Effectiveness and safety of extended treatment with direct oral anticoagulants for venous thromboembolism in Japan: A retrospective cohort study using claims data

Purpose

We aimed to evaluate the risk of recurrent venous thromboembolism (VTE) among patients receiving direct oral anticoagulant (DOAC) therapy (extended and non-extended) using claims data.

Methods

Patients treated with DOACs at the first VTE event were identified using Japanese claims data; Cox proportional hazard models were used to evaluate the risk of recurrence. Unadjusted, adjusted, and stabilized inverse probability treatment weighting (s-IPTW) analyses were used to assess the differences between patients treated for more than 90 days after the index date (extended treatment group) and those treated within 90 days (nonextended treatment group). Bleeding was assessed separately from recurrent VTE evaluation during the observation period.

Results

We included 4,010 patients (mean age, 69 years; 57.9% females; extended: 2,684, nonextended: 1,326). After IPTW, all patient characteristics were well balanced. When the follow-up was censored at 18 months, the hazard ratio (extended/nonextended) for unadjusted, adjusted, and two s-IPTW analyses were 1.31 (P = 0.2762), 1.25 (P = 0.372), 1.32 (P = 0.2579), and 1.33 (P = 0.2498), respectively. The proportions of those who experienced intracranial, gastrointestinal, respiratory, and renal/urinary tract bleeding during the observation period in both groups were 2.9% vs. 3.8%, 2.1% vs. 2.3%, 1.0% vs. 0.5%, and 1.4% vs. 0.7% (extended vs. nonextended), respectively.

Conclusions

There were no differences in VTE recurrence between the two groups. In clinical practice and with a risk-benefit evaluation, both VTE treatment types were well controlled. Further evaluation is required, considering patient background within the observation period of 90 days and safety of treatment for bleeding events.