Dabigatran is a direct thrombin inhibitor which was approved by the FDA in 2010 for the prevention of stroke and systemic embolism in patients with non-valvular atrial fibrillation.1 This approval was based on the results of the RE-LY trial.2 One of the incidental findings from the RE-LY trial is that significantly more myocardial infarctions (MI) were seen in the dabigatran arms than in the warfarin arm. After a review of the data from the RE-LY trial identifying additional events, more MIs were still observed in the dabigatran arms compared to the warfarin arm, but the difference was no longer statistically significant.3 Because of the possible increased risk of MIs with dabigatran, Uchino and Hernandez performed a meta-analysis to evaluate this association.4
The meta-analysis evaluated seven randomized controlled trials with dabigatran. The primary outcome for this study was MI or acute coronary syndrome (ACS) which was defined as confirmed unstable angina, MI, and cardiac death. The secondary outcome was overall mortality. Comparisons for the primary and secondary outcomes were made between a dabigatran and control arm. The dabigatran arm was comprised of patients receiving any dose of dabigatran. The control arm was comprised of patients receiving enoxaparin, warfarin, or placebo. Several additional sensitivity analyses were performed including using the revised RE-LY trial data, excluding short-term trials (≤1 month), and excluding low-quality trials to see what impact these factors would have on the primary analysis. The trials were also tested for heterogeneity.
There were 30,514 patients included in this meta-analysis. The dabigatran arm was shown to have a significant increase in risk of MI or ACS compared to the control arm in the primary analysis (1.19% vs. 0.79%, ORM-H 1.33, 95% CI 1.03-1.71, P=0.03). However, the absolute risk was small (0.27%) corresponding to a numbers needed to harm of 337. The sensitivity tests using the revised RE-LY trial data, excluding short-term trials, and excluding low-quality trials all yielded similar results showing a significant increase in MI or ACS with dabigatran. Heterogeneity among the trials was not detected. For the secondary outcome, dabigatran was shown to significantly lower overall mortality based on the six trials reporting overall mortality data (4.83% vs. 5.02%, ORM-H 0.89, 95% CI 0.80-0.99, P=0.04). However, heterogeneity was observed in risk differences for the trials in this analysis. In addition, the meta-analysis found no relationship between baseline risk of acute coronary events and the risk for acute coronary events with dabigatran.
This meta-analysis showed that the increased risk of MI or ACS seen with dabigatran can be detected beyond the RE-LY trial where this risk was initially observed. While this study highlights a potential safety issue with dabigatran, there are limitations to the meta-analysis. The RE-LY trial was weighted heavily in this meta-analysis encompassing 59% of the patients studied and 74% of the events reported.4 Another limitation of this meta-analysis is that most of the patients in the control arm received an active ...