Objective: Two novel oral anticoagulants recently have been investigated for the treatment of deep vein thrombosis and/or pulmonary embolism: dabigatran etexilate (dabigatran) and rivaroxaban. The aim of this analysis was to compare their efficacy and safety.
Methods: Randomized, controlled trials investigating dabigatran or rivaroxaban were identified by a systematic review. Direct meta-analyses and anchored (adjusted) indirect comparisons (AICs) were performed using aggregated results for the following endpoints for the overall treatment duration: first recurrent symptomatic venous thromboembolism (VTE) or VTE-related death, major bleeding events (MBEs), MBEs or clinically relevant bleeding events (CRBEs), and all-cause mortality.
Results: Four trials were identified; two compared dabigatran with warfarin and two compared rivaroxaban with vitamin K antagonists. The results of the trials, and the direct meta-analyses of the two dabigatran trials and the two rivaroxaban trials, are presented in Table 1. Overall, there was little evidence of heterogeneity in treatment effects among the RE-COVER trials (VTE or VTE-related death, MBEs, MBEs or CRBEs, all-cause mortality: P= 0.82, 0.67, 0.97, 0.97, respectively; I2= 0%). There was some evidence of heterogeneity among the EINSTEIN trials for VTE or VTE-related death (P=0.11; I2=61.9%) and all-cause mortality (P=0.16; I2=50%), but none for MBEs and MBEs/CRBEs with I2= 0% (P= 0.43, 0.47). In the AICs, sensitivity analyses were performed to explore potential trial heterogeneity arising from differences in the type of index VTE and time with an international normalized ratio between 2.0 and 3.0. AIC results suggested that dabigatran was associated with a lower risk of MBEs/CRBEs than rivaroxaban (upper 95% confidence limits for relative risks were less than 1.00). There was no evidence to suggest a difference between dabigatran and rivaroxaban with respect to prevention of recurrent symptomatic VTE or VTE-related death, MBEs, or all-cause mortality.
Conclusions: Dabigatran may be associated with a lower risk of major or clinically relevant bleeding; there was no evidence to suggest a difference among drugs in prevention of recurrent VTE, MBEs, or overall mortality.
