|Home | About | Journals | Submit | Contact Us | Français|
We read with great interest the article by Tack et al on the effect of the selective serotonin reuptake inhibitor (SSRI) citalopram on symptoms in patients with irritable bowel syndrome (IBS) (Gut 2006;55:1095–103). The usefulness of the results of this study are however debatable. Several previous studies have investigated the effect of tricyclic antidepressants and SSRIs on functional gastrointestinal symptoms. Because of errors or lack of clarity in study design, inclusion of very selected patient populations and, above all, small sample sizes, their role in the treatment of patients with IBS in daily clinical practice remains unclear.
The study of Tack et al, as already correctly pointed out by Creed in his commentary (Gut 2006;55:1065–7), also suffers from major shortcomings in study design, poor description of study population and no information on whether or not subjects and physicians/investigators were blinded and, if yes, how.
Nevertheless, Creed claims that this study provides useful information on the effect of citalopram on the primary outcome measure—number of days per week with abdominal pain. What is relevant for patients as well as physicians is the risk of reduction in the number of days with abdominal pain after citalopram treatment. From the results of this study a relative risk of abdominal pain can be calculated: using data from the parallel group only (the first treatment episode), patients that used citalopram reported 3.7/7=53% of the week with abdominal pain compared with 5.2/7=74% of the week in patients receiving placebo. This results in a relative risk of 0.72 (95% confidence interval 0.58–0.89) for abdominal pain when using citalopram compared with placebo. This sounds very promising. However, when performing a post hoc power analysis for this study, with alpha being 0.05 and a minimally appropriate power of 80%, each treatment group should have consisted of at least 82 subjects. This means that this study was heavily underpowered and results should therefore be interpreted as for a pilot study.
Considering that there are already many studies available investigating the potential benefits of antidepressants in small samples of patients with IBS, this study does not contribute to the ongoing discussion about the role of antidepressants in the treatment of patients with IBS in daily clinical practice. There is still a need for a large, well defined, randomised, double blind, placebo controlled clinical trial to investigate the true effect of an antidepressant on symptoms in patients with IBS.
Competing Interests: None declared.