Antibiotic treatment of constipation-predominant irritable bowel syndrome: the puzzle is yet to be solved.

Abstract

To the Editor, We read with keen interest the study by Pimental et al. [1] regarding the antibiotic treatment of constipation-predominant irritable bowel syndrome (IBS). We congratulate the authors for a well-conducted, multicenter double-blind randomized placebo-controlled trial that evaluated the effectiveness of neomycin plus placebo versus neomycin plus rifaximin in improving symptoms in methane-positive constipation-predominant IBS (C-IBS) subjects. The authors found that combination of neomycin and rifaximin was better to neomycin alone in improving severity of constipation as well as multiple C-IBS symptoms. They also found that in the neomycin plus rifaximin group, the improvement in constipation was more in those subjects who had successful reduction of their breath methane level below 3 parts per million (ppm). This study is clearly better designed than the previous publications on the contentious issue of use of antibiotics in treatment of C-IBS [2, 3]. This publication is truly a landmark one and has potential implications in the management of C-IBS. We have several pertinent comments and few questions that merit to be addressed to the authors of this study. First and foremost is the small number of recruited subjects. Total number of study subjects was only 31: 16 in the neomycin alone versus 15 in the neomycin plus rifaximin group. It is quite intriguing to notice this small sample size in spite of the fact that the study duration was 4 years, and it involved three tertiary care centers. The sample size was much lower than the originally planned number of 88, hence compromising on the power of the study, which in turn affects the reliability of the results. Small sample size also makes it more appropriate to use nonparametric tests for statistical analysis instead of parametric ones. It is also of interest to know whether colonic transit time evaluation was performed prior to inclusion of subjects in either of the two groups. The second issue is regarding post-treatment follow-up period which is much shorter than that in TARGET 1 and TARGET 2 studies [4] which addressed the issue of role of antibiotic therapy in patients with non-constipation IBS. Hence, the key question is whether the beneficial effect of antibiotics in C-IBS is durable or not. Recurrence of symptoms post-antibiotic therapy in IBS subjects remains as an elusive issue [5]. The third issue is regarding the parameters used to evaluate the efficacy of antibiotics in this study. Addition of Bristol stool form score and stool frequency would surely have added to the robustness of the results. Similarly, additional evaluation of global symptom score and quality of life assessment would have been desirable. It would have been interesting to know the exact severity scores of constipation, bloating and straining at the end of follow-up period as the graphs indicate that the difference between the scores of the two study groups gradually declined, particularly for severity of bloating. The fourth issue is regarding lack of any association between breath methane reduction and clinical response to antibiotic therapy. Number of subjects who achieved breath methane level of B3 ppm following antibiotic treatment was quite similar in the neomycin alone group [11 out of 16 (68.7 %)] as compared to neomycin plus rifaximin group [10 out of 15 (66.7 %)]. Moreover, the quantitative reduction in breath methane was not statistically different between the two study groups. These results put the puzzle back in the perspective: Is methane the sole or predominant S. Sachdeva (&) R. Kondala A. S. Dahale A. S. Puri Department of Gastroenterology, GB Pant Hospital, JL Nehru Marg, New Delhi 110002, India e-mail: sanjeevgastro@rediffmail.com