynama1; 1Radiology, Erasmus Medical Center, Rotterdam, Netherlands, 2Nephrology, Erasmus Medical Center, Rotterdam, Netherlands.

Purpose: To evaluate restenosis rate (RR) and variation of mean blood pressure (MBP), serum creatinine levels (SC), antihypertensive drugs number (AD) of non-diabetic and diabetic patients one year after PTRAS. Materials & Methods: Fifty-five significant RAS (>50%) in 47 patients were treated with balloon-expandable stents. Variations of MBP (mm/ Hg), SC (mg/dl), AD (n) and RR (%) were evaluated. Before PTRAS, the following values were recorded: MBP=114.07±2.09, SC=1.24±0.07, AD=2.06±0.15 in the whole study population; MBP=114.5±2.4, SC=1.25±0.08, and MBP=110.33±4.02, SC=1.21±0.12, considering nondiabetic (n=27) and diabetic patients (n=20), respectively. RR (³50%) was evaluated with CDU (ATL-Philips-HDI5000). Results: A statistically significant (p=0.001) reduction of MBP (105.20±1.60) was found with no significant differences between nondiabetic (105.3±1.9) and diabetic (104.4±1.7) patients. A statistically significant (p=0.05) increase of SC (1.35±0.08) was recorded, with a statistically difference between non-diabetic (1.29±0.09) and diabetic (1.56±0.18) patients. A slight reduction of AD (1.91±0.01) was observed. Considering the total study population, RR was 29.78%, with a marked difference between non-diabetic (16.6%) and diabetic (50%) patients. Conclusion: PTRAS can determine a MBP reduction, but a SC stabilization was recorded in non-diabetic patients only. Diabetic patients showed a worsening of the renal function and a restenosis risk three times higher than non-diabetic patients.

Purpose: Renovascular disease is reportedly the cause of end-stage renal failure in up to 20% of dialysis patients. This raises the question of whether more aggressive revascularization of renal artery stenosis (RAS) is necessary to prevent dialysis. Yet limited data is available about the most relevant issue in the discussion, i.e. the actual incidence of macrovascular RAS in dialysis patients. This study assessed the incidence of angiographically proven RAS. Materials & Methods: A cohort of consecutive dialysis patients registered over 12 years was studied. In each patient we recorded whenever in the medical history RAS of more than 50% was proven angiographically. Results: In 76 of 511 patients, renovascular disease was documented as the cause of dialysis (14.8%). In 69 of them angiography had been performed, which showed RAS in 26 patients. Additionally, RAS was found in nine patients who underwent angiography for peripheral vascular disease. The total incidence of angiographically proven RAS was therefore 6.8%. RAS was proven shortly before initiation of dialysis (median eight months, range 1-156). Baseline renal function at the time of angiography was poor (median serum creatinine: 245 micromol/L; range 98-410). Conclusion: A minority of dialysis patients due to renovascular disease had documented macrovascular RAS.

35.5.4

35.5.6

Five-year follow-up after renal angioplasty: clinical and functional outcomes H. Klingenstierna1, M. Annerstedt2, G. Jensen2, M. Delle1, L. Lonn1, H. Herlitz2, M. Aurell3, M. Hellstrom4; 1 Sahlgrenska University Hospital, Interventional Radiology, Goteborg, Sweden, 2Sahlgrenska University Hospital, Nephrology, Goteborg, Sweden, 3Sahlgrenska University Hospital, Dep of Nephrology, Goteborg, Sweden, 4Sahlgrenska University Hospital, Radiology, Goteborg, Sweden.

Endovascular repair of visceral artery aneurysms with stentgrafts P. P. Goffette1, P. F. Laterre2, X. Wittebolle2, J. F. Gigot3; 1Interventional Radiology, St-luc University Hospital, Brussels, Belgium, 2 Intensive Care, St-luc University Hospital, Brussels, Belgium, 3Digestive surgery, St-luc University Hospital, Brussels, Belgium.

Purpose: Long-term clinical and functional outcomes of renal angioplasty (PTRA). Materials & Methods: From 1994 to 1996, PTRA was performed in 107 patients; 59 of them (32 men, mean age: 64y, range: 26-81y) were available for a prospective five-year follow-up, with assessment of their life quality, blood pressure, renal function, DSA with intravascular pressure measurement (n=44) and MR-angiography (n=7). Seventeen (16%) patients had died and 31 (29%) were unwilling to participate. Results: At follow-up, 64% of patients reported a good physical, 78% a good mental, and 88% a good social well-being. Mean laboratory values before vs. five years after were: GFR: 64 vs. 66 ml/min; blood pressure 170/90 vs. 161/84 mm/Hg (54 patients had antihypertensive treatment). DSA showed a lumen reduction of >50% in 6/44 patients and >70% in 4/44. Six of them had significant pressure gradients. Four of the seven MR-angiographies showed a >50% restenosis; 41% of patients had suffered cardiovascular events (angina, MI, stroke) before or after PTRA. Conclusion: Over 2/3 of these patients reported good physical, mental and social well-being five years after PTRA. Their high patency rate (86%), blood pressure reduction, and preservation of kidney function indicate PTRA as an adequate treatment in selected patients.

Purpose: Assessing feasibility and clinical results of stentgraft implantation to exclude visceral artery aneurysms. Materials & Methods: Over two years, eight patients (median age, 53) underwent stentgraft placement as primary (7) or secondary (1) treatment of visceral arterial true (4) or false (4) aneurysms. Six saccular aneurysms (median diameter 24mm, range 9-70) arising from hepatic (2), superior mesenteric (2), celiac (1) or renal (1) arteries and two fusiform celiac aneurysms were included. Seven patients were symptomatic. In 4/7, stentgraft implantation was a rescue modality, to stop active bleeding (2) or to manage aneurysmal compression-induced mesenteric ischemia (2). Twelve balloon-expandable e-PTFE peripheral covered Jostents were initially implanted. Coil embolization of splenic and left gastric arteries was performed to avoid endoleak in fusiform aneurysms. Results: Initial success was achieved in six (75%) patients. In the remaining two, persistent aneurysmal perfusions due to too-short stent coverage or stent misplacement required additional stentgrafting. One proximal thrombosis and one distal embolus were easily managed by local lysis and thrombo-aspiration, respectively. During follow-up (median six months, 0.5-26), persistent exclusion of all aneurysms was observed. Conclusion: Endovascular repair of visceral aneuryms using stentgrafts seems safe and effective in both elective and emergent conditions. Longer-term follow-up is needed.

35.5.5

35.5.7

Effect of changing the definition of critical renal artery stenosis from 60 to 50% of lumen diameter reduction H. Eklöf1, H. Ahlström1, D. Bergqvist2, A. Boström-Ardin2, B. Andrén3, R. S. Nyman1; 1Radiology, ORKI, Uppsala, Sweden, 2Surgery, Uppsala, Sweden, 3Clinical physiology, Medical sciences, Uppsala, Sweden.

Treatment of peripheral aneurysms and pseudoaneurysms by stentgrafts D. Kuhelj, M. Šurlan, T. Kljucevsek, P. Berden, J. Klancar, T. Kunst, V. Salapura, P. Popovic, M. Stancovic; Clinical Radiology Institute, University Medical Center Ljubljana, Ljubljana, Slovenia.

Purpose: To evaluate the effect of changing the definition of critical renal artery stenosis from 60 to 50% of diameter reduction when screening for atherosclerotic renal artery stenoses (RAS). Materials & Methods: Three independent readers compared 3-D gadolinium-enhanced MR-angiography (MRA) with intraarterial pressure gradient measurement (PGM). From January 1996 to September 2000, 31 patients with a clinical suspicion of RAS were included in the study. Results: Fifty-seven main and seven accessory renal arteries were evaluated by both MRA and PGM. Each reader failed to define three accessory renal arteries. Six renal arteries were occluded and 23 had a stenosis with critical peak systolic gradient (15 mm Hg or more). Thirtytwo renal arteries had no critical gradient (<15 mm Hg). By changing the definition of critical stenosis from 60 to 50% of diameter reduction the average sensitivity improved from 85 to 92% and the specificity went down from 86 to 80%. Conclusion: When using MRA as a screening method, a high sensitivity is needed. A cut-off point defined at 50% of diameter stenosis is therefore recommended.

Purpose: To present our experience in treating peripheral aneurysms and pseudoaneurysms. Materials & Methods: From January 1997 to February 2003, 24 patients had stentgrafts implanted for 19 aneurysms and five pseudoaneurysms. Fourteen patients had aneurysms of the IA, three of the femoropopliteal artery, one of the brachial artery and one of the renal artery. Five patients had pseudoaneurysms, two of the IA and three of the FA. Three aneurysms were treated with Passager stentgrafts, two with Excluder stentgrafts. Two large aneurysms of the CIA were treated with specially made Dacron stentgrafts (Ella). The remaining cases were treated by Jostent peripheral stentgrafts. Follow-up continued with clinical examination and by Doppler US until January 31st 2003. Results: Stentgraft treatment was successful in all cases. In one case an additional stentgraft was needed due to a short Jostent stentgraft. Two SG in the femoropopliteal area kinked and became occluded, requiring surgery in one patient; in the other the SG reoccluded after additional selfexpandable stent placement. Conclusion: Peripheral intraarterial stentgrafting is an effective treatment for aneurysms and pseudoaneurysms except for those in the femoropopliteal area. Better results might be obtained with selfexpandable SG.

35.5.8

35.6.2

Endovascular treatment of life-threatening gastrointestinal bleeding due to giant superior mesenteric artery pseudoaneurysm using Monorail expanded PTFE-covered nitinol stents V. Pipitone, R. Gandini, G. Pendenza, D. Konda, S. Fabiano, C. A. Reale, G. Simonetti; Diagnostic Imaging and Interventional Radiology, University of "Tor Vergata", Roma, Italy.

Percutaneous stenting of lacrimal ducts in obstructive epiphora: our personal experience F. P. Florio1, W. Lauriola1, M. Nardella1, S. Balzano1, V. Strizzi1, F. Giordano2; 1 Vascular and Interventional Radiology, IRCCS Hospital "Casa Sollievo della Sofferenza", San Giovanni Rotondo, Italy, 2Oculistics, IRCCS Hospital "Casa Sollievo della Sofferenza", San Giovanni Rotondo, Italy.

Purpose: To evaluate the efficacy and safety of elective endovascular treatment of giant visceral artery pseudaneurysms in emergency conditions using Monorail expanded PTFE-covered nitinol stents. Materials & Methods: A 68-year old female patient presenting with gastro-intestinal (GI) bleeding (hematocrit 19.3%) and critical clinical conditions (ASA category IV) due to a superior mesenteric artery giant pseudoaneurysm (18.7 mm diameter) underwent emergency endovascular treatment. The supplying arterial fistula was sealed using a 5.0 mm diameter and 31 mm length Monorail expanded PTFE-covered nitinol stent (Symbiot; Boston Scientific) which was deployed in the superior mesenteric artery with complete exclusion of the pseudoaneurysm. Results: Postprocedurally the patient had gradual increase in hemoglobin and prompt improvement of clinical condition, and was discharged after 6 days. Control CT exams performed 1, 5, and 30 days, and 3 months after the procedure confirmed complete exclusion of the pseudoaneurysm and thrombosis of the sac. Conclusion: Endovascular treatment of giant visceral artery pseudaneurysms using Monorail expanded PTFE-covered nitinol stents is efficient, feasible and, owing to the Monorail system, less traumatic. Endovascular treatment should, therefore be considered in both emergency and non-emergency situations.

Purpose: Short- and mid-term evaluation of the efficacy of percutaneous stenting of lacrimal ducts in the treatment of obstructive epiphora. Materials & Methods: In 50 patients (18 men and 32 women, mean age 45 years, range 10-80) with long-lasting obstructive epiphora (with eight surgical relapses) 55 lacrimal stents were positioned percutaneously. After a mild sedation of the patients, the stents were positioned mainly using the Ho-Young Song's technique. Mean follow-up was 20 months (range 4-35 months). Results: Immediate technical success was 91%. At follow-up, 89% of cases showed a great clinical improvement of epiphora, with patency of the stents. Two cases showed a partial stent occlusion (with a partial improvement) and in other two cases the occlusion was total (at three and five months). One stent was accidentally removed and in another case the stent was removed for phlogosis. Conclusion: The treatment of obstructive epiphora by percutaneous stenting is easy to be performed, clinically effective, and with low complication rates.

35.6.1

35.6.3

Polyurethane stents in the treatment of lacrimal drainage system obstructions E. T. Ilgit, B. Önal, S. Özturk, O. Konuk, M. Unal; Department of Radiology, Gazi University, School of Medicine, Ankara, Turkey.

Stenting of lacrimal canal obstructions: a cost efficiency analysis F. Chiambaretta1, R. Mofid2, L. Gerbaud3, C. Privat2, H. Derumeaux4, J. Menerath1, J. Garcier2, L. Boyer2; 1Ophtalmology, CHU G Montpied, Clermont-Ferrand Cedex 1, France, 2Radiology B, CHU G Montpied, Clermont-Ferrand Cedex 1, France, 3Epidemiology, Hôtel-Dieu, Clermont-Ferrand Cedex 1, France, 4Epidemiology, Hôtel Dieu, Clermont-Ferrand Cedex 1, France.

Purpose: To describe the modified technique and assess the clinical efficacy of transluminal placement of home-made polyurethane Songtype stents in the treatment of lacrimal drainage system (LDS) obstructions. Materials & Methods: Fluoroscopically-guided polyurethane stent placement was attemped in 53 LDSs of 47 patients. The obstruction was complete in 34 LDSs and partial in 19. Stents were home-made from 6F polyurethane catheters; introducer sheats were used for the stent placement. Clinical examinations and saline irrigations of LDSs were performed at one week, one, three, six months and every six months thereafter. Dacryocystographic evaluations were performed at three or six months. Results: Stent placement was technically successful in 51 systems (98%). At one week, a complete resolution of epifora was observed in all stented patients. In a mean follow-up period of 18.9 months, 17 stented systems remained patent with no symptoms (32%). A periprocedural mild pain was the only complaint. Conclusion: The procedure--a simple and well-tolerated intervention-performed after an appropriate patients' selection, has a very high initial clinical success rate. Its high recurrence rate, however, necessitates secondary interventions and remains a problem. Home-made stents may also have a role on the treatment cost reduction.

Purpose: To compare three methods to treat naso-lacrimal canal obstructions. Materials & Methods: Thirty-two patients were treated in our hospital: 20 by endoscopic dacryocystorhinostomy (DCR Endos), five by dacryocystorhinostomy by external way (DCR Ext), and seven by percutaneous stenting. An average of the total cost per patient was calculated for all three methods and was then reported to the initial efficiency. Results: The average cost, using the French Sécurité Sociale financial rules, was up to 1366.80 Euros for stenting, up to 2001.53 for DCR Endos, and up to 2220.46 for DCR Ext. Analysis of variance gave significant results when comparing stenting with DCR Endos (p = 0.000007) and with DCR Ext (p = 0.02). Analysis of sensitivity concerning the success rate showed that stenting is potentially the most cost-effective and that a failure rate of 35.1% is rather acceptable in comparison with DCR Endos. If stenting is considered as an out-patient procedure, a failure rate of 50.0% is acceptable. Conclusion: If the failure rate stays under 35.1%, stenting may be considered as a short-term cost-efficient method.

35.6.4

35.6.6

Prospective evaluation of radiofrequency ablation of lung tumors: work in progress T. de Baere1, J. Palussiere2, A. Aupérin1, M. Kind1, V. Kuoch1, N. Teboune1, C. Dromain1, C. Le Hong1, P. Petrow1; 1 Institut Gustave Roussy, Villejuif, France, 2Institut Bergonié, Bordeaux, France.

Value of thoracocentesis for pneumothorax in CT-guided transthoracic needle biopsy J. E. Wildberger, L. Hülsmeier, T. Schmitz-Rode, R. W. Günther; Dept. of Diagnostic Radiology, University Hospital, Aachen, Germany.

Purpose: To define feasibility, safety, and local efficacy and to describe CT findings with radiofrequency (RF) ablation of lung tumors Materials & Methods: During one year, 36 patients aged 27 to 81 with inoperable lung tumors less than 4 cm, five or less in number, MEV superior to 1 l/sec, and no other tumor location, were treated under CT guidance with radiofrequency using 14G LeVeen needles and a RF 3000 generator with an algorithm dedicated to the lung. Follow-up included respiratory function testing at 30 days and CT 24 to 48 hours after RF and every two months Results: Six primary tumors and 57 metastases required one to six (mean=1.8) RF deliveries. Side effects were encountered in 45% of patients: pneumothorax, hemoptysis, alveolar hemorrhage, pneumopathy, atelectasia, and pleural effusion. Among the 27% with pneumothorax, only one required drainage. Mean tumor size increased from 19mm before radiofrequency, to 32mm immediately after, then 40mm early after RF, and decreased to 25, 23 and 18mm at 2, 4 and 9 months respectively. Only one of the 63 treated tumors recurred. Expiratory volume was not significantly modified after treatment Conclusion: RF ablation of lung tumors is feasible and safe. Local efficacy needs further investigation.

Purpose: To determine whether simple aspiration of air from the pleural space (thoracocentesis) obviates chest tube placement (CTP) in patients with large pneumothorax after transthoracic needle biopsy. Materials & Methods:During a four-year period, all 471 transthoracic needle biopsies performed under CT-control were retrospectively analysed (214 fine-needle aspiration biopsies; 257 core biopsies). Immediately after the intervention, control CT-scans were acquired; further follow-up was obtained with chest radiographs 2h and 4h after the procedure. Smaller pneumothoraces were followed clinically. In case of large pneumothorax (>30%), usually requiring CTP, percutaneous aspiration of the pneumothorax was performed. Thoracocentesis was performed by aspirating air from the pleural space using an 18-gauge intravenous catheter attached to a three-way stopcock and a 50-mL syringe. Results: Pneumothorax occured in 31.4% (148/471) of the punctures. In 82 cases further treatment was necessary, nine patients required immediate CTP. Thoracocentesis was performed in 73 patients. Finally, 64 of these patients did not require CTP, as near-complete resolution of the pneumothorax was documented on follow-up. The remaining nine patients (12.4%) had recurrence of pneumothorax, ultimately requiring CTP. Conclusion: Percutaneous catheter aspiration of a large biopsy-induced pneumothorax is safe and easy to perform and obviates CTP for the majority of patients.

35.6.5

35.6.7

Endobronchial stent placement in malignant or nonmalignant bronchial stenosis and/or bronchomalacia: results and complications B. S. Kapoor1, B. May2, J. Dunitz3, C. A. Dietz2, D. W. Hunter2; 1Medical Imaging, Royal University Hospital, Saskatoon, Canada, 2Radiology, University of Minnesota Medical School, Minneapolis, United States, 3Pulmonology, University of Minnesota Medical School, Minneapolis, United States.

Percutaneous mediastinal mass biopsy in children: techniques and outcomes A. M. Cahill1, C. R. Fitz2, R. D. Kaye1, R. B. Towbin1; 1Radiology, Children's Hospital of Philadelphia, Philadelphia, United States, 2Radiology, Children's Hospital of Pittsburgh, Pittsburgh, United States.

Purpose: From June 1992 to April 2002, 49 patients (25 male and 24 female, age 9-71 years) underwent stent placement for the treatment of bronchial stenoses and/or bronchomalacia. Materials & Methods: A total of 62 lesions were treated (26 right main stem bronchus, 7 bronchus intermedius, 17 left main stem bronchus and 12 trachea). Bronchoscopic indication for stenting was bronchial stenosis in 34 patients, bronchomalacia in 9 and both in 6. A total of 85 stents were used (10 Gianturco, 9 Palmaz, 23 Wallstent, and 43 Microvasive). Results: Significant improvement was noted in symptoms and pulmonary function tests following stent placement. Minimal self-limiting bleeding was noted during stent retrieval. 18 patients required repositioning and 13 required removal of the stents due to malpositioning or migration. Long-term complications included granulation tissue, infection, and migration, successfully treated with laser, antibiotics, repositioning or removal and replacement of stents respectively. Follow-up ranged from one month to 72 months. Eleven patients died by the end of the study due to the primary disease or unrelated causes. Conclusion: Endobronchial stent placement is a safe and effective procedure for bronchial stenosis and bronchomalacia, resulting in significant improvement in clinical symptoms and PFTs.

Purpose: To describe our techniques and results of mediastinal mass biopsy in children. Materials & Methods: Over a ten-year period, 31 mediastinal mass biopsies were performed on 31 children (16F, 15M), mean age 11.5 years, range 1.54-23 years. The indications included: malignancy (25), metastatic disease (4), infection (1), post-transplant lymphoproliferative disease (1). Seventeen biopsies were performed using a multiplepass core biopsy technique, ten biopsies using a coaxial core biopsy technique and four biopsies using an aspiration technique. Twenty-nine biopsies were performed with CT-guidance and two with ultrasoundguidance. Results: Twenty-six of 31 biopsies were diagnostic. Five specimens were non-diagnostic necessitating surgical biopsy. There were no major complications. A subclinical pneumothorax occurred in two patients which spontaneously resolved. Conclusion: Percutaneous mediastinal mass biopsy is a safe and accurate method for diagnosis in children. Percutaneous techniques obviate the need for thoracoscopic or open surgical biopsy in most children.

35.6.8

43.1.2

Allogenic islet transplantation by percutaneous trans-portal embolization in type-1 diabetes mellitus S. Terraz, T. Berney, F. Terrier, D. Didier, P. Morel, C. Becker; Diagnostic Imaging and Interventional Radiology, HUGE, Geneva, Switzerland.

European multicenter study on protected carotid stenting in high surgical-risk patients K. Mathias, E. Members of the Study Group; Radiologische Klinik, Klinikum Dortmund gGmbH, Dortmund, Germany.

Purpose: To evaluate efficacy and safety of allogenic islet transplantation (AIT) by a percutaneous trans-portal route. Materials & Methods: Thirty-one type-1 diabetic patients underwent percutaneous AIT in 54 sessions under conscious sedation. The right portal vein branch was punctured under US- and fluoroscopic-guidance, followed by portography. Pancreatic islets (5785±1870 islet equivalent/kg), previously isolated and purified, were administered by slow injection. Repeat portal manometry was performed. The percutaneous tract was embolized by gelatin sponge. The patients were then treated with an immunosuppressive regimen and closely followed by laboratory tests. Results: AIT was achieved in 53 cases (98%). There was no procedurerelated mortality. We observed two partial portal thrombi and three intraperitoneal bleedings in anticoagulated patients: one was treated by coil embolization and two underwent surgical hemostasis. Eight patients (27%) became totally insulin-independent, 13 (43%) kept a partial insulin secretion and nine patients (30%) lost all graft function. Of the nine last patients who received a new glucocorticoid-free immunosuppressive regimen, six (66%) became totally insulin-independent and three (33%) kept a partial insulin secretion. Conclusions: AIT can eliminate or decrease insulin requirements in the majority of type-1 diabetic patients. A particular care is mandatory in patients who require anticoagulation.

Purpose: To collect data on the clinical performance of the Wallstent™ for carotid artery stenting (CAS) associated with cerebral protection in the treatment of patients with carotid artery stenosis for whom preexisting conditions make surgery a technically difficult procedure or associated with a higher operative risk. Materials & Methods: This is a prospective, multi-center, single-arm study. Clinical follow-ups were performed after one, six, and 12 months post-procedure by Duplex ultrasound and NIH Stroke Scale. From September 2000, 161 symptomatic and asymptomatic patients with a de novo carotid stenosis or restenosis defined to be at high surgical risk were treated in 12 European centers. Primary endpoints were: perioperative stroke morbidity and mortality; secondary endpoints were: late stroke morbidity and mortality, ipsilateral stroke morbidity and mortality, perioperative minor neurologic events, technical success, angiographic success, and a restenosis of >60% lumen narrowing at six months. Results: A high technical success rate (99%) was achieved. At 30-day follow-ups, a combined morbidity and mortality rate of 7.6% was recorded: deaths 1.2%, strokes 5.8%, MI 0.6%. Conclusion: Initial results are consistent with recently reported results of CAS in high-risk patients. CAS has shown to be a viable option for treating these subgroup of patients.

43.1.1

43.1.3

Comparison of outcomes of unprotected and protected carotid stenting (NeuroShield™ filter) within a randomised trial usinga neuropsychometric test battery S. Macdonald1, T. J. Cleveland1, J. Graham2, P. D. Griffiths2, P. A. Gaines1; 1Sheffield Vascular Institute, Sheffield, United Kingdom, 2Academic Department of Radiology, Royal Hallamshire Hospital, Sheffield, United Kingdom.

Carotid stenting using the EPI-filter cerebral protection device in patients with a failed surgical attempt due to intraprocedural EEG abnormalities A. Spinelli1, R. Gandini1, C. A. Reale1, D. Konda1, C. Iani2, F. Arciprete2, G. Simonetti1; 1Diagnostic Imaging, University of Tor Vergata, Rome, Italy, 2UOC Neurology, Ospedale S.Eugenio, Rome, Italy.

Purpose: To use a sensitive means of assessing subtle neurological injury when comparing protected and unprotected carotid stenting. Materials & Methods: Thirty patients with symptomatic carotid artery stenosis =70% (NASCET) were randomised between the treatment limbs. The groups were comparable for age, sex, indication for treatment and lesion characteristics. A battery of 21 tests representing each cognitive domain and chosen for its sensitivity to unpredictable neuronal injury, subtle or overt was applied pre-stenting and at thirty-days post-stenting by an experienced neuropyschologist. The raw scores were compared with age-adjusted means for a large sample population and any deviation from baseline quantified. Parameters evaluated were: 1. Any deterioration (=1 and =2 standard deviations), 2. Deterioration =2 SDs, 3. Any improvement (=1 and =2 standard deviations), 4. Improvement =2 SDs. Results: Those who were significantly depressed (Beck Depression Inventory) were excluded from analysis. There was no significant difference between groups for each test for parameters 1, 2, and 4 above. There was a significant difference in the parameter "any improvement" (parameter 3, above) in the unprotected group (p=0.04, Fisher’s Exact Test). Conclusion: There is no significant difference in global cognitive function outcomes between unprotected and protected carotid stenting (NeuroShield™).

Purpose: Aim of this study is to assess the safety and efficacy of carotid stenting in high surgical risk patients using the EPI-FilterWire cerebral protection device. Materials & Methods: Out of a total of 28 patients treated with internal carotid artery stenting, nine were considered for endovascular treatment after a failed surgical attempt due to a significant depression of background activity on EEG in one or both hemispheres during clamping. The patients were treated using self-expandable monorail Carotid Wallstents under cerebral protection using the EPI-Filter device. Continuous EEG monitoring was performed during the procedure. Patients were followed with US performed seven days and three, six and 12 months after the procedure. Result: Technical success was obtained in 100% of cases. EEG monitoring showed no signs of cerebral dysfunction, and no complications occurred during the procedure. Post-procedural DSA revealed an optimal calibre of the treated artery and normality of the intracranial blood flow. At a mean follow-up of 7.4 months, all patients had patent stents. No neurological complications were reported during follow-up. Conclusion: Carotid stenting using self-expandable Carotid Wallstents and the EPI-Filter protection device is safe, efficient, and feasible in patients with EEG alterations during surgery.

43.1.4

43.1.6

Effective embolus diversion by the Diverter: a new device for preventing embolic strokes Y. Grad1, D. Harris1, D. Tanne2, B. Nishri1, O. Oz1, O. Yodfat1, Y. Assaf1; 1MindGuard Medical Devices Ltd., Caesarea, Israel, 2Stroke Unit, Dept. of Neurology, Sheba Medical Center, Tel Hashomer, Israel.

Protected carotid stenting: experience of the FilterWire™ (Boston Scientific) S. Macdonald1, T. J. Cleveland1, J. G. Moss2, M. Ablett3, D. Kessel4, D. Gould5, P. A. Gaines1; 1 Sheffield Vascular Institute, Sheffield, United Kingdom, 2Gartnavel General Hospital, Glasgow, United Kingdom, 3Ayr Hospital, Lanarkshire, United Kingdom, 4St James' Hospital, Leeds, United Kingdom, 5Royal Liverpool Hospital, Liverpool, United Kingdom.

Purpose: To test the efficacy of a new embolic stroke prevention device (the Diverter) to be implanted at the carotid bifurcation using a mock circulation and animal model. The Diverter is designed to divert emboli away from the internal into the external carotid artery (ECA). We hypothesized that emboli smaller than Diverter pores could be diverted. Materials & Methods: A compliant-transparent silicon model of the carotid bifurcation was used to investigate the capacity of the Diverter (450 micron pores) to redirect polystyrene particles sized 50-700 micron. A swine model was used to validate the in-vitro findings. After implantation of the Diverter between the right and bicarotid arteries, particles were injected and collected at both carotids using shunts with bleed valves. The weight of the diverted and non–diverted particles was compared. Results: Particles larger than pore size were fully diverted. WitAmonghin particles smaller than pore size, 98.4±0.9, 92.0±7.1 and 81.6±15.1% of the 350-420, 250-300 and 210-250 micron particles, respectively, were diverted into the right common carotid. Conclusions: The Diverter was highly effective for redirecting emboli smaller than its pore size. Our in-vitro and in-vivo data support the potential clinical use of this novel permanent protective device for prevention of embolic stroke.

Purpose: To report our experience on the safety and efficacy of the FilterWire™ (EX and EZ). Materials & Methods: Data were collected prospectively on 20 patients to 30-days. There were 14 males and six females, mean age 69 (range 54-78 years). Nineteen of them had symptomatic atherosclerotic lesions of the bifurcation/proximal internal carotid artery. The mean stenosis (NASCET criteria) was 82.5% (range 70-99%). Symptoms included eight TIAs, four non-disabling strokes, six amaurosis fugax and one hypoperfusion. All were treated within six months of onset of symptoms. The intention was to perform carotid artery stenting with cerebral protection using the FilterWire™. All were on optimal antiplatelet medication (including clopidogrel). Results: Technical success for filter placement was 20/20. Adjuvant manoeuvres to assist placement ("buddy wiring", pre-filter predilatation) were not required. Flow limiting spasm occurred in one and was treated by intra-arterial GTN and filter retrieval at completion. This was without clinical consequence. Events: There were no strokes or deaths to 30-days. Exclusions from analysis: one patient, due to incomplete antiplatelet therapy constituting a protocol violation. Conclusion: In appropriately selected patients this filter is technically feasible and will primarily cross lesions with =70% stenosis. We demonstrate no adverse events in this small series.

43.1.5

43.1.7

Stabilization of cellular proliferation into a steady state over the struts of the Diverter: a new technique for rerouting emboli to the external carotid artery Y. Assaf1, D. Harris1, D. Tanne2, B. Nishri1, D. Bushi1, O. Oz1, O. Yodfat1, Y. Grad1; 1MindGuard Medical Devices Ltd., Caesarea, Israel, 2Stroke Unit, Dept. of Neurology, Sheba Medical Center, Tel Hashomer, Israel.

Carotid primary stenting with cerebral protection: our experience in 158 consecutive cases N. Burdi1, M. Donatelli2, M. Di Palo1, M. C. Resta3, M. Resta1; 1Radiology, "SS. Annunziata" Hospital, Taranto, Italy, 2Radiology, Civil Hospital, Martina Franca, Italy, 3Institute of Radiology, University of Bari, Bari, Italy.

Objective: An innovative endovascular Diverter was designed to divert emboli away from the internal into the external carotid artery. Our aim was to test long-term patency and to determine cellular proliferation rates over the filtering wires. Methods: Fine-mesh Diverters (n=16) were implanted into swine illiofemoral bifurcations. After harvesting, the patency of the filtered ostium was calculated using computerized microscopy. For an additional 10 Diverters, animals were injected with BrdU 40mg/kg prior to sacrifice. Immunohistolabeling was performed at the filtering parts and proliferation rate calculated. Results: No occlusions or discernable stenoses were noted prior to sacrifice by angiography or ultrasound flowmetry. On morphometry, most of the filtered area of the specimens remained patent (99%, 89%, and 93% at 3, 10, and 17 weeks respectively). At the non-filtering portions, the Diverters were fully covered by a thin neointimal layer area after 3 weeks. The percent of BrdU labelled cells at the filtering portion was 18.7%, 12.8%, and 0.7% after 1, 3, and 23 weeks respectively. Conclusions: Implantation of an endovascular permanent arterial filtration diversion device in a swine model is feasible and the device remains patent. Our data supports the potential use of this emboli rerouting device (the Diverter) for stroke prevention.

Purpose: The availability of cerebral protection devices has made carotid primary stenting a valid alternative to surgical endarterectomy. Materials & Methods: From July 1999 to January 2003, 158/183 (in 140 patients) carotid stenting procedures were performed with cerebral protection. Since November 2001, 96 procedures have been recruited in the Italian Register for Carotid Stenting (RISC). We used 146 distal filters, 20 distal flow-block devices, two proximal and external flow-block devices. Stenoses (>70%) in both symptomatic and asymptomatic patients were treated. The patients were clinically evaluated by a neurologist the day before, during the procedure and immediately after. Follow-up was obtained with duplex scan at one, six and 12 months, and every year thereafter. Results: A technical success, with a residual stenosis <30%, was achieved in all cases. The 30-day major stroke/death rate was 1.9%. Cumulative primary patency rate was 99%. Twelve-month follow-up showed one case of intra-stent restenosis, treated with angioplasty. Conclusions: According to our experience, carotid stenting with cerebral protection revealed to be safe and feasible in the majority of patients and stenoses. Caution is however necessary in heavily calcified stenoses and in hard kinkings. Its low complication and high patency rates make it alternative to surgery.

43.1.8

43.2.1

A new concept for stent-protected carotid angioplasty: in-vitro and ex-vivo evaluations S. Müller-Hülsbeck, T. Jahnke, R. Wenke, H. Bolte, P. Stolzmann, C. Glass, M. Heller; Radiology, Universityhospital Schleswig-Holstein, Campus Kiel, Kiel, Germany.

Endovascular treatment of aortic pseudoaneurysms G. Carrafiello1, R. Caronno2, D. Laganà1, G. Piffaretti2, M. Carnini2, M. Canziani2, P. Castelli2; 1 Institute of Radiology, University of Insubria, Varese, Italy, 2Vascular Surgery, University of Insubria, Varese, Italy.

Purpose: Comparing the efficacy of a stent prototype for protected carotid angioplasty with cerebral filtration. Materials & Methods: In a bench-top model with human cadaver carotids (6) a nitinol stent (6/40 mm) coated with a membrane (Membrax™; porous size 80 µm) was implanted. Eight stents (8/32 mm) were then inserted in a silicon-made carotid bifurcation. When the stent was partially deployed, Contour™ particles (PVA) were injected separately (three sizes, 5 mg each [150-250 µm (n=10), 250-355 µm (n=10), 7101000 µm (n=10)]) into the internal carotid artery (ICA); the stent was finally deployed completely. Passed particles (mg) were determined in a 100-µm filter. Results: When stenting cadaveric carotids, the highest rate of particle embolization was measured while crossing the lesion (6.8 mg); embolization after additional balloon angioplasty was significantly lower as compared with particles measured in the effluent filter before stenting (3.8 mg; p<0.05). When using PVA, the added rate of passed emboli into the ICA after stenting measured 2.9%, for Emboshield™ 3.5% (n.s.) and for FilterWire EX™ 2.6% (n.s.), as compared with Angioguard™ (8%; p<0.001) and Trap™ (8.2%; p<0.001). Conclusion: Stent-protected carotid angioplasty with the Membrax™ proved to be a promising technique to reduce ex-vivo and in-vitro cerebral embolization.

Purpose: To describe safety and efficacy of primary endovascular repair of abdominal aortic pseudoaneurysms (AAPA). Materials & Methods: In the last year, AAPAs were repaired in five patients. Their mean age was 71 (range, 56-80), AAPA mean diameter was 65 mm (range, 33-90) and mean onset time was seven years (range, 1-10). Each patients underwent preoperative helical-CT with 3D reconstructions; preoperative angiographies were performed in iuxtarenal AAPAs. Bifurcated devices were implanted in four cases and aorto-uniliac grafts with crossover bypass in one. Each patient underwent spiral-CT follow-ups at one, three, six and 12 months after the procedure. Results: Endograft deployment was successful in 100% of cases; mortality and morbidity rates were 0. Average surgical time was 126 minutes (range, 80-195) and mean blood loss was 480 cc (range, 100-950). Patients were discharged after an average of five days (range, 3-11). Mean follow-up was nine months (range, 2-18). One iliac branch occlusion--successfully treated by thrombolisys--and two type-II endoleaks-one from the patent inferior mesenteric artery successfully excluded under US control by percutaneous injection of histoacryl and one from the lumbar artery by coil embolization--were recorded. Conclusion: Endovascular repair of AAPAs is a well-established option even in iuxtarenal AAPAs thanks to suprarenal devices.

43.1.9

43.2.2

Low versus high protection during carotid artery stenting H. K. Habboub1, S. A. Raouf1, M. A. Mufeed1, K. Hourani2, M. Qarioty3; 1Radiology, King Hussein Medical Center, Amman-Jordan, Jordan, 2Medicine, King Hussein Medical Center, Amman-Jordan, Jordan, 3Vascular surgery, King Hussein Medical Center, Amman-Jordan, Jordan.

Stent-graft treatment of mycotic aneurysms of the aorta and adjacent arteries: is it feasible and effective? S. J. Park1, D. Y. Lee2, J. Y. Won2; 1Diagnostic Radiology, Kang Dong Sung Shim Hospital, Hallym University, Seoul, Republic of Korea, 2Diagnostic Radiology, Severance Hospital, Yonsei University, College of Medicine, Seoul, Republic of Korea.

Purpose: Cerebral protection during carotid artery stenting seems to have potential to decrease incidence of thromboembolism and to reduce the rate of death from stroke. Methods & Materials: Carotid protection devices (CPD) were used in 32 patients who underwent carotid artery stenting for symptomatic carotid artery stenosis. The grade of stenosis ranged from 70-95%. The mean age of the patients was 68.5 years, and 80% were men. CPD were placed at the C1/C2 level (high) in 12 patients and at C2 or below in 20 patients. The 30-day incidence of stroke or death was recorded at follow up. Results: Carotid artery stenting was technically successful in 31 patients (97%). The thirty-day post-procedure combined stroke/death rate was 0/12 in patients with high protection (0%) and 1/19 (5.3%) with low protection. Conclusion: Carotid protection during carotid artery stenting appears to be associated with a lower incidence of strokes and death. A high position of protection device, when feasible, appears to be associated with a lower rate of complications.

Purpose: To evaluate the feasibility and effectiveness of stent-graft treatment in management of mycotic aneurysms in the aorta and adjacent arteries. Materials & Methods: Eight patients (five male and three female, age range from 30 to 85 years) with infected aneurysms who were treated with stent-grafting were included. Locations of mycotic aneurysms included the infra-renal and supra-renal abdominal aorta (n=5), the proximal descending thoracic aorta, the right innominate artery and the left common femoral artery. All patients were clinically considered to be at high surgical risk. The follow-up period ranged from 26 days to 53 months. Results: The overall technical success rate was 100%. Three patients (37.5%) died following stent-graft insertion, all due to infection; when stent-grafts were deployed there were indications of active infection in these patients. All five other patients are still alive; in none of them was there indication of active infection was present when the stent-graft was deployed. Conclusion: Despite a high initial technical success of endovascular stent-graft placement, mortality is high especially if infection is not adequately managed prior to stent-graft placement in treatment of mycotic aneurysms. Thorough evaluation and management of infection prior to stent-graft deployment are therefore mandatory.

43.2.3

43.2.5

Is initial aneurysm size predictive of outcome following endoluminal repair of infrarenal abdominal aortic aneurysm? D. Fay1, P. Davey2, L. Wilson2, M. Wyatt2, J. Rose1; 1Radiology, Freeman Hospital, Newcastle upon Tyne, United Kingdom, 2 Vascular Surgery, Freeman Hospital, Newcastle upon Tyne, United Kingdom.

The evolution of aneurysmal sac size after endovascular treatment with regard to the follow-up protocol M. Köcher1, P. Utikal2, E. Buriankova1, P. Bachleda2, M. Cerna1, Z. Kojecky2; 1 Department of Radiology, University Hospital, Olomouc, Czech Republic, 2Department of Surgery, University Hospital, Olomouc, Czech Republic.

Purpose: To determine whether pre-operative abdominal aortic aneurysm (AAA) diameter would correlate with differences in long-term outcome in patients undergoing endovascular aortic aneurysm repair (EVAR). Materials & Methods: One hundred seventy-one consecutive patients undergoing EVAR underwent abdominal CT pre-operatively and at 3, 6, 12, 24, and 36 months post-operatively. Patients were divided into those with an initial AAA size below (Group 1: n =82) and above (Group 2; n=89) 60mm. Cumulative intervention-free survival curves were generated using Kaplan-Meier estimations. Results: Median pre-operative aneurysm sizes were 55mm in Group 1 and 70mm in Group 2. There was no significant difference in 30-day mortality rates (Group 1, 3.6%; Group 2, 8.9%). Five patients in Group 1 developed an endoleak requiring intervention compared with 10 patients in Group 2 (p > 0.05). There was no significant difference in 3year intervention-free survival (Group 1, 0.83; Group 2, 0.70). Threeyear cumulative survival was better in Group 1 (0.76) than in Group 2 (0.69), p<0.05 (Wilcoxon). Conclusion: Late device-related complications following EVAR do not appear to be related to initial AAA diameter. The reduced survival of patients with large AAA does not appear to be related to device failure.

Purpose: To evaluate the evolution of aneurysmal sac size after endovascular treatment of abdominal aortic aneurysms with regard to the follow-up protocol. Materials & Methods: The follow-up protocol included CT and US at 3, 6, and 12 months after the procedure and yearly thereafter. In 85 patients the follow-up time was 12 months, in 51 patients 24 months, in 22 patients 36 months, and in 15 patients 48 months. Results: Shrinkage of the aneurysmal sac was found after one year in 51 % of patients, after two years in 67 %, after three years in 64 % and after four years in 68 %. Enlargement of the sac was found after one year in 5 % of patients, after two years in 15 %, after three years in 9 % and after four years in 13 %. No perigraft endoleak or aneurysmal sac rupture was found during follow-up. All retrograde endoleaks (n=9) were found during the first postopeative year after treatment. In only two of them was aneurysmal sac enlargement (3.5 % of all patients) seen. Conclusion: With regard to the development of the aneurysmal sac size and frequency of endoleaks, the first follow-up CT after 12 months is acceptable.

43.2.4

43.2.6

Suitability for endovascular aneurysm repair in a primary referral center: influence of experience, age, size, and gender P. Simons1, H. van Overhagen1, A. Nawijn1, B. Bruijninckx2, B. Knippenberg2; 1Radiology, Leyenburg Hospital, The Hague, Netherlands, 2Vascular Surgery, Leyenburg Hospital, The Hague, Netherlands.

Abdominal aortic aneurysm before and after stentgraft implantation: hemodynamic conditions simulated with computational fluid dynamics T. Frauenfelder, T. Schertler, B. Marincek, S. Wildermuth; Inst. Diagnostic Radiology, University Hospital Zurich, Zurich, Switzerland.

Purpose: To assess the suitability for endovascular aneurysm repair in an unselected patient population. Materials & Methods: During the period of February 1999 - May 2002 143 patients with non-emergency AAA were evaluated with contrastenhanced spiral CT. A panel of radiologists and vascular surgeons reviewed the clinical data as well as the vascular morphology and decided on the appropriateness of endovascular repair with the Zenith AAA Endovascular Graft. Results: Eighty-three (62%) patients had one or more contra-indications for endovascular repair. Unsuitability was most frequently (85%) related to the proximal neck morphology, with inadequate neck length the most common specific reason. Inadequate iliac morphology was the reason for unsuitability in 51% of patients. Unsuitability decreased from 67% during the first half of the study period to 55% during the second half. Unsuitability was more frequently seen in women (71%) than in men (61%). Age and aneurysm size did not differ between candidates and noncandidates. Conclusions: Fewer than half (38%) of the patients with AAA referred to a primary referral center are suitable for endovascular repair. Neck morphology is still the most frequent reason for rejection. Suitability is influenced by team experience and gender but not by age and aneurysm size.

Purpose: To evaluate the possibilities of computational fluid dynamics (CFD) in calculating pressures and flows in AAAs before and after stentgraft implantation based on multidetector CT-angiography (MDCT-A) data. Materials & Methods: Eleven patients with AAA undergoing MDCT-A before and after stent implantation were included. The aorta, including lumen, wall, thrombus, calcification, and the stent itself were segmented for generation of 3-D surface reconstructions of the aortic geometries. The patient-specific, anatomically accurate hemodynamics including pressure and pulsatile flow were simulated using a commercially available CFD-software and a standard PC after defining standardized inflow and structural values. Results: We present a feasible method to adapt and import complex aortic models based on MDCT-A into a commercially available software package for CFD. The simulations provide information about the distributions of blood velocities and wall pressures before and after stent implantation in AAA, including anchoring locations which often demonstrate high turbulences. Furthermore, the study shows a better flow in the before-minor-blood-supplied leg, as well as a more equal flow through both legs after intervention. Conclusion: CFD is a valuable tool to demonstrate the effect of stentgraft implantation and to predict possible complications; it is, therefore, a highly promising technique for the preinterventional planning.

43.2.7

43.2.9

Comparison of numerical and experimental flow simulation in patient-specific 3D-models of abdominal aortic aneurysm T. Frauenfelder1, E. Boutsianis2, Y. Ventikos2, B. Marincek1, S. Wildermuth1; 1 Institute of Diagnostic Radiology, University Hospital Zurich, Zurich, Switzerland, 2Ltnt, Swiss Federal Institute of Technology, Zuerich, Zurich, Switzerland.

Secondary interventions after endovascular aortic aneurysm repair (EVAR) at mid-term follow-up L. Hechelhammer1, M. Lachat2, N. Meier1, B. Marincek1, T. Pfammatter1; 1 Diagnostic Radiology, University Hospital, Zürich, Switzerland, 2 Vascular Surgery, University Hospital, Zürich, Switzerland.

Purpose: To validate numerical simulation using computational fluid dynamic (CFD) of a patient-specific abdominal aortic aneurysm (AAA) model using experimental studies of the same model for comparison. Materials & Methods: MDCT-A data of a patient with AAA were used for segmenting and generating a 3-D surface of aortic geometry. Based on the data of this model a 3-D silicon-model was built for experimental studies. Pulsatile flow was imitated with a computer-driven pump and measured with a synchronized high-speed laser camera. Commercially available CFD-software running on a standard PC was used to simulate patient-specific, anatomically accurate pulsatile flow corresponding to the experimental study. Results were compared with data received from the experimental study of the silicon model using the same input data. Results: The simulated flow pattern as well as the flow velocities of the numerical simulated blood flow using CFD correlated highly with the experimental studies. There were no significant discrepancies in the results in the areas adjacent to the ostiums of the visceral vessels due to lesser resolution in the numerical model. Conclusion: The high correlation of flow dynamics between CFD and the experimental study recommends use of CFD as a valuable tool for the functional evaluation of AAA.

Purpose: To report frequency and type of post-EVAR surgical/percutaneous interventions at mid-term follow-up. Materials & Methods: A single-center prospective database was used to assess a cohort of 332 patients (29 of whom had ruptured; mean age 68y; mean AAA diameter 6 cm) who had undergone EVAR between June 1997 and December 2002. Stent-grafts were: n=85 Vanguard II; n=192 Excluder; n=32 Zenith; n=20 Talent; n=3 Endologyx). Results: The mean follow-up was 32 months. There were 0.08 reinterventions per EVAR/year. Four patients had an AAA rupture associated with a Ia (2), a IIIa (1), or a IV (1) endoleak; 14 had limb thrombosis, all the others (36) were treated electively. Reinterventions were due to Ia (6) or Ib (11) endoleaks, type-II endoleaks in growing aneurysms (11), type-III endoleaks (14), type-IV endoleaks (3), stent-graft infection (1), symptomatic kinks (5), limb occlusion (17). Conversion to a surgical graft was performed in six patients, a thrombolysis in 15 limbs and thrombectomy or femoro-femoral bypass in two patients. Stent-graft extension or interposititon was performed in 37 patients and transluminal embolization in 16. Reintervention mortality was 1.9% (1/54). Conclusion: Although secondary interventions after EVAR are frequent, they can often be performed percutaneously and have a low mortality.

43.2.8

43.3.1

Endoluminal aortic aneurysm repair with the Zenith stent-graft J. Rose1, D. Fay1, P. Davey2, L. Wilson2, M. Wyatt2; 1Radiology, Freeman Hospital, Newcastle upon Tyne, United Kingdom, 2Vascular Surgery, Freeman Hospital, Newcastle upon Tyne, United Kingdom.

Uterine restoration after repeated sloughing of fibroids or vaginal expulsion following uterine artery embolization M. D. Kim1, W. K. Park2, J. H. Won3, D. Y. Lee4, J. H. Cho2, S. W. Youn1; 1Diagnostic Radiology, Pochon CHA University,Bundang CHA General Hospital, Sungnam city, Republic of Korea, 2Diagnostic Radiology, Yeungnam University, Daegu, Republic of Korea, 3Diagnostic Radiology, Ajou University, Suwon, Republic of Korea, 4Diagnostic Radiology, Yonsei University,Severance Hospital, Seoul, Republic of Korea.

Purpose: To evaluate the medium-term outcome following endovascular repair of abdominal aortic aneurysms with the Zenith device in a single center. Materials & Methods: Zenith devices were deployed electively in 105 patients (94 males; mean age 75 years, range 55–90) between September 1998 and January 2003. Peri-operative and delayed complications and secondary interventions were recorded prospectively. Cumulative survival, secondary endoleak-free survival and intervention-free survival curves were generated using Kaplan-Meier estimations. Results:13 aorto-monoiliac and 92 bifurcated systems were successfully inserted and deployed. 2 primary type I endoleaks related to poor proximal necks required early re-intervention. There were 14 peri-operative complications, including groin hematomas, renal artery occlusions, pseudo-aneurysms, and lymphoceles, and 3 early cardiac deaths (30day mortality: 2.9%). Nineteen late deaths occurred at a median of 7 months (1–36). The surviving patients have been followed for 21 months (range 1–52). Ten patients required late radiological or surgical re-intervention, and there has been one conversion to open repair. The estimated 4-year cumulative survival, endoleak-free survival, and intervention-free survival was 72%, 95% and 87% respectively. Conclusion: The Zenith stent graft has performed reliably over the medium term. Further data will be collected to confirm the longer-term results.

Purpose: To present our experience with uterine restoration after repeated sloughing of uterine fibroids or transvaginal expulsion following uterine artery embolization (UAE). Materials & Methods: Of 123 patients who underwent UAE, eight experienced uterine restoration following UAE. The mean size of their fibroids was 9 cm (3.5-18 cm) in diameter. Locations of uterine myomas were submucosal (n=7) and intramural (n=1). All patients underwent MRI before and after UAE. We reviewed clinical outcomes and MR imaging. Results: Between 12 days and five months later, uterine restorations occurred after UAE. One patient had incomplete transvaginal expulsion of a large myoma, and gentle abdominal compression was needed for complete expulsion of the myoma. Another patient became pregnant and delivered a full-term baby. All patients showed dramatic clinical response in menorrhagia, dysmenorrhea and bulk-related symptoms. On follow-up MRI, myomas were not seen or only a minimal scar was apparent. Conclusion: Repeated vaginal sloughing of fibroids or vaginal expulsion following UAE can eliminate submucosal myoma and, rarely, intramural myoma. Sloughing of fibroids or vaginal expulsion is to be considered as a continuation of treatment rather than a complication

43.3.2

43.3.4

Uterine fibroid embolisation (UFE): does choice of embolic material affect clinical and radiological outcome measures or complication rates? C. J. Johnston, S. M. Ford, J. McCann, E. Laffan, N. McEniff; Radiology, St James' Hospital, Dublin, Ireland.

Uterine fibroid embolization using calibrated microspheres F. Fanelli1, M. Bezzi2, F. M. Salvatori2, G. Marcelli2, S. Vagnarelli2, D. Pepino2, P. Rossi2, R. Passariello2; 1 Department of radiological sciences, University of rome, Rome, Italy, 2 Department of radiological sciences, University of Rome, Rome, Italy.

Purpose: To evaluate clinical outcomes, reduction in dominant uterine leiomyoma size and post-procedural complications in three patient groups undergoing UFE employing different embolic materials: polyvinyl alcohol particles (PVA) versus calibrated microspheres versus a combination of both. Materials & Methods: Fifty-five women were prospectively assessed over a 25-month period beginning January 2000. Indications for treatment were menorrhagia and uterine bulk-related symptoms. Embolisation was performed using PVA particles (21 patients), microspheres (27 patients) or both (seven patients). Ultrasound initially and then MRI was performed pre- and post-embolisation. Clinical evaluation was made at regular intervals up to one year after UFE. Results: Improvement in clinical symptoms and objective reduction in dominant leiomyoma size was similar across all three patient groups. Greater than 80% of patients reported symptom improvement: 18/22 patients in PVA group, 23/27 patients in microspheres group and 6/7 patients in combined group. Mean radiological reduction in dominant leiomyoma size in these groups was 49.2%, 47.1% and 53.8%, respectively. Complication rates for each group were comparable: one infective complication requiring hysterectomy occurred in the PVA only group. Conclusion: Both calibrated microspheres and PVA particles are effective embolic agents in UFE. Personal preference and cost will likely influence the choice of agent used.

