Bone Mineral Density (BMD) Assessment of Central Skeletal Sites From Peripheral BMD and Ultrasonographic Measurements: An Improved Solution Employing Age and Weight in Type 3 Regression

Abstract 

The objective of this study was to develop a method whereby bone mineral density measurements of the heel and finger, as well as ultrasonographic measurements of calcaneal sound transmission, could identify individuals with a diagnosis of osteoporosis or osteopenia by the World Health Organization criteria for these diagnoses in the central skeleton (i.e., the lumbar spine (LS) and hip [femoral neck] [FN]). Two hundred and forty-four women in a university hospital laboratory setting had dual-energy X-ray absorptiometry measurements of bone mineral density (BMD) in the calcaneus, finger, hip, and spine, and quantitative ultrasound of the calcaneus. Regression equations were developed to predict central bone mineral T-scores from T-scores of peripheral measurements, adjusted by age and weight. Equations were validated by predicting the cut point for osteopenia at the lumbar spine and hip (T-score=−1.0). Ninety-five percent confidence intervals of the mean predicted LS or FN T-score from each peripheral site included −1.0. We conclude that our derived regression equations (taking into account interaction of peripheral BMD with patient age and weight) are useful for predicting T-scores in the central skeleton. This approach reduces the potential for misdiagnosis, which can result if one uses unadjusted peripheral T-scores, which are only moderately correlated with the central measurements of BMD.

Key Words: Bone mineral density (BMD), dual-energy X-ray absorptiometry (DXA), osteoporosis, osteopenia, quantitative ultrasound (QUS)