Pediatr Cardiol 23:334±346, 2002 DOI: 10.1007/s00246-001-0197-6

Fetal Echocardiography: New Grounds to Explore During Fetal Cardiac Intervention T. Kohl Department of Obstetrics and Gynecology, Division of Prenatal Medicine, University of LuÈbeck Medical School, Ratzeburger Allee 60, 23538 LuÈbeck, Germany

Abstract. Over the past decade, revolutionary advances in ultrasound imaging technology have allowed the study of the evolution of congenital heart disease during fetal life. The frustration arising from watching the prenatal progression of severe semilunar valve obstructions and therapy-refractory fetal arrhythmias has prompted the interest in developing procedures for fetal cardiac intervention. Ultrasound techniques as the primary diagnostic and monitoring modalities in fetal medicine will play a key role in de®ning patient groups, peri-interventional assessment of fetal hemodynamics, and monitoring during these procedures. The purpose of this article is to provide pediatric cardiologists and perinatal medicine specialists an overview over the various technical approaches at fetal cardiac intervention and the special tasks that fetal echocardiography will have to accomplish. It also aims at illustrating the potential of fetal echocardiography for fetal selection. Key words: Fetal echocardiography Ð Cardiac surgery Ð Catheter intervention Ð Fetal surgery Ð Fetal cardiac intervention Over the past decade, revolutionary advances in ultrasound imaging technology have allowed the study of the evolution of congenital heart disease during fetal life. It has become clear that over the course of gestation, altered pressure and ¯ow relationships may result in signi®cant secondary damage to the developing heart. Particularly severe semilunar valvar obstructions may in¯ict progressive secondary damage to the a€ected ventricle and associated cardiovascular structures [1, 3, 16, 27±29, 47, 57, 60]. This

Email: [email protected]

damage signi®cantly increases postnatal mortality and morbidity and reduces treatment options and overall prognosis. Furthermore, life-threatening fetal cardiac failure, brain damage, or demise may result from fetal arrhythmias refractory to conventional treatment attempts [48, 54±56, 58, 59]. The frustration arising from watching the prenatal progression of severe semilunar valve obstructions and therapy-refractory fetal arrhythmias has prompted the interest in exploring the feasibility of fetal cardiac intervention [2, 5, 6, 9, 11, 13, 15, 19, 20, 23, 30, 32, 34, 36±38, 40, 42±46, 50, 53, 61, 63]. Shortening the prenatal disease course of semilunar valve obstructions during fetal life is intriguing because the severity of secondary cardiovascular injury may be reduced. If this goal is achieved, reconstructive (rather than palliative) surgical procedures might be performed postnatally. Normalization of fetal heart rate in cases of therapy-refractory fetal arrhythmias might be life-saving for fetuses in cardiac failure. Ultrasound techniques are the primary diagnostic and monitoring modalities in fetal cardiology, and they will play a key role in de®ning patient groups, peri-interventional assessment of fetal hemodynamics, and monitoring during any procedure. The purpose of this article is to provide pediatric and fetal cardiologists as well as perinatal medicine specialists an overview of the various technical approaches at fetal cardiac intervention and the special tasks that fetal echocardiography plays during these procedures. It also aims at illustrating the potential of fetal echocardiography for fetal selection. Fetal selection Before discussing the role that fetal echocardiography might play for fetal selection and peri-interventional monitoring, the bene®ts and drawbacks of fetal car-

Kohl: Fetal Echocardiography

diac interventions must be understood. As most fetuses with severe semilunar valve obstructions survive gestation, fetal cardiac intervention in these lesions will rarely serve as a life-saving procedure but rather aims at improving postnatal operative options and overall prognosis. Although more procedures will most likely be required postnatally, fetal cardiac intervention in these lesions may hold promise to ameliorate the natural disease course to such an extent that catheter interventions or reconstructive biventricular instead of palliative univentricular surgical procedures become possible [23, 33]. Because the potential success and risks of the prenatal procedures are dicult to predict and will di€er for each one of the interventional approaches, cardiac intervention in fetuses with severe semilunar valve obstructions in the worst case might be not only technically unsuccessful but also followed by fetal intrauterine death or very early preterm delivery, resulting in signi®cant and potentially lifelong additional morbidity. Therefore, even if the goal of a biventricular circulation is achieved by the prenatal procedure, the overall outcome and quality of life of these patients will ultimately determine whether fetal cardiac intervention will become a bene®cial therapeutic alternative to currently available postnatal procedures. This situation is entirely di€erent in fetuses with therapy-refractory arrhythmias, as the majority of them will not survive gestation or attempts at preterm delivery for postnatal treatment. Cardiac intervention in these fetuses will clearly be undertaken as a potentially life-saving procedure, justifying more aggressive interventional approaches. With these potential bene®ts and drawbacks in mind, the role of fetal echocardiography for fetal selection may be de®ned by analyzing and inferring from the growing body of pre- and postnatal studies in humans. In human fetuses with left heart obstructive lesions, retrospective studies suggest that serial measurements of left heart growth and assessment of ¯ow direction across the foramen ovale and distal aortic arch may predict postnatal left heart size [16, 27, 28, 47, 57]. Infants that required postnatal surgical univentricular repair showed decreased growth of the left ventricle, mitral valve, and ascending aorta. In addition, left-to-right shunting across the foramen ovale and retrograde ¯ow in the distal aortic arch were identi®ed by Doppler interrogation in this group. Although these ®ndings await con®rmation in prospective patient series, they appear physiologically sound to identify fetuses at risk for severe left heart hypoplasia that may bene®t from fetal cardiac intervention. Data from studies indicate that the severity of pulmonary valve obstructions may be assessed prenatally by the direction of ductus arteriosus ¯ow, the presence of pulmonary insuciency, the degree of

