Primary graft patency. Horizontal bars indicate values are given as mean ± SE.
Secondary graft patency. Horizontal bars indicate values are given as mean ± SE.
Limb salvage. Horizontal bars indicate values are given as mean ± SE.
Survival. Horizontal bars indicate values are given as mean ± SE.
Primary graft patency for diabetic and nondiabetic patients. Horizontal bars indicate values are given as mean ± SE.
Popliteal vs curural primary patency. Horizontal bars indicate values are given as mean ± SE.
Lauterbach SR, Torres GA, Andros G, Oblath RW. Infragenicular Polytetrafluoroethylene Bypass With Distal Vein Cuffs for Limb SalvageA Contemporary Series. Arch Surg. 2005;140(5):487-494. doi:10.1001/archsurg.140.5.487
Infragenicular polytetrafluoroethylene (PTFE)–venous cuff bypass grafting provides acceptable graft patency and limb salvage rates for limb salvage.
Retrospective clinical review of a consecutive series.
Vascular surgical practice during the interval October 1, 2000, to September 1, 2004.
Fifty-one male and 49 female patients whose mean age was 76.9 years were operated on for tissue loss (67%), chronic rest pain (28%), and severe claudication (6%). Fifty-two percent of patients were diabetic and 49% had undergone previous leg bypass surgery. All patients had absent or inadequate greater saphenous vein, and 84 patients had absent or inadequate arm vein.
One hundred five infragenicular PTFE bypasses were performed in these 100 patients. Distal targets were the infragenicular popliteal (40), posterior tibial (35), anterior tibial (16), and peroneal arteries (14). Sixty-eight venous cuffs were constructed from lesser saphenous vein.
Main Outcome Measures
Graft patency, limb salvage, and patient survival were analyzed.
Twelve early graft failures resulted in 7 leg amputations. The mean ± SE 3-year primary patency and limb salvage rates were 64.4% ± 12.8% and 74.4% ± 11.9%, respectively. Perioperative mortality was 2.9% and 3-year survival was 38%. Graft follow-up ranged from 1 to 47 months with a mean of 13 months using life-table methods.
For patients requiring arterial revascularization for limb salvage, in which autologous venous conduit is unavailable, distal venous cuff–PTFE bypass provides acceptable patency and limb salvage rates when viewed in the context of short life expectancy for these elderly patients.
When autogenous vein from the leg or arm is unavailable for lower extremity revascularization, alternative strategies for enhancing perfusion must be used for limb salvage. Should only a marginal increase in the circulation be required, localized procedures such as iliofemoral endarterectomy often suffice. If restoration of femoropopliteal continuity is required, a catheter-based minimally invasive procedure such as percutaneous balloon angioplasty, if feasible, can offer potential benefits such as reduced morbidity. In the vast majority of patients, however, in which the patient is handicapped by a TransAtlantic Inter-Society Consensus section D (TASC D)1 arterial lesion, a more extensive procedure such as a long bypass graft is nearly always required. The procedure and conduit will ideally provide effective and durable reperfusion with a low complication rate. We maintain a strong posture toward all-autologous venous conduit use for revascularization of the lower extremities. This includes, ideally, the greater saphenous vein2 and, additionally, arm veins, often spliced together from 2 to 3 segments.3,4 We have found that patients were coming to revascularization at an older age and after failed bypass without adequate autogenous conduit for limb salvage. Because they were classified as having TASC D arterial lesions and were not candidates for endoluminal therapy, we sought a different strategy previously unused by our group. In this article we describe our experience treating a contemporary consecutive series of patients with critical lower extremity ischemia using polytetrafluoroethylene (PTFE) grafts and distal vein cuffs.
All patients who underwent lower extremity bypass grafting using PTFE grafts with a distal venous cuff for chronic arterial ischemia from October 1, 2000, to September 1, 2004, in our practice were retrospectively reviewed for this article. The patients’ office and hospital medical records were reviewed. Patient history, demographics, comorbidities, anatomical data, operative details, postoperative complications, and follow-up dates were recorded. Life-table methods were used to calculate primary and secondary graft patency, limb salvage, and patient survival rates. Log-rank testing was used to compare life-table results.
