Introduction 5-Hydroxytryptamine three receptor antagonists (5HT3RA) are effective in diarrhoea predominant irritable bowel syndrome (IBS-D), with a number needed to treat (NNT) for Alosetron of 7.1 Due to safety concerns (constipation [25%] and ischaemic colitis [0.1%]) Alosetron is not licensed in the UK. Ondansetron (OND) is a 5HT3 RA, widely and safely used for nausea, with constipation as a side effect.
Methods We recruited 125 patients meeting the Rome III criteria from primary and secondary care to a two centre randomised, double-blind, placebo-controlled, crossover trial. Screening comprised of blood tests, rectal biopsy and a 1-week baseline Bristol stool form diary before randomisation to placebo (PLA) or OND for 5 weeks. Efficacy was optimised by dose titration during weeks 1–3 starting at 4 mg daily, increasing (max 8 mg three time a day) or decreasing as required. A 2-week washout preceded 5 weeks of the opposite therapy to that received in treatment 1. Symptom diaries including stool form, frequency, pain, bloating and urgency were completed daily. Transit using the Metcalf method was measured after each treatment period.2 The primary endpoint was the difference in average stool form between baseline and the last 2 weeks of each treatment. Stool consistency “responders” were defined as experiencing >50% reduction in the days/week with stool form 6 or 7. Analysis is presented by intention to treat.
Results 80 women and 28 men completed the study (mean age 40.8, range 18–72, and 41.5, range 25–60). The mode dose of OND was <4 mg a day. Stool form improved significantly in the OND arm, mean change 1.4 (95% CI 1.2 to 1.6), vs 0.5 (95% CI 0.3 to 0.7) compared to PLA, p=<0.0001. Stool frequency improved on OND, mean 0.86 (95% CI 0.8 to 1.4), vs 0.44 (95% CI 0.4 to 0.89) on PLA, p=<0.01, as did urgency score 0.6 (95% CI 0.8 to 0.4) vs 0.3 (95% CI 0.5 to 0.2) p=<0.001. There were no significant improvements in pain or bloating. 74% of patients preferred OND while 26% preferred PLA or had no preference, χ2 p=<0.0001. 70% were “stool consistency responders” while taking OND compared with 33% taking PLA giving a NNT of 2.7. Whole gut transit slowed significantly while taking OND, 25 (13.5–47.5) h compared to 16 (7–29) h with PLA p=<0.001. 9% receiving OND complained of constipation compared to 2% on PLA, all but one responded to dose reduction alone. 2% withdrew because of constipation. There was no case of ischaemic colitis.
Conclusion Using dose titration Ondansetron acts to slow whole gut transit and is highly effective in IBS-D with a low incidence of constipation.
Competing interests None declared.
References 1. Cremonini F. Neurogastroenterol Motil 2003.
2. Metcalf A. Gastroenterology 1987.