Abstract
BACKGROUND AND OBJECTIVES: Ischemic stroke and cancer often coexist. The aim of this study was to determine the prevalence of cancer among patients with stroke, their antithrombotic treatment patterns, and the risk of subsequent thromboembolic events in a large population-based cohort. METHODS: This retrospective cohort study used data from Optum's deidentified Clinformatics Data Mart Database to assess the 6-month risk of recurrent thromboembolic and major bleeding events identified using validated International Classification of Diseases code algorithms, among adults with active cancer hospitalized with ischemic stroke between October 2020 and September 2021. In addition, we explored the risk of recurrent thromboembolic events stratified by anticoagulant exposure. Landmark analyses were undertaken with anticoagulant exposure status based on dispensations in the week after hospital discharge. Propensity score matching aimed to minimize selection bias between the groups. Hazard ratios (HRs) were estimated using Cox proportional hazard models with robust variance estimators to account for clustering within propensity score-matched pairs. RESULTS: Among 86,365 patients identified with ischemic stroke, 10.2% had active cancer (most common types: 34% genitourinary, 25% gastrointestinal, 23% hematologic). After applying eligibility criteria, 4,781 patients were included in the main analysis (median age 74 [interquartile range 68-81] years; 48% female). Anticoagulants were dispensed to 14.7% of patients. The incidence of thromboembolic event recurrence and major bleeding events was 38.4 (95% CI 35.4-41.6) and 20.9 (95% CI 18.8-23.4) per 100 person-years, respectively. After propensity score matching, the risk of thromboembolism recurrence was not statistically different between those with and without anticoagulant prescription (HR 1.21; 95% CI 0.91-1.61). Anticoagulant prescription was also not associated with a higher risk of major bleeding events (HR 1.13; 95% CI 0.78-1.63). DISCUSSION: In this large-scale study, approximately 10% of patients with ischemic stroke had active cancer and they faced a markedly elevated short-term risk of recurrent thromboembolism and major bleeding events. Anticoagulants were infrequently prescribed after discharge and did not differentially affect the rate of ischemic or bleeding outcomes compared with alternative treatment strategies, although confounding by indication is likely. Clinical trials are required to further assess optimal antithrombotic strategies in this high-risk population.