- Email: [email protected]
Prospective controlled study of polytetrafluoroethylene versus saphenous vein in claudicant patients with bilateral above knee femoropopliteal bypasses
Prospective controlled study of polytetrafluoroethylene versus saphenous vein in claudicant patients with bilateral above knee femoropopliteal bypasses Ali F. AbuRahma, MD, Patrick A. Robinson, MD, and Steven M. Holt, MD, Charleston, WV
Background. Although several studies have compared the patency rates of polytetrafluoroethylene (PTFE) and saphenous vein grafts (SVG) for the above knee location, none have compared the 2 grafts when implanted in the same patient with claudication who needs bilateral above knee femoropopliteal bypasses. Methods. Forty-three patients (86 limbs) with bilateral disabling claudication who had superficial femoral artery occlusion and above knee reconstitution with 2- to 3-vessel runoff were analyzed. Patients were treated on one side with PTFE and on the other side with SVG. They were sequentially assigned to PTFE-SVG alternating with SVG-PTFE. All patients were followed using duplex ultrasound and ankle/brachial indexes at 1 month and every 6 months thereafter. Results. The perioperative complication rates were 5% for PTFE and 12% for SVG. There was no operative death or perioperative amputation for either procedure. The Kaplan-Meier estimate of primary, assisted primary, and secondary patency rates at 72 months were 68%, 68%, and 77% for PTFE and 76%, 83%, and 85% for SVG. There were no statistically significant differences between primary and secondary patency rates for both grafts; however, the assisted primary patency rates were higher for SVG (P < .05). The crude limb salvage rate at 72 months was 98% for PTFE and 98% for SVG. There were no risk factors identified that had an impact on graft patency. Conclusions. PTFE and SVG for above knee bypasses have comparable patency and limb salvage rates in claudicant patients with bilateral superficial femoral artery occlusion and 2- to 3-vessel runoff. This may justify the use of PTFE for above knee locations in these selected patients. (Surgery 1999;126:594602.) From the Department of Surgery, Robert C. Byrd Health Sciences Center Of West Virginia University, Charleston Area Medical Center, Charleston, WV
THROUGHOUT THE LITERATURE it has been widely recognized that, overall, autogenous saphenous vein is the best conduit for infrainguinal bypass. Although it is definitely established that infrageniculate arterial bypass with saphenous vein is the superior conduit, controversy still exists for the preferential use of polytetrafluoroethylene (PTFE) for above knee bypasses. This controversy is stirred by the fact that some authors advocate the use of the saphenous vein or an all-autogenous tissue policy for infrainguinal arterial reconstruction.1-4 Others have found that PTFE can be used for the above knee bypass with equal efficacy.5-12 Most studies that have compared PTFE with saphenous vein in above knee bypasses have studied grafts Presented at the 56th Annual Meeting of the Central Surgical Association, St Louis, Mo, Mar 4-6, 1999. Reprint requests: Ali F. AbuRahma, MD, 3100 MacCorkle Ave SE, Suite 603, Charleston, WV 25304. Copyright © 1999 by Mosby, Inc. 0039-6060/99/$8.00 + 0
594 SURGERY
11/6/99942
that were placed for multiple indications such as claudication, rest pain, or trophic changes, and most are retrospective studies. For many of these indications, runoff is usually less than ideal. When claudicant patients are looked at separately in these studies, most reveal a higher patency and limb salvage rates. In our previous study,13 when reconstruction was done for disabling claudication, a 5-year secondary patency rate of 64% was seen and the limb salvage rate was 97%. Prendiville et al8 showed similar long-term results, and Sterpetti et al5 reported slightly superior results. For those few select patients with disabling claudication that is refractory to medical therapy and in which bypass is performed, one may argue that the PTFE long-term patency rates are comparable to those of the saphenous vein. Rosenthal et al7 showed that for claudication alone, the preferential prosthetic above knee femoropopliteal bypass is a “safe and durable operation.” However, there are no studies in the literature that have compared PTFE with saphenous vein in above knee bypasses
AbuRahma, Robinson, and Holt 595
Surgery Volume 126, Number 4
Table I. Life table for Kaplan-Meier method for PTFE (primary patency) Time interval (mo) 0 1 6 12 18 24 30 36 42 48 54 60 66 72
No. at risk at start
No. failed
No. withdrawn (censored)
Interval patency rate
Cumulative patency rate (%)
43 41 39 39 33 29 27 26 25 23 19 14 10 6
2 0 0 6 4 1 0 0 0 0 0 0 0 0
0 2 0 0 0 1 1 1 2 4 5 4 4 6
0.953488 1 1 0.846154 0.878788 0.965517 1 1 1 1 1 1 1 1
95 95 95 81 71 68 68 68 68 68 68 68 68 68
in patients who need bilateral arterial reconstruction for disabling claudication. In this fashion, the patients become their own control when risk factors and demographics are analyzed for each type of graft to demonstrate whether PTFE can be preferentially used in this situation. Therefore, the purpose of this prospective controlled study is to compare the results of PTFE versus saphenous vein in claudicant patients with bilateral above knee femoropopliteal bypasses. PATIENT POPULATION AND METHODS This study includes 43 patients who had surgery between January 1992 and December 1994 at our institution. These patients had bilateral disabling claudication with failed medical therapy with long superficial femoral artery occlusion and above knee reconstitution with 2- to 3-vessel runoff. Patients were treated on one side with PTFE and on the other side with reversed saphenous vein grafts. These patients were sequentially assigned to PTFE/saphenous vein alternating with saphenous vein/PTFE. All PTFE grafts were 8-mm grafts (W L Gore & Associates, Inc, Flagstaff, Ariz). All grafts were done using regional anesthesia (epidural or spinal). All patients were given 325 mg aspirin in the immediate postoperative period and continued postoperatively. The hospital records were examined for perioperative complications, and all demographics/risk factors were collected. Graft surveillance. Preoperative and postoperative evaluations included lower extremity vascular laboratory studies with ankle/brachial indexes and duplex ultrasonography using Ultramark 9 equipment during the first part of the study and Ultramark 3000 equipment during the last 2 years
Standard error (%) 3.14 3.21 3.29 5.68 6.66 7.14 7.40 7.54 7.69 8.02 8.82 10.28 12.16 15.70
of follow-up (Advanced Technical Laboratory, Inc, Bothell, Wash). Graft surveillance protocol included an immediate postoperative duplex ultrasound with ankle/brachial indexes, and this was repeated at 1 month, 3 months, and every 6 months thereafter. All grafts were scanned throughout their length with systolic and diastolic peak velocities recorded every 5 to 10 cm and at any area of abnormality on B-mode or color interrogation. The proximal inflow and outflow native arteries were similarly scanned. Color duplex ultrasound studies were defined as abnormal if a peak systolic flow velocity was more than twice the peak systolic velocity in the adjacent graft, or if peak systolic velocity throughout the graft was uniformly less than 45 cm/sec. The absence of color flow in these grafts was indicative of a graft occlusion. All abnormal findings on duplex ultrasounds were confirmed using conventional arteriography. Patients with graft stenosis or occlusion were assigned to one of the above intervals at the time the abnormality was detected. All criteria used for this study attempted to conform to the standards suggested by the Ad Hoc Committee of the Joint Council of the Vascular Societies for Reports Dealing with Lower Extremity Ischemia.14 Primary patency was defined as uninterrupted patency of the original graft without any intervention; an assisted primary patency was defined as patency assisted by simple measures (primarily angioplasty); and secondary patency was defined as patency of the original graft that was thrombosed but maintained by the following measures: thrombectomy, thrombolytic therapy, percutaneous angioplasty, or patch graft angioplasty. Occluded grafts that were not salvaged, but were replaced with new grafts, were regarded as occlud-
596 AbuRahma, Robinson, and Holt
Surgery October 1999
Fig 1. Kaplan-Meier curve for time to loss of primary patency for both the PTFE and vein groups.
