Re-bleeding and Mortality After Lower Gastrointestinal Bleeding in Patients Taking Anti-platelets or Anti-coagulants.
Oakland K., Desborough MJ., Murphy MF., Schachter M., Jairath V.
BACKGROUND & AIMS: Patients who develop lower gastrointestinal bleeding (LGIB) while receiving anti-coagulants or anti-platelets have increased severity of bleeding and risk of re-bleeding. We compared outcomes of patients receiving anti-platelets, anti-coagulants, or direct oral anti-coagulants (DOACs) who develop LGIB, as well as the effects of withholding these drugs on their course of bleeding. METHODS: We performed a retrospective study of 2528 consecutive adult patients with LGIB at 143 hospitals in the United Kingdom, from September through December 2015; 917 were taking anti-coagulant or anti-platelet drugs and 1218 were taking neither (unexposed). We collected data on demographic features of patients, interventions or medications, outcomes, laboratory test results, and patient readmission until patient death, discharge, or 28 days after admission (whichever came first). Re-bleeding was defined as additional transfusion requirements and/or a decrease in hematocrit ≥20% after 24 hrs of clinical stability. Multivariate regression was used to examine the relationship between drug class on presentation with LGIB and re-bleeding, mortality, and cardiovascular events. Rates of re-bleeding and cardiovascular complications in patients who had these drugs withheld were also analyzed. RESULTS: Patients receiving anti-platelets, but not those receiving warfarin (n = 232) or DOACs (n = 102), had a higher risk of in-hospital re-bleeding (monotherapy hazard ratio [HR], 3.57; 95% CI, 1.13-11.28; n = 504 and dual anti-platelet therapy hazard ratio, 5.3; 95% CI, 1.56-18.54; n = 79) compared with the unexposed group. This risk was not lower in patients who received anti-platelets and had the drug withheld for fewer than 5 days, compared to those who continued the drug throughout admission (HR, 0.98; 95% CI, 0.45-2.17) No differences were observed in risk-adjusted mortality or re-admission with further bleeding for patients receiving anti-platelets, DOACs, or warfarin. Cardiovascular events were too few to allow meaningful comparison. CONCLUSIONS: In patients with LGIB, antiplatelet drugs, but not warfarin or DOACs, are associated with an increased risk of re-bleeding. Withholding anti-platelets during admission does not lead to reduction in re-bleeding.