Comparison of endovascular revascularization and bypass surgery for severe chronic lower limb ischemia: A prospective multicenter cohort study with propensity score matching

Objective: The aim of this study was to compare the results of lower limb bypass grafts with those of endovascular treatment in patients with severe chronic lower limb ischemia (SCI). Methods: This prospective, non-randomized, multicenter cohort study involved 520 patients with SCI. In this series, 239 patients underwent lower limb bypass surgery and 281 patients underwent endovascular treatment. The primary endpoint was 3-year amputation-free survival, the secondary endpoint was the percentage of ischemic wound healing at 90 days. The hemodynamic criteria associated with the severity of the wounds were analyzed by the WIfI classification and the extent of arterial lesions by the GLASS classification. We completed the study with a multidimensional analysis according to the Cox model and then by matching the groups by propensity score. Results: On the raw data, before pairing, the 3-year survival without amputation was not significantly different between the Bypass group (70 ± 3%) and the Endovascular group (67 ± 3%) (Kaplan-Meier Logrank: P = 32). The Cox model regression study found 5 variables independently associated with a risk of death and/or major amputation. After matching the two groups by propensity score, the 3-year amputation-free survival was found to be significantly higher in the Bypass group compared to the Endovascular group (75 ± 3% vs. 59 ± 4%, Logrank: P < 001). In addition, the healing rate was significantly higher in the Bypass group compared to the Endovascular group (Logrank: P < 001) with at 3 months 56 ± 6% complete healing in the Bypass group compared to 21 ± 5% in the Endovascular group. Finally, an interaction analysis by logistic regression, with the explanatory variable death/amputation, identified the WIfI score 2/3 (severity of ischemia and trophic disorder), the GLASS 3 score (Extensive arterial lesions), diabetes and severe chronic renal failure as more favorable factors for bypass revascularization. The presence of LVEF < 40% being more favorable for endovascular treatment. Conclusions: This prospective multicenter non-randomized study showed that patients treated with bypass surgery had more severe lesions than those treated with an endovascular technique. Matching groups by propensity score demonstrated the superiority of bypass revascularization for the primary outcome. Logistic regression interaction analysis suggested that the choice of revascularization technique should be tailored to the lesion profile (WIfI, GLASS) and the cardiovascular status of the patients.