Fig 2. Kaplan-Meier curve for time to loss of assisted patency for both the PTFE and vein groups.
ed or failed. Operative mortality (early death) and postoperative complications were considered if they occurred within 30 days of the operation. All patients were analyzed to determine limb salvage. When amputations were limited to the toes or transmetatarsal level, they were not considered failures; however, any amputation at the ankle or above was recorded as limb loss. Data analysis. The data were tabulated, and a statistical analysis was performed using the chi-square
test and the Student t test for group comparison. Life table analyses were used to calculate the cumulative patency and limb salvage rates, and groups were compared with the log-rank test. RESULTS The mean age of the group was 65 years (range 46 to 82 years). The demographic and clinical characteristics of the entire group were as follows: 25 (58%) were male, 31 (72%) were smokers, 18
AbuRahma, Robinson, and Holt 597
Surgery Volume 126, Number 4
Fig 3. Kaplan-Meier curve for time to loss of secondary patency for both the PTFE and vein groups.
Table II. Life table for Kaplan-Meier method for vein (primary patency) Time interval (mo) 0 1 6 12 18 24 30 36 42 48 54 60 66 72
No. at risk at start
No. failed
No. withdrawn (censored)
Interval patency rate
43 43 41 40 36 34 31 29 28 27 21 17 13 7
0 0 1 4 2 2 1 0 0 0 0 0 0 0
0 2 0 0 0 1 1 1 1 6 4 4 6 7
1 1 0.97561 0.9 0.944444 0.941176 0.967742 1 1 1 1 1 1 1
Cumulative patency rate (%) 100 100 98 88 83 78 76 76 76 76 76 76 76 76
Standard error (%) 0 0 2.38 4.85 5.71 6.27 6.71 6.94 7.06 7.19 8.15 9.06 10.36 14.12
Student’s t test = 0.60210. P not significant. Degree of freedom is 85.
(42%) had diabetes mellitus, 22 (51%) had hypertension, 17 (40%) had hypercholesterolemia, and 12 (28%) had coronary artery disease. The mean preoperative ankle/brachial indexes were similar for both the PTFE (0.56) and vein groups (0.55). Morbidity and mortality data. The perioperative complication rates were 5% for the PTFE group and 12% for the vein graft group (no statistically significant differences). Two patients (5%) in the vein group had perioperative nonfatal myocardial infarctions, 2 others (5%) had deep vein thrombosis, and 1 patient (2%) in the PTFE group had perioperative graft thrombosis necessitating an imme-
diate thrombectomy. One patient in each group (2%) had significant bleeding that required surgical exploration to control bleeding. There was no perioperative limb loss, with only 1 late limb loss in each group (ie, a limb salvage rate of 98%). There was no perioperative death in this series. However, there were 4 late deaths, 3 as a result of myocardial infarctions and 1 related to cancer (ie, a late survival rate of 91%). Long-term graft patency data. The crude primary patency, assisted primary patency, and secondary patency rates were similar in both the PTFE and vein grafts: 70%, 70%, and 79% and 77%, 84%,
598 AbuRahma, Robinson, and Holt
Surgery October 1999
Table III. Life table for Kaplan-Meier method for PTFE (assisted patency) Time interval (mo) 0 1 12 18 24 30 36 42 48 54 60 66 72
No. at risk at start 43 41 39 34 31 29 28 25 23 19 14 10 6
No. failed
No. withdrawn (censored)
2 0 5 3 1 0 2 0 0 0 0 0 0
0 2 0 0 1 1 1 2 4 5 4 4 6
Interval patency rate 0.953488 1 0.871795 0.911765 0.967742 1 0.928571 1 1 1 1 1 1
Cumulative patency rate (%) 95 95 83 76 73 73 68 68 68 68 68 68 68
Standard error (%) 3.14 3.21 5.47 6.40 6.80 7.03 7.27 7.69 8.02 8.82 10.28 12.16 15.70
Table IV. Life table for Kaplan-Meier method for vein (assisted patency) Time interval (mo) 0 1 12 18 24 30 36 42 48 54 60 66 72
No. at risk at start
No. failed
No. withdrawn (censored)
Interval patency rate
43 43 41 41 39 35 32 31 30 24 20 14 8
0 0 0 2 3 2 0 0 0 0 0 0 0
0 2 0 0 1 1 1 1 6 4 6 6 8
1 1 1 0.95122 0.923077 0.942857 1 1 1 1 1 1 1
Cumulative patency rate (%) 100 100 100 95 88 83 83 83 83 83 83 83 83
Standard error (%) 0 0 0 3.28 4.91 5.81 6.07 6.17 6.27 7.01 7.68 9.18 12.14
Student’s t test = 2.72838. P < .02. Degree of freedom is 85.
and 86%, respectively. The mean time in months to the loss of primary patency, assisted primary patency, and secondary patency were also similar for the PTFE and vein groups (12.9, 12.9, and 15.3 and 16.8, 18, and 18, respectively). The Kaplan-Meier estimate of primary, assisted primary, and secondary patency rates at 72 months were 68%, 68%, and 77% for PTFE and 76%, 83%, and 85% for vein grafts. There were no statistically significant differences between primary patency and secondary patency rates for both groups; however, the assisted primary patency rates were higher for the vein grafts (P < .05). Figs 1, 2, and 3 show the Kaplan-Meier curves for time to the loss of primary patency, assisted patency, and secondary patency for both the PTFE and vein groups. Tables 1 through 6 show the life table analysis for primary patency, assisted primary patency, and secondary patency for both PTFE and vein grafts. Risk factors and patency rates. No specific risk
factors (gender, smoking, diabetes mellitus, hypertension, hypercholesterolemia, or coronary artery disease) were identified that had an impact on graft primary, assisted primary or secondary patency. However, patients with a loss of patency on one side appeared to be at a higher risk for loss of patency on the opposite side. DISCUSSION Many investigators have concluded that PTFE above knee femoropopliteal bypass patency rates can be comparable to saphenous vein patency rates.5,6,10 Most of these studies have taken all types of vascular patients and indications for surgery into account. Authors continue to suggest that prosthetic above knee femoropopliteal bypass may be the preferential choice in a certain select group of patients, ie, claudicant patients with normal to near normal runoff (2- to 3-vessel), but argue that more studies need to be performed.5,8,12 No studies
AbuRahma, Robinson, and Holt 599
Surgery Volume 126, Number 4
Table V. Life table for Kaplan-Meier method for PTFE (secondary patency) Time interval (mo) 0 1 12 18 24 30 36 42 48 54 60 66 72
No. at risk at start 43 43 41 40 38 34 32 31 29 24 18 13 7
No. failed 0 0 1 2 3 1 0 0 1 1 0 0 0
No. withdrawn (censored) 0 2 0 0 1 1 1 2 4 5 5 6 7
before this current report have compared PTFE with saphenous vein in the same patient for the same indication for surgery with comparable runoff. Most vascular surgeons would agree that a patient with superficial artery disease with 2- to 3vessel runoff would be an “ideal” candidate for vascular reconstruction. In these ideal candidates, one could argue that PTFE could be preferentially selected over saphenous vein as the conduit of choice. Surgeons who favor preferential use of synthetic grafts in above knee femoropopliteal bypasses often base their argument on saving the vein for later use, either for coronary artery bypass revascularization or for ipsilateral below knee bypass when previous above or below knee bypasses have failed. With the advancement of cardiac revascularization techniques, ie, percutaneous treatment or use of internal mammary artery, authors disagree as to whether the argument for preserving the vein for future coronary bypass is valid3,5; however, the literature has shown a range of 14% to 28% of patients who undergo infrainguinal bypass will eventually need coronary artery reconstruction.7 Taylor et al2 support the practice of reversed saphenous vein as the procedure of choice for infrainguinal bypass, in spite of their findings that 28% of patients in their series had a missing greater saphenous vein on the side of need of revascularization, due to previous coronary artery bypass grafting that resulted in the use of an alternative conduit. Overall, the question remains whether it is beneficial to preserve the saphenous vein for future use, whether for distal bypass or coronary reconstruction. Some authors have shown that a staged procedure in which a preferential prosthetic above knee bypass followed by autogenous saphenous vein below the knee bypass for recurrent ischemia
Interval patency rate 1 1 0.97561 0.95 0.921053 0.970588 1 1 0.965517 0.958333 1 1 1
Cumulative patency rate (%) 100 100 98 93 85 83 83 83 80 77 77 77 77
Standard error (%) 0 0 2.38 3.96 5.30 5.88 6.06 6.16 6.64 7.56 8.73 10.27 14.00
