Objectives: To evaluate the risk of cerebrovascular events (CVEs) associated with selective cyclooxygenase-2 inhibitors (coxibs).
Method: Systematic review and meta-analysis of randomized controlled trials (RCTs). A fixed-effect model was used to estimate the odds ratios (ORs) for risk of CVE associated with coxibs compared against placebo, non-selective non-steroidal anti-inflammatory drugs (NSAIDs) and other coxibs.
Results: Forty trials (88 116 patients) were included in the meta-analysis. The overall pooled OR for CVE for any coxib against placebo was 1·03 (95% CI: 0·71, 1·50). Comparing individual coxibs against placebo, we found that celecoxib, rofecoxib, etoricoxib and lumiracoxib were associated with higher CVE risks and valdecoxib was associated with a lower CVE risk, although there were no significant differences detected. There was also no significant difference in risk of CVE when comparing coxibs against any non-selective NSAIDs; the corresponding pooled OR was 0·86 (95% CI: 0·64, 1·16).
Conclusion: On the basis of a detailed analysis of available RCTs, there does not appear to be any significant difference in risk of CVEs associated with coxibs when compared against placebo or non-selective NSAIDs. It is likely that the increased risk of thrombotic vascular events associated with coxibs is largely attributable to an increased risk of myocardial infarction, rather than CVEs.