335

tricuspid insuciency, and the size of the pulmonary valve and arteries [29]. Yet the hypoplastic right ventricle in patients with severe pulmonary valve stenosis shows signi®cant potential for catch-up growth following postnatal procedures. Therefore de®ning the population of fetuses with these lesions that will bene®t from fetal cardiac intervention remains speculative. In postnatal individuals with pulmonary atresia and intact septum, a simple echocardiographic measurement of the tricuspid valve diameter has been useful to predict the necessity of surgical univentricular repair in this lesion [24]. It remains to be seen if serial studies of prenatal tricuspid valve growth in fetal pulmonary valve obstructions may allow us to infer on the degree of postnatal right heart hypoplasia. In contrast to severe semilunar valvar obstructions, the recognition and selection of fetuses with therapy-refractory arrhythmias is more readily achieved with fetal echocardiography. Most fetuses with congenital complete heart block and normal cardiac anatomy will survive until term and throughout the neonatal period if their ventricular escape rate is higher than 55 beats per min [55]. If the ventricular escape rate falls below 50 beats per min or additional cardiac malformations are present, fetal hydrops from cardiac failure is likely to occur. Despite the anecdotal evidence that some fetuses in cardiac failure may bene®t from administration of bsympathomimetics, anticongestive drugs, steroids, or premature delivery [4, 12, 14, 18, 25, 51, 55], most fetuses with congenital complete heart block and hydrops from cardiac failure will die, regardless of whether they have normal or abnormal cardiac anatomy [21, 55]. Like those with congenital complete heart block, most fetuses with supraventricular tachycardia that do not exhibit signs of cardiac failure (i.e., hydrops) will survive gestation. In contrast, fetuses with supraventricular tachycardia and hydrops refractory to conventional treatment attempts die from severe cardiac failure in more than 25% of cases [56]. In survivors, an increased incidence of brain damage has been noted that might be related to the vastly abnormal fetal hemodynamics [54, 58]. Based on this poor outcome, fetuses with therapyrefractory arrhythmias in cardiac failure may be selected for life-saving intervention. Procedures Percutaneous Ultrasound-Guided Fetal Cardiac Interventions Pioneering attempts in human fetuses at balloon valvuloplasty in semilunar valvar obstructions, fetal

336

Pediatric Cardiology Vol. 23, No. 3, 2002

Fig. 1. Percutaneous direct ultrasound-guided puncture for fetal balloon valvuloplasty (courtesy of P. Zielinsky). Top left: External setup for procedure in a 32-week-old human fetus with severe aortic valve stenosis. Using a free-hand technique, the needle (left hand) has been inserted into the fetal heart under ultrasound monitoring (right hand). Top right: Dilated left ventricle with hyperechogenic mitral valve apparatus in this fetus. Bottom left: The arrows depict the percutaneous-transuterine-transthoracic needle course toward the left ventricular out¯ow tract. Bottom right: Following needle removal, signi®cant hemopericardium formation is observed which was accompanied by sustained fetal bradycardia. HP, hemopericardium. A, anterior; P, posterior; R, right; L, left; RV, right ventricle; LV, left ventricle.

cardiac pacing in congenital complete heart block, and overdrive stimulation in supraventricular tachycardia have been performed by percutaneous direct ultrasound-guided approaches [2, 13, 15, 17, 30, 36, 45, 46, 62, 63; personal communications from Zielinsky, Tworetzky, Gembruch, and Tulzer]. To perform these procedures, direct insertion of a needle through the maternal abdomen into the uterus and from there through the fetal chest wall into the cardiac region of interest is monitored by two-dimensional ultrasound imaging. Following intracardiac positioning of the needle, interventional devices (e.g., guide wire, valvuloplasty catheter, pacing lead) are delivered through the needle shaft toward the cardiac region of interest (Figs. 1 and 2). For the success of percutaneous ultrasoundguided fetal cardiac interventions, adequate fetal po-

sition, excellent imaging quality, and availability of sucient acoustic windows are critical [42]. A sucient amniotic ¯uid pocket between the uterine wall and the fetal chest facilitates recognition and ®netuning of the intra-amniotic course of the needle before it is advanced into the fetal chest. Patience is needed until the fetus attains the most favorable position as its lie is dicult to control from outside the womb. Unfortunately, hydropic fetuses in severe cardiac failure rarely move as a survival strategy to not waste oxygen. This problem may render the procedure not feasible if the fetal lie is dorsoanterior. Paralyzing and anesthetizing the fetus by intramuscular injection of nondepolarizing muscle relaxants under ultrasound guidance and maternal administration of opioid analgesics may help maintain the desired fetal lie and suppress unfavorable maternal and

