Introduction Whether bone quality is essentially modifiable or tracks predictably up to adulthood is unclear. We aimed to identify a biological timing when interventions may have optimal impact on future bone properties.
Methods Participants were 709 girls from a cohort of adolescents born in 1990 and evaluated at 13 and 17 years-old. Evaluations consisted of structured questionnaires including gynaecologic age (GA) and physical evaluations comprising height, weight, body composition (bioelectric impedance) and bone mineral density (BMD) at the forearm (dual-energy x-ray absorptiometry). Our outcome was BMD variation during follow-up. We used linear regression to estimate its associations with baseline BMD and anthropometry, in four GA classes at baseline: ≤−1, 0, 1 and ≥2 years relative to menarche.
Results Mean (95% CI) annual BMD variation from 13 to 17 years-old was highest in the earliest GA group (0.030 (0.028; 0.032) g/cm2-year) and lowest in the oldest GA (0.018 (0.016; 0.019)). Pearson's correlation between baseline BMD and its increase varied from −0.12 in the youngest to −0.45 in the oldest GA. After adjustment (baseline BMD, weight, fat mass and height) and standardisation, BMD variation in the lowest GA was associated with baseline BMD (−2.70 (−5.29; −0.108) per SD), but essentially with baseline weight (11.4 (0.523; 22.2)) and fat mass (−10.0 (−19.2; −0.912)). In the highest GA, BMD variation was strongly determined by baseline BMD (−5.44 (−7.20; −3.69)) but not by anthropometric variables.
Conclusion Bone quality tracking should be measured relative to GA. Bone accrual seems prone to modification, especially in earlier gynaecologic ages.