can improve limb salvage rates.4,11 Although Plecha et al4 showed that a secondary autogenous vein reconstruction can salvage a failed above knee PTFE graft, they did not believe there was an advantage to this approach. Rosenthal et al,11 however, reported an improved overall limb salvage rate for staged arterial reconstruction in the infrainguinal setting versus primary saphenous vein (86% versus 78%, respectively). One must also consider the possibility of affecting future saphenous vein bypass patency rates in a limb that has previously had a prosthetic graft. Some may argue that patency rates are decreased in this population, but one must consider that these patients have also had progression of their disease and the overall quality of the native artery and runoff is likely to be much poorer. Edwards et al,15 however, achieved a 5-year limb salvage rate of 90% when autogenous reversed saphenous vein was used as a secondary bypass for a previously failed infrainguinal bypass. No study using PTFE as a second conduit for limb salvage has come close to achieving these rates. A final argument can be made that the life expectancy of most patients with vascular disease, especially with critical ischemia, is decreased, and a single prosthetic above knee bypass may be the only procedure needed. Risk factors that have been shown to affect the patency rates of infrainguinal bypasses include habitual smoking, diabetes, poor runoff, indication for surgery, ankle/brachial index, and level of distal anastomosis. Our previous study reported on the correlation of ankle/brachial indexes to graft patency. When the ankle/brachial index was 0.5 or greater, primary and secondary patency rates were 57% and 68%, respectively, compared with a primary patency rate of 37% and a secondary patency rate of 56% for patients with an ankle/brachial
600 AbuRahma, Robinson, and Holt
Surgery October 1999
Table VI. Life table for Kaplan-Meier method for vein (secondary patency) Time interval (mo) 0 1 12 18 24 30 36 42 48 54 60 66 72
No. at risk at start
No. failed
43 43 41 41 41 38 34 32 31 24 20 14 8
0 0 0 0 2 3 1 0 0 0 0 0 0
No. withdrawn (censored) 0 2 0 0 1 1 1 1 7 4 6 6 8
Interval patency rate 1 1 1 1 0.95122 0.921053 0.970588 1 1 1 1 1 1
Cumulative patency rate (%) 100 100 100 100 95 88 85 85 85 85 85 85 85
Standard error (%) 0 0 0 0 3.28 5.00 5.64 5.82 5.91 6.71 7.36 8.79 11.63
Student’s t test = 0.81777. P not significant. Degree of freedom is 85.
index of less than 0.5.13 This present study is interesting in that ankle/brachial indexes had no correlation to patency. This is likely due to the fact that each patient had good runoff and the mean ankle/brachial index for the entire set was 0.55 with a low of 0.38 and high of 1.0. Prendiville et al8 reported that continued habitual smoking and patients with diabetes mellitus adversely affected primary patency rates. Smokers had a 5-year patency rate of 40% as compared with 66% for nonsmokers, and diabetic patients had a decreased 5year patency rate. Again, our study showed no correlation between these factors and patency of either graft. Because runoff, indication for surgery, and level of anastomosis were the same in each subset of patients, we found no risk factors in our study that had an impact on patency rates. Our study did reveal, however, that patients with a loss of patency on one side were at a higher risk for loss of patency on the other side. The evidence suggests that these patients may have an underlying defect that makes them a higher risk for graft failure. Edwards et al15 advocate using an aggressive approach when evaluating patients with failed grafts who need reoperation by checking for hypercoagulable states, blood lipid levels, and plasma homocysteine levels. By identifying these factors, surgeons could possibly improve reoperation patency rates in these high-risk patients. Recently, Allen et al16 reviewed their experience in 239 patients with claudication as a result of superficial femoral artery occlusion. Femoropopliteal reconstruction was performed with saphenous vein to below knee popliteal artery in 66 patients. PTFE was used in 128 patients as a bypass graft to the above knee popliteal artery, and 45 patients had a
PTFE graft to the below knee popliteal artery. All patients were involved in a postoperative graft surveillance program with graft revision, when appropriate. The primary patency rates at 5 years for above knee PTFE, below knee PTFE, and below knee vein were 58%, 55%, and 60%, respectively, and this was not significantly different among the graft groups. The 5-year secondary patency rates for above knee PTFE grafts were 79%, 73% for below knee PTFE, and 74% for below knee vein (not statistically significant). They concluded that long-term patency rates for femoropopliteal bypasses in patients with claudication were similar for PTFE and autogenous saphenous vein grafts. Also one must take into account the morbidity of each procedure. For prosthetic above knee bypasses, 2 small incisions are made for placement of the graft, in contrast to saphenous vein grafting where multiple incisions are made down the leg, often extending below the knee for harvesting adequate graft material. This may not only affect postoperative healing, but it also increases the time in the operating room, putting additional stress on the already sensitive vascular patient. As you can see in our results, the complication rates were higher for vein grafting as compared with PTFE, 12% to 5%, respectively, with 5% of these complications being of a cardiac origin. Even if these rates were not statistically significant, it may suggest the need for the simplest and fastest operation. As life expectancy increases because of improved medical therapy, more patients are going to need infrainguinal arterial reconstruction. As vascular advancements are made, the challenge of restoring blood flow to the lower extremity is going to continually grow. As endovascular techniques, throm-
AbuRahma, Robinson, and Holt 601
Surgery Volume 126, Number 4
bolysis, and new graft materials improve, all will play a role in vascular reconstruction, especially in superficial femoral artery disease. In conclusion, PTFE and saphenous vein grafts for above knee bypasses have comparable patency and limb salvage rates in claudicant patients with bilateral superficial femoral artery occlusion with 2to 3-vessel runoff. This may justify the use of PTFE for above knee locations in these selected patients. We believe that preferential use of PTFE in the above knee setting in claudicant patients is an acceptable practice. REFERENCES 1. Kent KC, Whittemore AD, Mannick JA. Short-term and midterm results of an all-autogenous tissue policy for infrainguinal reconstruction. J Vasc Surg 1989;9:107-14. 2. Taylor LM Jr, Edwards JM, Porter JM. Present status of reversed vein bypass grafting: five-year results of a modern series. J Vasc Surg 1990;4:193-206. 3. Michaels JA. Choice of material for above-knee femoropopliteal bypass graft. Br J Surg 1989;76:7-14. 4. Plecha EJ, Freischlag JA, Seabrook GR, Towne JB. Femoropopliteal bypass revisited: an analysis of 138 cases. Cardiovasc Surg 1996;4:195-9. 5. Sterpetti AV, Schultz RD, Feldhaus RJ, Peetz DJ Jr. Sevenyear experience with polytetrafluoroethylene as above-knee femoropopliteal bypass graft: is it worthwhile to preserve the autogenous saphenous vein? J Vasc Surg 1985;2:907-12. 6. Veith FJ, Gupta SK, Ascer E, White-Flores S, Samson RH, Soher LA, et al. Six-year prospective multicenter randomized comparison of autologous saphenous vein and expanded polytetrafluoroethylene grafts in infrainguinal arterial reconstructions. J Vasc Surg 1986;3:104-14. 7. Rosenthal D, Evans D, McKinsey J, et al. Prosthetic aboveknee femoropopliteal bypass for intermittent claudication. J Cardiovasc Surg 1990;31:462-8. 8. Prendiville EJ, Yeager A, O’Donnell TF Jr, Coleman JC, Jaworek A, Callow AD, et al. Long-term results with the above-knee popliteal expanded polytetrafluoroethylene graft. J Vasc Surg 1990;11:517-24. 9. Abbott WA. Prosthetic above-knee femoral-popliteal bypass: indications and choice of graft. Semin Vasc Surg 1997;10:3-7. 10. Bergan JJ, Veith FJ, Bernard VM, Yao JST, et al. Randomization of autogenous vein and polytetrafluoroethylene grafts in femoral-distal reconstruction. Surgery 1982;92:921-30. 11. Rosen RC, Johnson WC, Bush HL Jr, Cho SI, O’Hara ET, Nabseth DC. Staged infrainguinal revascularization: initial prosthetic above-knee bypass followed by a distal vein bypass for recurrent ischemia—a valid concept for extending limb salvage? Am J Surg 1986;152:224-30. 12. Quinones-Baldrich WJ, Prego AA, Ucelay-Gomez R, Freischlag JA, Ahn SS, Baker JD, et al. Long-term results of infrainguinal revascularization with polytetrafluoroethylene: a ten-year experience. J Vasc Surg 1992;16:209-17. 13. AbuRahma AF, Robinson PA, Stuart SP, Witsberger TA, Stewart WA, Boland JP. Polytetrafluoroethylene grafts in infrainguinal arterial revascularization: factors affecting outcome. Arch Surg 1993;128:417-22. 14. Ad Hoc Committee on Reporting Standards, Society for Vascular Surgery/North American Chapter, International Society for Cardiovascular Surgery. Suggested standards for reports dealing with lower extremity ischemia. J Vasc Surg 1986;4:80-94.
