Insulin therapy and outcomes of revascularization procedures for chronic limb-threatening ischemia in the BEST-CLI trial.

Abstract

Diabetes mellitus (DM) is a major risk factor for amputation in patients with chronic limb-threatening ischemia (CLTI) undergoing lower extremity revascularization. Observational studies comparing patients with DM based on insulin therapy have reported inconsistent findings. This study compares the outcomes of patients with insulin-requiring DM (IRDM) and non-IRDM (NIRDM) based on high-quality prospective data. Characteristics and outcomes of patients with IRDM and NIRDM enrolled in the BEST-CLI (Best Endovascular vs Best Surgical Therapy in Patients with CLTI) trial were compared. A Cox regression model was used to determine the association between insulin requirement and outcomes. The analysis included 1229 patients with DM, among whom 67.7% (N = 832) had IRDM. The mean age was 66.7 ± 9.9 years, with the IRDM group being younger (65.8 vs 68.6 years, P < .001) and more likely to be Hispanic (19% vs 13.1%, P = .01). Patients with IRDM were more likely to have renal disease and prior minor amputation (20.9% vs 13.9%, P = .003) but were less likely to be smokers (24.5% vs 38.3%, P < .001) compared with patients with NIRDM. Overall, only 26.8% of the cohort had adequate glucose control at study entry, with patients with NIRDM more likely to have a hemoglobin A1C (HbA1c) level <7 (19.2% vs 44%, P < .001). Patients with IRDM had significantly higher wound grade and more advanced overall WIfI (Wound, Ischemia, and foot Infection) stage compared with patients with NIRDM. At 3 years, patients with NIRDM were more likely to achieve resolution of CLTI, but there were no differences in other outcomes. Two years after revascularization, glucose control continued to be suboptimal, with only 37.3% of the cohort achieving HbA1c <7. Regression analysis showed that insulin requirement was associated with decreased CLTI resolution (hazard ratio [HR] = 0.86 [0.75-0.99]) but not with other outcomes. Compared with endovascular therapy, open revascularization was associated with a lower risk of major amputation (HR = 0.65 [0.48-0.88]), major reintervention (HR = 0.41 [0.30-0.56]), and major adverse limb events or death (HR = 0.78 [0.66-0.91]), irrespective of insulin therapy. A spline model demonstrated that open revascularization was associated with a lower HR of major adverse limb events or death compared with endovascular regardless of the baseline HbA1c level. Patients with IRDM undergoing lower extremity revascularization for CLTI present with worse glucose control and a more advanced CLTI stage compared with patients with NIRDM. Insulin requirement showed a modest association with the outcomes of revascularization in this cohort. Open revascularization was superior to endovascular in patients with DM, regardless of insulin therapy or the HbA1c level.