Kohl: Fetal Echocardiography

337

Fig. 2. Top row: The arrows depict the percutaneous-transuterine-transthoracic needle course toward the left ventricular out¯ow tract in a late gestation fetal sheep (left). As the needle is perfectly aligned with the out¯ow tract, a 0.014-inch guide wire is easily placed across the nonobstructed aortic valve in this experiment (right). Bottom row: A 4-mm balloon catheter (arrows) has been placed across the valve annulus (left). Note that even during real-time imaging the di€erentiation between needle shaft and catheter shaft is extremely dicult. This diculty results in balloon rupture as the balloon is torn by the edges of the catheter shaft during in¯ation, which has been observed in about half of the human cases. The arrow points at bubbles (right). A, anterior; P, posterior; R, right; L, left; RV, right ventricle; LV, left ventricle; AoV, aortic valve.

fetal stress responses. In addition, these steps are useful to avoid repeated attempts at cardiac puncture or fetal re¯ex movements that may result in bleeding complications or dislodging of interventional materials [46]. During percutaneous ultrasound-guided fetal cardiac interventions, ultrasound scattering at the surface of interventional devices might signi®cantly impair adequate visualization or even render the procedure unfeasible. Due to the limited clinical experience with which any of these procedures have been performed, the lack of tested interventional devices, and the selection of worst cases, the early clinical experience with percutaneous ultrasound-guided fetal cardiac interventions has been fraught with signi®cant fetal morbidity and mortality [13, 36, 62]. Between 1990 and May 2001 , percutaneous ultrasound-guided fetal balloon valvuloplasties were attempted in 18 human fetuses. Technically successful balloon valvuloplasties were achieved in eight of

these fetuses, two of whom had an atretic valve; only three of these remain alive today. Of the 10 remaining technical failures, 2 patients with severe aortic stenosis and 2 with pulmonary atresia with intact septum underwent successful postnatal interventions and remain alive. Six patients that survived prenatal intervention died from cardiac dysfunction or at surgery in the ®rst days or weeks after delivery. Five fetuses died early (within 24 h) after the procedure, one from a bleeding complication, three from persistent bradycardias, and one at valvotomy following emergency delivery. Between 1986 and June 2001, percutaneous ultrasound-guided insertion of pacing wires into the fetal heart was reported in three human fetuses, all of whom died within 36 h following the procedure [13, 62; Zielinsky personal communication]. Using the same approach for fetal cardiac access, overdrive stimulation of the left fetal atrium with an electrocatheter was unsuccessful in one fetus

338

Pediatric Cardiology Vol. 23, No. 3, 2002

Fig. 3. Top row: Maternal laparotomy and hysterotomy are required to access this human fetus for open fetal cardiac surgery (left) A pacing lead has been sutured onto its fetal heart to salvage this extremely hydropic fetus from demise due to congenital complete heart block (right) (courtesy of P. Zielinsky). Fetal echocardiography in this setting may aid in optimizing fetal hemodynamics by assessment of ventricular contraction and cardiac output at di€erent pacing rates. Bottom row: In this fetal lamb, pulmonary artery banding is performed by open fetal cardiac surgery at 60 days of gestation (left). The ecacy of band constriction (arrow) is monitored by fetal transesophageal echocardiography utilizing an intravascular ultrasound catheter.

with therapy-refractory tachycardia [Zielinsky Personal communication]. Open and Fetoscopic Fetal Cardiac Surgery The development of open and fetoscopic techniques for fetal cardiac surgery has signi®cant potential to improve the technical success rate of fetal valvotomies, pacemaker insertions, and other cardiac procedures not amenable to percutaneous ultrasoundguided approaches. Fetal cardiac access for open fetal cardiac surgery requires maternal laparotomy and hysterotomy, followed by fetal thoracotomy (Fig. 3). Employing the open approach, fetal cardiac bypass, pacemaker implantation, repair of simple cardiovascular lesions, and cryosurgical atrioventricular nodal

ablation have all been feasible in experimental studies in fetal sheep with survival to term [5, 6, 9, 10, 20, 23, 44, 50, 53]. Alternatively, the feasibility of various minimally invasive fetoscopic operations that aim at alleviation of severe fetal semilunar valvar obstructions and termination of therapy-resistant fetal arrhythmias in human fetuses has been tested in sheep [34, 37±43]. Fetoscopic fetal cardiac surgery is achieved by percutaneous intraamniotic placement of endoscopic cannula followed by fetal esophageal intubation for transesophageal echocardiography, thoracotomy, or cardiac catheterization (Figs. 4 and 5). Maternal transabdominal two-dimensional and Doppler color ultrasound scanning utilizing both linear and sector array transducers is critical during insertion of the ®rst trocar as injury to maternal and fetal organs and