15. Edwards JE, Taylor LM, Porter JM. Treatment of failed lower extremity bypass grafts with new autogenous vein bypass grafting. J Vasc Surg 1990;11:136-45. 16. Allen BR, Reilly JM, Rubin BG, Thompson RW, Anderson CB, Flye MW, Sicard GA. Femoropopliteal bypass for claudication: vein vs PTFE. Ann Vasc Surg 1996;10:178-85.
DISCUSSION Dr Gregorio A. Sicard (St Louis, Mo). I congratulate you for an important and landmark contribution to the surgical treatment of lower extremity claudication. The overall approach to the treatment of claudication continues to be the source of significant debate in the vascular literature. Many questions are debated. First, should claudicants undergo any form of surgical therapy? Should risk modification and a gradual exercise program be the gold standard from which to analyze and compare surgical results? Second, should the incapacitating lower extremity claudicant with isolated short segment superficial femoral artery stenosis have endoluminal techniques as another option to treat the disease? If a surgical option is chosen for superficial femoral artery disease, which should be the conduit of choice: saphenous vein, PTFE, or Dacron? In this prospective controlled trial, 43 patients were sequentially assigned to PTFE/saphenous vein alternating with saphenous vein PTFE above the knee bypass and followed longitudinally with a strict surveillance protocol to identify the so-called failing graft. They achieved excellent results and similar 72-month patency rates for saphenous vein and PTFE, respectively. I have a few questions for you. Do you have another cohort of patients with bilateral claudication caused by superficial femoral artery disease in which endoluminal treatment is compared with open surgical revascularization? As you well know, the endoluminal techniques are being done by many specialties, and the superficial femoral artery has been one of the sites that has been targeted for these procedures. If you do, what are the results and should a bypass be considered in patients with superficial femoral artery occlusion versus endoluminal techniques? The overall experience with surveillance protocols is not very good for prosthetic grafts compared with autologous vein. In your results, the mean time in months to loss of primary and primary assisted patency was approximately 12 months. This gives you a statistical difference for assisted primary patency of 68% for PTFE and 83% for vein, respectively. In other words, if looking at it in another way, there is a 6-month difference in which the graft fails. In your manuscript, the PTFE failed 6 months earlier than the saphenous vein bypasses. What other techniques do you suggest we use to evaluate the PTFE groups in this 12-month period, which seems to be very critical for
602 AbuRahma, Robinson, and Holt
failure of prosthetic grafts? Are exercise Dopplers and waveform evaluation of help in identifying the so-called failing PTFE graft between the 12- and 18-month period, or should arteriography be recommended? Last, could you expand on the observation that patients who occlude 1 graft are at greater risk to lose the contralateral graft? Does this observation apply to the initial loss of either conduit, and last, when one conduit thromboses, should the patient be systemically coagulated to protect the other? Dr AbuRahma. I totally agree with you that all patients with claudication should undergo risk modification and a gradual exercise program with or without pharmacologic therapy. If this fails in patients who have disabling claudication, an alternative therapy, such as a bypass or endovascular therapy, should be explored. We do not have another cohort of patients with bilateral claudication caused by superficial femoral artery disease in which endoluminal treatment is compared with open surgical revascularization. However, a significant number of our patients with superficial femoral artery disease are presently treated by some form of endovascular therapy, eg, percutaneous transluminal balloon angioplasty with or without stenting. I agree with your statement that the overall experience with surveillance protocol is not very good for prosthetic grafts compared with autologous veins. A significant number of PTFE grafts are discovered when they are thrombosed rather than at the stenosis stage. This may explain the differences between the assisted primary patency between the vein and PTFE grafts. Whether exercise Doppler and waveform evaluations would be helpful in identifying the so-called failing PTFE graft between a 12- and 18-month period is difficult to predict; however, it may be a good idea to try. We do not have any specific answers from our study as to why if one graft fails, the contralateral graft usually fails. However, this may be partly explained by unidentifiable hypercoagulable states or other risk factors that we could not document in our study. It may be a good idea to anticoagulate patients after one graft fails.
Surgery October 1999
Dr John D. Corson (Iowa City, Iowa). Were these procedures all done simultaneously or were they staged? If they were staged, how much time was there between the 2 procedures? Dr AbuRahma. They were staged, generally between 6 to 8 weeks. Dr William H. Baker (Maywood, Ill). If somebody comes into your hospital now with intermittent claudication available vein, which operation do you perform if they have intermittent claudication? Dr AbuRahma. If they are ideal candidates, similar to the patients described in our study, ie, claudicant with above knee reconstruction and good below knee runoff (2- to 3-vessel runoff), I would not hesitate to recommend a PTFE graft. Dr Eric D. Endean (Lexington, Ky). In grafts that failed, did you have to convert a vein graft to a PTFE or vice versa in any cases? Were any inflow procedures required for any of your grafts and did this affect patency? Dr AbuRahma. None of these had an inflow problem that required any type of correction for the graft that failed. In regards to converting vein graft to PTFE or vice versa, that was done only in patients with PTFE who had above knee grafts. If the graft failed using simple measures, such as patch angioplasty, they were converted into vein grafts for the below knee locations. Dr James R. Debord (Peoria, Ill). You randomized in these staged procedures PTFE-vein versus vein-PTFE. Did you do anything to randomize left versus right? There could be some differences in the runoff with a best-leg/worst-leg scenario. Could there have been any bias in which leg you picked for PTFE versus vein? Dr AbuRahma. We did not randomize regarding left verus right. In regards to your second question as to why we would measure limb salvage rate in patients who are all claudicant with limbs that are not in any jeopardy, this is true at the time they were initially seen for the first intervention. However, during the 5-year follow-up, some of these patients (a few percent) could have limb loss, even with intervention using a vein or PTFE. Therefore, we thought a limb salvage rate was appropriate for these patients.
