A Pediatric Bone Mass Scan has Poor Ability to Predict Peak Bone Mass: An 11-Year Prospective Study in 121 Children

This 11-year prospective longitudinal study examined how a pre-pubertal pediatric bone mass scan predicts peak bone mass. We measured bone mineral content (BMC; g), bone mineral density (BMD; g/cm(2)), and bone area (cm(2)) in femoral neck, total body and lumbar spine by dual-energy X-ray absorptiometry in a population-based cohort including 65 boys and 56 girls. At baseline all participants were pre-pubertal with a mean age of 8 years (range 6-9), they were re-measured at a mean 11 years (range 10-12) later. The participants were then mean 19 years (range 18-19), an age range that corresponds to peak bone mass in femoral neck in our population. We calculated individual BMC, BMD, and bone size Z scores, using all participants at each measurement as reference and evaluated correlations between the two measurements. Individual Z scores were also stratified in quartiles to register movements between quartiles from pre-pubertal age to peak bone mass. The correlation coefficients (r) between pre-pubertal and young adulthood measurements for femoral neck BMC, BMD, and bone area varied between 0.37 and 0.65. The reached BMC value at age 8 years explained 42 % of the variance in the BMC peak value; the corresponding values for BMD were 31 % and bone area 14 %. Among the participants with femoral neck BMD in the lowest childhood quartile, 52 % had left this quartile at peak bone mass. A pediatric bone scan with a femoral neck BMD value in the lowest quartile had a sensitivity of 47 % [95 % confidence interval (CI) 28, 66] and a specificity of 82 % (95 % CI 72, 89) to identify individuals who would remain in the lowest quartile at peak bone mass. The pre-pubertal femoral neck BMD explained only 31 % of the variance in femoral neck peak bone mass. A pre-pubertal BMD scan in a population-based sample has poor ability to predict individuals who are at risk of low peak bone mass.