Aortobifemoral Bypass Grafting Using Expanded Polytetrafluoroethylene Stretch Grafts in Patients with Occlusive Atherosclerotic Disease
Article Outline
The aim of this retrospective study was to evaluate the performance of bifurcated, longitudinally extensible (stretch), expanded polytetrafluoroethylene (ePTFE) grafts implanted in patients with aortoiliac or aortofemoral occlusive atherosclerotic disease. Between October 1991 and December 2005, 822 consecutive patients (708 men, 114 women; mean age 63.8 years) underwent aortoiliac or aortofemoral reconstruction using a bifurcated ePTFE stretch graft. Preoperatively, all patients had ultrasonographic and arteriographic evaluations and were divided into groups according to the TASC II morphological stratification of iliac lesions. Seventy-seven patients (9.4%) had type B lesions, 314 (38.2%) had type C lesions, and 431 (52.4%) were classified as type D lesions. Endarterectomy of the aorta was required in 172 patients (21%); femoral arteries were endarterectomized in 222 (27%). Femoropopliteal bypass grafting was performed in 18 patients, aortorenal bypass in 12, and mesenteric artery grafting in one. One patient died perioperatively of a myocardial infarction. Perioperative morbidity included cardiac (2.2% of patients), respiratory (0.9%), and gastrointestinal (1.2%) complications, as well as acute renal insufficiency (1.3%). Seven patients had bleeding requiring surgical revision within the first 24
hr after surgery. There were four cases (0.5%) of immediate graft thrombosis and five (0.6%) of single-limb graft thrombosis. During a mean follow-up time of 72 months (range 28-170), 58 patients (7.1%) were lost to follow-up and 205 patients (24.9%) died. The primary and secondary graft-patency rates during the observation period were 90.6% and 97.9%, respectively. Twelve late graft occlusions resolved after thrombectomy alone. Eleven cases of single-limb late thrombosis resolved after thrombectomy and profundoplasty. The limb-salvage rate during the observation period in patients who underwent operation for critical limb ischemia was 84.5%. There were nine postoperative graft infections (1.1% infection rate) in the series. Our long-term experience with ePTFE stretch grafts in aortoiliac and aortofemoral reconstruction shows that these prostheses have a high rate of patency and a low rate of graft-related complications.
Introduction
Although endovascular revascularization has become an optimal choice in the management of limited aortoiliac occlusive disease, surgical treatment remains the gold standard in the presence of extensive and complex aortoiliac disease. The recent evolution in surgical intervention for aortoiliac occlusive disease has produced satisfactory changes in approach, patency rates, and perioperative morbidity and mortality.
Longitudinally extensible (stretch) bifurcated grafts made of expanded polytetrafluoroethylene (ePTFE) have been used for aortoiliac and aortofemoral reconstruction in patients with occlusive atherosclerotic disease at our institution since 1991. Results in the first 242 patients in this series were reported previously.1 Here, we review the perioperative performance of ePTFE stretch grafts and long-term outcomes in the 822 patients in whom these grafts were implanted between October 1991 and December 2005, with special attention given to patency rate, graft infections, and other graft-related complications.
Methods
All 822 patients (708 men and 114 women, mean age 63.8 years, age range 43-88) who underwent an aortoiliac or aortofemoral reconstruction employing a bifurcated ePTFE stretch graft (Gore-Tex® Stretch Vascular Graft; W. L. Gore and Associates, Flagstaff, AZ) between October 1991 and December 2005 at our institution were included in this retrospective study. The patients' risk factors for vascular disease are listed in Table I. For elective surgery, the indications for operation were claudication in 777 patients (94.5%), including 57 total aortic occlusions, and critical ischemia (rest pain or tissue loss) in 45 (5.5%). For the six emergency cases (0.7%), the indication was acute critical limb ischemia.
