Age-related differences in hormonal and nutritional impact on lean anorexia nervosa bone turnover uncoupling.

INTRODUCTION: In anorexia nervosa (AN) patients osteoporosis occurs within a framework of multiple hormonal abnormalities as a result of bone turnover uncoupling, with decreased bone formation and increased bone resorption. The aim of study was to evaluate the hormonal and nutritional relationships with both of these bone remodeling compartments and their eventual modifications with age. PATIENTS AND MEASUREMENTS: In a cohort of 115 AN patients (mean BMI:14.6 kg/m2) that included 60 mature adolescents (age: 15.5-20 years) and 55 adult women (age: 20-37 years) and in 28 age-matched controls (12 mature adolescents and 16 adults) we assessed: bone markers [serum osteocalcin, skeletal alkaline phosphatase (sALP), C-telopeptide of type I collagen (sCTX) and tartrate-resistant acid phosphatase type 5b (TRAP 5b)], nutritional markers [ body mass index (BMI, fat and lean mass), hormones (free tri-iodothyronine (T3), free T4, thyroid stimulating hormone (TSH), luteinizing hormone (LH), follicle stimulating hormone (FSH), 17 beta estradiol, free testosterone index (FTI), dehydroepiandrosterone (DHEAS), insulin-like growth factor 1 (IGF-1), growth hormone (GH) and cortisol], plasma methoxyamines (metanephrine and normetanephrine) and calcium metabolism parameters [parathyroid hormone (PTH), Ca, vitamin D3].
RESULTS: Osteocalcin reached similar low levels in both AN age subgroups. sCTX levels were found to be elevated in all AN subjects and higher in mature adolescents than in adult AN (11,567+/-895 vs. 8976+/-805 pmol/l, p<0.05). sALP was significantly lower only in mature adolescent AN patients, while there were no significant differences in the levels of TRAP 5b between AN patients and age-matched control groups. Osteocalcin correlated with sCTX in the control subjects (r=0.65) but not in the AN patients, suggesting the independent regulation of these markers in AN patients. Osteocalcin levels strongly correlated with freeT3, IGF-I, 17 beta estradiol and cortisol, while sCTX correlated with IGF-I, GH and cortisol in both age subgroups of the AN patients. Other hormones or nutritional parameters displayed age-related correlations with bone markers, leading to different stepwise regression models for each age interval. In mature adolescent AN patients, up to 54% of the osteocalcin variance was due to BMI, cortisol and 17 beta estradiol, while 54% of the sCTX variance was determined by GH. In adult subjects, freeT3 and IGF-I accounted for 64% of osteocalcin variance, while 65% of the sCTX variance was due to GH, FTI and methoxyamines.
CONCLUSIONS: We suggest a more complex mechanism of AN bone uncoupling that includes not only "classical" influence elements like cortisol, IGF-I, GH or 17 beta estradiol but also freeT3, catecholamines and a "direct" hormone-independent impact of denutrition. Continuous changes of these influences with age should be considered within the therapeutic approach to AN bone loss.