Patients underwent preoperative venous duplex vein mapping of the upper and lower extremities in most cases and surgical bypass was performed in a standardized method after preoperative contrast angiography was performed and reviewed. Venous conduit was absent or inadequate by duplex mapping results and/or operative exploration. Criteria for venous inadequacy included small diameter (<2.5 mm), insufficient total available length, failure to distend, or simply sclerosis. The venous cuff was created at the distal arterial target as described by Miller et al5 and Siegman6 to which the PTFE graft was sutured end to end with continuous fine polypropylene (Prolene; Ethicon Inc, Somerville, NJ). Duplex graft surveillance protocols were not used postoperatively. Patients who were receiving warfarin sodium therapy preoperatively received warfarin therapy postoperatively. Clopidogrel and/or aspirin were routinely used postoperatively on all patients. Graft patency was determined by clinical examination (graft and distal pulse palpation) supplemented by duplex scanning, angiography, and/or operative findings. Amputation was considered major if it was below-knee or higher and was recorded for limb salvage calculations. Patients were considered lost to follow-up only after contact attempts were exhausted during the follow-up period ending September 1, 2004.
One hundred five bypass grafts in 100 patients were performed during this study period by 4 surgeons (all of us) in a single group. The patient cohort was composed of 51 males and 49 females, with a mean age of 76.9 years, and was operated on for chronic critical arterial ischemia in all cases (70 grade III, 29 grade II, and 6 grade I).7 Fifty-two patients had diabetes mellitus and 10 were undergoing hemodialysis. Forty-nine percent of these patients had a history of previous leg bypass surgery and 29% of previous inflow procedures; 15% had concomitant ipsilateral iliofemoral thromboendarterectomy. All cases involved absent (54 patients) or inadequate (51 patients) greater saphenous vein and 93% of cases had inadequate (80 patients) or absent (18 patients) arm vein.
Inflow sources for the bypass graft included common femoral (77 patients), superficial femoral (7 patients), deep femoral (7 patients), iliac artery (2 patients), or previous inflow graft (12 patients). The distal arterial targets were the infragenicular popliteal (40 patients), anterior tibial (16 patients), posterior tibial (35 patients), and peroneal (14 patients). All PTFE grafts were either 7 mm (71 patients) or 6 mm (34 patients) unringed stretch-PTFE (Gore-Tex; W. L. Gore, Flagstaff, Ariz). The venous cuff was constructed of lesser saphenous (68 patients), remnant greater saphenous (15 patients), basilic (8 patients), or other veins (14 patients). All patients underwent general anesthesia for their surgical bypass (Table 1).
The 30-day operative mortality rate was 2.9%. Perioperative complications included 14 local wound complications and 11 systemic complications. Four grafts became infected requiring removal in 3 patients resulting in 2 leg amputations and graft replacement in 1. All 4 of these patients died within the follow-up period of this study (Table 2).
The length of follow-up ranged from 1 to 1450 days, with a mean of 13 months. The mean (± SE) 1- and 3-year primary graft patency rates were 79.3% (± 5.6%) and 64.4% (± 12.8%), respectively (Figure 1). Seven of 21 grafts that thrombosed during this period were revised, yielding 1- and 3-year secondary patency rates of 81.2% (± 5.3%) and 65% (± 11.6%), respectively (Figure 2). There were 12 early graft failures resulting in 7 of 11 total leg amputations in the series. The mean (± SE) 1- and 3-year limb salvage rates were 90.2% (± 4.4%) and 74.4% (± 11.9%), respectively (Figure 3). The mean (± SE) life-table derived 1- and 3-year patient mortality rates were 31.6% (± 5.4%) and 62% (± 8.6%), respectively (Figure 4).
Log-rank testing showed no statistical significance between 3-year primary patency rates for diabetic patients and nondiabetic patients (60% vs 73%, P<.95). (Figure 5). Likewise, there was no statistical significance between 3-year primary patency rates between popliteal grafts and tibial or peroneal grafts (60% vs 69%, P<.95) (Figure 6).
