Updated meta‐analysis on the diagnostic accuracy of serum anti‐Mullerian hormone in poly cystic ovary syndrome involving 13 509 subjects. Issue 8 (8th April 2022)

• Record Type: Journal Article
• Authors: Anand, Shikha; Kumar, Amit; Prasad, Anupa; Trivedi, Kiran
• Is Part Of: Journal of obstetrics and gynaecology research. Volume 48:Issue 8(2022)
• Page Start: 2162
• Page End: 2174
• Publication Date: 2022-04-08
• Subjects: AMH -- hyperandrogenism -- PCOS -- polycystic ovarian morphology (PCOM) -- Rotterdam criteria; Gynecology -- Periodicals; Obstetrics -- Periodicals; 618.1005
• DOI: 10.1111/jog.15233
• Languages: English
• ISSNs: 1341-8076

Abstract

Abstract: Aim: To determine the diagnostic accuracy of anti‐Mullerian hormone (AMH) in the diagnosis of polycystic ovary syndrome (PCOS). Methods: Two independent reviewers searched the electronic databases and search engines using PubMed, Cochrane library, and Google Scholar systematically to retrieve relevant articles published from inception to September 2021. The diagnostic efficacy of AMH was computed using the random‐effects model in terms of pooled sensitivity, specificity, and diagnostic odds ratio (DOR) with 95% confidence interval (CI). A meta‐regression and subgroup analysis were performed to check for any source which could explain possible heterogeneity. Risk of bias assessment was conducted using the QUADAS‐2 tool recommended by Cochrane Library. Results: This meta‐analysis included a total of 41 studies involving 13 509 subjects. We observed promising pooled sensitivity 0.78 (95% CI 0.74 to 0.81), specificity 0.87 (95% CI 0.84 to 0.90), and diagnostic odds ratio (DOR) 24 (95% CI 15 to 37), for AMH in detecting PCOS and discriminatory power (summary receiver operating characteristic [SROC] curves, 0.89 [95% CI 0.86–0.92]). The most prominent bias was noted in the patient selection and index test assessment. Conclusions: With the findings of this current meta‐analysis, we conclude serum AMH to be a promising biomarker for the diagnosis of PCOS, however, substantial heterogeneity among studies needs individual patient data analysis in order to identify an optimal cut‐off value and homogenous findings. Registration Number and Guidelines: This meta‐analysis was performed according to constructed protocol registered in the PROSPERO database with registration number CRD42021246910.