Clinical Outcomes after Hospitalization for Oral Anticoagulant-Related Bleeding

Background: Use of oral anticoagulant (OAC) therapy for the prevention and treatment of thromboembolism is limited by serious bleeding complications, the most common OAC-related adverse event resulting in emergency department (ED) visits, hospitalization, and death. There are limited data regarding patient outcomes after hospitalization for OAC-related bleeding. Our objectives were to describe the risk of death, recurrent bleeding, and thromboembolism [deep vein thrombosis, pulmonary embolism, myocardial infarction, ischemic stroke] after hospitalization for OAC-related bleeding, and to identify risk factors for these outcomes.

Methods: Using administrative healthcare data from Ontario, Canada, we conducted a population-based cohort study of adults aged >65 years who were discharged after incident hospitalization for bleeding with OAC dispensed in the preceding 100 days (April 1, 2012 - March 31, 2020). The primary outcome was mortality within 100 days after hospital discharge. Secondary outcomes were the cumulative incidence of all-cause mortality within 1 year, and hospital and ED admissions for bleeding and thromboembolism. We examined associations between baseline covariates and mortality using multivariable Cox regression models to calculate adjusted hazard ratios (HR) and 95% confidence intervals (CI), and Fine-Gray regression models to calculate adjusted sub-distribution HR and 95%CI for thrombosis, bleeding, and mortality.

Results :Among 16180 cohort members, 14414 (89%) survived the index hospitalization and were included. Index bleeds were gastrointestinal (GI; n=9450, 66%), intracranial (n=1750, 12 %), genitourinary (GU; n=1178, 8%) or other (n=2036, 14%). The mean age was 81 years and 47% were female. Atrial fibrillation was the main indication for OAC (n=9,351, 65%). Patients were frequently prescribed factor Xa inhibitors (51%) and warfarin (39%). Within 100 days of discharge, OACs were dispensed to 9420 patients (65%), 87% of whom were prescribed a factor Xa inhibitor.

Within 100 days of discharge, 28% of patients were re-hospitalized. The estimated cumulative incidence of mortality was 12% over 100 days and 24% over 1 year. Mortality was highest among individuals hospitalized for intracranial or GI bleeding over both 100 days (16% and 12%) and 1 year (26% and 24%). Using competing risk methods, the cumulative incidence of rebleeding was 11% (highest for index GI bleeding [11%]). The cumulative incidence of thromboembolism was 3% (highest for index GI bleeding [3%]). Baseline covariates associated with an increased risk of mortality were: cancer (HR 2.82; 95%CI 2.36-3.36), discharge to long-term care (ref: home, HR 2.06; 95%CI 1.82-2.34), Elixhauser comorbidity index ≥4 (HR 1.46; 95%CI 1.31-1.64), venous thromboembolism (ref: atrial fibrillation, HR 1.36; 95%CI 1.13-1.64), intracranial bleeding (ref: GU, HR 1.34; 95%CI 1.06-1.68), congestive heart failure (HR 1.35; 95%CI 1.22-1.50), dementia (HR 1.33; 95%CI 1.19-1.50), and increasing age (1-year older HR 1.05; 95%CI 1.04-1.06).

Conclusion: Patients who survived hospitalization for OAC-related bleeding experienced substantial mortality, one-quarter dying within 1 year of hospital discharge. This confirms that older adults who are hospitalized for bleeding have a poor prognosis irrespective of index bleed site. The presence of cancer, discharge to long-term care and higher comorbidity burden had the strongest associations with mortality within 100 days. Evidence-based strategies are needed to identify and address risk factors to prevent OAC-related bleeding and to improve outcomes after serious bleeding.

Siegal:Servier: Other: honoraria paid indirectly to research institute; Roche: Other: honoraria paid indirectly to research institute; BMS-Pfizer: Other: honoraria paid indirectly to research institute; Astra Zeneca: Other: honoraria paid indirectly to research institute.