Whether the two drug-eluting stents approved by the US Food and Drug Administration—a sirolimus-eluting stent and a paclitaxel-eluting stent—are associated with increased risks of death, myocardial infarction, or stent thrombosis compared with bare-metal stents is uncertain. Our aim was to compare the safety and effectiveness of these stents.

We searched relevant sources from inception to March, 2007, and contacted investigators and manufacturers to identify randomised controlled trials in patients with coronary artery disease that compared drug-eluting with bare-metal stents, or that compared sirolimus-eluting stents head-to-head with paclitaxel-eluting stents. Safety outcomes included mortality, myocardial infarction, and definite stent thrombosis; the effectiveness outcome was target lesion revascularisation. We included 38 trials (18 023 patients) with a follow-up of up to 4 years. Trialists and manufacturers provided additional data on clinical outcomes for 29 trials. We did a network meta-analysis with a mixed-treatment comparison method to combine direct within-trial comparisons between stents with indirect evidence from other trials while maintaining randomisation.

Mortality was similar in the three groups: hazard ratios (HR) were 1·00 (95% credibility interval 0·82–1·25) for sirolimus-eluting versus bare-metal stents, 1·03 (0·84–1·22) for paclitaxel-eluting versus bare-metal stents, and 0·96 (0·83–1·24) for sirolimus-eluting versus paclitaxel-eluting stents. Sirolimus-eluting stents were associated with the lowest risk of myocardial infarction (HR 0·81, 95% credibility interval 0·66–0·97, p=0·030 vs bare-metal stents; 0·83, 0·71–1·00, p=0·045 vs paclitaxel-eluting stents). There were no significant differences in the risk of definite stent thrombosis (0 days to 4 years). However, the risk of late definite stent thrombosis (>30 days) was increased with paclitaxel-eluting stents (HR 2·11, 95% credibility interval 1·19–4·23, p=0·017 vs bare-metal stents; 1·85, 1·02–3·85, p=0·041 vs sirolimus-eluting stents). The reduction in target lesion revascularisation seen with drug-eluting stents compared with bare-metal stents was more pronounced with sirolimus-eluting stents than with paclitaxel-eluting stents (0·70, 0·56–0·84; p=0·0021).

The risks of mortality associated with drug-eluting and bare-metal stents are similar. Sirolimus-eluting stents seem to be clinically better than bare-metal and paclitaxel-eluting stents.