Increased Risk of Myocardial Infarction with Dabigatran Etexilate: Fact or Fiction? A Critical Meta-Analysis of over 580,000 Patients from Integrating Randomized Controlled Trials and Real-World Studies

An-Hua Wei,Zhi-Chun Gu,Chi Zhang,Yu-Feng Ding,Dong Liu,Juan Li,Xiao-Yan Liu,Hou-Wen Lin,Jun Pu
DOI: https://doi.org/10.1016/j.ijcard.2018.05.048
IF: 4.039
2018-01-01
International Journal of Cardiology
Abstract:Background: The question of whether the use of dabigatran etexilate is associated with a high risk of myocardial infarction (MI) remains unanswered owing to the lack of critical evidences. Methods: A comprehensive search of databases (Medline, Embase, Cochrane Library databases, and ClinicalTrials.gov Website) was performed for RCTs that reported MI events and observational nationwide database studies that reported adjusted hazard ratio (HR) with dabigatran treatment. Summary HRs and 95% confidence intervals (95% CI) were calculated using random-effects models. Cumulativemeta-analysis was conducted for evaluating the results as a continuum, and subgroup analyses were undertaken on the basis of study type, indication, controls, and dosage. Results: Finally, 24 studies including 588,047 patients (44,856 patients in 14 RCTs and 543,191 patients in 10 observational database studies) met the inclusion criteria, among which 222,352 (37.8%) patients receiving dabigatran and 365,695 (62.2%) patients receiving placebo/other anticoagulants. In comparison to controls, no significant association was detected between the use of dabigatran and the higher risk of MI (HR: 0.97, 95% CI: 0.87-1.06; I-2 for heterogeneity: 26.3%, P = 0.089). The results were consistent across the key subgroups (indication, controls, and dosage, P-interaction > 0.05 for each), with the exception of study type (RCTs or database studies, P-interaction = 0.046). Cumulative meta-analysis was not suggestive of a temporal trend in the effect of dabigatran on MI. Conclusions: This meta-analysis confirms a low risk of MI in patients exposed to dabigatran, which seems to be validated when pooling over 580,000 patients from RCTs and real-world studies. (C) 2018 Elsevier B.V. All rights reserved.
What problem does this paper attempt to address?