[Upper gastrointestinal hemorrhage caused by anti-inflammatory agents].

The aim of this study was to describe the clinical and evolutive characteristics of gastroduodenal bleeding occurring in patients receiving nonsteroidal anti-inflammatory (NSAI) drugs, containing salicylates or not, and to determine the relative toxicity of the NSAI drugs without salicylates. Eight hundred and fourty-five consecutive patients with upper gastrointestinal bleeding related to endoscopically proven peptic ulcer or gastroduodenal erosions were admitted between 1983 and June 1987 to an intensive care unit for digestive tract hemorrhage. Of these, 267 were using anti-inflammatory drugs; 151 (56 p. 100) were taking NSAI drugs other than salicylates, 97 salicylates (36 p. 100) and 10, steroids (4 p. 100). Patients taking nonsteroidal drugs without or with salicylates were compared with patients bleeding from gastroduodenal ulcer or erosion not receiving anti-inflammatory therapy. Patients receiving nonsteroidal drugs not containing salicylates were older (70 p. 100 over 65 years of age vs 46 p. 100, p less than 0.001) and the proportion of female patients was greater (54 p. 100 vs 33 p. 100, p less than 0.001) than in the other group. No significant difference was observed with regard to the following parameters: percentage of gastric lesions, concomitant anticoagulant therapy, need for surgical hemostasis, or mortality. Patients taking aspirin had more gastric lesions (75 p. 100 vs 64 p. 100, p less than 0.05) and less need for surgical hemostasis (7 p. 100 vs 15 p. 100, p less than 0.05); the other parameters did not differ. NSAI drugs other than salicylates were taken more often for osteoarthritis than salicylates (33.6 p. 100 vs 17.4 p. 100, p less than 0.05).(ABSTRACT TRUNCATED AT 250 WORDS)