Introduction: The gastro-intestinal (GI) toxicity associated with high dose aspirin has been fully demonstrated, but remains poorly elucidated at low doses i.e., less than 500 mg/day. Such toxicity is relatively difficult to study because lesional and/or bleeding GI complications are not always well described in studies. The objective of this review is to compile a documented inventory of GI complications induced by low-dose aspirin.
Methodology: This review is based on a detailed review of randomized studies, case-control and cohort studies which aim to study the GI toxicity of low-dose aspirin. These studies have been selected based on specific criteria from works published from 1983 to 2003 (PubMed).
Results: In 8 randomized, placebo-controlled or group-controlled studies, erosions, in particular gastric erosions, are more frequent in the elderly than in younger subjects. On the other hand, the prevalence of gastric and duodenal ulcers does not appear to be significantly increased except in specific cases. The most significant complications are GI bleeding described only in 3% of cases. Among randomized studies which have evaluated the cardiovascular and neurological usefulness of low-dose aspirin, 16 have evaluated GI safety and tolerability as a secondary end-point. These studies demonstrated a non-significant increase in esophageal, gastric and duodenal ulcers during treatment with low-dose aspirin (Odds Ratio = 1.22, P = 0.08), but a significant increase in bleeding ulcers (Odds Ratio = 1.77, P = 0.04). These complications seem to occur in rare cases (less than 3% of patients) and often seem minor. Nevertheless, ulcers have been less studied than bleeding because few endoscopic analyses have been performed in these studies. The gastro-duodenal bleeding related to erosions or ulcers are significantly more frequent with Odds Ratios between 1.3 and 3.3 (P < 0.05). There are 8 case-controlled studies and 1 cohort follow-up study which confirm the increased GI risk with low-dose aspirin. GI bleeding adverse effects related mainly to gastro-duodenal ulcers are more frequent in case of the regular use of aspirin.
Conclusion: The GI risk exists, starting with the lowest doses and appears to be dose-dependent. The lesional complications consist mainly of erosive lesions, most often gastric, and rarely true ulcers. Cases of bleeding appear more frequent, but generally are minor. This risk should be taken into account by the prescribing physician and the patient should be informed when treatment with low-dose aspirin is initiated.