The risks of thromboembolism vs. recurrent gastrointestinal bleeding after interruption of systemic anticoagulation in hospitalized inpatients with gastrointestinal bleeding: a prospective study

Abstract

Objectives: Anticoagulants carry a significant risk of gastrointestinal bleeding (GIB). Data regarding the safety of anticoagulation continuation/cessation after GIB are limited. We sought to determine the safety and risk of continuation of anticoagulation after GIB.

Methods: We conducted a prospective observational cohort study on consecutive patients admitted to the hospital who had GIB while on systemic anticoagulation. Patients were classified into two groups at hospital discharge after GIB: those who resumed anticoagulation and those who had anticoagulation discontinued. Patients in both groups were contacted by phone 90 days after discharge to determine the following outcomes: (i) thromboembolic events, (ii) hospital readmissions related to GIB, and (iii) mortality. Univariate and multivariate Cox proportional hazards were used to determine factors associated with thrombotic events, rebleeding, and death.

Results: We identified 197 patients who developed GIB while on systemic anticoagulation (n=145, 74% on warfarin). Following index GIB, anticoagulation was discontinued in 76 patients (39%) at discharge. In-hospital transfusion requirements, need for intensive care unit care, and etiology of GIB were similar between the two groups. During the follow-up period, 7 (4%) patients suffered a thrombotic event and 27 (14%) patients were readmitted for GIB. Anticoagulation continuation was independently associated on multivariate regression with a lower risk of major thrombotic episodes within 90 days (hazard ratio (HR)=0.121, 95% confidence interval (CI)=0.006-0.812, P=0.03). Patients with any malignancy at time of GIB had an increased risk of thromboembolism in follow-up (HR=6.1, 95% CI=1.18-28.3, P=0.03). Anticoagulation continuation at discharge was not significantly associated with an increased risk of recurrent GIB at 90 days (HR=2.17, 95% CI=0.861-6.67, P=0.10) or death within 90 days (HR=0.632, 95% CI=0.216-1.89, P=0.40).

Conclusions: Restarting anticoagulation at discharge after GIB was associated with fewer thromboembolic events without a significantly increased risk of recurrent GIB at 90 days. The benefits of continuing anticoagulation at discharge may outweigh the risks of recurrent GIB.