Abstract P375: Contemporary Rates and Predictors of Death in Adults with Warfarin-Associated Bleeding Treated with Plasma: The REVERSAL Study
Background: Warfarin increases the risk of bleeding that is frequently treated with plasma for acute reversal in the setting of life-threatening hemorrhage. However, limited contemporary data exist about the risk and predictors of death in the first 45 days after treatment with plasma.
Methods: We conducted a retrospective cohort of all adult members in Kaiser Permanente Northern and Southern California, two integrated healthcare delivery systems caring for >6.8 million persons who received plasma for acute warfarin reversal in the setting of major bleeding between 2008-2013. We identified deaths from any cause during the first 45 days after receipt of plasma based on comprehensive health plan databases, Social Security Administration vital status files and state death certificates. Data on patient demographics, comorbidities and baseline medication use was ascertained from health plan administrative, inpatient and outpatient diagnosis and procedure, lab and pharmacy databases. We used multivariable logistic regression to identify independent predictors of death.
Results: Among 4509 adults who received plasma for acute warfarin reversal in the setting of major bleeding, mean (SD) age was 75 (11) years, 56% were women, 11% were black, 12% were Asian and 15% Hispanic. Thirty-nine percent of bleeds were intracranial, 57% gastrointestinal and 5% other major extracranial. There was a high prevalence of pre-existing morbidity, including 42% heart failure, 87% hypertension, 38% diabetes, 84% dyslipidemia and 38% chronic lung disease. During the first 45 days after receipt of plasma, 15.2% (95% CI: 14.2%-16.3%) of patients died from any cause. Multivariable predictors of death are shown in the Table.
Conclusions: Among a large, diverse, contemporary population of patients receiving plasma for acute warfarin reversal for major bleeding, >15% died in the first 45 days. Type of bleed, degree of over-anticoagulation, and selected demographic and clinical characteristics can identify patients at higher risk for early death.
Author Disclosures: A.S. Go: B. Research Grant; Significant; CSL Behring. T.K. Leong: None. U.M. Forssen: A. Employment; Significant; CSL Behring. D.M. Watson: A. Employment; Significant; CSL Behring. S. Sung: None. T.N. Harrison: None. K. Reynolds: B. Research Grant; Significant; CSL Behring.
- © 2015 by American Heart Association, Inc.