Background: There is growing concern about the impact of eating disorders on the bone heath during adolescence where peak bone mass acquisition is of paramount importance.
Method: A total of 85 patients (77F/8M, 75% anorexia nervosa(AN) and 25% atypical eating disorder), median age 15.2 years (range, 10.919.8) and median BMI SDS −1.3 (−4.7 to 0.8) attended the bone densitometry service between Jan 2009 and Dec 2013 for total body (TB) and lumbar spine (LS) DXA scans. 13 patients (11F/2M, all AN) had a follow-up scan at an interval of 1.4 years (0.82.5). Bone size is reported as percent predicted bone area for age (ppBA-for-age), and for size adjustment, the extent of bone mineralisation within the bone is described as percent predicted bone mineral content for bone area (ppBMC-for-BA).
Results: Median ppBA-for-age was 89% (66124) at TB site, with 54% of patients presenting with ppBA-for-age ≤90%. Median ppBMC-for-BA was 99% (89116) at TB site, with only 5% being ≤90%. At LS site, median ppBA-for-age was 96% (65128), while median ppBMC-for-BA was 94% (73131). TB ppBA-for-age and LS ppBA-for-age correlated positively with BMI SDS(r=0.44, P<0.0001 and r=0.213, P=0.05). Median DXA software derived paediatric analysis centiles were 38th (091) for height age (shorter bones=less linear growth), 21st (078) for bone area for height (thinner bones=less periosteal expansion), 20th (095th) for lean mass for height (reduced muscle mass), and 52nd (1100th) for BMC for lean mass centile (adequate BMC for reduced lean mass). There was no significant difference between baseline and follow-up in ppBA-for-age or ppBMC-for-BA at TB or LS.
Conclusion: In young adults with an eating disorder, bones are small compared to the normal healthy population, but do not show reduced bone mineralisation. Muscle mass is low, but BMC appears preserved at the expense of a combination of periosteal expansion and linear growth. Long-term follow-up in these patients is required.
20 - 22 Sep 2014
European Society for Paediatric Endocrinology