Predictors of limb loss despite a patent endovascular-treated arterial segment
Article Outline
Objective
The goal of this study was to assess the frequency and predictors of major amputation with patent endovascular-treated arterial segments (PETAS) in patients with critical limb ischemia.
Methods
The study included 358 consecutive patients (412 limbs) who underwent endovascular (236 limbs) or open (176 limbs) revascularizations for critical limb ischemia from June 2001 through May 2007. Patients with limb loss despite PETAS were compared with the rest of the endovascular-treated group (EV-other, n = 212) and with those who underwent amputations with patent bypasses (APB).
Results
The EV group underwent 30 amputations (24 in PETAS, 6 in EV-other), and 37 occurred in the open group (14 in APB, 23 in open-other). Amputations occurring despite a patent revascularized segment constituted 38% of limb loss in open and 80% in EV-treated patients (P = .001). Limb loss occurred earlier in the PETAS group (58% vs 30% ≤3 months). Primary indications for limb loss in the PETAS group were extensive tissue loss or limb dysfunction after radical débridement of infection or gangrene (37%), recurrent infection (42%), and failure to reverse ischemia (21%). There were more patients with diabetes in PETAS group (96%) than in the APB group (64%, P = .018). Diabetes, dialysis-dependence, lower albumin level, gangrene, and infrapopliteal interventions were more likely in the PETAS group than in the EV-other group. Multivariate analysis showed diabetes (odds ratio [OR], 3.15; 95% confidence interval [CI], 1.22-8.13, P = .018), gangrene (OR, 3.33; 95% CI, 1.43-7.75; P = .005), and infrapopliteal interventions (OR, 3.09; 95% CI, 1.38-6.94; P = .006), predicted limb loss with patent open or EV-treated segments, whereas dialysis-dependence, peroneal artery-only runoff, albumin level <3 g/dL, location at the heel, and treatment type did not.
Conclusions
Amputation despite PETAS is the most common means of limb loss in patients undergoing endovascular revascularization for limb salvage. It is likely the result of aggressive attempts at limb salvage and usually occurs ≤3 months after the intervention. Patients with diabetes and gangrene undergoing infrapopliteal interventions are at a significantly high risk. Adjuncts to reduce tissue loss, preserve limb function, and prevent recurrent infection are needed to prevent limb loss despite PETAS, especially in diabetic patients.
Limb loss with patent bypass has been reported to occur in 4% to 9% of patients.1, 2, 3, 4 The relative incidence of amputations performed with patent bypass is higher, at up to 50%, in certain subgroups, including patients with end-stage renal disease,5, 6, 7 diabetes mellitus,2, 8 and those with limited runoff.9, 10 The causes of limb loss in patients despite patent grafts have included extensive infection, poor pedal runoff, failure to reverse ischemia at the site of tissue loss, heel necrosis (>4 cm), and forefoot gangrene, especially in patients with diabetes and end-stage renal disease,8 and primary amputation has been suggested in these patients.1, 5, 7, 8 The adoption of complex endovascular interventions has allowed limb salvage attempts in higher-risk patients who would otherwise not have been considered for revascularization, making the decision for primary amputation more complex.11, 12, 13, 14
Although the adoption of an aggressive endovascular-first approach for all comers with critical limb ischemia has decreased the primary amputation rate from 15% to 4%, accompanied by a decrease in limb loss in patients in whom limb salvage was attempted,13 we have found that most amputations occurred in patients with patent endovascular-treated segments (PETAS), which other authors have also described.15, 16 Despite numerous reports on limb loss despite patent bypass grafts, limb loss with PETAS has not been emphasized. The goal of this study was to assess the frequency and predictors of major amputation with PETAS in patients with critical limb ischemia.
Material and methods
All consecutive patients who presented to the Veterans' Administration Western New York Healthcare System between June 1, 2001, and May 31, 2007, with critical limb ischemia (Rutherford category 4 to 6)17 and who underwent a technically successful infrainguinal revascularization by endovascular or open bypass procedures were identified from our prospectively maintained database after we received Institutional Review Board approval for the study. The database recorded patient demographics, comorbidities, clinical presentation, noninvasive arterial studies, TransAtlantic Inter-Society Consensus (TASC) classification of the treated lesions,18 details of the procedures performed, the most distal level of intervention, number of runoff vessels, follow-up arterial studies, and status of the limb at the last follow-up.
