Previous use of non-steroidal anti-inflammatory drugs and anticoagulants: the influence on clinical outcome of bleeding gastroduodenal ulcers.

OBJECTIVE

To evaluate the relationship between prior non-steroidal anti-inflammatory drug (NSAID) or anticoagulant use and clinical outcome in bleeding gastric and duodenal ulcer patients.

DESIGN

Prospective cohort-study.

PARTICIPANTS

All patients (n = 132) admitted because of upper gastrointestinal bleeding during 3 months in the Amsterdam area.

METHODS

We compared clinical outcome (blood transfusion, rebleeding, surgery and mortality) between ulcer patients who used NSAIDs or anticoagulants and patients who did not use these drugs before the bleeding-episode.

RESULTS

Of the 132 patients admitted, 56 patients had gastric or duodenal ulcers. NSAIDs were used significantly more often before the bleeding episode in these ulcer patients than in the non-ulcer patients (n = 76), 21/56 (37.5%) vs. 15/76 (19.7%), respectively (P < 0.05), relative risk = 2.57, 95% confidence interval: 1.04-5.77). Stigmata of recent haemorrhage were found in 16/21 (76.2%) patients in the NSAID ulcer group, in 2/9 (22.2%) in the coumarin-ulcer patients, and in 12/24 (50%) in the no-medication ulcer group (not significant). Prior NSAID usage increased the in-hospital rebleeding rate from 16.7% to 42.9% (P = 0.05), leading to an increased need for surgical intervention from 16.7% to 42.9% (P = 0.05). In contrast prior usage of anticoagulants, which could be antagonized, did not affect the clinical outcome of the bleeding. Mortality was 9.5% in the NSAID group, 0% in the coumarin group, and 4.2% in the no-medication group.

CONCLUSION

Prior use of NSAIDs increases the risk of rebleeding in bleeding ulcer patients, and leads to a higher need for urgent surgery. In contrast, prior anticoagulant therapy does not raise the rebleeding risk.