Pelvic Blood Flow Following Aortobifemoral Bypass with Proximal End-to-Side Anastomosis

Pelvic Blood Flow Following Aortobifemoral Bypass with Proximal End-to-Side Anastomosis

Original articles Pelvic Blood Flow Following Aortobifemoral Bypass with Proximal End-to-Side Anastomosis Susan E. O'Connor, MD, Daniel B. Walsh, MD,...

2MB Sizes 0 Downloads 35 Views

Original articles

Pelvic Blood Flow Following Aortobifemoral Bypass with Proximal End-to-Side Anastomosis Susan E. O'Connor, MD, Daniel B. Walsh, MD, Robert M. Zwolak, MD, PhD, Joseph R. Schneider, MD, PhD, Jack L. Cronenwett, MD, Lebanon, New Hampshire

Nine patients with end-to-side aortobifemoral bypasses were studied in the first year after surgery using color duplex imaging to determine the source of pelvic blood flow. No patient had clinical evidence of postoperative pelvic ischemia. Six of nine patients were found to have occluded distal aortas by duplex studies performed at a mean of 4.4 months postoperatively (range 0.8-8.2 months). Of those six patients, postoperative duplex examination demonstrated two with no common or external lilac blood flow, two with bilateral retrograde external lilac flow, and two with unilateral retrograde external lilac flow. Of the three patients with patent distal aortas, two had no demonstrable external lilac blood flow, while the third had continued antegrade flow through one external lilac and retrograde flow through the other. Analysis of preoperative arteriograms failed to reveal accurate predictors of postoperative distal aortic patency or retrograde lilac blood flow. Despite the preoperative assumption that prograde common lilac artery blood was required to prevent pelvic ischemia, distal aortic patency was maintained in only three of nine patients. In the six patients with prograde lilac blood flow, no ischemic symptoms were present, including two patients with complete absence of antegrade aortic or retrograde external lilac blood flow. Our observations indicate that assumptions which underlie the decision to perform end-to-side aortic anastomoses are often not borne out in the months following aortobifemoral bypass. (Ann Vasc Surg 1992;6:493-498). KEY WORDS:

Aortobifemoral bypass; aortoiliac disease; lilac arteries; pelvic ische-

mia; aorta.

Aortobifemoral (ABF) bypass is the operation of choice for severe aortoiliac occlusive disease [1]. At our institution, end-to-end aortic anastomoses are From the Section of Vascular Surgery, DartmouthHitchcock Medical Center, Lebanon, New Hampshire. Reprint requests: Daniel B. Watsh, MD, Section of Vascular Surgery, Dartmouth-Hitchcock Medical Center, One Medical Center Drive, Lebanon, New Hampshire 03756.

preferred. However, end-to-side aortic grafts have been employed selectively in one-third of our patients in the presence of severe external iliac disease with patent common and internal iliac arteries to avoid inadequate pelvic blood flow [2-4]. The potential advantage of an end-to-side anastomosis presumes continued patency of the distal aorta with antegrade aortoiliac blood flow for some period of time after bypass grafting.

493

494

PELVIC BLOOD F L O W AFTER END-TO-SIDE AORTOBIFEMORAL B Y P A S S

Despite this practice, there has been little firm data substantiating the assumption. In fact, the early natural history of the bypassed native aortoiliac system is unknown. One series has documented a 92% occlusion rate by CT scan at 5-10 years, postoperatively [5]. We undertook the current study to identify the source of pelvic blood flow in the first year following end-to-side aortobifemoral bypass surgery to test the assumptions that underlie the performance of this type of proximal aortic anastomosis. Specifically, is distal aortic patency maintained after endto-side aortic anastomosis? In the presence of native aortic occlusion, is there retrograde filling of the iliac system? Can arteriography be used to predict postoperative distal aortic blood flow, and does the presence of patent inferior mesenteric artery affect distal aortic patency?

MATERIALS AND METHODS Eleven patients underwent end-to-side aortobifemoral bypass surgery at the Dartmouth-Hitchcock Medical Center during the 12 month period March 1990 to February 1991. Since most of our patients (68%) underwent end-to-end anastomoses, these 11 patients represented a subgroup in whom proximal aortic anastomoses were performed in an end-to-side manner to maximize pelvic blood flow. Prior to closure, distal aortic perfusion was confirmed in all patients by palpation of distal aortic and common iliac artery pulses. Two patients were lost to follow-up after operation and were excluded from further study. Of the nine patients remaining, six were male and three were female with a mean age at surgery of 58 years (range 45-70 years). All of these patients had a history of tobacco use and eight patients continued to smoke after surgery. Concurrent medical problems included hypertension (5), history of myocardial infarction (4) and diabetes mellitus (1). These nine patients were evaluated by color duplex scanning (Quad I Angiodynography*) postoperatively to determine patency and direction of blood flow in the native aortoiliac system. Scans were performed at various times, postoperatively, with the earliest at three weeks. Earlier scans were impaired by incisional pain, staples, and the presence of bowel gas. Experienced registered vascular laboratory technologists, whose accuracy in intraabdominal and femoral arterial imaging has been confirmed with compression arteriography, performed all duplex exams [6,7]. A 3.0 mHz transducer was used to image the aortic graft anastomosis. Using a 5.0 mHz transducer, the femoral anastomoses were identi*Quantum, Issaquah, Washington.

