Effects of aspirin and acetaminophen in gastrointestinal hemorrhage. Results from the Boston Collaborative Drug Surveillance Program.

All patients observed in this analysis of the Boston Collaborative Drug Surveillance Program data had no known disease predisposing them to gastrointestinal (GI) bleeding. Major GI bleeding occurred in hospitalized patients recently exposed to ethacrynic acid (5/111; 4.5%), heparin sodium (7/575; 1.2%), corticosteroids (7/1,484; 0.5%), aspirin (6/2,081; 0.3%), and warfarin sodium (1/423; 0.2%); combinations of two or more of these drugs also resulted in substantial incidences of bleeding. The rates for minor GI bleeding in hospitalized patients varied from 8.3% for heparin to 1.6% for aspirin. In outpatients admitted because of serious GI bleeding, 14/88 (16%) had a history of heavy, regular aspirin use compared with 1,015/14,813 (6.9%) control subjects admitted for other reasons (relative risk, 2.1). No association between light regular use of aspirin and GI bleeding was noted. No association between the use of acetaminophen and GI bleeding was noted in any of the patients studied.