Purpose: To evaluate the efficacy of uterine artery embolization using calibrated microspheres. Materials & Methods: Since July 2001, 23 patients underwent arterial embolization for symptomatic uterine fibroids. Indications were menorrhagia (n=16) or menorrhagia associated with bulk-related symptoms (n=7). Mean fibroid size was 5.5+/-2.4 cm (range 4-16 cm). Procedures were performed with bilateral femoral access using calibrated microspheres (Embosphere-Biosphere). Embospheres were injected in the right or left side with a flow-control injection technique. The procedural end-point was a “pruned tree” appearance of the uterine vasculature. Procedural pain was managed with continuous infusion therapy started 30 min before the procedure and continued for 24 hrs. Pre- and postprocedure fibroid evaluation was performed with MR imaging. Results: All procedures were technically successful. Mean follow-up was 8.7 months. Complete resolution of menorrhagia was observed in 23/23 patients. In one patient expulsion of a necrotic myoma was observed after 6 months. There was a significant reduction in uterine and dominant fibroid volume (average pre 55+/-25mm, post 20+/16mm). Two patients had transient and one permanent amenhorrea after embolization. No other complications were observed. Conclusion: Percutaneous uterine fibroid embolization seems to provide effective control of symptomatic disease with a high probability of significant fibroid size reduction.

43.3.3

43.3.5

Embolization of the uterus in post-partum bleeding R. Uflacker, J. B. Selby, C. Hannegan; Interventional Radiology, Medical University of South Carolina, Charleston, United States.

Diffusion and perfusion weighted MRI in the evaluation of leiomyomata tissue changes early after uterine aterial embolization P. Sipola1, Y. Liu1, P. Vainio1, M. Husso-Saastamoinen1, M. Hippeläinen2, R. Vanninen1, H. Manninen1; 1Clinical Radiology, Kuopio University Hospital, Kuopio, Finland, 2Obstetrics and Gynecology,, Kuopio University Hospital, Kuopio, Finland.

Purpose: To present experience with embolization in the treatment of post-partum uterine bleeding. Materials & Methods: Twenty-two patients, ranging in age from 20 to 34 years, with intra-abdominal/uterine bleeding, post-vaginal delivery (n=4) or C-section (n=18) were included in the review. The records and angiograms were reviewed. Two patients had embolization as the primary treatment, 13 patients had at least one surgical attempt to control the bleeding, before embolization, and the remaining patients had two or more surgeries. Five of the patients were unstable during the embolization. Blood transfusion was necessary in all patients, raging from two to 14 units. Transcatheter embolization was used in all patients, with coils and/or particles. Follow-up ranged from three months to nine years. Results: Control of the bleeding was achieved in all patients by embolization alone. No further surgery was necessary in any of the patients. Except for groin hematomas in two patients, no major complication developed from the procedure. Discharge was within three to six days of the procedure, in all patients. Conclusion: Transcatheter embolizaton is effective in the treatment of post-partum bleeding, with control of the hemorrhage and preservation of the uterus. Embolization allowed rapid recovery with no further surgical treatment necessary.

Purpose: To evaluate whether success of uterine arterial embolization (UAE) can be assessed early after the procedure using magnetic resonance imaging (MRI). Methods: Five symptomatic patients (age 45–63 years) underwent bilateral selective UAE with 550-700 mcg Embosphere particles. Anatomical sequences and diffusion and perfusion weighted MRI were obtained before and within 24 hours after UAE. Location and size of dominant leiomyomas were assessed. Signal intensity (SI) changes before and after UAE in diffusion weighted images were evaluated visually. SI versus time curves during first-pass of contrast agent from arterial input (aorta) and leiomyoma were calculated and residual perfusion was determined. Results: All dominant leiomyomas were intramural, with maximal diameter 7.9±1.3 cm (mean±SEM). The mean amount of injected embolization agent per patient was 9.3±3.4 ml (range 2-22 ml). In all leiomyomas, SI in diffusion-weighted images increased after UAE, reflecting cell edema. In one patient, high SI covered the area of leiomyoma only partially. The mean residual perfusion of leiomyomas was 2.4±1.3 % (range 0-7%). Conclusion: Diffusion and perfusion weighted imaging can detect changes indicative of cellular edema and extensive perfusion reduction in leiomyomas after UAE. The association between these early tissue changes and therapeutic response remains to be clarified.

43.3.6

43.3.8

Pathological findings two years after uterine embolization in sheep. Comparison of trisacryl-gelatin microspheres and non-spherical PVA particles J. P. Pelage1, M. Wassef2, J. Namur3, M. Bonneau4, J. Martal5, A. Laurent3; 1 Vascular Radiology Dept, A.Paré hospital, Boulogne, France, 2Pathology, Lariboisière hospital, Paris, France, 3Neuroradiology, Lariboisière hospital, Paris, France, 4Cr2i, APHP/INRA, Jouy en Josas, France, 5Physiology, INRA, Jouy en Josas, France.

Carbon dioxide as an alternative to iodinated contrast media for uterine fibroid embolization J. Golzarian, M. Laureys, F. Tannouri, A. Nazeri, J. Struyven; Radiology, Erasme hospital, Brussels, Belgium.

Purpose: To evaluate long-term efficacy of uterus embolization and particles location. Materials & Methods: Bilateral uterine artery embolization was performed in 39 sheep, in ten with PVA particles 600-1000 mum (PVA), in other ten with trisacryl microspheres 700-900 mum (MS), and in 19 controls. Animals were sacrificed after 26 months. Uterus and ovaries were fixed in formalin. HES- and orcein-stainings were used. Microscopical measurements included: vessel size, particle size, number of MS in each vessel, particle localization (a=uterine artery branches, b=periuterine, c=myometrium, d=endometrium). Results: No macroscopic lesions were recorded. The size of occluded vessels was larger (p<.0001) with PVA (1150±487, median 1140) than with MS (883±420, median 800). The mean number of particles was larger (p<.0001) with PVA (13.3±20.9, median 7) than with MS (2.5±2.8, median 2). PVA particles were located more proximally (p<.0001) (PVA a: 70.9%, b: 21.6%, c: 6.7%, d: 0.7%; MS a: 37.2%, b: 33.2%, c: 25.9%, d: 3.7%). Recanalization rate was 60 and 65% for MS and PVA, respectively (NS, Chi2). Conclusion: Macroscopic lesions were not recorded in the embolized groups. Both embolization agents mainly block in uterine and periuterine arteries; however, since PVA particles tend to form aggregates they occlude more proximally than microspheres.

Purpose: To evaluate feasibility and accuracy of carbon dioxide (CO2) injection for localizing uterine arteries (UAs) and guiding fibroid embolization. Materials & Methods: Fifteen consecutive patients underwent uterine fibroid embolization (UFE) using CO2. In all the patients an antero-posterior aorto-iliac arteriogram using a pigtail catheter (60cc of CO2/ 40 cc per sec) for the identification of UAs was carried out. Selective UA catheterization was performed using a 4-F Cobra catheter under CO2 guidance (30cc/15cc per sec); CO2 was used after UA catheterization during and at the end of embolization (20/10). Results: Carbon dioxide allowed the the depiction of UAs' origin in all the patients. Intra-uterine artery injection was associated with transient pain in nine patients; CO2 injections demonstrated well the fine uterine vascular structure, the anastomoses between UAs, and the contralateral UA in 12 patients. These findings were not reported by iodinated contrast agents. Carbon dioxide injections during embolization permitted to follow precisely the extent and the evolution of the embolization in 12 patients. Conclusion: Although CO2 injections in UAs can be painful, they represent a valuable alternative to iodinated contrast media. An accurate localization of UAs and the depiction of intra-uterine arterial structure and anastomoses was in fact possible.

43.3.7

43.4.1

Patient radiation exposure in uterine artery embolization (Study supported by the European Commission FIGM-CT 2000-00061) S. Vetter1, F. Schultz2, I. Kralj1, J. Zoetelief2, E. P. Strecker1; 1Interventional Radiology, Diakonissenkrankenhaus, Karlsruhe, Germany, 2Iri, Technical University, Delft, Netherlands.

Percutaneous treatment of hepatocellular carcinoma in patients with TIPS K. Tesdal1, C. Düber2; 1Radiology, Krankenhaus Friedrichshafen, Friedrichshafen, Germany, 2Radiology, Universitätsklinikum Mannheim, Mannheim, Germany.

Purpose: Evaluation of patient radiation exposure in uterine artery embolization (UAE) for fibroids and determination of organ doses and effective dose. Influence of radiographic techniques and radiologists' experience. Materials & Methods: Sixty-four consecutive patients (mean age 41y, mean body mass index 23 kg/m2) with symptomatic fibroids underwent UAE after a gynecological consultation. All patients underwent MRI including TOF-angiography of iliac arteries. For dosimetry radiographic and technical parameters were assessed. In five patients organ doses and effective doses were calculated using the Montecarlo method leading to a dose conversion factor (DCF) (mSv/Gycm2) to estimate effective doses of the procedure in general. During the series of procedures, optimization techniques like avoidance of oblique views and last image hold documentation instead of DSA were used. Results: Number of DSA images: mean 21 (0-89), median 10. Fluoroscopic time (min): mean 19 (2.5-42), median 17. Dose area product (Gycm2): mean 39 (3-318), median 26. DCF: 0.575 (mSv/Gycm2). Regarding the series of procedures, dose values decreased continuously: mean effective dose for patients 1-21: 39 mSv, patients 22-43: 19 mSv, patients 44-64: 6 mSv. Conclusion: Radiation exposure for UAE is at the level of diagnostic abdominal imaging. Radiation dose can be reduced considerably by dedicated radiographic techniques.

Purpose: To assess the role of transcatheter arterial chemoembolization (TACE) and percutaneous ethanol injection (PEI) in patients with hepatocellular carcinoma (HCC) and transjugular intrahepatic portosystemic shunt (TIPS). Materials & Methods: Five patients with HCC and TIPS were treated with either TACE (n=3) or TACE in combination with PEI (n=2). One patient had a known advanced, untreated HCC prior to TIPS. In four patients HCC was diagnosed eight, 51, 62 and 69 months after TIPS. TACE was performed using a mixture of 30-60 mg of epirubicin and 10 mL of Lipiodol. PEI was performed under CT-guidance. The mean followup after HCC treatment was 25.6 months (range, 7-41 months). Results: At follow-ups, all the patients were free of rebleeding. Two patients died seven and 38 months after one session of TACE and PEI (77 months after TIPS) and three sessions of TACE (91 months after TIPS), respectively. In the other patients, HCCs are well controlled 18, 24 and 41 months after repetitive percutaneous treatment. Shunt venograms revealed shunt dysfunctions in all patients due to stenosis (n=4) and occlusion (n=1), and were successfully reopened. Conclusion: TACE and PEI seem to be useful even in patients with TIPS, provided that liver function is good.

43.4.2

43.4.4

Histopathological evaluation of tissue extracted by the radiofrequency probe after liver tumor ablation. Preliminary findings C. T. Sofocleous; Radiology, Section of Vascular & Interventional, New Jersey Medical School UMDNJ-Newark, 150 Bergen Street UH C-320, United States.

A comparative study of changes in serum fibrosis indicators after transcatheter arterial chemoembolization with low-dose versus conventional-dose anticancer drugs in hepatocellular carcinoma W. Lu, Y. Li; Department of Interventional Radiology, NanFang Hospital, Guangzhou, China.

Purpose: To evaluate the characteristics of tissue extracted by the probe following hepatic malignant tumor radiofrequency ablation (RFA). Materials & Methods: The study was approved by the Institutional Review Board. From April to December 2001, 20 RFA sessions were performed in 18 patients with primary (16) and metastatic (2) liver masses. Track ablation according to device protocol was performed after each ablation. Tissue was extracted in all RFA probe passes. The specimens were collected and preserved in formalin. Results: Tissue was examined by the study pathologist. In 8/20 (45%) necrosis and coagulated tissue was present. In 5/20 (25%) "probably non-viable" tissue was extracted although some cell characteristics were identified. In 7/20 (35%) patients "possibly viable" tissue was found: five cases with hepatocellular carcinoma and two cirrhotic nodules. In one of the cirrhotic nodules, coagulation necrosis was also present. Conclusion: Histopathological evaluation of the tissue extracted by the RFA probe demonstrated that the expected coagulation effect is only present in less than 50% of the patients. Whether or not the pathologic exam has prognostic value is not yet known. We are therefore examining the clinical significance of these findings by continuing follow-up and adding more patients to the protocol.

Purpose: To study the changes in serum fibrosis indicators after transcatheter arterial chemoembolization (TACE) with low-dose versus conventional-dose anticancer drugs in hepatocellular carcinoma (HCC). Materials & Methods: Forty patients with HCC were divided into two groups to receive superselective TACE. Patients in group A (n=20) received low-dose anticancer drugs: mitomycin C (MMC) 2-8 mg, epirubicin (EPI) 5-10 mg, and carboplatin (CBP) 50-100 mg. Patients in group B (n=20) were given conventional-dose anticancer drugs (MMC 10 mg, EPI 40 mg, CBP 300 mg). Lipiodol anticancer drug emulsion was injected into the tumors feeding arteries followed by gelatin sponge or PVA particles embolization. Four serum fibrosis indicators (hyaluronate acid, HA), human procollagen type-III (hPC-III), laminin (LN), collagen type-IV (IVC)) were assessed before and seven days after TACE. Results: There were no significant differences between the two groups in the four indicators before TACE. After TACE, the concentrations of the four serum indicators had increased significantly in group B (p<0.05). The differences in the four indicators before and after TACE were not significant in group A (p>0.05). Conclusion: The formation of liver fibrosis after TACE in HCC was related to anticancer drugs dosage employed for chemoembolization. Low-dose anticancer drugs should be advocated.

43.4.3

43.4.5

A comparative study of damage to liver function after TACE with use of low-dose and conventional-dose anticancer drugs in hepatocellular carcinoma W. Lu; Dept. of Interventional Radiology, Nanfang Hospital, Guangzhou, China.

A comparative study of necrosis and apoptosis of hepatocellular carcinoma cells after transcatheter arterial chemoembolization with use of low-dose versus conventional-dose anticancer drugs Y. LI, W. Lu; Department of Radiology, NanFang Hospital, Guangzhou, China.

Purpose: To study liver function damage after transcatheter arterial chemoembolization (TACE) whith use of low-dose and conventionaldose anticancer drugs in patients with hepatocellular carcinoma (HCC). Materials & Methods: One hundred twelve patients with unresectable HCC were randomly divided into two groups to receive superselective TACE. Patients in group A (n=52) received low-dose anticancer drugs: mitomycin C (MMC) 2-8mg, epirubicin (EPI) 5-10mg and carboplatin (CBP)100mg were used. Patients in group B (n=60) were given conventional-dose of anticancer drugs (MMC 10mg, EPI 40mg, CBP 300mg). Liver function was evaluated with Child-Pugh scores, total bilirubin (TBIL), albumin (ALB) and alanine aminotransferase (ALT) before TACE, three days, one week and four weeks after procedures. Results: In both groups, TBIL, ALT, and Child-Pugh scores increased (p<0.001 or p<0.05) and ALB decreased (p<0.001 or p<0.01) three days and one week after TACE. The different between the parameters obtained four weeks after the procedure and baseline parameters was not siginificant in group A (p>0.05). In group B, however, significant difference(p<0.05) was found in all parameters except ALT. Conclusion: Superselective TACE with use of low-dose anticancer drugs induces transient impairment in liver function, but use of conventionaldose anticancer drugs can cause lasting, more serious worsening of liver function.

Purpose: To study necrosis and apoptosis of hepatocellular carcinoma (HCC) cells after transcatheter arterial chemoembolization (TACE) with use of low-dose and conventional-dose anticancer drugs. Materials & Methods: Thirty-four patients with HCC were randomly divided into two groups to receive superselective TACE. Patients in group A (n=16) received low-dose anticancer drugs: mitomycin C (MMC) 2-8mg, epirubicin (EPI) 5-10mg, carboplatin (CBP) 50-100mg. Patients in group B (n=18) were given conventional-dose anticancer drugs (MMC 10mg, EPI 40mg, CBP 300mg). Hepatic resection was performed subsequently. Twenty patients who were treated with resection alone (group C) were used as controls. The necrosis rates and apoptosis index of hepatocellular carcinoma cells were evaluated by pathologic examinations and TUNEL. Results: There was no significant difference in necrosis rates between groups A and B (p>0.05), while there was a statistically significant difference in necrosis rates between groups A and C (p<0.001) and between groups B and C (p<0.001). Apoptosis index in the three groups were 10.25±5.05%, 11.33±4.17%, and 4.29±2.02% respectively (group A vs. group C and group B vs. group C, p<0.005). No significant difference was observed between groups A and B. Conclusion: Superselective TACE with low-dose anticancer drugs was as efficacious for HCC as with conventional-dose drugs.

43.4.6

43.4.8

Intra-arterial hepatic oxalliplatinum through a percutaneously implanted port E. Elfassy, T. de Baere, V. Kuoch, C. Le Hong, V. Boige, C. Dromain, N. Teboune, M. Ducreux, D. Elias; Institut Gustave Roussy, Villejuif, France.

Arterial embolization of unresectable hepatocellular carcinoma with the use of microspheres, Lipiodol and cyanoacrylate T. Rand1, C. Loewe1, M. Schoder2, J. Kettenbach1, F. Lomoschitz1, M. Cejna1, J. Lammer1, T. Schmook1; 1 vienna university, dept of interventional radiology, Vienna, Austria, 2 Vienna University, Dept of Interventional Radiology, Vienna, Austria.

Purpose: To define feasibility, safety, and local efficacy of intra-arterial hepatic (IAH) oxaliplatinum through percutaneously implanted ports. Materials & Methods: Fifteen patients with liver metastases who had failed at least two types of systemic chemotherapy received 100mg of IAH oxaliplatinum bimonthly. A 4-French port/catheter was implanted through the right femoral artery after coil embolization of the extrahepatic arteries. The tips of 4 catheters were left free in the hepatic artery. Eleven catheters were inserted and blocked with coils in the GDA with a side hole placed in the common hepatic artery. Results: Catheter/ports could be implanted in all attempts, and were used every two weeks for 1-9 months. Chemotherapy was discontinued in one case due to catheter dysfunction. 2/4 free catheters dislodged from the hepatic artery, but one could be replaced. Gastric toxicity was encountered in 2/6 patients without coil occlusion of the right gastric artery (RGA), but resolved after RGA coil occlusion. 57% of evaluable patients responded to treatment, allowing liver resection or radiofrequency ablation in 30%. Conclusion: Percutaneous implantation of hepatic arterial ports is safe and effective in previously heavily pretreated patients. Blockage of the catheter in the GDA and embolization of the RGA seems to decrease complications.

Purpose: To assess the potential of transarterial permanent embolization (TAE) with use of microspheres, Lipiodol and cyanoacrylate. Materials & Methods: In a retrospective study, 40 patients with histologically proven hepatocellular carcinoma (HCC) were treated with TAE of the hepatic arteries. A mixture of microspheres (Embosphere; 100900µ) and ethiodized oil (Lipiodol Ultrafluide) was injected as an embolic agent to induce a permanent embolization. To inhibit the backflow of embolic agents cyanoacrylate (Glubran) was also injected. Cumulative survival rates were calculated. Results: Mortalities were not recorded during embolization or within the first 24 hours. Severe procedure-related complications occurred in two patients. At the time of analysis, 35 of the 40 patients were alive. The mean follow-up period was 5.9 months (range 0.25-19 months), with a median survival of 666 days. The 6-, 12-, and 18-month cumulative survival rates for all the patients were 80.6, 70.7, and 70.7%, respectively. Conclusion: TAE with the use of microspheres, Lipiodol and cyanoacylate for unresectable HCC is a feasible treatment modality. Our results indicate a potentially prolonged survival as compared with reports of other embolization techniques.

43.4.7

43.4.9

Different effectiveness of intraarterial hepatic chemotherapy between the right and the left lobe in case of a separated origin of the right and left hepatic arteries: can arterial ischemia of a single hepatic lobe contribute to a major therapeutic efficacy? M. Venturini, G. Crespi, E. Angeli, C. Losio, M. Salvioni, A. Del Maschio; Radiology, San Raffaele Hospital, Milan, Italy.

Chemoembolization in patients with liver metastases from neuroendocrine tumors R. D. Garcia Monaco1, H. G. Bertoni1, O. Peralta1, D. Cassina1, G. Pallotta2, M. Lopez Lincuez2, J. Lastiri2, M. Varela2; 1Interventional Radiology, Hospital Italiano, Buenos Aires, Argentina, 2Oncology, Hospital Italiano, Buenos Aires, Argentina.

Purpose: To evaluate the different effectiveness of intraarterial hepatic chemotherapy (IAHC) in patients with liver metastases from colorectal cancer between the right and the left hepatic lobe in presence of an anatomic variant: a separated origin of the right (RHA) and the left hepatic artery (LHA). Materials & Methods: After evaluation of metastases distribution, RHA/LHA calibers, and lobes dimensions a permanent catheter was percutaneosly placed in 27 RHAs and three LHAs. The proximal portion of the contralateral hepatic artery was embolized with Gianturco microcoils to obtain a distal vascularization through the collateral vessels arising from the catheterized artery. The different response of catheterized (group A) or embolized (group B) hepatic lobes was studied by spiral-CT, comparing the total area of the two major lesions for each lobe before and after three IAHC cycles (FUdR). Results: In 54.5% of cases, group B showed a better response, while a better response of group A was recorded in 18.1% of cases. The response was similar in 27.2% of cases. Conclusion: According to this preliminary study and in presence of an anatomic variant, arterial ischemia can be considered more effective than local chemotheraupeutic concentrations in the treatment of liver metastases from colorectal cancer.

Purpose: To evaluate response, survival, and toxicity in patients with liver metastases from neuroendocrine tumors (NET) treated with hepatic artery chemoembolization (CE). Materials & Methods: Between 1991 and 2002 133 CE procedures were performed in 49 patients with liver metastases from NET. Selective arterial transcatheter infusion with a lipiodol-doxorubicin solution was followed by selective embolization. Primary tumors included 35 carcinoids, four islet cell tumors, three gastrinomas, two vipomas, two glucagonomas, one mixed tumor, and two undetermined. Results: Surgical metastasectomy was performed in two patients. Twenty-one patients received systemic chemotherapy and 32 octeotride. In all patients with secretory tumors a humoral response was achieved, and 74% of them had a complete response. Symptomatic response was observed in 88% of all patients. Complete radiological response was seen in 19% of patients, partial response in 70%; in 11% no radiological response was observed. Median survival was 56.6 months from diagnosis and 31.46 months from treatment. One patient developed a hepatic abscess as a procedure-related complication and another developed a confusional syndrome. Conclusion: Hepatic artery CE is an effective and safe procedure for the treatment of unresectable, symptomatic liver metastases from NET, providing excellent symptomatic, humoral, and radiological responses.

43.5.1

43.5.3

Comparison of aortoiliac CT angiograms performed using 8- and 4channel MDCT M. Karçaaltýncaba1, D. W. Foley2; 1Department of Radiology, Hacettepe University School of Medicine, Ankara, Turkey, 2Department of radiology, Medical College of Wisconsin, Milwaukee, United States.

Accuracy of multirow CT angiography in the detection of hepatic artery stenosis after orthotopic liver transplantantion I. Bargellini1, R. Cioni1, L. Urbani2, P. Petruzzi1, C. Vignali1, F. Filipponi2, C. Bartolozzi1; 1 Diagnostic and Interventional Radiology, University of Pisa, Pisa, Italy, 2 Liver Transplant Unit, University of Pisa, Pisa, Italy.

Purpose: To compare performance parameters, clinical acqusition protocols, and contrast load of aortoiliac CT angiograms performed using 8- and 4-channel multidetector row CT (MDCT). Materials & Methods: Both the 8- and 4-channel systems utilized a matrix detector with 16 x 1.25 mm rows. Eighteen patients who had aortoiliac CT angiography with 8-channel MDCT and comparison studies with 4-channel MDCT were included in this study. Acqusition parameters for 8-channel aortoiliac CT angiography were image thickness 1.25 mm, pitch 1.35 or 1.65, table feed per rotation 19.3-33.5 mm/ s (dependent on scan rotation speed-0.5-0.7 sec). Results: 4-channel MDCT and 8-channel MDCT aortoiliac angiograms were performed with equivalent collimation. Mean contrast volumes used for 4- and 8-channel MDCT angiography were 144ml and 83ml respectively, for a 43% reduction of contrast dose in 8-channel studies. No image quality difference was noted despite decreased contrast dose. Mean aorta attenuation was 299 HU versus 307 HU for 4- and 8-channel MDCT angiograms respectively. Overall CT scanning with 8-channel MDCT was 2.2 times faster than 4-channel MDCT. Conclusion: CT angiography with 8-channel MDCT produces equivalent z-axis resolution with decreased contrast load and acquisition time when compared to 4-channel MDCT.

Purpose: To evaluate the accuracy of multidetector computed tomographic angiography (MDCTA) in detecting hepatic artery stenosis (HAS) after orthotopic liver transplantation (OLT). Materials & Methods: Twenty-one consecutive patients with OLT underwent MDCTA and DSA for evaluation of the hepatic arteries. Source images (Ax) were processed, obtaining MPRs, MIPs and VRs. Images were evaluated by two blinded radiologists to identify: a) arterial visualization (celiac axis, anastomosis, left (LHA) and right (RHA) hepatic arteries), b) detection of stenoses and c) grading of stenoses. Results: MDCTA was able to depict the celiac axis and anastomoses in all patients; the LHA and RHA were not clearly visualized in five cases with VRs and in one with the other modalities. All reconstruction modalities enabled correct diagnosis of celiac (n=3) and anastomotic lesions (n=14), as confirmed by DSA. Of six cases with LHA and RHA stenoses identified by DSA, four (66.7%) were visualized with Ax, MPRs, and VRs, and five (83.3%) with MIPs. Stenosis was overestimated in nine (47.4%) cases with VRs and in three (15.8%) with the other modalities. Conclusions: MDCTA enables accurate depiction of the arterial anatomy after OLT and detection of arterial stenosis; post-processing is required to increase diagnostic confidence.

43.5.2

43.5.4

Comparison of multislice spiral computed tomography and magnetic resonance imaging in the evaluation of living renal donors A. Chugtai1, D. Rix2, N. Soomro2, P. J. Haslam1; 1Radiology, Freeman Hospital, newcastle upon Tyne, United Kingdom, 2Urology, Freeman Hospital, newcastle upon Tyne, United Kingdom.

Radiation dose of 4-and 16-slice multi-detector spiral computed tomography angiography of the periferal run-off in comparison with digital subtraction angiography J. W. Kuiper1, T. Korteweg1, J. Geleijns2, P. M. Pattynama1; 1Radiology, Erasmus MC, Rotterdam, Netherlands, 2Radiology, Leiden Universitairy Medical Center, Leiden, Netherlands.

Purpose: To prospectively compare multislice computed tomographic (CT) angiography and gadolinium-enhanced magnetic resonance (MR) angiography for preoperative evaluation of laparoscopic live renal donors. Materials & Methods: Twenty patients underwent CT-angiography and MR-angiography prior to laparoscopic renal donation. Images were analysed by two radiologists on separate occasions. Surgical correlation was provided in 19 kidneys at laparoscopic nephrectomy. Results: CT demonstrated eight accessory renal arteries in eight patients. MR identified six accessory arteries. CT recognised six large left lumbar veins draining into the left renal vein, 15 gonadal veins, one retroaortic accessory lower pole renal vein and 13 adrenal veins. MR demonstrated large left lumbar veins draining into the left renal vein in only two patients, the accessory lower pole renal vein was confirmed, the gonadal vein was seen in five patients and the adrenal vein in one. Laparoscopic nephrectomy showed highest concordance with CT-angiographic findings. Conclusion: CT-angiography provides superior contrast and spatial resolution compared with MR-angiography. Variations in venous anatomy are better demonstrated with CT. Small accessory renal arteries may be missed with MR-angiography. This risk must be balanced with the additional radiation burden from CT.

Purpose: Multi-detector spiral CT (MDCT) angiography using 4- and 16slice scanners was compared with catheter-based digital subtraction angiography (DSA) for peripheral run-off from the aorta to the ankles. This study evaluates the dosimetric aspects of these techniques. Materials & Methods: The effective dose of MDCT-angiography was calculated in 65 consecutive patients: 48/65 were scanned by a 4detector CT scanner (Siemens Volume Zoom) and 17/65 by a 16-detector CT scanner (Siemens Sensation 16). DSA was carried out in 53 patients with a Philips Integris V-3000 DSA system. All studies were performed according to a standardized protocol. Radiation doses were calculated by CT dose index (CTDI) for MDCT and by dose-area product (DAP) meter for DSA, converted to a standard patient and expressed as effective dose in millisievert (mSv). Results: The mean effective dose for MDCT-angiography with the 4slice MDCT was 8.3 ± 1.1 mSv; with the 16-slice MDCT it was 9.6 ± 1.7 mSv. For DSA it was 6.7 ± 3.3 mSv (mean ± SD) Conclusion: The effective dose of MDCT-angiography for the peripheral run-off is significantly higher (8.3/9.6 mSv; p=0.002) than DSA, but is acceptable for a non-invasive diagnostic test that can replace DSA.

43.5.5

43.5.7

Diagnostic efficacy of peripheral CT-angiography as compared with DSA: a work in progress F. Lomoschitz1, M. Cejna1, A. Koechl1, D. Sandner1, I. Katzler1, A. Kanitsar2, J. Lammer1, D. Fleischmann1; 1 Radiology, University of Vienna, Vienna, Austria, 2Computergraphics, University of Vienna, Vienna, Austria.

Right renal artery as a marker of renal veins’ position: CT study J. Namyslowski1, A. M. Aisen1, J. A. Kaufman2; 1Radiology, Indiana University, Indianapolis, United States, 2Dotter Interventional Institute, OHSU, Portland, United States.

Purpose: To assess the diagnostic efficacy of peripheral CT-angiography (CTA) as compared with digital subtraction angiography (DSA) in the evaluation of peripheral arterial obstructive disease (PAOD). Materials & Methods: Twenty-five patients with PAOD (Fontaine IIb – IV) underwent lower extremity CT-angiography (collimation, 4x2.5 mm; feed, 4 mm per rotation; 3 mm section thickness, 1 mm reconstruction interval). Standard post-processing image included semi-automated bone removal, generation of maximum intensity projections (MIP) and rotating multi-path curved planar reformations (rmpCPR) which were generated by semi-automated vessel-tracking through the peripheral arterial tree (one aortic start point, six peripheral end points) and the subsequent automated composition of 21 views which simultaneously display curved planar reformations through all of the multiple paths over a range of 180°. CTA images were compared with DSA by two independent readers. Lesions were graded in four categories from 0-3 (fully patent, <50% stenoses, 50–99% stenoses, occlusion) for each arterial segment. Results: CTA depicted all vessel segments demonstrated by DSA. Overall image quality for CTA was ideal. Overall sensitivity/specificity/accuracy for CTA was 98/94/95% when MIP and rmpCPRs projections were used additionally to axial plane images. Conclusion: Peripheral CTA has the potential to replace DSA for the evaluation of PAOD.

Purpose: The right renal artery (RRA) is considered a reliable indicator of renal vein location during bedside US-guided caval filter placement. To test this we evaluated the distance between RRA and renal veins on CT scans. Materials & Methods: Contrast-enhanced abdominal CTs were analyzed. Table position was recorded at the level of the vessels of interest. The distances between the arteries and veins were calculated. Note was made of any anatomic vascular variants. Results: In 100 patients, the RRA-RRV distance varied from 0 to 31.5 mm. In 35% it was >10 mm; in 4% it was >20 mm. In 17% RRV inserted =10 mm cranial to the RRA; in 22% it inserted =10 mm caudal. The RRA-left renal vein (LRV) distance varied from 0 to 70.5 mm. In 42% it was >10 mm; in 10% it was >20 mm. In 43% LRV inserted =10 mm cranial to the RRA; in 9% it inserted =10 mm caudal. Venous vascular variants were identified in 13%; arterial variants in 16%. Of the 100 patients, 8% had circumaortic LRV. Conclusion: The anatomic relationship between the RRA and the renal veins is variable. This could have implications on the precision of USguided caval filter placements.

43.5.6

43.5.8

Contrast-enhanced magnetic resonance angiography in critical limb ischemia associated with diabetes M. Lapeyre1, P. Desgrange2, A. Luciani1, A. Rahmouni1, H. Kobeiter1; 1Imagerie medicale, Henri Mondor Hospital, Creteil, France, 2Chirurgie Vasculaire, Henri Mondor Hospital, Creteil, France.

Carbon dioxide in the angiographic detection of arterial hemorrhage A. Krajina, M. Lojik, J. Raupach, V. Chovanec; Department of Radiology, University Hospital, Hradec Kralove, Czech Republic.

Purpose: To evaluate the accuracy of dual-acquisition contrastenhanced magnetic resonance angiography (CE-MRA) in critical limb ischemia associated with diabetes. Materials & Methods: From May 2001 to May 2002, 16 consecutive patients with diabetic critical limb ischemia underwent stepping-table CE-MRA, beginning with calf acquisition. All patients had angiographic correlation. Lower extremity vessels were divided into 13 anatomic segments for review. Two blinded observers, a vascular surgeon (VS) and an interventional radiologist (IR), assessed maximum-intensity projection images and planned treatment. Disease per segment was graded as normal or non-significant stenosis, significant stenosis, or thrombosis. Digital substraction angiograms (DSA) were assessed similarly but separately . Results: One-hundred ninety segments were assessed. When compared to DSA, sensibility of MRA was 94% for stenosis and 98 % for thrombosis when reviewed by VS and 92% and 93.2% respectively by IR. Specificity was 95% for stenosis and 99% for thrombosis when reviewed by VS and 95% and 98% respectively when reviewed by IR. MRA led to four additional treatments for IR and five additional treatments for VS. Conclusion: Dual-acquisition CE-MRA may be accurate enough to replace DSA for critical limb ischemia in patients with diabetes and may unmask patent distal vessels reliably enough for treatment.

Purpose: To determinate the diagnostic efficacy of carbon dioxide (CO2) as an intraarterial contrast agent in the detection of arterial hemorrhage. Methods & Materials: Over a 36-month period 31 patients referred for angiography to detect arterial extravasation underwent CO2 angiography only when standard iodinated contrast medium (ICM) angiogram was indeterminate or negative. The CO2 digital subtraction arteriograms were performed using manual injection with a 60 ml CO2-filled syringe. Nineteen patients were referred because of arterial gastrointestinal bleeding, two patients with bleeding due to necrotizing pancreatitis, eight patients with liver trauma, and two patients with pelvic trauma. Results: Among 31 patients studied for arterial bleeding, primary ICM angiogram revealed extravasation in seven, and findings were assessed as dubious in another two. CO2 angiogram was performed in the remaining 24 patients and revealed clear-cut evidence of extravasation in another five patients. Conclusion: CO2 angiography can detect arterial extravasation when findings of routine arteriography are dubious or negative. The CO2 angiograms were unable to define detailed vascular anatomy.

43.5.9

43.6.2

Carbon dioxide angiography of lower limb arteries: results from a prospective, randomized comparison with iodinated contrast agents M. Kalinowski, A. Schmitt, H. Alfke, K. J. Klose, H. J. Wagner; Dep for Diagnostic Radiology, Philipps University Hospital, Marburg, Germany.

Is volume rendered image display feasible for three-dimensional (3D) US-guidance of interventional radiology procedures? S. C. Rose, T. R. Nelson; Radiology 8756, University of California, San Diego, San Diego, United States.

Purpose: Prospective, randomized comparison of carbon dioxide and non-ionic iodinated contrast material for intraarterial DSA of lower limb arteries. Materials & Methods: 102 patients were randomized to undergo DSA of the abdominal aorta and lower limb arteries with CO2 (n=51) or iopamidol(n=51). The primary outcome measure was diagnostic accuracy. Secondary outcome measures were diagnostic quality and safety. Angiograms were interpretated independently by two blinded angiographers. Results: In the iodine group 99% of the angiograms were adequate for diagnosis, whereas only 82% in the CO2 group achieved this category. Of the 357 angiograms obtained with iodine 83% were scored good or excellent, but only 10% of the 261 CO2 studies were good and none excellent. Below the knee CO2 did not achieve adequate images in 25% of the cases. No serious adverse events occurred with either contrast agent. Patients in the CO2 group reported significantly more pain. Conclusion: CO2 is a safe contrast agent for lower limb arteriography. However, its diagnostic performance was inferior to conventional iodinated contrast agents. In the popliteal and infrapopliteal regions CO2 did not achieve diagnostically adequate angiograms in 25% of cases.CO2 may be used as an adjunct, but not as a substitute for iodinated contrast material.

Purpose: To assess the conspicuity of various targets and devices using volume rendered (VR) display of 3-D US. If feasible, VR image display may be more intuitive for the operator than conventional multiplanar reformatted (MPR) display. Materials & Methods: Targets of petroleum gel with varying amounts of corn starch were implanted into fresh whole beef liver. A biopsy needle and a radiofrequency probe were introduced into the liver. Threedimensional US volume data sets were acquired over each target and device. Conspicuity was judged using a five-point visibility rating scale. The MPR image was held as the reference standard. VR image conspicuity was judged for combinations of seven different post processing rendering display parameters. Results: Of 351 possible combinations of target, devices, and various rendering mode pairings, VR caused conspicuity improvement, no change or degradation in 0, 3 (1%), and 348 (99%), respectively, relative to MPR. VR images were judged to be clinically useful, equivocal, or useless in 12 (3%), 14 (4%), and 325 (93%), respectively. Conclusion: VR display causes significant degradation of target and device conspicuity, and rarely produces clinically useful images in the liver. MPR is superior to VR for 3-D US-guidance of interventional procedures in similar acoustic environments.

43.6.1

43.6.3

Correlated live ultrasound-CT multiplanar reconstruction-guided interventions: a phantom study G. Bartal1, A. Belenky2, E. Brem3, A. Bahcebasi1, J. M. Gomori4; 1Diagnostic Imaging and Interventional Radiology, Hillel Yaffe Medical Center, Hadera, Israel, 2Angiography Unit, Rabin Medical Center, Petach-Tikva, Israel, 3Ultraguide Ltd., Yoqne'am, Israel, 4MRI Unit, Hadassah Medical Center, Jerusalem, Israel.

Is there a role for virtual reality in the interventional lab? R. Shoenfeld; Radiology, Newark Beth Israel Medical Center, Newark, New Jersey, United States.

Purpose: Evaluation of real-time co-planar multiplanar reconstructions (MPR's) based on pre-acquired CT data combined with live ultrasound for free-hand guided needle placement outside the CT suite. Materials & Methods: The TwinGuide system (UltraGuide®, Israel) provides online correlation between co-planar real-time MPR's of preacquired CT data and live ultrasound for interventional procedures. We scanned a Triple Modality 3D Abdominal Phantom with an attached magnetic localizer under standard spiral CT abdominal protocol. Needle placement was performed outside the CT suite with additional sensors attached to the 18G Chiba needle and the 3.5 MHz Ultrasound transducer. Four radiology residents selected targets in various organs of the phantom for needle placements. Needle positions and trajectories were simultaneously displayed over live ultrasound images and updated CTMPR's. Results: All targets were reached by live ultrasound assisted by the coplanar real-time CTMPR's. Insertion time required 5-60 seconds. In some cases more than one pass was necessary. Conclusion: Fusion of live ultrasound with real-time coaxial CTMPR's is feasible for abdominal target placement. Initial in-vitro results indicate that even inexperienced users can use this tool effectively outside the CT suite. TwinGuide can save expensive CT suite time and significantly reduce intra-procedural physician and patient exposure. Clinical evaluation is underway.

Purpose: Workstations capable of manipulating large datasets in realtime to render high-resolution 3-D images are available. Is there a practical use for them? Can they be integrated into the daily workflow in a meaningful way? Materials & Methods: Our interventional lab consists of a Siemens rotational angio-system with a Leonardo workstation calibrated to the C-arm. Unsubtracted or subtracted contrast-filled structures are imaged at isocenter over a 200-degree arc in five seconds. Contrast studies require a seven-second injection beginning two seconds before C-arm rotation commences. Volumes and their contents can be freely manipulated in space. Pixel size of 0.266 mm yields images of far higher resolution than CT or MR. Results: Studies were performed for 3D/quantitative definition of complex fractures/malformations, biliary anatomy/strictures, vertebral anatomy for vertebroplasty/intervention, staghorn calculi, tracheobronchial anatomy and stents, vascular anatomy, abnormalities, flow patterns/ collateral pathways. Conclusion: Virtual reality images help us better understand anatomic relationships, perform quantitative/qualitative analysis, minimize runs, views and contrast used, define optimal routes of access for intervention, assess adequacy of intervention and reduce procedure time. Tableside controls facilitate process integration and improve workflow.

43.6.4

43.6.6

Optimization of volume rendered three-dimensional ultrasound (3D US) post processing variables to improve conspicuity of targets and devices for interventional applications. Phantom models S. C. Rose, T. R. Nelson; Radiology 8756, University of California, San Diego, San Diego, United States.

Long-term results of the new PolyphosphazeneF nanocoat to reduce thrombogenicity and late restenosis of nitinol stents in a pig renal artery model G. M. Richter1, F. Thomas2, L. Grenacher1, P. Christoph1, M. Henn1, S. Satzl1, P. Kurz1, M. Grunze3; 1 Diagnostic Radiology, University Hospital Heidelberg, 69120 Heidelberg, Germany, 2Vascular Surgery, University Hospital Heidelberg, 69120 Heidelberg, Germany, 3Physical Chemistry, University Heidelberg, 69120 Heidelberg, Germany.

Purpose: To determine which post-processing display parameters yield greatest target or device conspicuity when using 3-D US volume rendered (VR) imaging display. Materials & Methods: Three US phantoms were made using a petroleum gel. Corn starch was used to vary echotexture. Each phantom had hypoechoic, isoechoic, and hyperechoic gel targets imbedded. A biopsy needle and a radiofrequency ablation probe were introduced into the background gel. Identical 3-D US volume data sets were acquired over each target/background and device/background combination. Data sets were displayed with combinations of post-processing display rendering modes. Target and device conspicuity was judged with a five-point visibility rating scale. Similar measurements were performed with beef liver as background. Results: Conspicuity of targets and devices varied widely. When background and target echotexture were similar, conspicuity was poor. When targets were less echogenic than the background, the transparent minimum parameter was best. When targets were more echogenic than the background or when imaging devices, transparent maximum parameter was best. Other rendering modes obscured all combinations of targets, background, and devices. Conclusion: When using 3-D US imaging, post-processing parameters selected dramatically affects target and device conspicuity in a relatively predictable fashion.

Purpose: To demonstrate the efficacy of a new nanocoating (Poly-TriFluorethoxy-Polyphosphazene) for nitinol stents to reduce thrombogenicity and restenosis in a pig renal artery model. Materials & Methods: Fifteen minipigs (25 kg) were divided into three groups (one-, eight-, and 20-week follow-up) with five animals/group. The new ultrathin (<50 nanom.) coat was applied on nitinol stents (5/20 mm, Optimed,) using a proprietary technique. Coated stents were randomly assigned to either the left or the right side. A non-coated stent was then placed contralaterally. Implantation was performed via a transfemoral approach applying heparin anticoagulation (150 U/Kg). Before necropsy, a magnification angiography was performed. Explanted stents were examined by electron scanning and light microscopy. Results: Coated stents showed no thrombus, the average intimal thickness measured 81µ ± 51 at four weeks and 78µ ± 50. In non-coated stents thrombus deposition occurred in three. A nearly total occlusion was found in two. Average intimal thickness measured 357µ ± 111 at four wks and 462µ ± 47 at 12 weeks (statistical significancy at all intervals, p< 0.01). Conclusions: The new coating provided an excellent efficacy in reducing intimal hyperplasia in a renal artery pig model. Regular nitinol stents cannot be recommended in human renal arteries.

43.6.5

43.6.7

Initial long-term results of the new PolyphosphazeneF nanocoat to reduce thrombogenicity and late restenosis of balloon-expandable stents in a pig renal artery model G. M. Richter1, F. Thomas2, L. Grenacher1, S. Satzl1, M. Henn1, P. Christoph1, P. Kurz1, M. Grunze3; 1Diagnostic Radiology, University Hospital Heidelberg, Heidelberg, Germany, 2Vascular Surgery, University Hospital Heidelberg, Heidelberg, Germany, 3Physical Chemistry, University Heidelberg, Heidelberg, Germany.

Intra-arterial prostaglandin E1 infusion in patients with rest pain: early results and scintigraphic correlation A. Chatziioannou1, C. Ladopoulos1, K. Katsenis2, D. Mourikis1, M. Glynos1, P. Dimakakos2, L. Vlachos1; 1Interventional Radiology, University of Athens - Areteion Hospital, Athens, Greece, 2Vascular Surgery, University of Athens - Areteion Hospital, Athens, Greece.

Purpose: To show the results of a new nanocoating (Poly-Tri-Fluorethoxy-Polyphosphazene) for vascular stents to reduce thrombogenicity and late restenosis in a pig renal artery model. Materials & Methods: Fifteen minipigs (25 kg) were divided into three groups (one-, eight-, and 20-week follow-up intervals) with five animals per group. The ultrathin (50 nanom.) coating was applied on balloon-expandable stents (5/20mm, 80 µ strut thickness) using a proprietary coating process. Coated stents were randomly assigned to either the left or right renal artery. A non-coated stent was placed contralaterally to serve as control. Implantation was performed via a transfemoral approach using 5-mm balloons and applying heparin anticoagulation (150 U/Kg). Before necropsy, a magnification angiography was performed. Explanted stents were examined by electron scanning and light microscopy. Results: Coated stents showed no early or late thrombosis; average intimal thickness measured 77µ ± 31 at four and 69µ ± 40 at 12 weeks. In the non-coated stent group, a thrombus deposition occurred in two. Average intimal thickness measured 157µ ± 51 at four and 162µ ± 47 at 12 weeks (p < 0.01). Conclusions: The new coating provided and excellent efficacy to reduce early thrombogenicity and intimal hyperplasia in a renal artery pig model.

Purpose: To present our results after short-term intra-arterial infusion therapy of PGE1-alprostadin via a port-system implanted in patients with severe rest pain. Materials & Methods: Fourteen patients with severe night rest pain and below-the-knee extensive peripheral vascular disease were included in this study. They received intra-arterial infusions of 20 mg PGE1–alprostadin daily via a port-catheter system for one month. The tip of the catheter was introduced with a retrograde puncture in the ipsilateral external iliac artery. Flow measurements of the distal extremity were performed with 99m Tc-pertechnetate immediately before and after the completion of the treatment. Results: Hours of continuous sleep and flow improvements evident in the scintigraphic studies were evaluated. A significant decrease of rest pain was observed in eight patients (57.1%) and an adequate reduction in four (28.6%); in two patients with Buerger disease who did not stop smoking during the therapy there was no response. No peripheral thrombosis or clinical deterioration was noticed. Conclusion: This method performed in selected patients with severe night rest pain was very effective, without any serious complications. Flow changes in the distal portion of the extremity, as measured with scintigraphies, may help us to withdraw those patients with no response.

43.6.8

P1

Use of chitosan gel as an embolic agent in experimental aneurysm models of rabbits: a focus on angiographic and histopathologic findings M. Lee1, B. Kwak2, J. Yoo3, H. Shim2, I. Kang1; 1Diagnostic radiology, Hallym University Hangang sacred Hospital, Seoul, Democratic People's Republic of Korea, 2Diagnostic radiology, Chung-Ang University, Seoul, Democratic People's Republic of Korea, 3 Diagnostic radiology, Ewha Womans University, Seoul, Democratic People's Republic of Korea.

Transcatheter arterial embolization versus surgery as salvage therapy for upper gastrointestinal peptic ulcer bleeding: a comparative study G. Rodriguez Rosales1, C. Ripoll2, J. Guzmán de Villoria1, M. Echenagusía Boyra1, N. Alava Echevarria1, R. Bañares2, G. Simó Muerza1, F. Camuñez Alonso1, A. Echenagusía Belda1; 1 Radiology, Hospital Gregorio Maranon, Madrid, Spain, 2Digestive, Hospital Gregorio Maranon, Madrid, Spain.

Purpose: To evaluate feasibility of the endovascular treatment in experimental aneurysms of rabbits using chitosan gel. Materials & Methods: An aneurysm was created in the common carotid artery of 15 rabbits and embolized with chitosan gel in 12. Three rabbits were not embolized as a control group. Two of the 12 rabbits died three days later. All 13 rabbits were evaluated by follow-ups with intravenous digital subtraction angiography (IV-DSA) at one (n=13), two (n=12), four (n=10) and six weeks (n=8) after embolization. Histopathologic evaluations were also done at three days (n=2), one (n=1), two (n=1), three (n=1), four (n=2), six (n=1), and eight weeks (n=1) after embolization. Results: At follow-ups with IV-DSA studies of embolized aneurysms, in seven rabbits more than 75% of decreased aneurysm size was noted, a decrease of 75%-50% was recorded in two rabbits, and less than 50% decrease in one rabbit. Histopathologic findings showed well-organized thrombi and neointima formation in the aneurysms. Conclusion: Chitosan gel may be a possible embolic agent in aneurysms because of its mechanical embolizing effect and its biological effect of eliciting fibrosis and neointima formation.

Purpose: To compare retrospectively the outcomes of transcatheter arterial embolization (TAE) and surgery as salvage therapy after failure of therapeutic endoscopy for upper gastrointestinal peptic ulcer bleeding. Material & Methods: From 1986 to 2001, 1,053 patients required emergency therapeutic endoscopy due to bleeding peptic ulcer. Of these patients, 85 (8%) were referred for alternative treatment because of recurrence or uncontrolled bleeding. Complete information for the purpose of this study was available in 70 patients (31 treated by TAE and 39 by surgery). Results: The patients who received TAE were older (p< 0.001), and had a higher proportion of cardiac disease (p< 0.001), and previous anticoagulation (p=0.018). There were no differences in other pre-treatment variables. No differences in outcomes were found between TAE and surgery: rebleeding (29% vs 23.1%), mortality (25.8% vs 20.5%), surgery after the initial alternative treatment (16.1% vs 30.8%), days of hospitalization (30.13 ± 24.6 vs. 25.77 ± 20.75) and global positive outcome (41.9% vs 48.7%). Conclusions: This study found no difference between TAE and surgery despite the older age and higher prevalence of heart disease in the former.

43.6.9

P2

Remodeling of suspended small intestinal submucosa venous valve: an experimental study in sheep to assess the host cells origin E. N. Brountzos1, D. Pavcnik2, H. A. Timmermans2, C. Corless2, B. T. Uchida2, M. Nakata2, M. Schoder2, J. A. Kaufman3, F. S. Keller2, J. Rösch2; 12nd Radiology, Athens University, Athens, Greece, 2Dotter Interventional Institute, Oregon Health & Science University, Portland, United States, 3Dotter Interventional Institute, Oregon Health & Science University, Portland, United Kingdom.

A soft tissue-anchored transcutaneous device (T-port) for longterm catheter bearers R. S. Nyman1, U. Haglund2, B. Fellström3, D. Nyholm4, L. Eriksson1, I. Rasmusen2, B. Karlsson2, D. Lundgren5; 1Department of diagnostic radiology, Uppsala, Sweden, 2Department of Surgery, Uppsala, Sweden, 3Department of Medicine, Uppsala, Sweden, 4Department of Neurology, Uppsala, Sweden, 5Department of Biomaterial & Handicap Research, Göteborg, Sweden.

Purpose: To investigate the origin of host cells during remodeling of small intestinal submucosa (SIS) venous valves (VV). Materials & Methods: Suspended valves (SVVs) were developed by suspending VVs within bare square stents. Eight SVVs were placed within the inferior vena cava (IVC) of four sheep. Eight standard VVs were implanted in the external jugular veins of these animals. Five weeks later the devices were examined for stability and patency, and the animals were sacrificed. Gross, histologic and scanning electron microscopic (SEM) examination were performed. Results: Follow-up showed no migration, venous occlusion or thrombus formation. All SVVs were intact without contact with the IVC wall. Histology showed remodeling of both SVVs and VVs with endothelial cells on the surface. SIS neovascularization was also present. There was no difference regarding SIS remodeling between SVVs and the free part of VVs leaflets. The VVs leaflets bases were thicker compared with their free parts (p<0.01). SEM showed endothelial cells on both sides of the SVVs and VVs. Endothelialization of the SVVs central leaflet surfaces was more complete than the peripheral surfaces of the SVVs leaflets. Conclusion: SIS-based valve remodeling occurs independently of vessel wall contact by recruitment of cells directly from the circulation.

Purpose: To evaluate safety, efficacy and clinical results of a transcutaneous soft tissue-anchored device (T-port, Transcutan AB) in patients requiring long-term use of catheters for biliary drainage, gastrostomies, or peritoneal dialysis. Materials & Methods: Twenty patients with either biliary drainage (12), gastrostomy tubes (4), or peritoneal dialysis catheter (4) were offered Tport implantation in local anesthesia. Results: There were no technical problems in implanting the T-ports. The wound around the T-port healed well in all patients. No signs of infection were observed around the ports during a follow-up period between three and 12 months. Leakage and infection alongside the tubes were not observed. The tubes were easy to exchange through the ports. The quality of life was considerably improved, with less local problems and improved cosmetic results. Conclusion: The T-port considerably improves the quality of life by preventing infection and leakage and by eliminating the catheter dangling outside the body.

P3

P5

Gastric outlet obstruction and postoperative strictures: treatment with covered metallic stents Y. Han1, D. Yang2, G. Jin1, G. Chung1; 1Radiology, Chonbuk National University Hospital, Chonju, Republic of Korea, 2Surgery, Chonbuk National University Hospital, Chonju, Republic of Korea.

Balloon dilatation of vigorous achalasia A. Ivanovic; Institute of Radiology, Clinical Center of Serbia, Belgrade, Yugoslavia.

