The review question and the initial inclusion criteria were clear. The authors searched three relevant databases and other sources. A systematic search for grey literature and a lack of language restrictions reduced risks of missed relevant studies and presence of language and publication biases. The authors reported using methods designed to reduce reviewer bias and error in the selection of studies and assessment of validity, but not in data extraction.
The validity assessment used appropriate criteria, but it was unclear what definition was used for good-quality studies eligible for inclusion in the review. The decision to use meta-analyses appeared appropriate. The assessment of statistical heterogeneity used reasonable tests, but with non-standard thresholds of significance.
The authors' conclusions and recommendations for practice reflected the significant difference in the composite outcome of total complications. However, it should be borne in mind that the included studies comprised a relatively small number of patients and that none of the individual outcomes showed a statistically significant difference between the groups.