ANNALS OF VASCULAR SURGERY

fled and direction of blood flow was determined in the external iliac artery proximal to the anastomosis. Preoperative arteriograms were reviewed to measure the severity of iliac arterial disease present at the time of operation. Based on the assumption that iliac artery outflow (especially common and internal) would determine postoperative distal aortic patency, we attempted to develop criteria to predict which aortas would occlude. Common iliac, internal iliac, and external iliac artery occlusive disease was graded using SVS/ISCVS criteria as 0 (minimal disease), 1.0 (20% to 49% stenosed), 2.0 (50% to 99% stenosed), 2.5 (occlusion of the vessel for less than half its length), or 3.0 (occlusion of the vessel for greater than half of its length) [8]. Since the contribution of each iliac artery to maintenance of distal aortic patency may differ, a weighting coefficient was assigned to each artery. Coefficients ranged from 1 to indicate least critical to 3 for most critical. Thus, the sum of the product of each vessel's occlusion grade and its weighting coefficient were used to derive an outflow score for each iliac limb which would describe the extent and location of a given patient's iliac occlusive disease. We calculated a total iliac outflow score based on all possible variations of disease and weighting coefficients for the common, internal, and external iliac arteries. In addition, the most and least diseased sides were analyzed separately. Status of the inferior mesenteric artery (IMA) was noted and compared to native aortic outcome to assess the potential correlation between preoperative IMA and patency and maintenance of distal aortic patency.

RESULTS No patients exhibited clinical evidence of pelvic or bowel ischemia in the perioperative period or during follow-up. The three patients with patent distal aortas were studied at 3.4, 5.3, and 8.2 months, postoperatively. Figure I illustrates the typical duplex image of the proximal aorto-graft anastomosis with a patent distal aorta. Of these three patients, two had no demonstrable external iliac blood flow (three of these four arteries were occluded at some level preoperatively). The third patient had continued antegrade flow through a preoperatively patent external iliac artery and retrograde flow in a proximally occluded external iliac artery. On preoperative arteriogram, two of these three patients had bilaterally patent internal iliac arteries. The third patient had bilaterally occluded internal iliac arteries and an occluded IMA.

VOLUME 6 N o 6 - 1992

PELVIC BLOOD F L O W AFTER END-TO-SIDE AORTOBIFEMORAL BYPASS

Fig. 1. Color duplex image of proximal aortic anastomosis following aortobifemoral bypass demonstrating antegrade blood flow in graft and distal aorta.

Six of nine patients (three men, three women) were found to have occluded distal aortas by duplex studies performed at a mean of 4.4 months after surgery (range 0.8-8.2 months). Figure 2 illustrates the typical color duplex appearance of the proximal aortic-graft anastomosis with an occluded distal aorta. Occluded aortas were discovered as early as three weeks preoperatively in two patients. The four remaining occluded aortas were detected during the initial duplex study performed postoperatively at 2.5, 5, 5.5 and 7.1 months. Among these six patients with distal aortic occlusion, duplex examination demonstrated two patients with no external iliac blood flow; two patients with bilateral retrograde external iliac flow; and two patients with unilateral, retrograde iliac blood flow. There were eight external iliac arteries in this group of six patients with preoperative arteriographic evidence of occlusion somewhere along their course. By postoperative duplex, four had retrograde flow suggestive of continued retrograde perfusion to the point of occlusion. One initially patent external iliac artery was occluded at the point of femoral anastomosis on postoperative exam, while two others had retrograde flow. All six patients with occluded distal aortas postoperatively had at least one patent internal iliac artery preoperatively.

495

Fig. 2. Color duplex image of proximal aortic anastomosis following aortobifemoral bypass demonstrating occlusion of distal aorta with antegrade blood flow in graft.