Table I. Patients' characteristics (total n=822)
| Characteristic | |
|---|---|
| Age (years) | Patients (%) |
| ≤60 | 302 (36.7) |
| >60 | 520 (63.3) |
| Male | 548 (66.7) |
| Risk factors | |
| Previous or current smoking | 669 (81.4) |
| Diabetes | 226 (27.5) |
| Hypertension | 548 (66.7) |
| Hyperlipemia | 503 (61.2) |
| Coronary artery disease | 512 (62.3) |
| Cerebrovascular disease | 89 (10.8) |
| Chronic renal insufficiency | 143 (17.4) |
| Previous endovascular procedure (aortoiliac) | 140 (17) |
| Poor runoff (superficial or profunda femoral artery obstruction) | 181 (22) |
All patients were evaluated preoperatively with duplex scanning and arteriography. All arteriograms were analyzed retrospectively in order to divide all patients into groups according to the Trans-Atlantic Inter-Society Consensus (TASC) II morphological stratification of iliac lesions.2 Seventy-seven patients (9.4%) had type B lesions, 314 (38.2%) had type C lesions, and 431 (52.4%) were classified as type D lesions. A transperitoneal approach through a midline incision was employed in 789 patients (96%) and an extraperitoneal approach in 33 (4%). Graft size was selected to optimize the match in size with native vessels. All proximal anastomoses were end-to-side, leaving a short graft body in an effort to enhance graft hemodynamics, and ePTFE suture (CV4 PT9, Gore-Tex® Suture; W. L. Gore and Associates) was used in all cases.
Our antimicrobial policy included administration of intravenous cefazolin intraoperatively and until postoperative day 3 or 4. If the patient had renal failure, amoxicillin-clavulanate was administered; intravenous amoxicillin-clavulanate or levofloxacin was used in association with cefazolin if the patient had a temperature >38°C. All patients were discharged with the indication to continue oral antimicrobial therapy with amoxicillin-clavulanate or ciprofloxacin until complete healing of surgical wounds.
After hospital discharge, all patients took an antiplatelet agent or, if necessary, an anticoagulant. Follow-up examinations were scheduled for within 2 months postoperatively, 6 months and 1 year after surgery, and annually thereafter. Graft patency was confirmed by the presence of a palpable femoral pulse and duplex scanning. A return of symptoms or a marked change in vascular laboratory test values led to additional evaluations and, if required, intervention.
Survival, graft-patency, and limb-salvage rates in the series were calculated using the life-table method. Graft failure was defined as an occluded graft. The limb-salvage (defined as freedom from major amputation) rate was assessed only in patients who underwent operation for critical limb ischemia. Primary patency and secondary patency were defined in accordance with the suggested reporting standards of the Ad Hoc Committee of the Society for Vascular Surgery and the North American Chapter of the International Society for Cardiovascular Surgery.3
Results
A total of 791 aortobifemoral reconstructions and 31 aortobi-iliac reconstructions (3.8%) were performed. Endarterectomy of the aorta was necessary in 172 patients (21%) and endarterectomy of the femoral artery in 222 (27%). The distal anastomosis was performed with the common femoral artery extending into the profunda in 106 patients (13%). An additional vascular procedure was performed concomitantly in 31 patients (3.7%): femoropopliteal bypass in 18, aortorenal bypass in 12, and mesenteric artery grafting for mesenteric ischemia in one. These procedures also used ePTFE stretch grafts. In four cases (0.4%), renal artery stenosis was treated intraoperatively with an endovascular procedure. Early postoperative assessment of the technical adequacy of renal revascularization was done using magnetic resonance imaging (MRI) or arteriography. Renal artery occlusion occurred in one patient. Intraoperative variables, including the graft sizes used, are listed in Table II. Only the 16 high-risk patients (2%) were assigned to the intensive care unit postoperatively. The mean length of postoperative hospital stay was 4.5 days (range 3-15).