Purpose: To evaluate the effectiveness and long-term results of covered metallic stent placement in inoperable malignant gastric outlet obstruction and postoperative anastomotic strictures. Materials & Methods: The placement of covered metallic stents was attempted in 43 patients: twenty-eight with gastric outlet obstruction (18 male, aged 47 to 87 years) and 15 with postoperative anastomotic stricture (13 male, aged 39 to 82 years). All patients had intolerance of oral alimentation and/or vomiting after eating. Success was defined both technically and clinically. Results: The placement of the stent was technically successful in all patients. After stent placement, 40 patients were able to ingest at least liquids and had a markedly decreased incidence of vomiting (clinical success, 93%). Follow-up duration was 1-41 weeks in the 28 with gastric outlet obstruction and 2-23 weeks in the 15 with postoperative anastomotic strictures. The only major complications were bowel obstruction in one patient and bowel perforation in another patient. Conclusion: Stenting appears to be effective and valuable in the treatment of gastric outlet obstruction and postoperative anastomotic strictures. Bowel obstruction and perforation must be considered possible delayed complications after stent placement.

Purpose: To show the long-term results of balloon dilatation of vigorous achalasia of the esophagus Materials & Methods: The study group consists of 16 patients with vigorous achalasia (11 male, 5 female, aged from 37 to 62 years). All patients presented with dysphagia of grade 1-4 according to the Atkins scale (grade 4 in nine, grade 3 in seven). Barium swallow was performed before the procedure, immediately afterward, and every 3 months subsequently. The mean follow-up period was 36 months, ranging from 25 to 42 months Results: In nine patients with preprocedural grade 4 dysphagia, immediately after the procedure dysphagia fell to grade 2 in five patients and to grade 1 in the other four. In all seven patients with preprocedural grade 3 dysphagia, immediately after the procedure dysphagia fell to grade 1. Six patients needed repeated dilation for reccurent dysphagia after 6, 9, and 12 months. Five needed surgical treatment (long myothomy) by 15 months. Conclusion: In patients with vigorous achalasia balloon dilatation is helpful only in patients with low-grade dysphagia and as an initial treatment method.

P4

P6

Use of radiological peroral gastrostomy in patients with failed percutaneous endoscopic gastrostomy: technique and results V. Prabhudesai, J. C. Varghese, J. K. Mordani, M. A. Siddiqui, M. Cowling, D. J. West; Radiology Department, North Staffordshire Hospital NHS Trust, Newcastle-u-Lyme, United Kingdom.

Long-term results of stent placement for hepatic artery stenosis or thrombosis after liver transplantation A. R. Cotroneo1, A. Cina2, C. Di Stasi2, A. De Gaetano2, G. Marano1; 1Radiology, Università degli studi del Piemonte Orientale, Novara, Italy, 2Radiology, Università Cattolica del Sacro Cuore, Rome, Italy.

Purpose: To evaluate the use of radiological peroral gastrostomy (RPG) in patients with a failed percutaneous endoscopic gastrostomy (PEG). Materials & Methods: Over a period of five years, RPG was performed in 12 patients (M:F, 8:4; age range, 44 to 88 years; mean age, 70.7 years) because of PEG failure due to obstructing lesions of the upper gastrointestinal tract (n=8) and technical failure of endoscopy (n=4). A15-F Freka gastrostomy tube (Freka-PEG Universal-Gastric-Set) was used for RPG. The technique will be described in full; it differs from other gastrostomy techniques because of the retrograde catheterisation of the oesophagus before antegrade placement of the gastrostomy tube through the mouth Results: RPG was technically successful in all patients. There were three (25%) minor complications due to temporary tube impaction within oesophageal tumour (n=1), wound infection (n=1) and aspiration pneumonia (n=1). There were no major complications or procedure-related deaths. Survival ranged from 132 days to 3.4 years (mean, 18.4 months) from gastrostomy tube placement. Conclusion: RPG is feasible in patients with failed endoscopy.

Purpose: Hepatic artery stenosis (HAS) and hepatic artery thrombosis (HAT) represent dangerous complications of liver transplant, with high mortality and morbidity rates. In the past, retransplantation was considered the first-choice therapy, while in recent times new revascularization techniques, both surgical and interventional, have been suggested.While extensive experience has been acquired using pecutaneous angioplasty (PTA) and fibrinolysis techniques, only sporadic cases of stenting in the hepatic artery of a transplanted liver have been reported, and no long-term results of this technique are available. Materials & Methods: We positioned seven stents (five Wallstent and two Palmaz) in four patients (two with stenosis and two with thrombosis). Stenting was performed in three cases after PTA and fibrinolysis, whereas in the fourth patient we preferred primary stenting. Results: In all cases technical success was obtained. During 24-32 months' follow up all the stents proved patent, with retreatment in one, and none of the patients required repeat transplantation. Conclusion: Although experience is still limited, in our experience the placement of a stent in the hepatic artery yields good medium/longterm success, improving the results obtained by fibrinolysis and PTA and consequently enabling the graft to survive and avoiding the need of retransplantation.

P7

P9

Application of a new guide-stylet set in abdominal lesions T. Petsas, I. Tsota, C. P. Kalogeropoulou, C. Giannakenas, M. Karamesini, D. Siablis; Department of Radiology, University of Patras, Patras, Greece.

Radiological drainage of abdominal abscesses using large (28-30 F) drains V. Prabhudesai1, K. Mitra1, M. Cowling1, M. Deakin2, D. J. West1; 1Radiology Department, North Staffordshire Hospital NHS Trust, Newcastle-u-Lyme, United Kingdom, 2Surgery Department, North Staffordshire Hospital NHS Trust, Newcastle-u-Lyme, United Kingdom.

Purpose: To evaluate a new guide-stylet set for technically difficult biopsies of abdominal lesions. Materials & Methods: Thirty-six patients underwent biopsy of paraaortic lymph nodes (n=13), pelvic (n=8), liver (n=7), adrenal (n=4), small bowel wall lesions (n=3) and neurolysis (n=1), with a coaxial technique using this new guide-stylet set. The set comprises a stainless steel guide-stylet, 0.41 mm in diameter (William Cook Europe), and a 21/22G Chiba needle. The needle is initially introduced and the stylet is inserted via the needle in the direction of the target. After the stylet has been correctly directed, the needle is advanced over the stylet towards the lesion. The stylet serves either as a guide for the biopsy needle, or as an exchange wire when a larger (e.g. Tru-Cut) needle is required. Results: The technique was easy to perform. It was effective in small lesions and, particularly, in those cases in whom the target was difficult to approach. Needle redirection was achieved, when required, using the guide-stylet. In some cases, the stylet was advanced beyond the lesion without consequences. No technique-related complications were observed. Conclusion: The guide-stylet is a useful complement for a better targeting of difficult abdominal lesions.

Purpose: Untreated intraabdominal abscesses have a high mortality (45% and 100%). Interventional radiological techniques have improved the outcome in such patients but drain size does not normally exceed 16 F. Percutaneous drainage therefore occasionally fails, and surgical drainage with insertion of drains up to 30 F is required. Materials & Methods: Three patients presented with pancreatitis. Two patients developed a pancreatic abscess and one a psoas abscess. One patient needed open necrosectomy on presentation, but continued to be septic and developed a further abscess, which was drained under CTguidance. In all cases large (28-30 F) drains were inserted percutaneously. Results: All three patients responded to catheter drainage and did not need subsequent surgery. There were no complications associated with the tube and all three recovered and were discharged. Most intraabdominal abscesses can be successfully drained percutaneously using drains of up to 16 F in diameter. We have used 30-F catheters in three patients who did not respond to such catheter drainage. All three responded well with no tube-related complications. The technique of insertion will be described. Conclusion: Intraabdominal abscesses with thick pus can be managed using large (28-30 F) bore catheters thus avoiding surgery.

P8

P10

Percutaneous transgastric pancreatic intervention to change a pancreaticocutaneous fistula to a pancreaticogastric fistula J. Matsumoto, E. Yamanouchi, R. Kumano, K. Kobayashi, T. Hattori, I. Tani; Radiology, St. Marianna University Yokohama Seibu Hospital, Yokohama, Japan.

Review of 83 cases of magnetic compression anastomosis E. Yamanouchi, R. Kumano, K. Kobayashi, T. Hattori, J. Matsumoto, I. Tani; Department of Radiology, St. Marianna Univerisity Yokohama Seibu Hospital, Yokohama, Japan.

Purpose: Surgery has been considered the only way to cure a pancreaticocutaneous fistula. Our purpose is to show as an alternative to surgery a new interventional method, a percutaneous transgastric approach, to treat pancreaticocutaneous fistula. Materials & Methods: Four female patients, seven to 48 years old, underwent this interventional treatment method. Percutaneous drainage of a pancreatic pseudocyst had caused a pancreaticocutaneous fistula in three cases. Postoperative leakage from an anastomosis between the pancreas and jejunum had caused a pancreaticocutaneous fistula in the other case. The present treatment method consists of percutaneous transgastric puncture of the pancreaticocutaneous fistula and balloon dilation of the newly formed fistula between the stomach and the old fistula. Results: The skin ostium of the pancreaticocutaneous fistula rapidly healed after percutaneous transgastric drainage. There has been no recurrence of the pancreaticocutaneous fistula or pancreatic pseudocyst after tract shift to the gastric cavity until the present (four years). No complications were observed during or after this procedure. Conclusion: This percutaneous transgastric pancreatic intervention to change a pancreaticocutaneous fistula to a pancreaticogastric fistula is quite safe and efficient. This new method instead of surgery may become the first choice to cure a pancreaticocutaneous fistula.

Purpose: Magnetic compression anastomosis (MCA) has developed satisfactorily since the first report at CIRSE 1998, and in Japan it has been done in a total of 83 patients. This poster reviews various cases of MCA and focuses on avoiding acute restenosis and other problems. Materials & Methods: Eighty-three patients (59 men and 24 women) with ileus or CBD obstruction underwent MCA. A pair of samariumcobalt rare earth magnets was placed in the organs undergoing anastomosis. Frequent balloon dilatation (FBD) was used to treat acute restenosis. Results: There were 23 bilo-enteric and 65 entero-enteric procedures. Bilo-enteric anastomosis was successful in all 23 cases, while enteroenteric anastomosis was achieved in 60 of the 65 patients who could be followed up. Emergency or correction laparotomy was needed in three cases of entero-enteric MCA. The biggest problem after MCA was acute restenosis, but this was largely solved by FBD. Conclusion: Restenosis or cholangitis are rare after bilo-enteric MCA, while acute restenosis after entero-enteric MCA can be handled by FBD. When entero-enteric MCA is done, barium enema and enterography via the ileus tube are necessary to avoid the trapping of other bowel loops between the mated magnets.

P11

P13

An assessment of type II endoleaks after endovascular abdominal aortic aneurysm repair M. J. Guiney1, J. A. Brookes1, M. Adiseshiah2; 1Radiology Dept.,, University College Hospital/ Middlesex Hospital, London, United Kingdom, 2Dept., of Vascular Surgery, University College Hospital/ Middlesex Hospital, London, United Kingdom.

Virtual-CT image fusion: its usefulness in the assessment of aortic prosthesis migration A. Mallouhi, B. V. Czermak, H. Gruber, W. R. Jaschke, M. Rieger; Department of Radiology, Innsbruck University Hospital, Innsbruck, Austria.

Purpose: We report our experience in the etiologic and therapeutic evaluation of type II endoleaks after endovascular abdominal aortic aneurysm repair (EVAR). Materials & Methods: Over a 5-year period EVAR was performed successfully in 147 patients. Follow up CT angiograms were performed prior to discharge after EVAR, at 3, 6, 12, and 18 months ,and thereafter at approximately yearly intervals. Results: Type II endoleaks were found in 8 patients (5.4%). All were secondary endoleaks detected at 3-36 months from stent insertion with a mean of 15.2 months. There were 6 simple (type IIa) and 2 complex (type IIb) endoleaks. Two of the type IIa also had type I leaks. A total of 4 were treated with transarterial coil embolisation, 1 with the combination of coil embolization and thrombin injection, 1 spontaneously resolved, 1 is currently under observation and 1 was lost to follow up. Of those treated, 4 resolved and 1 had a recurrent endoleak. Conclusion: Secondary intervention is effective in treating type II endoleaks, which almost exclusively involve the lumbar and/or inferior mesenteric arteries. However follow-up is required to determine long term outcome in both treated and untreated cases.

Purpose: Evaluating the usefulness of image fusion in a three-dimensional virtual reality environment to assess aortic prosthesis migration utilizing CT examinations. Materials & Methods: CT data-sets of 16 patients with aortic prosthesis (eight thoracic, eight abdominal) were evaluated. The bony structures and aortic prostheses were segmented and a three-dimensional model was created twice for each data-set. The two models of each set were transferred to an image fusion program, with which the bony structures were automatically fused resulting in a simultaneous overlapping of both prostheses. Data were evaluated by the limits-of-agreement method. Results: The software automatically recognized the regions of inconsistent overlapping and assigned them different colors according to error value enabling a comprehensive perception of fusion error. The maximal fusion error of the prosthesis in a data-set ranged from 0.2 to 1.2 mm. Limits-of-agreement analysis showed no clinically significant error values. The bias of mean maximal error values between fused models and standard reference (considered zero in each case) was 0.66 mm (95% CI, -0.04–1.36 mm). No significant differences were recorded between abdominal and thoracic prostheses. Conclusion: Image fusion in virtual reality environment and automatic calculations of fusion error is a promising method for assessing migration of aortic prostheses.

P12

P14

Combined treatment for thoracic aortic aneurysms involving the carotid artery M. Carnini1, R. Caronno1, D. Laganà2, G. Piffaretti1, M. Tozzi1, M. Canziani1, P. Castelli1, G. Carrafiello2; 1Vascular Surgery, University of Insubria, Varese, Italy, 2Institute of Radiology, University of Insubria, Varese, Italy.

Acute endovascular treatment of ruptured abdominal aortic aneurysms with aorto-uniiliac stent-grafts and subsequent crossover femoro-femoral bypass L. Mátyás1, I. Lázár2, G. Juhász1, J. Szarka1, G. Simon1, S. Szentesi1; 1Vascular Surgery, Borsod County Hospital, Miskolc, Hungary, 2Radiology, Borsod County Hospital, Miskolc, Hungary.

Purpose: The surgical treatment of descending thoracic aorta aneurysms have yet high mortality and morbidity rates. Materials & Methods: A 75-year old woman with an aspecific thoracic angina, backache, hypertension, and cardiac arrhythmia was diagnosed with a chronic thoracic type-B dissection. A thoracoabdominal spiral-CT and a MRA showed a 12-cm aneurysm of the descending thoracic aorta involving the left carotid and subclavian arteries. An endovascular treatment to exclude the aneurysm was planned. A preliminary angiography confirmed the absence of a proximal neck. To save the cerebral perfusion, a bypass between the right and the left carotid arteries was performed. With the insertion of three endoprostheses through a surgical prosthetic iliac artery anastomosis it was possible to exclude the aneurysm. Results: The patient was discharged ten days after the procedure with no clinical sign of neurological deficit. A follow-up CT-scan 12 months after the procedure documented the exclusion of the aneurysm and the patency of the carotid bypass. Conclusion: Although the thoracic aorta was completely reconstructed spinal cord failure or arm ischemia were not recorded. Stent-grafting seems to be feasible and minimally invasive and should be considered as an alternative approach for thoracic aneurysm repair.

Purpose: To present our results with the endovascular treatment of ruptured abdominal aortic aneurysms (AAA) in the acute setting. Materials & Methods: Between January 1998 and February 2003, we treated 46 patients with AAA with stent-grafts. Five had symptomatic aneurysms with abdominal pain but no CT evidence of rupture. Another six patients (five men, one woman, age: 59-73 years) had frank ruptured abdominal aortic aneurysms. All but one were treated with aortouniiliac stent-graft and one patient with bifurcated endograft (Talent, Medtronic) under epidural anesthesia. Results: All the procedures were technically successful. One procedural type-II endoleak was recognized. This patient suddenly died in the third postprocedural day but no autopsy was performed. Another patient underwent cardiac resuscitation on the operation table. She had good recovery. A third patient with aorto-enteric fistula underwent delayed laparotomy and small bowel resection five days after stent-graft insertion. He also is doing well. No other complications or aneurysm-related mortalities occurred. Two patients died in the follow-up due to mesenteric ischemia and bronchial carcinoma. Conclusion: Our results demonstrate the feasibility of AAA endovascular repair in an acute setting even with a relatively limited stock of stent-grafts using aorto-uniiliac prostheses with contralateral iliac occlusion and femoro-femoral crossover bypass.

P15

P17

Symptomatic saccular aneurysms of thoracic and abdominal aorta: management with stentgrafts B. Üstünsöz1, K. Ataç1, &. Somuncu1, H. Tatar2, A. Sönmez3, H. Bingöl2; 1Radiology, GATA, EtlÝk-Ankara, Turkey, 2Vascular Surgery, GATA, EtlÝkAnkara, Turkey, 3Internal Medicine, GATA, EtlÝk-Ankara, Turkey.

Medium-term renal function is not affected by supra-renal fixation during endoluminal repair of abdominal aortic aneurysm P. Davey1, D. Fay2, L. Wilson1, M. Wyatt1, J. Rose2; 1Vascular Surgery, Freeman Hospital, Newcastle upon Tyne, United Kingdom, 2Radiology, Freeman Hospital, Newcastle upon Tyne, United Kingdom.

Purpose: To present the management of two saccular aneurysms of the aorta with stentgrafts. Materials & Methods: An 81 year-old man with sudden back and abdominal pain underwent spiral-CT and DSA studies with the suspicion of a dissection of the aorta. Two huge saccular aneurysms were depicted in the thoracic and abdominal portions of the aorta. The thoracic aneurysm was located at the level of T7-8 vertebral body and measured 4x4 cm, the aortic aneurysm was located one centimeter above the celiac axis and measured 8x11cm. Both aneurysms were managed by a custom-made Talent stentgraft during the same session, by a right femoral approach. Results: No immediate or mid-term complications were observed. No migration, leak or protrusion of the stentgrafts into the lumen of the saccular aneurysms were observed. The patient was still asymptomatic at the four-month follow-up. Conclusion: Stentgrafts are a good options as compared with surgery in the management of saccular aneurysms.

Purpose: To compare medium-term renal function in patients receiving aortic endoluminal stent-grafts fixed above the renal arteries with renal function of patients receiving infra-renal devices. Materials & Methods: One hundred seventy-nine patients undergoing endoluminal repair of abdominal aortic aneurysms (AAA; EVAR) between December 1995 and December 2001 and alive at least 6 months post-EVAR were reviewed. Patients were divided into Group 1 (infra-renal fixation) or Group 2 (supra-renal fixation). Renal function pre- and post-operatively and at 6, 12, and 24 months was determined by measurement of serum creatinine (sCr) and blood urea nitrogen (BUN). Changes in renal function were compared using a Mann-Whitney U-test. Results: Paired data was available for 123 patients at a minimal followup of 6 months (Gp1, n=59; Gp2, n=64). Median values for sCr and BUN following EVAR were similar in the two groups. Mean contrast volumes were significantly higher in the earlier EVAR cases employing IR fixation, i.e. 306 ml in Group 1 compared with 266 ml in Group 2, p=0.01. Conclusion: These results suggest medium-term renal function remains unaffected following EVAR of AAA with either infra-renal or suprarenal fixation. Improved technical success does not appear to be at the expense of compromised renal function.

P16

P18

Implantation of stentgrafts in patients with traumatic rupture of the thoracic aorta J. Novotný1, J. H. Peregrin1, J. Pirk2; 1Dpt. of Diagnostic and Interventional Radiology, IKEM, Prague 4, Czech Republic, 2Dpt. of Cardiovascular Surgery, IKEM, Prague 4, Czech Republic.

Results of endovascular management of thoracic aortic aneurysms and dissection by stentgraft placement S. Sharma1, G. Gulati1, A. Bhan2, S. S. Kothari3, K. K. Talwar3, P. Venugopal2; 1Cardiovascular Radiology, All India Institute of Medical Sciences, New Delhi, India, 2Cardiothoracic & Vascular Surgery, All India Institute of Medical Sciences, New Delhi, India, 3Cardiology, All India Institute of Medical Sciences, New Delhi, India.

Purpose: To evaluate the feasibility and efficacy of stentgraft implantation in patients with traumatic rupture of the thoracic aorta. Material & Methods: In six patients (five males with acute traumatic thoracic aorta rupture, one female with chronic posttraumatic aortic pseudoaneurysm, mean age 47 years), a Talent stentgraft was implanted over the aortic tear. All ruptures were located at the thoracic aorta isthmus. All procedures were performed in the angiosuite (Multistar, Siemens) using a surgical cut-down under general anesthesia, with follow-up of the stentgraft implantation using transoesophageal sonography. Results: Implantation was technically successful in all cases. In two patients the stentgraft was implanted over the origin of the left subclavian artery without clinical sequelae. In all five cases of acute rupture the stentgraft completely covered the aortic tear, with no leak present after the procedure. In the chronic pseudoaneurysm, a small proximal endoleak was present after stentgraft implantation but disappeared spontaneously by one month. No major complications occurred. All patients are alive after 1-14 months, and all are symptom-free. Conclusion: Implantation of a stentgraft in patients with traumatic thoracic aorta rupture has a high technical and clinical success rate and offers an alternative to surgical therapy.

Purpose: To evaluate the results of endovascular stentgraft placement for treating thoracic aortic aneurysms and dissections. Materials & Methods: Eleven patients (five aneurysms, six type-B dissections; ten men, mean age 48 years) were treated by this method using a commercially available device. Follow-up was done by CTangiography at one week, three months and at six-month intervals thereafter. Results: The procedure was technically and clinically successful in 10/11 patients. In one patient, there was a distal migration of the stentgraft with a massive hemoptysis. He underwent surgery but had a recurrent pseudoaneurysm at the repair site two months later, treated by stentgrafting. One patient with an aneurysm developed a type-II endoleak after one month. He underwent surgery. The remaining patients were free of symptoms at a mean follow-up of seven months. Conclusion: Endovascular stentgraft placement is promising, safe and effective in treating selected patients with thoracic aortic aneurysms and dissections.

P19

P21

Creation of individual ideally shaped stents: in vitro results from the semi-automatic virtual stent designer H. Hyodoh1, Y. Katagiri1, T. Sakai2, H. Akiba1, M. Tamakawa1, K. Hyodoh1, N. Nakamura1, N. Yama1, T. Shonai1, M. Hareyama1; 1 Radiology, Sapporo Medical University, Sapporo, Japan, 2Radiology, Fukui Red Cross Hospital, Fukui, Japan.

Percutaneous bone biopsy in children: technical considerations and difficulties G. Papaioannou, L. Thanos, A. Nikita, E. Alexopoulou, I. Mitromaras, D. A. Kelekis; 2nd Department of Radiology, University of Athens, Athens, Greece.

Purpose: We created a virtual stent design system (Semi Automatic Virtual Stent (SAVS) designer ) using three-dimensional CT data. The usefulness of the SAVS designer was evaluated by measurement of transformed anatomical and straight stents. Materials & Methods: Curved model images were created, and a hollow virtual stent was produced by the SAVS designer. A homemade Nitinol stent (SPS) was transformed to match the curved configuration of the virtual stent. The accuracy of the anatomical SPS was evaluated by experimental strain phantom studies in comparison with the straight SPS. Results: Mean separation length was 0 mm in the anatomical SPS (22 mm outer diameter (OD)) and 5 mm in the straight SPS (22 mm OD). The strain voltage of the straight SPS was four times that of the anatomical SPS at the stent end. Conclusion: The anatomical SPS is useful because it fits the curved structure of the aorta and reduces the strain force when compared with the straight SPS. The SAVS designer can help to design and produce anatomical stents.

Purpose: To evaluate the feasibility and technical peculiarities of percutaneous CT-guided bone biopsy in children. Materials & Methods: Nine children with bone lesions were referred from a children's hospital for percutaneous CT-guided biopsy. Their age ranged from eight months to 14 years. Sedation was applied in the younger patients, while in three of the older children (12 - 14 years) local anesthesia was employed. One bone biopsy set (Trocar terphine needles of 14G) was used in all patients. Additionally, in the presence of an adjacent soft-tissue mass, biopsy guns of 12G were used. The macroscopic adequacy of extracted material was evaluated by the pathologist present in the CT-room. Results: In all cases the extracted material was adequate in quantity. Histological results indicated infection in five children, benign lesion in three, and a malignant lesion in one. No major complications were observed, early or late. Some patients experienced mild discomfort. During the procedure, special care was taken to avoid ossification nuclei and growth plates. In younger children, sedation overcame the difficulties resulting from lack of cooperation. Conclusion: Percutaneous CT-guided bone biopsy in children appeared a safe and effective method with no major complications. Under certain circumstances it may replace open biopsy.

P20

P22

Migration of various designs of stent-grafts after endoluminal repair of AAA: do new-generation stent grafts improve outcome? C. V. Walch1, R. J. Bale1, R. Perkmann2, G. Fraedrich2, P. Waldenberger1, W. R. Jaschke1, B. V. Czermak1; 1University Hospital of Innsbruck, Department of Radiology, Innsbruck, Austria, 2University Hospital of Innsbruck, Department of Vascular Surgery, Innsbruck, Austria.

Percutaneous intervention on hip lesions: CT-guided bone biopsy and CT-guided abscess drainage: a fast, safe, and highly effective, low cost method S. Mylona, V. Kalioras, L. Thanos, S. Lyra, N. Lepida, A. Fidiarakis, N. Batakis; Radiology, Korgialeneio-Benakeio Hellenic Red Cross Hospital of Athens, Athens, Greece.

Purpose: To evaluate stability of different bifurcated stent-graftdesigns using postoperative spiral-CT migration measurements. Materials & Methods: Among 43 patients (39 men, mean 69 years)undergoing endovascular AAA repair, 34 completed at least 12month follow–up including computerized CT-migration analysis by comparing postoperative datasets with CT-datasets obtained at 6, 12, 24, and 36 months. Spiral CT-studies were superimposed by means of image fusion software SNT. The device was Vanguard in 15 patients, Talent in eight, Zenith in seven, Excluder in four. Results: Migration was observed within 12 months after intervention, with no further substantial migration at 24 or 36 months. At 12 months mean cranio-caudal migration of the proximal end of the Vanguard and Talent devices was 5.6mm; of the Zenith device 2mm; of the Excluder device 2.55mm. Mean cranio-caudal migration of the distal ends of the Vanguard device was 3.9mm; Talent device 5.6mm; Zenith device 2mm; Excluder device 2.5mm. Mean axial migration of the bifurcation of the Vanguard device was 21.6mm; Talent device 14mm; Zenith device 0; Excluder device 2mm. Conclusion: Due to greater longitudinal stability "new generation" stent-grafts show less migration than first-generation stent-grafts. These results are important for the clinical outcome of patients treated for AAA.

Purpose: To present our experience with percutaneous CT-guided (PCTG) bone biopsy or abscess drainage of the hip joint and adjacent soft tissues. Materials & Methods: From June 2000 to December 2002 we performed PCTG bone biopsy of the hip or femur under local anesthesia in 31 patients, using a “bone biopsy set” for the bone segment, and an automatic16G needle to receive histological specimens. In cases with suspected inflammation, we aspirated material for microbiological examination and culture. We also performed PCTG drainage in 16 patients with abscesses adjacent to the hip joint. After local anesthesia at the point of puncture the catheter was inserted directly into the abscess for irrigation. Results: Adequate samples for histological examination were obtained in all biopsies. The material was diagnostic in 25 (80.6%) cases. In the remaining six (19.4%) cases a second PCTG biopsy was performed for final diagnosis. All abscesses were evacuated successfully. There were no major complications, only mild pain at the puncture area Conclusion: Percutaneous CT-guided bone biopsy and percutaneous CTguided drainage of abscesses located adjacent to the hip joint is a fast, safe and highly effective, low cost method.

P23

P25

Pre-operative embolization of hypervascular bone tumors H. Tung1, W. K. Tso1, J. Tsang1, Y. Wong1, P. Chien2, D. K. Luk2, F. L. Chan1; 1 Radiology, Queen Mary Hospital, Hong Kong, Hong Kong Special Administrative Region of China, 2Orthopaedics, Queen Mary Hospital, Hong Kong, Hong Kong Special Administrative Region of China.

Late MRI contrast enhancement after acute myocardial infarction: accuracy in predicting left ventricular function G. Puppini1, G. Destro2, S. Montemezzi1, G. Gasparini2, G. Gortenuti1; 1Servizio radiologia OCM B.go Trento, Verona, Italy, 2Divisione clinicizzata di cardiologia OCM B.go Trento, Verona, Italy.

Purpose: To evaluate the efficacy of pre-operative embolization of hypervascular bone tumors. Materials & Methods: Four patients with symptomatic hypervascular bone tumors were managed with pre-operative embolization. These included two giant cell tumors (both in the pelvis), one spindle cell tumour (in the pubic ramus) and one angiosarcoma (in the proximal femur). Tumor size ranged from 5cm to 14cm (mean 9.8cm). All were hypervascular, with multiple feeding vessels demonstrated on angiography. The feeding vessels were embolized with polyvinyl alchohol (PVA), gelfoam or coils. Results: Post-embolization angiography showed decrease in tumor vascularity in all cases. Surgical excision of the tumors was performed within 48 hours of embolization. The operations were uneventful in all cases, with operation duration ranging from four hours to 13 hours (mean= 8.7 hours). Only one patient had transient low grade fever postoperatively. The others recovered uneventfully. All patients were able to walk within 1 week after the operation. Conclusion: Pre-operative embolization of hypervascular bone tumors is an effective and safe method to reduce tumour vascularity and enhance surgical feasibility and outcome.

Purpose: To predict left ventricular function after acute myocardial infarction (AMI), using MRI contrast enhancement. Materials & Methods: Forty patients underwent MRI after AMI treated with reperfusion. Using TRUFI short-axis sequences we computed endsystolic (ESVI) and end-diastolic volumes (EDVI) and ejection fraction (EF). After contrast administration, IS images were acquired in 3 shortaxis slices subdivided into 6 segments. Two independent observers calculated the number of segments with late hyperenhancement. Total area (transmural + subendocardial) of hyperenhancement (IS) was expressed as percent of LV. In the same 3 slices systolic thickening was defined as normal, akinetic, or hypokinetic. Results: IS correlated with ESVI (r=.48; p=0.002), EDVI (r=.47; p=0.002) and EF (r= .34; p=0.03). The correlation between transmural vs. akinetic S was very high (r=0.51; p<0.001) and vs ESVI (r=0.47; p=0.002), EDVI (r=0.37; p=0.018), FE (r=0.37; p=0.019). Subendocardial infarctions were related neither to akinetic/hypokinetic segments, nor to ESVI, EDVI, or EF. Conclusions: The size of the areas of late enhancement correlates with left ventricular function, in particular ESVI, an index of myocardial contractility. In these areas, transmural instead of subendocardial necrosis seems to be decisive in predicting postinfarctional remodeling.

P24

P26

Closed reduction and percutaneous screw fixation of posterior pelvic ring fractures: methods for reducing dose exposure in CT-guided minimally invasive musculoskeletal interventions H. G. Staedele1, H. W. Roser2, S. Meckel1, A. L. Jacob1, P. Messmer3, W. Birkfellner4, J. Roth2; 1Radiology, University of Basel, Basel, Switzerland, 2Radiologic Physics, University of Basel, Basel, Switzerland, 3Traumatology, University of Basel, Basel, Switzerland, 4CARCAS research group, University of Basel, Basel, Switzerland.

Development of an extremely rare soft tissue tumor (chondrosarcoma) after occupational exposure to ionizing radiation in a cath lab. A case report R. Bongarzone1, S. Nicolini1, G. Gabrielli2; 1Area medico-legale, INAIL - Istituto Nazionale Assicurazione Infortuni sul Lavoro, Ancona, Italy, 2Cardioradiology - Haemodynamics, "G.M. Lancisi" Cardiological Hospital, Ancona, Italy.

Purpose: To measure organ doses and effective doses of patients treated with Closed reduction and percutaneous screw fixation (CRPF) and to find ways to lower X-ray doses. Materials & Methods: Pelvic screw fixation was done in a spiral CTscanner (Somatom plus 4vz, Siemens) with 4x2.5mm collimation, 140kV, 160mAx0.5s including a topogram, a pelvic scan for preoperative planning (scan length 15cm), and 6 control scans with a mean length of 3cm (n=132 in 22 patients). Doses were measured in an Alderson phantom with CaF2 thermo luminescence dosimeters including kidneys, bowel, ovaries, and testis. Additionally effective doses were calculated. Results: Organ doses up to 212mSv were measured. Mean effective doses are 3.3mSv for men and 5.6mSv for women. By reducing scan length up to 40%, using 120kV and a 5mm collimation it is possible to reduce effective doses up to 63% which leads to an average dose of 1.2mSv for males and 2.1mSv for females. Conclusion: A significant reduction of the radiation dose is possible by reducing scan duration for preoperative planning and kV for control scans. Preliminary experience suggests that mAS can be additionally lowered substantially without compromising the safety of the procedure.

Purpose: Ionizing radiations may induce soft tissue sarcomas in the irradiated field. There is not yet evidence of radiation-induced sarcomas after occupational exposure. We present the case of a chondrosarcoma related to occupational radiation exposure. Materials & Methods: An interventional cardioradiologist developed a soft tissue chondrosarcoma in the first metacarpal region of the left hand after 20 years of exposure. Mean radiation dose measured under lead apron was 4 mSv/year. Results: The highest radiation dose level for operators is found in catheterization laboratories; this is not uniform (relatively high doses to the extremities and low doses to the trunk). The hands--particularly the left one--of physicians undertaking angiographic procedures are very close to the x-ray beam, and not shielded. The mean radiation dose revealed by the film badge worn under the lead apron, therefore, is not representative of the radiation exposure at the extremities; unfortunately, thermoluminescent dosimetry measures were not available in our case. The adequate latency period, the high radiation dose at the left hand, the site of sarcoma, reasonably indicate that occupational exposure induced the neoplasm. Conclusion: Soft tissue tumors may be related to ionizing radiation occupational exposure. Interventional radiologists must use protective measures to reduce radiation exposure.

P27

P29

High-resolution 3-D free-breathing coronary MR angiography: clinical utility for the detection of coronary artery stenoses in 107 patients with suspected coronary artery disease T. Sommer, M. Hackenbroch, A. Schmiedel, U. Hofer, H. Schild; Radiology, University of Bonn, Bonn, Germany.

Myocardial ischemia in patients with thoracic aortic aneurysm F. Todua1, M. Razmadze2, M. Balavadze2, T. Vakhtangadze2; 1Director, Institute of Radiology, Tbilisi, Georgia, 2Ultrasound, Institute of Radiology, Tbilisi, Georgia.

Purpose: To evaluate the clinical utility of high-resolution free-breathing coronary MR angiography (coronary MRA) in patients with suspected coronary artery disease (CAD). Materials & Methods: One-hundred and seven patients with suspected CAD underwent free-breathing coronary MRA (Intera, 1.5 T, Philips). An ECG-gated, fat-suppressed, 3D segmented-k-space gradient echo sequence (acquisition window 82 ms, in plane resolution 0.70 x 0.79 mm) was used. Cardiac catheterization was performed in all the patients. MR visualization of coronary arteries (CA) was qualitatively assessed using a 4-point grading scale. Results: In coronary MRAs with a good image quality [grade 1 and 2, n=77/107 (72%)] sensitivity and specificity were 88 and 91%, respectively. Sensitivity, specificity, positive and negative predictive values for the detection of a high-risk subgroup of CAD patients (i.e. left main stenosis, left main equivalent, and three-vessel disease) were 81, 97, 87, and 95%, respectively. Conclusion: 1. High-resolution free-breathing 3D coronary MRA permitted a high-quality imaging of proximal and middle main CAs with a good sensitivity and specificity for the detection of >50% stenoses in most of the patients. However, in approximately 28% of them, the image quality was severely impaired. 2. Coronary MRA can reliably identify or exclude high-risk CAD.

Purpose: To determine the incidence of myocardial ishemia in patients with thoracic aortic aneurysm (TAA). Materials & Methods: We studied 97 patients with TAA of atherosclerotic aetiology. The diagnosis of TAA was established by spiral CT and MRI. Echocardiography and ECG were performed in all cases. 87 patients (90%) had arterial hypertension (AH), 29 (30%) diabetes mellitus (type II), 14(14.4%) previous myocardial infarction. In a 20% subgroup (19 pts) myocardial SPECT using 99Tc Sestamibi was performed, with isometric exercise test in the rest. Results: In 79 cases (81%) specific (37cases, 47%) or nonspecific (42cases, 53%) ST-T segment changes were detected. In the 19-patient subgroup, 14 (74%) showed specific ischemic changes on myocardial perfusion images. In 60% of them ischemic changes on perfusion images were accompanied with ischemic changes on ECG; the other patients had nonischemic changes on ECG. The patients who developed ischemia on perfusion images were found to have lower EF, ESV and EDV. Severe aortic regurgitation was found in 85% of cases. Conclusion: The incidence of myocardial ischemia in patients with TAA is quite high. Specific ischemic changes on ECG and low EF, ESV, and EDV are closely associated with myocardial perfusion changes.

P28

P30

Utility of gated SPECT Tc99m-Sestamibi in the evaluation of patients with coronary artery disease M. M. Khodjibekova, B. N. Myasnik; Nuclear Medicine, Scientific Center of Surgery, Tashkent, Uzbekistan.

Value of intravenous nitroglycerin infusion to improve visualization of coronary arteries in free breathing 3-D coronary MR angiography A. Schmiedel, M. Hackenbroch, U. Hofer, H. Schild, T. Sommer; Radiology, University of Bonn, Bonn, Germany.

Purpose: Assessment of the extent of coronary artery disease (CAD) and accuracy of ejection fraction (EF) estimation by Gated SPECT Tc99mSestamibi (GSPM). Material & Methods: Twenty-nine patients (mean age 54.4 yrs) underwent contrast angiography (CA), 2-D echocardiography (2-De) and GSPM study. Segmental myocardial uptake defects were compared with diseased vessels (>70% luminal narrowing) as shown at angiography. Visual assessment and a modified Simpson's rule was used for determining EF by CA and 2-De, respectively, and Gaussian myocardial count profile-fitting for EF estimation by GSPM. Results: The overall sensitivity of GSPM scintigraphy for detecting significant CAD was 91.3%, and the specificity was 83.3%. The sensitivity for the diagnosis of LAD, LCx, and RCA based on defects in at least one segment corresponding to vessels was 80%, 85.7%, and 86.7% respectively. The corresponding specificity was 55.6%, 73.3%, and 92.9% respectively. The mean EF calculated by CA was 49.09 ± 9.44%, by 2-De 46.74 ± 9.25%, and by gated SPECT 51.96 ± 11.86%. EF data correlated well between CA and GSPM (ra - 0.82) but relatively poorly between 2De and GSPM (re - 0.56). Conclusion: Gated SPECT Tc99m-Sestamibi is a useful method for simultaneous estimation of function and perfusion.

Purpose: Nitroglycerin is known to dilate epicardiac coronary vessels and to increase coronary blood flow. The aim of this study was to evaluate the value of nitroglycerin to improve visualization of coronary arteries in coronary MRA. Materials & Methods: Fifty-two patients with suspected coronary artery disease underwent cardiac catheterization and free-breathing coronary MRA (Intera, 1.5 T, Philips). An ECG-gated, 3-D TFE sequence (in plane resolution 0.79 x 0.70 mm) was used. Scans of the coronary arteries (CA) with and without intravenous nitroglycerin infusion (2.5 mg/h) were performed in random order. MR images were evaluated for 1) visualized CA length, 2) qualitative assessment of visualization of CA, and 3) coronary artery stenoses > 50%. Results: There were no significant differences in 1) the average contiguously visualized vessel lengths, 2) the qualitative assessment of the visualization of the LCA and RCA, and 3) sensitivity/specificity in the detection of coronary artery stenoses between the scans with and without nitroglycerin (p>0.05). Conclusion: Infusion of nitroglycerin does not improve the visualization of coronary arteries or detection of coronary artery stenoses in free breathing 3-D coronary MR angiography.

P31

P33

Incidence of cerebral embolism after retrograde catheterization of the aortic valve in degenerative aortic stenosis. A prospective and randomized study in 112 patients using diffusion-weighted imaging T. Sommer1, M. Hackenbroch1, A. Schmiedel1, H. Omran2, H. Schild1; 1Radiology, University of Bonn, Bonn, Germany, 2Cardiology, University of Bonn, Bonn, Germany.

Detection of new brain lesions by diffusion-weighted MR imaging after carotid artery stenting J. Gil-Romero, J. Palmero, J. Guijarro, J. Soler; Radiology, Clinic Universitary Hospital, Valencia, Spain.

Purpose: To assess the incidence of clinically apparent and silent cerebral embolisms in patients with degenerative aortic valve stenosis (AS) undergoing cardiac catheterization with transvalvular catheter passage. Materials & Methods: One-hundred and fifty-two consecutive patients with known or suspected AS were randomized to receive cardiac catheterization with (n=101) or without passage (n=51) through the aortic valve. Patients underwent cerebral MRI, including diffusion-weighted MR imaging (DWI), and neurological assessment within the 48 hours before and after the procedure to assess for embolic events. Thirty-two patients without AS who had undergone coronary angiography and ventriculography served as controls. Results: Twenty-two percent of patients with AS treated by retrograde catheterization of the aortic valve showed new acute cerebral embolic events after the procedure; three of them (3%) had clinically apparent neurological deficits. In contrast, none of the patients without passage through the stenotic aortic valve nor any of the controls had any evidence of new cerebral embolisms, as assessed by MRI (p<0.01). Conclusion: The risk of structural ischemic brain damage in patients undergoing retrograde catheterization through a stenotic aortic valve has been markedly underestimated in the past. This procedure should therefore be limited to those patients with suboptimal or unclear echocardiographic findings.

Purpose: Magnetic resonance (MR) evaluation of the incidence and clinical significance of cerebral ischemia lesions after carotid angioplasty and stent placement (CAS). Materials & Methods: Eighteen consecutive CAS procedures were performed in 17 patients with high-grade carotid stenoses. Patients underwent neurologic examination before, one day, and three months after CAS. Diffusion-weighted MR imaging (DW1) of the brain was performed before and after stent implantation. Any hyperintense signal was interpreted as a post-procedural ischemic lesion. Results: Thirteen patients (72.2%) showed an unchanged post-procedural DW1of the brain. New ipsilateral lesions were found in five patients (27.7%) and, in one of them, some lesions were also found in the contralateral side (5.5%). In four patients the lesions were clinically silent and one patient had a minor stroke. Conclusion: CAS is associated with new areas of cerebral ischemia. The majority of new lesions are clinically silent.

P32

P34

Nitinol stent for carotid artery stenting: a single center's one-year experience with Smart and Precise stents I. Lázár1, L. Mátyás2, G. Juhász2, J. Szarka2, G. Simon2, S. Szentesi2; 1Radiology, Borsod County Hospital, Miskolc, Hungary, 2Vascular Surgery, Borsod County Hospital, Miskolc, Hungary.

Initial experience with the endovascular treatment of carotid stenoses G. Passalacqua1, P. Filauri1, G. Bafile2, M. D'Elia2, G. L. Turco2, L. Scalisi2, S. Petitta2, M. Pinelli2, V. D'Amario2, G. De Blasis2; 1Radiology, Civil Hospital S.Filippo e Nicola Hospital, Avezzano, Italy, 2Vascular Surgery, Civil Hospital S.Filippo e Nicola Hospital, Avezzano, Italy.

Purpose: To assess the feasibility and early clinical results of stenting of the carotid bifurcation using self-expanding nitinol stents. Materials & Methods: Between September 2001 and September 2002, 30 patients (19 symptomatic, 11 asymptomatic) with high-grade stenosis according to NASCET criteria of the internal and/or common carotid artery were treated by implantation of nitinol stents (20 Precise, 10 Smart) without predilation. Twenty-nine patients were treated for primary atherosclerotic lesions and one for post-endarterectomy restenosis (age: 45-80 years, mean: 62.4). Five patients had occlusion and seven had significant stenosis of the contralateral internal carotid artery. In six cases postdilation was performed with cerebral protection (Angioguard). Results: All the implantations were successful without technical complications. Two patients had transient neurological symptoms during the intervention. One of them had repeated TIAs until successful common carotid artery stenting and a subclavian artery recanalisation was performed subsequently. After the procedure five patients were treated with dopamine infusions due to transient hypotension and/or bradycardia. All stents were patent at six-month duplex sonography without significant restenosis. Conclusion: Treatment of common and internal carotid artery stenoses with nitinol stents seems technically feasible and safe, and promises good patency.

Purpose: To report our initial results with carotid artery stenting (CAS). Materials & Methods: From May 1997 to December 2002, 62 patients underwent CAS (31 for restenosis and 31 for primary lesions). In all the cases Doppler-US studies showed plaque stability. In two cases only, a percutaneous transluminal angioplasty (PTA) without stenting was performed while, in the remaining 29 cases, cerebral protection devices were used. Results: The technical success rate was 90.6% (58/64). In two patients the lesion could not be crossed with a guidewire because of tortuosity of the common carotid artery; one patient underwent conversion because of the spasm of the internal carotid artery after stent deployment. In the remaining 59 cases, CAS reported good results (angiographic residual stenosis <10%). We recorded one minor stroke after 24 hours and one death (myocardial infarction). During follow-up (range: 1-68 months; mean: 14.23 months) one death for myocardial infarction occurred, there were no strokes, and in one patient a restenosis after PTA was recorded; he underwent CAS with good results. Conclusion: According to our experience, CAS is safe and effective. A higher number of cases and longer follow-ups are necessary to evaluate its effectiveness in the prevention of restenosis.

P35

P37

CT-guided brain tumor biopsy: a fast and effective intervention procedure in selected patients L. Thanos, S. Mylona, V. Kalioras, N. Lepida, P. Galani, N. Batakis; Radiology, Korgialeneio-Benakeio Hellenic Red Cross Hospital of Athens, Athens, Greece.

Intraventricular bleeding after embolization of a cerebral arteriovenous malformation with Onyx B. Üstünsöz1, S. Çekirge2, N. Bulakbasi3, F. Örs3, S. Sirin4, K. Oysul5, E. Timurkaynak4; 1 Radiology, GATA, EtlÝk-ankara, Turkey, 2Radiology, Hacettepe, SýhhiyeAnkara, Turkey, 3Radiology, GATA, EtlÝk-Ankara, Turkey, 4Neurosurgery, GATA, EtlÝk-Ankara, Turkey, 5Radiation therapy, GATA, EtlÝk-Ankara, Turkey.

Purpose: To evaluate the efficacy and safety of CT-guided needle biopsy of brain tumors and to determine the optimal indications for this technique. Materials & Methods: From December 2000 to December 2002, 18 patients (aged 61 to 82, male/female=11 / 7) underwent CT guided brain tumor biopsy. First CT brain examination was performed, to localize the lesion and to select the best route for needle insertion. After that, the patient was taken to the surgical suite and cranioanatrisis took place. Under CT-guidance, we performed the biopsy using a brain-biopsy needle with blind smooth end and lateral holes, 14G in diameter. Finally the biopsy region was checked by CT. Results: Brain tumors were diagnosed pathologically in 15 patients (83.3%), inflammatory process in two (11.1%), and no conclusive diagnosis was found in one (5.5%). Repeat procedures were required in two patients. Intracranial hematomas occurred after biopsy in one patient (5.5%). No deaths or other serious complications were induced by the biopsy. Conclusion: CT-guided biopsy is a reliable method for histopathological diagnosis of brain tumors in selected cases. For superficial and large tumors it is a simple, fast, and effective procedure.

Purpose: To present an intraventricular bleeding complication after cerebral embolization of an arteriovenous malformation (AVM). Materials & Methods: A 41-year-old man with a left temporal SpezlerMartin Grade-V AVM underwent Onyx embolization in a regular radiosurgery session to reduce the AVM size before radiosurgery. Onyx embolization was performed with a FlowRider microcatheter after pre-nidal catherization of the feeders in a wedged position. Corticosteroid was started and blood pressure was kept around 90-60 mm Hg. Results: An AVM reduction of 70-80% was obtained without any acute problem. Eight hours later, a motor aphasia developed and a CT-scan showed a tetra-ventricular bleeding. Mannitol was then started. No further neurological deficit was observed and a control CT-scan carried out three days later showed stabilization of the bleeding. A final control CTscan done three weeks later showed no ventricular dilatation or bleeding. The patient underwent a successful radiosurgical procedure four months later. Conclusion: The neuroradiology team must be aware of the risk of bleeding after AVM embolizations.

P36

P38

CT-guided placement of a catheter system for temporary thoracic sympathetic chain blockade T. Kagel1, C. M. Heyer1, S. Schulz2, S. Junger2, C. Maier2, V. Nicolas1; 1Dept. of Radiology and Nuclear Medicine, Bochum University Hospital, Bochum, Germany, 2Dept. of Anaesthesioloy, Bochum University Hospital, Bochum, Germany.

Transcatheter uterine artery embolization for the management of symptomatic uterine leiomyomas: one center's experience M. Gimeno, J. Medrano, M. Herrera, R. Alfonso, A. Mainar, M. De Gregorio; Interventional radiology, Clinico Universitario Lozano Blesa, Zaragoza, Spain.

Purpose: In the diagnosis of sympathetically-maintained pain, thoracic sympathic chain blockade is indicated if ganglionic opioid application or stellate ganglion block due to anatomic variants such as aberrant nerve fibers for arm supply (“Kuntz' nerve”) have failed. Continuous temporary thoracic sympathic chain blockade is an important tool in the therapy of complex regional pain syndrome (CRPS) Type I and Type II. Blockade of sympathetic supply of the upper extremity reduces ongoing pain, decreases allodynia, and increases acral circulation and temperature. Materials & Methods: CT-guided puncture of the thoracic ganglion region and catheter placement was performed in 24 patients at Th3. Initially a bolus of 1.5-2 ml Bupivacain 0.5% was injected followed by continuous injection of 1,5-2 ml per day within 4 to 32 days. Results: In 75% of patients, a significant benefit for assisted physical therapy and benefit in daily activities could be observed. Correct placement of the catheter could be obtained in 95.8%. In 95.8% of patients we observed temporary clinical success and in 83.3% of cases longterm clinical success. Conclusion: CT-guided placement of a catheter-system for temporary thoracic sympathetic chain blockade is a safe therapeutic option in the treatment of selected patients with CRPS.

Purpose. To evaluate the safety and efficacy of uterine fibroid embolization in the treatment of symptomatic uterine fibroids. Materials & Methods. From April 1998 to December 2002, uterine fibroid embolization was performed in 48 patients presenting bleeding (42), urinary symptoms (11), abdominal pain (31), sciatica (5), dispareunia (7), or sterility (1). Polyvinyl alcohol particles were used to embolize distally, closing with proximal coils when necessary. Results. Technical success was 100%, mean procedure duration 58 minutes, and oral analgesia required for 4.6 days (range 3-7). Mean clinical follow-up was 9.2 months (range 1-12). US was performed at one, three, six and 12 months of follow-up; MR was performed six months after treatment in 28 patients. All patients had improvement in symptoms, with complete resolution in 38. All patients had reductions in uterine and dominant fibroid volumes. Mean dominant fibroid volume decreased by 32% at three months, 47% at six months and 51% at 12 months. Complications: two patients presented severe abdominal pain due to problems with the epidural anaesthesia; nine had nausea and vomiting. Conclusion: Uterine artery embolization for the treatment of uterine fibroids is a minimally invasive technique with low complication rates and very good clinical efficacy

P39

P41

Hepatocellular carcinoma: correlation between the effect of embolotherapy and the time density curve with single-level dynamic computed tomography during hepatic arteriography H. Tanihata, M. Sato, S. Shirai, K. Kishi, S. Sahara, H. Ishikawa, S. Ishii, K. Yamada, M. Kimura, M. Terada; Department of Radiology, Wakayama Medical University, Wakayama, Japan.

Ultrasound guided contrast medium controlled percutaneous injection of human thrombin for the treatment of femoral pseudoaneurysms E. Pabst, M. Brodmann, G. Seinost, G. Stark, E. Pilger; Department of Internal Medicine, LKH Univ. Klinikum Graz, Graz, Austria.

Purpose: To examine the correlation between the effect of embolotherapy and the time density curve (TDC) of hepatocelluar carcinoma (HCC) with single-level dynamic computed tomography during hepatic arteriography (CTHA). Materials & Methods: Thirty-one patients with 37 hypervascular nodules (<3cm) underwent single-level dynamic CTHA. All patients were treated with transcatheter arterial chemoembolization (TACE), and 12 nodules were treated with percutaneous ethanol injection therapy (PEIT) after TACE. Single-level dynamic CTHA was performed by catheter insertion into the appropriate hepatic artery at the same level. Images were taken continuously one per second for 40 seconds during injection of contrast medium at 2ml/sec for 10 seconds. The CT value of nodules was measured from each image of 40, and TDC was created. Results: Patients with lower rates of local relapse had lower maximal CT values, lower CT values at 40 seconds, and shorter time to maximal CT values (p<.05). The group with higher survival rates had lower maximal CT values (p<.05). Patients with fewer distant relapses tended to have smaller size and to have additional PEIT. Conclusion: TACE effect is associated with maximal CT value, CT value at 40 seconds, and time to maximal CT value.

Purpose: To assess the effectiveness and safety of ultrasound guided contrast medium-controlled thrombin injection for the treatment of femoral artery pseudoaneurysms. Material & Methods: Sixteen patients who presented with postcatherization pseudoaneurysms (mean dimensions 34 mm x 19 mm x 19 mm) entered the study. In all cases ultrasound-guided compression as the first choice of therapy was unsuccessful. The chamber of the pseudoaneurysm located nearest the artery was punctured under ultrasound guidance. By injecting ultrasound contrast medium the flow pattern of the pseudoaneurysm was documented. In a position where no contrast medium wash out into the artery was detectable, human thrombin (200-1000 IU) was injected. The injection was interrupted when ultrasound showed thrombosis of the lesion. Results: Immediately successful thrombosis was achieved in 14 patients without complications. In one case a multichamber pseudoaneurysm was only partly occluded after thrombin injection; ultrasound examination 24 hours later showed complete thrombosis. In one patient there was a striking wash out of the ultrasound contrast medium, and therefore due to the potential risk of arterial thrombosis this patient was not treated. Conclusions: Ultrasound guided contrast medium-controlled thrombin injection is an effective, safe, and minimally invasive treatment for femoral artery pseudoaneurysms.

