The review addressed a clear research question and inclusion criteria appear appropriate. A number of relevant sources were searched for studies using appropriate search terms. Only published in the English language articles were sought, so language bias could not be ruled out. The search strategy did not include methods for identifying unpublished studies; formal methods were used to assess the likelihood of publication bias, which could not be ruled out. Appropriate methods to minimise error and bias were used to select studies and extract data.
An appropriate tool and method were used for quality assessment of the four included trials, which were of high quality. The L'Abbe plot included all four trials and demonstrated asymmetry. Participants in the included trials were mostly diagnosed by formal criteria (Rome and Manning), but a small proportion were diagnosed on the basis of bloating symptoms. The authors acknowledged that four studies were selected on the basis of the inclusion criteria, but only two studies had usable data on the key outcome (clinical response); these two studies, with only short-term data, were pooled in meta-analyses. Two meta-analyses were performed, one of all patients and another of patients who were exclusively diagnosed according to Rome criteria. Results from the two trials not included in the meta-analyses were not discussed in the results section of the publication, but referred to only in the discussion section. The authors acknowledged that, due to publication bias, their results may be an overestimate of the true effect.
The authors' cautious conclusions are appropriate and reflect the lack of available evidence.