The review addressed a focused question supported by clearly defined inclusion criteria. The literature search was adequate for published studies, but specific attempts were not made to locate unpublished studies, so publication bias was a possibility. This was assessed by the review, but the methods used were not appropriate for diagnostic studies. Appropriate steps were taken to minimise bias and errors in the selection of studies, but it was unclear whether such steps were also taken in the extraction of data or assessment of study quality.
Study quality was assessed using appropriate criteria, but the results of this were not reported in detail or considered in the synthesis of results. Methods used to pool data were appropriate and based on the most robust available models. Based on the summary receiver operating characteristic plots there appeared to have been substantial heterogeneity in both sensitivity and specificity, but results for the statistical assessment of heterogeneity were not reported and heterogeneity was not investigated.
The authors conclusions were supported by the data presented, but should be interpreted with caution given the possibility of publication bias and failure to investigate heterogeneity.
Four of the authors disclosed financial links with pharmaceutical companies.