Serum follicle stimulating hormone to oestradiol ratio as a biomarker for the menopausal transition: a secondary analysis on women in the Study of Women's Health Across the Nation

Abstract

Context: Previous studies have recognised that serum follicle stimulating hormone (FSH) and oestradiol (E2) levels show abrupt rise and fall respectively in the late menopausal transition. The ratio of the two parameters may be a better biomarker for the menopausal transition by combining the diagnostic power of the two, which has not been studied before. Objective: To evaluate the diagnostic performance of the FSH:E2 ratio in discriminating menopausal from premenopausal women. Methods: We carried out a retrospective secondary analysis on the Public Use Datasets of the Study of Women's Health Across the Nation (SWAN). Participants were followed up annually in the study, and publicly accessible data till visit 10 were retrieved. The trajectories of FSH (in IU/L), E2 (in pg/ml) and FSH:E2 ratio over time (in years) with respect to the final menstrual period (FMP) were observed. Blood was taken in the early follicular phase in menstruating women. For each participant, the visit preceding the one categorised as “postmenopausal” (defined by more than 12 months of amenorrhoea) was taken as the FMP visit. The dataset from Visit 5 was used as the reference visit to compare the diagnostic performance of FSH:E2 and FSH alone by the receiver-operator characteristic (ROC) curve. At this visit, women at FMP and beyond were taken as “menopausal”, while those at FMP -1 year or before were taken as “premenopausal”. Patient(s): A total of 930 women who participated since Visit 1 of the SWAN who had not used hormonal medication within one month before all study visits were studied. Intervention(s): No active intervention was involved in this secondary analysis. Main Outcome Measure(s): The area under the ROC curve for discriminating menopausal from premenopausal women, and the suggested cut-offs. Result(s): Serum FSH and the FSH:E2 ratio showed an abrupt rise, and E2 showed an abrupt decline, at FMP -1 year, and plateaued after FMP +1 year. FSH:E2 ratio had significantly higher area under the ROC curve (0.810, 95% CI 0.784 - 0.835) than FSH alone (0.783, 95% CI 0.755 - 0.809) (p< 0.001) in discriminating menopausal from premenopausal women. A fixed specificity of 80% could be achieved by the FSH:E2 cut-off at 4.7, with an associated sensitivity of 62%, positive likelihood ratio of 3.12 and negative likelihood ratio of 0.48. Conclusions: FSH:E2 ratio may be a biomarker for diagnosis of menopause. Its clinical use needs to be validated in further studies.