A patent runoff vessel was defined as an infrapopliteal vessel without a hemodynamically significant (>50%) angiographic stenosis distal to the treated site, and the number of adequately patent runoff vessels (0 to 3) was calculated after all interventions were completed for that limb. In-line flow was defined as reinstitution of uninterrupted flow by at least one nondiseased infrapopliteal runoff vessel (anterior tibial, posterior tibial, or peroneal artery) to the foot after revascularization. If the only runoff vessel was the peroneal artery, this was noted.
Patients who underwent supramalleolar amputations were identified, and the patency of their endovascular-treated segments or the bypass grafts was recorded. The 24 patients with limb loss despite a patent endovascular-treated segment (PETAS group) were compared with the 212 remaining in the endovascular-treated group (EV-other), with the 14 patients who underwent amputations with patent bypasses (APB), and with the 23 bypass patients who had amputation due to other causes (open-other).
All patients were monitored by clinical assessment and by our vascular laboratory during the first postoperative visit at 1 to 4 weeks, at 3 and 6 months, and every 6 months thereafter for ankle-brachial index (ABI) measurements, graft or stent velocities, and duplex imaging. Angiography was performed when clinically indicated.
All patients with open wounds underwent very close follow-up in our vascular surgery wound clinic until wounds were completely healed. The time to complete healing was recorded. Repeat sharp débridements were performed as necessary. Enzymatic débridement was used when deemed necessary. Various adjuncts, including vacuum-assisted closure (VAC, Kinetic Concepts Inc, San Antonio, Tex) and various skin or collagen tissue substrates, were used when deemed appropriate by the vascular surgeon. Patients were referred for hyperbaric oxygen therapy when other modalities failed. Autologous skin grafts were used when an adequate wound base was achieved.
The patency of the endovascular-treated segments was routinely checked using duplex examination. An angiogram was performed when wound healing was not observed, with the intention of performing additional interventions if needed. Additional endovascular or bypass procedures were performed if limb salvage was thought to be feasible.
The débridements were mostly performed immediately after the revascularization procedure. Patients with extensive infection in whom débridement was thought to result in extensive tissue loss or who had systemic sepsis initially underwent wound débridement, and revascularization (endovascular or open) was attempted immediately after sepsis was controlled. The 18 patients who had extensive foot sepsis and did not have adequate foot for salvage underwent primary guillotine amputation without a revascularization attempt during the study period. In our hospital, patients who are bedridden, especially with flexion contractures, are offered primary amputation without revascularization.
All endovascular procedures were performed by vascular surgeons in the operating room using the OEC 9800 system (General Electric Medical Systems, Salt Lake City, Utah). Most infrainguinal interventions were performed by a contralateral femoral artery approach using 6F sheaths. Society for Vascular Surgery (SVS) reporting standards for lower extremity arterial procedures were followed.17
Data analysis was performed using SPSS 16.0 software (SPSS Inc, Chicago, Ill). Kaplan-Meier analysis and the log-rank test were used to compare groups for limb salvage and overall survival rates. Demographic comparisons were made using the two-tailed Fisher exact test for categoric variables and by the t test for continuous variables. Univariate analyses were performed for identifying factors predicting limb loss, and multivariate analysis was performed using Cox proportional regression to identify the independent predictors of limb loss. Statistical significance was set at P < .05.
Results
The analysis included 358 patients (99.2% men, 412 limbs) with critical limb ischemia. There were 197 patients (236 limbs) in the endovascular-treated group and 161 patients (176 limbs) in the open group. When compared with the open group, the patients in the endovascular group were significantly older (71.5 ± 10.6 vs 69.1 ± 11.0, P = .022), were more frequently diabetic (59.3% vs 48.3%, P = .028), and had hyperlipidemia (66.1% vs 55.7%, P = .04). There were more patients with gangrene (43.6% vs 34.1%, P = .053) and more had tissue loss at the heel (14.4% vs 5.1%, P = .002) in the endovascular group as well.