Extensive evaluation of preoperative arteriographic data failed to reveal accurate predictors of postoperative distal aortic patency. All possible combinations of weighting coefficients for the iliac arteries were examined; however, no combination could accurately separate patients whose distal aortas occluded from patients whose aortas remained patent. The formula which best separated these groups emphasized common and external iliac artery patency on the "least diseased" side. Outflow score = [(common iliac grade x 3) + (internal iliac grade x l) + (external iliac grade x 2)]. If we had applied this predictive system clinically and considered all patients with one iliac artery outflow score of >4 for end-to-end anastomosis (under the supposition that they would proceed to aortic occlusion), all of our patients who then went on to aortic occlusion would have been correctly identified. However, one patient who maintained aortic patency would have been incorrectly classified and possibly relegated to the wrong operation. Preoperative arteriograms from two patients are shown in Figures 3 and 4 ("least diseased" outflow scores were calculated at 6 and 8, respectively). Although these patients have similar patterns of disease and thus, similar outflow scores, Patient 1 (Fig. 3) maintained distal aortic patency postoperatively

496

PELVIC BLOOD F L O W AFTER END-TO-SIDE AORTOB1FEMORAL B Y P A S S

ANNALS OF VASCULAR SURGERY

Fig. 3. Preoperative pelvic arteriogram (Patient 1) with outflow score of 6. Patent distal aorta documented postoperatively at 8.2 months,

Fig. 4. Preoperative pelvic arteriogram (Patient 4) with outflow score of 8. Occluded distal aorta documented at 5 months.

while Patient 4 (Fig. 4) suffered distal aortic occlusion. Seven of nine patients had a patent IMA documented on preoperative arteriogram or during operation. Five of these patients had occluded distal aortas postoperatively. Of the two patients with occluded IMAs preoperatively, one retained distal aortic patency and the other proceeded to occlusion of the distal native aorta.

cated comparable patency rates for the two techniques. We perform the aortic end-to-side anastomosis only when preoperative arteriography indicates a pattern of disease that might critically limit pelvic blood flow after an end-to-end aortic anastomosis, namely, significantly more external than common iliac artery disease, with a patent internal iliac artery. In this study, we found that the assumption which directed our choice of aortic anastomosis, i.e., the need for continued antegrade aortic blood flow, was often not borne out over the months following aortic bypass. Sixty-seven percent of the patients studied were found to have occluded distal aortas despite an end-to-side aortobifemoral bypass performed to preserve distal aortic flow. The mean date of postoperative duplex examination for those patients with occluded distal aortas was 3.6 months compared to a mean study date of 5.6 months for patients with patent aortas. Thus it appears likely that two distinct populations exist for postoperative end-to-side ABF graft recipients: those with sufficient common iliac artery outflow to maintain distal

DISCUSSION Several studies have shown significantly increased graft patency for aortobifemoral bypass performed with end-to-end aortic anastomosis. Pierce and associates found 100% end-to-end ABF graft patency at 3.5 years contrasted with 91% for grafts utilizing end-to-side anastomoses [9]. Brewster and colleagues reported a five-year failure rate of 44% for end-to-side ABF versus 15% for end-toend grafts [1]. Other reports from Rutherford [10], Melliere [11], and Ameti [12], however, have indi-

VOLUME 6 No 6 - 1992

PELVIC BLOOD F L O W AFTER END-TO-SIDE AORTOBIFEMORAL B Y P A S S

aortic patency and those without. In two patients, thrombosis of the native distal aorta was detected as early as three weeks after surgery with no evidence of symptomatic pelvic ischemia. Although aortic occlusions were documented over a wide range of postoperative intervals (one to seven months), the presumption that continued aortic patency beyond this early period is necessary to prevent pelvic ischemia appears unwarranted. However, it is still possible that a critical period of distal aortic patency exists, but occurs in the immediate postoperative period, which was not evaluated in this study. Pelvic ischemia is considered a very rare but devastating complication of aortobifemoral bypass. Given the relative infrequency of this complication and the small number of patients in this study, the absence of pelvic ischemia in this study population is not surprising. However, two patients with conditions thought most likely to lead to pelvic ischemia, i.e., no antegrade aortic inflow and no retrograde iliac pelvic flow, showed no clinical ischemia. This certainly suggests that collateral circulation plays a critical role in supplying the pelvic blood flow following aortobifemoral bypass. Pre- and postoperative penile pressures, which were not measured in this study, might provide further insight into operation-related alterations in pelvic blood flow which do not result in overt symptoms of pelvic ischemia. Antegrade filling of a previously patent IMA was presumably interrupted in the five patients with occluded distal aortas. None of these patients developed clinical evidence of bowel ischemia. The increased outflow afforded by a patent IMA did not effect distal aortic patency. Preoperative arteriographic predictors of postoperative distal aortic patency after end-to-side aortic anastomosis would be helpful if one prefers end-toend anastomoses. Those patients with continued distal aortic patency might represent a subgroup that truly requires antegrade blood flow to preserve pelvic circulation, at least in the initial postoperative period. Unfortunately, evaluation of our patients' arteriograms failed to identify predictors which could clearly separate patients who maintained distal aortic patency from those whose aortas occluded following end-to-side anastomosis. Statistical analysis between groups was not useful due to the small number of patients in each group. Evaluation of external iliac blood flow showed no clear trends. Several external iliac arteries that appeared occluded on preoperative arteriography had duplex evidence of some degree of retrograde flow from the distal femoral anastomosis after surgery. In most cases it was not possible to follow this retrograde flow in the native iliac vessels proximally to the point of obstruction. Due to technical limitations, the geometry of the pelvis limits duplex