Fig. 4. The many roles of ultrasound during fetoscopic transventricular fetal cardiac catheterization. Top row: Percutaneous access is achieved by carefully scanning for a suciently large amniotic ¯uid pocket that allows for save insertion of the ®rst trocar. Using color Doppler imaging with low Nyquist limits is useful to avoid injury to maternal, fetal, or umbilicoplacental structures during this critical step. To avoid fetal stabbing or maternal injury, two or three T-fasteners (nylon suture attached to 5-mm stainless steel bars) are inserted into the amniotic cavity (left). The right image shows the released steel bar (arrow) inside the amniotic sac. Middle row: Then the ultrasound transducer is placed between the T-fastener sutures, which are pulled up (left). The operator now scans from this site in all directions to de®ne the deepest amniotic ¯uid pocket. Once the pocket is found, he or she aligns the sutures with the transducer. While the sutures are being held by an assistant, the transducer is removed and the surgeon inserts the trocar into the amniotic cavity maintaining strict alignment with the sutures (right). Bottom row: Following placement of two or three more 3±5-mm trocars and gas insu‚ation of the amniotic cavity, an intravascular ultrasound catheter is inserted into the fetal esophagus. Fetoscopic view on the fetal lower mandible, which has been suspended using a stay suture. A 10F catheter sheath has been placed into the fetal oropharynx to allow for rapid exchange of imaging catheters (left). The catheter permits two-dimensional real-time assessment of the fetal heart (right).

Fig. 5. The many roles of ultrasound during fetoscopic transventricular fetal cardiac catheterization (continued). Top row: Following intraesophageal insertion of the ultrasound imaging catheter, the insu‚ation gas is vented and the amniotic cavity re®lled with warmed sterile saline (left). This approach permits fetal posturing and ultrasound-guided marking of the incision site for fetoscopic fetal thoracotomy (right). Middle row: Following evacuation of most amniotic ¯uid and gas insu‚ation of the amniotic cavity, fetal thoracotomy can be carried out precisely at the desired site (left, outside view; right, fetoscopic image). Bottom row: In this instance, a 16-gauge needle shaft serves as the cardiac access route for transventricular fetal cardiac catheterization (left). The needle shaft enters the left ventricular cavity right next to the descending coronary vessels. The shaft has been oriented toward the left ventricular out¯ow tract. Placement of a 0.014 inch guide wire (arrowheads) across the aortic valve toward the aortic arch (AoA) is achieved under fetal transesophageal echocardiographic guidance (right). DA, ductus arteriosus; T, trachea.

Kohl: Fetal Echocardiography

vessels as well as placental damage must be avoided [41, 43]. During open and fetoscopic fetal cardiac procedures, conventional maternal transabdominal fetal echocardiography is precluded because it interferes with the operative approach or may not be possible after gaseous insu‚ation of the amniotic cavity [37, 42]. Alternatively, imaging of the fetal heart from the fetal esophagus can be performed utilizing smallcaliber intravascular ultrasound catheters. By being advanced and withdrawn within the esophagus, these catheters permit accurate de®nition of fetal cardiac anatomy (Figs. 6 and 7). Shaft diameters of ultrasound catheters suitable for transesophageal fetal echocardiography range of 6.5±10 French and can be inserted in fetal sheep as small as 300 g. This translates to a human gestational age as young as 20 weeks. The various purchasable mechanical ultrasound catheters carry rotating assemblies equipped with 9±12.5 MHz transducers that create a 360° imaging plane perpendicular to the long axis of the catheter (Fig. 6). The disadvantage of this design for fetal transesophageal imaging is that only close to 15% of the displayed image contains the fetal heart and that any increment in magni®cation is bought at the expense of image depth display. Realtime images are obtained at a rate of 30 frames per s with a tissue penetration of little more than 40 mm. During fetal transesophageal echocardiography, a mechanical intravascular ultrasound system permits unobstructed high-quality imaging of infra- and supracardiac vessels and cardiac structures in the near ®eld of the catheter (0±2.5 cm), which allows measurement of the fetal great vessels and of cardiac valve dimensions with high measurement agreement and acceptable variability to conventional maternal transabdominal fetal echocardiography [39]. Conversely, the imaging of ventricular structures or interventional devices in the far ®eld of the catheter (>2.5 cm) is substantially worse. As the mechanical intravascular ultrasound catheter lacks Doppler capabilities, assessment of fetal hemodynamics during fetal cardiac interventions is limited to observation of changes in heart rate, size of the arterial duct, and foramen ovale patency. The decreased depth penetration and the lack of Doppler capabilities limit the application of the available mechanical intravascular ultrasound systems in the human fetus. Fortunately, the apparent limitations of the mechanical intravascular ultrasound systems for fetal transesophageal echocardiography have recently been overcome by dramatic improvements in intravascular ultrasound technology. A novel single-plane intravascular catheter tipped with a frequency agile 5.5±10-MHz vector phased-array ultrasound transducer permits two-dimensional real-time imaging