Comparison of endovascular vs open-treated patients
In the endovascular group, 30 amputations were performed, giving an overall limb salvage of 87.3%. In the open group, 37 amputations were performed, giving an overall limb salvage of 79.0%. The 36-month limb salvage rates by Kaplan-Meier analysis were 81% ± 4% in the endovascular-treated patients and 75% ± 4% in the open group (P = .1). In the endovascular group, 24 amputations occurred despite a patent endovascular-treated segment (PETAS group), and six occurred due to failed endovascular-treated segments. Of these amputations, 21 were below the knee and 3 were above the knee. In the open group, 14 amputations occurred in patients with patent bypasses (APB group), and 23 occurred in patients with occluded grafts. Half of the amputations with patent bypasses were below the knee. Amputations occurring despite patent revascularized segments constituted 38% of limb loss in the open patients and 80% in the endovascular-treated patients (P = .001).
PETAS vs EV-other
Primary causes for limb loss in the PETAS group were extensive tissue loss and limb dysfunction after radical débridement of infection or gangrene (37%), recurrent infection (42%), and failure to reverse ischemia (21%). All patients underwent duplex examination and noninvasive studies. Eight patients in the PETAS group (33%) had a repeat angiogram, three of whom underwent reintervention for restenosis or treatment of runoff vessels at 1, 3, and 21 months after the initial intervention. One patient underwent a popliteal-posterior tibial bypass 1 month after the endovascular intervention. Although the patient had a patent graft and stented segments, recurrent infection resulted in a below knee amputation 7 months later. None of the other patients who lost their limbs for failure to reverse ischemia were eligible for a bypass procedure because of lack of target vessels (2 patients) or poor medical condition (2 patients). The six patients who underwent amputation with occluded endovascular-treated segments underwent an angiogram, and thrombolysis was attempted without success in three patients. The remaining patients had either advanced infection or ischemia, and no further revascularization was attempted.
The comorbidities and demographic characteristics of patients in the PETAS group and all other patients in the endovascular group are summarized in Table I. The PETAS patients were more likely to have diabetes, dialysis-dependence, gangrene, a history of contralateral amputation, and lower preoperative albumin levels. Overall, 80% of the endovascular-treated patients had TASC C or D lesions, with no difference between the PETAS group and the EV-other group (76% vs 81%, P = .188). The mean preoperative and postoperative ABI values were similar between the PETAS and the EV-other groups (Table I). Although in-line flow to the foot, mean number of runoff vessels, and hindfoot tissue loss were similar, more infrapopliteal interventions were done in the PETAS group, and the peroneal artery was the only runoff vessel to the foot in more patients. In the PETAS group, limb loss occurred ≤1 month postoperatively in 7 patients, between 1 and 3 months in 7 patients, between 3 and 12 months in 7 patients, and at 14, 17, and 21 months in the remaining 3 patients.
Table I. Patients with limb loss despite a patent endovascular-treated arterial segment compared with all other endovascular-treated patients
| Variable | EV-other (n =212) | PETAS (n=24) | P |
|---|---|---|---|
| Age, mean±SD, y | 71.9±10.5 | 68.0±11.1 | .09 |
| CAD, % | 60 | 67 | .661 |
| Hypertension, % | 63 | 76 | .214 |
| Diabetes, % | 55 | 96 | <.001 |
| CVD, % | 26 | 13 | .212 |
| Hyperlipidemia, % | 67 | 54 | .255 |
| COPD, % | 24 | 8 | .119 |
| Active smoker, % | 39 | 38 | 1.0 |
| Ever smoked, % | 92 | 96 | 1.0 |
| Renal insufficiency, % | 25 | 38 | .217 |
| Dialysis-dependent, % | 7 | 21 | .038 |
| Contralateral amputation, % | 3 | 29 | <.001 |
| Albumin, mean±SD, g/dL | 3.3±0.6 | 2.7±0.5 | <.001 |
| Tissue loss (gangrene), % | 75(40) | 96(79) | <.001 |
| Heel ulcer/gangrene, % | 14 | 21 | .356 |
| Infrapopliteal, % | 20 | 54 | .001 |
| Femoropopliteal, % | 53 | 46 | .678 |
| Aortoiliac, % | 27 | 0 | .001 |
| ABI, mean±SD | |||
| Preoperative | 0.40±0.25 | 0.43±0.26 | .593 |
| Postoperative | 0.82±0.20 | 0.83±0.26 | .875 |
| In-line flow, % | 91 | 96 | .687 |
| Runoff vessels, mean±SD, No. | 1.5±0.8 | 1.4±0.6 | .382 |
| Peroneal runoff, % | 17 | 38 | .025 |
PETAS vs APB
The comorbidities and other characteristics of patients in the PETAS group and the APB group are compared in Table II. There were significantly more patients with diabetes in the PETAS group than in the APB group (96% vs 64%, P = .018). Although more distal revascularizations were performed in the APB group than in the PETAS group, this was not statistically significant (79% vs 54%, P = .175). The mean preoperative and postoperative ABIs and the number of runoff vessels were similar between the PETAS and APB groups (Table II). In the PETAS group, 88% of all the amputations were performed below the knee, whereas this was 50% in the APB group (P = .021). Primary causes for limb loss in the APB group were extensive tissue loss or limb dysfunction after radical débridement of infection or gangrene (7%), recurrent infection (21%), failure to reverse ischemia (50%), and prosthetic graft infection (21%), which were significantly different from PETAS group (P = .009).
