Background: The recently proposed VTE-BLEED score to predict the risk of major bleeding in patients with venous thromboembolisms (VTEs) on prolonged anticoagulation therapy was validated externally in randomized controlled trials and in a selected cohort, but not applied in non-selective cohorts.
Objectives: The present study aimed to evaluate the generalizability of the VTE-BLEED score in a non-selective cohort.
Patients/methods: The COMMAND VTE Registry is a multicenter retrospective registry enrolling consecutive patients with acute symptomatic VTEs among 29 centers in Japan. The present study population consisted of 2124 patients with prolonged anticoagulation therapy beyond 30 days, including 2008 (95%) patients with a first VTE episode and 1075 (51%) with unprovoked VTEs, who were divided into 1445 patients (68%) with a VTE-BLEED score of ≥2 (high-risk group) and 679 (32%) with a VTE-BLEED score of <2 (low-risk group).
Results: During a median follow-up period of 672 days, major bleeding events occurred in 121 patients. The cumulative 5-year incidence of major bleeding beyond 30 days was significantly higher in the high-risk group than low-risk group (13.2% versus 5.4%, P < .001). The hazard function curves demonstrated that the hazards of the high-risk group were consistently higher over time compared with the low-risk group, which suggested the long-term predictive ability of the score.
Conclusions: In the present real-world VTE registry, the VTE-BLEED score had a long-term predictive ability for high-risk patients with major bleeding during prolonged anticoagulation therapy, which could be useful in determining the optimal duration of anticoagulation therapy in individual patients.
Keywords: anticoagulants; cohort studies; hemorrhage; risk assessment; venous thromboembolism.
© 2019 International Society on Thrombosis and Haemostasis.