Causes of Death in Patients with Venous Thromboembolism Anticoagulated with Direct Oral Anticoagulants: A Systematic Review and Meta-Analysis

Abstract

Death is more frequent than nonfatal recurrent venous thromboembolism (VTE) and major bleeding after acute VTE. The analysis of the causes of death is fundamental to explore new strategies to reduce mortality rates in these patients. The authors performed a meta-analysis to analyze mortality and independently adjudicated causes of death in anticoagulated patients due to VTE, and to evaluate potential differences between different anticoagulant schemes. They searched MEDLINE and CENTRAL, from January 1, 2000, to January 31, 2017, and performed additional searches in Web sites of regulatory agencies, clinical trial registers, and conference proceedings. Two investigators independently selected studies and extracted the data. Study quality was assessed with the Cochrane Collaboration's tool for assessing the risk of bias in randomized studies. Seven prospective randomized trials in 29,844 patients (22,025 patient-year follow-up) were included, comparing dabigatran, rivaroxaban, apixaban, and edoxaban with the standard anticoagulant treatment of VTE. A total of 718 patients died during the follow-up (3.4% per year; 95% confidence interval [CI]: 2.3-4.8). The most frequent causes of death were cancer (42%), followed by VTE (20%), infections (13%), hemorrhage (6%), heart disease (4%), and stroke (2%). There were no differences in the overall survival and causes of death according to the anticoagulant type. Concomitant active cancer during the study was significantly associated with death (odds ratio: 15.2; 95% CI: 9.2-25.1). Cancer is the leading cause of death in contemporary VTE trials. Interventions beyond anticoagulation, particularly in patients with active cancer, are needed.

Arguments for Using Direct Oral Anticoagulants in Cancer-Related Venous Thromboembolism

“…Unfortunately, DOACs had a low safety profile as a result of increasing the risk of major bleeding and CRNMB, even though the absolute risk differences were small (2-3%). Better compliance with DOACs than LMWHs was hypothesized to explain the differences in bleeding events [18].…”

Meta-regression of randomized control trials with antithrombotics: weak correlation between net clinical benefit and all cause-mortality

“…Similarly, cancer has been found to be the most common cause of death (42%) in a meta-analysis of seven randomized trials evaluating DOAC for the treatment of VTE 29 , whereas recurrent VTE and fatal bleeding have been estimated to be responsible for only 20% and 6% of deaths, respectively.…”