341

with a high penetration (up to 10 cm from the lens) as well as multimodal Doppler imaging (Fig. 7). Once this technology is cleared for human application, it may hold promise to o€er the tiny unborn patient the full range of echocardiographic modalities that has become standard during postnatal operations. Many questions regarding the most e€ective operative techniques and their safety for mother and fetus have not been answered yet. As gaseous or ¯uid emboli might originate from the amniotic cavity or placental abruption occur during fetoscopic or open fetal surgery, the mother must be monitored meticulously for these potentially dangerous complications. The role that ultrasound monitoring might play for early recognition and de®nition of severity during these operations still needs to be de®ned. Therefore, apart from a few attempts at pacemaker implantation in human fetuses by open fetal cardiac surgery [Harrison personal communication, Zielinsky personal communication], all other open and fetoscopic approaches still dwell in various stages of maturation. Peri-interventional Monitoring Following fetal cardiac intervention, serial ultrasound studies are required to assess the maternal and fetal e€ects of the intervention. To address all relevant maternal and fetal aspects, these studies should be undertaken by an interdisciplinary team consisting of surgeon, obstetrician, perinatologist, and fetal cardiologist. The mother must be monitored closely for clinical and ultrasound evidence of placental abruption, intra-abdominal bleeding, or amniotic leakage. If possible, the fetus should be monitored for signi®cant pericardial e€usions as well as ®rst-time occurrence, regression, or worsening of fetal hydrops, which might be the most useful indicator of overall fetal hemodynamics. Particularly after the open and fetoscopic operative procedures, maternal intra-abdominal air or skin emphysema, dorsoanterior fetal lie, and decreased amniotic ¯uid amount may render attempts at imaging unfeasible. Despite these unfavorable imaging conditions, umbilical cord blood ¯ow signals can in most instances be obtained by pulsed Doppler interrogation. Prospective studies are, however, required to corroborate retrospective data from a small cohort of human fetuses (with a noncardiac lesion, i.e., diaphragmatic hernia) that low postoperative umbilical ¯ow velocities may be early indicators of poor outcome [35]. Although useful in fetuses with intrauterine growth failure [26], umbilical venous pulsations and retrograde ¯ow in the ductus venosus with atrial contraction do not necessarily re¯ect poor hemodynamic status in fetuses with congenital heart disease.

Fig. 6. Demonstration of fetal cardiac anatomy in horizontal planes by fetal transesophageal echocardiography in a sheep fetus utilizing a mechanical intravascular ultrasound catheter. Top: Four-chamber plane (left), aortic valve plane (right). Middle: Pulmonary valve plane (left), great vessel plane (right). Bottom: Arrows point at 0.014 inch guide wire in aortic arch (left), arrows point at in¯ated 3.5-mm balloon catheter across aortic valve (right). A, anterior; P, posterior; R, right; L, left. RA, right atrium; LA, left atrium; RV, right ventricle; LV, left ventricle; AoV, aortic valve; MPA, main pulmonary artery; PV, pulmonary valve; RVOT, right ventricular out¯ow tract; T, trachea; DA, ductus arteriosus; AA, aortic arch.

Kohl: Fetal Echocardiography

343

Fig. 7. Fetal transesophageal echocardiography in vertical planes in a sheep fetus utilizing a phased-array intravascular ultrasound catheter. Top row: Demonstration of fetal cardiac anatomy. Bottom row: Color Doppler imaging of forward ¯ow in the ascending aorta during left ventricular systolic ejection (left) and of retrograde ¯ow from aortic valve regurgitation during diastole (middle). Pulsed Doppler interrogation of the ductus venosus. A, anterior; P, posterior; I, inferior; S, superior; RA, right atrium; RV, right ventricle; SVC, superior vena cava; IVC, inferior vena cava; VS, ventricular septum; RVOT, right ventricular out¯ow tract; AoV, aortic valve; LV, left ventricle; LA, left atrium; AAo, ascending aorta; MPA, main pulmonary artery; DV, ductus venosus.

Following attempts at transumbilical fetal cardiac catheterization, the catheterized umbilical artery and its distal placental bed need to be examined for patency and changes in vascular resistance. After fetal cardiac pacing, fetal echocardiography is required to assess ventricular capture, cardiac function, and hydrops. Similarly, in fetuses with supraventricular tachycardia, fetal echocardiography aims at assessing the e€ects of overdrive stimulation and antiarrhythmic drug therapy on cardiac rhythm and function. Oligohydramnios or ahydramnios following fetal surgery may be the result of poor hemodynamic status of the fetus or renal dysfunction due to tocolysis with indomethacin. In all fetuses (particularly those with severe semilunar valve obstructions) maternal administration of indomethacin for tocolysis

should be closely followed because of its potential to induce ductal constriction [31, 49, 52]. For the same reason, steroid treatment to enhance fetal lung maturation will require careful echocardiographic monitoring in fetuses with ductus-dependent heart disease [7]. Cerebral damage from intraventricular hemorrhage or periventricular leukomalacia has been observed after fetal surgery for diaphragmatic hernia in human fetuses [8, 22]. It is not clear if this complication results from hemodynamic changes that occur during either the procedure or the perioperative period. In the future, prospective serial multimodal ultrasound studies of the fetal brain, starting with at least one baseline scan before the operation, may provide a better understanding of the timing of events

