[Critical reading of systematic reviews and meta-analyses about diagnostic imaging]

Systematic reviews of diagnostic validity have been proposed as the best methodological tool to integrate all the available evidence and to help physicians decide whether to use a given diagnostic test. These studies aim to synthesize the results obtained in different primary studies into a couple of indices, generally sensitivity and specificity, or into a summary receiver operating characteristic (ROC) curve. Although there is a certain parallelism with reviews about the efficacy of therapeutic interventions, reviews of diagnostic validity have certain peculiarities that add complexity to the analysis and interpretation of the results. This article emphasizes the methodological aspects that make it possible to critically assess the extent to which the results of a review of the validity of diagnostic tests are valid and provides rudimentary knowledge of the statistics necessary to understand the results.