Announcement
Thank you for attending THSNA 2026. The virtual meeting is now closed.
Thank you for attending THSNA 2026. The virtual meeting is now closed.
Presentation Details
| A Novel Approach to Estimating the Risk of Gastrointestinal Bleeding for Patients on Oral Anticoagulants and Other Medications Including Non-Steroidal Anti-Inflammatory Drugs and Antidepressants Daniel C.Malone1, Ainhoa Gomez-Lumbreras1, Abdelrahman G.Tawfik1, Daniel M.Witt1, Guilherme Del Fiol2, Thomas Reese3. 1Department of Pharmacotherapy, College of Pharmacy, University of Utah, Salt Lake City, UT, USA.23Department of Biomedical Informatics, School of Medicine, University of Utah, Salt Lake City, UT, USA.3Department of Biomedical Informatics, Vanderbilt University Medical Center, Nashville, TN, USA |
Abstract
Background: Drug-drug interactions (DDI) in patients receiving oral anticoagulants (OAC) may increase the risk of gastrointestinal bleeding (GIB). Objectives: To develop and validate a risk model to predict GIB among individuals receiving OAC and interacting medications. Methods: A model (Drug Interactions in Oral AntiCoagulants and risk of GastroIntestinal Bleeding (DIOAC-GIB)) was developed to predict GIB based on interacting medications and other patient-specific risk factors. Inputs for the model were derived from a synthesis of evidence for the following attributes: exposure to nonsteroidal anti-inflammatory drugs (NSAIDs), selective serotonin reuptake inhibitors (SSRIs)/ selective norepinephrine reuptake inhibitors (SNRIs), aspirin, antiplatelets, and proton pump inhibitors (PPIs); age; and previous GIB history. GIB risk estimates for these attributes were calculated using random-effect meta-analysis or obtained from the literature. Estimates of GIB risk for OACs (i.e., apixaban, dabigatran, edoxaban, rivaroxaban, and warfarin) were based on pivotal trials. The risk model was incorporated into a dynamic clinical decision support tool called DDInteract and was integrated into the EPIC electronic health records (EHR) system at three academic medical centers. For patients receiving an OAC, DDInteract populates risk factors and provides clinicians with an estimate of the number of bleeds per 100 patients based on the patient-specific risk factors. Clinicians can explore selecting different NDSAIDs and/or SSRI/SNRI agents to estimate patient-specific bleeding risk. DIOAC-GIB model was also compared via correlation to existing OAC models (HAS-BLED and DOAC score) for a cohort of patients receiving apixaban, rivaroxaban, or warfarin at a single academic medical center. Results: Risk of GIB for OAC products obtained from pivotal studies were: 1.99% for apixaban; 3.15% for rivaroxaban; 2.99% for dabigatran; 3.31% for edoxaban; and 2.00% for warfarin. The odds of GIB for NSAIDs ranged from 1.16 (95% CI:0.84-1.61) for celecoxib to 20.67 (95% CI:14.56-29.34) for ketorolac. Odd ratios for GIB among SSRI/SNRI ranged from 1.31 (95% CI:1.07-1.62) for paroxetine to 1.50 (95% CI:1.32-1.70) for venlafaxine. Risk of GIB for other factors included: persons over 65 years of age (OR=2.50, 95% CI:1.2–5.5); previous history of GIB (OR=5.13, 95% CI:3.95–6.67); aspirin use (OR=1.31, 95% CI:0.93-1.85); antiplatelets use (OR=1.95, 95% CI:1.68-2.27); and corticosteroids use (OR=1.68, 95% CI:1.49-1.90). PPIs reduced GIB risk (OR=0.67, 95% CI=0.62–0.74). To validate the model, data for 15,357 persons receiving either apixaban (n=10,526, 68.6%), rivaroxaban (n=2509, 16.3%), or warfarin (n=2322, 15.1%) was evaluated. Among warfarin users, 15.0% were receiving at least one NSAID, as compared to 29.0% for apixaban or rivaroxaban (p<0.0001). The use of SSRIs/SNRIs were similar for warfarin (36.2%) and apixaban/rivaroxaban (36.6%) (p=0.77). The average (SD) GIB risk scores for DIOAC-GIB were 5.48 (SD 5.51) for warfarin users and, 4.95 (SD 5.52) for DOAC users, while the mean HAS-BLED (warfarin only) and DOAC scores were 1.65 (SD 0.97) and 3.86 (SD 2.57), respectively. Correlations between the DIOAC-GIB and both the HAS-BLED and DOAC scores were 0.5 for both (p<0.05). Conclusion: The DIOAC-GIB incorporates drug interactions with OACs and had good correlation with other bleeding risk scores. DIOAC-GIB may help identify high-risk patients that benefit from tailored prescribing.
No part of this publication may be reproduced, distributed, or transmitted in any form or by any means, including photocopying, recording, or other electronic or mechanical methods, without the prior written permission of the author.
No part of this publication may be reproduced, distributed, or transmitted in any form or by any means, including photocopying, recording, or other electronic or mechanical methods, without the prior written permission of the author.