The high cost of drug-eluting stents (DESs) has made identification of patients who are at low risk for subsequent revascularization after treatment with bare metal stents (BMSs) highly desirable. Previous reports from randomized trials suffer from biases induced by restricted entry criteria and protocol-mandated angiographic follow-up. Between 1994 and 2001, 5,239 consecutive BMS patients, excluding those with coil stents, technical failure, brachytherapy, staged procedure, or stent thrombosis within 30 days, were prospectively identified from a large single-center tertiary-referral-center prospective registry for long-term follow-up. We sought to identify characteristics of patients with very low (< or = 4%) or low (4-10%) likelihood of coronary revascularization 9 months after BMS. Nine-month clinical follow-up was obtained in 98.2% of patients. Coronary revascularization was required in 13.4% and did not differ significantly by stent type. On the basis of multivariate analysis identifying 11 independent correlates and previous reports, 20 potential low-risk patient and lesion groups (228 +/- 356 patients/groups) were identified (e.g, patients with all of the following: native vessel, de novo, reference diameter > or = 3.5 mm, lesion length < 5 mm, no diabetes, not ostial in location). Actual and model-based outcomes were analyzed. No group had both predicted and observed 9-month revascularization < or = 4% (very low risk). Conversely, 19 of 20 groups had a predicted and observed revascularization rate of 4-10% (low risk). In the real-world setting, the need for intermediate-term revascularization after BMS may be lower than expected, but it may be very difficult to identify patients at very low risk. Conversely, if the benefits of DESs are attenuated in routine practice, many groups of patients treated with BMSs may have nearly comparable results.
(c) 2004 Wiley-Liss, Inc.