The value of a new computerized radiogrammetric method of assessment of the second metacarpal has been evaluated, and its results have been compared with those of single (SPA) and dual photon absorptiometry measurements (forearm and spine) in 74 and 79 postmenopausal women, respectively. Standard hand X-rays were digitized by a video-camera connected to a microcomputer. The combined cortical thickness (CCT) was automatically calculated in a zone of 10 mm around the midpart of the second metacarpal. The intra- and interobserver coefficients of variation were close to 1%. The correlation between CCT and SPA of the proximal and midforearm (with a significant amount of cortical bone) was satisfactory (r = 0.62 and 0.50, respectively; P < 0.001). The correlation between CCT and osteodensitometry of sites comprising more trabecular bone were not so good (0.50 for lumbar and 0.44 for distal forearm bone mineral density, respectively), but still significant (P < 0.001). Radiogrammetry proved unaffected by a change of X-ray apparatus, as measurements of 48 metacarpals radiographed by two different kinds of X-ray apparatus were not significantly different. Radiogrammetry is by no means the best method to evaluate bone mass. Its automation did not improve the correlation with osteodensitometric values. Radiogrammetry is still of interest in mass screening, particularly when other more expensive techniques such as osteodensitometric methods of bone mass measurement are not readily available. Its automation makes it simpler, faster, and more precise, rendering its use easier on a larger scale.

No abstract