Background: The efficacy of antidepressants in irritable bowel syndrome (IBS) is controversial. No trials have directly compared a tricyclic with a selective serotonin reuptake inhibitor. Our aim was to determine whether imipramine and citalopram are efficacious in IBS. Methods: This was a randomized, double-blind, placebo-controlled, parallel group pilot trial with imipramine (50 mg) and citalopram (40 mg). Results: Of 51 IBS patients randomized, baseline characteristics were comparable among the treatment arms; the majority was diarrhea-predominant. Adequate relief of IBS symptoms (primary endpoint) was similar for each treatment arm. Improvements in bowel symptom severity rating for interference (P = 0.05) and distress (P = 0.02) were greater with imipramine versus placebo, but improvements in abdominal pain were not. There was a greater improvement in depression score (P = 0.08) and in the SF-36 Mental Component Score (P = 0.07), with imipramine. Citalopram was not superior to placebo. Approximately 20% of the variance in scores was explained by treatment differences for abdominal pain, bowel symptom severity disability, depression and the mental component of the SF-36. Conclusion: Neither imipramine nor citalopram significantly improved global IBS endpoints over placebo.