A Propensity-Matched Analysis of Endovascular Intervention versus Open Nonautologous Bypass as Initial Therapy in Patients with Chronic Limb-Threatening Ischemia.

OBJECTIVES Endovascular therapies are increasingly used in patients with complex multilevel disease and chronic limb-threatening ischemia (CLTI). Infrageniculate bypass with autologous vein conduit is considered the gold standard in these patients. However, many patients often lack optimal saphenous vein, leading to the use of nonautologous prosthetic conduit. We compared limb salvage and survival rates for patients with CLTI undergoing first time revascularization with either open nonautologous conduit or endovascular intervention. METHODS We retrospectively reviewed consecutive patients undergoing first time endovascular or open surgical revascularization at our institution between 2009 and 2016. Patients were divided endovascular intervention or open bypass with nonautologous conduit (NAC) cohorts. Primary endpoints were amputation-free survival (AFS), freedom from reintervention (FFR), primary patency (PP), and overall survival (OS). Propensity scoring was used to construct matched cohorts. Outcomes were evaluated using Kaplan-Meier and Cox Proportional Hazards models. RESULTS A total of 125 revascularizations were identified. There were 65 endovascular interventions and 60 NAC bypasses. In unmatched analysis, there was an elevated risk of perioperative MI (7% vs 0%, p=0.05) and amputation (10% vs 2%, p=0.04) for the NAC groups compared to the endovascular group. In matched analysis, endovascular patients had a lower incidence of 30-day amputation (1.5% vs 10%p=0.04) and length of stay (median days, 1 vs 9, p<0.01) compared to the open cohort. While not statistically significant, the endovascular group trended towards increased rates of two-year AFS (76% vs 65%, p=0.07) compared to the NAC group. There was no significant difference in OS when the endovascular cohort was compared to NAC (85% vs 77%, p=0.29) patients. In matched Cox analysis, nonautologous conduit use was associated with an increased risk of limb loss (HR 2.03, 95% CI 0.94-4.38, p=0.07) compared to endovascular revascularization. CONCLUSIONS An "endovascular first" approach offers favorable perioperative outcomes and comparable AFS compared to NAC and may be preferable when autologous conduit is unavailable.