Objective: The calcaneus has been proposed as an alternative site for assessment of bone mineral density (BMD) in women with osteoporosis. The objectives of this study were to evaluate the clinical usefulness of dual-energy X-ray absorptiometry (DXA) of the calcaneus for predicting spinal fracture and to compare the predictive value of assessment of BMD by DXA of the calcaneus, quantitative CT or DXA of the lumbar spine, and DXA of the most distal part of the radius.
Subjects and methods: The study group consisted of 399 healthy volunteer women (aged 19-81 years) and 27 women (aged 44-85 years) who had osteoporosis. They had DXA of the calcaneus, quantitative CT and DXA of the lumbar spine, and DXA of the most distal part of the radius. T-score analyses were used to compare decreases in BMD in postmenopausal women with decreases in premenopausal women, and to compare decreases in women with and without osteoporosis.
Results: T-score analysis of age-related changes in BMD in the control subjects showed that densities in the calcaneus, the lumbar spine, and the most distal part of the radius decrease significantly with age in postmenopausal women. In women with osteoporosis (n = 27), BMDs in the calcaneus (DXA, p < .0002), L1-L3 (quantitative CT, p < .0001), L1-L4 (DXA, p < .0005), and the most distal part of the radius (DXA, p < .005) were significantly lower than those in healthy control subjects (n = 77). T-score analysis indicated that in women with osteoporosis, DXA of the calcaneus showed BMD losses greater than those shown by DXA of L1-L4 and the most distal part of the radius and similar to those shown by quantitative CT of L1-L3. Receiver-operating characteristic (ROC) analysis showed that findings on DXA of the calcaneus were as good a predictor of the risk of spinal fracture as findings on DXA of the lumbar spine and not as good as findings on quantitative CT of the lumbar spine.
Conclusion: The results suggest that DXA assessment of BMD in the calcaneus is not a useful adjunct to other measurements. However, it can be used to predict the risk of fracture when better measures, such as quantitative CT, are unavailable.