2022-RA-1389-ESGO Diagnostic accuracy of ultrasound O-RADS for classifying adnexal mass: systematic review and meta-analysis

Introduction/BackgroundThe objetive is evaluate the diagnostic accuracy of ultrasound O-RADS classification for discriminating bening from malignant adnexal masses.MethodologyA search was performed in PubMed/MEDLINE, CINAHL, Scopus, Cochrane, ClinicalTrials.gov and Web of Science databases (January 2018 to January 2022) for studies evaluating ultrasound O-RADS classification (index test) for discriminating benign from malignant adnexal masses, using histology or adequate follow-up as reference test. The Quality Assessment of Diagnostic Accuracy Studies-2 (QUADAS-2) tool was used to evaluate the quality of the studies. Pooled sensitivity, specificity, positive and negative likelihood ratios for detecting adnexal malignancy were calculated separately considering O-RADS groups 4 and 5 as suspicious for malignancyResultsThe search identified 185 citations after excluding duplicates, papers not related to the topic, reviews and paper assessing MRI O-RADS and papers with no 2x2 tables available, six studies comprising 3063 adnexal masses in 3006 patients were ultimately included in the qualitative and quantitative syntheses. The mean prevalence of adnexal malignancy on surgery was 29% (range: (8% to 59%). All studies were retrospective and four of them were considered of high risk of bias in patient selection due to inadequate exclusions. All studies were considered as low risk of bias for index test, reference test and flow and timing. Overall, pooled sensitivity, specificity, positive and negative likelihood ratios and DOR for O-RADS classification were 97% (95%CI 93%-99%), 76% (95%CI 58%-87%), 4.0 (95%CI 2.2–7.3), 0.04 (95%CI 0.02- 0.09) and 100 (95%CI 35–280). Heterogeneity as high for specificity and moderate for sensitivity. Meta- regression showed that neither sample size nor malignancy prevalence explained this heterogeneity.ConclusionUltrasound O-RADS classification offers a high sensitivity and moderate specificity for discriminating malignant from benign adnexal masses.