Table II. Patients with limb loss despite patent endovascular-treated segments compared with patients who had amputations with patent bypasses
| Variable | APB (n=14) | PETAS (n=24) | P |
|---|---|---|---|
| Age, mean±SD, y | 63.6±12.0 | 68.0±11.1 | .254 |
| CAD, % | 57 | 67 | .729 |
| Hypertension, % | 86 | 76 | .160 |
| Diabetes, % | 64 | 96 | .018 |
| CVD, % | 21 | 13 | .650 |
| Hyperlipidemia, % | 43 | 54 | .737 |
| COPD, % | 14 | 8 | .616 |
| Active smoker, % | 64 | 38 | .179 |
| Ever smoked, % | 86 | 96 | .542 |
| Renal insufficiency, % | 50 | 38 | .510 |
| Dialysis-dependent, % | 21 | 21 | .999 |
| Contralateral amputation, % | 14 | 29 | .438 |
| Albumin, mean±SD, g/dL | 3.0±0.8 | 2.7±0.5 | .321 |
| Tissue loss (gangrene), % | 86(71) | 96(79) | .392 |
| Heel ulcer/gangrene, % | 7 | 21 | .383 |
| Infrapopliteal, % | 79 | 54 | .175 |
| Femoropopliteal, % | 14 | 46 | .077 |
| Aortoiliac, % | 7 | 0 | .368 |
| ABI, mean±SD | |||
| Preoperative | 0.39±0.22 | 0.43±0.26 | .679 |
| Postoperative | 0.84±0.25 | 0.83±0.26 | .944 |
| In-line flow, % | 93 | 96 | .615 |
| Runoff vessels, mean±SD, No. | 1.2±0.6 | 1.4±0.6 | .450 |
| Peroneal runoff, % | 21 | 38 | .472 |
All patients underwent duplex examination and noninvasive studies, and graft patency was confirmed. Graft occlusion in two patients was successfully treated using thrombolysis at and 9 months before the amputation. Angiograms in two additional patients were not able to identify a correctable lesion. Limb loss occurred ≤1 month postoperatively in 1 patient, between 1 and 3 months in 4 patients, between 3 and 12 months in 7 patients, and at 38 and 46 months in the remaining 2 patients. Limb loss occurred earlier in the PETAS group (58% vs 36% ≤3 months, P = .313). Most amputations occurred ≤12 months in both groups (PETAS, 88%; APB, 86%).
The 12-month survival rate in the PETAS group was worse than in the EV-other (67% ± 10% vs 78% ± 3%, P = .065) and APB groups (86% ± 9%, P = .113), but these rates were not statistically different owing to the small number of patients in each group.
Predictors of PETAS/APB
Multivariate analysis showed diabetes (odds ratio [OR], 3.15; 95% confidence interval [CI], 1.22-8.13, P = .018), gangrene (OR, 3.33; 95% CI, 1.43-7.75; P = .005), and infrapopliteal interventions (OR, 3.09; 95% CI, 1.38-6.94; P = .006) predicted limb loss with patent open or endovascular-treated segments, whereas dialysis-dependence, peroneal artery-only runoff, albumin level <3 g/dL, location at the heel, and type of treatment did not (Table III). However, the 12-month limb salvage rate in the 26 patients with diabetes and gangrene after infrapopliteal endovascular interventions was 61% ± 11%, and 12-month survival was 64% ± 10%. Nine of the 10 amputations in this subgroup occurred despite PETAS. On the other hand, the 12-month limb salvage rate in open-treated diabetic patients with gangrene who underwent infrapopliteal bypasses was 58% ± 8% (P = .865 vs endovascular), and the 12-month survival was 71% ± 7% (P = .295 vs endovascular).
