Noninvasive Diagnosis of Adenomyosis: A Structured Review and Meta-analysis of Diagnostic Accuracy in Imaging

OBJECTIVE: To conduct a systematic review and meta-analysis to evaluate imaging methods used to diagnose adenomyosis. DATA SOURCES: A thorough search was completed through the Cochrane Central Register of Controlled Trials, EMBASE, and PubMed or MEDLINE databases from January 2000 to June 2019. METHODS OF STUDY SELECTION: Following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines and the Quality Assessment of Diagnostic Accuracy Studies 2 tool, studies reporting the diagnostic accuracy of an imaging method and histopathology as a reference standard were selected and prospectively registered in the International Prospective Register of Systematic Reviews. Statistical analysis was performed using the R-package Meta-Analysis of Diagnostic Accuracy. TABULATION, INTEGRATION, AND RESULTS: Of the 1168 records identified, 10 studies were included in the meta-analysis, and these included 827 patients undergoing 2- or 3-dimensional transvaginal ultrasound (TVUS) and 317 patients undergoing magnetic resonance imaging (MRI). The risk of bias was low or unclear for all Quality Assessment of Diagnostic Accuracy Studies domains. The overall prevalence of adenomyosis was 40%. Overall MRI, 2-dimensional TVUS, 3-dimensional TVUS, and TVUS had the following aggregated diagnostic qualities (95% confidence interval): sensitivity of 78% (70%-84%), 74% (68%-79%), 84% (77%-89%), 78% (73%-82%); specificity of 88% (83%-92%), 76% (71%-79%), 84% (77%-89%), 78% (74%-81%); positive likelihood ratio of 6.8 (4.5%-10%), 3 (2.5%-3.7%), 5.2 (3.6%-7.4%), 3.5 (3%-4.2%); negative likelihood ratio of 0.25 (0.18%-0.35%), 0.34 (0.27%-0.43%), 0.19 (0.13%-0.28%), 0.28 (0.23%-0.34%); and pooled area under the operator curve of 0.77, 0.7, 0.83, 0.73, respectively. The pooled area under the operator curve for all modalities was not significantly different (all: p ≥.4). CONCLUSION: As a result of the systemic review and meta-analysis, we identified TVUS and MRI as good and comparable noninvasive imaging methods for diagnosing adenomyosis, leading us to recommend TVUS as the first-line diagnostic imaging method and MRI as a second-line method, if TVUS is inconclusive.