P40

P42

Transarterial embolization in patients with acute bleeding from hepatic artery injuries P. Petruzzi, I. Bargellini, R. Cioni, A. Cicorelli, C. Vignali, C. Bartolozzi; Diagnostic and Interventional Radiology, University of Pisa, Pisa, Italy.

Comparative survival analysis of radiosurgical treatment in embolized and non-embolized cerebral arteriovenous malformations in a large cohort of patients K. Burton1, G. Tomlinson2, C. Wallace3, M. Schwartz4; 1Public Health Sciences, University of Toronto, Toronto, Canada, 2Clinical Epidemiology, Toronto General Hospital, Toronto, Canada, 3Neurosurgery, Toronto Western Hospital, Toronto, Canada, 4Health Science Centre, Sunnybrook & Women's College, Toronto, Canada.

Purpose: To evaluate the results of transarterial embolization in acute hepatic arterial injuries. Materials & Methods: Sixteen patients underwent hyperselective transarterial embolization of a bleeding hepatic artery from 1998 to January 2003. Patients had symptoms and laboratory analyses suggestive of intraabdominal blood loss. They underwent Computer Tomography (CT), followed by Digital Subtraction Angiography (DSA) and hyperselective embolization of the affected hepatic artery. Immediate DSA control and 1-7-day CT follow-up were performed. Results: CT and DSA showed six bleeding hepatocellular carcinomas, seven arterial pseudoaneurysms (three after trauma, two after biopsy, one following surgery, one caused by neoplastic infiltration), two arterial ruptures (one after left hepatectomy and pancreatoduodenectomy, one after trauma), and two arterio-venous fistulas. Lesions were selectively catheterized and embolized, using microcatheters and microcoils; in one case of a 15-cm post-traumatic pseudoaneurysm, catheterization was interrupted because of rupture of the lesion during the procedure; the patient died during surgery. DSA and CT controls demonstrated complete lesion devascularization, persisting patency of the main arteries, and acceptable liver arterial perfusion. Conclusion: Transarterial hepatic embolization is feasible and safe and enables immediate bleeding resolution. The use of microcatheters and microcoils allows hyperselective embolization, reducing the risks of ischemia and loss of functioning parenchyma.

Purpose: Stereotactic radiosurgery is a desirable and effective treatment option for patients diagnosed with AVMs; however it is not free from significant morbidity and mortality upon treatment failure. Given the range of treatment options available to AVM patients (endovascular embolization, surgical resection, radiosurgery, or a multimodal combination of them), it is essential that the risks associated with treatment failure be exposed. To do so helps ensure that radiosurgery and the multimodal treatment of embolization and radiosurgery are prescribed in the most advantageous contexts and safest manner; thus, the quality of patient care and AVM management may be improved. Materials & Methods: A retrospective cohort analysis of 150 patients who received stereotactic radiosurgery (linear accelerator, LINAC) treatment at the Toronto Sunnybrook Regional Cancer Centre (TSRCC)—one of the largest databases of arteriovenous malformation patients North America—between 1989 and 2000. Survival analysis (with AVM hemorrhage as the outcome of interest) for radiosurgically- and multimodally-treated AVMs was performed. Results: Preliminary results may indicate that multimodal endovascular embolization and radiosurgery treatment was more successful than radiosurgery alone in the treatment of cerebral AVMs. Conclusion: Final results and conclusion will be available shortly, and for inclusion in a poster presentation at CIRSE 2003.

P43

P45

Selective transcatheter embolization in idiopathic and traumatic epistaxis J. Nickel1, W. Sachs2, G. Reusch3, R. Andresen1; 1Department of diagnosic and interventional Radiology, Guestrow Municipal Hospital, Teaching Hospital, University of Rostock, Guestrow, Germany, 2Department of ENT, Guestrow Municipal Hospital, Teaching Hospital, University of Rostock, Guestrow, Germany, 3Department of Radiology, Heart and Bloodcycle Hospital Bevensen, Bad Bevensen, Germany.

Retroperitoneal bleeding in pelvic trauma: emergency treatment with selective transcatheter arterial embolization R. Corso, G. Carrafiello, M. Intotero, M. Solcia, M. C. Castoldi, A. Rampoldi; Vascular and Interventional Radiology, Niguarda Hospital, Milan, Italy.

Purpose: Rarely nosebleeds can be intractable and unfrequently conservative measures fail. The available therapeutic options are either surgical ligature or endovascular occlusion. We report on our experience with supraselective catheter embolization to achieve hemostasis in epistaxis. Materials & Methods: In three patients with idiopathic and two patients with trauma-related uncontrollable nosebleeds, the maxillary artery was imaged supraselectively using a 5-F Vitek catheter via a transfemoral incision. The bleeding source was attributed to the area supplying the sphenopalatine artery in each case. A tracker system was then used to place heliacal mini-coils in the peripheral branches of the sphenopalatine artery as embolic material. Results: After intravascular occlusion of the unilateral branches of the area supplying the sphenopalatine artery, the bleeding stopped in all patients within a few minutes. A single unilateral occlusion was sufficient in the two patients with post-traumatic epistaxis, while a further embolization of the contralateral side was necessary in one of the patients with idiopathic epistaxis three months after the initial intervention. Conclusion: Supraselective embolization is an easy, effective and lowcomplication method for the definitive treatment of idiopathic and severe post-traumatic nosebleeds. This therapeutic option should be considered especially in emergency situations.

Purpose: We report our experience in patients presenting severe pelvic fractures and retroperitoneal hematoma treated on an emergency basis with selective arterial embolization. Materials & Methods: Since 1999 we have treated 16 patients (mean age 44 years) admitted to the emergency department with major pelvic trauma complicated by active retroperitoneal bleeding detected by CTscan. The indication for therapeutic angiography was hemodynamic instability. In 90% of cases treatment was performed via a 3 French microcatheter, using coils and/or gelfoam. Results: In six patients we observed bleeding from only one vessel; the other 10 had bilateral bleeding sources from hypogastric artery branches. In 13 patients we performed selective embolization using coils 3-6 mm in diameter plus gelfoam in nine cases; three patients were embolized only with gelfoam. Thirteen patients had a prompt and permanent recovery of hemodynamic parameters which allowed elective surgical fixation of the pelvic fractures. One patient, successfully embolized, died almost immediately because of a clinically unsuspected thoracic aorta lesion. Two patients had an irreversible hemorrhagic state and died within four hours of embolization. Conclusion: We confirm emergency selective embolization as first therapeutic intervention in severe pelvic trauma with active retroperitoneal bleeding, which permits hemostasis even in hemodynamically unstable patients.

P44

P46

Splenic artery aneurysms: endovascular exclusion D. Laganà, G. Carrafiello, F. Fontana, M. Mangini, P. Castelli, C. Fugazzola; Univeristy Hospital, Varese, Italy.

Non-surgical management of grade III-IV splenic injuries in adult patients: does transcatheter proximal splenic artery embolization reduce the risk of secondary rupture and bleeding? B. Bessoud1, S. D. Qanadli1, F. Doenz1, M. Gillet2, P. Schnyder1, A. Denys1; 1Radiology and Interventional radiology, Centre Hospitalier Universitaire Vaudois, Lausanne, Switzerland, 2Surgery, Centre Hospitalier Universitaire Vaudois, Lausanne, Switzerland.

Purpose. To verify the safety and feasibility of different modalities of percutaneous treatment of splenic artery aneurysms (SAA). Materials & Methods. Ten SAA in eight patients (seven female and one male) aged between 51 and 69 years, all asymptomatic, were observed. Color Doppler Ultrasound (CDU) diagnosis was confirmed by spiral CT. Eight were saccular and 2 fusiform, 3 to 6 cm in size; two patients had a double localization. Eight SAA were treated with micro coils (the two fusiform aneurysms were treated by endovascular ligature), in 2 patients with association of trans-catheter Histoacryl injection; two peripheral aneurysms were excluded only by trans-catheter Thrombin injection. Results. The first follow-up was performed by CDU one day after the procedure. All patients were discharged after three days. No surgical treatment was required. Exclusion was confirmed at follow-up examination (CDU and spiral CT) in all patients at 3 months. A small splenic infarction was observed only in two patients with double aneurysm. Conclusion. Endovascular exclusion of SAA is a safe and effective procedure in asymptomatic patients, with a 100% primary success rate.

Purpose: To evaluate the feasability and efficacy of transcatheter proximal splenic artery embolization (TPSAE) in the management of grade III and IV splenic injuries. Materials & Methods: Thirty-five patients (28 men, 7 women) aged from 19 to 89 years (mean 39) were referred for splenic artery embolization after splenic trauma. All patients were hemodynamically stable, with grade III or IV injuries according to the AAST Organ Injury Scale for the spleen. We evaluated from medical records the rate of rebleeding, the need for blood transfusion, and the rate of secondary splenectomy. Results: TPSAE was technically feasible in all patients. After the procedure, all patients except one showed stable hemodynamic parameters and hematocrit levels. None of them required further red blood cell transfusions. One patient underwent splenectomy two days after embolization for splenic rebleeding associated with hemodynamic instability. Conclusion: Even in severe splenic injuries PSAE is feasible and allows non-operative management for the majority of patients.

P47

P49

Glubran use in endovascular treatment of aortoiliac aneurysms J. Urbano, M. A. Arjonilla; Radiology, Fundación Jimenez Diaz, Madrid, Spain.

Trysacryl gelatin microspheres versus polyvinyl alcohol particles in the preoperative embolization of bone neoplasms A. Basile1, T. Rand1, F. Lomoschitz1, C. Toma2, J. Lammer1; 1Angiography and Interventional Radiology, University Clinic, Vienna, Austria, 2Orthopedy, University Clinic, Vienna, Austria.

Purpose: To demonstrate an alternative and/or complementary endovascular treatment of aortoiliac aneurysms using surgical glue (Glubran, N-butyl-2-cyanoacrylato-monomer). Materials & Methods: Case 1: a 5 cm saccular aneurysm on the left side of the supraceliac abdominal aorta. Usual endovascular techniques were not possible due to the risk of occlusion of the celiac trunk and SMA. Surgical treatment had been rejected. Through an aortic wallstent the aneurysmal sac was filled with coils and treatment was completed by injecting Glubran between the coils. Case 2: AAA enlarging despite previous bifurcated stentgraft treatment. Type III endoleak due to a fabric hole. The aneurysmal sac was filled with Glubran and the lumbar arteries blocked with coils. Case 3: a 4 cm isolated aneurysm in the left hypogastric artery. It was filled with Glubran after occluding the distal branches of the aneurysm with coils. Results: There were no distant embolizations or other complications . Treatment was effective without sign of leakage. After nine months aneurysm size remains stable in case 1 and has decreased by 5 and 10 mm after six and four months respectively in cases 2 and 3. Conclusion: Treatment of selected cases of aortoiliac aneurysms with cyanoacrylate is feasible and safe.

Purpose: The efficacy of polyvinyl alcohol particles versus trisacryl gelatin microspheres in the preoperative embolization of bone neoplasms was evaluated. Materials & Methods: Preoperative embolization of bone tumors (primary or secondary) was carried out in 49 patients (age range 12/83). Twenty patients were treated using microspheres (group A) and 29 using PVA particles (group B). The delay between embolization and surgery ranged between one and 13 days in group A and one and four days in group B. The difference in pre- and post-operative hematic levels concerning hemoglobin, hematocrit and erythrocites count, obtained immediately before and after surgery, were considered indirect signs of intraoperative bleeding and used as statistical comparative parameters. For a homogeneous comparison we evaluated the hematic data of those patients treated in the first three days after embolization (group A n=17; group B n=26). Results: According to the Student t-test, the differences of hematic parameters in the two groups were significant (group A= Hb 1.43g/dl, Hk 4.52%, Eryt. 0.53T/l; group B: Hb 2.28 g/dl, Hk 6.76%, Eryt.0.84 T/l). Conclusion: Microspheres are more effective than PVA particles in preoperative embolization of bone tumors.

P48

P50

Ultrasound-guided glue injection in the treatment of femoral pseudoaneurysms C. Aytekin, A. Firat, E. Yildirim, I. Kirbas, F. Boyvat; Dept. of Radiology, Baskent University, Ankara, Turkey.

Hepatic arterial embolization for treatment of massive bleeding following hepatic and pancreatic surgery A. Sato, T. Ishibashi, H. Saito, T. Matsuhashi, S. Takahashi; Diagnostic radiology, Tohoku university school of medicine, Sendai, Japan.

Purpose: To evaluate the efficacy of ultrasound-guided glue injection in the treatment of femoral pseudoaneurysms. Materials & Methods: In 14 patients, 15 femoral pseudoaneurysms were embolized with direct percutaneous glue injection under ultrasonographic guidance. We compressed the aneurysm neck during the injection to prevent distal embolization. Results: All of the patients were treated successfully. In each patient, complete occlusion of the pseudoaneurysm was achieved without complication. The procedure time varied between five and 20 minutes. Conclusion: Direct percutaneous glue injection under ultrasonographic guidance is an effective and easily applied technique in the treatment of pseudoaneurysm.

Purpose: Massive intraperitoneal or gastrointestinal bleeding is a lethal complication after pancreaticoduodenectomy and hepatic lobectomy, often caused by inflammatory rupture of the hepatic artery. Hepatic arterial embolization (HAE) may bring hepatic failure because of the reduction of collateral arteries. The purpose is to evaluate the efficacy of HAE for post-surgical bleeding Materials & Methods: We selected 10 patients who required HAE using coils and/or n-butyl cyanoacrylate after pancreaticoduodenectomy (n = 8) and hepatic lobectomy (n = 2). Embolized hepatic arteries were as follows: proper and/or common hepatic arteries in seven; main hepatic branches in three. Results: In all patients, the bleeding was controlled by HAE alone (technical success rate of 100%). Five patients with collateral hepatic branches such as aberrant right hepatic artery had favorable outcomes. Liver failure occurred in three of other five patients without collateral hepatic branches. Two of three patients with hepatic failure were rescued by intensive medical care, but one died (clinical success rate of 90%). Conclusion: HAE is often needed in patients following hepatic and pancreatic surgery to control bleeding. Though hepatic failure sometimes occurred in patients without collateral hepatic branches, the clinical success rate of embolization is higher than that of other treatments.

P51

P53

Transluminal radiofrequency thermal ablation using a new stenttype electrode: an experimental pilot study H. Rhim1, Y. Kim1, H. Song2, J. Shin2, T. Seo3, B. Koh1, O. Cho1, H. Seo1, Y. Kim1; 1 Diagnostic Radiology, Hanyang University Hospital, Seoul, Republic of Korea, 2Diagnostic Radiology, Asan Medical Center, Seoul, Republic of Korea, 3Diagnostic Radiology, Gachon University Hospital, Inchon, Republic of Korea.

Histological comparison of a vibrating guidewire with conventional guidewire technique in an in-vivo experimental coronary model C. Katsouras1, L. Michalis1, V. Malamou-Mitsi2, M. Rees3, D. Niokou4, D. Nikas1, D. Tsetis5, D. Sideris1; 1 Cardiology, University Hospital Ioannina, Ioannina, Greece, 2Pathology, University Hospital Ioannina, Ioannina, Greece, 3Radiology, Bristol Royal Infirmary, Bristol, United Kingdom, 4Anaesthesiology, University Hospital Ioannina, Ioannina, Greece, 5Radiology, University Hospital Heraklion, Heraklion, Greece.

Purpose: To evaluate the feasibility of transluminal radiofrequency thermal ablation using a new stent-type electrode. Materials & Methods: In-vitro study: the radiofrequency electrode is a self-expandable nitinol stent with a bilateral 1-cm insulation. Four ablation sessions at the target temperatures of 70, 80, 90, and 100°C were performed in bovine portal veins. The ablated lesions were then measured. In-vivo study: in four mongrel dogs, ablation was performed in the common bile duct and esophagus. Dogs were sacrificed immediately after and bioptic specimens were obtained. Results: In-vitro study: the sizes of the ablated lesions showed significant correlations with the target temperatures (r>0.4, p<0.05). The ablated sizes of the insulated segment in all the cases of the 70- and 80°C- groups were less than 1 cm. In-vivo study: on microscopic examination, central tissues of the bile duct stent showed more prominent pathological changes than peripheral tissues, and those tissues adjacent to the stent showed minimal or no changes. Microscopy showed mucosal destruction and shedding, submucosal edema and coagulation necrosis, and destruction of submucosal glands. The muscle layers, however, did not show any abnormalities. Conclusion: Transluminal radiofrequency thermal ablation using the stent-type electrode may be useful for prolonging patent stent duration.

Purpose: To compare the arterial damage caused by vibrational angioplasty (VA) and conventional guidewire manipulation (CGM) in normal coronary arteries. Materials & Methods: The coronary arteries of six lambs were randomised to VA, CGM or no treatment. At the end of the procedure, 3-5 samples from different arterial segments were examined histologically and scored from 1 (no damage) to 4 (severe damage). The arterial damage was evaluated by means of the percentage of damaged samples (PDS), the mean score of damage(MSD) [sum of the damage score (DS) of each sample/total number of samples], the mean maximum DS per artery (MMDSPA) (sum of the maximum DS per artery/6) and the percentage of samples with DS 3 or 4. Results: Sixty-two samples were scored (VA:25, conventional angioplasty:18). There was no difference in the PDS between the two groups (VA: 40.0%; CGM: 66.6%). Using VA, the MSD (1.52±0.71) and the MMDSPA (1.33±0.51) were smaller than after CGM (2.33±1.08, p=0.005 and 2.66±1.03, p=0.018, respectively). Also the percentage of samples with DS 3 to 4 was lower in the VA group (12.0% vs. 55.0%, p=0.006). Conclusion: Vibrational angioplasty may be a safer way of passing a guidewire through a vessel than conventional manipulation.

P52

P54

Feasibility of a highly porous graft in the stent-graft treatment of aortic dissection: an experimental study in dogs Y. Ohuchi, M. Hashimoto, T. Kaminou, K. Nakamura, K. Sugiura, T. Ogawa; Radiology, Faculty of Medicine, Tottori University, Yonago, Japan.

Developement of a bovine liver perfusion model for experimental radiofrequency ablation A. Lubienski, R. Bitsch, C. Dechow, K. Lubienski, M. Düx; Diagnostic Radiology, University Hospital Heidelberg, Heidelberg, Germany.

Purpose: Highly porous graft material has an advantage in biocompatibility and downsizing of stent-graft introducing systems. The aim of this study is to evaluate the feasibility of highly porous mesh as a graft material for sealing off the entry site in aortic dissection. Materials & Methods: Hand-made drum-type stents were used in this study. Highly porous polyester mesh (HPPM, porosity 10,000ml) was attached to the distal end of Gianturco Z stent struts. The stents were placed into seven beagles' aortas and arterial blood pressure was monitored continuously above and below the stents. The animals were sacrificed one month after placement, and specimens of the aorta with the mesh grafts were prepared for histopathological examination. Results: After placement, systolic pressure gradient between above and below the graft increased gradually. By one month the graft mesh was completely occluded, and it was histologically proved that the grafts were sealed with neointima in all cases. Conclusion: The highly porous polyester mesh graft was capable of blocking aortic blood flow and it sealed completely with neointima. From these results it was considered that the highly porous mesh stentgraft has potential clinical application in aortic dissection.

Purpose: To optimize technical features and clinical application of radiofrequency ablation systems a bovine liver perfusion model was developed. Material & Methods: In a liver perfusion tank, perfusion of bovine livers directly from the slaughterhouse was simulated. The fluid used was an oxygenated Tyrode's solution prepared according to physiological criteria (as for liver transplants) heated to 36-37°C by means of a heat pump. Two additional pumps controlled the portal vein and the hepatic artery inflows. The flow rate was adjusted as for the physiology of a human liver converted to bovine liver conditions. The Tyrode's solution discharged from the liver was returned into the perfusion system through the vena cava. Results: Swivels mounted on the liver perfusion tank allowed the reproducible generation of several RFA lesions of standardized depths and angles with respect to the liver surface. Access to defined liver areas was given with ultrasound guidance. Additional puncture needles allowed the independent injection of different liquids. According to the liver size (usually 6 to 8 kg) and the RF system, it was possible to generate at least 25 to 40 RFA lesions/liver. Conclusion: Our liver perfusion model allowed extensive, standardized evaluation of technical or procedure-related developments of radiofrequency systems.

W

N W RA D ITH

P55

P57

Initial long-term results of the new PolyphosphazeneF nanocoat to reduce thrombogenicity and late restenosis of balloon-expandable stents in a pig external iliac artery model G. M. Richter1, F. Thomas2, L. Grenacher1, M. Henn1, S. Satzl1, P. Christoph1, P. Kurz1, M. Grunze3; 1 Diagnostic Radiology, University Hospital Heidelberg, Heidelberg, Germany, 2Vascular Surgery, University Hospital Heidelberg, Heidelberg, Germany, 3Physical Chemistry, University Heidelberg, Heidelberg, Germany.

High-field-strength MR imaging and coronary artery stents: in vitro evaluation of magnetic attraction forces and heating at 3.0 Tesla A. Schmiedel, M. Hackenbroch, H. Schild, T. Sommer; Radiology, University of Bonn, Bonn, Germany.

Purpose: To report the results of a new nanocoating (Poly-Tri-Fluorethoxy-Polyphosphazene) to reduce stent thrombogenicity and late restenosis in a pig iliac artery model. Materials & Methods: Fifteen minipigs (25 kg) were divided into three groups (one-, eight-, and 20-week follow-up intervals) with five animals per group. The new ultrathin (50 nanom.) coating was applied on balloon-expandable stents (5/20mm, 80µ struts) using a proprietary coating process. Coated stents were randomly assigned to either the left or the right side. A non-coated stent was placed contralaterally to serve as a control. Implantation was performed via a transfemoral approach using 5-mm balloons and applying heparin anticoagulation (150 U/Kg). Before necropsy, a magnification angiography was performed. Explanted stents were examined by electron scanning and light microscopy. Results: Coated stents showed no thrombus, the average intimal thickness measured 51µ ± 44 at four and 59µ ± 47 at 12 weeks. In noncoated stents, a thrombus deposition occurred in four. Average intimal thickness measured 141µ ± 50 at four and 155µ ± 55 at 12 weeks (statistical significancy at all intervals, p < 0.01). Conclusions: The new coating provided an excellent efficacy to reduce early thrombogenicity and intimal hyperplasia in an external iliac artery pig model.

Purpose: To evaluate the MR compatibility of various coronary artery stents with respect to magnetic attraction, torque, and heating at 3.0 Tesla. Materials & Methods: Twenty commonly used coronary artery stents from various companies were studied in a 3.0 T system (Intera, Philips Medical Systems). Each stent was evaluated for: 1) magnetic deflection forces, 2) magnetic field-induced torque, and 3) RF-induced heating. Results: Magnetic deflection force: In all devices tested the deflection force (range 0.01 -0.18 mN, mean 0.07 mN +/- 0.02 [SD]) was less than the gravitational force (i.e. the clips weight). Torque: None of the clips aligned to or rotated in the magnetic field at any of the various 45°-interval positions. Heating: Maximum temperature increases of the clips ranged from 0.1°C to 0.8°C (mean 0.2°C +/- 0.19). Conclusion: MR imaging at 3.0 Tesla may be performed safely in patients with any of the 20 different coronary artery stents evaluated in this study.

P56

P58

Controlled release of doxorubicin from new chemoembolic microspheres prepared with chitosan and doxorubicin B. Kwak1, H. Shim1, J. Kim1, M. Yoon1, H. Baik1, M. Lee2, S. Han3, S. Son3; 1Radiology, Yongsan hospital, Chung-Ang university, Seoul, Republic of Korea, 2Radiology, Hanlym university, Seoul, Republic of Korea, 3Environmental Health, Seoul Health College, Sungnam, Republic of Korea.

MR-guided renal embolization using intra-arterial CE-MRA and active catheter tracking C. Fink1, P. Hallscheidt2, S. Zuehlsdorff3, S. Volz3, R. Umathum3, R. Galmbacher4, W. Semmler3, H. U. Kauczor1, M. Bock3; 1Radiology, DKFZ, Heidelberg, Germany, 2Radiology, University of Heidelberg, Heidelberg, Germany, 3Biophysics, DKFZ, Heidelberg, Germany, 4University of Heidelberg, Heidelberg, Germany.

Purpose: To evaluate the release of doxorubicin from chitosan microspheres as a chemoembolic material. Materials & Methods: Doxorubicin was physically trapped in chitosan microspheres by expansion-drug loading-shrinking process. The doxorubicin release profile was investigated by dipping it into 50 ml of variously conditioned solutions with different NaCl concentrations (distilled water, 0.9% NaCl, 2.0% NaCl, 5.0% NaCl) and different pH ranges (pH 3, 5, 7, 8). The solution was changed at 8, 16, 24, 32, 40, 48 hours, and 3, 4, 5, 6, 7, 8, 9, 10, 11, 12 days. Released doxorubicin was measured by a UV/VIS spectrophotometer at 480nm. Results: The doxorubicin was released from chitosan microspheres for as long as more than 12 days. Although about 45% of added doxorubicin was rapidly released within 48 hours in normal saline solution, about 55% of doxorubicin remained in the molecular network of the chitosan microspheres. When release profiles were examined, the trapped doxorubicin released more slowly in NaCl solution than distilled water, and more slowly in alkali solution than acidic solution. Conclusion: Doxorubicin can be trapped in chitosan microspheres. It is released more slowly in NaCl solution than distilled water and in alkali solution than acidic solution.

Purpose: To assess the feasibility of MR-guided renal embolization in a porcine model using active catheter tracking. Materials & Methods: All experiments were performed in a porcine model in a conventional 1.5-Tesla whole-body MR scanner (Siemens, Symphony). Dedicated active 5.5-F catheters were used in combination with a conventional guidewire (Terumo). MR imaging with real-time catheter tracking was performed with a temporal resolution of 400 ms. An interactive user interface enabled to change MR imaging parameters. The active catheter was positioned in the intrarenal arterial branches and embolization material (acrylic microspheres) was injected. The success of the embolization was validated by intra-arterial CEMRA-angiography and pathology. Results: Renal embolization was successful in 60% of cases. Intra-arterial CE-MRA allowed a reliable control of embolization results. Unsuccessful catheterization was exclusively due to inappropiate catheter dimensions or configurations, as well as to the lack of MR visible guidewires. Conclusion: MR-guided tumor embolization of the kidney is feasible in a conventional MR scanner. Intra-arterial CE-MRA allows a reliable control of embolization results. MR-visible guidewires and optimized catheters are required for clinical embolizations. Potential advantages of MR-guided vascular interventions are the lack of radiation and iodinated contrast material and the combination with functional measurements (flow and perfusion).

P59

P61

Endovascular treatment of symptomatic uterine leiomyomas: effect of MRI signal and contrast-enhancement pattern on treatment M. Harman1, S. Zeteroglu2, M. Sengul2, H. Arslan1, M. Kamaci2; 1Radiology, University of yuzuncu yil, Van, Turkey, 2Obstetric and gynecology, University of yuzuncu yil, Van, Turkey.

Ultraflex endourethral prostheses in treatment of bulbar urethral strictures of iatrogenic and postinflammatory origin: long-term effects D. Mašulovic1, Z. Markovic1, C. Tulic2, J. Hadzidjokic2, D. Saranovic1, J. Šaponjski1, K. Kacar1; 1 Interventional Radiology, University Clinical Center, Belgrade, Yugoslavia, 2Institute of Urology and Nephrology, University Clinical Center, Belgrade, Yugoslavia.

Purpose: We assessed the MRI signal and contrast enhancement features of uterine leiomyomas before and after embolization to determine whether there are pre-embolization MRI characteristics predictive of a successful outcome. Materials & Methods: We performed bilateral uterine artery embolization on 20 patients (28 leiomyomas) who had symptomatic uterine leiomyomas. These patients were evaluated with T1-, T2-weighted and gadolinium-enhanced MRI sequences both before and at 6th months after embolization. The relationship between the MRI signal characteristics and contrast involvement features of leiomyomas before the procedure and the change in size of the lesions after treatment was investigated. Results: Before embolization, mean leiomyoma volume was 280 cm³ (8560 cm³). The decrease in leiomyoma volume averaged 47 % (mean volume of leiomyoma: 148 cm³) six months after embolization. The response to treatment was better in leiomyomas with hyperintense T2weighted images and with more contrast involvement according to the myometrium (p<0.002). The response to treatment was insufficient in leiomyomas with high signal characteristic in T1-weighted images (p<0.001). Conclusion: MRI signal characteristics and contrast enhancement of leiomyomas before embolization can help predict subsequent response to treatment

Purpose: Long-term results of Ultraflex endourethral prosthesis in treatment of recurrent bulbar stricture are evaluated. Material/Methods: During a 2.5-year period a total of 22 patients (aged from 43 to 67) with bulbar urethral lesions underwent stent placement. Their lesions had been previously treated using endoscopic urethrotomy at least twice, and they also underwent dilatations. The strictures were up to 1.5 cm in length. Criteria governing the decision on stent placement included history, urethroscopic, urethrographic and uroflowmetric findings. Stents up to 3.5 cm long and 14 mm in diameter were applied. Results: Six months subsequent to the interventions satisfactory results, based on subjective patients assessment, uroflowmetry and series of endoscopic examinations, were confirmed in 20 (91%) patients. After another six months, the success rate was 86%, a rate unchanged after 18 months. In all patients, postprocedural perineal discomfort and erection-associated pain decreased with time, while in seven cases post-micturition dribbling developed. These complaints resolved spontaneously in all patients in two weeks to three months. Conclusion: The long-term results obtained indicate high therapeutic efficacy of the implanted Ultraflex metallic prostheses in this urethral segment with strictures of the above mentioned etiologies.

P60

P62

Percutaneous nephrostomy without hydronephrosis J. Irurzun, S. Gil, P. de la Iglesia, F. de España, J. Verdu; Vascular and Interventional Radiology, Hospital General Universitario de Alicante, Alicante, Spain.

Fluoroscopy-guided transurethral exchange or removal of ureteral stents: technical notes S. Park, I. Cha, K. Kim, S. Hong, H. Chung, S. Cho, Y. Kim; Diagnostic Radiology, Korea University, Guro Hospital, Seoul, Republic of Korea.

Purpose: To present our experience in percutaneous nephrostomies without hydronephrosis. Materials & Methods: In a five-year period we performed 11 percutaneous nephrostomies without excretory tract dilatation in three women and eight men aged between 39 and 82 years (mean, 60.2). Ultrasoundguided nephrostomies were performed for: ureteral iatrogenic fistula (6), endoluminal treatment with drugs (3), other reasons (2). In five cases the nephrostomy was performed in renal transplantation. In five cases, a 22-G Chiba needle and an Accustick system introducer were used, in the remaining cases an18-G needle, a 0.035-inch guidewire, and a Seldinger technique were employed. In the last five patients, hyperhydratation and diuretic administration were carried out, to permit visualization of the excretory tract. In all cases, an 8-F pig-tail catheter was used. Results: In one case we failed to place the catheter. One arterio-venous fistula managed conservatively occurred. In those cases in whom hyperhydratation and diuretic administration were used, a minimum dilatation of the excretory tract was achieved, which facilitated the renal tract punction. Conclusion: Percutaneous nephrostomy in non-dilated excretory renal tract is a procedure sometimes necessary and technically feasibly. Quick hyperhydratation together with administration of diuretics and ultrasound-guidance helps the realization of this procedure and minimizes complications.

Purpose: To evaluate the various methods and the usefulness of transurethral replacement or removal of ureteral stents under fluoroscopic guidance. Materials & Methods: A total of 17 female patients (38 cases) were included. They had a history of cervical cancer with surgery and radiation therapy (n=12), stomach cancer with carcinoma peritonei (n=4), or transitional cell carcinoma (n=1). An 8 Fr Nelaton catheter was inserted in the urinary bladder, and the bladder was fully distended with contrast media. The distal end of the ureteral stent was extracted to the urethral orifice using a gooseneck snare, guide wire, angiographic catheter or grasping forcep. Results: Ureteral stents were successfully exchanged or removed in all cases. Four methods were used: direct snare technique (19 cases), second snare technique (13 cases), wire loop technique (3 cases) and direct grasping technique (3 cases). Some patients showed mild pain controlled with analgesics (n=6) or self-limited hematuria (n=8). Conclusion: Fluoroscopy-guided transurethral exchange or removal of ureteral stents is a useful alternative to cystoscopy .

P63

P65

Antegrade insertion of ureteric stents using a pull-through technique S. Park1, H. Shin1, H. Kim1, C. Lee2; 1Diagnostic Radiology, Soonchunhyang University Chonan Hospital, Chonan, Republic of Korea, 2Urology, Soonchunhyang University Chonan Hospital, Chonan, Republic of Korea.

Percutaneous management of obstructive uropathy due to pelvic malignancy: our experience V. D. Souftas, I. Tsitouridis; Radiology - Interventional Radiology, Papageorgiou General Hospital, Thessaloniki, Greece.

Purpose: To evaluate the efficacy of antegrade insertion of ureteric stents using a pull-through technique in tight ureteric strictures when a stent cannot follow a passed guidewire. Materials & Methods: Six patients (M:F=2:4, mean age, 57 years) underwent antegrade ureteric stent insertion using the pull-through technique. The strictures were due to impacted ureter stone (n=2), metastasis from rectal cancer (n=1), and idiopathic ureter stricture (n=3). After failing to negotiate the ureteric stent through the stricture, a 7-F introducer sheath was inserted into the bladder through the urethra. Using a snare technique, the distal end of the guidewire was retrieved through the introducer sheath. Then, using the pull-through technique, an antegrade ureteric stent insertion was performed. Results: Technical success rate was 100% with an adequate internal drainage. Complications included abdominal/flank pain (n=5), gross hematuria (n=3), elevated blood pressure (n=1), nausea (n=1), and pain shock (n=1). Pain shock occurred due to back pain in a patient with ureter and spinal metastasis from rectal cancer. All complications relieved within three days with medical care. Conclusion: Antegrade ureteric stent insertion using the pull-through technique is feasible, and can be used in tight stenoses when a ureteric stent cannot follow the passed guidewire.

Purpose: To evaluate the indications and the effectiveness of percutaneous nephrostomy (PN) in obstructive uropathy due to pelvic malignancy. Materials & Methods: In 182 patients (23 – 94 years; average 64 years) with obstructive uropathy due to pelvic malignancy (73 in terminal stage), PN was performed under fluoroscopic guidance. Percutaneous ureteral stenting was attempted in 35 patients and was successful in 32. Results: In 78.6% of the patients renal function was restored to normal; in 12.1% it improved, and in 9.3% no change was noticed. In a period between 48h and 37 months post PN, 177 patients died (median survival 8.7 months). 32.8% patients were able to return to their usual personal activities, 16.4% needed minimal support, and 13% needed full support. There was no clinical improvement in 37.8% (elderly, with ascites, limb edema, or metastatic bone disease). Five patients with non-Hodgkin's lymphoma survived, because they could be treated for their underlying disease. Conclusion: PN/stenting for management of obstructive uropathy due to pelvic malignancy is an easy and effective procedure. It offers prolongation of survival with an acceptable quality of life in the majority of patients, including those in a terminal stage.

P64

P66

MR findings three months after uterine artery embolization for symptomatic leiomyomas M. Janczarek1, R. Pietura1, W. Bednarek2, P. Bernat1, J. Kotarski2, M. Szczerbo-Trojanowska1; 1Interventional Radiology & Neuroradiology, University Medical School of Lublin, Lublin, Poland, 2Gynecology, University Medical School of Lublin, Lublin, Poland.

Uterine leiomyoma embolization: the role of power-Doppler ultrasonography M. Harman1, S. Zeteroglu2, M. Sengul2, O. Etlik1, H. Arslan1; 1Radiology, University of Yuzuncu Yil, Van, Turkey, 2Obstetric and gynecology, University of Yuzuncu Yil, Van, Turkey.

Purpose: MR findings evaluation three months after uterine artery embolization (UAE) for leiomyomas. Materials & Methods: Between November 2001 and January 2003, MR was performed in 110 patients. T1- and T2-weighted spin-echo sagittal, transverse, and coronal MR pelvic images were obtained before and three months after UAE. Results: Before UAE, 58% of leiomyomas had iso- or hypointense signal in T2-weighted images. The mean volume of the uterus was 496 cm3, that of the dominant fibroid 198 cm3 and that of the non-dominant leiomyoma 90 cm3. Leiomyomas were submucosal (63%), transmural (13%), and subserosal (24%). At three-month post-UAE MR, the mean uterine volume had diminished by 36%, the mean dominant leiomyoma volume by 57%, and the mean non-dominant leiomyoma volume by 65% (p<0.05). Three months post-UAE, the signal in T2-weighted images in 103/110 patients was lower than before the procedure (p=0.0014). The volume of those leiomyomas with a high signal in T2weighted images before the procedure had decreased more than that of leiomyomas with a low signal (p<0.05). Conclusion: Three-month post-UAE MR showed that leiomyomas' volume had reduced by more than 50%. The volume of submucosal leiomyomas and those with a high signal in T2-weighted images had decreased more than others.

Purpose: Vascularity of uterine leiomyomas was evaluated by powerDoppler ultrasonography (PDUS) before and after bilateral uterine artery embolization (UAE) to establish the efficacy of the procedure and its contribution to the treatment. Materials & Methos: Patients were evaluated by PDUS to assess uterus and leiomyomas vascularity before and after embolization; findings were then compared with angiography. Relationships between PDUS findings and treatment results were statistically analyzed. Results: Before UAE, vascularity was defined as mild (group I, five cases), moderate (group II, seven cases) and remarkable (group III, eight cases). At the sixth month after the procedure, decreases in the volume of leiomyomas were determined as 37.2, 44.9, and 59.4% in group I, II, and III, respectively. A more significant volume regression was recorded in those leiomyomas with a remarkable vascularity before UAE and in those cases with a longer period of low vascularity in the follow-up period (p<0.005). Conclusion: PDUS can be helpful in the selection of patients candidate for UAE. It is a non-invasive and non-expensive method to determine the efficacy of the procedure in early and late periods after embolization.

P67

P69

Intra-arterial Actinomycin-D infusion chemotherapy for cervical pregnancy K. Ogawa1, Y. Tanami2, M. Shinohara1, S. Kuribayashi2, Z. Kondoh3, J. Sakai4, K. Kamei4; 1 Diagnostic Radiology, Nippon Kokan Hospital, Kawasaki, Japan, 2 Diagnostic Radiology, Keio university hospital, Tokyo, Japan, 3 Gynecolory, Nippon Kokan Hospital, Kawasaki, Japan, 4Gynecology, Nippon Kokan Hospital, Kawasaki, Japan.

Hemodialysis cathether management: current problems and potential solutions M. D. Ferrer-Puchol1, E. Esteban-Hernandez1, M. Gonzalez-Añon2, A. Pastor1, P. Sanfelix1, S. Ferragud1, A. Nacher1; 1 Radiology, Hospital de la Ribera., Valencia, Spain, 2Radiology, Fundacion Instituto Valenciano de Oncologia, Valencia, Spain.

Purpose: To introduce the use of intra-arterial Actinomycin-D infusion therapy for cervical pregnancy. Materials & Methods: Four cases of cervical pregnancy were diagnosed. All diagnoses were based on HCG titers, transvaginal ultrasound, and MRI. The cases were diagnosed between the 5th and 8th week of gestation. The gestational sacs were between 12 and 18 mm in size, and the HCG titers were between 8000 and 32000 IU/l. Because of patient desires for future fertility and to minimize the adverse reactions of chemotheraputic agents, conservative management using an intraarterial Actinomycin-D injection was selected instead of hysterectomy or systemic injection. Actinomycin-D to a total dosage of 3.0 to 4.0 mg was selectively injected into the uterine artery in each case. HCG levels and changes in the size of the gestational sac were used to monitor the effects of the treatment. Results: In all cases, the HCG titer and the size of the gestational sac rapidly decreased, with titers negative within 4 weeks after the start of chemotherapy. Endocervical curettage was performed in one case, but the other three cases did not require any additional therapy. Conclusion: Intra-arterial Actinomycin-D infusion therapy is a useful alternative treatment for cervical pregnancy.

Purpose: To discuss current problems with hemodialysis catheters and show potential solutions. Materials & Methods: From October 1999 to December 2002, we placed 115 tunnelled Ash Split-Caths. The procedure was divided into four steps: 1. central vein access; 2. creation of the subcutaneous tunnel and placement of the catheter through the tunnel; 3. insertion of the catheter into the central venous system; 4. follow-up and evaluation of the catheter patency. Results: In 113/115 cases, catheter placement was successful. 1. The patients with limited available sites for access were 15/113. The catheter was placed through an IVC stent in 1/113. 2. 2/113 patients received pre-procedure blood products. A neck and chest hematoma occurred in 1/113. 3. Air embolism occurred in 2/113 with spontaneous resolution; catheter kinking was recorded in 4/113, catheter malpositioning in 2/ 113. 4. Catheter exchange was done in 17/113. Infection occurred in 4/ 113, fibrin sheath formation and mechanical problems were present in 12/113, venous stenosis occurred in 3/113. All procedures were described with images and drawing schemes to show the various solutions' techniques. Conclusion: The placement of hemodialysis catheter is a common procedure in interventional radiology but with many complications. A valuable experience of the operator is therefore essential.

P68

P70

Ovarian dysfunction and amenorrhea as a complication of uterine artery embolization A. Szymañska1, R. Pietura1, M. Janczarek1, D. Swatowski2, M. SzczerboTrojanowska1; 1Department of Interventional Radiology and Neuroradiology, University Medical School of Lublin, Lublin, Poland, 2Department of Gynecology, University Medical School of Lublin, Lublin, Poland.

Clinical and endovascular treatment parameters of hemodialysis angio-access do not modify patient radiation entrance skin dose, mean fluoroscopy time and room time T. Cabrera, M. Sapoval, I. Papet, L. Turmel-Rodriguez, A. Raynaud, J. Gaux; Plateau technique invasive, Hôpital Européen georges Pompidou, Paris, France.

Purpose: Uterine artery embolization (UAE) is an efficient and safe treatment for uterine fibroids, although a potential complication is represented by a presumed menopause resulting from ovarian failure. Amenorrhea rate in general population <40 is 1%. The aim of this study was to assess the incidence of presumed menopause in patients undergoing UAE. Materials & Methods: Over a two-year period, 110 patients with symptomatic uterine fibroids underwent UAE. Their median age was 44 years (29-53). Patients were divided into three groups: A) <40 years (24 patients); B) 40-45 years (36 patients); C) >45 years (50 patients). Three months post-UAE, a questionnaire was administered and amenorrhea, sign of ovarian dysfunction, was defined as presumed menopause. All the patients had a normal estradiol serum level before embolization. In all the cases, an angiogram evaluated the presence of anastomoses between the uterine and the ovarian artery. Results: A collateral flow was present in 33% of patients. A presumed menopause occurred in seven group-C patients. In three of them, an angiogram showed uterine to ovarian artery anastomoses. No correlation between the presence of anastomoses and menstrual dysfunction was found. Conclusion: After UAE, ovarian dysfunction occurred in 6% of our group-C patients.

Purpose: To evaluate the relationship between clinical and endovascular treatment parameters that could affect mean patient radiation entrance skin dose (ESD), fluoroscopy time (FT) and procedure mean time during percutaneous declottiong and/or angioplasty of hemodialysis angio-access (HAA). Materials & Methods: From a prospective database, all consecutive cases of percutaneous declotting and/or angioplasty of HAA performed over a ten-month period were evaluated. Mean fluoroscopy duration, patients’ radiation entrance skin dose and room time, as well as fistulae type, anatomic location, technical procedure and demographic variables were retrieved using the built-in database and the statistical package AIR-online. Results were analyzed by multiple linear regression Results: One-hundred ten procedures (angioplasty n=90, declotting n=9, stent-placement n=11) were performed in 53 men and 57 women (mean age: 59y). Mean room time was 88min (20-210min), mean duration fluoroscopy time was 15min (3-77min). Mean ESD was 2758.66 cGy/cm² (11.6-27742.0 cGy/cm²). No clinical or technical procedure predicted radiation exposure or procedure time. Age showed a stronger but non-significant association which could be interpreted as a bias. Conclusion: HAA percutaneous angioplasty or declotting is carried out with variable and widely ranging mean fluoroscopy times and dosages. Neither clinical nor technical parameters were predictive of radiation exposure or operative times.

P71

P73

PTA/stent of central venous stenoses/occlusions in hemodialysis patients D. Krajecková, J. H. Peregrin; Dpt. of Diagnostic and Interventional Radiology, IKEM, Prague, Czech Republic.

Percutaneous placement of tunneled hemodialysis catheters: AshSplit and Permcath A. N. Keeling, H. M. O'Dwyer, S. Lyon, M. J. Lee; Department of Radiology, Beaumont Hospital, Dublin 9, Ireland.

Purpose: To compare the effectiveness of percutaneous central venous dilatation with primary stent implantation in hemodialysis patients. Materials & Methods: From 1999-2001 we performed 45 central vein PTA in 35 patients (15 with occlusions, 20 with stenoses). 29 pts were available for follow-up (six failed attempted occlusions), including: Stenoses with favorable post-PTA results (stenosis < 30%): randomization to stent (six pts) or simple PTA (nine pts). Suboptimal post PTA results (stenosis > 30 %): stent implantation (nine pts). Occlusions: Primary stent implantation (nine pts). Results: Technical success: stenoses 100%, occlusions 60%. We performed 46 procedures in 29 patients (1.6 proc./pt). No serious complications. Six-month/one-year patency rates, primary and secondary: Overall : 69%/46% and 100%/85% Simple PTA: 66% /66% and 100%/100% Stent implantation for stenoses: 80%/40% and 100 %/67 % Stent implantation for occlusions: 58%/29% and 100%/100% Conclusion: Stenoses dilated by simple PTA had higher one'year patency rates than stenoses with primary stent implantation. Patency rate for stented stenoses was higher than for stented occlusions. PTA should be performed first, reserving stent implantation for suboptimal post PTA results, restenosis or occlusions. Primary stent implantation does not increase the long-term patency rate.

Purpose: The AshSplit catheter was compared with the Gold Standard Permcath for hemodialysis efficiency. Materials & Methods: A retrospective analysis of 127 consecutive patients requiring radiologically-monitored insertion of hemodialysis catheters was performed. The information obtained included catheter insertion sites, catheter duration, and final outcome. Results: Ninety-two AshSplit catheters and 81 Permcath catheters were placed in 69 and 58 patients, respectively. Access by the RIJ was used in 62%. Complications occurred in 12 patients (9.4%): one cardiac arrest (successfully resuscitated), local hemorrhage (n=4) and sepsis (n=7). Mean catheter duration was 254 days (range 4-786 days) for AshSplit and 244 days (range 1-1,396 days) for Permcath. 33% (AshSplit) versus 31% (Permcath) had functioning catheters at 42-month follow-up, whereas 22% versus 21% had died with a functioning catheter. The remaining catheters were electively removed for a variety of reasons. Conclusion: This study confirms that the AshSplit catheter is a safe and effective dialysis catheter, with a similar mean catheter duration to the Permcath.

P72

P74

Radiologic placement of Optiflow PC 2 catheters for dialysis access: prospective study of 24 patients D. Henroteaux1, C. Sherrington1, A. Sepul2, P. Cambier2, L. Rausin1; 1Medical Imaging, CHR de la Citadelle, Liege, Belgium, 2Nephrology, CHR de la Citadelle, Liege, Belgium.

The apex puncture technique for mechanical thrombectomy of loop hemodialysis grafts S. Park1, D. Goo2; 1Diagnostic Radiology, Soonchunhyang University Chonan Hospital, Chonan, Republic of Korea, 2Diagnostic Radiology, Soonchunhyang University Seoul Hospital, Seoul, Republic of Korea.

Purpose: To evaluate the efficacy and safety of placement of a dual lumen catheter (DLC) under imaging guidance and to assess the effectiveness of DLC to achieve an efficient hemodialysis. Materials & Methods: Twenty-four 14-F Optiflow PC2 catheters (Bard Inc., USA) were percutaneously placed under ultrasound and fluoroscopic guidance into the right (13) or the left (11) internal jugular vein of 22 patients. Immediate and late complication rates were recorded. DLC functioning were assessed, with flow rates and adverse events. Results: The technical success rate was 100%, with a prolonged blood oozing in one patient only. Two (8%) DLC were removed for late complications: one dislocation occurring after 35 days and one case of infection after 59 days. Two DLC needed urokinase injection to restore patency. Mean DLC time was 132 days (44 to 282 days). Mean measured dialysis flow rate was 300 mL/min (222 to 345 mL/min). Conclusion: Placement of DLC under imaging control is a safe and effective technique to achieve short- and long-term hemodialysis accesses.

Purpose: Bidirectional access is usually performed for percutaneous thrombectomy of a thrombosed loop hemodialysis graft. The efficacy of an alternative method of access, the apex-puncture technique, is evaluated. Materials & Methods: 101 hemodialysis patients (M:F=45:56) underwent 131 mechanical thrombectomy procedures for thrombosed loop hemodialysis grafts using the apex puncture technique. The apex of the loop was punctured horizontally toward the medial equator. The puncture needle was directed toward the venous limb for thrombectomy and angioplasty. Then the introducer sheath was withdrawn, rotated horizontally, and redirected toward the arterial limb for thrombectomy. Mechanical thrombectomy was performed in 96, lyse and wait in 29, and balloon thrombectomy in six procedures. Results: The technical success rate was 90.8 % in mechanical thrombectomy and 80 % in angioplasty of venous stenosis. Mean procedure time was 70 minutes. During the procedures, the perisheath protected area was eliminated, and radiaton exposure to the operator's hand was kept minimal. Complications were venous rupture (n=25), bleeding (n=2), arterial occlusion (n=2), ruptured graft (n=1), and cardiac arrest (n=1). Conclusion: Apex puncture technique is an effective and time-saving method, alternative to bi-directional access, with elimination of perisheath protected areas and with less radiation exposure to the operator's hand.

P75

P77

Multivariate analysis of factors predicting biliary stricture formation in living-donor liver transplant recipients J. Susman, Z. J. Haskal, J. L. Weintraub, J. H. Rundback; Radiology, New York Presbyterian Hospital/Columbia, New York, United States.

A recent CT or MRI-MRCP scan facilitates a percutaneous interventional approach to the obstructed billiary system V. D. Souftas, A. Filippidis, I. Tsitouridis; Radiology-Interventional Radiology, Papageorgiou General Hospital, Thessaloniki, Greece.

Purpose: To identify factors predicting development of biliary strictures in recipients of living-donor liver transplants (LDLT). Materials & Methods: Retrospective analysis of 85 consecutive patients who underwent LDLT between 1/98 and 7/02. Mean age: 34.8y (r 0.3-67.4); 58 adult, 27 pediatric (<18y); 43 women. Donor grafts: 50 right lobes, 25 left. Twenty-four developed biliary strictures: 23 anastomotic, one intrahepatic. Parameters tested: gender, age, preexisting liver disease, donor lobe, anastamosis type, donor duct to anastamosis ratio, stent placement during transplant, and hepatic artery thrombosis or stenosis. Results: Risk factors predicting stricture formation: biliary atresia (56% vs. other etiologies of liver disease (p=0.05), and 2:1 donor duct to anastamosis ratio (53.3% vs. 17.6% of 2:2 ratio, p=0.03). Biliary atresia predicted stricture independent of age or transplant type (odds ratio 5.0). No cases of hepatic artery thrombosis/stenosis were found in stricture patients (vs. three in non-stricture cohort). Mean time for stricture development: 213d ± 279 (r, 15-1150). Stent placement correlated with shorter time to stricture (89.7d vs. 301d for non-intubated, p=0.038). Conclusion: Biliary atresia and a 2:1 donor duct to anastamosis ratio were most predictive of development of biliary strictures following LDLT. Duct intubation correlated with shorter times to stricture formation.

Purpose: To investigate the value of CT or MRI-MRCP scanning before performing percutaneous billiary drainage and stenting. Materials & Methods: One hundred thirty-eight patients (65 male, 73 female, average age 58 years) with obstructive jaundice due to nonoperable neoplasms were treated by percutaneous billiary stenting or external drainage. All patients had undergone either CE-CT or MRIMRCP as part of their clinical investigation within10 days before the intervention.We used the printed images to plan our procedures and to choose the site, the direction and the depth of the puncture. A selected site on the spine was the guiding anatomical point. Appropriate distances were measured by using the cm scale on each printed image. Results: In 75 patients (54.3%) we could access the desired billiary branch on the first puncture and in 48 (34.8%) on a second or a further try. Stents were placed in 121 patients (87.7%). The only complication was a subdiaphragmatic abscess which developed in one patient one week after the procedure and required a CT-guided drainage. Conclusion: A recent CT or MRI-MRCP scan helps greatly in the practical planning of percutaneous billiary interventions.

P76

P78

Radiofrequency thermal ablation of benign cystic lesions: early clinical experience H. Rhim1, J. Baek2, Y. Kim1, B. Koh1, O. Cho1, H. Seo1, Y. Kim1; 1Diagnostic Radiology, Hanyang University Hospital, Seoul, Republic of Korea, 2Diagnostic Radiology, Daerim St. Mary Hospital, Seoul, Republic of Korea.

Efficacy of percutaneous dilation of anastomotic biliary strictures in living-donor liver transplant recipients J. Susman, Z. J. Haskal, J. L. Weintraub, J. H. Rundback; Radiology, New York Presbyterian Hospital/Columbia, New York, United States.