Table II. Intraoperative variables
| Variable | Value∗ |
|---|---|
| Duration of operation (min) | 161 ± 21 |
| Aortic clamping time (min) | 18 ± 6 |
| Intraoperative bleeding (mL) | 342 ± 313 |
| ePTFE stretch graft size (mm) | |
| 12 x 6 | 1.6% |
| 14 x 7 | 54.9% |
| 16 x 8 | 40.8% |
| 18 x 9 | 2.2% |
| 20 x 10 | 0.6% |
∗Plus-minus values are means ± standard deviation. Percentages are the proportions of patients in the series given a graft of that size. |
Perioperative complications are listed in Table III. One patient (0.1%) died of an acute myocardial infarction on the third postoperative day. Acute renal insufficiency occurred primarily in patients who had elevated creatinine levels before surgery. Cases of excessive bleeding required surgical revision to obtain hemostasis within 24hr postoperatively. In all nine patients with immediate graft thrombosis or occlusion, patency was restored by surgical thrombectomy. In two patients, profundaplasty and ePTFE patching of the anastomosis were also performed. In four cases, endovascular stenting was necessary at the end of the procedure to improve blood flow in a limb of the prosthesis.
Table III. Perioperative complications (within 30 days of surgery)
| Complication | Patients (%) |
|---|---|
| Death | 1 (0.1) |
| Cardiac (arrhythmia, MI, CHF) | 18 (2.2) |
| GI (prolonged ileus, ischemic colon, GI bleeding, hepatic failure) | 10 (1.2) |
| Pulmonary failure (respiratory assistance for 48-72hr) | 7 (0.9) |
| Stroke | 3 (0.4) |
| Acute renal insufficiency (dialysis not required) | 11 (1.3) |
| Excessive bleeding | 7 (0.9) |
| Thrombosis of one graft limb within 24hr | 5 (0.6) |
| Graft occlusion within 24hr | 4 (0.5) |
The mean follow-up time in the series was 72 months (range 28-170). Fifty-eight patients (7.1%) were lost to follow-up. A total of 205 patients (24.9%) died during follow-up (Fig. 1): 124 (15.1%) of heart disease, 39 (4.8%) of cerebrovascular disease, 16 (1.9%) of cancer, and 26 (3.2%) of unknown causes. Figure 1 documents the survival rate.
Fig. 1
Year-by-year analysis of the overall survival rates for the 822 patients treated. Values below the x axis represent the numbers of patients available for follow-up at the start of the study and at 12, 48, 72, 96, and 120 months of the study.
Anastomotic stenosis was treated with an endovascular procedure if the proximal anastomosis was affected (six patients, 0.8%) and with endarterectomy of intimal hyperplasia and patching of the anastomosis if the stenosis was distal (26 patients, 3.4%; eight bilateral cases). Twelve late graft occlusions (1.5%) were managed by thrombectomy, with a concomitant profundaplasty performed in five patients and axillofemoral bypass in three. Three late occlusions were managed conservatively; however, one of the patients affected underwent amputation at another institution and subsequently died of complications from the operation. In another patient, a thoracobifemoral bypass was performed after thrombectomy failed to restore blood flow.
In 11 cases of single-limb thrombosis, graft patency was restored by thrombectomy and profundaplasty. In three patients, both thrombolysis and stenting were used; in four cases, a crossover was done. The primary graft-patency rate during the observation period was 90.6%; the secondary rate was 97.9% (Fig. 2).
Fig. 2
Life-table analysis of overall 11-year primary patency of the bifurcated Gore-Tex ePTFE stretch grafts. Values below the x axis represent the numbers of limbs at risk at the start of the study and at 12, 48, 72, 96, and 120 months of the study.
In 14 patients, another operation was required because of progression of atherosclerotic disease. Two of these patients underwent amputation. The limb-salvage rate during the observation period in patients who underwent operation for critical limb ischemia was 84.5% (Fig. 3).
Fig. 3
Life-table analysis of limb salvage among the 45 patients (limbs) in the series with chronic critical limb ischemia (CLI). Values below the x axis represent the numbers of limbs at risk at the start of the study and at 24, 48, 72, 96, and 120 months of the study.
