Patients with diabetes mellitus are particularly prone to restenosis after percutaneous coronary intervention. An exploratory, nested, case-control study was undertaken to identify clinical, lesional, and procedural predictors of angiographic restenosis in these patients.

Seventy-five patients with diabetes mellitus with 86 coronary lesions were selected from a larger population of 217 patients who had undergone 6-month angiographic follow-up after a first, successful balloon angioplasty (PTCA) or stent implantation procedure. Data collection was by patient interview and review of hospital database and other medical records. All angiograms were analyzed with quantitative coronary angiography, and restenosis was defined as a ≥50% diameter reduction at the treated site. A multivariate analysis of 10 prespecified explanatory variables, derived from a literature review, was performed on a per-lesion basis.

There were 45 patients (53 lesions) with restenosis and 30 patients (33 lesions) without restenosis. Univariate predictors of binary restenosis were periprocedural glycosylated hemoglobin level, vessel reference diameter, PTCA, and larger final balloon size to reference artery diameter ratio. Multiple logistic regression identified poor glycemic control (odds ratio [OR] 3.03, 95% CI 1.06–8.65, P = .038), small vessel reference diameter (OR 3.41, 95% CI 1.17–9.95, P = .025), and mode of intervention (OR 3.12, 95% CI 1.08–9.00, P = .036) as independent risk factors. Vessel reference diameter appeared to be an important effect modifier of the association between type of intervention and angiographic outcome, with stenting no longer superior to PTCA in patients with diabetes mellitus who had vessels <2.87 mm in diameter (P = .054).

Poor glycemic control, vessel size, and PTCA were independent predictors of restenosis in patients with diabetes mellitus. It is possible that improved periprocedural glycemic control, in addition to stenting, may reduce the restenosis rate in these patients.