Purpose: To report early clinical experience with radiofrequency thermal ablation of benign cystic lesions. Materials & Methods: Percutaneous radiofrequency thermal ablation was performed in six patients with benign cystic lesions of the liver (n=5) and kidney (n=3). The size of the cysts ranged from 4cm to 20cm (mean: 8.2cm). Using a RITA 1500 generator and 15G electrodes, USguided RF ablation of the cystic lesions was performed at 80°C for 1015 mins. Cyst fluid was aspirated during or before the ablation through the side arm of the RFelectrode or the 18 G needles, respectively. Clinical follow-up including imaging studies was available for at least three months (mean: eight months). Results: The procedure was tolerable without immediate or delayed complications in all patients, although all complained of moderate pain during ablations. The contraction ratio of the maximum diameter of the cystic lesion ranged from 10% to 100% (mean; 67%). The contraction ratio of the estimated volume ranged from 27% to 100% (mean; 85%). A common technical problem was difficulty in tracing the needle tip due to microbubble artifacts from heated cyst fluid. Conclusion: Radiofrequency thermal ablation may be useful as a new and effective interventional procedure for treating benign cystic lesions.

Purpose: To assess the efficacy of percutaneous biliary stricture dilation in living-donor liver transplant recipients (LDLT). Materials & Methods: Eighty-five consecutive patients underwent LDLT between 1/98 and 7/02; 23 developed anastomotic biliary strictures at mean 202.3 days ± 280 (r, 15-1150). Mean age: 27.2 yr (r, 0.3-67.4); 13 adult /10 pediatric (<18 yr); 14 female/9 male. Donor grafts: ten right lobes, four left, and nine left lateral segments. Etiologies of liver disease: biliary atresia (n=10), hepatitis-C cirrhosis (5), hepatoma (2), autoimmune hepatitis (2), other (4). Fifteen patients successfully completed the course of percutaneous dilation; four were primarily revised surgically; four were referred for surgery after 1-2 dilations. Successful treatment (and catheter removal) was determined by clinical trial and/ or biliary stress manometry. Results: A mean of 2.3 biliary stricture dilations (r, 1-4) were performed over an average period of 61 days (r, 20-106), with a mean 21 days (r, 655) between sessions. Mean follow-up time for those completing dilations was 12.9 m ± 10.9 (r, 0.4-35.7), with 93% remaining obstructionfree. Using Kaplan-Meier analysis, obstruction-free survival for these patients was 93%, at six, 12, and 24 months. Conclusion: Anastomotic strictures in LDLT patients can be effectively managed in a percutaneous fashion.

P79

P81

Percutaneous treatment versus surgery for liver hydatic cysts: preliminary results of a prospective randomized trial O. Akhan1, O. Abbasoglu2, M. Köroglu1, D. Akinci1, M. Özmen1, &. Sayek2, C. Basaran1, E. Akpinar1; 1 Department of Radiology, Hacettepe University School of Medicine, Ankara, Turkey, 2Department of Surgery, Hacettepe University School of Medicine, Ankara, Turkey.

Multi-detector row helical CT in patients with self-expandable metallic stents for malignant biliary obstruction: how useful is it in detection of stent occlusion? C. Kim, S. Kim, S. Kim, S. Lee; Busan National University Hospital, Busan, Republic of Korea.

Purpose: To compare percutaneous and the surgical treatments of liver hydatid cysts in a prospective randomized trial. Materials & Methods: Twenty-two patients with 42 (Type I, II, and III) liver hydatid cysts were randomized into percutaneous and surgical treatment groups. Twenty-six cysts in 11 patients were treated percutaneously, 16 cysts in 11 patients were treated surgically. Mean follow-up was 12.4 months. PAIR and catheterization techniques were used for the percutaneous treatment. Partial cystectomy and omentopexy was preferred for the surgical treatment. Results: Volume reductions of the cysts after percutaneous and surgical treatments were 73.8 and 71.4%, respectively (p >0.05). Mean hospital stay was 3.3 days for the percutaneous and 16.3 days for the surgical treatment (p<0.05). No major complications were observed in both groups. No serious minor complications (abscess, biliary fistula, etc.) were encountered in the percutaneous treatment group. These complications, however, were observed in two (18%) patients of the surgery group and treated percutaneously. No statistically significant difference was observed between the two groups in respect to costs. Conclusion: The percutaneous treatment of liver hydatid cysts is associated with a shorter hospitalization and a lower complication rate when compared with surgery. Long-time follow-up is needed to evaluate recurrences.

Purpose: To evaluate the usefulness of multi-detector row helical CT (MDCT) in detecting stent occlusion in patients with self-expandable metallic stenting for malignant biliary obstruction. Materials & Methods: Among 79 consecutive patients who underwent biliary stent insertion due to malignant biliary obstruction between August 1999 and July 2002, 20 had MDCT studies for suspicion of stent malfunction. Fifteen cases in 14 patients who underwent percutaneous transhepatic biliary drainage were studied. The presence of stent occlusion, enhancing intraluminal or peristent soft tissue masses, and the location of stent occlusion were reviewed in MDCT. These findings were compared with direct cholangiography. Results: MDCT correctly diagnosed the presence of stent occlusion in all 15 cases in 14 patients, and contrast-enhanced MDCT detected enhancing intraluminal or peristent soft tissue masses in all cases. Of 15 cases, the location of occlusion site in MDCT was correctly estimated in 12 cases (80%), underestimated in two cases (13.3%), and overestimated in one case (6.6%), respectively. Conclusion: MDCT is a useful imaging modality for detecting the occlusion site and evaluation of the cause of occlusion when stent occlusion is suspected. It can be used for planning palliative percutaneous therapy in these patients.

P80

P82

Predictive value of biliary stress manometry in living-donor liver transplant recipients with biliary strictures J. Susman, Z. J. Haskal, J. L. Weintraub, J. H. Rundback; Radiology, New York Presbyterian Hospital/Columbia, New York, United States.

Use of metallic stents in biliary complications after liver transplantation A. Doros, V. Weszelits, D. Görög, I. Fehérvári; Department of Transplantation and Surgery, Semmelweis University, Budapest, Hungary.

Purpose: To assess predictive value of biliary stress manometry (BST) for determining clinical success after percutaneous dilation of anastomotic biliary strictures in living-donor liver transplantation (LDLT) recipients. Materials & Methods: Ten LDLT recipients with biliary anastomotic strictures were treated with balloon dilation, interim ductal intubation, and stress biliary manometry when cholangiography suggested stricture improvement. A solution of 320 mgI/ml iodixanol diluted in normal saline was infused above the anastomosis at increasing incremental rates until maximum of 20 cc/min for 2 min. Biliary pressures were measured at baseline and at each infusion level. Manometric success was defined as biliary pressures <20 cm H2O after the 20 cc/min infusion. After successful manometry, external drainage catheters were removed. Results: All patients passed stress manometry after a mean of 2.5 (r, 14) stricture dilations during a mean treatment period of 62 days (r, 36106). Repeat dilations were performed at an average interval of 23 d (r, 6-55). Mean follow-up was 8.4 months +/- 4.8 (r, 4.1-15.9) with 90% of patients remaining obstruction-free. Using Kaplan-Meier analysis, obstruction-free survival was 90% at three, six, and 12 months. Conclusions: Biliary stress manometry is useful in predicting prolonged clinical patency following anastomotic stricture balloon dilation in LDLT recipients.

Purpose: In the Semmelweis University Department of Transplantation and Surgery 136 liver transplantations were performed. Fifty-eight biliary complications occurred in 29 patients, requiring surgical therapy or a combination of minimally invasive methods. Although the use of metallic stents is not widely accepted, in some cases we implanted such stents. We analysed our results after stent placement. Materials & Methods: We observed seven intrahepatic and 21 anastomotic stenoses in 22 patients. Surgery was performed in seven, retransplantation in five patients. Four endoscopic stents were used. We implanted 13 metallic stents to manage anastomotic and combined stenosis. In two cases we implanted two short stent grafts Results: We implanted metallic stents in cases when surgical or endoscopic therapy was ruled out. Three patients died with a patent stent within one month after stent implantation, one patient required reintervention, and the other patients are well. The two-stent grafts managed the bile leak and the stenosis sufficiently. Mean follow-up time is 25 months (1-74). Conclusion: Metallic stents are generally not recommended in benign biliary diseases, however, our long-term results are good. We recommend the use of such stents in cases when surgery, endoscopic stenting, or percutaneous dilatation is not an option.

P83

P85

Percutaneous treatment of hydatid liver cysts in pregnancy B. Üstünsöz1, N. K. Duru2, N. Bulakbasi1, M. Kocaoglu1, S. Görgülü3; 1Radiology, GATA, EtlÝk-Ankara, Turkey, 2Obstetric and Gynaecology, GATA, EtlÝk-Ankara, Turkey, 3General Surgery, GATA, EtlÝk-Ankara, Turkey.

Percutaneous management of biliary lithiasis C. Lanciego, L. García García; Unidad de Radiología Intervencionista, Hospital Virgen de la Salud, Toledo, Spain.

Purpose: To describe percutaneous management of hydatid liver cysts in pregnancy. Materials & Methods: A 26-year-old and a 28-year-old primigravida had abdominal ultrasound examination (Accuson 128 HP10), performed because of vomiting and right upper abdominal pain, which showed hydatid liver cysts. Both cases were treated with the PAIR (Puncture, Aspiration, Injection of hypertonic saline and Reaspiration) technique. Results: Both cases gave birth to healthy babies. The first case, who had a 20-cm Type 3 hydatid cyst, had no immediate or late complication. The second case, who had a 15-cm Type 1 hydatid cyst, had fever and headache probably due to the hypertonic saline for the first two days of therapy, which resolved with only supportive treatment. Conclusion: Percutaneous treatment of liver hydatid cysts is a good option compared with the surgical risks in pregnancy.

Purpose: To describe our experience in the percutaneous expulsion of biliary calculi into the duodenum by dilating the papilla with a ballooncatheter. Materials & Methods: Between January 1994 and January 2003, 212 patients received percutaneous treatment for calculi in the bile duct (101 men; mean age 73y; range, 31-95y); 73 of them (residual lithiasis) via a T-tube in situ. One-hundred thirty-nine patients had native or primary lithiasis and were treated with a transhepatic or transcystic route. Results: Initial technical success was 90.4% (n=169) at the first attempt, with five more cases added at a second attempt (93%). The 73 patients with residual lithiasis were resolved in 98.6% of cases; effectiveness reduced slightly to 92% in the group of native or non-residual lithiasis. Conclusions: Percutaneous anterograde elimination of bile duct lithiasis with dilation of the papilla and evacuation of the calculi using an occluding balloon is an effective, non-traumatic and safe technique. It can be considered in selected patients as an alternative therapy in the management of biliary lithiasis, especially in those cases in whom access to the bile duct is via an indwelling T-tube or a transhepatic catheter.

P84

P86

Percutaneous transhepatic thin cholangioscopy: experience with 80 cases R. Kumano, E. Yamanouchi, K. Kobayashi, T. Hattori, J. Matsumoto, I. Tani; Radiology, St. Marianna University Yokohama Seibu Hospital, Yokohama, Japan.

Feasibility of CT fluoroscopy-guided transhepatic biliary drainage: application to emergency patients with obstructive jaundice J. Kim; Diagnostic Radiology, Konyang University Hospital, Daejeon, Republic of Korea.

Purpose: Conventional percutaneous transhepatic cholangioscopy has been performed infrequently because it is necessary to dilate by a percutaneous route up to 16-18F, and several weeks are necessary to resolve the percutaneous access. We aimed to evaluate the efficacy of percutaneous transhepatic thin cholangioscopy, which does not need to dilate the route. Materials & Methods: Our thin cholangioscope is 8.4F in caliber, with 6000 crystal fibers, a 3.6F working channel, and a two-way movable tip. This cholangioscope has been used in 80 patients (37 males and 43 females). Results: Thirty-one of 80 patients had choledocholithiasis. Electrohydraulic lithotripsy was successfully performed in 19 of 31. Balloon lithotomy was successfully performed after balloon dilatation of the ampulla of Vater in 12 of 31 patients. Thirty-two of 80 patients had malignancy, confirmed by biopsy or surgery (25 of 32 patients), or diagnosed by ERCP, CT or MRI (seven of 32 patients). One case was a foreign bodies (surgical thread). The others were unknown cases. Conclusion: Thin cholangioscopy was clearly less invasive and quite efficient. It could be of value as a new instrument of interventional radiology to access the biliary tract.

Purpose: To evaluate the feasibility of CT fluoroscopy (CTF)-guided PTBD in patients with obstructive jaundice. Materials & Methods: Twenty-eight jaundiced emergency room patients were included. Bilirubin level was highly elevated, requiring emergency biliary drainage. All patients underwent CT to evaluate biliary dilatation and its causes. Immediately after CT scanning, CTF guided PTBD was performed without additional fluoroscopic guidance. The puncture site was the left lobe in 21 patients and the right lobe in seven patients. Moving the CT table to monitor the bile duct, a guidewire and drainage catheter were inserted. Success rates, number of needle insertions for bile duct puncture, and complications were evaluated. Results: CTF-guided PTBD was successful in 24 patients. Failure in three patients was due to poor respiratory control. One patient was moved to the fluoroscopy room to complete the procedure due to unusual guide wire twisting. In 15 patients, bile duct puncture was achieved on the first attempt. The others were successful within three trials. There were no significant complications. Conclusion: CTF-guided PTBD following diagnostic CT scan is useful in saving time and cost. It can be done safely with a high success rate, and can be an alternative method to conventional PTBD.

P87

P89

Efficacy of pre-operative portal vein embolization using polyvinyl alcohol particles R. T. Gandhi, K. T. Brown, A. M. Covey, H. V. Tengg-Kobligk, L. Brody, G. I. Getrajdman; Department of Radiology, Memorial Sloan-Kettering Cancer Center, New York, United States.

Management of hemobilia due to percutaneous biliary drainage L. Pichler, J. Malcher, K. Lomoschitz, W. Anzböck, W. Hruby; Department of radiology, Danube Hospital, Vienna, Austria.

Purpose: To determine the efficacy of pre-operative portal vein embolization (PVE) with polyvinyl alcohol (PVA) particles for inducing contralateral hepatic hypertrophy. Materials & Methods: From September 1999 to February 2003, 40 patients with primary or secondary hepatic malignancy but not known underlying cirrhotic liver disease underwent PVE with PVA prior to hepatectomy. Pre- and post-embolization liver volumes were calculated utilizing a 3-D workstation with manual and semiautomated segmentation techniques. Volumetric follow-up data for this group of patients is currently ongoing. Results: At the time of this abstract, volumetric analysis has been completed for 19 patients. Mean volume of the contralateral liver increased from a median volume of 647 cc (range 249-1809) before PVE to 800 cc (range 497-1699) after embolization. Mean increase in volume was 38%. Fifteen of these patients ultimately underwent surgery with curative intent without post-operative mortality or liver failure. Evidence of advanced intrahepatic or extrahepatic disease precluded curative resection in four patients. Conclusion: PVE with PVA is a simple and effective means of achieving contralateral liver hypertrophy in patients with small predicted future liver remnants following major hepatic resection. Hypertrophy of the remnant liver volume may reduce the risk of post-operative liver failure.

Purpose: To report frequency and treatment of severe vascular complications due to percutaneous transhepatic biliary drainage. Materials & Methods: Between January 1995 and December 2002, 396 patients underwent percutaneous biliary drainage (PBD). The most frequent indication was malignant obstruction. In 44 cases PBD followed complications after postoperative biliary or pancreatic surgery. Twentyone out of 396 patients developed hemobilia. Results: Intrahepatic arterial lesions were encountered in six cases and successfully managed by coil embolization. Serious venous bleeding was seen in 15 cases and was managed by upsizing the drainage catheter. Conclusion: Hemobilia after PBD is usually transient and without clinical consequences, but can occasionally be quite serious. Arterial bleeding is a rare but severe complication of PBD and can easily be treated by selective arterial embolization. Venous or portal venous bleeding can in most cases be managed by upsizing the drainage catheter.

P88

P90

Percutaneous placement of metallic stents in malignant hilar obstructions: do we need to place more than one stent? M. Inal, E. Aksungur, E. Akgül; Radiology, Cukurova University Faculty of Medicine, Adana, Turkey.

A covered stent for malignant biliary obstruction: five-year experience T. Yoshioka1, K. Furuichi2, H. Sakaguchi3, K. Kichikawa3, H. Ohishi4; 1Radiology, Nara Prefectural Nara Hospital, Nara, Japan, 2Radiology, Nara Prefectural Mimuro Hospital, Nara, Japan, 3Radiology, Nara Medical University, Nara, Japan, 4Oncoradiology, Nara Medical University, Nara, Japan.

Purpose: To evaluate the necessity of draining more than one hepatic duct in malignant hilar obstructions. Materials & Methods: Two-hundred and seventeen self-expandable uncovered metallic stents were placed percutaneously in 138 patients with unresectable malignant hilar obstruction. Uniductal drainage was achieved in 74 (54%) patients. In 64 (46%) patients, to maintain a biductal drainage, two stents were inserted either through dual transhepatic tracts in a "Y" configuration (n=41) or a single transhepatic tract in a "T" configuration (n=23). Long-term stent patency rates were calculated using the Kaplan-Meier method and log-rank tests were used to compare the patency rates among different groups. Results: There was no statistically significant difference between uniductal and biductal drainage in Bismuth type I, II and III obstructions. In Bismuth type IV obstructions, the patency rate was significantly higher in those patients in whom two stents were deployed in a "Y" configuration. Conclusion: There is no need to place more than one stent in Bismuth type I, II and III hilar obstructions. In Bismuth type IV obstructions, the deployment of two parallel stents through dual transhepatic tracts should be preferred.

Purpose: To assess retrospectively the use of a polytetrafluoloethylene covered stent (NT stent) in patients with malignant biliary obstruction. Materials & Methods: From February 1996 to March 2001, NT stents were placed transhepatically in 57 patients with malignant biliary obstruction (pancreatic cancer in 30, lymph node metastasis in 12, cholangiocarcinoma in 9, gallbladder cancer in 4, and others in 2). NT stent is constructed of elastic alloy with gold coating and is completely covered with the PTFE membrane on both inner and outer surfaces. The biliary ostructive site was extrahepatic in 55 patients and the anastomosis of the choledochojejunostomy in two. Results: Patency rates at six months was 88%, at one year 53%. Survival rate was37% at six months and 17% at one year. Obstructive jaundice recurred in 10 patients at 24 to 449 days after stenting. Causes of reobstruction were cholecystitis in three patients and tumor growth in seven including, overgrowth in four, ingrowth in one, and in- and overgrowth in two. Ten complications were observed. Conclusion: Our results suggest that use of NT stent reestablishes bile flow in the occluded biliary tree and that its efficacy and patency rate are also adequate.

P91

P93

Long-term results in the management of benign biliary strictures with long-term coaxial dilatation B. Üstünsöz1, Y. Pabusçu1, &. Ugurel1, &. Aslan2, D. Sen2, S. Bagci3; 1Radiology, GATA, EtlÝk-Ankara, Turkey, 2General Surgery, GATA, EtlÝkAnkara, Turkey, 3Gastroenterology, GATA, EtlÝk-Ankara, Turkey.

Self-expandable nitinol stent placement for malignant hilar biliary obstruction using a single access method T. Ihaya, J. Nakanishi, T. Ookawa; Department of radiology, Tottori Red Cross Hospital, Tottori, Japan.

Purpose: To present our result of long-term coaxial dilatations in the management of benign biliary strictures. Materials & Methods: Nine cases (six women, three men) with an average age of 51 underwent long-term coaxial biliary stricture dilatation with an increasing size of the external-internal biliary drainage catheter, beginning with 8 up to14 F every two months. All initial catheterizations were performed with Accustick sets and all catheters were secured with sutures and locking mechanisms. All catheters were exchanged with exchange guidewires under fluoroscopy guidance. The reasons for biliary strictures were video-laparascopic cholecistectomy in six cases, open surgery in two and chronic pancreatitis in one case. Results: Six cases were cured. Two cases showed a partial improvement but needed surgery. One case required a permanent catheterization. The mean follow-up time was 2.8 years. Most common problems were catheter-related complications (23) and infections (11). Life-threatining conditions did not occur and all the cases were followed as outpatients. Conclusion: Long-term coaxial dilatation of bening biliary strictures can be resolutive in most of the cases.

Purpose: To asscess the feasibility of biliary endoprosthesis in malignant hilar obstructions using self-expandable nitinol stents with a single access method. Materials & Methods: Sixteen patients with malignant hilar biliary obstruction were enrolled in this study. They consisted of eight cases of hilar cholangioma, three cases of nodal meatstases, three cases of gallbladder cancer, and one case of dissemination. Percutanous biliary drainage was performed followed by self-expandable nitinol stent placement. A hydrophilic wire was passed to the contralateral undrained lobe through the first stent mesh. The second metallic stent was placed end to side as a partial stent in stent after balloon dilation of the route. Results: Most of the cases were successfully treated with a single access method. Contralateral approach was required in only one case, because of the steep angle of hepatic ducts. The external drainage catheter could be removed in all cases. No major complications were encountered. During a mean follow-up of 3.2 months, recurrent jaundice was observed in three cases because of tumor ingrowth and/or overgrowth. Conclusion: Self-expandable nitinol stent placement by a single access method may be a feasible and less invasive technique in the treatment of malignant hilar biliary obstruction.

P92

P94

Portal embolization prior to surgery to induce hypertrophy of the future remaining liver R. D. Garcia Monaco, H. G. Bertoni, O. Peralta, M. M. Buzzi, D. Cassina, E. De Santibañez; Interventional Radiology, Hospital Italiano, Buenos Aires, Argentina.

Liver abscess secondary to hepatic arterial chemoembolization of hepatocellular carcinoma: percutaneous CT-guided drainage L. Thanos, G. Papaioannou, A. Nikita, E. N. Brountzos, K. Malagari, A. Sissopoulos, D. A. Kelekis; 2nd Department of Radiology, University of Athens, Athens, Greece.

Purpose: To report our experience with preoperative portal embolization (PPE) to induce hepatic regeneration and hypertrophy of the future remaining liver. Materials & Methods: Between 1999 and 2002, 12 PPE procedures were performed. The procedure was indicated when hepatectomy had an increased risk of postoperative hepatic insufficiency. Three patients had primary tumors and nine had hepatic metastases. PPE was performed by selective catheterization of each intrahepatic portal branch with distal occlusion. The procedures were done percutaneously under ultrasound guidance in three cases, and by laparotomy with catheterization of a colonic vein in nine patients. The embolization material was cyanocrilate mixed with ultrafluid lipiodol. After four weeks all patients underwent CT scan to evaluate hypertrophy of the remaining liver. Results: The procedure was technically successful in all cases. The average volume increase of non embolized hepatic segments was equivalent to 35% growth of the future remaining liver. In three patients, tumors could not be removed because of disease progession. Conclusion: Preoperative portal embolization leads to contralateral liver hypertrophy, decreasing the incidence of postoperative liver failure. Both percutaneous and surgical approches of PPE have their indications; a larger number of patients are required to define the best approach.

Purpose: To assess the role of percutaneous CT-guided drainage of liver abscesses which develop secondary to hepatic arterial chemoembolization. Materials & Methods: During a six-year period, 327 hepatic arterial chemoembolizations (HACE) were performed in 147 patients with hepatocellular carcinoma.. Five patients developed liver abscess inside the embolized tumor after one to six weeks. Four of them were managed with percutaneous CT-guided drainage. Results: Abscesses were successfully drained in all patients, using pigtail 12F and 14F catheters. No complications were noted. Mean drainage time was 10 days, during which time patients presented clinical improvement and normalization of laboratory tests. Aspiration of the lesions produced purulent material that was sent for culture. The latter revealed a causative microrganism in 4/5 patients and enabled targeted antibiotic therapy. Conclusion: Liver abscess that complicates HACE can be successfully treated with percutaneous CT-guided drainage.

P95

P97

Treatment of malignant biliary obstruction with self-expandable metallic stents: retrospective comparison of Wallstents and Memotherm stents S. Geyik, D. Akinci, V. Gelebek, M. Özmen, O. Akhan; Department of Radiology, Hacettepe University School of Medicine, Ankara, Turkey.

Transcatheter subsegmental arterial chemoembolization under balloon occlusion of the hepatic vein M. Okazaki1, H. Higashihara1, T. Sakamoto2, S. Furui3; 1Department of Radiology, Fukuoka University, Fukuoka, Japan, 2 Department of Radiology, Kohga Hospital, Shiga, Japan, 3Department of Radiology, Teikyo University, Tokyo, Japan.

Purpose: Retrospective analysis of the outcomes of treatment of malignant biliary obstruction with two different self expandable metallic stents in a single center. Materials & Methods: Between 1995 and 2001, 179 self-expandable metallic biliary stents were successfully placed in 161 patients with malignant biliary obstruction. Wallstents and Memotherm stents were used in 125 and 36 patients respectively. Stent patency, complications, and patient survival rates were analyzed retrospectively Results: Mean follow-up was 6.5 months (range: 1-26) and 4.3 months ( range: 1-12) for Wallstent and Memotherm stent groups, respectively; mean stent patency durations were 6.7 months ( range: 1-18) and 4.1 months (range: 1-10) respectively. Eleven patients in the Wallstent group and six patients in the Memotherm group required re-interventions due to reocclusion. All except two were treated using second stents. Balloon dilatation was performed in two of them. Thirty-day mortality rate was 6 %. There was no procedure-related mortality. Major and minor complications rates were 8 % and 18 .5 %, respectively, for all patients. Conclusion: Self-expandable metallic stents provide good palliation in patients with malignant biliary obstruction. Wallstent stents provides a longer patency than Memotherm stents.

Purpose: The efficacy and pitfall of a newly devised subsegmental chemoembolization under balloon-occlusion of the hepatic vein (NDSTACE) to achieve complete necrosis with the simultaneous embolization of both the hepatic artery and the portal vein were evaluated. Materials & Methods: Ninety-three patients with 105 hepatocellular carcinomas limited to a single subsegment were studied. At angiography, all the lesions appeared as nodular stains. A microcatheter was introduced into the tumor-bearing subsegmental feeding artery while a balloon-occlusion of the hepatic vein was carried out. NDS-TACE was then performed through the microcatheter. Embolic materials used were the same as those employed in conventional TACE. Results: Lipiodol accumulations in the tumor and the portal vein were found in all patients. Regrowth of the embolized tumors was recorded within three years in all cases. (Five-year recurrence-free: 80.1%). Survival rates of the 86 patients treated with this method alone were 84.8% at three years and 70.3% at five years. Conclusion: Our results suggest that with this procedure it is possible to easily embolize both the hepatic artery and the portal vein branch in the tumor-bearing subsegment alone.

P96

P98

Radiofrequency thermal ablation of malignant hepatic tumors guided by contrast-enhanced ultrasonography G. Ferreri, G. Di Giambattista, D. Lupoi, P. Di Renzi, C. Simonetti, A. Orlacchio; Radiology, Fatebenefratelli Hospital, Rome, Italy.

Better pain control by blocking the celiac nerve bundle in laserinduced thermal therapy of liver tumors/metastases A. N. Beck, J. Ricke, M. Werk, M. Pech, B. Spors, M. Schäfer; Department of Radiology, Charité, Berlin, Germany.

Purpose: To evaluate the use of contrast media injection during ultrasonography (US) to: 1. improve detectability of hepatic lesions; 2. guide the percutaneous approach for thermal ablation; 3. identify residual viable tumors for a second approach. Materials & Methods: Thirty patients with hepatic tumors (12 metastases and 18 hepatocarcinomas) underwent thermal ablation with contrast media injection to identify those lesions not detected with conventional exams. The patients were also evaluated by contrastenhanced US at follow-ups. Sonovue (Bracco SpA) contrast medium (2.4 to 4.8 ml) was used. Results: The use of contrast medium allowed the detection of all lesions with a good accuracy, thus facilitating thermoablation and placement of the electrode needle. In those patients in whom a persistent vascular flow signal within the tumors was demonstrated by contrast medium at follow-up, a second thermoablation was performed. Conclusion: According to our experience, contrast-enhanced US has improved the detection of many hepatic lesions, especially hypervascular ones. Contrast media can be also very useful in detecting residual or recurrent tumors after radiofrequency ablation and in guiding a second approach.

Purpose: To demonstrate a significant pain reduction obtained by blocking the celiac nerve bundle before laser-induced thermal therapy (LITT), thus minimizing interventional times and maximizing local tumor control rates and enabling a complete tumor ablation. Materials & Methods: We compared pain indices during and after the intervention, consumption of pain killers, and intervention times. Until today, 49 consecutive patients have been recruited; 24/49 have been randomized to receive blockage of the celiac nerve bundle. All the patients received additional pain killers, if needed. Results: By blocking the celiac nerve bundle before LITT, the need of opiates reduces drastically thus lowering the pain level. This procedure also shortened significantly intervention times. The need of additional pain killers lowered significantly in those patients receiving the blockage (average, 250yg of Fentanyl) than in those who did not (average, 348 yg Fentanyl). The intervention time was significantly shorter when the blocking of the celiac nerve bundle was performed. Conclusion: Blocking the celiac nerve bundle before performing a LITT drastically lowers the pain level and the need of additional pain killers. It makes the intervention more comfortable for the patients and might provide a further improvement in LITT local tumor control rate.

P99

P101

Two-year follow-up of percutaneous CT-guided radiofrequency thermal ablation of primary and metastatic liver tumors L. Thanos, S. Mylona, M. Pomoni, V. Kalioras, A. Fidiarakis, N. Batakis; Radiology, Korgialeneio-Benakeio Hellenic Red Cross Hospital of Athens, Athens, Greece.

Fluoroscopy-guided hepatic radiofrequency ablation J. Urbano1, M. A. Arjonilla1, C. Arocena2, B. Polo2, A. Gonzalez-Guirado2, O. Bosch2; 1 Radiology, Fundación Jimenez Diaz, Madrid, Spain, 2Hepatology, Fundación Jimenez Diaz, Madrid, Spain.

Purpose: A retrospective study of 43 patients with primary or metastatic liver tumors, treated with radiofrequency thermal ablation (RFA) in a 2-year period. Materials & Methods: We applied CT-guided RFA in 24 patients with HCC and 19 with metastases. Eighty-eight sessions required (36 for HCC and 56 for metastases). We used a Rita device with 7-9 unfoldable electrodes. Each session lasts 15-17 min with a target temperature set at 950C. Lesion diameter was 1-6 cm. Follow-up with dual-phase spiral CT was made immediately after the procedure and 1,3,6,12, and 24 months later, also checking alpha-fetoprotein and CEA. Results: Twenty (83.3%) patients with HCC had total necrosis and 4(16.7%) of them needed a new session because of tumor recurrence. Total necrosis occurred in 90% (9/10) of unifocal and in 81.8% (18/22) of multifocal metastases. New metastases were detected in 8 (42.2%) patients Alpha-fetoprotein decreased in 14 (87.5%) of 16 patients and CEA in 9(81.8%) of 11 patients with high pre-treatment levels. There were no major complications. Conclusion: Percutaneous CT-guided RFA of primary and metastatic liver tumors is a safe and feasible alternative treatment in patients with inoperable lesions.

Purpose: US or CT have classically guided radiofrequency ablation. We present our experience with hepatic radiofrequency guided exclusively by fluoroscopy. Materials & Methods: Twenty-six hepatocellular carcinomas in 17 patients were treated. Mean nodule size was 45 mm (10-90 mm). Cool tip electrodes were used in all cases and a mean of 2,4 (2-6) ablations were performed in each nodule. All tumors had been previously embolized with lipiodol and were easily visualized on fluoroscopy. Movement of the C-arm confirmed the location of the needle inside the nodule. Results: The electrode was placed inside the nodule in all cases. No complications arose during the procedure or the outcome. Six nodules showed signs of incomplete ablation and two of those were retreated. All the rest were completely treated and there have been no signs of recurrence during follow up for a mean period of 13 months (1-24) Conclusion: Fluoroscopy guided radiofrequency ablation is a feasible and safe technique in the treatment of hepatocellular carcinoma following embolization with lipiodol. Advantages include an easier approach to deep lesions, accurate relocation of the needle avoiding bubble artifact seen when using US; previous embolization also improves the therapeutic effect.

P100

P102

The effects of the dosage of anticancer agents on T-cell subsets during transcatheter arterial chemoembolization in hepatocellular carcinoma Y. LI, W. Lu; Department of Interventional Radiology, NanFang Hospital,Guangzhou,China, Guangzhou, China.

Chemoembolization in one case of gallbladder unresectable carcinoma M. Cerna1, M. Köcher1, H. Svebisova2, R. Havlik3, E. Buriankova1; 1Department of Radiology, University Hospital, Olomouc, Czech Republic, 2Department of Oncology, University Hospital, Olomouc, Czech Republic, 3Department of Surgery, University Hospital, Olomouc, Czech Republic.

Purpose: To investigate the effects of the dosage of anticancer agents on the changes of host T-cell subsets during transcatheter arterial chemoembolization (TACE) in hepatocellular carcinoma (HCC). Materials & Methods: Thirty-six patients with unresectable HCC were randomly divided into two groups to receive superselective TACE. Patients in group A (n=18) received low-dose anticancer drugs: mitomycin C (MMC)2-8 mg, epirubicin (EPI) 5-10 mg and carboplatin (CBP) 50-100 mg. Patients in group B (n=18) were given conventional-dose anticancer drugs (MMC 10 mg, EPI 40 mg and CBP 300 mg). The peripheral blood T cell subsets including CD3+-CD4+-CD8+-NK-CD4+/CD8+ ratio-CD4+CD45+-CD4+CD29+-CD8+CD28+-CD8+CD28- in both groups were measured by flow cytometry before and one week after treatment. Results: There was no difference between the two groups in T-cell subsets before the treatment. After TACE, no significant changes were found in CD3+-CD4+-CD8+-NK-CD4+/CD8+-CD4+CD29+-CD8+CD28- in group A, while significant decreases in CD4+CD45+ and increases in CD8+ CD28+ were observed (p<0.05 and 0.001, respectively). In group B, CD4+, CD4+CD29+ level and CD4+/CD8+ ratio were significant lower than before the treatment (p<0.05), and CD8+ and CD8+CD28- subsets significantly higher (p<0.05). Conclusion: The cellular immune function of HCC patients is weakened by conventional-dose anticancer drug TACE. Low-dose anticancer drugs should therefore be recommended.

Purpose: To report the case of a 60-year-old man with unresectable adenocarcinoma of the gallbladder treated by chemoembolization. Materials & Methods: Chemoembolization is not indicated in patients with unresectable adenocarcinomas of the gallbladder; chemoembolization of the cystic artery, in fact, represents a complication of this procedure performed for unresectable malignant liver tumors. After considering all possible difficulties, the patient was however treated by superselective chemoembolization of the cystic artery with a suspension of doxorubicine and Lipiodol. Results: The procedure was technically successful. Serious complications due to the embolization of the cystic artery were not recorded. After the first chemoembolization, the patient showed a decrease in serum tumor markers level and a regression in the tumor size was depicted by CT scans. Conclusion: Chemoembolization of inoperable gallbladder carcinomas is technically possible and, in selected patients, it could be considered as a palliative treatment which can also provide long-term survival.

P103

P105

Percutaneous radiofrequency thermal ablation combined with transcatheter arterial chemoembolization for non-resectable hepatocellular carcinoma larger than 3cm in diameter J. Nakanishi1, T. Ihaya2, T. Ogawa1; 1Radiology, Tottori University Hospital, Yonago, Japan, 2Radiology, Tottori Red Cross Hospital, Tottori, Japan.

CT-guided radiofrequency thermal ablation of primary lung tumors: initial experience L. Thanos1, S. Mylona1, V. Kalioras1, L. Zoganas2, M. Pomoni1, N. Batakis1; 1 Radiology, Korgialeneio-Benakeio Hellenic Red Cross Hospital of Athens, Athens, Greece, 2Thoracosurgery, Korgialeneio-Benakeio Hellenic Red Cross Hospital of Athens, Athens, Greece.

Purpose: To assess the clinical efficacy of the combination of percutaneous radiofrequency thermal ablation (RFA) and transcatheter arterial chemoembolization (TACE) in the treatment of hepatocellular carcinoma (HCC) larger than 3cm in diameter Materials & Methods: Seventeen HCC nodules ranging from 3.2 to 8 cm in diameter have been treated with TACE followed by a single session of RFA. We performed superselective TACE with gelatin sponges after injection of an EpiADM/Lipiodol mixture. RFA was performed with an expandable needle electrode under US and/or CT guidance. Response to therapy was evaluated by imaging studies four weeks after the treatment. Results: During a mean follow-up of 7.9 months, complete necrosis was obtained in cases except two local recurrences in nodules larger than 7 cm in diameter. One-year survival rate was 88.9% and mean survival period was 486.1 days. There were no major complications, but minor complications such as abdominal pain, fever and transient elevation of liver enzymes did occur. Conclusion: RFA combined with TACE might contribute to improved local control for non-resectable HCC larger than 3 cm. Some modifications may be necessary in the treatment of HCC larger than 7 cm diameter by single RFA procedures after TACE.

Purpose: To present the technique, effectiveness and possible complications of radiofrequency thermal ablation (RF) of primary inoperable lung tumors. Materials & Methods: During a six-month period, we applied RF under CT-guidance on eight patients with non-small cell lung carcinoma. We used a RITA device and a trocar with seven electrodes. During the ablation, which lasts 15-18 minutes, the target tissue temperature was set at 95oC. The procedure was performed after local anaesthesia at the puncture point. Follow-up with dual-phase spiral CT after IV administration was performed immediately after the procedure and 1, 3, and 6 months later. Results: In all patients, the first CT control showed increased attenuation around the needle tract and decreased attenuation of the tumor. At one and three months, we noticed a clearly outlined region without contrast uptake, with peripheral annular contrast medium enhancement. Tumor necrosis was complete in five patients and segmental in three. At the six-month follow-up, tumor size was stable in five, smaller in two, and larger in one patient. There was one pneumothorax and one hemorrhage but no major complications ). Conclusion: Radiofrequency thermal ablation is an effective and safe alternative method for the treatment of primary inoperable lung tumors.

P104

P106

The role of chemoembolization and radiofrequency thermal ablation in percutaneous treatment of inoperable primary hepatocellular carcinoma and liver metastases M. Pomoni, L. Thanos, S. Mylona, V. Kalioras, A. Tsanis, N. Batakis; Radiology, Korgialeneio-Benakeio Hellenic Red Cross Hospital of Athens, Athens, Greece.

Unconventional application of Angio-Seal devices in vascular interventional procedures M. Manzi, A. Marangon, G. Mansi-Montenegro, G. Torsello; Department of diagnostic imaging, Institute of Radiology, S. Bortolo Hospital, Vicenza, Italy.

Purpose: To present an interventional therapeutic protocol for inoperable primary hepatocellular carcinoma (HCC) or liver metastases, including radiofrequency thermal ablation (RF) and chemoembolization according to lesion characteristics. Materials & Methods: In a period of 3 years, we treated 51 patients with HCC or liver metastases, selected according to the following criteria: - Unifocal HCC under 6 cm in diameter: RF; - Unifocal HCC larger than 6 cm in diameter: Chemoembolization and, in selected cases, RF; - Multifocal HCC: Chemoembolization of large lesions and RF of smaller ones; - Liver metastases under 5 cm in diameter: RF; - Liver metastases larger than 5 cm (especially hypervascular ones): Chemoembolization Results: All lesions had good response in both procedures, with total or segmentally necrosis at 6-month follow-up using dual-phase spiral CT. Eighty-five percent of treated lesions were stable or smaller in size. There were no major complications (mild right-upper-quadrant pain in the puncture region after RF was treated by analgesics; one hepatic abcess required drainage, after chemoemolization). Conclusion: Percutaneous treatment of inoperable primary and metastatic liver diseases using a combination of chemoembolization and radiofrequency thermal ablation is another alternative treatment in cancer patients

Purpose: To verify the efficacy of using Angio-Seal devices in closing unconventional vascular approaches for interventional procedures. Materials & Methods: From February 2001 to December 2002, 226 consecutive Angio-Seal devices were deployed in antegrade CFA approaches,14 in Cimino fistula arterialized veins, 16 in the popliteal artery (eight via the retrogade and eight via the antegrade approach). Two devices were deployed both in CFA and two in the popliteal artery during the same procedure. Results: All the vascular accesses were promptly sealed, with no acute thrombosis. Hemathomas were evident in 0.5% after one week. One patient had persistent pain for 10 days after CFA deployment. Conclusion: Angio-Seal is in our experience, a safe and effective device even in unconventional vascular approaches, leading to good blood flow maintenance in revascularized districts.

P107

P109

Application of a specifically designed guide-stylet set in thoracic biopsies T. Petsas1, I. Tsota1, C. P. Kalogeropoulou1, A. C. Tsamandas2, Z. Papathanasiou1, I. Dimopoulos1; 1 Department of Radiology, University of Patras, Patras, Greece, 2 Department of Pathology, University of Patras, Patras, Greece.

Treatment of epiphora with nasolacrimal stents: final results of a multicenter study with six years of experience C. Lanciego1, S. De Miguel2, M. Padilla3, M. Perea2, C. Cano4, I. Ruiz5, N. Toledano6, J. Dávila7, R. Rodriguez-Merlo8, L. García García1; 1 Unidad de Radiología Intervencionista, Hospital Virgen de la Salud, Toledo, Spain, 2Oftalmología, Hosp.V.de la Salud, Toledo, Spain, 3Orl, Hosp.V.de la Salud, Toledo, Spain, 4Unidad de Radiología Intervencionista, H.General Yagüe, Burgos, Spain, 5Oftalmología, H.General Yagüe, Burgos, Spain, 6Oftalmología, H.General Móstoles, Madrid, Spain, 7 Unidad de Radiología Intervencionista, H.General Móstoles, Madrid, Spain, 8Anat.Patológica, Hosp.V.de la Salud, Toledo, Spain.

Purpose: Presenting our experience with the use of a new guide-stylet set for thoracic biopsies. Materials & Methods: Sixty-six patients underwent chest biopsies of lung (n=41), mediastinal (n=13), spinal and rib lesions (n=12) using this new guide-stylet set. The guide-stylet, made of stainless steel, has a 26G diameter and is combined with a 22-G Chiba needle (William Cook Europe). Following needle insertion, the stylet is directed through it towards the lesion. After the stylet has been correctly directed, the needle is advanced over the stylet to reach the target. The stylet serves either as a guide for the needle, or as an exchange wire when a larger needle is required. It can be curved before insertion inside the 22-G needle for reaching those lesions which require a biopsy needle redirection. Results: The technique was easily performed in all cases. It was effective in small lung tumors and especially in mediastinal and vertebral lesions. Six patients with lung biopsies developed pneumothorax (14.63%). The stylet was advanced beyond the lesion in seven cases; in three cases this led to a mild parenchymal hemorrhage without consequences. Conclusion: The guide-stylet is a useful complementary tool for better targeting difficult thoracic lesions.

Purpose: To report our experience with polyurethane stents for epiphora. Materials & Methods: In 435 patients from three hospitals, over a sixyear period, nasolacrimal stents have been inserted for severe epiphora. Five-hundred and ten stents were implanted in 480 lacrimal systems to treat complete (n=460) or partial (n=20) obstruction of the sac or the nasolacrimal duct. In this latter case, a modification of the Song stent (by creation of an opening in the proximal end) was used. Results: Technical success was 96%. At a mean follow-up of 28 months (one week-69 months), 349 stents remained patent. Mean primary patency was 68.5%, (75.6, 68.4, 64.8, 60.1, 59.2 at 1, 2, 3, 4, 5, and >5 years, respectively). Obstructions were 161 (31.4%) and 119 stents were retrieved at a mean follow-up of 27 months (1.5-53); during this time, 32% of patients had remained asymptomatic. We employed 24 "modified" stents with excellent results. Conclusions: Results are acceptable in relation to primary patency. However, obstruction of the stent remains a problem which reduces the efficacy of the technique. A better stent design and anti-fibrin agents might increase the confidence in a technique which, we believe, continues to be a valid alternative to conventional dacryocystorhinostomy.

P108

P110

Estimating vasculogenesis following the method of Ilizarov in an experimental sheep model D. Siablis1, M. Tillianakis2, I. Tsota1, A. Papadopoulos2, A. Karageorgos2, D. Karnabatidis1, G. Sakellaropoulos3, G. Nikiforidis3; 1Department of Radiology, University of Patras, Patras, Greece, 2Department of Orthopaedics, University of Patras, Patras, Greece, 3Department of Medical Physics, University of Patras, Patras, Greece.

Percutaneous radiologic gastrostomy in patients with respiratory difficulty K. Lee, J. Y. Won, D. Y. Lee; Diagnostic Radiology, Yonsei University College of Medicine, Seoul, Republic of Korea.

Purpose: The Ilizarov technique--an established method for limb elongation--has been proposed as a strategy to generate vasculogenesis for the salvage of soft tissues following critical ischemia. The effectiveness of this method to induce vasculogenesis in an experimental animal model was evaluated. Materials & Methods: Tibial shaft osteotomy and callus distraction using an Ilizarov external fixator was performed in 12 skeletally mature female sheep. All sheep were evaluated by color-Doppler sonography on the 30th distraction day. Studies of blood flow parameters were performed in the vessels of the distracted and the contralateral normal limb, as a control. The external fixator was then removed. Forty days later, blood flow parameters were reevaluated and DSA were performed. Results: Blood flow parameters (peak systolic velocity and volume flow) on the 30th day had increased in the elongated limb as compared with the control limb.On the 70th day, flow parameters of the distracted limb had decreased, but still remained higher than those of the control limb. DSA revealed an increase of the total vessel area and length in the distracted limb. Conclusion: Our results suggest that the Ilizarov method increases the vascularity of the elongated limb, which remains active 40 days after distraction completion.

Purpose: To evaluate the safety and feasibility of percutaneous radiologic gastrostomy in patients with respiratory difficulty. Materials & Methods: During two years, 22 patients with respiratory difficulty underwent percutaneous radiologic gastrostomy. The causes of respiratory difficulty were: amyotrophic lateral sclerosis in 15 patients, post-tracheostomy in four, tracheo-esophageal fistula in two, and myasthenia gravis in one patient. Results: A technical success was achieved in all the patients. Pneumoperitoneum occurred in two patients; one of them expired due to the stopping of the respiratory assistance aimed at eliminating the possible source of the aggravating pneumoperitoneum. The other patient with pneumoperitoneum showed a spontaneous absorption. A transient gastric bleeding occurred in one patient and ceased after cold saline irrigation. Conclusion: Percutaneous radiologic gastrostomy is a feasible procedure, althought it should be kept in mind that uncontrolled pneumoperitoneum might aggravate the respiratory difficulty.

P111

P113

In vivo comparative study of hind limb ischemia using volume flow

Large digital flat panel technology in interventional radiology: first clinical experiences H. Rousseau, P. Duthil, P. Otal, K. Benali, A. Lowe, L. Despond, P. Gobert, F. Joffre; Radiology, CHU Rangueil, Toulouse, France.

measurements and quantification of digital angiography D. Siablis1, G. Sakellaropoulos2, N. Samaras1, D. Karnabatidis1, I. Maroulis3, E. Sourgiadaki4, G. Nikiforidis2, I. Dimopoulos1; 1 Department of Radiology, University of Patras, Patras, Greece, 2 Department of Medical Physics, University of Patras, Patras, Greece, 3 Department of Surgery, University of Patras, Patras, Greece, 4 Department of Anaesthesiology, University of Patras, Patras, Greece. Purpose: Animal models of hind limb ischemia are very important for understanding peripheral vascular diseases and for estimating the therapeutic methods for the improvement of collateral circulation. A rabbit’s hind limb model was used to estimate revascularization following ischemia and to evaluate the capability of volume flow (VF) measurements to monitor this process in correlation with the angiographic quantification of the vascular area and length. Materials & Methods: In 24 male New Zealand white rabbits (4 kg) the femoral artery was resected and VFs of the external iliac artery were measured (every five days for 40 days). Angiographies were performed at days 10, 20, and 40. After computerized analysis of the angiographic images and quantification of the vascular area and length, the results were compared with VF measurements. Results: VF measurements were comparable with angiographic findings. The peak of angiogenesis after operation was recorded from day 0-15, it decreased during days 15-25 and remained stable during days 25-40, when minimal changes of the vascular bed and VF rates were observed. Conclusion: VF measurement is an efficient non-invasive tool for monitoring the ischemic limb revascularization process; it can be used for clinical purposes and for estimating the efficiency of revascularization strategies.

Purpose: Since the first angiography procedure in the fifties, cardiovascular imaging has benefited from consistent but relatively modest technological advances. The demand for the highest possible image quality was never as heavy as today. Materials & Methods: GE Medical Systems has recently introduced in the market Innova 4100, the world’s first large flat panel angiography system equipped with a 41-x-41 cm digital detector. By converting the incident x-ray photons into digital images at the detector, the all-digital technology replaces the image chain components of conventional systems. Results: The intent of this work is both to provide an overview of the digital flat panel technology, its advantages and potential limitations in radiology, and to introduce the reader to the new clinical applications this emerging technology will bring into the cath lab. Conclusion: Additionally, we report the results of our first investigations on dose-monitoring and image-quality based on our routine daily practice.

P112

P114

Assessment of microembolism during percutaneous radiofrequency ablation for lung tumors by diffusion-weighted and ultrasound imagings A. Yamamoto1, T. Matsuoka1, M. Toyoshima1, T. Okuma1, Y. Oyama1, K. Kono1, K. Nakayama1, K. Inoue2, K. Nakamura1, Y. Inoue1; 1Radiology, Osaka City University, Osaka, Japan, 22nd surgery, Osaka City University, Osaka, Japan.

Embolization of refractory epistaxis M. Gimeno, J. Medrano, R. Alfonso, M. Herrera, A. Mainar, M. De Gregorio; Interventional radiology, Clinico Universitario Lozano Blesa, Zaragoza, Spain.

Purpose: Percutaneous radiofrequency ablation (RFA) is reported to generate microbubbles that may cause embolism. The purpose of the present study was to assess the risk of cerebral microembolism during percutaneous RFA for lung tumors. Materials & Methods: From June 2000 to January 2003, 40 tumors in 20 patients were treated with RFA using CT-guidance with LeVeen needles. In 14 of them both ultrasound (US) imaging and diffusionweighted imaging (DWI) were carried out. With US, we continuously monitored the right or the left carotid artery during the RFA procedure for the detection of possible microemboli. DWI was performed before and after a median of 20 hours post-procedure. DWIs were independently interpreted by two radiologists. Results: Microemboli were detected by US in three patients. Post-operative DWIs did not show abnormal high-intensity areas in any patient. After the RFA procedure, clinical neurologic symptoms were not recorded. Conclusion: In percutaneous RFA, microemboli were detected by US in the carotid artery, but cerebral infarctions and clinical symptoms were not observed. According to the present study, therefore, RFA for lung tumors seems to be safe.

Purpuse: To evaluate the efficacy and safety of transcatheter embolization in the treatment of intractable epistaxis. Materials & Methods: From October 1999 to January 2003, 15 patients (12 men, 3 women) with intractable epistaxis were refered to our service for performance of endovascular therapy. All patients were treated initially with anterior and posterior nasal packing. Ten had idiopathic recurrent epistaxis, three Osler-Weber-Rendu syndrome, one HIV thrombocytopenia, and one nasopharyngeal carcinoma. Embolization materials included microcoils and polyvinyl alcohol particles and embospheres. Results: Technical success was 100% and immediate clinical success 100%. Sphenopalatine arteries were embolized in 13 patients and the superior meningeal artery in one, with immediate cessation of nasal hemorrhage. The oncologic patient presented a jaw fistulous tract with direct bleeding of facial and lingual arteries, ceasing after embolization. Because of acute recurrent bleeding in this case, a second embolization was performed in the external carotid, with cessation of bleeding. Two patients presented neck swelling and facial pain, treated with analgesics. Conclusion: Transcatheter embolization of refractory posterior epistaxis carries a low morbidity rate when performed with anatomical considerations. It is an excellent, safe, and less invasive alternative to surgery in patients with intractable epistaxis.

P115

P117

Three-dimensional gadolinium-enhanced MR-angiography for establishing venous thrombo-embolic disease: "one-stop-shop" imaging of pulmonary arteries, vena cava, pelvic and both lower extremity veins in 30 minutes M. Aschauer1, A. Obernosterer2, H. R. Portugaller3, H. Deutschmann3, R. Stollberger1, F. Ebner1; 1 Magnetic Resonance Institute, University Hospital Graz, Graz, Austria, 2 Department of Internal Medicine, University Hospital Graz, Graz, Austria, 3Department of Radiology, University Hospital Graz, Graz, Austria.

Quantitative assessment of the humeral head vascular integrity with selective DSA and correlation with the risk of avascular necrosis in displaced fractures D. Siablis1, M. Tillianakis2, A. Panagopoulos2, P. Kraniotis1, G. Sakellaropoulos3, D. Karnabatidis1, I. Tsota1, G. Nikiforidis3; 1 Department of Radiology, University of Patras, Patras, Greece, 2 Department of Orthopaedics, University of Patras, Patras, Greece, 3 Department of Medical Physics, University of Patras, Patras, Greece.

Purpose: To assess feasibility and image quality in the diagnosis of pulmonary air embolism (PAE) and venous thrombosis in a "one-stop-shop" fashion after that PAE has been diagnosed maximum 48 hours before. Materials & Methods: Fifteen patients underwent a "one-stop-shop" MR pulmonary angiography (PA) and a repeated three-station MRangiography (MRA/MRV). Parameters: 1.5T; 1. PA: TR/TE/flip angle: 2.4 ms/1.04 ms/20°, THK 3 mm, 32-40 slices, duration 5-6.4 sec., scan repeated five times (15 ml GdDTPA). 2. Peripheral MRA/MRV: pelvic, thigh and calf. Gd-DTPA volume: 0.6 ml/kg minus 15 ml, 1 ml/sec. for 10 ml, 0.5 ml for the rest. Scan was repeated four times using automatic table movements. Results: The detection of thrombi confirmed the diagnosis in 13/14 patients. One patient had fibrinolysis and thrombi resolved completely within 48 hours, as confirmed by an additional CT after MRA. MRV was more sensitive in detecting thrombi: in two patients additional large thrombi were found in the internal iliac vein, in three, gastrocnemius vein thrombi were found. The detection of pelvic thrombi was much more reliable with MRV than with conventional venography. Conclusion: High sensitivity and specificity in detecting acute pulmonary emboli by MRA can be confirmed in a "one-stop-shop" fashion with MRV.