There were nine graft infections in the series, for an overall graft-infection rate of 1.1%. All four patients with an aortic infection had a secondary aortoenteric fistula. The mean time between graft implantation and the diagnosis of aortic graft infection was 3 years (range 1-8). All grafts were removed and axillofemoral reconstruction was performed. Three of the patients with aortoenteric fistula died 10 months after surgery. The other patient was well 15 months postoperatively.
Three inguinal infections with no proximal involvement of the graft were treated locally. One sterile asymptomatic pseudoaneurysm, at the proximal anastomosis, was observed during a routine follow-up assessment 35 months postoperatively and was treated successfully with ePTFE reconstruction. A sterile aortic pseudoaneurysm was treated endovascularly. In seven patients (0.8%), a femoral pseudoaneurysm detected at follow-up was treated surgically with resection of the lesion and reconstruction of the anastomosis using ePTFE material. Antibiotic therapy was administered for 3 months. One year later, duplex ultrasonography and clinical and hematological studies showed no signs of residual infection. Routine ultrasonographic studies showed no other periprosthetic collections and no graft dilatations during follow-up.
Discussion
In general, the recent evolution in surgical intervention for aortoiliac occlusive disease has produced satisfactory changes in approach, patency rates, and perioperative morbidity and mortality. Endovascular techniques have had an important impact on the treatment of aortoiliac disease, and the use of these approaches continues to increase, primarily because of their minimally invasive nature. Percutaneous angioplasty with selective stent placement has been found to be effective in treating isolated, focal iliac stenosis. In contrast, the results of iliac stenting in patients with diffuse or multisegment iliac occlusive disease have not been promising.4 Thus, aortobifemoral bypass, with its durability and safety, remains the gold-standard treatment for diffuse iliac occlusive disease.
The TASC I and II document states that TASC C and D iliac disease, which, by definition, involves multiple iliac segments, is best treated with open revascularization.2, 5 Moreover, a study by Powell et al.6 found that in patients with diffuse iliac occlusive disease also involving the external iliac arteries, the 1-year primary and secondary patency rates after stenting were only 30% and 53%, respectively. Our series included a large number of patients with multisegment iliac occlusive disease who had previously undergone stent grafting; however, percutaneous endovascular therapy is typically advocated as the initial treatment choice, with bypass surgery the secondary treatment.7
The availability of both the endovascular and the surgical approaches does provide new treatment options for patients with aortoiliac disease. In selected patients with unilateral iliac artery occlusion and contralateral atherosclerotic stenosis, crossover femorofemoral bypass may be performed together with endovascular treatment. Axillobifemoral bypass is an acceptable alternative vascular procedure in high-risk patients.8 Surgical intervention is sometimes necessary to treat complications from a percutaneous endovascular procedure.
Our series illustrates the complexity of surgical management of total occlusion of the abdominal aorta. Juxtarenal aortic occlusion was observed in 12 of our patients, all of whom underwent renal artery reconstruction. Improvements in surgical protocols have reduced the risk associated with open revascularization in such patients. Organ protection must be optimized with the collaboration of the anesthetist and systemic administration of nephroprotective drugs.
Patients with aortoiliac occlusive disease who have undergone aortobifemoral bypass grafting must be monitored regularly and permanently after surgery to decrease the risk of sudden graft failure associated with progressive arteriosclerotic disease or late anastomotic aneurysm. These patients have a high rate of distal anastomotic stenosis caused primarily by neointimal hyperplasia, and they should be studied angiographically before undergoing reoperation.
Patients with poor femoral runoff because of infrainguinal disease, such as patients with occlusion of the superficial femoral and popliteal arteries, have lower cumulative aortic graft-patency rates than patients with good runoff. The adverse effects of concomitant superficial femoral artery disease may be eliminated completely by an adequate profundaplasty. Moreover, a simultaneous femoropopliteal bypass has been reported to enhance aortic graft patency.9
The consequences of graft failure and the need for subsequent complex interventions to treat it must always be considered, particularly in younger patients with a longer life expectancy. Recent studies have shown that aortobifemoral grafting yields less encouraging results when performed in patients with premature atherosclerosis, with 5-year patency rates as low as 50%.10, 11
In our series of 822 procedures using ePTFE stretch grafts, there were no perigraft hematomas or seromas, graft dilatations, or aneurysms during long-term follow-up. These results are in agreement with those of our previous study of these prostheses, in which MRI assessments during follow-up revealed no dilatations or perigraft seromas.12 In contrast, aortic grafts made of Dacron dilate immediately after placement, and dilatation continues slowly over time. Another earlier study by our group1 showed that these grafts have acceptable handling characteristics and long-term patency. The current large, long-term investigation provides solid support for these findings: the primary patency rate during a follow-up period of up to 162 months was 90.3%.
