Steroid reference intervals in women: influence of menopause, age and metabolism

Objective. To investigate the impact of age, obesity and metabolic parameters on thirteen circulating steroids in reproductive and menopausal age. To define reference intervals (RI). Design. Cross-sectional. Methods. 325 drug-free, healthy and eumenorrheic women were selected from the general population. Independent relationships of LC-MS/MS-determined steroid levels with age, body mass index (BMI) and metabolic parameters were estimated. Reference sub-cohorts were defined for calculating upper and lower limits in reproductive age, menstrual phases and menopause, and these were compared with limits in dysmetabolic sub-cohorts. Results. Lower androgens, pro-androgens and estrogens, but higher cortisol and metabolites were found in menopausal compared to reproductive age women. Androgens and precursors decreased during reproductive age (P<0.001–P=0.002) but not after menopause. 17OH-progesterone decreased with BMI (P=0.006) and glucocorticoids with waist circumference (P<0.001–P=0.002) in reproductive age, but increased with triglycerides (P=0.011-P=0.038) after menopause. Inverse associations of dihydrotestosterone with BMI (P=0.004) and HDL-cholesterol (P=0.010), estrone with total cholesterol (P=0.033) and estradiol with triglycerides (P=0.011) were found in reproductive age. After menopause, estrone increased with waist circumference (P<0.001) and decreased with insulin resistance (P=0.012). Ovarian steroid RI were estimated in menstrual phases and menopause. Age- and reproductive status-specific RI were generated for androgens, precursors and corticosteroids. Lower limits for reproductive age cortisol (P=0.020) and menopausal 11-deoxycortisol (P=0.003) in dysmetabolic sub-cohorts were reduced and increased, respectively, compared to reference limits. Conclusions: Obesity and dysmetabolism differently influence circulating steroids in reproductive and menopausal status. Age, menstrual and menopausal status-specific RI were provided by LC-MS/MS for a broad steroid panel.