Chronic lower extremity arterial occlusive disease all too often poses a therapeutic challenge for limb salvage when adequate venous conduit is unavailable and long arterial segments with TASC D lesions are occluded. We routinely perform percutaneous interventions such as percutaneous transluminal angioplasty (PTA) or subintimal angioplasty for TASC A and B lesions. None of these patients described herein qualified for endovascular therapy; in fact, all patients were treated for TASC D lesions with most lesions exceeding 20 cm. Clinical indication was rarely claudication likely because of age and disability, whereas the majority had ischemic lesions and or distal gangrene, ulceration, or rest pain. The 105 bypass cases in this series is approximately 11% of all lower extremity arterial bypass cases performed in our practice during the same 47-month interval.
This aged population reported with a mean age of 76.9 years is nearly a full decade older than the populations in many series previously reported in the literature focusing on infrapopliteal PTFE arterial bypassing with and without venous cuffs or patches.8- 11 These patients, 52% diabetic, with significant other comorbidities, highlight the fragility of our cohort. What is striking is the calculated estimated mortality rate for all patients—nearly a third of patients are expected to have died at 1 year and nearly two thirds at 3 years despite an operative mortality rate of less than 3%. Perhaps we are observing the fact that these patients’ overall life expectancies are approaching their estimated norm and survival would be small for this cohort irrespective of surgical intervention for limb salvage. Data from the Centers for Disease Control and Prevention demonstrated the average remaining life expectancy for 75-year-olds in the United States for males and females for all races in 2001 was 11½ years.12
Attempting lower extremity bypass surgery vs primary amputation in medically challenged patients was studied by Ouriel et al.13 The patients who underwent bypass surgery experienced reduced operative mortality, shorter lengths of hospital stay, and longer overall survival than patients whose limbs were amputated.13 Raviola et al14 demonstrated no cost benefit to primary amputation when compared with arterial leg bypass surgery for limb salvage and recommended that cost should not be used as a factor to deny attempts at limb salvage. Even though the limb salvage rates are acceptable, our calculated survival rates are sobering and make decision making with patients and their families even more complex. It has been our experience that the vast majority of patients and their families consistently request attempts at limb salvage rather than primary amputation.
One- and 3-year graft primary patency rates of 79% and 64%, respectively, with 1- and 3-year limb salvage rates of 90% and 74%, respectively, compare favorably with modern series of similar bypass procedures. Stonebridge et al10 reported 1- and 2-year primary graft patency rates of 80% and 52%, respectively, on 57 PTFE–infrapopliteal venous cuffed grafts (49 popliteal and 8 tibial) in 1997 from the United Kingdom. This translated into a 2-year limb salvage rate of 62%. Stonebridge and colleagues15 later reported on 89 tibial grafts with a 2-year primary patency rate of 32% and a 2-year 44% limb salvage rate. In contrast, we observed no statistically significant difference in graft patency between popliteal and infrapopliteal distal anastomotic sites (P<.95) (Figure 6). Perhaps this was due to the fact that most popliteal grafts had single-vessel runoff and behaved more like grafts placed directly to the sole tibial or peroneal vessel.
Our data digress from the findings of Kreienberg et al16 and Miller et al17 with respect to the discrepancy between primary patency and limb salvage. Kreienberg et al16 reported a 3-year primary patency rate of 38% and a 3-year limb salvage rate of 92% for 59 PTFE grafts with a venous cuff in 2000. In 2002, Kreienberg and colleagues9 reported on 20 similar grafts with 2-year primary patency of 49% and 2-year limb salvage of 85%. Additionally, Miller and colleagues reported a 3-year primary patency rate of 29% and limb salvage rate of 64% in 201 grafts to crural targets in 199317 and a 57% 3-year primary patency rate for 177 popliteal cuffed grafts in 1995.18 We observed only 10% greater 3-year limb salvage rates than graft patency rates vs a difference of some 40% in the aforementioned studies. It appears that failed grafts in our series led to more leg amputations despite an improved primary patency rate. Patient selection may play a role as 4 of 11 leg amputations were in diabetic patients with patent grafts. Perhaps these limbs were simply not salvageable owing to extensive necrosis and should have been amputated primarily.
