Minimal Incision Aortic Surgery (MIAS) William D. Turnipseed, MD, Sandra C. Carr, MD, John R. Hoch, MD, and Jon R. Cohen, MD, Madison, Wisconsin

This study evaluates the clinical and economic impact of using less extensive minimal invasive aortic surgery (MIAS) for elective treatment of infrarenal aortic aneurysms (AAA) and aortoiliac occlusive disease (AIOD) in two independent surgical departments. Surgeons from two institutions conducted a prospective consecutive, nonrandomized analysis of MIAS electively performed in 80 patients. MIAS outcomes were compared with 80 consecutive elective standard open aortic procedures (40 from each institution), which were performed during the same time period. Cost analyses for MIAS and standard open repair were performed at each institution. Our results indicated that MIAS is as safe as standard open repair, is more cost-effective, and has significantly shorter hospital stays than with standard open repair.

INTRODUCTION Contemporary efforts to improve outcomes for standard open abdominal surgery by use of alternative imaging and laparoscopic techniques or less extreme methods of aortic exposure have demonstrated the capacity to shorten hospital stay, reduce morbidity, and facilitate patient recovery. This report summarizes the early clinical experience of two independent vascular surgery departments using minimal invasive aortic surgery (MIAS) as an alternative to standard open aortic surgery in the elective treatment of patients with infrarenal aortic aneurysms (AAA) and/or aortoiliac occlusive disease (AIOD).

MATERIAL AND METHODS Consecutive elective MIAS cases performed at the University of Wisconsin (UW) and at the Long

Correspondence to: W.D. Turnipseed, MD, 600 Highland Avenue, G5/325, Madison, WI 53792, USA, E-mail: [email protected]. Ann Vasc Surg 2003; 17: 180-184 DOI: 10.1007/s10016-001-0197-2  Annals of Vascular Surgery Inc. Published online: 14 March 2003 180

Island Jewish Medical Center in Long Island, New York, between June of 1999 and June 2000 were included in this study. Critical components of the MIAS procedure include a small midline periumbilical incision ( £ 10 cm), intracavitary retraction of small bowel, and traditional hand-sewn vascular anastomoses. The abdominal incision length and configuration are based on the presence or absence of iliac disease involvement. The periumbilical incision is extended cephalad for aneurysms limited to the infrarenal aorta for easier exposure of the neck and caudad if there is disease extension into the iliac arteries. When aneurysm or occlusive disease extends beyond the distal common iliac artery into the pelvis, the common femoral arteries are used as target vessels for distal anastomoses. There was no attempt to select MIAS candidates based on risk category, body size, aneurysm size, or previous abdominal surgery. Exclusionary criteria included acutely ruptured aneurysms, suprarenal aneurysms, and patients with aneurysm or aortoiliac occlusive disease that required concomitant renal, mesenteric, or infrainguinal vascular reconstruction. Patients who were candidates for standard open, MIAS, or endograft repair of abdominal aneurysms had each procedure explained and patients were allowed to select one over the other. With few exceptions patients deferred to the sur-

Vol. 17, No. 2, 2003

geon’s preference. Eighty consecutive elective standard open or retroperitoneal aortic surgical procedures for aneurysm or iliac occlusive disease (40 from each institution) performed during the same time period were used for comparison with MIAS. The MIAS exposure technique and standard open aortic repairs at the two surgical institutions were performed in similar fashion. Postoperative management was standardized in each institution by care pathways. Demographics for the MIAS and standard repair groups were compared. Outcome analysis included operating room (OR) time, intraoperative fluid resuscitation, transfusion requirement, intensive care unit (ICU) stay, time to general dietary feeding, hospital length of stay (LOS), as well as morbidity and mortality (less than 30 days). MIAS and standard open repair patients were extubated in the operating room after surgery. At the University of Wisconsin, epidural catheters were used in less than 20% of all MIAS procedures and in 85% of all standard open procedures. At the Long Island Jewish Medical Center, 85% of both the MIAS and standard open procedures were treated with epidural anesthesia. This was the only significant difference in patient management between the two surgical groups. Autotransfusion was used in MIAS and standard open cases at both institutions. Nasogastric tubes for MIAS patients were removed in recovery and on the second or third postoperative day for open repairs depending on return of bowel sounds and reduced nasogastric drainage. Patient demographics were compared with a Fisher’s exact test. Operative parameters were compared with a two-tailed Student t-test, performed on NSTAT software. Average time interval to regular diet, ICU stay, and hospital LOS were compared using the Wilcoxan rank/sum test. Other analyses were performed using SAS statistical software (SAS Institute, Inc.). For the data analyzed by Student t-test, p-values £ 0.05 were considered to be significant. Medical reimbursement at the University of Wisconsin Hospital recorded as total standard cost per patient for the principle diagnosis of unruptured abdominal aortic aneurysm (ICD-9 #441.4) was calculated based on metric mean LOS, which is a statistically adjusted value for all cases of a given diagnosis related group (DRG) (110, 111 = AAA with or without complications). Cost data were provided by the University Health System Consortium, which is the national agency that provides cost-related data for all major university health care providers in the United States. Information from

