Weight gain reverses bone turnover and restores circadian variation of bone resorption in anorexic patients.

Abstract

The present study was conducted in order to describe the variations and circadian rhythm of biochemical markers of bone remodelling at baseline and after weight gain in patients with anorexia nervosa (AN). We studied 9 women (mean age 21 years, range: 16-30) with established AN who remained amenorrhoeic during the study and with a low body mass index (BMI) after refeeding and 6 female controls (mean age 20 years, range, 18-24 and BMI: 20.6 +/- 1.1 kg/m2). Refeeding was not associated with any other intervention or treatment, especially oestrogen replacement or hormonal contraception. Serum levels of oestradiol remained below 70 pmol/l before and after refeeding. During the study, PTH and 25-hydroxyvitamin D measurements were performed. Markers of bone formation: serum intact osteocalcin (iBGP) and serum intact BGP + fragments (iBGP+F) and markers of bone resorption: urine C-teloptide of type I collagen (uCTX) and serum C-telopeptide ofvtype 1 collagen (s-CTX) were measured. At baseline, PTH and 25 OH-vitamin D concentrations were within the normal range in AN patients and no significant variation was observed after refeeding. Bone formation markers were found to be significantly different at baseline between AN patients and controls. After refeeding, iBGP and iBGP+F levels increased by 172% and 154%, respectively, to values no different from controls. Intact BGP and iBGP+F exhibited a significant circadian variation in controls (P < 0.05 and P < 0.002, respectively), whereas we did not find any such circadian rhythm in AN patients. After refeeding no significant circadian variation was observed; however, iGBP+F tended to peak in early morning and exhibited a nadir in the afternoon. At baseline, sCTX was 2-fold higher in AN patients than in controls. After weight gain sCTX decreased significantly and reached control values. Refeeding induced a non-significant 40% decrease in uCTX. We found positive correlations between uCTX and the 24-h mean value of sCTX levels (r2 = 0.93, P < 0.0001) and between uCTX and the mean value of sCTX peak levels at 0800 h (r2 = 0.65, P < 0.0003). Serum CTX exhibited a significant circadian variation in controls (P < 0.001) with a peak at 0800 h and a nadir at 1600 h with a 60% decrease between peak and nadir values. We found that anorexia nervosa suppressed the sCTX circadian variation which was restored by refeeding. We found a significant non-linear relationship between BMI and sCTX/iBGP ratio in AN (r2 = 0.6, P < 0.0001), thus illustrating the influence of nutritional status on bone remodelling. In this study we found that weight gain, related to refeeding only, reversed the anorexia nervosa-induced uncoupling of bone remodelling and restored circadian variation of a bone resorption marker.