344

and the e€ects of surgery for the development of these lesions. Perspectives Due to further technical advances, fetal transesophageal echocardiography utilizing intravascular ultrasound catheters now o€ers great potential to become the monitoring tool of choice during open and fetoscopic fetal cardiac interventions. Further research using the conventional maternal transvaginal and transabdominal fetal imaging techniques, however, is required to achieve adequate selection of fetuses with semilunar valve obstructions for these procedures. Prospective serial fetal echocardiographic studies may aid in either con®rming or refuting retrospective data that changes in fetal cardiovascular growth and ¯ow patterns may predict whether an a€ected fetus will or will not be able to sustain a biventricular circulation postnatally, and in the latter case might become a candidate for fetal cardiac intervention. In contrast, hydropic fetuses with therapy-refractory arrhythmias can already be selected for potentially life-saving intervention by multimodal fetal echocardiography because of their high risk of prenatal demise from these disorders. Acknowledgments. This work was supported by a grant (Ko 1484/ 3-2) of the Deutsche Forschungsgemeinschaft, Bonn, Germany. I cordially thank Sibyl Storz of Karl Storz Incorporation, Tuttlingen, Germany, for philanthropically providing the endoscopic equipment for the development of fetoscopic fetal cardiac interventions. I appreciate the ultrasound equipment support by Acuson, NuÈrnberg, Germany.

Pediatric Cardiology Vol. 23, No. 3, 2002

8. 9.

10.

11. 12.

13. 14.

15.

16.

17.

References 18. 1. Allan LD, Crawford DC, Tynan MJ (1986) Pulmonary atresia in prenatal life. J Am Coll Cardiol 8(5):1131±1136 2. Allan LD, Maxwell DJ, Carminati M, Tynan MJ (1995) Survival after fetal aortic balloon valvoplasty. Ultrasound Obstet Gynecol 5(2):90±91 3. Allan LD, Sharland G, Tynan MJ (1989) The natural history of the hypoplastic left heart syndrome. Int J Cardiol 25(3):34l± 343 4. Anandakumar C, Biswas A, Chew SS, et al. (1996) Direct fetal therapy for hydrops secondary to congenital atrioventricular heart block. Obstet Gynecol 87(5 Pt 2):835±837 5. Assad RS, Aiello VD, Jatene MB, et al. (1995) Cryosurgical ablation of fetal atrioventricular node: new model to treat fetal malignant tachyarrhythmias. Ann Thorac Surg 60(6 Suppl):S629±S632 6. Assad RS, Jatene MB, Moreira LF, et al. (1994) Fetal heart block: a new experimental model to assess fetal pacing. Pacing Clin Electrophysiol 17(7):1256±1263 7. Azancot-Benisty A, Beni¯a JL, Matias A, De Crepy A, Madelenat P (1995) Constriction of the fetal ductus arteriosus

19.

20. 21.

22. 23.

during prenatal betamethasone therapy. Obstet Gynecol 85(5 Pt 2):874±876 Bealer JF, Raisanen J, Skarsgard ED, et al. (1995) The incidence and spectrum of neurological injury after open fetal surgery. J Pediatr Surg 30(8):1150±1154 Bical O, Gallix P, Donzeau-Gouge P, et al. (1985) Antenatal cardiac surgery. Creation of an experimental model of pulmonary stenosis in the fetus and repair in utero. Arch Mal Coeur Vaiss 78(3):445±449 Bical O, Gallix P, Toussaint M, et al. (1990) Intrauterine versus postnatal repair of created pulmonary artery stenosis in the lamb. Morphologic comparison. J Thorac Cardiovasc Surg 99(4):685±690 Bruch SW, Adzick NS, Reiss R, Harrison MR (1997) Prenatal therapy for pericardial teratomas. J Pediatr Surg 32(7):1113± 1115 Buyon JP, Swersky SH, Fox HE, Bierman FZ, Winchester RJ (1987) Intrauterine therapy for presumptive fetal myocarditis with acquired heart block due to systemic lupus erythematosus. Experience in a mother with a predominance of SS-B (La) antibodies. Arthritis Rheum 30(1):44±49 Carpenter R Jr., Strasburger JF, Garson A Jr., et al. (1986) Fetal ventricular pacing for hydrons secondary to complete antrioventricular block. J Am Coll Cardiol 8(6):1434±1436 Carreira PE, Gutierrez-Larraya F, Gomez-Reino JJ (1993) Successful intrauterine therapy with dexamethasone for fetal myocarditis and heart block in a woman with systemic lupus erythematosus (see comments]. J Rheumatol 20(7):1204± 1207 Chaoui R, Bollmann R, GoÈldner B, Rogalsky V (1994) Aortic balloon valvuloplasty in the human fetus under ultrasound guidance: a report of two cases. Ultrasound Obstet Gynecol 4:162A Danford DA, Cronican P (1992) Hypoplastic left heart syndrome: progression of left ventricular dilation and dysfunction to left ventricular hypoplasia in utero. Am Heart J 123:1712± 1713 Daubeney PE, Sharland GK, Cook AC, et al. (1998) Pulmonary atresia with intact ventricular septum: impact of fetal echocardiography on incidence at birth and postnatal outcome. UK and Eire Collaborative Study of Pulmonary Atresia with Intact Ventricular Septum. Circulation 98(6):562± 566 Deloof E, Devlieger H, Van Hoestenberghe R, et al. (1997) Management with a staged approach of the premature hydropic fetus due to complete congenital heart block Eur J Pediatr 156(7):521±523 Dunn JM, Weil SR, Russo P (1993) Management of fetal cardiac anomalies. In: Hanson MA, Spencer JAD, Rodeck CH (eds) Fetus and Neonate. Physiology and Clinical Applications Ð The Circulation. Cambridge University Press, Cambridge, pp.377±395 Fenton KN, Zinn HE, Heinemann MK, Liddicoat JR, Hanley FL (1994) Long-term survivors of fetal cardiac bypass in lambs. J Thorac Cardiovasc Surg 107(6):1423±1427 Gembruch U, Hansmann M, Redel DA, Bald R, Knop¯e G (1989) Fetal complete heart block: antenatal diagnosis, signi®cance and management. Eur J Obstet Gynecol Reprod Biol 3(1):9±22 Gibbs DL, Piecuch RE, Graf JL, et al. (1998) Neurodevelopmental outcome after open fetal surgery. J Pediatr Surg 33(8):1254±1256 Hanley FL (1994) Fetal cardiac surgery. Adv card Surg 5:47± 74