Minimal incision aortic surgery 181

Table I. Prospective nonrandomized study demographics Factor

MIAS

Standard open repair

Patients (n) Age (years) Male (n) Female (n) AAA size (cm) Median Body weight (kg) Median Comorbid risks Coronary artery disease High blood pressure Adult-onset diabets Chronic obstructive pulmonary disease Previous abdominal surgery

80 65 + 10 57 23 4.0-8.2 5.7 52-110 87.5

80 67 + 12 61 19 4.2-9.0 6.2 60-105 85.0

52% 70% 9% 26%

33% 67% 10% 24%

20%

24%

p ‡ 0.130.

the Hospital Cost Report Information System (HCRIS) was used to estimate cost from reported hospital charges. Cost determination (total direct and indirect cost) for patient populations is built from costs of individual encounters; cost of an individual patient encounter is estimated from the sum of costs assigned to each billable item or service that the patient received during the stay. The following cost measures are employed when reporting on populations, individual cases, or specific services. Total actual cost is an approximate measure of fully allocated costs of providing services to patients including all direct costs of care (patient care staff, disposable medical supplies, drugs, medical equipment) plus an allocation of indirect or support costs (administration, medical records, information systems, facility cost such as building depreciation and utilities, etc.). Actual direct cost plus actual indirect cost equals actual total cost. Net hospital revenue is a product of the total hospital reimbursement minus the total actual cost per patient. A less comprehensive cost analysis of MIAS and standard open surgical procedures performed at the Long Island Jewish Hospital was based on the mean sum of service charges from the operating room, intensive care, and floor care units.

RESULTS There was no difference between the MIAS and standard open repair groups with regard to age, gender distribution, aneurysm size, or body weight. Comorbid risk factors were comparable between

182 Turnipseed et al.

Annals of Vascular Surgery

Table II. Prospective nonrandomized study: combined resultsa Factor

MIAS

Standard open repair

Intraoperative OR time (min) IV fluid (cc) Transfusion (units) Postoperative ICU stay (days) Postoperative general diet (days)b Total LOS (days)b Morbidity Major complications Ileus Increased VENT. assist Wound complications Leg embolus Mortality

157 ± 37 4375 ± 1583 1.1 ± 1.3 1.5 ± 1.2 3.4 ± 1.2 4.8 ± 1.4 (3.4 ± 1.0) 12 (15%) 5 2 1 1 1 1 (1.3%)c

200 ± 44 5304 ± 2433 1.7 ± 2.0 1.8 ± 1.5 5.4 ± 4.7 7.7 ± 3.4 (5.8 + 1.2) 21 (26%) 12 6 5 1 — 2 (2.5%)c

a

From the University of Wisconsin and Long Island Jewish Hospital. p < 0.001. c Myocardial infarct. b

the two groups (Table I). Sixty tube grafts were performed in the MIAS group (75%) and 64 tube grafts (80%) were performed in the standard open repair group. Of the 20 bifurcation grafts performed in the MIAS group, 5 had distal anastomoses performed at the common iliac level and 15 had distal anastomoses done at the common femoral level. Of the 16 bifurcation grafts performed in the standard open repair group, 9 had distal anastomoses performed in the iliac artery system and 7 had common femoral level distal anastomoses. Of the 30 groin wounds in the MIAS group, 1 developed a seroma (3.3%) and 1 of 14 groin wounds developed a hematoma (7%) in the standard repair group. Although OR time tended to be shorter for MIAS than for standard open repair, the differences are not statistically significant. There was no difference in the intraoperative fluid use or transfusion requirement between the two groups. Autotransfusion was used in both MIAS and standard open repairs. ICU stay, return to general dietary feeding, and hospital LOS were significantly shorter for the MIAS procedure. This difference was even more apparent when MIAS was performed in good-risk patients without a postoperative complication. Overall, morbidity and mortality rates were not significantly different (Table II). At the University of Wisconsin, OR costs were nearly equivalent for both treatment groups. Total standard cost was lower and net hospital reimbursement (profit margin) was higher for the MIAS procedure than for standard open repair (Table III). The major reason for cost saving with MIAS was lower ICU utilization and shorter hospital stay. At Long Island Jewish Hospital mean service charges were signif-

icantly lower for MIAS cases when compared with standard open repair.

