Category: Federal Forum Posters
Purpose: Oral anticoagulation, either in the form of vitamin k antagonists or direct oral anticoagulants (DOAC), is the recommended therapy for stroke prevention in atrial fibrillation and treatment of venous thromboembolism. The major safety concern while receiving oral anticoagulation is hemorrhage. Bleeding risk schemas exist but are validated in warfarin only. This study aims to determine risk factors associated with bleeding events while taking DOACs.
Methods: This pilot, retrospective case-control analysis was conducted at a single VA health care system. Patients with an active outpatient order for apixaban, rivaroxaban, or dabigatran for at least 90 days with a medication possession ratio of at least 80 percent, plus documented bleeding event comprised the case cohort. A bleeding event was defined as either an admission or discharge associated with any diagnosis code for bleeding. Patients were excluded if receiving edoxaban, receiving dabigatran 75mg twice daily, or if a bleed occurred within 72 hours after a procedure. Control patients were randomly selected in a 1:4 case-control ratio from remaining identified DOAC recipients. The chosen variables for evaluation were: age 65 or older, age 75 or older, history of prior bleed, anemia, thrombocytopenia, diabetes mellitus, uncontrolled hypertension, history of stroke, tobacco use, alcohol abuse, malignancy, kidney dysfunction, liver disease, gastrointestinal disorder, concomitant non-steroidal anti-inflammatory drug (NSAID) use, concomitant NSAID with proton pump inhibitor use, and concomitant antiplatelet use. Univariate analysis and logistic regression were used to identify potential predictors significantly associated with bleeding event occurrence. This study was approved by the institutional review board at the Department of Veterans Affairs Saint Louis Health Care System and was exempt from informed consent.
Results: A total of 68 bleeds were discovered, however only 50 patients with bleeding events met inclusion criteria. Of the remaining 1,094 patients, 200 were randomly chosen to make up the control cohort. Overall, the bleed rate was 5.9 percent. Baseline demographic information was evaluated across the type of DOAC the patient received. In general, there were not significant differences amongst the various DOACs, however significant differences were found between groups in patients age 65 or greater, age 75 or greater, DOAC indication, duration on DOAC, DOAC dosed according to the package insert, and kidney dysfunction. Univariate analysis identified the following as significant possible predictors: history of prior bleed, liver disease, anemia, and history of stroke. Age 65 or greater, per protocol, and the listed variables were entered into the regression. History of prior bleed (Odds ratio 2.91, 95 percent confidence interval (1.42-5.98);p=0.004), liver disease (Odds ratio 2.97, 95 percent confidence interval (1.05-8.35);p=0.039), and anemia (Odds ratio 1.99, 95 percent confidence interval (1.02-3.93);p=0.045) were identified as independently associated with bleeding events in this population.
Conclusion: This is the first independent trial to analyze the predictive role of various factors on bleeding events in patients receiving a DOAC. History of prior bleed, anemia, and liver disease were associated with a bleeding event in patients receiving a DOAC. The results of this analysis may aid in identification of factors associated with higher bleeding rates in patients receiving DOACs.
Kaitlen Shumate– Pharmacist, Saint Louis VA Health Care System, Vancouver, WA