Kohl: Fetal Echocardiography

24. Hanley FL, Sade RM, Blackstone EH, et al. (1993) Outcomes in neonatal pulmonary atresia with intact ventricular septum. A multiinstitutional study. J Thorac Cardiovasc Surg 105(3):406±423 25. Harris JP, Alexson CG, Manning JA, Thompson HO (1993) Medical therapy for the hydropic fetus with congenital complete atrioventricular block. Am J Perinatol 10(3):217± 219 26. Hecher K, Campbell S, Doyle P, Harrington K, Nicolaides K (1995) Assessment of fetal compromise by Doppler ultrasound investigation of the fetal circulation. Arterial, intracardiac, and venous blood ¯ow velocity studies. Circulation 91(1):129± 138 27. Hornberger LK, Need L, Benacerraf BR (1996) Development of signi®cant left and right ventricular hypoplasia in the second and third trimester fetus. J Ultrasound Med 15(9):655±659 28. Hornberger LK, Sanders SP, Rein AJ, et al. (1995) Left heart obstructive lesions and left ventricular growth in the midtrimester fetus. A longitudinal study. Circulation 92(6):1531± 1538 29. Hornberger LK, Sanders SP, Sahn DJ, et al. (1995) In utero pulmonary artery and aortic growth and potential for progression of pulmonary out¯ow tract obstruction in tetralogy of Fallot. J Am Coll Cardiol 25(3):739±745 30. Huhta J (1996) Fetal echocardiography in the detection and management of fetal heart disease. In: Spitzer AR (ede.) Intensive Care of the Fetus and Neonate 1996. Mosby, St. Louis, pp. 772±786 31. Huhta JC, Cohen AW, Wood DC (1990) Premature constriction of the ductus arteriosus. J Am Soc Echocardiogr 3(1):30±34 32. Kikuchi Y, Shiraishi H, Igarashi H, Chunfeng L, Yanagisawa M (1995) Cardiac pacing in fetal lambs: intrauterine transvenous cardiac pacing for fetal complete heart block. Pacing Clin Electrophysiol 18(3 Pt 1):417±423 33. Kohl T, (1996) Fetal cardiac intervention in congenital heart disease. In: Chervenak FA, Kurjak A (eds.) Current Perspectives on the fetus as a Patient 1996. Parthenon, New York, pp. 297±315 34. Kohl T, Kirchhof PF, Gogarten W, et al. (1999) Fetoscopic transesophageal electrocardiography and stimulation in fetal sheep: a minimally invasive approach aimed at diagnosis and termination of therapy-refractory supraventricular tachycardias in human fetuses. Circulation 100(7): 772±776 35. Kohl T, McElhinney DB, Farrel J, et al. (1999) Impact of fetoscopic versus open fetal surgery on fetoplacental blood ¯ow and outcome in human fetuses. Eur Heart J 20:644 A 36. Kohl T, Sharland G, Allan LD, et al. (2000) World experience of percutaneous ultrasound-guided balloon valvuloplasty in human fetuses with severe aortic valve obstruction. Am J Cardiol 85(10):1230±1233 37. Kohl T, Stelnicki EJ, VanderWall KJ, et al. (1996) Transesophageal echocardiography in fetal sheep. A monitoring tool for open and fetoscopic cardiac procedures. Surg Endosc 10 (8):820±824 38. Kohl T, Strumper D, Witteler R, et al. (2000) Fetoscopic direct fetal cardiac access in sheep: an important experimental milestone along the route to human fetal cardiac intervention. Circulation 102(14):1602±1604 39. Kohl T, Suda K, Reckers J, et al. (1999) Fetal transesophageal echocardiography utilizing a 10-F, 10-MHz intravascular ultrasound catheter Ð comparison with conventional maternal

345

40.

41.

42.

43.

44.