DISCUSSION Although there is little question that endovascular stent grafts will prevail as an option for the repair of aortic aneurysms, midterm performance standards have not matched expectations and give rise to concern about the durability of currently available device technology.1-4 Certainly, the early results of endograft clinical trials have demonstrated morbidity and mortality outcomes that were as good or even better than standard open repair.5-7 However, the fact that both supported and unsupported endograft designs are beginning to demonstrate specific failure patterns between 18 and 24 months of follow-up has led to the conclusion that long-term, post-procedural CT surveillance is absolutely mandatory.4,8,9 Unfortunately, human nature and our mobile society do not readily lend themselves to the dictum of comprehensive long-term followup. This becomes an issue when recent reports suggest that at least 25% of all implanted endografts can be classified as failures by the SVS/ISCVS reporting standards. The purpose of this collaborative study is to demonstrate that the skills required to perform alternative methods of aortic exposure and repair such as the MIAS technique can be mastered and duplicated in comparable clinical settings. We also wanted to show that such techniques can improve patient recovery and hence cost efficiency compared to results from traditional abdominal vascular surgery performed within the same

Vol. 17, No. 2, 2003

Minimal incision aortic surgery 183

Table III. Prospective nonrandomized study: fiscal analysis Factor

University of Wisconsin OR cost ICU cost Total hospital cost Total reimbursement Net revenue Long Island Jewish Hospital OR charges ICU charges Floor care charges Total hospital charges

MIAS ($ U.S.)

Standard open repair ($ U.S.)

3068 3134 12,585 21,030 +8445

3805 3132 18,445 23,434 +4989

7225 ± 831 9477 ± 4626 5990 ± 1309 22,692 ± 5246

10,195 28,629 23,501 62,324

± ± ± ±

3032 42,447 27,468 70,618

p = 0.0326.

institution.10,11 In selected circumstances this technique may be an attractive alternative to endoaortic stent grafting. The concept of using less extensive methods of aortic exposure and direct graft repair is gaining credibility among vascular surgeons. Numerous clinical reports suggest that small midline or retroperitoneal incisions used in concert with direct aortic grafting can achieve quicker patient mobilization and easier pain control while reducing postoperative ileus and hospital LOS.12-15 We prefer the midline minilaparotomy because it does not traverse muscle and as a consequence, wound discomfort and complications are reduced. The midline incision allows for easy extension and better visualization when necessary, and it makes access to both common iliac arteries much easier than with retroperitoneal exposure. The midline minilaparotomy can be used for tube graft or bifurcation graft repair. Although we utilize groin incisions more frequently with the MIAS procedure (15/80 = 19%), 100% of endograft repairs require bilateral groin incisions with subsequent risk exposure. The minilaparotomy technique is ideal for use in younger good-risk patients because they can be quickly mobilized, fed earlier, and sent home quicker. This technique is more difficult in very obese patients, in patients with very large aneurysms (>10 cm), and in patients with truly hostile surgical abdomens. Slightly larger incisions can be used in obese patients (10-15 cm) or in patients with very large aneurysms, preserving the clinical advantage of ileus prevention as long as intracavitary retraction of small bowel can be achieved. Twenty-five percent of the MIAS patients in this series had had at least one previous major abdominal procedure before vascular surgery, which emphasizes the fact that this technique can be performed in lieu of previous operations. The sur-

geons from Long Island Jewish Hospital only included patients with aneurysms that could be repaired with tube grafts, while the UW group included patients with aneurysms and/or AIOD that required tube or bifurcation graft repairs. Although epidural anesthesia was used for MIAS and standard open repair in both clinical practices, it has been less commonly employed in the UW experience. The major concern has been that mobilization and Foley catheter removal are slowed with its use. This may explain the minor LOS difference between the UW and Long Island Jewish Medical Center patients (4.8 vs. 5.2 days). Some reports in the literature suggest that hospital LOS for standard open repair can be significantly reduced by pathway criteria and preoperative patient conditioning alone. We were able to shorten LOS from 9.2 to 7.7 days by using pathway criteria, but we could not achieve more significant reductions in LOS until introduction of the MIAS exposure technique. Outcomes similar to those documented in this report can be achieved by using standard retroperitoneal aortic exposure and repair. Although extended retroperitoneal exposures can be used for treatment of complex juxtarenal and periviseral aortic aneurysms, many surgeons reserve this exposure for instances when the aneurysm is limited to the infrarenal aorta and when tube graft repair is possible. The advantage of using the small midline incision is that patient selection is less critical. More recently, laparoscopic and video-assisted methods of aortic exposure and repair have been perfected. Major advantages of MIAS over these techniques are that laparoscopic skills and equipment are not required, anastomoses are easier to perform, and operating times are shorter. The major advantages for MIAS over endovascular stent graft repair are that there are fewer patient rejections because of

184 Turnipseed et al.

Annals of Vascular Surgery

Table IV. Less invasive aortic repair versus standard repair Patients (n)

