Tracking of bone mass from childhood to adolescence and factors that predict deviation from tracking

It has been hypothesized that bone density tracks but long term studies in children are lacking. As such, the aim of this study was to describe tracking of dual X-ray absorptiometry measures from age 8 to age 16 years, whether this was independent of change in body size and whether deviation from tracking could be predicted. 116 males and 67 females had anthropometric (height and weight), questionnaire (medication use, sports, breastfeeding), fitness (PWC(170)) and DXA measures (bone free lean mass [LM], fat mass [FM] and bone mass) at baseline and follow-up. BMC and aBMD were assessed at the spine and hip and total body and bone mineral apparent density (BMAD) at the spine and hip. We found all DXA measures tracked significantly after adjustment for change in height and change in weight (males: R(2): BMC 24-62%; aBMD 41-48%; BMAD 30-37%, females: R(2): BMC 45-72%; aBMD 36-56%; BMAD 30-48%). Factors that predicted subjects would deviate positively, that is improve in tertile or remain in the highest tertile of spine and hip aBMD included having been breastfed, increase in LM, PWC(170) at age 8 and sport participation in males. LM at age 8 was beneficial in males while in females; FM at age 8 predicted subjects would deviate positively. Boys who gained absolute and percent FM and girls who gained percent FM, were more likely to deviate negatively, that is, decrease in tertile or remain in the lowest tertile of spine and hip aBMD. ICS use at age 8 also predicted subjects, particularly males would not improve in bone mass relative to their peers. In conclusion, DXA measures track moderately to strongly from childhood to adolescence. This was independent of linear growth and sex indicating bone development and physical growth are under largely separate mechanistic control. Body composition was the main predictor of altered tracking but environmental factors also appear important.