Background. Although several studies have compared the patency rates of polytetrafluoroethylene (PTFE) and saphenous vein grafts (SVG) for the above knee location, none have compared the 2 grafts when implanted in the same patient with claudication who needs bilateral above knee femoropopliteal bypasses. Methods. Forty-three patients (86 limbs) with bilateral disabling claudication who had superficial femoral artery occlusion and above knee reconstitution with 2- to 3-vessel runoff were analyzed. Patients were treated on one side with PTFE and on the other side with SVG. They were sequentially assigned to PTFE-SVG alternating with SVG-PTFE. All patients were followed using duplex ultrasound and ankle/brachial indexes at 1 month and every 6 months thereafter. Results. The perioperative complication rates were 5% for PTFE and 12% for SVG. There was no operative death or perioperative amputation for either procedure. The Kaplan-Meier estimate of primary, assisted primary, and secondary patency rates at 72 months were 68%, 68%, and 77% for PTFE and 76%, 83%, and 85% for SVG. There were no statistically significant differences between primary and secondary patency rates for both grafts; however, the assisted primary patency rates were higher for SVG (P < .05). The crude limb salvage rate at 72 months was 98% for PTFE and 98% for SVG. There were no risk factors identified that had an impact on graft patency. Conclusions. PTFE and SVG for above knee bypasses have comparable patency and limb salvage rates in claudicant patients with bilateral superficial femoral artery occlusion and 2- to 3-vessel runoff. This may justify the use of PTFE for above knee locations in these selected patients. (Surgery 1999;126:594602.) From the Department of Surgery, Robert C. Byrd Health Sciences Center Of West Virginia University, Charleston Area Medical Center, Charleston, WV
THROUGHOUT THE LITERATURE it has been widely recognized that, overall, autogenous saphenous vein is the best conduit for infrainguinal bypass. Although it is definitely established that infrageniculate arterial bypass with saphenous vein is the superior conduit, controversy still exists for the preferential use of polytetrafluoroethylene (PTFE) for above knee bypasses. This controversy is stirred by the fact that some authors advocate the use of the saphenous vein or an all-autogenous tissue policy for infrainguinal arterial reconstruction.1-4 Others have found that PTFE can be used for the above knee bypass with equal efficacy.5-12 Most studies that have compared PTFE with saphenous vein in above knee bypasses have studied grafts Presented at the 56th Annual Meeting of the Central Surgical Association, St Louis, Mo, Mar 4-6, 1999. Reprint requests: Ali F. AbuRahma, MD, 3100 MacCorkle Ave SE, Suite 603, Charleston, WV 25304. Copyright © 1999 by Mosby, Inc. 0039-6060/99/$8.00 + 0
594 SURGERY
11/6/99942
that were placed for multiple indications such as claudication, rest pain, or trophic changes, and most are retrospective studies. For many of these indications, runoff is usually less than ideal. When claudicant patients are looked at separately in these studies, most reveal a higher patency and limb salvage rates. In our previous study,13 when reconstruction was done for disabling claudication, a 5-year secondary patency rate of 64% was seen and the limb salvage rate was 97%. Prendiville et al8 showed similar long-term results, and Sterpetti et al5 reported slightly superior results. For those few select patients with disabling claudication that is refractory to medical therapy and in which bypass is performed, one may argue that the PTFE long-term patency rates are comparable to those of the saphenous vein. Rosenthal et al7 showed that for claudication alone, the preferential prosthetic above knee femoropopliteal bypass is a “safe and durable operation.” However, there are no studies in the literature that have compared PTFE with saphenous vein in above knee bypasses
AbuRahma, Robinson, and Holt 595
Surgery Volume 126, Number 4
Table I. Life table for Kaplan-Meier method for PTFE (primary patency) Time interval (mo) 0 1 6 12 18 24 30 36 42 48 54 60 66 72
No. at risk at start
No. failed
No. withdrawn (censored)
Interval patency rate
Cumulative patency rate (%)
43 41 39 39 33 29 27 26 25 23 19 14 10 6
2 0 0 6 4 1 0 0 0 0 0 0 0 0
0 2 0 0 0 1 1 1 2 4 5 4 4 6
0.953488 1 1 0.846154 0.878788 0.965517 1 1 1 1 1 1 1 1
95 95 95 81 71 68 68 68 68 68 68 68 68 68
in patients who need bilateral arterial reconstruction for disabling claudication. In this fashion, the patients become their own control when risk factors and demographics are analyzed for each type of graft to demonstrate whether PTFE can be preferentially used in this situation. Therefore, the purpose of this prospective controlled study is to compare the results of PTFE versus saphenous vein in claudicant patients with bilateral above knee femoropopliteal bypasses. PATIENT POPULATION AND METHODS This study includes 43 patients who had surgery between January 1992 and December 1994 at our institution. These patients had bilateral disabling claudication with failed medical therapy with long superficial femoral artery occlusion and above knee reconstitution with 2- to 3-vessel runoff. Patients were treated on one side with PTFE and on the other side with reversed saphenous vein grafts. These patients were sequentially assigned to PTFE/saphenous vein alternating with saphenous vein/PTFE. All PTFE grafts were 8-mm grafts (W L Gore & Associates, Inc, Flagstaff, Ariz). All grafts were done using regional anesthesia (epidural or spinal). All patients were given 325 mg aspirin in the immediate postoperative period and continued postoperatively. The hospital records were examined for perioperative complications, and all demographics/risk factors were collected. Graft surveillance. Preoperative and postoperative evaluations included lower extremity vascular laboratory studies with ankle/brachial indexes and duplex ultrasonography using Ultramark 9 equipment during the first part of the study and Ultramark 3000 equipment during the last 2 years
Standard error (%) 3.14 3.21 3.29 5.68 6.66 7.14 7.40 7.54 7.69 8.02 8.82 10.28 12.16 15.70
of follow-up (Advanced Technical Laboratory, Inc, Bothell, Wash). Graft surveillance protocol included an immediate postoperative duplex ultrasound with ankle/brachial indexes, and this was repeated at 1 month, 3 months, and every 6 months thereafter. All grafts were scanned throughout their length with systolic and diastolic peak velocities recorded every 5 to 10 cm and at any area of abnormality on B-mode or color interrogation. The proximal inflow and outflow native arteries were similarly scanned. Color duplex ultrasound studies were defined as abnormal if a peak systolic flow velocity was more than twice the peak systolic velocity in the adjacent graft, or if peak systolic velocity throughout the graft was uniformly less than 45 cm/sec. The absence of color flow in these grafts was indicative of a graft occlusion. All abnormal findings on duplex ultrasounds were confirmed using conventional arteriography. Patients with graft stenosis or occlusion were assigned to one of the above intervals at the time the abnormality was detected. All criteria used for this study attempted to conform to the standards suggested by the Ad Hoc Committee of the Joint Council of the Vascular Societies for Reports Dealing with Lower Extremity Ischemia.14 Primary patency was defined as uninterrupted patency of the original graft without any intervention; an assisted primary patency was defined as patency assisted by simple measures (primarily angioplasty); and secondary patency was defined as patency of the original graft that was thrombosed but maintained by the following measures: thrombectomy, thrombolytic therapy, percutaneous angioplasty, or patch graft angioplasty. Occluded grafts that were not salvaged, but were replaced with new grafts, were regarded as occlud-
596 AbuRahma, Robinson, and Holt
Surgery October 1999
Fig 1. Kaplan-Meier curve for time to loss of primary patency for both the PTFE and vein groups.
