Abstract TP216: Ischemic Stroke In Patients With Cancer Compared To Ischemic Stroke In Patients Without Cancer - A Cohort Study Using Synthetic Data

Abstract

Background: Patients with cancer are at an increased risk for ischemic stroke (IS) compared to those without cancer. The objectives of this study are 1) to examine risk factors for stroke in cancer and non-cancer patients, and 2) to identify predictive factors for recurrent IS in cancer patients. Methods: We performed a retrospective cohort study using MDClone, a platform that produces synthetic datasets based on real health system data from the Ottawa Hospital Data Warehouse. We analyzed all subjects with a diagnosis of cancer (excluding non-melanoma skin cancer or primary central nervous system malignancies) and IS within a 2-year period preceding and following their cancer diagnosis, and all IS patients without cancer, from the same time period (2000-2019). Patients were followed until May 2019. A forward selection, stepwise multivariate logistic regression model was used to assess the association between recurrent IS (primary outcome) and baseline characteristics. A sensitivity analysis was performed with only survivors in the cancer cohort. Results: We analyzed 10,875 subjects with IS: 1,250 had cancer and 9,625 did not. In cancer subjects, there was a higher prevalence of chronic obstructive pulmonary disease (8.4% vs 4.7%), previous IS (1.9% vs 0.1%), and previous venous thromboembolism (VTE) (8.3% vs 1.5%); the prevalence of atrial fibrillation and vascular risk factors was similar between the two groups. Recurrent IS occurred in 11.0% of cancer subjects and 12.1% of non-cancer subjects. In cancer subjects, the only significant predictor of recurrent IS was previous IS (OR 3.8, 95%CI 2.6 - 5.6). A sensitivity analysis of survivors amongst cancer subjects revealed an even stronger relationship between previous IS and recurrent IS (OR 4.4 95%CI 2.7 - 7.0). In the non-cancer subjects, significant predictors for recurrent IS include older age, while hypertension, dyslipidemia, and history of hemorrhagic stroke were negative predictors of recurrent IS. Conclusion: We found cancer patients with IS have a higher prevalence of COPD, previous IS and VTE, and that previous IS is an important predictor of recurrent IS in cancer patients. These results highlight the importance of identifying optimal secondary prevention treatments in cancer patients.