Long-term limb salvage and survival after endovascular and open revascularization for critical limb ischemia after adoption of endovascular-first approach by vascular surgeons.

OBJECTIVE: The adoption of endovascular interventions has been reported to lower amputation rates, but patients who undergo endovascular and open revascularization are not directly comparable. We have adopted an endovascular-first approach but individualize the revascularization technique according to patient characteristics. This study compared characteristics of patients who had endovascular and open procedures and assessed the long-term outcomes.

METHODS: From December 2002 to September 2010, 433 patients underwent infrainguinal revascularization for critical limb ischemia (CLI; Rutherford IV-VI) of 514 limbs (endovascular: 295 patients, 363 limbs; open: 138 patients, 151 limbs). Patency rates, limb salvage (LS), and survival, as also their predictors, were calculated using Kaplan-Meier and multivariate analysis.

RESULTS: The endovascular group was older, with more diabetes, renal insufficiency, and tissue loss. More reconstructions were multilevel (72% vs 39%; P < .001) and the most distal level of intervention was infrapopliteal in the open group (64% vs 49%; P = .001). The 30-day mortality was 2.8% in the endovascular and 6.0% in the open group (P = .079). Mean follow-up was 28.4 ± 23.1 months (0-100). In the endovascular vs open groups, 7% needed open, and 24% needed inflow/runoff endovascular reinterventions with or without thrombolysis vs 6% and 17%. In the endovascular vs open group, 5-year LS was 78% ± 3% vs 78% ± 4% (P = .992), amputation-free survival was 30% ± 3% vs 39% ± 5% (P = .227), and survival was 36% ± 4% vs 46% ± 5% (P = .146). Five-year primary patency (PP), assisted-primary patency (APP), and secondary patency (SP) rates were 50 ± 5%, 70 ± 5% and 73 ± 6% in endovascular, and 48 ± 6%, 59 ± 6% and 64 ± 6% in the open group, respectively (P = .800 for PP, 0.037 for APP, 0.022 for SP). Multivariate analysis identified poor functional capacity (hazard ratio, 3.5 [95% confidence interval, 1.9-6.5]; P < .001), dialysis dependence (2.2 [1.3-3.8]; P = .003), gangrene (2.2 [1.4-3.4]; P < .001), need for infrapopliteal intervention (2.0 [1.2-3.1]; P = .004), and diabetes (1.8 [1.1-3.1]; P = .031) as predictors of limb loss. Poor functional capacity (3.3 [2.4-4.6]; P < .001), coronary artery disease (1.5 [1.1-2.1]; P = .006), and gangrene (1.4 [1.1-1.9]; P = .007) predicted poorer survival. Statin use predicted improved survival (0.6 [0.5-0.8]; P = .001). Need for infrapopliteal interventions predicted poorer PP (0.6 [0.5-0.9-2.2]; P = .007), whereas use of autologous vein predicted better PP (1.8 [1.1-2.9]; P = .017).

CONCLUSIONS: Patients who undergo endovascular revascularization for CLI are medically higher-risk patients. Those who have bypass have more complex disease and are more likely to require multilevel reconstruction and infrapopliteal intervention. Individualizing revascularization results in optimization of early and late outcomes with acceptable LS, although survival remains low in those with poor health status.