In addition, the structure of ePTFE material may help to reduce graft-related complications. The surface characteristics of graft fibers, the relative degree of graft material hydrophobia, and the presence of an anionic compared with a cationic surface charge affect initial bacterial adherence to a prosthesis. Dacron vascular grafts have a greater propensity for bacterial adherence than ePTFE grafts.13 Because ePTFE is relatively nonporous and more hydrophobic than Dacron, it has theoretical advantages with respect to resistance to bacterial adhesion and growth.13 In our previous study of aortioiliac and aortofemoral reconstructions1 ePTFE stretch grafts demonstrated good tissue incorporation, with no fluid collection between the prostheses and surrounding tissues. In the current study, the graft-infection rate remained within acceptable limits during long-term follow-up.
Conclusions
Aortoiliac or aortofemoral reconstructive surgery is a proven treatment for aortoiliac occlusive disease. We found that ePTFE stretch grafts have excellent long-term patency in this application, along with a low rate of graft-related complications.
After aortobifemoral bypass grafting, patients must be monitored closely to decrease the risk of sudden graft failure associated with progressive arteriosclerotic disease or late anastomotic aneurysm.
References
- . Extensible expanded polytetrafluoroethylene vascular grafts for aortoiliac and aortofemoral reconstruction. Cardiovasc Surg. 2000;8:538–544
- . on behalf of the TASC II Working Group. Inter-Society Consensus for the Management of Peripheral Arterial Disease (TASC II). J Vasc Surg. 2007;45(Suppl. S):S5–S67
- Recommended standard for reports dealing with lower extremity ischemia: revised version. J Vasc Surg. 1997;26:517–538
- Early results of stent-grafting to treat diffuse aortoiliac occlusive disease. J Vasc Surg. 2003;37:1175–1180
- . Management of peripheral arterial disease (PAD). TASC Working Group. TransAtlantic Inter-Society Consensus (TASC). J Vasc Surg. 2000;31:S1–S296
- . Predicting outcome of angioplasty and selective stenting of multisegment iliac artery occlusive disease. J Vasc Surg. 2000;32:564–569
- . Aortoiliac occlusive disease: surgical versus interventional therapy. Curr Interv Cardiol Rep. 2001;3:109–116
- Multivariate analysis of long-term results after an axillobifemoral and aortobifemoral bypass in patients with aortoiliac occlusive disease. J Cardiovasc Surg (Torino). 2000;41:905–910
- . Aortobifemoral grafting: factors influencing long-term results. Vascular. 2004;12:171–178
- Aortofemoral bypass in young patients with premature atherosclerosis: is superficial femoral vein superior to Dacron?. J Vasc Surg. 2004;40:17–23
- . The impact of patient age and aortic size on the results of aortobifemoral bypass grafting. J Vasc Surg. 2003;37:1219–1225
- A new ePTFE stretch graft for aorto-iliac reconstructions. Surgical evaluation and one year follow-up with magnetic resonance imaging. J Cardiovasc Surg (Torino). 1995;36:135–141
- . Physical characteristics of expanded polytetrafluoroethylene grafts. In: Stanley JC editors. Biologic and Synthetic Vascular Prostheses. Orlando, FL: Grune & Stratton; 1982;p. 553–561
PII: S0890-5096(09)00042-9
doi:10.1016/j.avsg.2009.01.004
© 2009 Annals of Vascular Surgery Inc. Published by Elsevier Inc All rights reserved.