Table III. Multivariate analysis of factors that predict limb loss with patent open or endovascular-treated segments
| Factor | OR | 95% CI | P |
|---|---|---|---|
| Diabetes | 3.15 | 1.22-8.13 | .018 |
| Dialysis-dependence | 2.24 | 0.85-5.91 | .103 |
| Albumin <3 g/dL | 1.78 | 0.83-3.88 | .142 |
| Peroneal-only runoff | 1.25 | 0.56-2.75 | .588 |
| Gangrene | 3.33 | 1.43-7.75 | .005 |
| Open vs endovascular | 1.85 | 0.81-4.21 | .144 |
| Infrapopliteal | 3.09 | 1.38-6.94 | .006 |
| Heel location | 1.87 | 0.61-5.79 | .275 |
Discussion
Limb loss with patent bypass grafts is well-documented and is generally a reflection of the vascular surgeon's threshold for attempting limb salvage in patients with adverse features such as end-stage renal disease, diabetes, extensive tissue loss, advanced infection, and poor functional status. As the incidence of primary amputation decreases, the chances of having limb loss with patent revascularization increases.
The incidence of limb loss with patent bypass has varied from 3%19 to as high as 59%,6 depending on the group being studied; however, not all of these studies with high limb loss described the at-risk patient subgroups. Carsten et al4 reported 17 limb losses with patent grafts (57%), and found that black race, renal failure, and infrapopliteal bypasses were independently associated with this occurrence, but diabetes was not. Seeger et al3 reported that limb loss with a patent graft occurred in 46% of patients, which constituted the largest subgroup of patients with limb loss, and found that diabetes, extensive tissue loss, dialysis-dependence, and poor runoff were predictors of this occurrence. Other reports of patients with critical limb ischemia and dialysis-dependence also reported a high incidence of up to 59% and also reported heel necrosis >4 cm, diabetes, extensive tissue loss, infection, and poor runoff were significant contributory factors for predicting limb loss with patent grafts.5, 6, 7, 20
We found that the incidence and other characteristics of our open-treated group with patent grafts who had limb loss were not too dissimilar from those reporting aggressive limb salvage attempts.3, 4 The relative incidence of limb loss with patent grafts constituted 38% of all amputations in this subgroup, and the clinical characteristics of this group were similar to previous reports, with diabetes, 64%; end-stage renal disease, 21%; undergoing infrapopliteal revascularizations, 79%; peroneal artery-only runoff, 21%; presenting with gangrene, 71% (Table II). The incidence of limb loss with PETAS was significantly higher (80%) in the endovascular-treated group, however, possibly reflecting an even more aggressive approach used in this subgroup of patients, although the relative incidence of limb loss was somewhat less than in the open-treated patients. Significantly more diabetic patients were in the PETAS group (96%) than in the APB group, and gangrene, contralateral amputation, peroneal-only runoff, and heel location were also more frequent in PETAS group. All these observations suggest that patients with disadvantaged anatomic and physiologic features were more likely to be offered endovascular interventions rather than open interventions, resulting in an increased incidence of limb loss with PETAS.
Another reason for the excessive occurrence of limb loss with PETAS is the result of attempting limb salvage in patients who would otherwise undergo primary amputation. Abou-Zamzam et al21 recently analyzed the risk factors that led to primary amputations vs limb salvage attempts in 224 consecutive patients who presented with critical limb ischemia. Major tissue loss, end-stage renal disease, diabetes, and nonambulatory status were associated with primary amputation, which occurred in 43% of their patients. These risk factors are similar to ours and other reports1, 2, 3, 4 that predict limb loss with patent revascularized extremities. This suggests that as the primary amputation rate increases, there will be less limb loss with patent revascularized extremities, and vice versa. We have previously reported13 that adoption of endovascular interventions has resulted in a decreased rate of primary amputation from 15% to 4% in addition to increased limb salvage in those with attempted limb salvage. Our observation that most endovascular-treated patients had limb loss despite PETAS is likely a direct reflection of this change in practice and our overall increased aggressiveness of limb salvage attempt in higher risk patients. Interestingly, the overall number of limbs lost actually was less in the endovascular-treated group as a whole, and the increased aggressiveness resulted in a better limb salvage rate than the open group despite worse clinical characteristics.
