The review question and selection criteria were clear. Several bibliographic sources were searched. Only studies published in English were included so it was possible that some studies were missed. Appropriate measures were taken to minimise error and bias during study selection, data extraction and quality assessment.
Reporting of design characteristics was limited and this made the quality of the studies unclear. Trials were limited in number and size. Reporting of analysis methods had several limitations. The authors did not clearly specify how odds ratios were calculated. Given the nature of the primary outcome (which appeared to be continuous), use of odds ratios appeared inappropriate. The authors did not report the findings clearly and this made interpretation difficult. Given the variation in intervention techniques, patient characteristics and outcome measurement methods, a narrative synthesis may have been more appropriate. Very high levels of heterogeneity made the applicability of the review findings unclear.
The authors recommended that their findings be interpreted with caution due to variation between the studies. Poor reporting and unclear methods of analysis made it difficult to know whether the conclusions reflected the evidence and the review may not be reliable.