Purpose: To assess the vascular integrity of the humeral head in displaced 4-part valgus impacted fractures of the proximal humerus treated with osteosuture. Materials & Methods: Fourteen consecutive patients with 4-part valgus impacted fractures of the proximal humerus were studied. The average impaction angle was 42.4° while the lateral displacement of the humeral head ranged between 1 and 7 mm. Preoperative selective DSA of the brachial artery was performed six to 12 hours after admission using three standard views. Fracture fixation was achieved with osteosuture. Angiography was performed again, in 12 patients, eight to ten weeks following surgery. Results: Digital image processing, using the segmentation technique, showed no statistical difference in the humeral head blood supply preand post-operatively. The average lengths of the vessels and the overall area of blood supply (vessels/mm2) were almost the same (p<0.01) in 11 patients. During follow-up, no signs of avascular necrosis were observed 18-20 months post-operatively. One patient presented with partial avascular necrosis. Conclusion: Selective DSA of the brachial artery is an effective method for the evaluation of the vascular integrity of the proximal humerus and correlates well with the risk of avascular necrosis in displaced fractures.

P116

P118

Renal apoptosis changes after CT and MR contrast medium injections in rats with acute renal failure Y. Han, Y. Lee, G. Jin, G. Chung; Radiology, Chonbuk National University Hospital, Chonju, Republic of Korea.

Sonographic and Doppler spectrum analysis of percutaneous arterial closure (Angio-Seal) devices H. Chun, B. Choi, J. Byun; Department of Radiology, Kangnam St. Mary's Hospital, Seoul, Republic of Korea.

Purpose: To evaluate renal toxicity through the renal function changes and the degree of apoptosis in renal tissue after contrast injection in rats with acute renal failure (ARF). Materials & Methods: We divided CT and MR groups into three subgroups: low dose; standard dose; and high dose, for a total of six groups with ten rats in each group. The 48-hour creatinine level was compared in each group after contrast injection, using one-way ANOVA tests in phase I. Apoptosis was examined in the renal tissues of the 21 ARF rats and in three rats in the control, CT contrast medium, and MR contrast medium groups in the phase II experiments. Apoptosis was compared in each group by one-way ANOVA tests. Results: Both images showed good corticomedullary delineation in the standard and high-dose injections. Creatinine levels did not differ significantly between the CT and MR groups (p=0.116). The CT contrast medium group showed severe apoptosis in the renal cortex and medulla as compared to the control group and the MR group (p<0.05). Conclusion: In apoptosis of the renal cortex and medulla, MR contrast medium is a safe agent compared to CT contrast medium in cases of renal insufficiency.

Purpose: To assess sonographic findings after use of a percutaneous arterial closure device (Angio-Seal) and evaluate its hemodynamic effect on arterial flow with Doppler spectrum analysis. Materials & Methods: The study population consisted of 24 patients (M:F=19:5, mean 64 years) in whom Angio-Seals were deployed after transarterial chemoembolization via a transfemoral approach for treatment of known hepatocellular carcinoma. We performed sonographic examination two days (group 1), one month (group 2) and two months (group 3) after the procedures, focusing on morphologic and size changes of the Angio-Seal anchor. We also performed Doppler spectral analysis proximal to, at, and distal to the anchor. Results: The initial volume of the anchor was 33.9 mm3. In group 1, peak systolic velocity increased at the anchor site in all patients, and the peak systolic velocity ratio was 1.22±0.18. The volume of the anchor was variable (49.7±25.6mm3) and the shape was globular in group 2. The peak systolic velocity ratio was 1.30±0.39. In group 3, all anchors could not be found. Conclusion: The anchor of the Angio-Seal may increase peak systolic velocity locally at the anchor site but it is not hemodynamically significant. In addition, the anchor is completely resolved between one and two months.

P119

P121

A new contrast injector for computed tomography and angiography U. Erikson, A. Wahlberg, A. Wessman, A. Nygren, A. Magnusson, H. Ulfendahl; Radiology, Akademiska Sjukhuset, Uppsala, Sweden.

In vivo study of the effects of intravenous contrast media on red

Purpose: A new contrast injector has been constructed, with a new disposable pump attached to the motor unit (patented). Materials & Methods: It consists of a powerful motor part which is guided by a programme unit. Results: The programme includes safety rules which enable high-safety level to prevent undesired injections. The contrast flows through an airbubble detector of very high accuracy (patented), and the injection stops immediately when the bubbles appear. Conclusion: The injector can be used in any medical area and a battery driven type has also been developed.

blood cell filterability D. Siablis1, G. Athanasiou2, T. Kordonouri1, P. Angele1, P. Zabakis1, E. Tzorakolefterakis3; 1 Department of Radiology, University of Patras, Patras, Greece, 2 Polytechnical School, University of Patras, Patras, Greece, 3Department of Surgery, University of Patras, Patras, Greece. Purpose: Erythrocyte filterability during interventional procedures to preserve the microvascular bed is an important factor. The aim of this work was the in-vivo study of the deformability of the erythrocyte membrane in patients who received contrast media and compare their outcome. Materials & Methods: Twenty-two patients were enrolled. Blood samples were collected before and 5, 30, and 60 minutes after the injection of contrast medium. Iodixanol, Iopentol and Iopromid were compared. Red blood cell deformability was studied by the initial filtration flow rate method. Filtration measurements were performed using the Hemorheometer and, subsequently, the Index of Rigidity (IR) was estimated. Results: IR had increased immediately after the injection of contrast media and was found to be close to the normal range one hour later. Iopentol had influenced the IR more than other contrast media. Iodixanol, instead, had no significant effects on the filterability of red blood cells. Conclusion: The present study showed that radiographic contrast media exhert a significant influence on the deformability of the erythrocytes via rheological effects on its membrane.

P120

P122

Treatment of femoral pseudoaneurysms: a prospective randomised study comparing thrombin injection versus ultrasound-guided compression L. Lonn1, A. Olmarker1, K. Geterud1, B. Risberg2; 1Sahlgrenska University Hospital, Interventional Radiology, Goteborg, Sweden, 2Sahlgrenska University Hospital, Dep of Surgery, Goteborg, Sweden.

Solitary pulmonary nodule less than 2 cm in diameter: histologic identification with percutaneous CT-guided biopsy L. Thanos, A. Nikita, G. Papaioannou, M. Papathanassiou, D. Schizas, D. A. Kelekis; 2nd Department of Radiology, University of Athens, Athens, Greece.

Purpose: To perform a randomized prospective study aimed at comparing success rate, hospitalization, and complications of compression and thrombin treatments. Materials & Methods: Thirty consecutive patients with iatrogenic femoral pseudoaneurysms detected at our department were randomized for treatment with either ultrasound-guided compression or ultrasoundguided thrombin injection. In case of failure of the compression treatment, each patient was offered a thrombin injection the second day. Results: Of the 15 patients treated with thrombin, 13 could be mobilized after four hours and only one required a new thrombin injection. In the compression group, only two patients were successful at 24-hour controls, while in 13 cases extra treatments with compression or thrombin were needed, requiring at least two additional days of immobilization. Conclusion: Ultrasound-guided thrombin injection provides an effective and safe embolization of pseudoaneurysms. The technique is superior to the compression treatment, with a shorter treatment duration and a faster mobilization of the patients.

Purpose: To evaluate the effectiveness and safety of percutaneous CTguided biopsy in histologic identification of solitary pulmonary nodules sized less than 2 cm. Materials & Methods: We studied 61 patients with solitary pulmonary nodule ess than 2 cm in size who underwent percutaneous CT-guided biopsy. Initial CT scan was performed to stage a known primary malignancy or to investigate thoracic clinical signs. Biopsy guns of 18G with 2 cm cutting edge were used in all patients. One or two tissue specimens were extracted in every case. Post-procedural scan was performed to rule out complications. Results: The extracted material was sufficient to establish diagnosis in 54 cases (88.3%), of which 46 were positive for malignancy (primary or secondary). Verification of diagnoses included surgical resection in all cases with positive result, thoracoscopy in negative cases with known primary malignancy or predisposing factors, and follow-up with repetition of CT scan in negative cases with no predisposing factors (patients younger than 35 years, non-smokers, no primary malignancy). Pulmonary contusion was observed in 17 patients and limited pneumothorax in six. Complications resolved uneventfully. Conclusion: Percutaneous CT-guided biopsy of solitary pulmonary nodules appears to be of great assistance in initial histologic characterization and patient management.

P123

P125

Percutaneous transluminal angioplasty of infrapopliteal arteries in patients with intermittent claudication. Acute and one-year results H. Krankenberg, T. Tübler; Angiology, Center for Cardiology and Vascular Intervention, Hamburg, Germany.

Endoluminal treatment of traumatic arteriovenous fistulae with Hemobahn covered stents S. Gil, F. de España, J. Irurzun, P. de la Iglesia, J. Verdu; Vascular and Interventional Radiology, Hospital General Universitario de Alicante, Alicante, Spain.

Purpose: PTA is a well-established technique to treat infrapopliteal arterial occlusive disease with critical ischemia. However, PTA is currently not the therapy of choice in intermittent claudication (IC). In this prospective study technical aspects and long-term results are presented. Materials & Methods: In 81 patients (61 men, 75%, age 71±11 years) with IC (Rutherford 2/3), 104 interventions (23 patients with bilateral PTA) were performed. At baseline, the relative/absolute claudication distance (RCD/ACD) was 49±34/102±57m, the ankle-brachial index (ABI) was 0.61±0.2 before and 0.49±0.2 after exercise. In 19 interventions/18.3%, the laser technique was used, in 26 cases/25% a total of 39 stents were implanted and in 59 legs PTA was done. Results: Primary angiographic success rate was 89.4%. RCD/ACD improved to 107±67 m and 167±74 m (p<0.05 vs. baseline each), respectively, and the ABI at rest/after exercise increased to 0.88±0.13 and 0.72±0.19 (p<0.05 vs. baseline each). One groin hematoma requiring surgical repair occurred. After 12 months, primary patency rate was 66.3%, cumulative primary assisted patency rate was 81.9% and secondary patency rate was 91.5%. Conclusion: Revascularization of infrapopliteal arteries in patients with IC is feasible and associated with good clinical results and an encouraging long-term patency rate. The complication rate is low.

Purpose: To present our experience with the endoluminal treatment of traumatic arteriovenous fistulae (AVF) with Hemobahn, an e-PTFE nitinol covered stent. Materials & Methods: In a three-year period, eight patients (six men and two women), aged between 48 and 77 years (mean 53, years) have been treated with this stent. AVFs were: in the common femoral artery (3), in the superficial femoral artery (4), and in the deep femoral artery (1). Color-Doppler follow-ups were performed in six patients. In seven cases AFV was iatrogenic after a femoral puncture, in the remaining case, it was due to a stab wound. Results: In all cases, the sealing of the arteriovenous communication was achieved. In one case, a second stent was needed, while in another case percutaneous drainage and fibrinolysis with Urokinase of a large hematoma was performed. In the six color-Doppler studies performed at follow-ups between three and 25 months (mean, 12 months), AFVs were closed, stents were patent, and artery and vein permeability was observed. Conclusion: The endoluminal treatment of traumatic AVFs with covered stents is a valid, safe, and effective option.

P124

P126

Limb salvage in patients with type 2 diabetes: preliminary results with a combined balloon angioplasty and Rotablator technique M. Manzi, A. Marangon, G. Mansi-Montenegro, G. Torsello; Department of diagnostic imaging, Institute of Radiology, S. Bortolo Hospital, Vicenza, Italy.

Application of peripheral cutting balloons to conventional balloonresistant lesions. Preliminary results N. Hirai, R. Tanaka, M. Higashi, K. Kimura, H. Naito; Radiology, National Cardiovascular Center, Osaka, Japan.

Purpose: To reduce the amputation rate in type 2 diabetes patients. Materials & Methods: From October 2001 to December 2002, 39 limb salvage procedures were performed in 21 type 2 diabetic patients. All the patients were scheduled for major amputation, presented severe cutaneous ulcerating lesions or infected gangrene, and had cutaneous O2 tension measurements at the foot of <20 mm Hg. All patients had Rotablator debulking procedures in at least one artery below the knee, eight via a popliteal antegrade approach and 31 with CFA approaches. All the patients were treated with PTA using 1.5-3 mm balloons after the Rotablator debulking procedure. All the vascular accesses were closed with Angio-Seal devices. Results: One major amputation was due to an infected periprocedural poplital hematoma. There were no other amputations and no distal embolizatons. The technical success rate was 100 % , and 100 % of patients had O2 tension of >35 mmHg on measurement at two weeks, one month, six months, and one year; cutaneous ulcerations and infections healed 89 % , whereas 9 % showed no significant clinical improvement. Conclusion: A combined Rotablator/ballon PTA technique seems to be safe and effective for limb salvage in type 2 diabetes.

Purpose: To evaluate the efficacy of peripheral cutting balloons versus conventional balloon-resistant lesions. Materials & Methods: In the period between July 2002 and February 2003, percutaneous transluminal angioplasty (PTA) using peripheral cutting balloons was performed in eight patients with eight lesions. All the lesions were resistant to conventional angioplasty balloons. Definitions of initial success were the followings: a) a residual stenosis rate less than 25%, b) a residual pressure gradient less than 10 mm Hg, c) an effective luminal gain in complex lesions as compared with conventional balloon angioplasty. Results: In all eight patients, an initial success was obtained without any complication. Three of the eight lesions were balloon-resistant elastic lesions, two were anastomotic stenoses of by-pass grafts, two were stenoses near the joints, and one was an in-stent lesion. Mean followup period was 3.5±2.0 months. In one of the two anastomotic stenoses of a Y-shaped graft, a thrombotic occlusion occurred after one month. In the other seven patients, no clinical restenosis occurred after the procedure during follow-up. Conclusion: PTA using peripheral cutting balloons is initially effective as compared with conventional balloon-resistant lesions.

P127

P129

Endovascular treatment of vascular complications developing after liver transplantation A. Memis1, S. Demir1, B. Güney1, M. Parildar1, I. Oran1, M. Kılıç2, Y. Tokat2; 1 Dept. of Interventional Radiology, Ege University Medical School, Izmir, Turkey, 2Dept. of Surgery, Ege University Medical School, Izmir, Turkey.

Angiography and distal leg reconstruction by reversed saphenous flap: clinical analysis F. Perona1, A. Zerbi1, F. W. Baruffaldi Preis2, M. Cavallini2, M. Nava2, M. Cugnasca3; 1 Diagnostic & Interventional Radiology, Galeazzi Institute, Milan, Italy, 2 Plastic Surgery, Galeazzi Institute, Milan, Italy, 3Vascular Surgery, Galeazzi Institute, Milan, Italy.

Purpose: To evaluate the results of endovascular procedures performed for the treatment of vascular complications in liver transplant recipients. Materials & Methods: Vascular complications developed in 19 of 130 liver transplant recipients between 1994 and 2002 (hepatic artery thrombosis=6, hepatic artery stenosis=2, portal vein stenosis=2, portal vein thrombosis=3, inferior vena cava stenosis=2 and hepatic vein stenosis=4). Color Doppler ultrasonography was very useful in the diagnosis. Endovascular intervention was performed in eight patients. Results: Due to restenosis after repeated multisession balloon dilatation, two vena cava anastomotic stenoses were dilated with metallic stents. In two of four hepatic vein stenoses, balloon dilatation was successful alone; in the other two, metallic stents were implanted due to restenosis after PTA during follow-up. In the hepatic artery stenoses, balloon dilatation was performed in one patient. Endovascular intervention was unsuccessful in the other high-grade filiform stenosis. Surgical thrombectomy was performed in four acute thromboses and no therapy was considered in another two with clinically silent thrombosis in a chronic phase. Conclusion: Endovascular intervention has a major role in the treatment of vascular complications, considering the difficulty of surgical treatment in liver transplant recipients. Percutaneous treatment of vascular stenoses is an efficient primary treatment method.

Purpose: Reconstruction of skin degloving lesions remains a significant problem for plastic surgeons. A reliable and vital saphenous reversed flap represents an excellent therapeutic option. The aim of the study is to evaluate the microvascular pattern improvement and flap survival. Material & Methods: During December 2001 - December 2002 twentyeight patients (17 male - 11 female) were treated by a sural or saphenous reversed flap. Surgical indications for lower leg, heel and proximal foot were chronic ulcer (10), open fractures (8), neoplastic ulcer ( 2 ), diabetic ulcer (4 ), and ischemic ulcer ( 4 ).One- or two-step techniques of flap preparation were considered according to patient conditions.Six patients underwent peripheral PTA to improve distal circulation. Results: All patients were evaluated by DSA preoperatively and three months after surgery. The flap survived completely in 20/28 patients (71.4%), and in all cases microcirculation improvement was achieved. Eight patients with severe systemic diseases (diabetes 4, peripheral vasculopathy 4 ) suffered from delayed healing due to venous congestion, angiographically confirmed. PTA of tibio-peroneal arteries was successful in all cases. Conclusion: The evaluation of these patients confirmed the role of microvascular pattern improvement after plastic surgery, enhanced in selected cases by peripheral endovascular treatment

P128

P130

Post-catheterization femoral arteriovenous fistulas: endovascular treatment with stent-grafts B. Önal, S. Kosar, T. Gumus, E. T. Ilgit, S. Akpek; Department of Radiology, Gazi University, School of Medicine, Ankara, Turkey.

FIRE infrapopliteal PTA on-line registry. Results 14 months after launching M. Sapoval1, T. Nicholson2, C. Zollikofer3, A. Adam4, J. Struyven5, D. Allison6, D. Vorwerk7; 1HEGP, Paris, France, 2Hull Royal Infirmary, Hull, United Kingdom, 3Kantonsspital, Winterthur, Switzerland, 4St Mary, London, United Kingdom, 5Hôpital Erasme, Bruxelles, Belgium, 6St Georges Hospital, London, United Kingdom, 7Hospital, Ingolstadt, Germany.

Purpose: To evaluate efficacy and safety of stent-grafts for the endovascular treatment of iatrogenic femoral arteriovenous fistulas. Materials & Methods: Endovascular treatment of iatrogenic femoral arteriovenous fistulas as a result of arterial puncture for coronary angiography and/or angioplasty was attemped in ten cases. Balloonexpandable stent-grafts (Jomed), one for each lesion, were used to repair the fistulas which were in the deep femoral artery to the femoral vein in all cases. Stent-graft implantation into the deep femoral artery was performed via a contralateral retrograde approach. Results: All stent-grafts were deployed successfully. Closure of the fistulas occurred immediately in nine of the ten cases. In one case, a complete closure could not be obtained. No complication was observed after implantation. Follow-up for 3-26 months (mean, 16 months) with color-Doppler-ultrasonography and intravenous DSA revealed patency of the stented arterial segments and no arteriovenous shunting. Conclusion: Stent-graft implantation is an alternative and reliable method for the closure of post-catheterization of femoral arteriovenous fistulas.

Purpose: To report the results of infra-popliteal PTA performed for critical limb ischemia using the FIRE e-registry one year after the first patient was entered in the registry. Materials & Methods: In each center, consecutive patients undergoing infra-popliteal PTA for critical limb ischemia were entered in the registry. Data retrieval was performed using the built-in statistical package, as defined by the FIRE working group. Results: As of February 2003, 235 patients from 21 centers in 11 countries were recorded. Risk factors were: diabetes in 74%, tobacco in 33%, and creatinine clearance <50 ml/min in 29%. Clinical stages were: Rutherford 4 (26.5%), 5 (48.5%), and 6 (25%). Interventions concerned 1.6 sites/patients and total occlusion was present in 38%. Lesion length was >10 cm in 17%. In-flow stenosis was treated in 63% of the patients in the same session. Immediate complications were 15% and in-hospital mortality was 4.4%. Mean ABI was 0.5 before PTA and 0.8 at three months. Mean follow-up is now four months and limb salvage rate at three months is 88%. Conclusion: Preliminary results of infra-popliteal PTA for critical limb ischemia as retrieved from the e-registry are within previously reported figures.

P131

P133

PTA of infrapopliteal arteries: early, mid-term, and long-term lower limb salvage B. Koznar, J. Novotný, D. Krajíckova, M. Rocek, J. Kovác, J. Lastovicková; Diagnostic and Interventional Radiology, IKEM, Prague 4, Czech Republic.

Percutaneous balloon angioplasty for treating stenosis of the aorta caused by nonspecific aortitis S. Sharma1, G. Gulati1, S. S. Kothari2, R. Juneja2, A. Saxena2, K. K. Talwar2; 1 Cardiovascular Radiology, All India Institute of Medical Sciences, New Delhi, India, 2Cardiology, All India Institute of Medical Sciences, New Delhi, India.

Purpose: To evaluate the efficacy of infrapopliteal PTA after one, three and five years. Materials & Methods: From 1985 to the end of 2002, infrapopliteal PTA was performed in 638 patients (805 left limbs, 1614 arteries, including 96 re-PTAs and 72 PTAs of both left limbs) in our department. Patients were men in 67.5% of cases, 82% of them had DM and their average age was 66 years. Indications were: disabling claudication in 7%, phlegmone in 3%, non-healing amputation wound in 7%, rest pain in 12%, non-healing ulcer in 11%, gangrene in 60%. Results: A technical success of PTA was achieved in 84% of the arteries intended to treat. Clinical success was defined in presence of symptoms-free patients or toe or up to TMT amputations. One-year followup was available for 315 patients, with 227 left limb salvages (72%); three-year follow-up was available for 136 patients, with 91 left limb salvages (67%); and five-year follow-up was available for 42 patients, with 27 left limb salvages (65%). Conclusion: PTA of infrapopliteal arteries is a method with a high technical efficacy and a satisfactory long-term left limb salvage rate.

Purpose: To evaluate results of percutaneous balloon angioplasty (PTA) in treating aortic stenosis caused by nonspecific aortitis, and the influence of angiographic features on outcome Materials & Methods: 78 patients with hypertension (75) or claudication (three) and inactive disease were treated. Bail-out stents were placed in patients with obstructive dissection. Influence on outcome of features including stenosis location, centricity, length (>4 cm), presence of adjacent disease, calcification, residual stenosis and gradient, was assessed. Multivariate and life-table analysis were performed. Results: Technical success was found in 74 and clinical success in 76 patients. Stenosis decreased from 80+9% to 18+17 %, gradient fell from 78+21 to 20+17 mm, blood pressure improved from 184+21/ 114+13 to 130+10/84+15 mm and drug requirement decreased from 4.3+0.5 to 1.5+0.8 (p<0.001). Complications included obstructive dissection in 19 and local hematoma in two patients. At 19+11 months' follow-up, restenosis was seen in eight and PTA siteaneurysm in three patients. Cumulative five-year patency rate was 83%. Juxta-diaphragmatic location, greater length, and adjacent disease correlated with obstructive dissection. Conclusion: PTA is successful in treating these patients. Specific angiographic features predispose to an unfavorable outcome. Stents are useful in treating such patients, in addition to those with sub-optimal PTA

P132

P134

Efficacy of the percutaneous treatment of chronic aorto-iliac occlusions M. Tozzi1, D. Laganà2, R. Caronno1, G. Carrafiello2, M. Carnini1, G. Piffaretti1, M. Canziani1, C. Fugazzola2, P. Castelli1; 1Vascular Surgery, University of Insubria, Varese, Italy, 2Institute of Radiology, University of Insubria, Varese, Italy.

Reconstruction of the aortic bifurcation with percutaneous stent placement: long-term results G. Papaioannou, E. N. Brountzos, E. Alexopoulou, K. Malagari, A. Sissopoulos, D. A. Kelekis; 2nd Department of Radiology, University of Athens, Athens, Greece.

Purpose: To update mid-term results of a retrospective analysis on aortoiliac occlusions. Materials & Methods: Through the last three years, 53 patients (40 men and 13 women, mean age 52 years, range: 48-75) with intermittent claudication due to chronic aorto-iliac occlusion were treated. Patients underwent preoperative Duplex scan and/or helical-CT. Ten (19%) PTA-stentings of the infrarenal aorta, 14 (26%) of the common iliac artery bilaterally, and 29 (56%) of one iliac artery were carried out. Follow-up Duplex scans were performed at one, six, and 12 months after the procedure; an angiographic control in case of reocclusion was always done. Results: Mortality and morbidity rates were 0; primary technical success rate was 100%. Mean follow-up was 18 months (range, 3-36). Every patient improved clinically (from class III to class II of LericheFontaine classification). We documented two (4%) stent thromboses eight months after the procedure, which were successfully treated by loco-regional thrombolysis and PTA of the stent distal portion. Conclusion: The endovascular treatment should be considered the first treatment choice of aorto-iliac occlusive diseases.

Purpose: To present long-term results of percutaneous reconstruction of the aortic bifurcation with stent placement. Materials & Methods: During a three-year period, 36 stents were implanted, using the kissing-stent technique, at the aortic bifurcation of 36 limbs. Rest pain or tissue loss were present in 11.1%. KaplanMeier patency distribution curves were plotted, and Log rank test and Cox's Regression model were employed to assess the impact of each parameter on long-term patency. The power of the study was >80%. Results: Angiographically successful stent deployment was achieved in 91.7%. Complications (8.3%) included stent misplacement (1/3) and groin hematoma at the puncture site (2/3). 65.6% of limbs had initial clinical and imaging improvement. Nevertheless, during long-term follow-up (mean 37.8 months), 38.9% of limbs deteriorated clinically, and 27.8% showed imaging (duplex ultrasonographic) signs of restenosis. Secondary intervention was performed in four cases. Primary one-year, two-year and three-year patency was 80.38%, 62.17% and 55.26% respectively. Of the parameters studied, exercise and hypertension reduction positively influenced long-term patency, while age was of borderline significance. Conclusion: Follow-up of percutaneously restored aortic bifurcations with stent placement reveals satisfying rates of long-term patency. Clinical deterioration seems to be associated with lack of exercise and the presence of hypertension.

P135

P137

Retrospective comparison of self-expanding stents for superficial femoral artery stenoses and occlusions D. Laganà1, R. Caronno2, G. Carrafiello1, G. Piffaretti2, M. Carnini2, M. Tozzi2, M. Canziani2, F. Fontana1, P. Castelli2; 1 Institute of Radiology, University of Insubria, Varese, Italy, 2Vascular Surgery, University of Insubria, Varese, Italy.

Endovascular treatment of renal artery stenosis resulting from aortic dissection M. Januszewicz1, O. Rowinski1, K. Milczarek1, Z. Galazka2, J. Szmidt3; 1II Department of Clinical Radiology, The Medical University of Warsaw, Warsaw, Poland, 2Department of General ,Vascular and Transplant Surgery, The Medical University of Warsaw, Warsaw, Poland, 3Department Of General ,Vascular and Transplat Surgery, The Medical University of Warsaw, Warsaw, Poland.

Purpose: Updating the effectiveness of PTA and stenting in chronic occlusions of the superficial femoral artery (SFA). Materials & Methods: Through the last three years, self-expanding stents (Memotherm, Smart, Wallstent) were implanted in 77 patients (43 men). Twenty-three of them (30%) had critical limb ischemia; preoperative angiography was carried out in all the patients. Patency rates were compared between: stenoses and occlusions, good and poor runoffs, diabetic and nondiabetic patients. Duplex follow-up was performed at one, six, and 12 months post-procedure. Results: Primary technical success rate was 100%. Mortality and morbidity rates were 0. Mean lesion length was 6 cm (2-20), 23 (30%) were 5-10 cm in length, 44 of them (53%) were occlusions. Forty-three lesions (56%) involved the Hunterian channel. The mean length of hospitalization was two days; the median, one (1-10). Loco-regional thrombolysis was used before stenting in ten (13%) patients, while adjunctive PTA of the iliac arteries was done in 12 (16%). One-year primary patency rate was 71% and mean follow-up was 18 months (3-36). Wallstents had better results than other stents. Conclusion: Stenting of the SFA has mid-term results comparable with traditional surgery; diabetis, occlusion >10 cm, and poor run-off seem to have poorest results.

Purpose: We report the endovascular treatment of three patients, one with type-A and two with type-B chronic aortic dissection, who presented with a poor management of hypertension and a rise of their serum creatinine levels. Materials & Methods: The patient with a type-A dissection (a 63-yearold man) had undergone surgery of the ascending aorta three years earlier; the two patients with a type-B dissection were two women aged 67 and 77. A Duplex-Doppler sonography and a spiral-CT were carried out to confirm, in all cases, a left renal artery stenosis due to an aortic dissection. A puncture of the true lumen at the level of the femoral artery was performed in all cases. After a septal fenestration, angioplasty and stent placement were carried out to treat those complications consequence of the aortic dissection. Results: After the treatment, hypertension resolved and serum creatinine levels diminished. The patient with a type-A dissection was discharged after prescribing a baseline drug regimen. The two patients with a type-B dissection underwent a successful endovascular aortic stent graft placement. Conclusion: The employment of endovascular techniques in the treatment of renal ischemia improves the clinical outcome in high-risk patients with aortic dissection.

P136

P138

Renal artery embolisation. A first-line treatment option for endstage hydronephrosis K. Mitra1, V. Prabhudesai1, R. L. James2, M. Cowling1, M. E. French3, D. J. West1; 1Radiology Department, North Staffordshire Hospital NHS Trust, Newcastle-u-Lyme, United Kingdom, 2Urology Department, Stafford General Hospital, Stafford, United Kingdom, 3Urology Department, North Staffordshire Hospital NHS Trust, Newcastle-u-Lyme, United Kingdom.

Renal stenting: effect on the renal function in chronic renal failure R. Rivolta1, F. Nizzola1, L. Rendina1, L. Dellafiore1, C. Pinerolo de Septis2; 1Radiology, Ospedale San Carlo-Milano, Milano, Italy, 2Nephrology, Ospedale San Carlo-Milano, Milano, Italy.

Purpose: Patients with poorly functioning hydronephrotic kidneys traditionally undergo nephrectomy if they are symptomatic. In our unit patients are offered renal artery embolisation as an alternative. Materials & Methods: Fifteen patients with a mean age of 32.9 years have undergone renal artery embolisation for symptomatic hydronephrosis with poor function. Mean follow-up was 38.9 weeks (range 8104). All patients had loin pain and hydronephrosis. Twelve patients had primary pelvi-ureteric junction obstruction. Two patients had poorly functioning hydronephrotic kidneys secondary to chronic calculous obstruction. One patient had chronic pain in an obstructed but reasonably functioning kidney following a previous pyeloplasty for PUJO. Selective renal artery embolisation was carried out under antibiotic cover using a 7-F balloon occlusion catheter and a combination of steel coils, absolute alcohol and poly vinyl alcohol. Results: Nine patients developed post-embolisation syndrome with no evidence of sepsis. One patient required readmission with this condition. One patient developed a haematoma at the puncture site. Mean hospital stay was 2.3 days. One patient had minor discomfort. Nine patients have had follow-up ultrasound confirming involution of the hydronephrotic renal unit. Conclusion: Renal artery embolisation is an effective, safe, well tolerated, minimally invasive treatment option in end-stage hydronephrosis.

Purpose: To evaluate the renal function after renal stenting in those patients with proximal renal artery stenosis (RAS) and chronic renal failure. Materials & Methods: Forty patients (mean age: 67±13), with a serum creatinine >1.8 mg/dL (range, 1.8-8.1), underwent renal stenting because of atherosclerotic RAS. A successful procedure was estabilished when the residual stenosis was less than 30%; colorDoppler follow-ups at scheduled times were performed. Patients had a post-procedure median follow-up of 15 months. Creatinine levels before the procedure were collected for a median time of three months. Regression analysis was used to establish evidence of significant changes in the renal function. Results: A significant reduction of serum creatinine levels was recorded in 32.5% of patients (95% confidence interval (CI): 18.0-47.0); 47.5% of patients (CI: 32.5-62.5) did not show any significant change, while 20.% of them (CI: 7.6-32.4) presented a significant increase. An indipendent factor associated with the renal function decrease was an initial elevated serum creatinine level (p<0.05). Conclusion: Renal stenting of RAS can ameliorate or stabilize the renal function in the majority of patients with chronic renal failure. Patients with initial elevated creatinine levels have a bad prognosis.

P139

P141

Kinks of the transplant renal artery with no intra-arterial pressure gradient do not require endovascular correction. Five-year outcome study G. C. Chua1, U. Patel1, S. Snowden2; 1Department of Radiology, St George's Hospital, London, United Kingdom, 2Department of Urology, St George's Hospital, London, United Kingdom.

Renal artery stenting for the treatment of atherosclerotic renal artery stenosis: experience in one center M. Gimeno, J. Medrano, R. Alfonso, R. Tobio, M. Herrera, M. De Gregorio; Interventional radiology, Clinico Universitario Lozano Blesa, Zaragoza, Spain.

Purpose: Significant transplant renal artery narrowing results in an intra-arterial pressure gradient and graft dysfunction correctable by endovascular therapy. Kinks of the transplant artery cause velocity gradients on Doppler-ultrasound, but some will have no gradient across the kink on intra-arterial measurement. It is unknown whether these should undergo endovascular therapy. This is a study of non-flow limiting kinks managed conservatively. Materials & Methods: Fourteen cases with kinked arteries (on DopplerUS and angiography) were studied over five years. The intra-arterial pressures were measured in all cases. Those with a significant gradient (>10% pressure change across the kink) underwent stenting. The rest were managed conservatively. Outcome was determined by creatinine levels, blood pressure (BP) control and histology over a 3-5 year period (median, 4yrs). Results: Eleven of the 14 cases had no pressure gradient. At the median follow-up of four years, creatinine and BP recordings were steady on conservative treatment [median values 124 µmol/l and 126/88 mmHg, respectively]; 2/11 grafts were lost, both had chronic rejection on histology. The three stented patients also had a good outcome at four years. Conclusion: Kinks of the transplant renal artery with normal intra-arterials do not require endovascular correction, with satisfactory graft outcome on conservative therapy.

Purpose: To evaluate the efficacy and safety of renal artery stents (Herculink, Guidant) in renovascular disease Materials & Methods: From October 2000 to January 2003 we implanted 37 renal Herculink stents in 34 patients, 27 men and 7 women. All patients were diagnosed as having renovascular hypertension, with mean systolic pressures of 188 mm Hg and mean diastolic arterial pressures of 95 mm Hg. Renal failure was present in 29 patients. 30 lesions were located in the ostium, with a mean degree of stenosis of 80.3%. Results: Stent placement was successful in all arteries. There were complications in six procedures: four post angioplasty artery dissections and two puncture site hematomas. Mean follow-up was 7.3 months. Two stents presented thrombosis six months after implantation (resolved by angioplasty). Hypertension was cured in 27 patients (taking drugs) and improved in five; two had no change. Mean post-treatment systolic pressure was 138.6mmHg and mean diastolic arterial pressure 77.2 mm Hg. Renal failure was absent in 11 patients, improved in 11, and worsened in seven. Conclusion: Renal artery Herculink stenting has a high technical success rate, a low complication rate, and a low rate of restenosis, and is a effective method for treating renovascular hypertension.

P140

P142

Stent placement in ostial and nonostial atherosclerotic renal arterial lesions: a prospective study E. N. Brountzos, P. Argyriou, K. Malagari, E. Alexopoulou, V. Nikolaou, L. Thanos, D. A. Kelekis; 2nd Radiology, Athens University, Athens, Greece.

Ultrasound ablation in peripheral arterial occlusion in diabetes mellitus A. Wang, D. Jing, J. Du; Vascular Surgery Department, Hei Long Jiang Provincal Hospital, Harbin 150036, China.

Purpose: Outcome evaluation of stent treatment in renal artery lesions. Materials & Methods: Prospective evaluation of 41 symptomatic patients (mean age 63.8±8.4) with 54 renal artery atheromatous lesions treated with stenting. All patients were hypertensive, and 16 (39%) also had renal insufficiency. Follow up included Duplex sonography, CT angiography, and DSA. Patients' serum creatinine levels and blood pressure were regularly recorded. Mean follow up time was 20.56±15.4 months. Results: Technical success was achieved in 54 of 54 lesions (100%). Complications were encountered in four patients. Ten patients were cured of hypertension (24.4%), 24 patients improved (58.5%), five patients remained stable (12.1%) and two were treatment failures (4.8%). Renal insufficiency improved in 14 patients (87.5%); two patients deteriorated, one requiring hemodialysis. Conclusion: In our experience renal artery stenting is a safe procedure with a high clinical success rate.

Purpose: To evalute the efficiency of ultrasound ablation for the treatment of peripheral arterial occlusion in diabetes mellitus. Materials & Methods: Twenty-six patients received ultrasound ablation treatment. Through a perculaneous or surgical approach the ultrasound ablation catheter was inserted into occluded arterial segments. Ultrasound frequency was 42KHz, and energy was 45 watts. Results: The procedure was successful in 22 of 26 patients, for a success rate of 86.4%. Pre-existing gangrene was mitigated in six of nine patients after the treatment. Rest pain was mitigated in ten of 15 patients. Long-term effiency was 64%. Conclusion: Ultrasound ablation constitutes a new approach for peripheral arterial occlusion in diabetes mellitus.

P143

P145

Interventional treatment of difficult-to-treat varices: development of a modified percutaneous transhepatic obliteration T. Kaminou1, M. Hashimoto1, Y. Ohuchi1, T. Ninoi2, M. Hamuro2, N. Nishida2, Y. Sakai2, K. Nakamura2, T. Ogawa1, R. Yamada2; 1 Radiology, Tottori University, Yonago, Japan, 2Radiology, Osaka City University, Osaka, Japan.

Successful treatment of acute and subacute Budd-Chiari syndrome with transjugular intrahepatic portosystemic shunt A. Rampoldi, M. Intotero, M. Solcia, G. Carrafiello, M. C. Castoldi, R. Corso; Vascular and Interventional Radiology, Niguarda Hospital, Milan, Italy.

Purpose: To develop an m-PTO (modified percutaneous transhepatic obliteration of varices) and evaluate its efficacy for difficult-to-treat varices. Materials & Methods: Thirty-nine patients with gastric and duodenal varices, difficult to treat by endoscopic management, underwent mPTO. Antegrade injection of EOI with metallic coils was performed through a microcatheter which was transhepatically advanced into the varices. During injection, the outflow tract was occluded by a ballooncatheter or an S-B tube. In one case with multiple out-flow tracts, metallic coils were placed into the varices to reduce blood flow. The effect on varices was evaluated by contrast-CT or Doppler-US one week after the procedure. Results: The procedure was successful in 92% of patients. No technique-related complications were noted. Complete or near-complete thrombosis was obtained in 86% of cases. In four patients, a recurrence of varices occurred. No rupture of the treated varices was noted. In about half of the cases esophageal varices worsened after the procedure, but most of them were treatable by endoscopic management. Conclusion: Modified PTO was considered to be very useful for difficult-to-treat variceal varices using only balloon-occluded retrograde trasvenous obliteration.

Purpose: To evaluate the safety and efficacy of transjugular intrahepatic portosystemic shunting (TIPS) in treatment of acute and subacute Budd-Chiari syndrome (BCS). Materials & Methods: From January 1999 to December 2002, nine patients (four males, aged 18-49 years) with acute or subacute BCS, uncontrolled by medical therapy, were referred for TIPS. In two cases BCS was caused by myeloproliferative disorders; the other cases had an unknown etiology. In one patient portal vein thrombosis was also present. Results: Technical success was achieved in all patients without major complications; marked relief of symptoms and clinical parameters was always observed . Follow-up lasted 4-35 months (mean 22 months). All patients but one are alive at present, and none has requested or been scheduled for liver transplantation. Acute leukemia was the cause of that single death. The patient with associated portal vein thrombosis underwent ultrasound-thrombolysis to regain complete patency, but the TIPS occluded after four months. All other patients maintain full patency of the TIPS, with significant improvement of liver function. Conclusion: TIPS is an effective and safe treatment in acute and subacute BCS and it could be considered as an alternative to traditional surgical portosystemic shunting for liver transplantation.

P144

P146

Variation of gastrorenal shunts and collateral veins from gastric varices; radiographical investigation H. Maeda, S. Hirota, K. Kobayashi, Y. Kako, S. Yamamoto, K. Arai, Y. Miyamoto, K. Miura, N. Nakao; Radiology, Hyogo college of Medicine, Nishinomiya, Japan.

Balloon-occluded retrograde transvenous obliteration of gastric varices with/without hepatic encephalopathy: clinical results of 145 patients S. Hirota1, H. Maeda1, S. Yamamoto1, K. Arai1, Y. Kako1, K. Miura1, N. Nakao1, K. Sugimoto2, T. Fukuda3; 1Radiology, Hyogo College of Medicine, Nishinomiya, Japan, 2Radiology, Kobe Univertsity, Kobe, Japan, 3Radiology, Kobe University, Kobe, Japan.

Purpose: To evaluate variations of gastrorenal shunts (GRS) and collateral veins on balloon-occluded retrograde venography (BRTV). Materials & Methods: Since 1993, BRTV, superior mesenteric and splenic arteriography have been performed before balloon-occluded retrograde transvenous obliterance (BRTO) in 104 patients. Variations of GRS and collateral veins were analyzed precisely. Results: The indication for GRS was gastric varices in 96.2% (100/104). Types of GRS were as follows: only one main GRS, 92% (92/100); two main GRS with ring-like appearance, 3% (3/100); one main GRS with other small draining veins to left renal vein, 5% (5/100). The rates of various collateral veins were as follows: inferior phrenic veins, 76% (76/ 100); pericardiacophrenic veins, 41% (41/100); gonadal veins, 15% (15/ 100); retroperitoneal veins (ascending lumbar veins and hemiazygos veins), 69% (69/100). Conclusion: GRS variations and the rate of various collateral veins from gastric varices were investigated in 104 patients. It is very important to know the precise variation when passing the balloon catheter into gastrorenal shunts and in avoiding unexpected injection of sclerosing agent.

Purpose: To evaluate the clinical results of balloon-occluded retrograde transvenous obliteration (BRTO) for gastric varices and/or hepatic encephalopathy in 145 patients. Materials & Methods: From December 1994, BRTO has been performed in 145 patients with gastric varices (GV) with/without hepatic encephalopathy. Twenty-five of the 145 patients were emergency cases. BRTO was performed by transvenous obliteration of gastric varices, using 5% ethanolamine oleate (sclerosing agent), under balloon occlusion of the gastrorenal shunt. Effects on GV, hepatic encephalopathy, hepatic function, and in emergency cases were evaluated. Results: Gastric varices disappeared or markedly decreased in 107 of 115 patients (93%). Hepatic encephalopathy improved in 13 of 14 patients. The hepatic function reserve improved in 48% of patients, but remained stationary or was exacerbated in those patients with hepatic tumors. Aggravation of esophageal varices was found in 23%. In emergency cases with rupture of GV, the bleeding stopped completely in all the 25 patients. Conclusion: The very high success rate in GV obliteration obtained by BRTO suggests this to be the first choice treatment for GV. The improvement in the hepatic function was probably due to the interruption of portosystemic shunting.

P147

P149

Interventional procedures for portosystemic shunts in cirrhotic patients with chronic recurrent hepatic encephalopathy N. Nishida1, T. Kaminou2, Y. Sakai1, T. Ninoi1, M. Hamuro1, T. Kitayama1, K. Sato1, S. Tanaka1, K. Nakamura1, Y. Inoue1; 1 Radiology, Osaka city university, Osaka, Japan, 2Radiology, Tottori university, Yonago, Japan.

Effectiveness of dexamethasone-releasing nitinol stents on pseudointimal hyperplasia in transjugular intrahepatic portosystemic shunt: an experimental study in swines T. Seo1, J. Oh2, Y. Yoon2, H. Song3; 1Radiology, Gachon Medical School, Gil Medical Center, Incheon, Republic of Korea, 2Radiology, Kyung Hee University Hospital, Seoul, Republic of Korea, 3Radiology, University of Ulsan, College of Medicine, Asan Medical Center, Seoul, Republic of Korea.

Purpose: To evaluate the efficacy of interventional procedures of portosystemic shunts in cirrhotic patients with chronic recurrent hepatic encephalopathy (CRHE). Materials & Methods: Six cirrhotic patients with CRHE refractory to medical treatment were studied. One intrahepatic portocaval, one superior mesenteric-caval, two inferior mesenteric-caval, and two splenorenal shunts were identified. Stentgraft placement into the hepatic vein to occlude the intrahepatic portocaval shunt (1), percutaneous transhepatic obliteration (PTO) of the mesocaval shunt (1), balloon-occluded retrograde transvenous obliteration (B-RTO) of the mesocaval shunt (2), B-RTO of the splenorenal shunt (1), and splenic vein embolization for disconnection between portal and systemic circulation preserving the shunt (1) were performed. Results: All the procedures were successful without complications. Post-treatment, encephalopathy disappeared and serum ammonia levels decreased in all the patients. The two patients in whom PTO and splenic vein embolization were performed showed a recurrent encephalopathy with collateral veins seven and nine months after the procedure; a new PTO was successfully performed. In one of the two patients treated by B-RTO, an esophageal varix developed four months after. Conclusion: Embolization procedures for portosystemic shunts with CRHE had an excellent outcome, but a close clinical follow-up is mandatory to detect recurrent encephalopathy and visceral venous varices.

Purpose: To evaluate the effectiveness of dexamethasone (DM)-releasing nitinol stents in inhibiting pseudointimal hyperplasia after transjugular intrahepatic portosystemic shunt (TIPS). Materials & Methods: Metallic stents used for TIPSs, made of a nitinol wire, measured 50 mm in length and 10 mm in diameter. Stents were classified into three types: types 1 and 2 were coated by 12% and 20% of DM solution, respectively, mixed with polyurethane (PU); type 3 was a bare stent used for the control study. TIPSs were successfully performed in 15 swines with portal hypertension by injecting n-butyl-2cyanoacrylate into the portal vein. After 14 days, follow-up portograms were performed and, then, all the animals were sacrificed. Histologic analyses were carried out and the maximum pseudointimal hyperplasia as a percentage of the stent radius was calculated. Results: On follow-up portograms, one type-1 and two type-2 stents had maintained the patency while all the others had occluded. Histopathologic analyses showed that the mean of maximum pseudointimal hyperplasia, as percentages of the stent radius, was: 51.2, 50, and 76% in type-1, type-2, and type-3 stents, respectively. Conclusion: Although patency did not improve, TIPS with DM-releasing nitinol stents seemed to reduce pseudointimal hyperplasia in swines with portal hypertension.

P148

P150

Transjugular intrahepatic portosystemic shunt for refractory ascites T. Ninoi1, K. Nakamura1, N. Nishida1, Y. Sakai1, T. Kitayama1, M. Hamuro1, M. Nakata1, Y. Inoue1, R. Yamada1, T. Kaminou2; 1Radiology, Osaka City University, Osaka, Japan, 2Radiology, Tottori University, Yonago, Japan.

Balloon-occluded retrograde transvenous obliteration for gastric varices draining into gastro-inferior phrenic shunts T. Kitayama1, N. Nishida1, T. Kaminou2, T. Ninoi1, Y. Sakai1, M. Hamuro1, M. Isota1, K. Nakamura1, Y. Inoue1, R. Yamada1; 1Radiology, Osaka City University, Osaka, Japan, 2Radiology, Tottori University, Yonago, Japan.

Purpose: Transjugular intrahepatic portosystemic shunt (TIPS) is a useful method to improve the quality of life (QOL) in those patients with refractory ascites. The aim of this study was to evaluate efficacy and prognosis of TIPS for refractory ascites. Materials & Methods: TIPS was performed in 38 patients (29 men and nine women; age 55±9 years) with refractory ascites due to portal hypertension. The change of ascites volume, complications and prognosis were evaluated. Results: TIPS procedures were successfully performed in 36 patients (95%) and the portal pressure significantly decreased from 33±8 to 22±7 mm Hg. Ascites disappeared in 39% and decreased in 25%. Hepatic encephalopathy occurred in 36% of cases and shunt obstructions within one month occurred in 8%. Five patients with renal failure died within one month after the procedure. Cumulative survival rate was 73% at six months, 63% at one year and 53% at two years. There was a significant difference in the cumulative survival rate between the effective and the ineffective group (p<0.05). Conclusion: TIPS is a useful method for refractory ascites and contributes to the prognosis.

Purpose: Balloon-occluded retrograde transvenous obliteration (B-RTO) for gastric varices draining into gastro-renal shunts have been described in many reports, but few reports exist about gastro-inferior phrenic (G-IP) shunts. The purpose of this study is to evaluate the efficacy of B-RTO for gastric varices draining into G-IP shunts. Materials & Methods: Ten patients with gastric varices mainly draining into G-IP shunts were treated by B-RTO only or by B-RTO combined with modified percutaneous transhepatic obliteration (m-PTO). All the patients were followed for at least two months (average 20 months). Varices' flows were evaluated by color-Doppler endoscopic ultrasonography and by enhanced-CT after two weeks, three months, six months and every six months thereafter. Results: B-RTO was technically successful in all the patients without major complications. In six patients, an improvement of gastric varices was obtained by B-RTO only, but in the remaining four, m-PTO was required because of too many other collateral draining veins. The gastric varices flow disappeared in all the patients two weeks after the treatment and no recurrences were recorded in the follow-up period. Conclusion: B-RTO is a safe and effective method in treating gastric varices mainly draining into G-IP shunts.

P151

P153

Are transjugular intrahepatic portosystemic shunt stent-grafts a contraindication for future liver transplantation? G. Maleux1, J. Pirenne2, J. Vaninbroukx1, R. Aerts2, F. Nevens3; 1Radiology, University Hospitals, Leuven, Belgium, 2Transplantation surgery, University Hospitals, Leuven, Belgium, 3Hepatology, University Hospitals, Leuven, Belgium.

Transjugular intrahepatic porto-systemic shunts with PTFE-covered stents (VIATORR): comparative study versus uncovered stents (Wallstent) G. Rodriguez Rosales1, M. Echenagusía Boyra1, N. Alava Echevarria1, J. Guzmán de Villoria1, R. Bañares2, G. Simó Muerza1, F. Camuñez Alonso1, A. Echenagusía Belda1; 1 Radiology, Hospital Gregorio Maranon, Madrid, Spain, 2Digestive, Hospital Gregorio Maranon, Madrid, Spain.

Purpose: To evaluate the feasibility and safety of orthotopic liver transplantation (OLT) in patients with prior transjugular intrahepatic portosystemic shunt (TIPS) using a commercially available stent-graft advanced into the inferior vena cava (IVC). Materials & Methods: Over a 26-month period we performed 48 TIPS procedures using a commercially available stent-graft. In the same time period 123 OLTs were performed. The records of eight of these 123 patients who underwent TIPS prior to OLT were reviewed from initial admission, through TIPS, to OLT and current postoperative status. Results: In all eight patients the TIPS stent-graft was accurately placed. OLT was performed four to 279 days (mean interval: 80.8 days) after the TIPS procedure. In all cases it was possible to place a suprahepatic clamp on the suprahepatic vena cava without "entrapping" the stentgraft inside the clamp. No other difficulties during OLT were reported. The postoperative follow-up has been uneventful for all patients except one. Conclusion: Accurate placement of a TIPS stent-graft into the inferior vena cava was feasible in all patients. No surgical difficulties or complications were encountered during later OLT. TIPS stent-grafts should not be considered a contraindication in patients who need TIPS as a bridge to OLT.

Purpose: To compare the outcomes of transjugular intrahepatic portosystemic shunts (TIPS) with covered stents (CS) versus uncovered stents (US). Materials & Methods: From September 1998 to February 2002, 70 consecutive patients were treated with TIPS (50 with US and 20 with CS). In both groups we analyzed evolution of the portocaval gradient (PCG) after TIPS, frequency of TIPS dysfunction (PCG > 12 mm Hg), appearance of encephalopathy, early liver failure, and mortality. Results: The two groups were similar in age, Child-Pugh score, etiology of cirrhosis, portal thrombosis, and previous PCG. PCG was similar in both groups after TIPS and in the first month but was lower in the sixth month in the CS group (7 ± 1 vs 14.2 ± 5.6 mm Hg, p< 0.0001). No patients in the CS group had TIPS dysfunction during follow-up vs 69% in group US. No differences were found in the probability of present encephalopathy. Three patients in the CS group had early liver failure vs. no patients in the US group. The mortality was similar in both groups (31 % vs 28 %). Conclusions: The use of CS significantly decreases TIPS dysfunction. However, early liver failure is more likely with CS.

P152

P154

Treatment of transjugular intrahepatic portosystemic shunt malfunction with VIATORR stents. Early experience G. Rodriguez Rosales, N. Alava Echevarria, M. Echenagusía Boyra, J. Guzmán de Villoria, G. Simó Muerza, F. Camuñez Alonso, A. Echenagusía Belda; Radiology, Hospital Gregorio Maranon, Madrid, Spain.

Pre-operative portal vein embolization in patients with hepatobiliary malignancy W. K. Tso, H. Tung, J. Yuen, F. L. Chan; Radiology, Queen Mary Hospital, Hong Kong, Hong Kong Special Administrative Region of China.

Purpose: To present our experience in the treatment of transjugular intrahepatic portosystemic shunt (TIPS) malfunction with e-PTFE covered stents (VIATORR). Materials & Methods: Nine patients with TIPS malfunction (portocaval gradient >12 mm Hg) were treated with covered stents (CS). The midterm between TIPS and CS placement was 35.4 months (average: 2-79 months). Causes of TIPS malfunction were: parenchymal tract stenoses in four patients, hepatic vein stenoses in four patients and shunt thrombosis in one patient. Results: In all the cases, CSs were successfully placed. The portocaval gradient decreased from 17.4±2.9 mm Hg to 6.0±2.6 mm Hg. No complications were observed during or after procedure. During follow-up (average: 7.7 months) no episodes of TIPS malfunction were observed in eight patients. One patient presented shunt thrombosis 14 months after CS implantation, successfully treated by placement of a new CS. Conclusion: In our preliminary experience CSs are a valid alternative in the treatment of TIPS dysfunction.

