Author + information
VTE is regarded as the third general cause of cardiovascular disease beyond myocardial infarction and stroke. Vitamin K antagonists (VKA) are the mainstay of long-term anticoagulant therapy. Warfarin remains the primary approach which efficacy is well established. Although the emerging alternatives-novel oral anticoagulants for VTE have been investigated in some large randomized controlled trials, the advantages of VKA cannot be ignored. Warfarin is available to regular laboratory monitoring and may be reversed by administering vitamin K for the duration of its residence in the body. The effectiveness of VKA has been widely investigated, while the impact of bleeding risk is lacking. Duration of anticoagulation is determined by the risk of recurrence and then the risk of bleeding events. Several systematic reviews have investigated the effectiveness of different durations of treatment but none have focused on the aspect of bleeding risk. We performed this meta-analysis of randomized controlled trials (RCTs) enrolling patients with DVT or PE to compare the impact of major bleeding regarding the duration of VKA therapy.
We systematically searched PUBMED, Web of Science, Embase and Cochrane Library to identify all RCTs that involved patients with venous thromboembolism who were receiving VKA for at least 6 weeks. Two reviewers independently extracted population data and the reported major bleeding episodes. Durations of anticoagulant therapy compared in the articles and presented the outcome of the bleeding episodes of the three categories: short durations (6 weeks or 3 months), intermediate durations (6 or 12 months); indefinite therapy (≥24 months). Fixed effects model was performed to get estimates as relative risk (RR) and 95 % CI. Heterogeneity was evaluated with I2 test. All statistical analyses were conducted using Stata 12.0.
12 RCTs involving 4755 patients were analyzed. 9 trials evaluated intermediate or indefinite durations to short durations (6 weeks to 3 months) which mainly enrolled patients with idiopathic VTE. The longer durations of therapy were associated with risk of major bleeding (RR 3.02; 95% CI 1.58-5.78). Subgroup analyses of 5 trials (6 months vs. short durations) suggested that an increased risk of major bleeds (RR 3.27; 95% CI 1.45-7.38). Four studies have compared indefinite therapy with patients who stopped treatment at 6 months. The result (RR 1.90; 95% CI 0.95-3.77) shows that the risk of therapy beyond 24 months is lower than that of intermediate or indefinite duration to short duration.
The major bleeding risk in patients with VTE who is receiving VKA is an important clinical factor, and long-term oral anticoagulant therapy especially over 6 months is needed. Nevertheless, the safety impact of anticoagulation must be taken into account for the majority of patients.