Management of Direct Acting Oral Anticoagulants in Hospitalized Patients With Upper Gastrointestinal Bleeding: A Real-World Observational Study

Introduction: Upper gastrointestinal bleeding (UGIB) remains a common medical emergency leading to hospitalization. Management of antithrombotic agents including direct acting oral anticoagulants (DOACs) in patients presenting with UGIB has become an increasingly common clinical challenge. Limited data exists to guide clinicians in weighing the risk benefit tradeoff of resuming DOAC therapy. We sought to describe post-procedural DOAC prescribing practices in a large, academic health system and to identify rates of rebleeding and thrombosis relative to when DOACs are restarted. Methods: We performed a retrospective cohort study of adults ( >18 years) between 2014-2018 who were taking a DOAC and underwent esophagogastroduodenoscopy (EGD) for signs or symptoms of overt UGIB within 3 hospitals in a large academic hospital system. Demographic information, current medications, past medical history, initial labs, vitals, endoscopic findings, and timing of DOAC re-initiation were extracted from the chart. Outcomes of interest were 30-day rebleeding and thrombotic events (Table 1). Results: Among 141 patients included in the study, 136 (96%) had DOAC held prior to endoscopy. 62 (44%) had an identifiable source on EGD, and of these, the most common etiology was peptic ulcer disease (18/62, 29%). Mechanical clips were placed in 12 patients while thermal therapy was performed in 10 patients. Of 136 patients who had DOAC held prior to endoscopy, 80 (59%) were restarted while 56 (41%) had their DOAC discontinued permanently. Overall, 30 patients developed rebleeding (21%). Regarding timing of DOAC re-initiation, 20/80 (25%) were restarted within 24 hours post-EGD; 47/80 (59%) between 24 hours and 7 days; 13/80 (14%) were restarted after 7 days. All rebleeding events occurred in patients who had their DOAC restarted within 7 days. 2 patients suffered strokes, both occurring within 10 days of DOAC being held. All-cause 30-day mortality occurred in 7/141 (5%) patients. Conclusion: In this cohort of hospitalized patients with UGIB, we observed substantial variability in DOAC prescribing practices, particularly in the timing of DOAC initiation after EGD. Rebleeding occurred in patients who had the DOAC restarted within 7 days, though thrombotic events occurred within a 10-day window in patients who had the DOAC held. Further work to expand this cohort with pre-defined subgroup analysis based on bleeding etiology is planned. Table 1. - Baseline Characteristics of Patients on DOACs Presenting with Acute Upper Gastrointestinal Bleeding (n=141) Mean Age (years) 73.8 Proportion (N=141) Gender Male 51% Female 49% Race White/Caucasian 70% Black/African American 18% Asian 1% Other 11% Ethnicity Non-Hispanic 91% Hispanic/Latino 9% Type of DOAC Apixaban 55% Rivaroxaban 45% Other Prior Medications Aspirin 39% Clopidogrel 11% Prasugrel 1% NSAID 6% Proton Pump Inhibitor 32% H2 Receptor Blocker 8% Presentation Out of Hospital Bleed 90% Inpatient Bleed 10% Past Medical History Stroke 13% Transient Ischemic Attack 4% Deep Vein Thrombosis 14% Pulmonary Embolism 16% Atrial Fibrillation 48% Gastrointestinal Bleed 8% Peptic Ulcer Disease 3% H. pylori 2% Coronary Artery Disease 44% Heart Failure 32% Hypertension 78% Type 2 Diabetes 40% Chronic Kidney Disease 11% Alcohol Use Disorder 4% Cirrhosis 1% Timing of DOAC Reinitiation Post-EGD Proportion (N=80) Within 24 Hours 25% 24 Hours - 7 Days 59% >7 Days 16% Rebleed Within 24 Hours 1% 24 Hours - 7 Days 19% >7 Days 0% Abbreviations: DOAC: Direct Acting Oral Anticoagulation; EGD: Esophagogastroduodenoscopy.