Limitations of forearm bone densitometry as an index of vertebral or femoral neck osteopenia

Osteoporosis of the spine and femoral neck is a major problem in aging populations, but detection prior to a fracture remains a challenge. Forearm bone mineral content has been advocated as a useful screening test in this situation. We have examined the correlation between distal forearm bone mineral content, by single photon absorptiometry, and lumbar vertebral and femoral neck bone mineral density, by dual photon absorptiometry. Eighty women aged 20 to 76 years were studied and significant correlations (p less than .001) were found between the measurements on the forearm and at the two axial sites (r = 0.66 and 0.69, respectively). However, forearm bone mineral content was an unreliable predictor of axial bone mineral density. For prediction of lumbar spine osteopenia with a sensitivity of 88%, the false positive rate was 91%. Conversely, to achieve a specificity of 82%, the false negative rate was 65%. Similarly, for prediction of femoral neck osteopenia at a sensitivity of 92%, the false positive rate was 87%, and at a specificity of 90%, the false negative rate was 33%. These data demonstrate that forearm bone densitometry cannot be used as a screening procedure for osteopenia of the lumbar spine or femoral neck.