Diabetic patients seemed to do less well with these grafts in terms of patency and limb salvage although not statistically different (Figure 5). Nine of 11 patients who had leg amputation were diabetic, 4 of these patients had patent grafts. The diabetic patients commonly had ischemic necrotizing lesions that were more extensive than would be anticipated on clinical grounds alone. These neuro-ischemic lesions were often in challenging locations such as the heel that failed to achieve direct reperfusion because the posterior tibial–plantar axis was occluded. Moreover, the presence of clinical infection was commonplace and added to the severity of neuro-ischemic gangrene and ulceration.
When compared with PTFE infrapopliteal bypass without a venous cuff, the present results suggest more favorable outcomes. The series by Hamdan et al19 of 47 tibial grafts had 3-year primary patency and limb salvage rates of 58% and 63%, respectively. Schweiger et al8 reported a 5-year primary patency rate and limb salvage rate of 23% and 51%, respectively, on 211 grafts, and Klinkert et al20 had an 18% 5-year primary patency and 62% limb salvage rates with 83 tibial grafts. Additionally, Quinones-Baldrich et al21 reported 3-year primary patency rates of 22% with limb salvage at that same interval of 37% in 28 PTFE infrapopliteal grafts done for limb salvage out of a large series of 322 PTFE infrainguinal revascularizations and Parsons et al22 reported 3-year primary patency and limb salvage rates of 39% and 71%, respectively, for 66 bypass grafts to infrapopliteal arteries.
Synthetic bypass grafts including PTFE are used with surprising frequently for revascularization of the lower extremities despite well-documented inferior patency and limb salvage rates compared with autologous vein, including arm vein.23,24 Merrell and Gusberg25 reported data over a recent 3-year period from a national perspective within the United States and found 41% of all infrainguinal grafts were synthetic: 60% of femoropopliteal grafts and 14% of infrapopliteal grafts. There was considerable geographic variance throughout the country. In some practices, the use of ectopic vein is not performed because of either unfamiliarity with the technique or possibly because of a lack of manpower. We believe the PTFE–venous cuffed graft is a better strategy than the uncuffed graft, and it should be considered by surgeons performing lower extremity arterial surgery for limb salvage.
Limb salvage is unquestionably the primary goal for patients requiring these grafts, and multiple interventions are often required. When many of these venous cuffed grafts thrombose, the runoff arteries are preserved because the venous cuff remains patent. Raptis and Miller18 observed this in approximately 50% of thrombosed grafts from their series of nearly 600 cases. Indeed, that has been our experience at surgical exploration; however, we only revised 7 thrombosed grafts. More commonly, if another attempt at bypass was considered, a new graft was constructed to a distal target rather than thrombectomy and extension grafting from the primary graft’s venous cuff because sufficient venous conduit was most often unavailable. We observed secondary patency rates of 81% at 1 year 81% and 65% at 3 years, which were not significantly different from our primary patency rates (P>.95), most likely caused by the lack of revisions done in this series. Five of 7 patients experienced extended patency from the secondary intervention and 1 patient had limb amputation despite successful graft thrombectomy. Because the grafts do not demonstrate a clinically evident “failing phase” prior to thrombosis, we do not use graft surveillance techniques as is routinely done with vein grafts.26
In our opinion, the advantages of the PTFE–venous cuff technique include reduced surgical trauma to the leg and a potentially better compliance match of vein to artery at the distal anastomosis with improved flow dynamics.27,28 Most importantly, the ability to inspect the distal anastomosis through the venous cuff ensures optimal construction. There is an available synthetic distal cuff graft (Distaflo; Impra Inc, Tempe, Ariz) that resembles a venous cuff PTFE graft but that does not allow complete inspection of the distal anastomosis. Recent 1-year patency of 39% with 50% limb salvage for 50 grafts (20 popliteal and 30 tibial) have been reported by Fisher et al.29 More favorable results have recently been reported by Panneton et al30 who observed no significant difference between the vascular prosthetic and PTFE–distal venous cuffed grafts as measured by 1- and 2-year patency rates and limb salvage. Our data, at 3 years’ follow-up, are similar to their early data with regard to limb salvage.
Despite commitment to autogenous venous conduits for complex, distal, and multiple redo lower extremity revascularizations, we recognize the need for nonautogenous options for a small percentage of cases. Among the patients who require an infrageniculate bypass for limb salvage in whom a long autogenous vein is unavailable, we believe the PTFE–distal venous cuff is the best option. It has the advantage of allowing unimpeded visualization of the vein-to-artery suture line, conforms well to deep crural arteries, and is durable in a patient cohort with a reduced life expectancy.