Fig 2. Kaplan-Meier curve for time to loss of assisted patency for both the PTFE and vein groups.
ed or failed. Operative mortality (early death) and postoperative complications were considered if they occurred within 30 days of the operation. All patients were analyzed to determine limb salvage. When amputations were limited to the toes or transmetatarsal level, they were not considered failures; however, any amputation at the ankle or above was recorded as limb loss. Data analysis. The data were tabulated, and a statistical analysis was performed using the chi-square
test and the Student t test for group comparison. Life table analyses were used to calculate the cumulative patency and limb salvage rates, and groups were compared with the log-rank test. RESULTS The mean age of the group was 65 years (range 46 to 82 years). The demographic and clinical characteristics of the entire group were as follows: 25 (58%) were male, 31 (72%) were smokers, 18
AbuRahma, Robinson, and Holt 597
Surgery Volume 126, Number 4
Fig 3. Kaplan-Meier curve for time to loss of secondary patency for both the PTFE and vein groups.
Table II. Life table for Kaplan-Meier method for vein (primary patency) Time interval (mo) 0 1 6 12 18 24 30 36 42 48 54 60 66 72
No. at risk at start
No. failed
No. withdrawn (censored)
Interval patency rate
43 43 41 40 36 34 31 29 28 27 21 17 13 7
0 0 1 4 2 2 1 0 0 0 0 0 0 0
0 2 0 0 0 1 1 1 1 6 4 4 6 7
1 1 0.97561 0.9 0.944444 0.941176 0.967742 1 1 1 1 1 1 1
Cumulative patency rate (%) 100 100 98 88 83 78 76 76 76 76 76 76 76 76
Standard error (%) 0 0 2.38 4.85 5.71 6.27 6.71 6.94 7.06 7.19 8.15 9.06 10.36 14.12
Student’s t test = 0.60210. P not significant. Degree of freedom is 85.
(42%) had diabetes mellitus, 22 (51%) had hypertension, 17 (40%) had hypercholesterolemia, and 12 (28%) had coronary artery disease. The mean preoperative ankle/brachial indexes were similar for both the PTFE (0.56) and vein groups (0.55). Morbidity and mortality data. The perioperative complication rates were 5% for the PTFE group and 12% for the vein graft group (no statistically significant differences). Two patients (5%) in the vein group had perioperative nonfatal myocardial infarctions, 2 others (5%) had deep vein thrombosis, and 1 patient (2%) in the PTFE group had perioperative graft thrombosis necessitating an imme-
diate thrombectomy. One patient in each group (2%) had significant bleeding that required surgical exploration to control bleeding. There was no perioperative limb loss, with only 1 late limb loss in each group (ie, a limb salvage rate of 98%). There was no perioperative death in this series. However, there were 4 late deaths, 3 as a result of myocardial infarctions and 1 related to cancer (ie, a late survival rate of 91%). Long-term graft patency data. The crude primary patency, assisted primary patency, and secondary patency rates were similar in both the PTFE and vein grafts: 70%, 70%, and 79% and 77%, 84%,
598 AbuRahma, Robinson, and Holt
Surgery October 1999
Table III. Life table for Kaplan-Meier method for PTFE (assisted patency) Time interval (mo) 0 1 12 18 24 30 36 42 48 54 60 66 72
No. at risk at start 43 41 39 34 31 29 28 25 23 19 14 10 6
No. failed
No. withdrawn (censored)
2 0 5 3 1 0 2 0 0 0 0 0 0
0 2 0 0 1 1 1 2 4 5 4 4 6
Interval patency rate 0.953488 1 0.871795 0.911765 0.967742 1 0.928571 1 1 1 1 1 1
Cumulative patency rate (%) 95 95 83 76 73 73 68 68 68 68 68 68 68
Standard error (%) 3.14 3.21 5.47 6.40 6.80 7.03 7.27 7.69 8.02 8.82 10.28 12.16 15.70
Table IV. Life table for Kaplan-Meier method for vein (assisted patency) Time interval (mo) 0 1 12 18 24 30 36 42 48 54 60 66 72
No. at risk at start
No. failed
No. withdrawn (censored)
Interval patency rate
43 43 41 41 39 35 32 31 30 24 20 14 8
0 0 0 2 3 2 0 0 0 0 0 0 0
0 2 0 0 1 1 1 1 6 4 6 6 8
1 1 1 0.95122 0.923077 0.942857 1 1 1 1 1 1 1
Cumulative patency rate (%) 100 100 100 95 88 83 83 83 83 83 83 83 83
Standard error (%) 0 0 0 3.28 4.91 5.81 6.07 6.17 6.27 7.01 7.68 9.18 12.14
Student’s t test = 2.72838. P < .02. Degree of freedom is 85.
and 86%, respectively. The mean time in months to the loss of primary patency, assisted primary patency, and secondary patency were also similar for the PTFE and vein groups (12.9, 12.9, and 15.3 and 16.8, 18, and 18, respectively). The Kaplan-Meier estimate of primary, assisted primary, and secondary patency rates at 72 months were 68%, 68%, and 77% for PTFE and 76%, 83%, and 85% for vein grafts. There were no statistically significant differences between primary patency and secondary patency rates for both groups; however, the assisted primary patency rates were higher for the vein grafts (P < .05). Figs 1, 2, and 3 show the Kaplan-Meier curves for time to the loss of primary patency, assisted patency, and secondary patency for both the PTFE and vein groups. Tables 1 through 6 show the life table analysis for primary patency, assisted primary patency, and secondary patency for both PTFE and vein grafts. Risk factors and patency rates. No specific risk
factors (gender, smoking, diabetes mellitus, hypertension, hypercholesterolemia, or coronary artery disease) were identified that had an impact on graft primary, assisted primary or secondary patency. However, patients with a loss of patency on one side appeared to be at a higher risk for loss of patency on the opposite side. DISCUSSION Many investigators have concluded that PTFE above knee femoropopliteal bypass patency rates can be comparable to saphenous vein patency rates.5,6,10 Most of these studies have taken all types of vascular patients and indications for surgery into account. Authors continue to suggest that prosthetic above knee femoropopliteal bypass may be the preferential choice in a certain select group of patients, ie, claudicant patients with normal to near normal runoff (2- to 3-vessel), but argue that more studies need to be performed.5,8,12 No studies
AbuRahma, Robinson, and Holt 599
Surgery Volume 126, Number 4
Table V. Life table for Kaplan-Meier method for PTFE (secondary patency) Time interval (mo) 0 1 12 18 24 30 36 42 48 54 60 66 72
No. at risk at start 43 43 41 40 38 34 32 31 29 24 18 13 7
No. failed 0 0 1 2 3 1 0 0 1 1 0 0 0
No. withdrawn (censored) 0 2 0 0 1 1 1 2 4 5 5 6 7
before this current report have compared PTFE with saphenous vein in the same patient for the same indication for surgery with comparable runoff. Most vascular surgeons would agree that a patient with superficial artery disease with 2- to 3vessel runoff would be an “ideal” candidate for vascular reconstruction. In these ideal candidates, one could argue that PTFE could be preferentially selected over saphenous vein as the conduit of choice. Surgeons who favor preferential use of synthetic grafts in above knee femoropopliteal bypasses often base their argument on saving the vein for later use, either for coronary artery bypass revascularization or for ipsilateral below knee bypass when previous above or below knee bypasses have failed. With the advancement of cardiac revascularization techniques, ie, percutaneous treatment or use of internal mammary artery, authors disagree as to whether the argument for preserving the vein for future coronary bypass is valid3,5; however, the literature has shown a range of 14% to 28% of patients who undergo infrainguinal bypass will eventually need coronary artery reconstruction.7 Taylor et al2 support the practice of reversed saphenous vein as the procedure of choice for infrainguinal bypass, in spite of their findings that 28% of patients in their series had a missing greater saphenous vein on the side of need of revascularization, due to previous coronary artery bypass grafting that resulted in the use of an alternative conduit. Overall, the question remains whether it is beneficial to preserve the saphenous vein for future use, whether for distal bypass or coronary reconstruction. Some authors have shown that a staged procedure in which a preferential prosthetic above knee bypass followed by autogenous saphenous vein below the knee bypass for recurrent ischemia
Interval patency rate 1 1 0.97561 0.95 0.921053 0.970588 1 1 0.965517 0.958333 1 1 1
Cumulative patency rate (%) 100 100 98 93 85 83 83 83 80 77 77 77 77
Standard error (%) 0 0 2.38 3.96 5.30 5.88 6.06 6.16 6.64 7.56 8.73 10.27 14.00