45. 46. 47. 48. 49.

50.

51.

52. 53. 54. 55.

56. 57.

transabdominal fetal echocardiography in sheep. Ultrasound Med Biol 25(6):939±946 Kohl T, Szabo Z, Suda K, et al. (1997) Fetoscopic and open transumbilical fetal cardiac catheterization in sheep. Potential approaches for human fetal cardiac intervention. Circulation 95(4):1048±1053 Kohl T, Szabo Z, Suda K, et al. (1997) Percutaneous fetal access and uterine closure for fetoscopic surgery. Lessons learned from 16 consecutive procedures in pregnant sheep. Surg Endos 11(8):819±824 Kohl T, Szabo Z, VanderWall KJ, et al. (1996) Experimental fetal transesophageal and intracardiac echocardiography utilizing intravascular ultrasound technology. Am J Cardiol 77(10):899±903 Kohl T, Witteler R, StruÈmper D, et al. (2000) Operative techniques and strategies for minimally invasive fetoscopic fetal cardiac interventions in sheep. Surg Endosc 14:424± 430 Liddicoat JR, Klein JR, Reddy VM, et al. Hemodynamic effects of chronic prenatal ventricular pacing for the treatment of complete atrioventricular block. Circulation 96(3):1025± 1030 Lopes LM, Cha SC, Kajita LJ, et al. (1996) Balloon dilatation of the aortic valve in the fetus. A case report. Fetal Diagn Ther 11(4):296±300 Maxwell D, Allan L, Tynan MJ (1991) Balloon dilatation of the aortic valve in the fetus: a report of two cases. Br Heart J 65(5):256±258 McCa€rey FM, Sherman FS (1997) Prenatal diagnosis of severe aortic stenosis. Pediatr Cardiol 18(4):276±281 Naheed ZJ, Strasburger JF, Deal BJ, Benson DW Jr., Gidding SS (1996) Fetal tachycardia: mechanisms and predictors of hydrops fetalis. J Am Coll Cardiol 27(7):1736±1740 Rasanen J, Jouppila P (1995) Fetal cardiac function and ductus arteriosus during indomethacin and sulindac therapy for threatened preterm labor: a randomized study. Am J obstet Gynecol 173(1):20±25 Reddy VM, Liddicoat JR, Klein JR, Wampler RK, Hanley FL (1996) Long-term outcome after fetal cardiac bypass: fetal survival to full term and organ abnormalities. J Thorac Cardiovasc Surg 111(3):536±544 Rosenthal D, Druzin M, Chin C, Dubin A (1998) A new therapeutic approach to the fetus with congenital complete heart block: preemptive, targeted therapy with dexamethasone. Obstet Gynecol 92(4 Pt 2):689±691 Saenger JS, Mayer DC, D'Angelo LJ, Manci EA (1992) Ductus-dependent fetal cardiac defects contraindicate indomethacin tocolysis. J Perinatol 12(1):41±47 Scagliotti D, Shimokochi DD, Pringle KC (1987) Permanent cardiac pacemaker implant in the fetal lamb. Pacing Clin Electrophysiol 10(6):1253±1261 Schade R, De Bryun D, Hutter P, et al. (1998) Neurological morbidity after fetal supraventricular tachyarrhythmia. Cardiol Young 9(Suppl 1):1A Schmidt KG, Ulmer HE, Silverman NH, Kleinman CS, Copel JA (1991) Perinatal outcome of fetal complete atrioventricular block: a multicenter experience. J Am Coll Cardiol 17(6):1360± 1366 Simpson JM, Sharland GK (1998) Fetal tachycardias: management and outcome of 127 consecutive cases. Heart 79(6):576±581 Simpson JM, Sharland GK (1997) Natural history and outcome of aortic stenosis diagnosed prenatally. Heart 77(3):205± 210

346

58. Sonesson SE, Winberg P, Lidegran M, Westgren M (1996) Foetal Supraventricular tachycardia and cerebral complications. Acta Paediatr 85(10):1249±1252 59. Stevens DC, Hilliard JK, Schreiner RL, et al. (1982) Supraventricular tachycardia with edema, ascites, and hydrops in fetal sheep. Am J Obstet Gynecol 142:316±322 60. Todros T, Presbitero P, Gaglioti P, Demarie P (1988) Pulmonary stenosis with intact ventricular septum: documentation of development of the lesion echocardiographically during fetal life. Int J Cardiol 8:1131± 1136

Pediatric Cardiology Vol. 23, No. 3, 2002

61. Turley K, Vlahakes GJ, Harrison MR, et al. (1982) Intrauterine cardiothoracic surgery: the fetal lamb model. Ann Thorac Surg 34(4):422±426 62. Walkinshaw SA, Welch CR, McCormack J, Walsh K (1994) In utero pacing for fetal congenital heart block. Fetal Diagn Ther 9(3):183±185 63. Wright JGC, Skinner JR, StuÈmper O (1994) Radiofrequency assisted pulmonary valvotomy in a fetus with pulmonary atresia and intact ventricular septum. Paper presented at the 13th Meeting of the International Fetal Medicine and Surgery Society, Antwerp, Belgium.