Evaluation of oral anticoagulants for the extended treatment of venous thromboembolism using a mixed-treatment comparison, meta-analytic approach.

PURPOSE: Target-specific oral anticoagulants (apixaban, rivaroxaban, and dabigatran) are widely available for the treatment of venous thromboembolism (VTE). Although analyses comparing these agents to placebo or warfarin exist, direct comparisons of these agents for extended VTE treatment have not been conducted. Therefore, this network meta-analysis aimed to evaluate the efficacy and tolerability of VKA and target-specific oral anticoagulants for extended VTE treatment using a mixed-treatment comparison, meta-analytic approach.
METHODS: A comprehensive literature search of EMBASE and MEDLINE was conducted to identify relevant randomized, controlled trials published in English between 1960 and November 2013. Eligible studies investigated the extended use (≥6 months) of oral anticoagulants (apixaban, dabigatran, rivaroxaban, and/or warfarin [conventional or low dose]) and placebo in patients with confirmed VTE. Search terms included extension or extended treatment or therapy, venous thromboembolism (or VTE), deep vein thrombosis (or DVT), pulmonary embolism (or PE), and anticoagulant or anticoagulant agent. Key articles were cross-referenced for additional studies. The efficacy end points evaluated were recurrent VTE or death from any cause, DVT, and nonfatal pulmonary embolism PE. Tolerability end points included major bleeding and nonmajor or clinically relevant bleeding. The data were screened, evaluated, and entered into statistical software to generate direct and indirect comparisons of the various anticoagulants across each study. The data are reported as rate ratios and 95% credible intervals.
FINDINGS: Ten trials were analyzed and aggregated, representing data from >14,000 patients. With respect to efficacy end points, no statistically significant between-treatment differences in the composite end point of VTE or death, nonfatal PE, or DVT were found. Major bleeding was significantly greater with warfarin versus apixaban (rate ratio, 4.24; credible interval, 1.28-25.0), and the risk for major bleeding varied somewhat with warfarin and greatly with rivaroxaban. The assessment of nonmajor or clinically relevant bleeding did not identify any meaningful differences between these agents. IMPLICATIONS: The majority of the data represented in this study were derived from noninferiority trials. In the present meta-analysis, efficacy end points in the extended treatment of VTE with apixaban, dabigatran, rivaroxaban, warfarin (conventional and low dose), and placebo were not significantly different. Elevated bleeding risks were identified with rivaroxaban and warfarin; however, the wide credible intervals with rivaroxaban prevent the interpretation of these increased risks.