can improve limb salvage rates.4,11 Although Plecha et al4 showed that a secondary autogenous vein reconstruction can salvage a failed above knee PTFE graft, they did not believe there was an advantage to this approach. Rosenthal et al,11 however, reported an improved overall limb salvage rate for staged arterial reconstruction in the infrainguinal setting versus primary saphenous vein (86% versus 78%, respectively). One must also consider the possibility of affecting future saphenous vein bypass patency rates in a limb that has previously had a prosthetic graft. Some may argue that patency rates are decreased in this population, but one must consider that these patients have also had progression of their disease and the overall quality of the native artery and runoff is likely to be much poorer. Edwards et al,15 however, achieved a 5-year limb salvage rate of 90% when autogenous reversed saphenous vein was used as a secondary bypass for a previously failed infrainguinal bypass. No study using PTFE as a second conduit for limb salvage has come close to achieving these rates. A final argument can be made that the life expectancy of most patients with vascular disease, especially with critical ischemia, is decreased, and a single prosthetic above knee bypass may be the only procedure needed. Risk factors that have been shown to affect the patency rates of infrainguinal bypasses include habitual smoking, diabetes, poor runoff, indication for surgery, ankle/brachial index, and level of distal anastomosis. Our previous study reported on the correlation of ankle/brachial indexes to graft patency. When the ankle/brachial index was 0.5 or greater, primary and secondary patency rates were 57% and 68%, respectively, compared with a primary patency rate of 37% and a secondary patency rate of 56% for patients with an ankle/brachial
600 AbuRahma, Robinson, and Holt
Surgery October 1999
Table VI. Life table for Kaplan-Meier method for vein (secondary patency) Time interval (mo) 0 1 12 18 24 30 36 42 48 54 60 66 72
No. at risk at start
No. failed
43 43 41 41 41 38 34 32 31 24 20 14 8
0 0 0 0 2 3 1 0 0 0 0 0 0
No. withdrawn (censored) 0 2 0 0 1 1 1 1 7 4 6 6 8
Interval patency rate 1 1 1 1 0.95122 0.921053 0.970588 1 1 1 1 1 1
Cumulative patency rate (%) 100 100 100 100 95 88 85 85 85 85 85 85 85
Standard error (%) 0 0 0 0 3.28 5.00 5.64 5.82 5.91 6.71 7.36 8.79 11.63
Student’s t test = 0.81777. P not significant. Degree of freedom is 85.
index of less than 0.5.13 This present study is interesting in that ankle/brachial indexes had no correlation to patency. This is likely due to the fact that each patient had good runoff and the mean ankle/brachial index for the entire set was 0.55 with a low of 0.38 and high of 1.0. Prendiville et al8 reported that continued habitual smoking and patients with diabetes mellitus adversely affected primary patency rates. Smokers had a 5-year patency rate of 40% as compared with 66% for nonsmokers, and diabetic patients had a decreased 5year patency rate. Again, our study showed no correlation between these factors and patency of either graft. Because runoff, indication for surgery, and level of anastomosis were the same in each subset of patients, we found no risk factors in our study that had an impact on patency rates. Our study did reveal, however, that patients with a loss of patency on one side were at a higher risk for loss of patency on the other side. The evidence suggests that these patients may have an underlying defect that makes them a higher risk for graft failure. Edwards et al15 advocate using an aggressive approach when evaluating patients with failed grafts who need reoperation by checking for hypercoagulable states, blood lipid levels, and plasma homocysteine levels. By identifying these factors, surgeons could possibly improve reoperation patency rates in these high-risk patients. Recently, Allen et al16 reviewed their experience in 239 patients with claudication as a result of superficial femoral artery occlusion. Femoropopliteal reconstruction was performed with saphenous vein to below knee popliteal artery in 66 patients. PTFE was used in 128 patients as a bypass graft to the above knee popliteal artery, and 45 patients had a
PTFE graft to the below knee popliteal artery. All patients were involved in a postoperative graft surveillance program with graft revision, when appropriate. The primary patency rates at 5 years for above knee PTFE, below knee PTFE, and below knee vein were 58%, 55%, and 60%, respectively, and this was not significantly different among the graft groups. The 5-year secondary patency rates for above knee PTFE grafts were 79%, 73% for below knee PTFE, and 74% for below knee vein (not statistically significant). They concluded that long-term patency rates for femoropopliteal bypasses in patients with claudication were similar for PTFE and autogenous saphenous vein grafts. Also one must take into account the morbidity of each procedure. For prosthetic above knee bypasses, 2 small incisions are made for placement of the graft, in contrast to saphenous vein grafting where multiple incisions are made down the leg, often extending below the knee for harvesting adequate graft material. This may not only affect postoperative healing, but it also increases the time in the operating room, putting additional stress on the already sensitive vascular patient. As you can see in our results, the complication rates were higher for vein grafting as compared with PTFE, 12% to 5%, respectively, with 5% of these complications being of a cardiac origin. Even if these rates were not statistically significant, it may suggest the need for the simplest and fastest operation. As life expectancy increases because of improved medical therapy, more patients are going to need infrainguinal arterial reconstruction. As vascular advancements are made, the challenge of restoring blood flow to the lower extremity is going to continually grow. As endovascular techniques, throm-
AbuRahma, Robinson, and Holt 601
Surgery Volume 126, Number 4
bolysis, and new graft materials improve, all will play a role in vascular reconstruction, especially in superficial femoral artery disease. In conclusion, PTFE and saphenous vein grafts for above knee bypasses have comparable patency and limb salvage rates in claudicant patients with bilateral superficial femoral artery occlusion with 2to 3-vessel runoff. This may justify the use of PTFE for above knee locations in these selected patients. We believe that preferential use of PTFE in the above knee setting in claudicant patients is an acceptable practice. REFERENCES 1. Kent KC, Whittemore AD, Mannick JA. Short-term and midterm results of an all-autogenous tissue policy for infrainguinal reconstruction. J Vasc Surg 1989;9:107-14. 2. Taylor LM Jr, Edwards JM, Porter JM. Present status of reversed vein bypass grafting: five-year results of a modern series. J Vasc Surg 1990;4:193-206. 3. Michaels JA. Choice of material for above-knee femoropopliteal bypass graft. Br J Surg 1989;76:7-14. 4. Plecha EJ, Freischlag JA, Seabrook GR, Towne JB. Femoropopliteal bypass revisited: an analysis of 138 cases. Cardiovasc Surg 1996;4:195-9. 5. Sterpetti AV, Schultz RD, Feldhaus RJ, Peetz DJ Jr. Sevenyear experience with polytetrafluoroethylene as above-knee femoropopliteal bypass graft: is it worthwhile to preserve the autogenous saphenous vein? J Vasc Surg 1985;2:907-12. 6. Veith FJ, Gupta SK, Ascer E, White-Flores S, Samson RH, Soher LA, et al. Six-year prospective multicenter randomized comparison of autologous saphenous vein and expanded polytetrafluoroethylene grafts in infrainguinal arterial reconstructions. J Vasc Surg 1986;3:104-14. 7. Rosenthal D, Evans D, McKinsey J, et al. Prosthetic aboveknee femoropopliteal bypass for intermittent claudication. J Cardiovasc Surg 1990;31:462-8. 8. Prendiville EJ, Yeager A, O’Donnell TF Jr, Coleman JC, Jaworek A, Callow AD, et al. Long-term results with the above-knee popliteal expanded polytetrafluoroethylene graft. J Vasc Surg 1990;11:517-24. 9. Abbott WA. Prosthetic above-knee femoral-popliteal bypass: indications and choice of graft. Semin Vasc Surg 1997;10:3-7. 10. Bergan JJ, Veith FJ, Bernard VM, Yao JST, et al. Randomization of autogenous vein and polytetrafluoroethylene grafts in femoral-distal reconstruction. Surgery 1982;92:921-30. 11. Rosen RC, Johnson WC, Bush HL Jr, Cho SI, O’Hara ET, Nabseth DC. Staged infrainguinal revascularization: initial prosthetic above-knee bypass followed by a distal vein bypass for recurrent ischemia—a valid concept for extending limb salvage? Am J Surg 1986;152:224-30. 12. Quinones-Baldrich WJ, Prego AA, Ucelay-Gomez R, Freischlag JA, Ahn SS, Baker JD, et al. Long-term results of infrainguinal revascularization with polytetrafluoroethylene: a ten-year experience. J Vasc Surg 1992;16:209-17. 13. AbuRahma AF, Robinson PA, Stuart SP, Witsberger TA, Stewart WA, Boland JP. Polytetrafluoroethylene grafts in infrainguinal arterial revascularization: factors affecting outcome. Arch Surg 1993;128:417-22. 14. Ad Hoc Committee on Reporting Standards, Society for Vascular Surgery/North American Chapter, International Society for Cardiovascular Surgery. Suggested standards for reports dealing with lower extremity ischemia. J Vasc Surg 1986;4:80-94.
