Twelve studies (3,182 patients) were included in the review (three RCTs, five RCT subgroups and four observational studies). The mean follow-up period was 377 days.
Target vessel failure (ten studies): Rates of target vessel failure were significantly lower with drug-eluting than with bare-metal stents (OR 0.37, 95% CI 0.29 to 0.47). There was moderate heterogeneity (Ι²=46.9%), but no evidence of publication bias.
Major adverse cardiovascular events (ten studies): Rates of major adverse cardiovascular events were were significantly lower with drug-eluting than with bare metal stents (OR 0.36, 95% CI 0.29 to 0.45). There was low heterogeneity (Ι²=31.9%).
Target vessel revascularisation (nine studies): Rates of target vessel revascularisation were significantly lower with drug-eluting than with bare-metal stents (OR 0.28, 95% CI 0.21 to 0.37). There was moderate heterogeneity (Ι²=49.9%). Event rates appeared to be lower for sirolimus stents (OR 0.20, 95% CI 0.14 to 0.30) than for paclitaxel stents (OR 0.49, 95% CI 0.32 to 0.74).
Binary restenosis (nine studies): Rates of binary restenosis were significantly lower with drug-eluting compared than with metal-stents (OR 0.16, 95% CI 0.12 to 0.20). There was moderate heterogeneity (Ι²=42.7%).
Late lumen loss (nine studies): This was significantly smaller for drug-eluting stents (mean difference -0.46mm, 95% CI -0.55 to -0.38), although heterogeneity was high (Ι²=81.5%).
No statistically significant differences between stent types were seen for mortality, myocardial infarction or stent thrombosis for pooled studies or in stratified analyses.