Correspondence: Stephen R. Lauterbach, MD, 2701 W Alameda Ave, Suite 606, Burbank, CA 91505 (firstname.lastname@example.org).
Accepted for Publication: January 18, 2005.
Previous Presentation: This paper was presented at the 112th Scientific Session of the Western Surgical Association; November 10, 2004; Las Vegas, Nev; and is published after peer review and revision. The discussions that follow this article are based on the originally submitted manuscript and not the revised manuscript.
Acknowledgment: We thank Michael Lauterbach of Oslo, Norway, for his assistance with statistical analysis and data presentation.
R. James Valentine, MD, Dallas, Tex: The authors are to be congratulated for identifying and successfully treating a subgroup of elderly patients with challenging vascular problems. These patients had a high prevalence of diabetes mellitus and prior vascular operations, but their most relevant common feature was a lack of adequate vein conduit. Prosthetic leg bypass was the only option to avoid major amputation in most subjects, and the authors used distal vein cuffs in the hopes of improving the overall patency rates. The use of cuffs, patches, boots, and other vein configurations at the distal anastomosis is not new. The cuff technique used in this study was first reported in 1984 by Justin Miller from Australia. All of these distal anastomotic configurations appear to improve patency rates of bypasses performed to arteries below the knee, probably by reducing the effects of intimalhyperplasia. The present results at 1 year of follow-up are in keeping with previous reports of cuffed bypasses. The number of patients at risk at 3 years is too small to evaluate, but the results at 900 days (2½ years) are comparable with other studies.
The authors did not perform routine duplex graft surveillance in these patients. This is within acceptable standards of care. Unlike vein bypasses, less than 20% of patients with prosthetic grafts will have an identifiable lesion prior to graft thrombosis. Unfortunately, most prosthetic graft occlusions are sudden and result in severe, limb-threatening ischemia. The use of warfarin therapy has been associated with improved patency and a reduction in ischemic consequences of prosthetic graft failure. It has been our practice to maintain our patients with infrainguinal prosthetic grafts on both aspirin and warfarin therapy. This brings me to my first question: what was your postoperative routine? Did your patients receive routine anticoagulation therapy or were their conditions maintained on antiplatelet therapy?
The patients in this series represented a bit of a mixed bag, in that there were a multitude of different anastomotic sites, bypass lengths, and operative indications. Runoff scores were not reported. Most grafts presumably crossed the knee joint, but the authors did not use externally supported PTFE grafts. The authors found no differences in the patency rate between diabetic and nondiabetic patients, or between bypasses created to the popliteal and crural arteries. Other markers for failure such as previous graft thrombosis were not evaluated. My second question is whether there was a difference in graft patency rate between patients who had prior bypasses compared with those who had not.
My final comment and question refer to the late mortality in this series. The low operative mortality and substantial late death rate attest to excellent clinical selection. These results underscore the importance of fitting an operation to an indication. The cuffed prosthesis appears to be an ideal option for a population with limited life expectancy, regardless of vein availability. Would the authors consider expanding their indications to patients with limited life expectancy?
Dr Lauterbach: Dr Valentine, with respect to our management of these patients postoperatively, patients receiving warfarin therapy preoperatively for other causes were maintained on warfarin therapy postoperatively—approximately 15% of patients. These patients did not receive anticoagulation therapy specifically for their grafts. In most cases, we started a postoperative regimen of clopidogrel (Plavix) for these patients. Additionally, the majority of patients in this series were receiving aspirin therapy. I agree that warfarin anticoagulation therapy should be studied more specifically. I think those patients with synthetic grafts who received anticoagulation therapy and who did thrombose are in less jeopardy at the time of thrombosis. I do not know whether starting these patients on a regimen of warfarin therapy will improve the patency rate.
