Long-term outcomes of diabetic patients undergoing endovascular infrainguinal interventions.

Abstract

Diabetes mellitus (DM) has traditionally predicted poor outcomes after lower extremity revascularization for peripheral vascular disease (PVD). This study assessed the influence of DM on long-term outcomes of percutaneous transluminal angioplasty, with or without stenting (PTA/stent), in patients with PVD. From January 2002 to December 2007, 920 patients underwent 1075 PTA/stent procedures. Patients were stratified into DM and non-DM cohorts. Study end points included primary patency (PP), assisted patency (AP), limb salvage, and survival and were evaluated using Kaplan-Meier and Cox regression analyses. There were 533 DM and 542 non-DM limbs. Median follow-up was 34 months. Overall, the 5-year actuarial PP was 42% +/- 2.4%, AP was 81% +/- 2.0%, limb salvage was 89% +/- 1.6%, and survival was 60% +/- 2.4%. On univariate analysis, DM vs non-DM was associated with inferior 5-year PP (37% +/- 3.4% vs 46% +/- 3.3%; P = .009), limb salvage (84% +/- 2.6% vs 93% +/- 1.8%, P < .0001), and survival (52% +/- 3.5% vs 68% +/- 3.1%, P = .0001). AP did not differ between DM and non-DM patients (P = .18). In the entire cohort, DM (hazard ratio [HR], 1.25; 95% confidence interval [CI], 1.01-1.54; P = .04), single-vessel peroneal runoff (HR, 1.54; 95% CI, 1.16-2.08; P = .003), and dialysis (HR, 1.59; 95% CI, 1.10-2.33; P = .02) were associated with decreased PP on multivariate analysis. The only variables on multivariate analysis to predict limb loss and death were critical limb ischemia (HR, 9.09; 95% CI, 4.17-20.00; P < . 0001; HR, 2.99; 95% CI, 2.01-4.44; P < .0001, respectively) and dialysis (HR, 2.94; 95% CI, 1.39-5.00; P = .003; HR, 4.24; 95% CI 2.80-6.45; P < .0001, respectively). DM is an independent predictor of decreased long-term primary patency after PTA/stent. Although acceptable assisted patency rates can be achieved with close surveillance and reintervention, long-term limb salvage remains inferior in diabetic patients compared with non-diabetic patients due to a more severe clinical presentation and poor runoff.