MIAS 80 Laparoscopic assisted operative repair AAA (Cohen) 20 AIOD (Alimi) 6 AAA (Alimi) 13 Laparoscopic assisted (closed) Castronuovo 60 Alimi 7 Stent endograft (bifurcated) Aneurex 190 Ancure 268 University of Wisconsin 40 Standard open repair 80

OR time (min)

ICU stay (days)

Regular diet (days)

LOS (days)

Mortality (%)

Morbidity (%)

157

1.5

3.4

4.8 [3.4]

1.3

15

246 235 393

2.2 NS NS

3.0 3.0 NS

5.8 3.3 9 [6]

0 0 7

23 16 28

462 351

2.4 NS

NS NS

6.5 10.7 [7.2]

0 0

23 29

186 211 257 200

0.9 0.5 1.0 3.5

1.4 2.0 2.0 5.4

3.4 3 4.3 7.7 [5.8]

2 2.6 3 2.5

17a 28.7b 16 26

NS, not stated. Numbers in brackets indicate complication-free recovery. a Phase II trial only. No recent data available. b Composite safety risk.

vascular morphology or limited device availability, advanced catheter management skills and expensive imaging and device delivery accessories are not required, and less economic stress is exerted on hospital resources. Common characteristics of all of the less invasive methods of treating aortic disease include reduced postoperative morbidity and shortened hospital stay (Table IV). Mortality statistics, however, have not been significantly affected by use of any of these methods. MIAS appears to be more cost-efficient than standard open aortic repair and the other less extensive aortic exposure repair techniques because shorter hospital stay can be achieved without expensive imaging or device delivery costs. In summary, MIAS should be considered for treatment of patients with aortic disease because it is safe and cost-efficient, recovery time is reduced, and quality outcomes associated with standard open repair are maintained. REFERENCES 1. Prinssen M, Wever JJ, Mali WPTM, Eikelboom BC, Blankensteijn JD. Concerns for the durability of the proximal endograft fixation from a 2-year and 3-year longitudinal computed tomography angiography study. J Vasc Surg 2001;33:64-69. 2. Bush RL, Lumsden AB, Dodson TF, et al. Mid-term results after endovascular repair of the abdominal aortic aneurysm. J Vasc Surg 2001;33:70-76. 3. Ho¨ lzenbein TJ, Kretschmer G, Thurnher S, et al. Mid-term durability of abdominal aortic aneurysm endograft repair: a word of caution. J Vasc Surg 2001;33:46-54.

4. Jacobs TS, Won J, Gravereaux EC, et al. Mechanical failure of prosthetic human implants: a 10-year experience with aortic stent graft devices. J Vasc Surg 2003;37: 16-26. 5. May J, White GH, Waugh R, et al. Improved survival after endoluminal repair with second-generation prostheses compared with open repair in the treatment of abdominal aortic aneurysms: a 5-year concurrent comparison using life-table method. J Vasc Surg 2001;33:21-26. 6. Zarins CK, White RA, Moll FL, et al. The AneuRx stent graft: four-year results and worldwide experience 2000. J Vasc Surg 2001;33:135-145. 7. Makaroun MS. The Ancure endografting system: an update. J Vasc Surg 2001;33:129-134. 8. Carpenter JP, Neschis DG, Fairman RM, et al. Failure of endovascular abdominal aortic aneurysm graft limbs. J Vasc Surg 2001;33:296-303. 9. Beebe HG, Cronenwett JL, Katzen BT, Brewster DC, Green RM. Results of an aortic endograft trial: impact of device failure beyond 12 months. J Vasc Surg 2001;33:55-63. 10. Turnipseed WD. A less-invasive minilaparotomy technique for repair of aortic aneurysm and occlusive disease. J Vasc Surg 2001;33:431-434. 11. Cerveira JJ, Halpern VJ, Faust G, Cohen JR. Minimal incision abdominal aortic aneurysm repair. J Vasc Surg 1999;30:977-984. 12. Nakajima T, Kawazoe K, Komoda K, Sasaki T, Ohsawa S, Kamada T. Midline retroperitoneal versus midline transperitoneal approach for abdominal aortic aneurysm repair. J Vasc Surg 2000;32:219-223. 13. Castronuovo JJ, James KV, Resnikoff M, McLean ER, Edoga JK. Laparoscopic-assisted abdominal aortic aneurysmectomy. J Vasc Surg 2000;32:224-233. 14. Barbera L, Mumme A, Metin S, Zumtobel V, Kemen M. Operative results and outcome of twenty-four totally laparoscopic vascular procedures for aortoiliac occlusive disease. J Vasc Surg 1998;28:136-142. 15. Kline RG, D’Angelo AJ, Chen MHM, Halpern VJ, Cohen JR. Laparoscopically assisted abdominal aortic aneurysm repair: first 20 cases. J Vasc Surg 1998;27:81-88.