Utility of iron biomarkers in differentiating menopausal status: Findings from CoLaus and PREVEND

Abstract

Aim: To examine the association of iron biomarkers with menopausal status and assess whether these biomarkers can help differentiate menopausal status beyond age. Methods: In this cross-sectional study we included 1679 women from the CoLaus and 2133 from the PREVEND cohorts, with CoLaus used as primary cohort and PREVEND for replication. Ferritin, transferrin, iron, and transferrin saturation (TSAT) were used to assess iron status. Hepcidin and soluble transferrin receptor were assessed only in PREVEND. Menopausal status was self-reported and defined as menopausal or non-menopausal. Logistic regressions were used to explore the association of these iron biomarkers with menopause status. Sensitivity, specificity, area under the receiver operating characteristic curves (AUC), positive and negative predictive values as well as cut-off points for the iron biomarkers were calculated. The model with the highest AUC was defined as the best. Results: In the CoLaus and PREVEND cohorts, respectively, 513 (30.6 %) and 988 (46.3 %) women were postmenopausal. Ferritin (OR, 2.20; 95 % CI 1.72–2.90), transferrin (OR, 0.03; 95 % CI 0.01–0.10), and TSAT (OR, 1.28; 95 % CI 1.06–1.54) were significantly associated with menopausal status in CoLaus, with the findings replicated in PREVEND. AUC of age alone was 0.971. The best model resulted from combining age, ferritin, and transferrin, with an AUC of 0.976, and sensitivity and specificity of 87.1 % and 96.5 %, respectively. Adding transferrin and ferritin to a model with age improved menopause classification by up to 7.5 %. In PREVEND, a model with age and hepcidin outperformed a model with age, ferritin, and transferrin. Conclusion: Iron biomarkers were consistently associated with menopausal status in both cohorts, and modestly improved a model with age alone for differentiating menopause status. Our findings on hepcidin need replication.