Purpose: To study the efficacy of pre-operative portal vein embolization (PVE) in inducing hypertrophy of the liver remnant (FLR) in patients with small livers who are marginal candidates for major hepatic resection. Materials & Methods: Over a 27-month period, September 2000 to November 2002, PVE were performed in eleven patients (eight male, three female, age 37-83, mean age 66.9, 10 with cholangiocarcinoma, one with hepatocellular carcinoma) by a percutaneous or transileocolic approach, using a specially designed triple-lumen balloon catheter and fibrin-glue containing thrombin-fibrinogen compound as the embolic agents. CT liver volumetry was obtained before PVE and 3-4 weeks after PVE. The absolute volume and percentage volume of the FLR were measured in each case. Results: PVE was technically successful in all patients, without significant procedure-related complications. PVE induced a significant increase in mean absolute volume and mean percentage volume of FLR (591.34 cc versus 467.25 cc, p<0.01 and 47.97% versus 37.73%, p<0.05, respectively). All patients proceeded to surgery 6-8 weeks after PVE. Seven had major hepatic resection, but found were found to be inoperable due to peritoneal metastases. There was no post-operative mortality. Conclusion: PVE is an effective and safe method in inducing hypertrophy for patients undergoing extensive hepatic resection.

P155

P157

Transjugular intrahepatic portosystemic shunts: experience at a single center B. Çil, B. Peynircioglu, S. Çekirge, F. Balkancý, A. Besim; Department of Radiology, Hacettepe University, Ankara, Turkey.

Duodenal varices: diagnosis and endovascular therapy J. Koizumi1, T. Hashimoto1, Y. Imai1, I. Kawamata1, I. Muro1, S. Kojima2, T. Kanai2, K. Watanabe2, I. Watanabe2, T. Mine2; 1 Diagnostic Radiology, Tokai University School of Medicine, Isehara-city, Kanagawa-pref., Japan, 2Internal Medicine, Tokai University School of Medicine, Isehara-city, Kanagawa-pref., Japan.

Purpose: Transjugular intrahepatic portosystemic shunting, introduced into clinical practice in 1988, is a safe and minimally invasive procedure that effectively provides portal decompression and an alternative method of treating complications of portal hypertension. Herein present our experience with TIPS. Materials & Methods: Sixty-four TIPS shunts were created in 60 patients (41 male, 19 female) in the 10-year period between 1993 and 2003. The patients were between 19 and 70 years old (mean 46). Fourteen were Child class C and 46 were Child class B. The follow up period was between six months and 10 years, using Doppler US and angiography as indicated. Results: Our technical success rate was 96.9% (the TIPS procedure failed in two patients). Our procedure related mortality was 1.6% (one patient died immediately after the procedure due to bleeding). In two patients biliary fistulas occurred and were treated with covered stents. Twelve patients were lost to follow-up, and 14 died during follow-up. Complications included shunt stenoses (n=21) and early shunt occlusion (n= 4). Two malpositioned stents were restented during follow up. Conclusion: The risk associated with TIPS placement is reasonable, and it is an effective procedure for the decompression of portal hypertension.

Purpose: To compare multidetector-computed tomography (MDCT) and magnetic resonance (MR) for planning the most suitable interventional procedure to treat duodenal varices. Materials & Methods: In two rare cases of varices in the third to the fourth duodenal portion, multiplanar reformations (MPR) using both MDCT and MR were performed. Based on anatomical data, the access route for the endovascular therapy was determined. Angiographic findings and procedures were compared with imaging findings. Results: MPR depicted the location of those vulnerable-to-rupture varices projecting into the lumen. MR with cine bFFE sequences visualized both the feeding and the draining veins of the target varices. Varices in the third duodenal portion (fed by the inferior pancreaticoduodenal vein and draining into the left gonadal vein) were mainly embolized by a transhepatic approach because closer to target varices. Duodenal varices in the fourth portion (fed by the first jejunal vein and draining into the left gonadal vein) were embolized by a transfemoral approach because closer to target varices. Conclusion: MDCT and MR were highly effective in visualizing target varices with both feeders and drainers.

P156

P158

Incidental visualization of the bile duct during transjugular intrahepatic portosystemic shunt. Can this be used as landmark for the portal vein puncture? K. Lee1, D. Y. Lee1, J. Y. Won1, K. Sung2, H. Yoon2, G. Ko2; 1Diagnostic Radiology, Yonsei University College of Medicine, Seoul, Republic of Korea, 2Radiology, Ulsan University College of Medicine, Seoul, Republic of Korea.

Multidetector CT-angiography of lower limb arteries: a useful tool for the vascular radiologist? E. Paci, S. Alborino, M. Maggi, R. Candelari, G. Centini, E. Antico; Dept of Radiology, Regional Hospital Umberto I°- Torrette, Torrette, Ancona, Italy.

Purpose: To assess whether the biliary tree can be used as a landmark for the portal vein (PV) puncture during transjugular intrahepatic portosystemic shunt (TIPS). Materials & Methods: Fifty-seven TIPS procedures were evaluated. MR images using true-FISP sequence for evaluating anatomic relationships between the PV and the bile duct were obtained from 200 patients. Results: The bile duct was visualized by tractogram in eight of the 57 (14%) patients. Using the bile duct as a landmark, we successfully accessed the right PV in three patients and the PV at the bifurcation level in five patients. On MR images, the PV bifurcation was located postero-medially to the bile duct in 179 cases (89.5%), posteriorly in 13 (6.5%) and medially in eight(4%) at the hepatic hilar level. The right PV was located postero-inferiorly to the right bile duct in 80 cases (40%) and posteriorly in 70 (35%). In the remaining patients (n=50, 25%), the bile duct crossed the right PV. Conclusion: A thorough understanding of the anatomic relationship between the PV and bile duct at the liver hilum is useful for successfully performing TIPS, and the incidentally visualized bile duct during TIPS can be used as landmark for PV puncture.

Purpose: To report our experience with multidetector computed tomography (MDCT), a new technique for evaluating peripheral artery diseases. Materials & Methods: Since February 2002, 50 patients (29 men, age 64±10y; 37 in Rutherford category 2 and 13 in category 3) were investigated by MDCT and, within ten days, by conventional arteriography of the legs. A 4-8 channel GE CT-scanner was used, 18-G needle-cannula, flow rate 4 ml/sec, 150 ml iodinated contrast medium 300/350 mgI/ml, bolus tracking program. A dedicated operator processed axial images into MIP and VR reformations. Two experienced vascular radiologists evaluated conventional and MDCT images in a blinded fashion, regarding: presence of stenoses (lumen reduction 50-99%); occlusions (>99%), or absence of lesions (<50%) in 950 arterial segments. Twentyfive patients underwent percutaneous treatment after MDCT: implications of MDCT images to choose the percutaneous approach were evaluated. Results: After data evaluation, the concordance between MDCT and conventional angiogram was 92.3% in lesions detection, 95.9% in lesions exclusion, 64.4 % in stenoses definition, and 88.8% in occlusions definition. In 18/25 patients, the percutaneous approach was modified according to the preliminary evaluation of CT images. Conclusion: MDCT arteriography is effective in evaluating peripheral artery diseases and can guide vascular radiologist's decisions.

P159

P161

Risk-factor profiles and changes in the anterior arterial circulation in patients with multiple lacunar brain infarctions F. Todua1, D. Gachechiladze2, M. Beraia3; 1Director, Institute of Radiology and Interventional Diagnostics, Tbilisi, Georgia, 2Ultrasound, Institute of Radiology and Interventional Diagnostics, Tbilisi, Georgia, 3Mrt, Institute of Radiology and Interventional Diagnostics, Tbilisi, Georgia.

Virtual endoscopy as a diagnostic or follow-up tool in renal artery stenosis D. Siablis1, G. C. Kagadis2, E. Liatsikos3, D. Karnabatidis1, C. P. Kalogeropoulou1, T. Petsas1, G. Nikiforidis2, G. Barbalias3; 1 Department of Radiology, University of Patras, Rion, Greece, 2 Department of Medical Physics, University of Patras, Rion, Greece, 3 Department of Urology, University of Patras, Rion, Greece.

Purpose: To study risk factor profiles, carotid stenosis severity, and atherosclerotic plaque stability in patients with lacunar brain infarctions (LI). Materials & Methods: 62 patients (mean age 57.2y) with MRT evidence of LI were examined. All underwent brain MRT, MRA of extra-intracranial vessels, extracranial Color Doppler (CD), transcranial Doppler (TCD), and cardiologic examination to exclude cardiogenic embolism. Thirtyseven patients had symptomatic cerebral ischemia. Results: LI were located in basal ganglia, periventricular white metter, or cortex. Investigation revealed frequent CA stenosis (45.2% mild, 32.2% moderate) but infrequent high-grade lesions (17.7% severe, 4.8%occlusion). There were embologenic plaques in 53.3% (8.1% soft, 45.2% heterogenous). CD showed increase of mean CCA intima-media thickening (1.12mm). TCD revealed decreased flow in anterior circulation arteries compared to age-matched normal subjects and an increased pulsatile/resistive index (mean flow velocity in MCA-48.8sm/ s, PI-1.02). Hypertension was positively associated with LI, while frequency of IHD and peripheral vascular disease was comparatively low. Conclusion: These results suggest that LI patients have only mild to moderate CA stenosis, so LI is more likely to be due to intracranial small-vessel disease. Embolism due to unstable plaque may be a cause of multiple LI as of non-lacunar infarctions. Arterial hypertension is the main risk factor.

Purpose: We propose the use of virtual endoscopy (VE) in patients with renal artery stenosis as a non-invasive diagnostic and follow-up tool especially when there is the the need to define vascular patency. Materials & Methods: We performed the proposed technique in 35 patients with ostial atherosclerotic renal artery stenosis and correlated the results with conventional angiography. The stenotic segment was successfully depicted with the aid of VE. These patients were treated successfully by placement of metal stents, and VE was performed as a follow-up study for patency control. The patients were followed for a mean period of 12 months. Results: In all patients the stenotic segment was observed, and VE findings were concurrent with angiography examination. In addition, the 12-month patency rate after metal stent insertion was 85%. Angiography and VE findings were also concordant during the follow up period. Moreover, VE proceeds beyond the stenotic part, allowing visualization of the arterial lumen both cephalad and caudal to the obstruction point. Conclusion: VE provided a more dynamic, direct approach to evaluation of the arterial lumen as well as in evaluation of a vascular stricture both cephalad and caudal to the obstruction point.

P160

P162

Contrast-enhanced color Doppler sonography evaluation of internal jugular venous valve competence M. S. Ilic; Dpt for Vascular and Interventional Radiology, Institute of Radiology UMC Nis, Nis, Yugoslavia.

Automated curved planar reformatted multislice CT-angiography: usefulness in the evaluation of the carotid syphon. Initial assessment A. Mallouhi, M. Rieger, P. Waldenberger, W. R. Jaschke; Department of Radiology, Innsbruck University Hospital, Innsbruck, Austria.

Purpose: The internal jugular vein valve is the only valve between the heart and the brain, preventing venous reflux into the IJ vein. The absence or incompetence of the IJV has been associated with many diferent neurologic symptoms. The purpose of this study was to demonstrate the sonographic appearance and competence of the IJV using color Doppler sonography and contrast-enhanced color Doppler sonography. Materials & Methods: Evaluation of the color Doppler sonographic appearance of these structures was performed in cohort of 150 adults. Twenty-five were observed by CDE using an ultrasound contrast agent. Results: During sonographic examination we were able to detect venous valves in 87.3% of patients (bilaterally in 62% and unilaterally in 25.3% of the patinets). Retrograde jugular venous flow was seen in 11 patients without IJ valves and in 14 patients where valves were presentò they were observed using contrast-enhanced color Doppler sonography with Levovist to confirm incompetence Conclusion: Contrast-enhanced color Doppler sonography is a reliable noninvasive method for evaluating incompetence of IJ valves. Potential clinical applications include patients with increased central venous pressure, those with morning headaches, and those with transient global amnesia.

Purpose: To assess the diagnostic potential of automated curved planar reformatted (ACPR) multislice CT-angiography (CTA) in the evaluation of the carotid syphon. Materials & Methods: Thirty-four patients who had undergone multislice CTA and DSA of intracranial arteries at the same moment were retrospectively recruited. CPRs of the carotid syphon were automatically generated by the "Vessel Analysis" software. Two independent radiologists rendered the images of petrous, supraclinoid, and intracavernous segments in a blinded fashion. Cut-off points for stenoses classification were 30% and 70%. Results: For all the 68 carotid arteries, diagnostic CPRs could be automatically performed. The surrounding bony structures resulted in no artifacts leading to image degradation. Thanks to the ability of an interactive rotation of ACPR images, the arterial lumen of the syphon segments could be comprehensively evaluated from arbitrary perspectives. Two intracavernous aneurysms were detected by ACPR. ACPR enabled the automatic quantification of the lumen by means of vessel diameters and area. Linear regression analysis showed that ACPRs yielded a significant correlation with DSA in all syphon segments (petrous, r2=0.84; supraclinoid, r2=0.81; intracavernous, r2=0.89). Conclusion: ACPR, as a complement to post-processing techniques of multislice CTA of intracranial vasculature, adds valuable information on the carotid syphon.

P163

P165

Renal magnetic resonance angiography reliably estimates the severity of renal artery stenosis S. B. Habib, K. S. Blanshard; Radiology, Leicester General Hospital, Leicester, United Kingdom.

Characterization of cardiovascular structural changes in Fabry's disease with magnetic resonance and high-resolution ultrasonography E. Rizzo1, F. Barbey2, D. Hayoz3, E. Heasler3, D. Teta4, S. D. Qanadli1; 1Radiology, CHUV University of Lausanne, Lausanne, Switzerland, 2 CHUV University of Lausanne, Lausanne, Switzerland, 3Angiology, CHUV University of Lausanne, Lausanne, Switzerland, 4Nephrology, CHUV University of Lausanne, Lausanne, Switzerland.

Purpose: To evaluate the reliability of magnetic resonance angiography (MRA) of the renal arteries in estimating the severity of stenosis. Materials & Methods: We have retrospectively reviewed 15 MRA patients who also had digital subtraction angiography (DSA) upon MRA recommendation. There were nine men, mean ages of 66 years. All patients underwent contrast-enhanced 3-D Time-of-Flight MRA. Results: Twelve patients (80%) with both MRA and DSA had a 90% stenosis, involving at least one renal artery. Ten patients had successful stent deployment and two patients continued on medical treatment. One (6.6%) renal transplant artery had normal MRA and DSA, but DSA was undertaken to investigate the severe atheroma seen in the iliac arteries on MRAs. There were two discrepancies between MRA and DSA results (13.3%). There were one false positive and one false negative. Conclusion: MRA has high sensitivity and specificity in detecting significant renal artery stenoses.

Purpose: To characterize structural changes of the myocardium and the vascular wall in Fabry's disease and to determine how this is an inherited lysosomal storage disorder caused by a deficiency of the alphagalactosidase which induces cardiovascular morphologic changes. Materials & Methods: Eleven patients (six men) were investigated with high-resolution ultrasound (US) of the common carotid artery and cardiac MRI for structural cardiovascular alterations. Mean age was 43±16 (range, 17-63 years). The common carotid artery thickness (IMT) and the interventricular septal thickness (IST) of the myocardium were measured in each patient. Results: A homogenous hypertrophy was observed in 10/11 patients. Mean carotid IMT and mean myocardium IST were 786±89 mm and 16.3±1.2 mm, respectively. These values were significantly higher than those obtained in an age matched control population. Significant and positive correlation between carotid IST as well as myocardium IMT and age was observed (r=0.84; p<0.004 and r=0.71; p<0.023, respectively). Furthermore, a good correlation was observed between carotid IMT and myocardium IST (r=0.88, p=0.0014). Conclusion: Fabry's disease can be characterized by a homogenous hypertrophy of the myocardium and the vascular wall. Our findings, however, suggest that hypertrophy may be the consequence of a growth factor effect rather than its storage.

P164

P166

Detection of coronary artery calcifications with multislice CT: optimization with a 3-D soft segmentation method S. D. Qanadli1, J. Costello2, E. Rizzo1, B. Bessoud1, P. Schnyder1, J. Dehmeshki3; 1Radiology, CHUV University of Lausanne, Lausanne, Switzerland, 2Kings College Hospital, London, United Kingdom, 3Medicsight PLC, London, United Kingdom.

Peripheral arterial disease: evaluation with MR-angiography versus conventional angiography G. Puppini1, S. Montemezzi1, F. Furlan1, G. Veraldi2, G. Gortenuti1; 1Servizio Radiologia OCM B.go Trento, Verona, Italy, 21° Chirurgia clinicizzata OCM B.go Trento, Verona, Italy.

Purpose: To define a new concept for the detection of coronary artery calcifications (CAC) with multislice CT based on a 3-D soft segmentation method and in vivo attenuation x-ray values of blood and myocardium. Material & Methods: Radiological attenuation values of the blood (Ab) and myocardium (Am) were determined in a population of 124 patients (89 men, 35 women, mean age 63±12 years) who had non-enhanced cardiac CT. Ab and Am were measured in the ascending aorta and the myocardium. The mean and maximum values as well as the distribution of these parameters were analyzed regarding the thresholds (90 HU [T90], 130 HU [T130]) usually used to detect CAC. Results: Mean Ab was 45±10 HU and maximum Ab was 72 HU. Mean Am and maximum Am were 49±11 HU and 87 HU, respectively. Based on these data, the thresholds for detecting CAC could be expressed as T90=[(lb+lm)/2+4SD] and T130=[(lb+lm)/2+8SD]. T130 is probably greater than needed to accurately detect CAC. A new approach using 3D soft segmentation to optimize calcium detection is presented here. Conclusion: The authors present a new approach to optimize detection of CAC with non-enhanced multislice CT.

Purpose: To assess the value of MR-angiography (MRA) with automatic table movement in patients with peripheral artery disease compared with conventional angiography (CA). Materials & Methods: Twenty-five patients underwent both CA and MRA for peripheral arterial occlusive disease. A flash 3-D FS angiographic sequence (TR 3.5, TE 1.2 slice thickness, 1.5 mm FOV 500 Flip Angle 25 time of acquisition 75 sec) with a dedicated lower-estremity coil was performed and--within seven days--CA was referred as the gold standard. The vascular tree was divided into 21 segments from the aortic bifurcation to the ankle. For every segment, two indipendent observers estimated the degree of stenosis using six different classifications: 1) evaluation impossible, 2) negative, 3) stenosis between 1 and 50%, 4) 51-70% stenosis, 5) 71-99% stenosis, and 6) occlusion. Results: Sensitivity, specificity, positive and negative predictive values for the detection of stenoses were 91, 99.3, 99, and 93.1%, respectively. Sensitivity, specificity, positive and negative predictive values for the grading of hemodynamic stenoses were 62.5, 93.7, 64.8, and 93.1%, respectively. Conclusion: Gadolinium-enhanced MRA with automatic table movement can be accurately performed in patients with peripheral artery disease in a short time and with good results.

P167

P169

Rotational digital subtraction angiography versus conventional digital angiography of the ascending aorta in post-CABG angina patients V. Klimov, A. Ardashev; The Burdenko Main Veterans Clinical Hospital, Moscow, Russian Federation.

Peripheral vascular disease: comparison of digital subtraction angiography with multistep MR-angiography optimized for the calf station N. Ikonomopoulos, N. L. Kelekis, E. N. Brountzos, M. Papathanassiou, E. Alexopoulou, D. A. Kelekis; 2nd Radiology, Athens University, Athens, Greece.

Purpose: To compare rotational DSA of the ascending aorta with conventional DA in the assessment of aortocoronary bypass graft patency in post-CABG angina patients. Materials & Methods: One hundred nine angina patients three months to 14 years post-CABG were studied. The examination was carried out in two stages: (1) conventional selective coronary graft DA followed by conventional ascending aorta DA in all patients; (2) RDSA of the ascending aorta of 64 patients with 113 bypass grafts. The parameters of interest were image quality and diagnostic value of RDSA and conventional DA, and the number and volume of contrast injections. Results: In 95.6% of cases (108 of 113 grafts) a single RDSA procedure was sufficient to demonstrate all proximal graft anastomoses. An average number of 3.5 DA procedures were informative in 39.8% (74 of 186 grafts). There were 7.7% false-negative results obtained using conventional DA. In contrast, RDSA had no false-negatives. RDSA required three times less contrast medium than DA . Conclusion: RDSA of ascending aorta is superior to conventional DA technique in the assessment of proximal anastomoses of either patent or occluded grafts in post-CABG angina patients.

Purpose: To assess an optimized MR-angiography (MRA) protocol for the visualization of distal vessels and compare it with digital subtraction angiography (DSA). Materials & Methods: Nine symptomatic patients were prospectively evaluated for peripheral vascular disease with DSA and MRA. Multistep MRA was performed on a 1.5-T MR equipment using a phased-array body coil and a submillimeter resolution for the distal station--which was acquired first--and a body coil for the two upper stations. Each limb was examined in 19 vascular segments using a 4-point lesion severity scale. Results: Three hundred forty-two segments were evaluated by both modalities. MRA depicted 46 segments not visualized by DSA. Pedal vessels were consistently depicted by MRA. In 280 segments both modalities were concordant for lesion classification. In 15 segments, the two modalities differed of no more than one category; MRA diagnosed one false obstruction. Conclusion: MRA correlates well with DSA and is accurate in the evaluation of peripheral vascular disease. Dedicated calf station scanning improves visualization of small vessels and prevents image deterioration by venous contamination.

P168

P170

Indications for application of data catheter arteriography for contrast-enhanced magnetic resonance angiography M. Y. Voloduckin, M. M. Ibatullin, M. N. Malinovskiy; Angiography, Interregional Clinic Diagnostical Centre, Kazan, Russian Federation.

Illustrated history of angiography, Part Iii D. Novak; Medical Faculty University J. J. Strossmayer, Osijek, Croatia, Cakovec, Croatia.

Purpose: To determine the time to appearance of contrast material in various segments of lower limb arteries on the data catheter arteriography, and to apply this data in performing contrast-enhanced magnetic resonance (MR) angiography Materials & Methods: Ratings were performed of 92 catheter arteriography images of patients with peripheral arterial occlusive disease. We assessed the following parameters: the time of appearance of contrast material in the segment, the symmetrical appearance of the contrast material in left and right lower limbs. Results: The time to appearance of contrast medium below the femoral level showed wide scatter, from 6 to 36 sec. Asymmetrical appearance of contrast substance below the femoral level was found in 35% of patients. Conclusion: These data indicate wide scatter in time and asymmetrical appearance of contrast substance in lower limb arteries below the level of the femoral artery. Therefore, interpretation of contrast-enhanced MR angiography is complicated by the combination of time to contrast appearance and K-space below this level. As a result, contrast enhanced MR angiography of these areas will be not be interpretable.

Purpose: To achieve an overview of the historic development of angiography the original publications were studied and systematized in chronological order. The biographies of angiography pioneers are illustrated with portrait photographs. Angiographic images from the original papers demonstrate the first results published. Results : Historical landmarks: 1.Pioneers of Angiography in Sweden : S Radner, J Bergstrand, S Cronquist, E Lindgren, T Greitz, G Joensson, T Olsson, E Boijsen, P Oedman, B Nordenstroem, T Olin, a Gidlun, L Ekelund, J Goethlin, C Hellekant, A Lunderquist, G Nylander, U Tylen . 2. Further Development of Angiocardiography : SR Kjellberg 1955), HL Abrams (1958), 3. Selective Renal Arteriography: O Olsson (1955), P Oedman (1956), P Edholm , SI Seldinger (1956), E Boijsen (1959), 4. Selective Visceral Arteriography: P Oedman (1958), JA Evans (1964), E Boijsen ; SR Reuter (1966), 5. Superselective Arteriography: M Viamonte (1964), RE Paul (1965), PC Kahn (1967), SR Reuter (1969), 6. Pharmaco-angiography: HR Bierman (1951), HL Abrams (1962, 1964), 7. Digital Subtraction Angiography : J Steinberb (1962), CA Mistretta (1973), AB Crummy (1981).

P171

P173

Detection of leakages after endovascular repair of aortic aneurysms prior and post-embolization: comparison of MRI and multislice-CT S. C. Krämer1, J. Görich2, R. Pamler3, R. Fischbach1, H. Brambs2, W. Heindel1; 1 Diagnostic Radiology, University of Münster, Münster, Germany, 2 Diagnostic Radiology, University of Ulm, Ulm, Germany, 3Thoracic and Vascular Surgery, University of Ulm, Ulm, Germany.

Contrast enhanced magnetic resonance angiography during followup of kidney transplants A. Stecco1, P. Oronzo1, G. Marano1, E. Turello2, S. Stangalini1, A. Carriero1; 1 Istituto di Radiologia, A.O. "Maggiore della Carità", Novara, Italy, 2 Divisione di Nefrologia e Dialisi, A.O. "Maggiore della Carità", Novara, Italy.

Purpose: Comparison of MRI and multislice-CT (MSCT) in the follow-up of endovascular repair of aortic aneurysms for the detection and postinterventional control of endoleaks. Materials & Methods: After endovascular repair of aortic aneurysms ten consecutive patients were suspected to have endoleaks. As a clinical routine, MSCT was done prior and after the treatment. After having obtained an informed consent, MR-imaging and MRA were also performed. Evaluations of MSCT and MRI were done using a score from 0 to 3 for leak detection, imaging of the feeding vessel, and artefacts. Statistics were done by Wilcoxon testing (p<0.05). Results: Pre-intervention, MRI was able to detect all ten leakages and MSCT nine. MRA visualized the feeding vessel in about 60% of cases. After embolization, MRI showed a successful treatment in five cases and persisting leakages in the remaining five. MSCT detected four persisting leaks whereas six examinations were not reliable due to strong artefacts. A statistical significance could be achieved for the feeding vessel, successful leak occlusion and artefacts. Conclusion: MRI is comparable with MSCT in the detection of endoleaks. In the feeding vessel imaging and post-interventional control, MR is superior to MSCT and should be considered the golden standard.

Purpose: To evaluate the arterial complications following kidney transplantation by mean of Contrast enhanced magnetic resonance angiography (CE-MRA). Materials & Methods: Between January 1999 and December 2002, 30 consecutive transplanted patients with clinical suspected vascular complications (hypertension, laboratory signs of graft failure) were evaluated with CE-MRA. All studies were performed on a 0.5T unit, with a body-flex coil. After a pre test-bolus 2D sequence, we acquired images with enhanced 3D fast gradient echo (GRE 3D) pulse sequence before and after injection of 30 cc of GD-BOPTA at 2.0 ml/sec. The acquisition data were post-processed with digital subtraction and MIP reconstruction. Two independent observers retrospectively evaluated the MR angiography sequences to detect or exclude significant (>= 50%) arterial stenosis. Each evaluation was performed on both MIP and raw data images. Results: In 18 of 30 patients (60%) CE-MRA was positive, showing significant arterial stenosis (extrinsic or intrinsic). All 18 patients underwent Digital Subtraction Angiography (DSA), which confirmed arterial lesions in all cases. CE-MRA and DSA had a concordance rate of 100%. Conclusion: MR imaging protocols allow complete noninvasive anatomical and functional evaluation during follow-up of kidney graft complications, with little nephrotoxicity. CE-MRA demonstrates high sensitivity in detecting significant arterial stenosis.

P172

P174

Evaluation of iliac vein compression in healthy persons B. Roh; Radiology, Wonkwang University Hospital, Iksan City, Republic of Korea.

Branch involvement in descending aorta dissections: radiologic evaluation with spiral CT and angiography K. Milczarek, O. Rowinski, M. Januszewicz, E. Maj; Radiology, SPCSK AM, Warsaw, Poland.

Purpose: To evaluate iliac vein compression by the overlying iliac artery in healthy persons Materials & Methods: Fifty-eight healthy adult volunteers (24 men and 34 women, average age 47 years) were studied. MR venography of transverse 2-D TOF images covering the both iliac veins were obtained, and 3-D reconstruction images were produced by MIP. We described the degree of iliac vein compression and the anatomical structure of the IVC and both iliac veins and correlated the degree of iliac vein compression with sex, age, body weight. Results: There was no indentation of the venous lumen in 10 persons (17%), mild indentation in 26 (45%), severe compression with flow disturbance in 15 (26%), and complete occlusion in 7 (12%). The anatomical structures were left iliac vein compression by right iliac artery in 38(79%), left iliac vein by left iliac artery in 5(10%), left iliac vein by right and left iliac artery in 5(10%). The degree of iliac vein compression correlated significantly with body weight (p<.05). Conclusion: The incidence of iliac vein compression by the overlying iliac artery is very high in healthy persons, and there are various degrees and anatomical structures of iliac vein compression.

Purpose: To evaluate branch involvement in aortic dissection and describe pitfalls and artifacts in spiral CT and angiography. Materials & Methods: From January 2000, 51 patients (16 women) were diagnosed at our institution. Mean age was 52 years (18-81). Thirty-six patients had type B dissection, and 15 had type A dissection previously treated surgically. Radiologic evaluation included spiral CT and angiography. All examinations were reviewed for aortic branch involvement. Results: The most frequently dissected branch was the left common iliac artery (n=21), corresponding to the most frequent distal extension of dissection. The LRA was the most frequently affected aortic branch (supplied by the false lumen in 14, dissected in five, occluded in two). The RRA caused most problems in evaluation (difficulty assessing the supplying channel in four cases). TC and SMA dissections were the most frequent pathology of these branches, with six cases each. Angiography was superior in detecting obstruction of true branch lumen (26 versus 19 for CT) and demonstrating flow patterns. Conclusion: We found the typical course of aortic dissection to determine the frequency of involvement of different branches. CT is very sensitive in detecting involvement, but angiography remains the examination of choice for defining aortic branch compromise.

P175

P177

Gadopentate dimeglumine: an alternative contrast agent for threedimensional computed tomography angiography with multidetector-row helical scanning A. K. Gupta, D. Engin, R. Alberico, Z. Grossman; Diagnostic Imaging, Roswell Park Cancer Institute, Buffalo, United States.

Diverticula of Kommerell in association with anomalous subclavian arteries R. G. Fisher, C. J. Whigham, C. C. Trinh; Department of Radiology, Baylor College of Medicine, Houston, United States.

Purpose: To demonstrate that gadopentate dimeglumine is an alternative contrast medium for computed tomographic angiography (CTA) of the cervical vessels. Materials & Methods: The cervical vessels of three adult human patients were imaged for presurgical planning of the neck. Gadopentate dimeglumine, 0.5 mol/l (Berlex Laboratories, Wayne, NJ, USA), a Light Speed QX/i CT (General Electric Medical Systems, Milwaukee, WI, USA) and an Ultra Sparc II (SUN Microsystems, Santa Clara, CA, USA) running Advantage Windows 3.1 (General Electric Medical Systems) were used. Results: A sufficient enhancement for CTA of the cervical vessels in all three patients was obtained. Conclusion: In that subset of patients with contraindications to iodinated contrast medium and for whom magnetic resonance angiography is inappropriate, gadopentate dimeglumine was an alternative contrast medium for CTA of the cervical vessels.

Purpose: The purpose of this exhibit is to share our experience with diverticula of Kommerell in association with anomalous subclavian arteries. Materials & Methods: Over the past 12 years in a major metropolitan level-I trauma center, approximately 150-200 arch aortograms were performed annually for trauma and/or cerebral vascular disease. Vascular anomalies involving the subclavian arteries were noted relatively infrequently. Approximately, one or two cases with anomalous subclavian arteries were seen annually with the majority having a diverticulum of Kommerell. Results: Nineteen patients are included in this series with anomalous right (15) or left (4) subclavian arteries. Diverticula of Kommerell were recognizable in 15 patients. Three of these had aneurysms involving the diverticulum which were demonstrated by angiography and computed tomography. One contained thrombus, another was associated with a localized dissection which originated in the diverticulum, and the third was ruptured. All three were managed surgically. Conclusion: Anomalous right or left suclavian arteries are relatively uncommon and frequently are associated with diverticula of Kommerell. Rarely, aneurysms involve the diverticula and carry a very high mortality rate if they are not discovered before rupture. Early diagnosis and surgical management can be lifesaving.

P176

P178

Magnetic resonance imaging of renal artery stenoses compared with intraarterial pressure gradient measurement H. Eklöf1, H. Ahlström1, D. Bergqvist2, A. Boström-Ardin2, B. Andrén3, R. S. Nyman1; 1Radiology, ORKI, Uppsala, Sweden, 2Surgery, Uppsala, Sweden, 3Clinical physiology, Medical sciences, Uppsala, Sweden.

MRA findings of pulmonary hypoplasia and aplasia C. Basekim1, H. Mutlu1, E. Silit1, Z. Pekkafalý1, Ö. Deniz2, E. Öztürk1, E. Kýzýlkaya1; 1Radiology, GATA Haydarpaþa Teaching Hospital, Ýstanbul, Turkey, 2Chest Diseases, GATA Haydarpaþa Teaching Hospital, Ýstanbul, Turkey.

Purpose: To evaluate 3-D gadolinium-enhanced MR-angiography (MRA) in the screening of renal artery stenoses (RAS). Materials & Methods: Three independent readers compared MRAs with intraarterial pressure gradient measurements. From January 1996 to September 2000, 31 patients with a clinical suspicion of RAS were included in the study. Results: Sixty-one main renal arteries and nine accessory renal arteries were visualized by digital subtraction angiography. Three (33%) accessory renal arteries were identified by all three readers. Of the 29 arteries with a critical gradient (15 mm Hg or more), the three readers correctly classified 27/23/24 stenoses to be 60% or greater on MRA. Sensitivity was 93, 79, and 83%, respectively. Of the 32 renal arteries with no critical gradient (peak systolic gradient <15 mm Hg), the three readers correctly classified 25/28/30 stenoses to be less than 60% on MRA. Specificity was 78, 87, and 94%, respectively. Conclusion: 3-D gadolinium-enhanced MRA provided good anatomical images of the renal arteries from the aorta to the renal hilus, but accessory renal arteries were not always visualized. Evaluation of renal artery stenoses is more hampered by the individual reader’s subjective opinion than by the technical shortcomings of MRA.

Purpose: To describe MRA findings in those patients with pulmonary hypoplasia and aplasia. Materials & Methods: In 11 patients MRA was performed because of radiological suspicion of pulmonary hypoplasia or aplasia. Additionally, five patients were evaluated by angiography, seven patients were evaluated by CT and four patients were evaluated by scintigraphy. Results: MRA showed pulmonary hypoplasia in seven patients (in four patients on the left side and in three on the right side) and aplasia in the remaining four patients (in three on the left side and in one on the right side). These findings were confirmed by the other radiological studies. Conclusion: MRA is a valuable and non-invasive method in the detection of pulmonary hypoplasia and aplasia.

P179

P181

The European project AQUATICS: aneurysm quantification through an internet collaborative system E. Neri1, I. Bargellini1, A. Giachetti2, M. Tuveri2, A. Mallouhi3, D. Caramella1, G. Zanetti2, W. R. Jaschke3, C. Bartolozzi1; 1 Diagnostic and Interventional Radiology, University of Pisa, Pisa, Italy, 2 CRS4, Cagliari, Italy, 3Diagnostic Radiology, University of Innsbruck, Innsbruck, Austria.

The role of 3-D CTA in endovascular treatment of superior vena cava syndrome J. Kim, W. Yoon, N. Lim, Y. Moon, E. Lee, J. Park; Department of Radiology, Chonnam National University School of Medicine, Gwangju, Republic of Korea.

Purpose: The endovascular treatment of abdominal aortic aneurysms (AAA) has become an appealing alternative to standard surgical repair; it requires an accurate analysis of the aorta and its branches. Aim of this trial was to introduce a web-based application for the planning of AAA endovascular treatment. Material/Methods: In the European project AQUATICS (aneurysm quantification through an internet collaborative system), a web server has been developed to manage remotely virtual reality modeling language (VRML) 3-D models of the aorta, based on multirow CT-angiography datasets. Models are created by a software through aortic semi-automatic segmentation. Models were created by using an AAA phantom of and CT datasets of patients with AAA. The software provides several tools: measurements of diameters, angles and centerline lenghts, testing of the vessel patency, virtual navigation. Results: Trained radiologists tested the server--by connecting to the Internet and performing a detailed analysis of VRML models--to depict aneurysm's features and facilitate the selection of the appropriate stent-graft. Measurements obtained from phantom models proved to be accurate and reproducible. Conclusion: The main expected result of the AQUATICS project was to show that sophisticated image processing of the AAA to facilitate treatment planning through an Internet service is possible.

Purpose: To evaluate the role of 3-D CTA for endovascular treatment of SVC obstruction by lung malignancies. Materials & Methods: Ten patients with SVC syndrome due to malignancy of the lung were treated by stent placement. Prior to the endovascular procedure, 3-D CTA was performed to evaluate the level and extension of SVC obstruction in all patients. The etiology was non-small cell lung cancer in nine patients and tracheal adenocarcinoma in one patient. Venography and the endovascular procedure were performed via the jugular vein in eight patients and via the femoral vein in two patients. Results: The pre-procedure 3D-CTA clearly depicted the level and extension of SVC obstruction in all cases. 3-D CTA correlated well with DSA angiography for evaluation of the lesion. Recanalization was successful in all patients. Thirteen stents were implanted in the 10 patients. Resolution of symptoms was achieved in eight patients within 48 hours and in the other two within a week. Recurrent thrombotic occlusion of the stent occurred after several days later in one patient who was successfully treated by secondary stenting. Conclusion: 3D-CTA is helpful in planning endovascular treatment. Such treatment is safe and effective in the management of malignant SVC obstruction.

P180

P182

Color-guided duplex scanning, MRA and multislice CTA versus DSA in lower limb peripheral arterial disease G. Marano1, G. Guzzardi1, A. R. Cotroneo1, D. Savio2, C. Rabbia2, F. Nessi3, A. Carriero1; 1Istituto di Radiologia, A.O. "Maggiore della Carità", Novara, Italy, 2Servizio di Radiologia Vascolare ed Interventistica, A.O. "San Giovanni Battista", Torino, Italy, 3Divisione di Chirurgia Vascolare, A.O. "Maggiore della Carità", Novara, Italy.

Transluminal biopsy of cavo-atrial tumors with the Quick-Core needle L. Hechelhammer1, M. Kurrer2, J. Seebach3, T. Schertler1, B. Marincek1, T. Pfammatter1; 1Diagnostic Radiology, University Hospital, Zürich, Switzerland, 2Diagnostic Pathology, University Hospital, Zürich, Switzerland, 3Internal Medicine, University Hospital, Zürich, Switzerland.

Purpose: To compare the diagnostic accuracy of color-guided duplex scanning (CGDS), MR-angiography (MRA) and multislice CT-angiography (CTA) compared with digital subtraction angiography (DSA) in lower limb peripheral arterial occlusive disease. Materials & Methods: Fifteen consecutive patients (13 men, 2 women; average age 70.6 years) with symptomatic lower limb arterial disease were imaged with CGDS, MRA, multislice CTA, and DSA (gold standard). All studies were performed using the same protocol and evaluated blindly by four readers for vessel stenosis and occlusion. Sensitivity, specificity, and positive and negative predictive values were analysed. Agreement among techniques was evaluated with Cohen's K test. Results: Overall 223 to 314 lesions were observed. Concordance of CGDS, MRA and CTA with DSA in detecting significant stenosis (> 70%) and occlusion was higher in iliac (K 0.64 to 0.72) and femoro-popliteal districts (K 0.71 to 0.75), with comparable image quality. CGDS and MRA (K 0.41 and 0.46 respectively) had better agreement with DSA than CTA (K 0.38) in evaluating below-knee segments. Conclusion: Our data suggest that when compared with DSA, CGDS can be considered as the best imaging modality in early evaluation of peripheral vascular occlusive disease; MRA and CTA may be helpful only in clinically selected cases.

Purpose: To report on the variant use of the transjugular liver or kidney Quick-Core needle for tissue sampling of cavo-atrial tumors. Materials & Methods: The Quick-Core system was used in three patients, aged between 51 and 76, with tumors extending from the superior (n=2) or inferior (n=1) vena cava into the right atrium. All these masses were initially diagnosed by CT. The tumors were biopsied under fluoroscopic guidance with the Quick-Core system using transjugular (n=1) or transfemoral (n=2) access. Results: All the specimens were diagnostic as proved by subsequent tumorectomy (primary leiomyosarcoma, metastatic adenocarcinoma of unknown origin) or response to chemotherapy (non-Hodgkin's lymphoma. There were no procedural complications. Conclusion: Initial experience of transluminal cavoatrial biopsies with the Quick-Core system suggests feasibility associated with high diagnostic yield.

P183

P185

Retrievable vena cava filters: our early experience P. Filauri1, G. De Blasis2, G. Bafile2, M. D'Elia2, G. L. Turco2, L. Scalisi2, M. Pinelli2, G. Passalacqua1; 1 Radiology, Civil Hospital S.Filippo e Nicola Hospital, Avezzano, Italy, 2 U.O. Vascular Surgery, Civil Hospital S.Filippo e Nicola Hospital, Avezzano, Italy.

Radiologic placement of central venous access devices: a single center's experience B. Çil, S. Geyik, D. Akinci, F. Balkancý; Department of Radiology, Hacettepe University Faculty of Medicine, Ankara, Turkey.

Purpose: To evaluate preliminary clinical experience with retrievable ALN inferior vena cava (IVC) filters, including facility and safety of retrieval. Materials & Methods: From August 2001 to January 2003 eight patients, mean age 43 years (range 29-77), were treated with ALNIVC filters. All had deep venous thrombosis, with extensive ilio-femoral thrombus in two and ilio-caval thrombus in one. Indications were pulmonary embolism (PE) during anticoagulant therapy in three, contraindication to or complication of anticoagulant therapy in four, and PE prophylaxis in one. All filters were introduced with a brachial approach and retrieved without complications with a right jugular approach. The mean implantation period was 85 days (range 46-153). The mean radioscopy time was 5.5 minutes (range1.6-15.3). Follow-up was medical and radiologic, with abdominal radiography,CT and cavography in all patients before filter removal to estimate filter position and presence of thrombus trapped within the filter. Results: In all patients the filters were correctly placed. Tilting was seen in 3/8 patients. There were no complications during filter positioning. No patient developed new episodes of pulmonary embolism. In all cases the filters were retrieved without complications Conclusion: Retrievable ALN filters are effective in preventing PE; insertion and retrieval are safe and feasible.

Purpose: To report a single center's experience with the placement, complications, and outcome of central venous access devices. Materials & Methods: Between April 2002 and October 2002 150 central venous access devices, 60 tunneled internal jugular catheters, 40 chest ports, and 50 non-tunneled jugular catheters were placed in adult patients.Venipunctures were all performed under sonographic guidance. The right internal jugular vein was used for access in 92 % (140) of procedures Results: All catheters except one were successfully inserted. One chest port procedure was terminated due to development of pectoral subcutaneous hematoma during local anaesthetic injection for port pocket incision. Asymptomatic air embolism occurred during insertion in two patients and inadvertent arterial puncture occured in three patients, both complications were managed conservatively. The initial complication rate was 3 %, with no infectious complication..Catheter removal was done electively. Follow-up was 21,890 catheter-days in total (mean time: 268 catheter-days, 154 catheter-days, and 19 catheter- days for chest ports, tunneled catheters and non-tunneled catheters respectively. Conclusion: Using sonographic guidance, inserting central venous access devices is a safe and effective technique that has high rates of technical success and low rates of complication.

P184

P186

Iliac vein compression syndrome: diagnosis and endovascular treatments in Japan H. Minamiguchi, S. Sahara, M. Kimura, M. Sato; Radiology, Wakayama Medical University, Wakayama, Japan.

Bilateral simultaneous adrenal vein sampling in the differential diagnosis of primary aldosteronism T. Ishibashi, A. Sato, H. Saito, S. Ono, T. Matsuhashi, K. Takase, S. Takahashi; Diagnostic Radiology, Tohoku University School of Medicine, Sendai, Japan.

Purpose: To review the iliac vein compression syndrome in Japan from the viewpoints of imaging diagnosis and endovascular treatments. Materials & Methods: From January 2001 to January 2003, 35 patients with leg swelling were referred to our department. The diagnosis was achieved by ultrasound sonography, conventional venography, and CT venography. Iliac vein stenosis was diagnosed by CT, venography and intravascular US. Results: The diagnosis in 28 patients was deep vein thrombosis (DVT), in seven patients no DVT (hematomas in two, lymphedema in two, injuryrelated swelling in two, hypoalbuminemia in one). Iliac vein compression syndrome was observed in 14 patients. Twelve patients with acute DVT were treated with catheter-directed thrombolysis and stenting. No recurrence was found in this group. One patient with chronic DVT was treated in the same manner as the acute DVT group, but recurrent DVT occurred. ONe patient with lymphedema and retroperitoneal fibrosis due to malignant metastasis was treated by stenting for venous stenosis, but leg swelling persisted. Conclusion: Diagnosis and accurate treatment of iliac vein compression syndrome are important and provide good results.

Purpose: Multidetector CT (MDCT) scanning may not detect small aldosterone-producing adenomas (APA), while CT-detected tumors do not necessarily produce aldosterone. Adrenal venous sampling (AVS) is the most reliable test to distinguish APA from idiopathic hyperaldosteronism (IHA). This study investigated the diagnostic accuracy of AVS for the preoperative diagnosis in primary aldosteronism (PA). Materials & Methods: Forty-six patients with hypertension and hypokalemia were referred with the diagnosis of PA based on elevations of aldosterone and suppression of PRA. All patients underwent CT (1.252mm sections). AVS for measurement of aldosterone and cortisol levels were performed before and after stimulation with ACTH. Three patients with iodine allergy were studied using Gd or CO2. Results: Catheterization of the right adrenal vein was unsuccessful in two patients; catheterization of the left adrenal vein was successful in all. Twenty-nine patients were diagnosed with APA by AVS, confirmed by adrenalectomy. Four of 29 were not detected by CT. In 29 of 29 patients with APA, unilateral elevation of aldosterone was noted after stimulation with ACTH. Conclusion: Adrenal CT imaging is not a satisfactory method for the differential diagnosis of PA. Bilateral simultaneous AVS was safe and essential to establish the correct diagnosis of PA.

P187

P189

Factors affecting therapeutic efficacy in balloon-occluded transvenous obliteration of gastric varices with gastrorenal shunts W. Hwang1, B. Kim1, H. Kim1, I. Hwang1, D. Kim2, K. Paek2; 1Radiology, Hallym University Hospital, Chunchon, Republic of Korea, 2 Internal Medicine, Hallym University Hospital, Chunchon, Republic of Korea.

Primary stenting of central venous stenosis in hemodialysis patients A. Rampoldi, M. Solcia, M. Intotero, G. Carrafiello, M. C. Castoldi, R. Corso; Vascular and Interventional Radiology, Niguarda Hospital, Milan, Italy.

Purpose: To know the factors affecting the therapeutic efficacy of balloon-occluded transvenous obliteration (BRTO) in patients with gastric varices and gastrorenal shunts. Materials & Methods: From March 2000 to January 2003 we treated 14 cases with BRTO, puncturing the right femoral vein and introducing the balloon catheter into the gastrorenal shunt. After balloon occlusion of the shunt we introduced 20% glucose and 20-60ml of the sclerosing agent into gastric varices. After 1-3 hours we deflated the balloon. The sclerosing agent was mixed with two methods: ethanolamine oleate:contrast media=1:1 and ethanolamine oleate:lipiodol =1:4. We analyzed success rates according to Hirota grade of gastric varices, diameter of gastrorenal shunts, amount and mixing method of ethanolamine oleate, and ballooning time. Results: Gastric varices were completely eradicated in 43%, decreased in 43%, and unchanged in 16%. The success rate was proportional to the amount of ethanolamine oleate, inversely related to Hirota grade (IV) and the diameter of gastrorenal shunt, and not related to the ballooning time. In all three cases of 1:4 solution, gastric varices disappeared. Conclusion: We suggested increase in the quantity of the sclerosing agent and introduction of 1:4 solutions in BRTO for more effective eradication of gastric varices.

Purpose: To report our experience with primary stenting in hemodialysis patients with central venous stenosis. Materials & Methods: Since 1996, 51 hemodialysis patients with symptomatic central vein obstructions involving arm swelling or dialysis problems were referred. Forty (78%) had a Brescia-Cimino hemodialysis fistula while 11 (22%) had a graft. In the follow-up chiefly clinical indicators were examined for detection of hemodialysis access dysfunction. Results: The patients were treated using 58 Wallstents, 15 Smart, three SelfX, four Palmaz, and two “Z” stents with diameter ranging from 10 to 14 mm. The stents were deployed primarily (82%) or following initial predilatation of the obstruction (18%). Technical success was 100% without complications. Swelling of the upper limbs disappeared within 48 hours after initial treatment in all patients. In seventeen patients (33%) we obtained primary patency at 1 year. In the remaining 34 patients (67%) repeated procedures were required during follow-up because of intra-stent stenosis or thrombosis. Conclusion: In our opinion treatment of central venous stenoses using primary stenting is effective, with a high rate of immediate technical success. Howewer in some cases multiple reinterventions, including angioplasty, stenting or both, are necessary for maintaining hemodynamically adequate patency.

P188

P190

Long-term results of deep venous thrombosis in the lower extremity B. Roh; Radiology, Wonkwang University Hospital, Iksan City, Republic of Korea.

Percutaneous endovenous radiofrequency closure of a refluxing greater saphenous vein S. Cho1, Y. Kim2, E. Chung3, S. Kim4, J. Cho5, S. Paik6, Y. Yoon4, Y. Kim1; 1Radiology, College of Medicine, Korea University, Seoul, Republic of Korea, 2Dermatology, S&U Clinic, Seoul, Republic of Korea, 3Dermatology, Eulji Hospital, Eulji School of Medicine, Seoul, Republic of Korea, 4Radiology, Eulji Hospital, Eulji School of Medicine, Seoul, Republic of Korea, 5Radiology, Samsung Medical Center, Sungkyunkwan University School of Medicine, Seoul, Republic of Korea, 6Radiology, Bucheon Hospital, Soonchunhyang University, Seoul, Republic of Korea.

Purpose: To evaluate the long-term follow-up results of deep venous thrombosis in the lower extremity after treatment with interventional procedures Materials & Methods: Fifty-six patients (27 men and 29 women, mean age=55.6 years) with deep venous thrombosis in lower limb were treated with interventional procedures and anticoagulation. There were acute 39 DVT and 17 chronic DVT. We monitored the long-term followup results of clinical response and patent deep veins by US, CT and catheter venography for five years Results: The clinical response showed improvement in 39(78%) and recurrence or no improvement in 11(22%) among the 50 surviving patients (total 56, dead 6). In acute DVT, the venous lumen was patent in 30 (88%) and occluded in 4 (12%). In chronic DVT, nine (56%) showed improvement and seven (44%) showed no improvement. There were three re-thromboses during the two weeks after treatment, retreated with catheter-directed thrombolysis. During follow-up, there were four occlusions of iliac stents, with aggravated clinical symptoms, and three occlusions of femoral veins, without aggravated symptoms Conclusion: Long-term results of deep vein thrombosis in lower extremity following treatment with interventional procedures show a high patency rate of the venous lumen and a good clinical response

Purpose: To evaluate the efficacy of percutaneous endovenous radiofrequency (RF) closure of the incompetent greater saphenous vein (GSV). Materials & Methods: Forty limbs with saphenofemoral junction (SFJ)reflux on Doppler-US were treated percutaneously with an endovenous RF closure catheter on an out-patient basis. Additional ambulatory phlebectomy or sclerotherapy of varicose veins were performed. Follow-ups included Doppler-US and clinical examinations (620 months). Results: Twenty-two of the 30 limbs followed for one year or longer (mean: 15 months) and ten limbs followed for six months or longer (mean: seven months) underwent Doppler-US. SFJ reflux-free rate was 97% (31/32). Three recanalizations occurred (occlusion rate: 29/32, 91%); one total recanalization of the GSV without SFJ reflux and two segmental recanalizations with one minimal SFJ reflux were recorded. One minimal reflux occurred after 19 months and was confined to the recanalized proximal segment with completely obliterated mid, distal segments and no recurrent varicosities. One limb showed a recurrent varicosity at six-month follow-up without SFJ reflux, probably due to incompetent perforating vein. Two paresthesias present after six months disappeared after one year. Conclusion: Percutaneous endovenous RF closure of incompetent GSV is an effective and less invasive technique to eliminate SFJ reflux and occlude the GSV.

P191

P193

Chest port implantation in neutropenic pediatric patients. Is it a safe technique? A preliminary report B. Çil, S. Geyik, D. Akinci, A. Besim; Radiology, Hacettepe University Faculty of Medicine, Ankara, Turkey.

In-vitro evaluation of a self-sealing hemostatic flow-valve G. Bartal1, A. Palestrant2, E. Haytman3, J. M. Gomori4; 1Diagnostic Imaging and Interventional Radiology, Hillel Yaffe Medical Center, Hadera, Israel, 2Interventional Radiology, Vascular & Interventional Physicians, Phoenix, AZ, United States, 3Elcam Medical, Kibbutz Bar'am, D.N. Merom Hagalil, Israel, 4MRI Unit, Hadassah Medical Center, Jerusalem, Israel.

Purpose: To evaluate safety and clinical outcomes of implanting chest ports in neutropenic pediatric patients with malignant diseases. Materials & Methods: Between August and December 2002, 15 chest ports were implanted in seven girls and eight boys (age range: six months to 12 years; mean, 6.7) who needed a long-term central venous access for chemotheraphy. All the patients had neutropenia secondary to chemotheraphy. Venipunctures were all performed under sonographic guidance and the right internal jugular vein was used for the access in all cases. Results: The technical success was 100% without any complication during insertion. Infections or delayed wound healings were not recorded. Follow-up was 1263 catheter-days, with a mean of 84.2 catheter-days. Conclusion: The radiological placement of chest ports in neutropenic pediatric patients is a safe technique and can be an alternative to surgical ports placement.

