The review question and inclusion criteria were clear. A limited number of sources were searched. Studies not written in English were excluded. The author acknowledged that some studies may have been missed for this reason. The potential impact of this limitation on the review findings was unclear. The author did not state how many reviewers selected the studies and extracted the data. Only one reviewer performed the quality assessment, but this was on two separate occasions.
Studies were generally scored as being high quality, but there were no details on which criteria were problematic. The choice of a narrative synthesis appeared appropriate, given the differences between the studies (notably in outcomes and follow-up duration). The author did not state what statistical methods were used to calculate effect sizes within the included studies, or whether he attempted to extract or calculate confidence intervals, which made the interpretation of the results difficult.
Studies were generally small so some may have been underpowered (as noted by the author). Follow-up duration was short, which limited the reliability of the results (notably side effects and toxicities).
The conclusions generally reflected the evidence. Due to the relatively small size and number of studies, imperfections in the conduct of the review, lack of clarity around the validity and limited comparability of the studies, the reliability of the conclusions is unclear.