With regard to comparing patients who had undergone previous bypass surgery with those who did not, I thank you for bringing that to my attention for I had the opportunity to review the data. With regard to the actual patency rate, at 1 year those patients who did not have previous bypass surgery had an 83% primary patency rate and at 3 years that again was 83%, accepting, as you pointed out, by the time we got out to 3 years we had fewer than 10 patients that were at risk, so I think we do not have enough patients as we move out on the timetable toward 3 years to reach a significant conclusion. Those who had had prior bypass surgery did have lower patency rates; 74% at 1 year and 56% at 3 years, so there appeared to be differences in the results. Neither one was statistically significant.
Your final question, should we expand the indications for this technique to elderly patients accepting their reduced life expectancy? It has been my opinion that for elective bypasses in patients with chronic ischemia, if acceptable autogenous conduit is available, every effort should be made to use it. The results are simply far superior to synthetic grafts even with the distal venous cuff. Fortunately, for most of these cases, we have 2 experienced surgeons working simultaneously. I do not think arm vein harvest and use in our experience adds much operative time nor have we found significant arm vein harvest morbidity. I think in patients who present more acutely, who perhaps do not have the opportunity to undergo adequate 4-extremity vein mapping, we are more inclined to use the cuffed PTFE technique if the greater saphenous vein is not readily available and adequate.
James Debord, MD, Peoria, Ill: One of the approaches to this problem when you have an inadequate but at least a reasonable section of vein is a composite graft with a more traditional anastomosis between the prosthetic graft and the vein segment and then a vein to target artery anastomosis. Do you use that in your practice, or do you believe the hemodynamics of the Miller cuff are just as beneficial as a composite graft?
Dr Lauterbach: In my practice, I do not use these composite conduit strategies. I do not favor the synthetic graft to vein end-to-end anastomosis. My fear is that thrombosis of the synthetic component results in thrombosis of the venous component, which at times can be a significant length of vein. I would rather use spliced venous composites. Additionally, the benefit of a wide, cuffed anastomosis is lost for the prosthetic graft, but I have no data comparing the venous cuffed PTFE graft to composite grafts. I like the Miller cuff from a technical perspective because we can interrogate the vein-to-artery anastomosis through the cuff. Also, I think there is some hemodynamic advantage provided by the cuff. I do use sequential strategies such as synthetic grafting to the above knee popliteal artery with a vein graft originating just distal to the synthetic anastomosis as a separate anastomosis if the segment of patent artery is long enough. This technique can enable us to reach our desired distal target when inflow is from the common femoral artery and adequate vein length is absent.
Sachinder Hans, MD, Warren, Mich: Do you have any data on the outcomes? Did they all go home? How many of them went to extended care facilities? How many patients are ambulatory after this bypass? Also, I would like to know the role of cryo-preserved vein. Do you ever use cryo-preserved veins? I think that these procedures should be rarely used. In my experience of over 25 years of having performed more than 500 infrainguinal reconstructions, I have only crossed the knee joint with a prosthetic graft 4 times and I want to caution people regarding overuse of multiple procedures in these very sick patients as they may end up spending 3 or 4 months in the hospital. Otherwise, I enjoyed the paper. I think there is definite use of these procedures; however, it should be under rare circumstances.
Dr Lauterbach: With regard to ambulatory status, I think patients who are ambulatory before the operation are going to be ambulatory after the operation in most cases if limb salvage is achieved. Approximately 15% of the patients were nonambulatory and remained that way despite limb salvage. Most of these patients required rehabilitation in the form of physical therapy, some as inpatients in rehabilitation facilities.
We practice in a large-volume referral center and see many patients who have had multiple prior bypass procedures—half of all patients in this cohort. These 105 cases represent only 11% of our entire lower extremity revascularization procedures over the same 47-month interval. The age of this cohort of nearly 80 years coupled with the number of patients with a history of prior vascular bypass surgery including coronary surgery equates to patients requiring long bypasses with inadequate venous conduit. The progression of atherosclerosis as a function of aging results in failed primary grafts and requires further intervention for limb salvage. Despite the comorbidities of this cohort, the operative mortality rate of approximately 3% is acceptable in our opinion.
I have used cryopreserved vein as conduit in the past, and found reasonable limb salvage but poor patency rates. Cost and aneurysmal dilatation are major disadvantages in my opinion. In my experience, the overall results are inferior to the PTFE-venous cuff technique.