Abstract
INTRODUCTION: Warfarin is a widely used and easily reversible anticoagulant. Although bleeding is more likely in warfarin users, it may also be more readily treated. This retrospective observational case-control study compares the outcome of acute nonvariceal upper gastrointestinal hemorrhage in warfarin users with a supratherapeutic international normalized ratio (INR) and outcome in non - warfarin users. PATIENTS AND METHODS: Clinical and endoscopic data for patients presenting with overt upper gastrointestinal hemorrhage were collected between 23rd February 2001 and 12 October 2010. Patients with variceal hemorrhage were excluded. Warfarin users with a supratherapeutic INR (≥ 3.0) at presentation (supratherapeutic anticoagulation [SA] group) were matched to a cohort with upper gastrointestinal hemorrhage not taking warfarin at presentation (control group). Patients were matched by age, sex, Rockall score, year of endoscopy, inpatient or outpatient status, and the presence of disseminated cancer at presentation. The incidence rates of major outcomes in the two groups were compared. RESULTS: A total of 128 patients (SA group) were matched to 135 control patients. The SA group patients were less likely to die within 30 days (6.25 % vs. 15.5 %, odds ratio = 0.36, P = 0.028 by Test for Equality of Proportions). There was a trend toward more surgery in the control group (5 % vs. 2 %), and rates of blood transfusion (77 % vs. 70 %) were similar in the two groups. CONCLUSION: In patients presenting with nonvariceal upper gastrointestinal hemorrhage, a supratherapeutic INR at presentation due to warfarin use is associated with reduced mortality.