Accuracy of imaging modalities for adnexal torsion: a systematic review and meta-analysis.

Abstract

Adnexal torsion (AT), a serious gynaecological emergency, often presents with non-specific symptoms leading to delayed diagnosis. To compare the test accuracy of ultrasound, computed tomography (CT) and magnetic resonance imaging (MRI) to diagnose AT. We searched EMBASE, MEDLINE and Cochrane CENTRAL until December 2019. Studies reporting on the accuracy of any imaging modality (Index Test) in female patients (paediatric and adult) suspected of AT compared with surgical diagnosis and/or standard clinical/radiological follow-up period until resolution of symptoms (Reference Standard). We assessed study quality using QUADAS-2. We conducted test accuracy meta-analysis using a univariate model or a hierarchical model. We screened 3836 citations, included 18 studies (1654 women, 665 cases), and included 15 in the meta-analyses. Ultrasound pooled sensitivity (n = 12, 1187 women) was 0.79 (95% CI 0.63-0.92) and specificity was 0.76 (95% CI 0.54-0.93), with negative and positive likelihood ratios of 0.29 (95% CI 0.13-0.66) and 4.35 (95% CI 2.03-9.32), respectively. Using Doppler with ultrasound (n = 7, 845 women) yielded similar sensitivity (0.80, 95% CI 0.67-0.93) and specificity (0.88, 95% CI 0.72-1.00). For MRI (n = 3, 99 women), the pooled sensitivity was 0.81 (95% CI 0.63-0.91) and specificity was 0.91 (95% CI 0.80-0.96). A meta-analysis for CT was not possible with two case-control studies and one cohort study (n = 3, 232 women). Its sensitivity range was 0.74-0.95 and specificity was 0.80-0.90. Ultrasound has good performance as a first-line diagnostic test for suspected AT. Magnetic resonance imaging could offer improved specificity to investigate complex ovarian morphology, but more evidence is needed. To investigate adnexal torsion, ultrasound is a good first-line diagnostic test with a pooled sensitivity of 0.79 and specificity of 0.76.