Randomized controlled trial of remote endarterectomy versus endovascular intervention for TransAtlantic Inter-Society Consensus II D femoropopliteal lesions

Objective: This study evaluated outcomes of remote endarterectomy (RE) vs endovascular (ENDO) interventions on TransAtlantic Inter-Societal Consensus (TASC)-II D femoropopliteal lesions and identified factors predictive of restenosis. Methods: From October 2004 to December 2008, 95 patients with TASC-II D lesions were randomized 1: 1 to receive RE of the superficial femoral artery (SFA) with end point stenting (51 patients) or ENDO, consisting of subintimal angioplasty with stenting (44 patients). The groups were balanced for age, sex, atherosclerotic risk factors, and comorbidities. Categoric data were analyzed with chi(2) tests, and time to event provided two-sided P values with a level of significance at .05 and 95% confidence intervals (CIs). Survival curves for primary patency were plotted using the Kaplan-Meier method. Univariate analysis for diabetes, hypertension, dyslipidemia, smoking, and critical ischemia was performed according to the Cox proportional hazards model. Results: The mean follow-up was 52.5 months (range, 35-75 months). Five RE patients and four ENDO patients were lost to follow-up (censored). Primary patency was 76.5% (39 of 51) in RE and 56.8% (25 of 44) in ENDO (hazard ratio [HR], 2.6; 95% CI, 0.99-4.2; P = .05) at 24 months and was 62.7% (32 of 46) in RE and 47.7% (21 of 40) in ENDO (HR, 1.89; 95% CI, 0.94-3.78; P = .07) at 36 months. Assisted primary patency was 70.6% (36 of 51) in RE and 52.3% (23 of 44) in ENDO (HR, 2.45; 95% CI, 1.20-5.02; P = .01). Secondary patency overlapped the primary comparison data at 12 and 24 months; at 36 months, there was a slight but significative advantage for RE (HR, 2.26; 95% CI, 1.05-4.86; P = .03). Univariate analysis demonstrated that hypercholesterolemia and critical limb ischemia (CLI) were significantly related to patency failure, whereas diabetes was significant only in ENDO. These factors (hypercholesterolemia and CLI) were independent predictors of patency on Cox multivariate analysis. Conclusions: RE is a safe, effective, and durable procedure for TASC-II D lesions. Our data demonstrate a significantly higher primary, assisted primary, and secondary patency of RE vs ENDO procedures. Furthermore, overall secondary patency rates remain within the standard limits, although preoperative CLI and dyslipidemia continue to be associated with worse outcomes. Taken together, these data suggest that RE should be considered better than an endovascular procedure in SFA long-segment occlusion treatment. (J Vasc Surg 2012;56:1598-605.)