Hirudin is a selective thrombin inhibitor with strong anticoagulant properties which could elicit gastro-intestinal bleeding. A double-blind cross-over study of the effects of hirudin on gastro-intestinal bleeding was therefore conducted on 12 healthy, consenting males. After labelling erythrocytes with 51Cr and returning them intravenously, stools were collected for 2 days to measure radioactivity and hence baseline faecal blood loss. After injection of hirudin or placebo stools were collected for 3 days. Partial thromboplastin time was measured sequentially after medication with hirudin or placebo. This procedure was repeated after injection of the alternate medication 1 week later. Hirudin was tolerated well. Mean faecal blood loss associated with hirudin was slightly higher than with placebo (1.63 ml vs 1.15 ml over 3 days; 95% confidence interval for the difference between hirudin and placebo was -0.68 to 1.63) but these differences are clinically irrelevant. After hirudin injection PTT was elevated to about twice the baseline values but returned to baseline within 12 h after the last hirudin injection.