Correction: ARF Confers a Context-Dependent Response to Chemotherapy in Muscle-Invasive Bladder Cancer.
Article
Overview
abstract
BACKGROUND: Each year, about half a million Americans experience a venous thromboembolism (VTE), which includes deep vein thrombosis and pulmonary embolism. Oral anticoagulants (OACs) have shown to reduce the recurrence of VTE and decrease mortality in patients with VTE. Yet, the optimal duration of anticoagulation after VTE is unknown. Guidelines recommend anticoagulation for at least 3 months, with treatment beyond 3 months for patients in certain clinical scenarios. Extending therapy beyond the initial 6 to 12 months has been shown to decrease the cumulative risk of VTE recurrence but comes with the risk of bleeding. Currently, few data are available on the optimal duration of therapy and the benefit-to-risk profile of extended anticoagulation for patients with VTE. OBJECTIVES: The primary objective was to evaluate the effectiveness and safety of extended anticoagulant treatment for the secondary prevention of VTE. More specifically, we evaluated the safety and effectiveness associated with extending OAC treatment beyond the initial 3 months compared with stopping therapy; among patients who extended treatment, we compared the safety and effectiveness of the 3 most common agents (apixaban, rivaroxaban, and warfarin). In addition, we examined heterogeneity in treatment effects across subgroups of age, sex, and relevant clinical factors. METHODS: The study used a parallel-group, new-user cohort design with propensity score–based adjustment for confounding. The study was conducted with data from 3 electronic health care claim databases (Medicare, Optum Clinformatics, and IBM MarketScan). The study population consisted of patients with an initial hospitalization for deep vein thrombosis or pulmonary embolism (2010 to March 2021) who had at least 365 days of insurance enrollment before VTE hospitalization and no OAC use during that year. To be included, patients were required to fill a prescription for an OAC within 30 days after hospital discharge following VTE. For the evaluation of extended treatment, patients must have completed at least 90 days (180, 270, or 360 days in secondary analyses) of OAC treatment and follow-up for outcomes started after completion of the required initial treatment. Primary outcomes were recurrent VTE, defined as hospitalization for a VTE diagnosis as the principal discharge diagnosis, and major bleeding, defined as hospitalization with discharge diagnoses for an intracranial bleed, gastrointestinal bleeding, or other major bleeding in any position. For the comparison of extending therapy with stopping therapy, patients could have been on any OAC. For the comparison of individual agents in patients who extended treatment, we compared the 3 most commonly used anticoagulants: apixaban, rivaroxaban, and warfarin. Propensity score weighting was used to account for between-group differences in measured characteristics, including VTE and bleeding risk factors. Weighted Cox proportional hazards models with robust standard errors were used to estimate hazard ratios (HRs) and 95% CIs. Several prespecified subgroup analyses were conducted for each analysis; for each subgroup, propensity scores were reestimated within that subgroup, and patients were reweighted within each database. RESULTS: The final study sample included 90 921 eligible patients who completed at least 90 days of initial anticoagulant treatment (mean age, 70 years; 57% female), of whom 67 720 continued treatment and 23 201 stopped treatment after 90 days. Patients who continued treatment beyond 90 days experienced substantially lower rates of VTE recurrence than did patients who stopped after 90 days (HR, 0.30 [95% CI, 0.25-0.35]), with somewhat higher rates of major bleeding (HR, 1.16 [95% CI, 0.99-1.37]) relative to those who stopped. The results were consistent across analyses of patients who extended therapy beyond the initial 180 days, 270 days, or 360 days. In the comparative analysis of individual anticoagulants, among eligible patients who continued therapy beyond the initial 90 days (N = 64 642), those who were on apixaban experienced lower rates of VTE recurrence than those on warfarin (HR, 0.69 [95% CI, 0.49-0.99]), although rates were comparable with rivaroxaban (HR, 0.80 [95% CI, 0.53-1.19]). There were no differences in rates of major bleeding across the 3 agents among patients who extended therapy beyond the initial 90 days. Subgroup analyses revealed similar results across subgroups of age, sex, and clinically relevant factors. CONCLUSIONS: In patients with VTE, extending therapy past 90 days is associated with substantially lower rates of VTE recurrence and higher rates of bleeding. Among patients who extended therapy beyond the initial 90 days, apixaban appeared to be associated with lower rates of VTE recurrence than warfarin; rates of bleeding were similar across the 3 agents. LIMITATIONS: Because patients were not randomly assigned to groups, confounding by unmeasured factors cannot be ruled out. Although we measured many risk factors for both VTE and bleeding, some variables that may have confounded our analyses, such as laboratory values, socioeconomic status, and smoking, are not recorded in the health care claim databases. Findings may have also been affected by exposure and outcome misclassification.