15. Edwards JE, Taylor LM, Porter JM. Treatment of failed lower extremity bypass grafts with new autogenous vein bypass grafting. J Vasc Surg 1990;11:136-45. 16. Allen BR, Reilly JM, Rubin BG, Thompson RW, Anderson CB, Flye MW, Sicard GA. Femoropopliteal bypass for claudication: vein vs PTFE. Ann Vasc Surg 1996;10:178-85.
DISCUSSION Dr Gregorio A. Sicard (St Louis, Mo). I congratulate you for an important and landmark contribution to the surgical treatment of lower extremity claudication. The overall approach to the treatment of claudication continues to be the source of significant debate in the vascular literature. Many questions are debated. First, should claudicants undergo any form of surgical therapy? Should risk modification and a gradual exercise program be the gold standard from which to analyze and compare surgical results? Second, should the incapacitating lower extremity claudicant with isolated short segment superficial femoral artery stenosis have endoluminal techniques as another option to treat the disease? If a surgical option is chosen for superficial femoral artery disease, which should be the conduit of choice: saphenous vein, PTFE, or Dacron? In this prospective controlled trial, 43 patients were sequentially assigned to PTFE/saphenous vein alternating with saphenous vein PTFE above the knee bypass and followed longitudinally with a strict surveillance protocol to identify the so-called failing graft. They achieved excellent results and similar 72-month patency rates for saphenous vein and PTFE, respectively. I have a few questions for you. Do you have another cohort of patients with bilateral claudication caused by superficial femoral artery disease in which endoluminal treatment is compared with open surgical revascularization? As you well know, the endoluminal techniques are being done by many specialties, and the superficial femoral artery has been one of the sites that has been targeted for these procedures. If you do, what are the results and should a bypass be considered in patients with superficial femoral artery occlusion versus endoluminal techniques? The overall experience with surveillance protocols is not very good for prosthetic grafts compared with autologous vein. In your results, the mean time in months to loss of primary and primary assisted patency was approximately 12 months. This gives you a statistical difference for assisted primary patency of 68% for PTFE and 83% for vein, respectively. In other words, if looking at it in another way, there is a 6-month difference in which the graft fails. In your manuscript, the PTFE failed 6 months earlier than the saphenous vein bypasses. What other techniques do you suggest we use to evaluate the PTFE groups in this 12-month period, which seems to be very critical for
602 AbuRahma, Robinson, and Holt
failure of prosthetic grafts? Are exercise Dopplers and waveform evaluation of help in identifying the so-called failing PTFE graft between the 12- and 18-month period, or should arteriography be recommended? Last, could you expand on the observation that patients who occlude 1 graft are at greater risk to lose the contralateral graft? Does this observation apply to the initial loss of either conduit, and last, when one conduit thromboses, should the patient be systemically coagulated to protect the other? Dr AbuRahma. I totally agree with you that all patients with claudication should undergo risk modification and a gradual exercise program with or without pharmacologic therapy. If this fails in patients who have disabling claudication, an alternative therapy, such as a bypass or endovascular therapy, should be explored. We do not have another cohort of patients with bilateral claudication caused by superficial femoral artery disease in which endoluminal treatment is compared with open surgical revascularization. However, a significant number of our patients with superficial femoral artery disease are presently treated by some form of endovascular therapy, eg, percutaneous transluminal balloon angioplasty with or without stenting. I agree with your statement that the overall experience with surveillance protocol is not very good for prosthetic grafts compared with autologous veins. A significant number of PTFE grafts are discovered when they are thrombosed rather than at the stenosis stage. This may explain the differences between the assisted primary patency between the vein and PTFE grafts. Whether exercise Doppler and waveform evaluations would be helpful in identifying the so-called failing PTFE graft between a 12- and 18-month period is difficult to predict; however, it may be a good idea to try. We do not have any specific answers from our study as to why if one graft fails, the contralateral graft usually fails. However, this may be partly explained by unidentifiable hypercoagulable states or other risk factors that we could not document in our study. It may be a good idea to anticoagulate patients after one graft fails.
Surgery October 1999
Dr John D. Corson (Iowa City, Iowa). Were these procedures all done simultaneously or were they staged? If they were staged, how much time was there between the 2 procedures? Dr AbuRahma. They were staged, generally between 6 to 8 weeks. Dr William H. Baker (Maywood, Ill). If somebody comes into your hospital now with intermittent claudication available vein, which operation do you perform if they have intermittent claudication? Dr AbuRahma. If they are ideal candidates, similar to the patients described in our study, ie, claudicant with above knee reconstruction and good below knee runoff (2- to 3-vessel runoff), I would not hesitate to recommend a PTFE graft. Dr Eric D. Endean (Lexington, Ky). In grafts that failed, did you have to convert a vein graft to a PTFE or vice versa in any cases? Were any inflow procedures required for any of your grafts and did this affect patency? Dr AbuRahma. None of these had an inflow problem that required any type of correction for the graft that failed. In regards to converting vein graft to PTFE or vice versa, that was done only in patients with PTFE who had above knee grafts. If the graft failed using simple measures, such as patch angioplasty, they were converted into vein grafts for the below knee locations. Dr James R. Debord (Peoria, Ill). You randomized in these staged procedures PTFE-vein versus vein-PTFE. Did you do anything to randomize left versus right? There could be some differences in the runoff with a best-leg/worst-leg scenario. Could there have been any bias in which leg you picked for PTFE versus vein? Dr AbuRahma. We did not randomize regarding left verus right. In regards to your second question as to why we would measure limb salvage rate in patients who are all claudicant with limbs that are not in any jeopardy, this is true at the time they were initially seen for the first intervention. However, during the 5-year follow-up, some of these patients (a few percent) could have limb loss, even with intervention using a vein or PTFE. Therefore, we thought a limb salvage rate was appropriate for these patients.