Diagnostic accuracy of transvaginal ultrasound and magnetic resonance imaging for the diagnosis of adenomyosis: Systematic review and meta-analysis

Abstract

Objective The clinical diagnosis of adenomyosis can be challenging due to the signs and symptoms of being nonspecific and often mimic those of other uterine pathology such as fibroids and endometriosis. It is therefore paramount that a diagnosis be achievable based on imaging, particularly in symptomatic women who wish to opt for conservative management. The aim of this study is to systematically review the current literature and use meta-analysis techniques to determine the accuracy of transvaginal ultrasound (TVS) and magnetic resonance imaging (MRI) in diagnosing adenomyosis. Methods An extensive search was carried out in PubMed and EMBASE for articles from inception to February 2019, for studies that compared TVS and/or MRI in the diagnosis of adenomyosis with histopathology evidence of adenomyosis via hysterectomy as the reference standard. Studies were eligible if the TVS or MRI was performed within one year of hysterectomy. Quality of the studies was assessed using the QUADAS-2 tool for risk of bias and assessment of applicability. Results The search yielded 664 citations, in which 31 studies were considered eligible. 25 studies were used to calculate the accuracy of TVS and 8 studies were used to calculate MRI, 3 studies included both TVS and MRI. The pooled sensitivity for TVS was 0.79, specificity was 0.83, LR+ was 4.63, LR- was 0.25. For MRI, the pooled sensitivity was 0.78, specificity was 0.93, LR+ was 11.98, LR- was 0.23. In accordance to the QUADAS-2 tool, the quality of the studies included was considered good in most domains. Conclusion The diagnostic performance of TVS is similar and comparable to the performance of MRI in diagnosing adenomyosis.