Endocrine Abstracts

Background: Steroid profiling by LC-MS/MS is increasingly used in endocrine research, and is being introduced in the clinical practice. Due to the large disagreement with previous immunoassays, and to the enlarged panel of measurable steroids, novel and robust reference intervals (RIs) need to be established. Steroid circulating levels are known to be influenced by several physio-pathological factors, such as age, gender, stress and metabolic status. Therefore, generating unbiased reference cohorts is a non-trivial task.

Objective. Aiming at estimating robust and age-specific RIs for a broad steroid panel, our study evaluated the impact of aging and of the metabolic status on steroid levels in a cohort of disease-free adult men.

Methods. After excluding previous or current diseases, 318 drug-free men aged 18-90y were selected from the Italian general population. Blood was withdrawn in fasting conditions at 8-10am after 10min saline infusion. Cortisol (F) 11-deoxycortisol (11S), 17-hydroxyprogesterone (17OHP4), 17-hydroxypregnenolone (17OHP5), testosterone (T), androstenedione (A4), dheidroepiandrosterone (DHEA), dihydrotestosterone (DHT), progesterone (P4), corticosterone (B), estrone (E1) and estradiol (E2),were measured by two validated in-house LC-MS/MS assays. Anthropometric and metabolic parameters were analyzed by multiple stepwise-regression to estimate their independent impact of on each steroid. For steroids varying with ageing, decade-specific 2.5-97.5 centile RIs were calculated by fractional polynomial regression. For steroids influenced by metabolic parameters, we excluded subjects displaying BMI≥25.0kg/m², waist circumference>88 cm, systolic/diastolic blood pressure≥90/140mmHg, HOMA-IR>2.5 or total-cholesterol/HDL-cholesterol>5.

Results. Aging negatively impacted 17OHP4, 17OHP5, A4, DHEA, P4 and B (all P<0.001), F (P=0.015) and 11S (P=0.006), while not influencing T, DHT, E1 and E2 levels. BMI positively influenced E2 (P=0.031), while negatively associating with 17OHP4 (P=0.002), T (P=0.045), DHT (P=0.003) and P4 (P=0.023). Waist circumference directly associated with E1 (P=0.012), while inversely associated with F (P=0.004). HOMA-IR negatively associated with T (P=0.002). Blood pressure positively associated with E1 (P=0.034). Finally, total-cholesterol/HDL-cholesterol ratio negatively associated with B (P=0.001) and 17OHP5 (P=0.024). According to regression results, age-specific RIs were generated on the whole cohort for 11S, A4 and DHEA, and after excluding subjects displaying metabolic abnormalities for F, B, 17OHP4, 17OHP5 and P4. At variance, age-independent RIs were estimated for T, DHT, E1 and E2 in the sub-cohort with no metabolic abnormalities. This study provided LC-MS/MS-based, age and metabolism-adjusted male RIs for a panel of circulating steroids, allowing the effective application of this technique in clinical investigations and in the diagnostic workup of diseases related to steroidogenesis imbalances.