P192 Benign and malignant central venous stenoses: ten years' experience F. Perona1, M. Cugnasca2, A. Morbidelli2, A. De Troia2, A. Barile3, A. Zerbi1; 1Interventional Radiology, Galeazzi Institute, Milan, Italy, 2Vascular Surgery, Galeazzi Institute, Milan, Italy, 3Radiology, ASL2, Savona, Italy. Purpose: To assess the effectiveness of endovascular treatment of benign and malignant venous stenoses. Materials & Methods: Fifty-four patients (31 male, 23 female, aged 47-74 years ) underwent therapy for benign stenoses. Forty-one patients ( 29 male, 12 female, 42-69 years old ) were treated for malignant stenoses. All cases were treated between December 1992 and December 2002. Wallstent (Boston Scientific, USA),10-14 mm calibre, was used for subclavian and innominate veins and Gianturco-Rosch “ Z” stent (Cook, USA) for SVC stenoses. Results: In benign conditions technical success was achieved in 49/54 patients (90.7%). Two of four patients with early obstruction were successfully recanalized; in the other two cases all therapies were unsuccessful. Late obstructions were observed in 11 patients and stent patency was achieved in six cases. In malignant conditions technical success was achieved in 36/41 patients (87.8%); two early occlusions occurred, and in one case all therapies were unsuccessful. Three patients died due to early complications. Conclusion: These positive results, the contribution to prognosis, and the reduced invasivity demonstrate that stent placement is a safe and feasible approach. This technique is at the present moment the best therapeutic option.

Purpose:Evaluation of a new self-sealing hemostatic flow-valve that allows bloodless guide-wire handling and catheter flushing without opening or closing. Material/Methods:An angiographic self-sealing hemostatic flow-valve (Elcam Medical, Israel) was evaluated in-vitro. It was attached to standard 5F angiographic catheter, which was connected to the flow-system that generates continuous backpressure. The valve has proximal large-lumen housing and actuator that controls hemostatic slit septum valve with gradual lumen diameter adjustment. The distal housing is a standard Luer-lock that fits all catheters. Common guide-wires (0.014” to 0.038”) and 3F microcatheters were introduced through the valve against 250mm Hg backpressure. We assessed ease of insertion, handling, friction and leakage. A Luer-lock 10ml syringe was used for evaluating ease of aspiration, injection of saline and leakage. Results:Luer-lock syringe aspiration and injection of saline showed no resistance or leakage during connection or disconnection. All evaluated guide-wires and microcatheters were easily inserted via the 5F catheter under continuous 250mm Hg backpressure and were handled without excessive friction or any leakage of saline. Conclusion:The self-sealing hemostatic flow-valve allows convenient coaxial guide-wire and microcatheter placement without leakage of saline. Aspiration or injection of saline through the valve is easy without leakage or opening – closing requirement. Clinical evaluation is underway.

This index lists the names of authors of all submitted abstracts. The presentation number is bolded for presenting authors. (The letter "P" before the presentation number refers to the Poster Section) Abbasoglu, O. Ablett, M. Acunas, B. Adam, A. Adiseshiah, M. Aerts, R. Ahlström, H. Air-online user group, A. Aisen, A. M. Akaileh, a. Akgül, E. Akinci, D.

Akhan, O. Akiba, H. Akpek, S. Akpinar, E. Aksungur, E. Alava Echevarria, N. Alberico, R. Alborino, S. Alexopoulou, E. Alfke, H. Alfonso, R. Allison, D. Almeida, P. A. Aly, S. Andrén, B. Andreou, A. Andresen, R. Andrews, J. C. Andrews, R. T. Androulakakis, Z. Angele, P. Angeletti, C. A. Angeli, E. Annerstedt, M. Antico, E. Anzböck, W. Arai, K. Arciprete, F. Ardashev, A. Argentos, S. I. Argyriou, P. Arjonilla, M. A. Arocena, C. Arslan, H. Aschauer, M.

P79 43.1.6 17.1 6.3.2, 41.1, P130 6.2.5, P11 P151 35.5.5, P176 6.6.1 43.5.7 15.4.1 P88 6.4.3, 15.5.7, P79, P95, P185, P191 6.4.3, 15.5.7, 38.1, P79, P95 P19 P128 P79 P88 P1, P152, P153 P175 P158 P21, P134, P140, P169 43.5.9 P38, P114 P141 P130 6.4.9 6.2.5 35.5.5, P176 15.4.5 35.3.5, P43 6.5.3 15.6.2 6.2.2, 15.3.3 P121 6.6.3 15.2.8, 35.5.1, 43.4.7 35.5.4 P158 P89 P144, P146 43.1.3 P167 15.1.4, 31.1 P140 P47, P101 P101 P59, P66 P115

Aslan, &. Assaf, Y. Ataç, K. Atar, E. Athanasiou, G. Aupérin, A. Aurell, M. Aytekin, C. Babunashvili, A. M. Bachar, G. N. Bachleda, P. Baek, J. Bafile, G. Bagci, S. Bahcebasi, A. Baik, H. Balavadze, M. Baldi, I. Bale, R. J. Balkancý, F. Balzano, S. Bañares, R. Barbalias, G. Barbey, F. Bargellini, I. Barile, A. Bartal, G. Bartolozzi, C. Baruffaldi Preis, F. W. Basaran, C. Basekim, C. Basile, A. Batakis, N. Baus, S. Bauso Toselli, P. L. Bayindir, O. Beck, A. N. Becker, C. Becker, T. Bednarek, W. Behrens, P. Belenky, A. Bellini, F. Bellomi, M. Benali, K. Beraia, M. Bérczi, V. Berden, P. Beregi, J. Bergqvist, D. Bernat, P.

P91 43.1.4, 43.1.5 P15 15.5.5 P121 35.6.4 35.5.4 P48 35.3.8 15.5.5 43.2.5 P76 P34, P183 P91 43.6.1 P56 P29 6.2.6 P20 P155, P185 35.6.2 P1, P153 P161 P165 15.2.7, 43.5.3, P40, P179 6.4.2, P192 15.5.5, 43.6.1, P193 6.6.3, 43.5.3, P40, P179 P129 P79 P178 P49 P22, P35, P99, P104, P105 6.6.6, 35.1.7 6.1.7 35.5.2 P98 6.3.1, 35.6.8 6.3.5 35.2.8, P64 35.3.5 15.5.5, 43.6.1 35.3.7 6.3.6 P113 P159 6.1.8 15.2.1, 35.5.7 35.3.2 35.5.5, P176 P64

Berney, T. Bertoni, H. G. Besim, A. Bessoud, B. Betschart, T. Bezzi, M. Bhan, A. Bilgen, C. Bingöl, H. Binkert, C. A. Birkfellner, W. Bitsch, R. Bitzer, M. Bjarnason, H. Blanshard, K. S. Blecher, B. Boccardi, E. Bock, M. Boekkooi, F. P. Boige, V. Bolte, H. Bongarzone, R. Bonneau, M. Bonomo, G. Bosch, O. Bosiers, M. Boström-Ardin, A. Boutsianis, E. Bovelander, J. Boyer, L. Boyvat, F. Bracco, D. Brakowiecki, F. Brambs, H. Brehme, U. Brem, E. Brennan, P. Brodmann, M. Brody, L. Broe, D. Brookes, J. A. Brothers, T. Brountzos, E. N.

Brown, K. T. Bruijninckx, B. Bruit, J. Brzek, V. Buchmann, P. Bulakbasi, N. Burdi, N.

35.6.8 6.1.7, 35.1.3, 43.4.9, P92 P155, P191 P46, P164 6.4.1 6.1.1, 15.1.8, 15.2.6, 43.3.4 P18 15.6.6 P15 6.1.9 P24 15.6.1, P54 15.3.1 6.5.3 P163 15.1.6 15.4.2 P58 35.2.4 43.4.6 43.1.8 P26 43.3.6 6.3.6, 12.1 P101 6.1.6, 35.3.1 35.5.5, P176 35.4.1, 43.2.7 6.5.5 16.3, 35.6.3 P48 35.1.3 15.1.7 P171 35.3.4 43.6.1 6.2.9, 15.4.3 P41 6.3.8, P87 15.5.3 6.2.5, P11 15.1.2 6.1.9, 7.2, 15.1.4, 43.6.9, P94, P134, P140, P169 6.3.8, P87 43.2.4 27.1 35.1.2 6.4.1 P37, P83 43.1.7

Buriankova, E. Burton, K. Bushi, D. Buzzi, M. M. Bylov, K. V. Byun, J. Cabrera, T. Cademartiri, F. Cahill, A. M. Cambier, P. Camuñez Alonso, F. Candelari, R. Canevet, G. P. Cano, C. Canziani, M.

Cappelli, A. Caramella, D. Carnini, M.

Caronno, R.

Carrafiello, G.

Carriero, A. Carrilho, F. Carvalheira, M. Caselli, A. Casey, S. Cassina, D. Castelli, P.

Casteneda, F. Castoldi, M. C. Castriota, F. Cavallini, M. Cejna, M. Çekirge, S. Centini, G. Cerna, M. Cesareo, V. Cha, I. Chabbert, V. M. Chagnon, S. Chan, F. L. Chapot, R. Chatziioannou, A.

43.2.5, P102 P42 43.1.5 P92 35.3.8 P118 P70 35.4.6 6.6.2, 6.6.8, 35.6.7 P72 P1, P152, P153 P158 1.1 P109 6.2.8, 15.1.5, 43.2.1, P12, P132, P135 6.2.6 P179 6.2.8, 15.1.5, 43.2.1, P12, P132, P135 6.2.8, 15.1.5, 43.2.1, P12, P132, P135 15.1.5, 43.2.1, P12, P44, P45, P132, P135, P145, P189 35.3.7, P173, P180 6.4.9 6.4.9 6.2.7 36.2 6.1.7, 35.1.3, 43.4.9, P92 6.2.8, 15.1.5, 43.2.1, P12, P44, P132, P135 15.3.2 P45, P145, P189 6.1.2 P129 35.3.6, 43.4.8, 43.5.5 P37, P155 P158 43.2.5, P102 35.1.3 P62 1.1 6.6.9 P23, P154 6.6.4, 15.6.7 43.6.7

Chavan, A. Chiambaretta, F. Chien, P. Cho, J. Cho, J. H. Cho, O. Cho, S. Choi, B. Choi, W. Chovanec, V. Christoph, P. Chua, G. C. Chugtai, A. Chun, H. Chung, E. Chung, G. Chung, H. Cicarelli, O. Cicorelli, A. Çil, B. E.

Cina, A. Cioni, R. Claussen, C. C. Cleveland, T. J. Coleman, J. Colville, J. Corless, C. Corso, R. Costantini, C. Costello, J. Cotroneo, A. R. Covey, A. M. Cowling, M. Cremonesi, A. Crespi, G. Crocetti, L. Cugnasca, M. Cwikiel, W. Czermak, B. V.

D'Agostino, H. D'Amario, V. d'Archambeau, O. C. D'Elia, M. Dave, H. Davey, P. Dávila, J. de Baere, T.

6.6.6, 35.1.7 35.6.3 P23 P190 43.3.1 P51, P76 P62, P190 P118 15.6.3 10.2, 43.5.8 43.6.5, 43.6.6, P55 P139 43.5.2 P118 P190 P3, P116 P62 15.2.3 P40 6.4.3, 15.5.7, 26.2, P155, P185, P191 6.2.1, P6 6.6.3, 43.5.3, P40 15.3.1, 35.3.4 6.1.4, 6.1.5, 43.1.1, 43.1.6 6.2.9 15.5.3 43.6.9 P45, P145, P189 35.4.5 P164 6.2.1, 35.3.7, P6, P180 6.3.8, P87 P4, P9, P136 6.1.2 43.4.7 6.6.3 6.4.2, P129, P192 15.2.4 6.4.1, 35.1.4, 35.1.5, 35.1.8, P13, P20 17.3 P34 35.2.6, 35.3.3 P34, P183 35.4.1 43.2.3, 43.2.8, P17 P109 6.3.3, 35.6.4, 43.4.6

De Blasis, G. De Cobelli, F. de Donato, G. de España, F. De Gaetano, A. De Gregorio, M. A.

de la Iglesia, P. De Miguel, S. De Santibañez, E. De Santis, A. De Schepper, A. M. De Troia, A. Deakin, M. Dechow, C. Dehmeshki, J. Del Maschio, A. Delavelle, J. Della Vigna, P. Dellafiore, L. Delle, M. Deloose, K. Demir, A. Demir, S. Deniz, Ö. Denys, A. Derumeaux, H. Desgrange, P. Despond, L. Destro, G. Deutschmann, H. Di Giambattista, G. Di Palo, M. Di Renzi, P. Di Stasi, C. Dias, N. Diaz-Daza, O. Didier, D. Dietrich, T. Dietz, C. A. Dimakakos, P. Dimopoulos, I. Dincler, S. Doenz, F. Dominik, J. Donatelli, M. Doros, A. Doundoua, D. P. Drelich-Zbroja, A. Dromain, C. Drozdov, S. A. Du, J. Düber, C. Ducreux, M. Duda, S. H.

P34, P183 35.5.1 6.2.6 P60, P125 P6 6.5.2, 17.3, P38, P114, P141, P60, P125 P109 P92 15.4.2 35.3.3 6.4.2, P192 P9 15.6.1, P54 P164 15.2.8, 35.5.1, 43.4.7 6.4.7 6.3.6 P138 35.5.4 6.1.6, 35.3.1 6.6.9 P127 P178 P46 35.6.3 43.5.6 P113 P25 P115 P96 43.1.7 P96 6.2.1, P6 28.3 15.4.4 35.6.8 15.3.1 35.6.5 43.6.7 P107, P111 6.4.1 P46 35.1.2 43.1.7 P82 35.3.8 6.1.3 6.3.3, 35.6.4, 43.4.6 35.3.8 P142 15.1.3, 43.4.1 43.4.6 15.3.1, 15.3.2, 35.3.2, 35.3.4

Duke, D. M. Dunitz, J. Duru, N. K. Duthil, P. Düx, M. Ebner, F. Echenagusía Belda, A. Echenagusía Boyra, M. Eklöf, H. El Hajjam, M. Elfassy, E. Elias, D. Engin, D. Eracleous, E. A. Erikson, U. Eriksson, L. Esteban-Hernandez, E. Etlik, O. Evans, D. Fabiano, S. Fanelli, F. Farkas, J. Farrell, M. Fava, M. Fay, D. Fazel, A. Fehérvári, I. Fellström, B. Feng, L. Ferragud, S. Ferreira, M. Ferrer-Puchol, M. D. Ferreri, G. FIBROID Registry Steering Committee Fidiarakis, A. Filauri, P. Filippidis, A. Filipponi, F. Fink, C. Fiorina, P. Firat, A. Fischbach, R. Fisher, R. G. Fitz, C. R. Fleischmann, D. Florio, F. P. Foley, D. W. Fontana, F. Ford, S. M. Fornari, M. Fotheringham, T. Foudali, A. Fraedrich, G.

6.2.9, 15.5.3 35.6.5 P83 P113 15.6.1, P54 P115 P1, P152, P153 P1, P152, P153 35.5.5, P176 6.6.9 43.4.6 6.3.3, 43.4.6 P175 3.1 P119 P2 P69 P66 6.1.4 6.2.7, 6.5.1, 6.6.5, 35.5.8 6.1.1, 15.1.8, 15.2.6, 43.3.4 15.4.7 6.4.8, 15.4.3 6.1.7, 35.1.3 15.5.4, 43.2.3, 43.2.8, P17 35.2.2 P82 P2 6.6.4 P69 6.4.9 P69 P96 35.2.1 P22, P99 P34, P183 P77 43.5.3 P58 15.2.8 P48 P171 P177 6.6.2, 6.6.8, 35.6.7 43.5.5 35.6.2 43.5.1 P44, P135 35.2.7, 43.3.2 15.4.2 15.5.3 6.6.9 35.1.4, 35.1.5, 35.1.8, P20

Franchini, C. François, O. G. Frauenfelder, T. French, M. E. Frericks, B. B. Frigo, E. Fugazzola, C. Fukuda, T. Funovic, M. Furlan, F. Furui, S. Furuichi, K. Gabrielli, G. Gachechiladze, D. Gaines, P. A. Galani, P. Galanski, M. Galazka, Z. Galmbacher, R. Gamboa, P. Gandhi, R. T. Gandini, R.

Gangi, A. García García, L. Garcia Monaco, R. D. Garcier, J. Gasparini, G. Gaux, J. Gelebek, V. Geleijns, J. Georgieva, B. Gerbaud, L. Geterud, K. Getrajdman, G. I. Geubel, A. P. Geyik, S. Giachetti, A. Giannakenas, C. Gigot, J. F. Gil, S. Gil-Romero, J. Gill-Leertouwer, T. C. Gillams, A. R. Gillet, M. Gimeno, M. J.

Giordano, F. Glass, C. Glynos, M. Gobert, P.

6.6.3 35.3.3 35.4.1, 43.2.6, 43.2.7 P136 6.3.5 15.5.2 P44, P132 P146 35.3.6 P166 P97 P90 35.4.5, P26 P159 6.1.4, 6.1.5, 43.1.1, 43.1.6 P35 6.3.5, 6.6.6, 35.1.7 P137 P58 6.5.2 6.3.8, P87 6.2.7, 6.5.1, 6.6.5, 15.5.1, 35.4.3, 35.4.4, 35.5.8, 43.1.3 6.3.2 P85, P109 6.1.7, 35.1.3, 43.4.9, P92 16.3, 35.6.3 P25 P70 P95 43.5.4 15.2.5 35.6.3 P120 6.3.8, P87 15.2.2 6.4.3, P95, P185, P191 P179 P7 35.5.6 P60, P125 P33 35.5.3 6.3.7 P46 6.5.2, 17.3, P38, P114, P141 35.6.2 43.1.8 43.6.7 P113

Godoy-Gepert, A. Goffette, P. P. Goldberg, N. Golzarian, J. Gomori, J. M. Gonzalez-Añon, M. Gonzalez-Guirado, A. Goo, D. Gor, D. M. Görgülü, S. Görich, J. Gorman, G. Görög, D. Gortenuti, G. Gougoulakis, A. Gould, D. Gourtsoyiannis, N. Grad, Y. Graham, J. Grenacher, L. Griffiths, P. D. Grossman, Z. Grosso, M. Gruber, H. Gruenenfelder, J. Grunze, M. Guijarro, J. Guiney, M. J. Gujrathi, S. Gulati, G. Gümüs, B. Gumus, T. Güney, B. Günther, R. W. Gupta, A. K. Guzmán de Villoria, J. Guzzardi, G. Haage, P. Habahbeh, m. Habboub, H. K. Habib, S. B. Hackenbroch, M. Hadzidjokic, J. Hagl, C. Haglund, U. Hahn, T. Haliloglu, M. Hallscheidt, P. Hamady, M. Hamuro, M.

15.2.2 15.2.2, 15.2.3, 35.5.6 15.5.5 35.2.5, 35.2.6, 43.3.8 43.6.1, P193 P69 P101 P74 15.4.7 P83 P171 6.4.8 P82 P25, P166 15.1.4 43.1.6 6.2.2, 15.3.3 43.1.4, 43.1.5 43.1.1 43.6.5, 43.6.6, P55 6.1.4, 6.1.5, 43.1.1 P175 15.2.7 P13 35.4.1 43.6.5, 43.6.6, P55 P33 6.2.5, P11 15.4.7 35.4.2, P18, P133 15.5.7 P128 P127 6.5.5, 29.1, 35.6.6 P175 P1, P152, P153 35.3.7, P180 6.5.5 15.4.1 15.4.1, 43.1.9 P163 P27, P30, P31, P57 P61 35.1.7 P2 15.6.1 15.5.7 P58 6.3.2 P143, P147, P148, P150

Han, S. Han, Y. Hannegan, C. Hareyama, M. Harman, M. Harris, D. Hasan, T. Hashimoto, M. Hashimoto, T. Haskal, Z. J. Haslam, P. J. Hattori, T. Hatzidakis, A. A. Hausegger, K. A. Haverich, A. Havlik, R. Hayoz, D. Haytman, E. He, X. Heasler, E. Hechelhammer, L. Heindel, W. Heller, M. Hellstrom, M. Helmberger, T. K. Hendriks, J. Henn, M. Henroteaux, D. Herber, S. Herlitz, H. Herrera, M.

Herrmann, K. Herzog, P. Heyer, C. M. Higashi, M. Higashihara, H. Higgins, C. B. Hillmeier, J. Hippeläinen, M. Hirai, N. Hirota, S. Hobbs, G. P. Hofer, U. Hong, S. Houdart, E. Hourani, K. Hruby, W. Hulek, P. Hülsmeier, L. Hunter, D. W. Husso-Saastamoinen, M. Hüttl, K.

P56 P3, P116 6.5.6, 15.1.2, 43.3.3 P19 P59, P66 43.1.4, 43.1.5 15.5.4 P52, P143 P157 P75, P78, P80 15.5.4, 43.5.2 P8, P10, P84 3.2, 6.2.2, 6.3.2, 15.3.3 15.2.5 35.1.7 P102 P165 P193 15.6.3 P165 15.2.9, 43.2.9, P182 P171 43.1.8 35.5.4 35.4.7 35.3.3 43.6.5, 43.6.6, P55 P72 6.3.4, 15.1.3, 15.1.6 35.5.4 6.5.2, 17.3, P38, P114, P141 35.4.7 35.4.7 P36 P126 P97 29.3 6.4.5 43.3.5 P126 P144, P146 15.4.4 P27, P30 P62 6.6.4 43.1.9 15.5.2, P89 10.2 35.6.6 35.6.5 43.3.5 6.1.8

Hwang, I. Hwang, W. Hyodoh, H. Hyodoh, K. Iani, C. Ibatullin, M. M. Ihaya, T. Ikonomopoulos, N. Ilgit, E. T. Ilic, M. S. Imai, Y. Inal, M. Inoue, K. Inoue, Y. Intotero, M. Ioannidis, I. E. Ioannou, C. Irurzun, J. Ishibashi, T. Ishii, S. Ishikawa, H. Islak, C. Isota, M. Ivancev, K. Ivanovic, A. Jacob, A. L. Jacob, D. Jahnke, T. James, R. L. Janczarek, M. Januszewicz, M. Jargiello, T. Jaschke, W. R.

Jensen, G. Jin, G. Jing, D. Jing, Z. P. Joffre, F. G. Johnson, M. C. Johnston, C. J. Jordaen, v. Juhász, G. Juneja, R. Jung, T. Junger, S. Kýzýlkaya, E. Kabaroglu, C. Kacar, K. Kagadis, G. C. Kagel, T. Kako, Y. Kalinowski, M.

P187 P187 P19 P19 43.1.3 P168 P93, P103 P169 35.6.1, P128 P160 P157 P88 P112 P112, P147, P148, P150 P45, P145, P189 15.4.5 15.3.3 P60, P125 P50, P186 P39 P39 11.2 P150 28.3 P5 P24 35.2.2 2.3, 2.4, 43.1.8 P136 35.2.8, P64, P68 P137, P174 6.1.3, 15.1.7 35.1.4, 35.1.5, 35.1.8, P13, P20, P162, P179 35.5.4 P3, P116 15.3.4, P142 35.1.6 1.1, P113 6.5.3 35.2.7, 43.3.2 35.4.6 P14, P32 35.4.2, P133 6.4.1, 35.1.4, 35.1.5, 35.1.8 P36 P178 35.5.2 P61 P161 P36 P144, P146 43.5.9

Kalioras, V. Kalogeropoulou, C. P. Kamaci, M. Kamei, K. Kaminou, T.

Kanai, T. Kandarpa, K. Kang, B. Kang, I. Kanitsar, A. Kapoor, B. S. Karabulut, N. Karageorgos, A. Karagiannakidis, E. Karamesini, M. Karçaaltincaba, M. Karck, M. Karlsson, B. Karnabatidis, D.

Karnel, F. Kartashov, D. S. Kasperk, C. Katagiri, Y. Katsamouris, A. Katsanos, K. Katsenis, K. Katsouras, C. Katzler, I. Kauczor, H. U. Kauffmann, G. W. Kaufman, J. A. Kavteladze, Z. A. Kawamata, I. Kaye, R. D. Keeling, A. N. Kelekis, A. D. Kelekis, D. A.

Kelekis, N. L. Keller, F. S. Kesik, J. Kessel, D. Kettenbach, J. Khodjibekova, M. M. Kichikawa, K. Kiliç, M. Kim, B. Kim, C. Kim, D.

P22, P35, P99, P104, P105 P7, P107, P161 P59 P67 P52, P143, P147, P148, P150 P157 36.1 15.6.4 43.6.8 43.5.5 35.6.5 6.4.3 P108 15.3.3 P7 6.4.3, 43.5.1 35.1.7 P2 6.2.4, 15.3.5, P108, P111, P117, P161 15.2.5 35.3.8 6.4.5 P19 6.2.2, 15.3.3 6.2.4, 15.3.5 43.6.7 P53 43.5.5 P58 6.4.5, 15.6.1, 35.2.3 6.1.9, 15.6.2, 43.5.7, 43.6.9 35.3.8 P157 6.6.2, 6.6.8, 35.6.7 P73 6.4.6, 6.4.7 7.2, 15.1.4, 31.1, P21, P94, P122, P134, P140, P169 15.1.4, P169 15.6.2, 43.6.9 15.1.7 43.1.6 35.3.6, 43.4.8 P28 P90 P127 P187 P81 P187

Kim, H. Kim, J. Kim, K. Kim, M. D. Kim, S. Kim, T. Kim, Y. Kimura, K. Kimura, M. Kind, M. Kinney, T. B. Kirazli, T. Kirbas, I. Kirchhoff, T. D. Kishi, K. Kitayama, T. Klancar, J. Klimov, V. Klingenstierna, H. Kljucevšek, T. Klos, G. Klose, K. J. Knippenberg, B. Ko, G. Kobayashi, K. Kobeiter, H. Kocaoglu, M. Köcher, M. Koechl, A. Koh, B. Koizumi, J. Kojecky, Z. Kojima, S. Konda, D. Kondoh, Z. Kono, K. Konuk, O. Kopp, A. F. Kordonouri, T. Köroglu, M. Korteweg, T. Kos, X. Kosar, S. Kostas, T. Kotarski, J. Kothari, S. S. Kotsarini, C. Kovác, J. Koznar, B. Krajícková, D. Krajina, A.

P63, P187 15.6.5, P56, P86, P181 P62 43.3.1 P81, P190 15.6.3, 15.6.5 P51, P62, P76, P190 P126 P39, P184 35.6.4 6.6.7, 15.1.1 15.6.6 P48 6.3.5 P39 P147, P148, P150 15.2.1, 35.5.7 P167 35.5.4 15.2.1, 35.5.7 15.1.6 43.5.9 43.2.4 P156 P8, P10, P84, P144 43.5.6 15.5.6, P83 43.2.5, P102 43.5.5 P51, P76 P157 43.2.5 P157 15.5.1, 35.5.8, 43.1.3 P67 P112 35.6.1 20.3 6.2.4, 15.3.5, P121 P79 43.5.4 35.2.6 P128 15.3.3 35.2.8, P64 35.4.2, P18, P133 6.2.4 P131 P131 P71, P131 10.2, 35.1.2, 43.5.8

Kralj, I. Krämer, S. C. Kraniotis, P. Krankenberg, H. Kuhelj, D. Kuiper, J. W. Kumano, R. Kumins, N. H. Kunst, T. Kuoch, V. Kuribayashi, S. Kurrer, M. Kurz, P. Kwak, B. Lachat, M. Lacombe, P. Ladopoulos, C. Laffan, E. Laganà, D.

Lagrange, C. Lammer, J. Lampmann, L. E. Lanciego, C. Landwehr, P. Lapeyre, M. Lastiri, J. Lastrovicková, J. Laterre, P. F. Latini, V. Laurent, A.

Laureys, M. Lauriola, W. Lázár, I. Le Dref, O. Le Hong, C. Lee, C. Lee, D. Y.

Lee, E. Lee, K. Lee, M. Lee, M. J. Lee, S. Lee, Y. Lees, W. R. Lehmann, R. Lencioni, R. Lepida, N.

43.3.7 P171 15.3.5, P117 6.2.3, 15.3.6, P123 15.2.1, 35.5.7 43.5.4 P8, P10, P84 15.1.1 35.5.7 6.3.3, 35.6.4, 43.4.6 P67 P182 43.6.5, 43.6.6, P55 43.6.8, P56 43.2.9 6.6.9 43.6.7 35.2.7, 43.3.2 6.2.8, 15.1.5, 43.2.1, P12, P44, P132, P135 6.6.9 35.3.6, 43.4.8, 43.5.5, P49 35.2.4 P85, P109 35.3.2 43.5.6 43.4.9 P131 15.2.3, 35.5.6 6.2.7 6.6.4, 15.4.6, 15.6.7, 35.2.2, 43.3.6 35.2.5, 43.3.8 35.6.2 P14, P32 35.2.2 6.3.3, 35.6.4, 43.4.6 P63 35.1.1, 43.2.2, 43.3.1, P110, P156 P181 P110, P156 43.6.8, P56 15.5.3, P73 P81 P116 6.3.7 15.2.9 6.6.3, 27.2 P22, P35

Lerut, J. P. Letoublon, C. Li, Y. Liatsikos, E. Lim, J. Lim, N. Liso, A. Liu, J. Liu, L. Liu, Y. Llovlet, J. Loewe, C. Logan, M. P. Lohle, P. N. Lojik, M. Lomoschitz, F. Lomoschitz, K. Lonn, L. Lopez Lincuez, M. Losio, C. Lovblad, K. Lowe, A. Lu, W. Lubienski, A. Lubienski, K. Luciani, A. Luk, D. K. Lundgren, D. Lupoi, D. Lyon, S. Lyra, S. Macdonald, S. Madariaga, B. Maeda, H. Maffi, P. Maggi, M. Maglione, F. Magnusson, A. Magotteaux, P. Maier, C. Mainar, A. Maj, E. Majno, P. Maksimovic, R. Malagari, K. Malamou-Mitsi, V. Malcher, J. Maleux, G. Malina, M. Malinovskiy, M. N. Mallouhi, A.

15.2.2, 15.2.3 6.3.3 43.4.4, 43.4.5, P100 P161 15.6.3 P181 6.1.2 15.3.4 15.3.4 43.3.5 27.1 35.3.6, 43.4.8 6.4.8 35.2.4 10.2, 35.1.2, 43.5.8 43.4.8, 43.5.5, P49 P89 35.5.4, P120 43.4.9 43.4.7 6.4.6, 6.4.7 P113 43.4.3, 43.4.4, 43.4.5, P100 15.6.1, P54 15.6.1, P54 43.5.6 P23 P2 P96 P73 P22 6.1.4, 6.1.5, 43.1.1, 43.1.6 6.5.2 P144, P146 15.2.8 P158 15.2.7 P119 35.2.6 P36 17.3, P38, P114 P174 6.3.1 35.4.6 15.1.4, P94, P134, P140 P53 P89 P151 28.3 P168 35.1.4, 35.1.5, 35.1.8, P13, P162, P179

Manetti, R. Mangini, M. Manninen, H. Mansi-Montenegro, G. Manzi, M. Marangon, A. Marano, G. Marcelli, G. Marincek, B.

Markovic, Z. Maroulis, I. Martal, J. Martin, C. Martin, J. Mašulovic D. Mathias, K. Mathieu, N. Matsagas, M. Matsuhashi, T. Matsumoto, J. Matsuoka, T. Mátyás, L. May, B. Maynar, M. McCann, J. McEniff, N. McKusick, M. A. McNamara, T. Meckel, S. Medrano, J.

Meeder, H. Mehdizade, A. Meier, N. Members of the Study Group, E. Memis, A. Menerath, J. Menzinger, G. Merkesdal, S. Merland, J. Messmer, P. Michalis, L. Migazzo, E. Mikol, J. Milczarek, K. Miller, F. J. Minamiguchi, H. Mine, T. Misra, S. Mitra, K. Mitromaras, I.

6.1.2 P44 43.3.5 P106, P124 P106, P124 P106, P124 6.2.1, 35.3.7, P6, P173, P180 6.1.1, 15.1.8, 15.2.6, 43.3.4 6.5.4, 15.2.9, 35.4.1, 43.2.6, 43.2.7, 43.2.9, P182 P61 P111 43.3.6 6.2.9 6.4.6, 6.4.7 P61 43.1.2 35.2.5 6.2.2 P50, P186 P8, P10, P84 P112 P14, P32 35.6.5 36.2 35.2.7, 43.3.2 35.2.7, 43.3.2 6.5.3 15.3.2 P24 6.5.2, 17.3, P38, P114, P141 6.4.5 6.4.6 43.2.9 43.1.2 15.6.6, 35.5.2, P127 35.6.3 6.2.7 6.3.5 15.4.6 P24 6.2.2, P53 35.3.7 6.6.4 P137, P174 6.6.7 P184 P157 6.5.3 P9, P136 P21

Mitsos, A. Miura, K. Miyamoto, Y. Mofid, R. Molnár, A. Á. Monfardini, L. Montana, X. Montemezzi, S. Moon, Y. Mor, E. Morbidelli, A. Mordani, J. K. Morel, P. Moss, J. G. Mourikis, D. Mufeed, M. A. Müller-Hülsbeck, S. Murgo, S. Muro, I. Mutlu, H. Myasnik, B. N. Mylona, S. Nacher, A. Naef, R. Naik, S. A. Naito, H. Nakamura, K.

Nakamura, N. Nakanishi, J. Nakao, N. Nakata, M. Nakayama, K. Namur, J. Namyslowski, J. Nardella, M. Nava, M. Nawijn, A. Nazeri, A. Nelson, T. R. Nemes, B. Neri, E. Nessi, F. Neufang, A. Neuhauser, B. Nevens, F. Nice, C. Nicholson, T. Nickel, J. Nicolas, V. Nicolini, S. Niedercorn, J. Nikas, D.

15.4.5 P144, P146 P144 35.6.3 6.1.8 6.3.6 27.1 P25, P166 P181 15.5.5 6.4.2, P192 P4 35.6.8 43.1.6 43.6.7 43.1.9 2.3, 2.4, 43.1.8 35.2.5, 35.2.6 P157 P178 P28 P22, P35, P99, P104, P105 P69 6.4.1 15.4.4 P126 P52, P112, P143, P147, P148, P150 P19 P93, P103 P144, P146 15.6.2, 43.6.9, P148 P112 15.4.6, 35.2.2, 43.3.6 43.5.7 35.6.2 P129 43.2.4 43.3.8 6.6.7, 43.6.2, 43.6.4 6.1.8 P179 35.3.7, P180 15.1.3 35.1.8 P151 15.5.4 P130 P43 P36 P26 35.2.6 P53

Nikiforidis, G. Nikita, A. Nikolaou, V. Ninoi, T. Niokou, D. Nishida, N. Nishri, B. Nisini, A. Nizzola, F. Noeldge, G. Novak, D. Novotný, J. Nygren, A. Nyholm, D. Nyman, R. S. O'Connell, A. O'Dwyer, H. M. O'Sullivan, P. J. Obernosterer, A. Ogawa, K. Ogawa, T. Oh, J. Ohishi, H. Ohuchi, Y. Okazaki, M. Okuma, T. Oliva, V. Olmarker, A. Omran, H. Önal, B. Oner, K. Ono, S. Ookawa, T. Oran, I. Orfanidis, G. Orlacchio, A. Oronzo, P. Örs, F. Orsi, F. Oshoala, K. Otal, P. P. Otto, G. Ouriel, K. Owens, E. Oyama, Y. Oysul, K. Oz, O. Özmen, M. N.

Öztürk, A. Öztürk, E.

P108, P111, P117, P161 P21, P94, P122 P140 P143, P147, P148, P150 P53 P143, P147, P148, P150 43.1.4, 43.1.5 35.4.4 P138 6.4.5 P170 (see P194-P217) P16, P131 P119 P2 35.5.5, P2, P176 15.4.3 P73 6.4.8 P115 P67 P52, P103, P143 P149 P90 P52, P143 P97 P112 35.3.2 P120 P31 35.6.1, P128 15.6.6 P186 P93 15.6.6, 35.5.2, P127 15.4.5 P96 P173 15.5.6, P37 6.3.6, 12.1 6.1.2 1.1, P113 6.3.4 15.3.2 15.1.1 P112 P37 43.1.4, 43.1.5 6.4.3, 8.2, 15.5.7, P79, P95 6.4.3 P178

Özturk, S. Pabst, E. Pabusçu, Y. Paci, E. Padilla, M. Paek, K. Paik, S. Paiva, I. Palestrant, A. Pallotta, G. Palmero, J. Palussiere, J. Pamler, R. Panagiotou, I. E. Panagopoulos, A. Pangrazi, A. Papadopoulos, A. Papaioannou, G. Papapetrou, K. Papathanasiou, Z. Papathanassiou, M. Papet, I. Parildar, M. Parildar, Z. Park, H. Park, J. Park, S. Park, S. J. Park, W. K. Passalacqua, G. Passariello, R. Pastor, A. Patel, U. Pattynama, P. M. Pavcnik, D. Pech, M. Peeters, P. Pêgo, M. Pekkafalý, Z. Pelage, J. P. Penco, S. Pendenza, G. Pepino, D. Peralta, O. Perea, M. Peregrin, J. H. Perkmann, R. Perona, F. Petersen, B. T.

35.6.1 P41 P91 P158 P109 P187 P190 6.4.9 P193 43.4.9 P33 35.6.4 P171 6.1.9 P117 35.4.5 P108 7.2, P21, P94, P122, P134 15.4.5 P107 P122, P169 P70 35.5.2, P127 35.5.2 15.6.3 P181 P62, P63, P74 43.2.2 43.3.1 15.2.7, P34, P183 6.1.1, 15.1.8, 15.2.6, 43.3.4 P69 P139 35.4.6, 35.5.3, 43.5.4 15.6.2, 43.6.9 P98 6.1.6, 35.3.1 6.4.9 P178 6.6.9, 25.2, 35.2.2, 43.3.6 6.3.6 6.6.5, 15.5.1, 35.5.8 6.1.1, 15.1.8, 15.2.6, 43.3.4 43.4.9, P92 P109 6.5.7, 31.3, P16, P71 35.1.4, 35.1.5, P20 6.4.2, 15.4.2, P129, P192 6.1.9

Petitta, S. Petrow, P. Petruzzi, P. Petsas, T. Peynircioglu, B. Pfammatter, T. Pfeffer, J. G. Pichler, L. Pichlmaier, M. Pieraccini, M. Pieters, J. J. Pietura, R. Piffaretti, G.

Pilger, E. Pinelli, M. Pinerolo de Septis, C. Pinto Pabón, I. Pipitone, V. Pirenne, J. Pirk, J. Pitton, M. B. Piva, R. Piva, T. Polo, B. Pomoni, M. Poniewierski, J. Popovic, P. Portugaller, H. R. Prabhudesai, V. Pretorius, D. H. Privat, C. Providência, L. Psomas, M. Puppini, G. Qanadli, S. D. Qarioty, M. Qian, Z. Quartagno, R. Rabbia, C. Radeleff, B. Radi, F. Rahmouni, A. Rampoldi, A. Rand, T. Raouf, S. A. Rasmusen, I. Raupach, J. Rausin, L.

P34 35.6.4 43.5.3, P40 P7, P107, P161 P155 6.5.4, 15.2.9, 43.2.9, P182 6.5.5 P89 6.6.6 6.2.6 35.2.4 35.2.8, P64, P68 6.2.8, 15.1.5, 43.2.1, P12, P132, P135 P41 P34, P183 P138 25.1 6.5.1, 6.6.5, 35.5.8 P151 P16 6.3.4, 15.1.3, 15.1.6 35.4.5 35.4.5 P101 P99, P104, P105 35.3.3 15.2.1, 35.5.7 15.2.5, P115 P4, P9, P136 6.6.7 35.6.3 6.4.9 15.4.5 P25, P166 P46, P164, P165 43.1.9 36.2 35.5.1 P180 35.2.3 15.4.1 43.5.6 P45, P145, P189 35.3.6, 43.4.8, P49 43.1.9 P2 10.2, 35.1.2, 43.5.8 P72

Ravel, A. Raynaud, A. Razmadze, M. Real, M. Reale, C. A.

Reb, P. Rees, M. Reiser, M. F. Rendina, L. Resta, M. Resta, M. C. Reusch, G. Rhim, H. Ricci, E. Richter, G. M.

Ricke, J. Rieger, M. Rimbach, S. Ring, E. J. Ripoll, C. Risberg, B. Risse, O. Rivera SanFeliz, G. M. Rivolta, R. Rix, D. Rizzo, E. Rocek, M. Roberts, A. C. Rodriguez Rosales, G. Rodriguez-Merlo, R. Roh, B. Rokitansky, A. Romagnoli, A. Rösch, J. Rose, J. Rose, S. C. Rosenthal, H. Roser, H. W. Rosset, E. Rossi, P. Roth, J. Rousseau, H. P. Rowinski, O. Ruefenacht, D. A. Ruehm, S. G. Ruiz, I. Rundback, J. H. Sabater, E. A. Sabharwal, T. Sachs, W.

16.3 P70 P29 27.1 6.5.1, 6.6.5, 15.5.1, 35.5.8, 43.1.3 15.6.7 6.2.2, P53 35.4.7 P138 43.1.7 43.1.7 P43 P51, P76 6.1.2 35.2.3, 35.3.2, 43.6.5, 43.6.6, P55 P98 35.1.4, P13, P162 35.2.3 6.3.8 P1 P120 6.3.3 6.6.7, 15.1.1 P138 43.5.2 P164, P165 6.5.7, P131 6.6.7, 15.1.1 P1, P152, P153 P109 P172, P188 15.5.2 35.4.3, 35.4.4 15.6.2, 43.6.9 43.2.3, 43.2.8, P17 6.6.7, 15.1.1, 43.6.2, 43.6.4 6.3.5 P24 16.3 6.1.1, 15.1.8, 15.2.6, 43.3.4 P24 1.1, P113 P137, P174 6.4.6, 6.4.7 33.3 P109 P75, P78, P80 6.5.3 6.3.2 P43

Sahara, S. Saint-Maurice, J. Saito, H. Sakaguchi, H. Sakai, J. Sakai, T. Sakai, Y. Sakamoto, T. Sakellaropoulos, G. Salapura, V. Salvatori, F. M. Salvioni, M. Samaras, N. Sandner, D. Sanfelix, P. Šaponjski, J. Sapoval, M. Šaranovic, D. Sato, A. Sato, K. Sato, M. Satzl, S. Savio, D. Saxena, A. Sayek, &. Scalisi, L. Schäfer, M. Schertler, T. Schicchi, F. Schild, H. Schizas, D. Schmehl, J. Schmenger, P. Schmid, C. Schmidt, W. Schmiedel, A. Schmiedt, W. Schmitt, A. Schmitz, K. P. Schmitz-Rode, T. Schmook, T. Schneider, J. Schnyder, P. Schoder, M. Scholten, M. Schöpke, W. Schultz, F. Schulz, S.

P39, P184 15.4.6 P50, P186 P90 P67 P19 P143, P147, P148, P150 P97 P108, P111, P117 15.2.1, 35.5.7 6.1.1, 15.1.8, 15.2.6, 43.3.4 43.4.7 P111 43.5.5 P69 P61 6.6.1, 16.2, P70, P130 P61 P50, P186 P147 P39, P184 43.6.5, 43.6.6, P55 P180 35.4.2, P133 P79 P34, P183 P98 35.4.1, 43.2.6, P182 35.4.5 P27, P30, P31, P57 P122 15.3.1 15.1.6 6.5.4 35.3.5 P27, P30, P31, P57 15.1.3 43.5.9 35.3.5 1.3, 6.5.5, 35.6.6 43.4.8 6.3.4, 15.1.3, 15.1.6 P46, P164 35.3.6, 43.4.8, 43.6.9 35.4.6 6.4.1 43.3.7 P36

Schwartz, M. Schwarz, A. Schweitzer, H. Secchi, A. Seebach, J. Seinost, G. Selby, J. B. Semmler, W. Sen, D. Sengel, C. Sengul, M. Seo, H. Seo, J. Seo, T. Sepul, A. Setacci, C. Shah, K. Sharma, S. Sharma, V. Sherrington, C. Shim, H. Shin, H. Shin, J. Shinohara, M. Shirai, S. Shoenfeld, R. Shonai, T. Siablis, D.

Siddiqui, M. A. Sideris, D. Silit, E. Simó Muerza, G. Simon, G. Simonetti, C. Simonetti, G.

Simonffy, Á. Simons, P. Sipola, P. Sirin, S. Sissopoulos, A. Smeets, A. J. Smith, J. Snowden, S. Sofocleous, C. T. Solcia, M. Soler, J. Sommer, T. Somon, T.

P42 15.6.7 15.1.3 15.2.8 P182 P41 6.5.6, 15.1.2, 43.3.3 P58 P91 6.3.3 P59, P66 P51, P76 15.6.5 P51, P149 P72 6.2.6 15.4.7 35.4.2, P18, P133 15.4.7 P72 43.6.8, P56 P63 15.6.3, 15.6.5, P51 P67 P39 43.6.3 P19 6.2.4, 15.3.5, P7, P108, P111, P117, P121, P161 P4 6.2.2, P53 P178 P1, P152, P153 P14, P32 P96 6.2.7, 6.5.1, 6.6.5, 15.5.1, 35.4.3, 35.4.4, 35.5.8, 43.1.3 6.1.8 43.2.4 43.3.5 P37 P94, P134 35.2.4 15.3.2 P139 6.3.8, 43.4.2 P45, P145, P189 P33 P27, P30, P31, P57 6.4.7

Somuncu, &. Son, S. Sonesson, B. Song, H.

15.5.6, P15 P56 28.3 15.6.3, 15.6.5, 38.2, P51, P149 Sönmez, A. P15 Soomro, N. 43.5.2 Souftas, V. D. P65, P77 Sourgiadaki, E. P111 Sperandio, M. 35.4.3 Spinelli, A. 6.5.1, 15.5.1, 43.1.3 Spors, B. P98 Staedele, H. G. P24 Stangalini, S. P173 Stankovic, M. 35.5.7 Stark, G. P41 Stecco, A. P173 Stefanini, M. 6.2.7, 6.5.1, 6.6.5 Steingruber, I. E. 35.1.4, 35.1.5, 35.1.8 Steinkamp, H. J. 2.3 Stollberger, R. P115 Stolzmann, P. 43.1.8 Strecker, E. P. 43.3.7 Strizzi, V. 35.6.2 Struyven, J. 43.3.8, P130 Sugimoto, K. P146 Sugiura, K. P52 Suh, S. 35.1.1 Sung, K. P156 Šurlan, M. 15.2.1, 35.5.7 Susman, J. P75, P78, P80 Svebisova, H. P102 Swatowski, D. P68 Szajner, M. 6.1.3 Szarka, J. P14, P32 Szczerbo-Trojanowska, M. 6.1.3, 15.1.7, 35.2.8, P64, P68 Szentesi, S. P14, P32 Szmidt, J. P137 Szymañska, A. 35.2.8, P68 Takahashi, S. P50, P186 Takase, K. P186 Talwar, K. K. P18, P133 Tamakawa, M. P19 Tanaka, R. P126 Tanaka, S. P147 Tanami, Y. P67 Tani, I. P8, P10, P84 Tanihata, H. P39 Tanne, D. 43.1.4, 43.1.5 Tannouri, F. 35.2.5, 43.3.8 Tatar, H. P15 Tauss, J. 15.2.5 Teboune, N. 35.6.4, 43.4.6

Tedoldi, S. Tengg-Kobligk, H. V. Tepe, G. Tepe, S. M. Terada, M. Terraz, S. Terrier, F. Tesdal, K. Teta, D. Thanos, L.

Thelen, M. Thomas, F. Thompson, D. Thomson, K. R. Thornton, J. Tielbeek, A. Tielemans, H. Tillianakis, M. Timmermans, H. A. Timurkaynak, E. Tobio, R. Todua, F. Tokat, Y. Toledano, N. Toma, C. Tomassini, M. Tomlinson, G. Torsello, G. Towbin, R. B. Toyoshima, M. Toyota, N. Tozzi, M.

Tran Ba Huy, P. Trerotola, S. Trinh, C. C. Trück, J. Tsamandas, A. C. Tsang, J. Tsanis, A. Tsetis, D. K.

Tsitouridis, I. Tso, W. K. Tsolakis, I. Tsota, I. Tübler, T. Tulic, C.

35.5.1 P87 15.3.1, 15.3.2, 35.3.4 20.1 P39 6.3.1, 35.6.8 35.6.8 43.4.1 P165 P21, P22, P35, P94, P99, P104, P105, P122, P140 6.3.4, 15.1.3, 15.1.6 43.6.5, 43.6.6, P55 6.1.5 24.1 6.2.9, 15.4.3 35.3.2 35.3.2 P108, P117 15.6.2, 43.6.9 P37 17.3, P141 P29, P159 P127 P109 P49 35.4.3, 35.4.4 P42 P106, P124 6.6.2, 6.6.8 35.6.7 P112 15.6.2 6.2.8, 15.1.5, P12, P132, P135 6.6.4 9.2, 32.4 P177 35.3.4 P107 P23 P104 6.2.2, 6.3.2, 7.1, 15.3.3, P53 P65, P77 P23, P154 6.2.4, 15.3.5 6.2.4, P7, P107, P108, P117 6.2.3, 15.3.6, P123 P61

Tung, H. Turbin, R. Turco, G. L. Turello, E. Turmel-Rodrigues, L. A. Tuveri, M. Tzorakolefterakis, E. Uccioli, L. Uchida, B. T. Uflacker, R. Ugurel, &. Ulfendahl, H. Umathum, R. Unal, M. Urban, M. Urbani, L. Urbano, J. Üstünsöz, B. Utikal, P. Vagnarelli, S. Vainio, P. Vakhtangadze, T. Valiensi, S. M. Valji, K. Valvassori, L. van Overhagen, H. van Vliet, M. Vaninbroukx, J. Vanninen, R. Varela, M. Varghese, J. C. Ventikos, Y. Venturini, M. Venugopal, P. Veraldi, G. Verbist, J. Verdu, J. Vervest, H. A. Vetter, S. Videla, C. G. Vignali, C. Vlachos, L. Voloduckin, M. Y. Volz, S. Vorwerk, D. Wagner, H. J. Wahlberg, A. Walch, C. V. Waldenberger, P. Wallace, C. Wang, A. Wang, W. Wang, Y.

P23, P154 15.4.7 P34, P183 P173 32.2, P70 P179 P121 6.2.7 15.6.2, 43.6.9 6.5.6, 15.1.2, 43.3.3 15.5.6, P91 P119 P58 35.6.1 15.5.2 43.5.3 P47, P101 15.5.6, P15, P37, P83, P91 43.2.5 6.1.1, 15.1.8, 15.2.6, 43.3.4 43.3.5 P29 6.1.7 6.6.7, 15.1.1 15.4.2 43.2.4 35.5.3 P151 43.3.5 43.4.9 P4 35.4.1, 43.2.7 15.2.8, 35.5.1, 43.4.7 35.4.2, P18 P166 6.1.6, 35.3.1 P60, P125 35.2.4 43.3.7 6.1.7 15.2.7, 43.5.3, P40 43.6.7 P168 P58 6.5.5, P130 43.5.9 P119 P20 P20, P162 P42 15.3.4, P142 6.5.3 6.6.4

Wassef, M.

6.6.4, 15.4.6, 15.6.7, 43.3.6 Watanabe, I. P157 Watanabe, K. P157 Weintraub, J. L. P75, P78, P80 Weiss, T. 6.6.6 Weiß, C. 6.5.5 Wendel, H. J. 35.3.4 Wenke, R. 43.1.8 Werk, M. P98 Wessman, A. P119 West, D. J. P4, P9, P136 Weszelits, V. P82 Whigham, C. J. 15.4.4, P177 Wiesinger, B. 15.3.1, 35.3.2 Wiesli, P. 6.5.4 Wildberger, J. E. 6.5.5, 35.6.6 Wildermuth, S. 35.4.1, 43.2.6, 43.2.7 Williams, L. 6.5.6 Wilputte, J. Y. 15.2.2 Wilson, L. 43.2.3, 43.2.8, P17 Wirth-Jaworski, L. 15.6.1 Wiskirchen, J. 15.3.1 Wittebolle, X. 15.2.3, 35.5.6 Won, J. H. 43.3.1 Won, J. Y. 35.1.1, 43.2.2, P110, P156 Wong, Y. P23 Worthington-Kirsch, R. L. 35.2.1 Wyatt, M. 43.2.3, 43.2.8, P17 Yama, N. P19 Yamada, K. P39 Yamada, R. P143, P148, P150 Yamamoto, A. P112 Yamamoto, S. P144, P146 Yamanouchi, E. P8, P10, P84 Yang, D. P3 Yildirim, E. P48 Yilmaz, H. 6.4.6, 6.4.7 Yin, Q. 15.6.2 Yodfat, O. 43.1.4, 43.1.5 Yoo, J. 43.6.8 Yoon, C. 15.6.3, 15.6.5 Yoon, H. P156 Yoon, M. P56 Yoon, W. P181 Yoon, Y. P149, P190 Yoshioka, T. P90 Youn, S. W. 43.3.1 Yuen, J. P154 Yuk, S. 15.6.5 Zabakis, P. 6.2.4, 15.3.5, P121 Zanetti, G. P179 Zech, F. 15.2.2

Zerbi, A. Zeteroglu, S. Zietse, B. Zoetelief, J. Zoganas, L. Zollikofer, C. Zubilewicz, T. Zuehlsdorff, S.

6.4.2, P129, P192 P59, P66 35.5.3 43.3.7 P105 1.2, P130 15.1.7 P58