Risk of Fatal Bleeding in Episodes of Major Bleeding with New Oral Anticoagulants and Vitamin K Antagonists: A Systematic Review and Meta-Analysis

BACKGROUND The reversibility of new/novel oral anticoagulants (NOAC) is not well understood, whereas the reversal strategies for bleeding associated with vitamin k antagonists (VKA), such as warfarin, is well established. It is unknown whether outcomes are different between bleeds occurring with NOAC compared to VKA use. OBJECTIVES This systematic review and meta-analysis of randomized controlled trials determines the relative odds of fatal bleeding given that a patient suffered a major bleed while on NOAC versus VKA therapy. SEARCH METHODS Data on major and fatal bleeding events was sought from randomized controlled trials of NOAC agents compared to VKAs. MAIN RESULTS 20 trials were included in the meta-analysis. From which, 4056 first-time, major bleeding events were reported and included in the primary analysis. The summary odds ratio for the conditional odds of fatal bleeding given that a major bleeding event occurred was 0.65 [0.52, 0.81] favoring the NOAC agents (p = 0.0001). The reduced odds of fatal bleeding with NOACs was not demonstrated after controlling for bleeding location. Given that an intracranial bleeding event occurred, the summary odds ratio for the conditional odds of fatal bleeding was 0.96 [0.70, 1.32]. For extracranial bleeding events, the summary odds ratio was also statistically insignificant at 0.945 [0.66, 1.35]. AUTHOR’S CONCLUSIONS: The odds ratio calculated in this meta-analysis showed a reduced odds of death in major bleeding associated with NOAC use. This risk reduction was due to a disproportionate amount of intracranial bleeding in the VKA arms. For any given bleeding site, there was no evidence of a significant difference in fatal outcomes from bleeds associated with NOAC versus VKA use. PROTOCOL REGISTRATION Protocol registered on PROSPERO under CRD42014013294.