The risk factors that were associated with limb loss despite PETAS and open-treated patients were diabetes, gangrene, and infrapopliteal interventions in the multivariate analysis. We also found that patients with end-stage renal disease, peroneal-only runoff, and contralateral amputations were significantly more in PETAS group than the remainder of the endovascular-treated patients. These findings are similar to the previous reports on limb loss with patent grafts.3, 4, 5 We also found several other factors that were common, including contralateral amputation, poor nutritional status, and peroneal-only runoff in patients with limb loss despite PETAS. Although nonsignificant in our series, contralateral amputation as an independent predictor has been reported, with a 20% to 24% incidence of contralateral amputation in patients with limb loss,1, 2, 22 and is likely a reflection of the overall poor general condition of these patients. The peroneal artery as the only runoff has been reported as being adequate after open and endovascular revascularizations for limb salvage23, 24, 25, 26 and was not associated with limb loss despite a patent revascularized segment in the multivariate analysis in our series.
The timing of limb loss with patent revascularized extremities seems to be similar in our study as in those reporting on bypasses.1, 2, 3, 4 In our PETAS group, 58% of the limb loss occurred ≤3 months, and 79% were infection-related. Patients requiring early amputation had overwhelming infection that resulted in extensive tissue loss despite repeated débridements in an attempt to control infection. Late infection-related amputations were mostly in diabetic patients with recurrent infections in a refractory wound, despite adequate perfusion and aggressive wound care. In the series by Carsten et al,4 53% of amputations despite patent grafts occurred ≤30 days and the remaining after 45 days. In other series, 50% to 66% of such amputations occurred early, mainly due to infections,3, 5, 6 with a few cause by poor runoff resulting in failure to reverse ischemia at the tissue level.
Although selection of patients before attempting revascularization is an attractive idea,27 our study fails to identify specific parameters that would predict limb loss and allow us to recommend primary amputation in a subset of patients. In fact, we found that limb salvage was achieved in 61% of patients in the endovascular group and in 58% in the open-treated groups in whom all three predictors of limb loss with patent revascularized limbs were identified in the multivariate analysis. Our results, however, prompt us to recommend careful patient selection for revascularization, especially among those with diabetes mellitus, gangrene, and significant infrapopliteal disease.
Infection seems to be a significant cause of early limb loss despite revascularization; thus, one could argue that revascularization should be considered after adequate débridement and infection control, which might possibly select out those with advanced infections and nonsalvageable limbs. Because the endovascular interventions allow revascularization with remote access sites and the risk of infection of the endovascular-treated segment, even in the face of pedal infection is extremely low, angiograms with attempted endovascular interventions will continue to be used by aggressive vascular surgeons and other interventionalists. Better assessment of the extent of infection and improvements in postoperative wound care and control of infection may potentially decrease the rate of early limb loss despite patent revascularized extremities.6
Survival of patients who had limb loss with PETAS was marginally worse than the rest of the endovascular-treated patients, which was likely a reflection of the poorer medical condition of this subgroup or was possibly due to the negative effect of amputation on survival in some patients. In the Reifsnyder et al1 series of 67 patients who underwent major amputations after bypass procedures, the 2-year survival of 21% was much worse than our rates of 56% in PETAS and 57% in the APB groups. We do not think that attempting limb salvage in these patients had a negative effect on their survival.
Another potential advantage of an aggressive endovascular approach is the possibility that the revascularization may have enabled the amputation to be performed at a more distal level for the PETAS group. In our series, 88% of the amputations were below the knee in the PETAS group, whereas this was only 50% in the APB group (P = .021).
Our study has certain limitations, including the retrospective nature and the small sample size treated by a single vascular surgeon in a single center with a nearly all male population. In addition, we do not have any objective information of the skin perfusion pressure or transcutaneous oxygen pressure, and pedal angiography was not routine in all patients because of concerns about contrast use or patient movement.
Conclusions
Aggressive use of endovascular interventions has resulted in limb salvage attempts in higher-risk patients who would otherwise have primary amputations. This resulted in amputations despite PETAS being the most common means of limb loss in patients undergoing endovascular revascularization for limb salvage. This occurs mostly in the first 3 months after the intervention. Patients with diabetes and gangrene who undergo infrapopliteal interventions are at significantly higher risk. Adjuncts to reduce tissue loss, preserve limb function, and prevent recurrent infection are needed to prevent limb loss despite PETAS, especially in diabetic patients.