497

imaging of portions of the common, internal and most proximal external iliac vessels. Presumably, retrograde blood flow observed in these cases reaches the internal iliac circulation by means of external iliac collaterals. However, the preoperative arteriogram was not a useful predictor of which external iliac arteries would retain retrograde flow. Complete occlusion of iliofemoral vessels proximal to the femoral anastomosis occurred frequently in both those patients with patent distal aortas and in those with occluded aortas after end-to-side aortofemoral bypass. Two patients with no evidence of antegrade distal aortic blood flow or retrograde external iliac blood flow at the time of their studies (5 and 5.5 months, postoperatively) also had no clinical evidence of pelvic ischemia. The pelvic blood supply from multiple collaterals must be sufficient by this time to prevent this complication. However, the interval necessary to allow for sufficient collateralization remains unclear. For this reason, we continue to advocate use of the aortic end-to-side anastomosis in selected cases with the understanding that (1) antegrade blood flow is not usually preserved beyond the first several months following operation; and (2) it is not possible to accurately predict postoperative flow patterns in the bypassed iliac arteries. Prospective studies, possibly employing CT scans, might better address these issues at earlier time points. However, it may not be possible to predict with certainty the occasional patient with predominantly external iliac disease who requires at least temporary pelvic blood flow from a patent distal aorta. Until they are, however, we believe it is still safest to selectively employ end-to-side anastomoses in these patients, especially since the end-to-side anastomoses have not been proven superior to end-to-end anastomosis. ACKNOWLEDGMENTS The authors gratefully acknowledge the technical assistance of Anne Musson, RVTF, Elizabeth LaBombard, RVT, and MaryAnne Waters, RVT. REFERENCES 1. BREWSTER DC, DARLING RC. Optimal methods of aortoiliac reconstruction. Surgery 1978;84:739-748. 2. SCHNEIDER JR, ZWOLAK RM, WALSH DB, et al. Lack of diameter effect on short-term patency of size-matched Dacron aortobifemoral grafts. J Vasc Surg 1991;13:785-791. 3. PICONE AL, GREEN RM, RICOTTA JR, et al. Spinal cord ischemia following operations on the abdominal aorta. J Vasc Surg 1986;3:94-103. 4. ILIOPOULOS JI, HOWANITZ PE, PIERCE GE, et al. The critical hypogastric circulation. A m J Surg 1987;154:671675. 5. MIKATI A, MARACHE P, WATEL A, et al. End-to-side aortoprosthetic anastomosis: long-term computed tomography assessment. Ann Vasc Surg 1990;4:584-591.

498

PELVIC BLOOD F L O W AFTER END-TO-SIDE AORTOBIFEMORAL B Y P A S S

Evaluation of end-to-side vs end-to-end anastomosis in aortobifemoral bypass. Arch Surg 1982;117:1580-1588. 10. RUTHERFORD RB, JONES DN, MARTIN MS, et al. Serial hemodynamic assessment of aortobifemoral bypass. J Vasc Surg 1986;4:428--435. 11. MELLIERE D, LABASTIE J, BECQUEMIN J-P, et al. Proximal anastomosis in aortobifemoral bypass: end-to-end or end-to-side? J Cardiovasc Surg 1990;31:77-80. 12. AMEL! FM, STEIN M, ARO L, et al. End-to-end versus end-to-side proximal anastomosis in aortobifemoral bypass surgery: does it matter? Can J Surg 1991;34:243-246.

6. WALSH DB, GILBERTSON J J, ZWOLAK RM, et al. The natural history of superficial femoral artery stenoses. J Vasc Surg 1991 ;•4:299-304. 7. BOWERSOX JC, ZWOLAK RM, WALSH DB, et al. Duplex ultrasonography in the diagnosis of celiac and mesenteric artery occlusive disease. J Vasc Surg 1991;14:780788. 8. RUTHERFORD RB, F L A N I G A N DP, GUPTA SK, et al. Suggested standards for reports dealing with lower extremity ischemia. J Vasc Surg 1986;4:80-94. 9. PIERCE GE, TURRENTINE M, STRINGFIELD S, et al.

II

ANNALS OF VASCULARSURGERY

II

II