| Abstract: | The ability of stents to reduce restenosis was established in larger coronary arteries. Clinical trials of stenting in smaller vessels have yielded conflicting results due in part to their sample sizes. The aim of this meta‐analysis was to increase the statistical power by pooling data from these clinical trials. Trials were identified from Medline search, review of recent cardiology meetings' abstracts, and manual review of bibliographies. Studies were included if they were prospective randomized controlled trials. Endpoints examined included a dichotomized definition of angiographic restenosis, target lesion revascularization (TLR), target vessel revascularization (TVR), or any repeat revascularization. Pooling of data was performed by calculating a Mantel‐Haenszel odds ratio (OR). The analysis included 2,598 patients enrolled in eight clinical trials. Stenting significantly reduced restenosis (OR = 0.62; 95% CI = 0.61–0.63). Concordantly, stenting reduced TLR (OR = 0.49), TVR (OR = 0.90), and any revascularization (OR = 0.48). This meta‐analysis supports the hypothesis that stenting reduces restenosis in small coronary arteries as well as in larger coronary arteries. The apparent discordant result of individual clinical trials was due in part to underpowering related to small sample sizes. Catheter Cardiovasc Interv 2004;62:425–429. © 2004 Wiley‐Liss, Inc. |
