Background : The incidence and severity of non-steroidal anti-inflammatory drugs (NSAIDs)-induced gastro-duodenal ulcer have not been extensively studied in Japan.

Aim : We performed a prospective study to clarify NSAIDs-induced gastro-duodenal injury, focusing especially on low-dose aspirin (L-A).

Methods : Two hundred and thirty-eight patients with bleeding peptic ulcers admitted to our hospital. History of taking NSAIDs and anti-ulcer drugs was obtained from all patients who underwent endoscopic examinations. The lesion scores of patients taking L-A were classified numerically from zero (no lesion) to five (ulcer).

Results : The NSAIDs were associated with 28.2% of hemorrhagic ulcers. The rates of patients using L-A, loxoprofen, diclofenac, and combination of two of these drugs were 27, 16, 10 and 9%, respectively. Co-administered anti-ulcer drugs were cytoprotective anti-ulcer drugs (27%), H2 receptor antagonists (16%), PPI (4%), and none (53%). In patients taking L-A, H2 receptor antagonists were used most frequently. The HP was positive in 63% of L-A-induced ulcer cases and in 69% of NSAIDs other than low-dose aspirin-induced ulcer cases. The lesion scores of patients taking L-A with H2 receptor antagonists or PPI were significantly lower than those of patients who were taking only L-A (P < 0.05).

Conclusions : Approximately one-third of hospitalized patients with NSAIDs-induced hemorrhagic ulcer showed an association with L-A. Prospective randomized controlled trials including H2 receptor antagonists are required to establish preventive efforts aimed at L-A-induced gastro-duodenal injury.