Author contributions
References
- . Limb loss after lower extremity bypass. Am J Surg. 1997;174:149–151
- . Limb loss with patent infra-inguinal bypasses. Eur J Vasc Surg. 1990;4:413–417
- . Potential predictors of outcome in patients with tissue loss who undergo infrainguinal vein bypass grafting. J Vasc Surg. 1999;30:427–435
- . Factors associated with limb loss despite a patent infrainguinal bypass graft. Am Surg. 1998;64:33–38
- . Realistic expectations for pedal bypass grafts in patients with end-stage renal disease. J Vasc Surg. 1998;28:976–983
- . Failure of foot salvage in patients with end-stage renal disease after surgical revascularization. J Vasc Surg. 1995;22:280–286
- . An updated meta-analysis of infrainguinal arterial reconstruction in patients with end-stage renal disease. J Vasc Surg. 2007;45:536–542
- . Limb salvage in end-stage renal disease (ESRD): comparison of modern results in patients with and without ESRD. Arch Surg. 1988;123:116–148
- . Limb loss despite functioning distal bypass. J Cardiovasc Surg (Torino). 1989;30:473–478
- . Limitations of peroneal artery bypass grafting for limb salvage. J Vasc Surg. 1993;18:881–888
- BASIL Trial participants (Bypass versus angioplasty in severe ischemia of the leg (BASIL): multicentre, randomized controlled trial). Lancet. 2005;366:1925–1934
- . The increasing role of percutaneous transluminal angioplasty in the primary management of critical limb ischemia. Eur J Vasc Endovasc Surg. 2002;23:398–403
- . Does preferential use of endovascular interventions by vascular surgeons improve limb salvage, control of symptoms and survival in patients presenting with critical limb ischemia?. Am J Surg. 2006;192:572–576
- . The influence of subintimal angioplasty on level of amputation and limb salvage rates in lower limb critical limb ischemia. Eur J Vasc Endovasc Surg. 2005;30:291–299
- . Angioplasty and primary stenting of high-grade, long-segment disease: is it worthwhile?. Ann Vasc Surg. 2003;17:430–437
- Intermediate results of percutaneous endovascular therapy of femoropopliteal occlusive disease: a contemporary series. J Vasc Surg. 2006;44:762–769
- Reporting standards for lower extremity arterial endovascular procedures (Society for Vascular Surgery/International Society for Cardiovascular Surgery). J Vasc Surg. 1993;17:1103–1107
- . Management of peripheral arterial disease (PAD) (TASC Working Group. TransAtlantic Inter-Society Consensus (TASC)). J Vasc Surg. 2000;31:S1–S296
- . Results of infrainguinal revascularization with reversed vein conduits: a modern control series. Ann Vasc Surg. 1991;5:156–162
- . Infrainguinal bypass grafting in patients with end-stage renal disease: improving outcomes?. J Vasc Surg. 2001;33:1171–1178
- A prospective analysis of sritical limb ischemia: Factors leading to major primary amputation versus revascularization. Ann Vasc Surg. 2007;21:458–463
- . The management of ischemic heel ulcers and gangrene in the endovascular era. Am J Surg. 2007;194:600–605
- . Peroneal bypass is equivalent to inframalleolar bypass for ischemic pedal gangrene. Arch Surg. 1996;131:894–898
- . Arterial reconstruction for limb salvage: is the terminal peroneal artery a disadvantaged outflow tract?. Surgery. 1995;118:763–767
- . Are peroneal artery bypass grafts hemodynamically inferior to other tibial artery bypass grafts?. J Vasc Surg. 1994;19:964–968
- . Peroneal artery-only runoff following endovascular interventions is effective for limb salvage in patients with tissue loss. J Vasc Surg. 2008;48:137–143
- Invasive treatment of chronic limb ischemia according to the Lower Extremity Grading System (LEGS) score: a 6-month report. J Vasc Surg. 2004;39:1268–1276
Competition of interest: none.
PII: S0741-5214(09)00526-6
doi:10.1016/j.jvs.2009.02.226
© 2009 Society for Vascular Surgery. All rights reserved.
