Dual-energy x-ray absorptiometry derived body composition trajectories across adulthood: reference values and associations with body roundness index and body mass index

Background Population-specific reference values are needed to accurately contextualise age-related changes in body composition. This study aimed to a) establish age- and sex-specific reference values and cut-points for a range of dual-energy x-ray absorptiometry (DXA) derived metrics of lean mass (LM), fat mass (FM) and bone mineral density (BMD), across adulthood in a large adult cohort; and b) determine the association between DXA-derived body composition, body roundness index (BRI), and body mass index (BMI). Methods Cross-sectional data were collected from 10033 men and women aged from 18-92 years. Whole-body DXA scans were performed, and a range of metrics were calculated for LM (total LM, arm LM, leg LM, appendicular lean mass: ALM, skeletal muscle index: SMI), FM (total FM: kg and %, FMI, android to gynoid: A/G ratio) and bone (BMD). Cut-points equivalent to Z-scores of 1.0 - 2.5 SDs from the mean of a young reference population were established for each body composition metric. Results Detailed age- and sex-specific percentile curves were generated using the LMS method. Metrics of LM, central adiposity and BMD were higher in men, compared to women, whereas metrics of general FM accumulation were higher in women, compared to men. In both sexes, all LM metrics remained broadly stable during early and middle adulthood, after which progressively lower quantities were shown, whereas progressively higher FM metrics were shown from early adulthood through to late adulthood. In men, BMD was broadly stable across adulthood, whereas in women, markedly lower BMD was observed from the fifth decade of life. Significantly higher quantities of LM were shown across BMI categories, but not across BRI categories. The BRI was better correlated with FM%, FMI, and A/G ratio, compared to the BMI. Conclusion The reference values presented herein may support the interpretation of body composition in public health settings and the identification of people who may benefit from intervention to improve musculoskeletal and metabolic health. The BRI better reflects DXA-derived body composition and may provide screening utility beyond that of the BMI.