Vertebral bone mineral density measured laterally by dual-energy X-ray absorptiometry

Abstract

The bone mineral density (BMD) of lumbar vertebrae in the anteroposterior (AP) view may be overestimated in osteoarthritis or with aortic calcification, which are common in elderly. Furthermore, the risk of spinal crush fracture should be more closely related inversely to the BMD of the vertebral body than to that of the posterior arch. Therefore, we measured BMD of lumbar vertebrae in lateral (LAT) view (L2–L3), using a standard dual-energy X-ray absorptiometer (DEXA), thus eliminating most of the posterior spinal elements. The precision of BMD LAT measurement was determined both in vitro and in healthy volunteers. Then, we compared the capability of BMD LAT and BMD AP scans for monitoring bone loss related to age and for discriminating the BMD of postmenopausal women with nontraumatic vertebral fractures from that of young subjects. In vitro, when a spine phantom was placed in lateral position in the middle of 26 cm of water in order to simulate both soft-tissue thickness and X-ray source remoteness, the coefficient of variation (CV) of six repeated determinations of BMD was 1.0%. In vivo, the CV of paired BMD LAT measurements obtained in 20 healthy volunteers after repositioning was 2.8%. The age-related difference between a peak bone mass group estimated in a group of 27 healthy women aged 20 to 35 years and a group of 50 women aged 60 to 75 years, in whom neither vertebral fracture nor osteoporosis risk factors could be detected, were 21.7% and 37.6% in AP and LAT view, respectively. An arbitrary BMD fracture threshold was defined in AP and LAT views as the 90th percentile of the BMD value of a group of 22 osteoporotic women with vertebral fractures. The distribution of BMD AP and LAT above and below this threshold in 169 consecutively screened women without vertebral fracture was then analysed. In both AP and LAT views, 39.1% and 31.3% had BMD values above and below this threshold, respectively. Of the remaining, 16.0% had a BMD below this threshold only in AP and 13.6% only in LAT view. Thus, if BMD LAT was a better reflection of vertebral body bone mass than BMD AP, and thereby a better predictor of the resistance to crush fracture, our results would suggest that only the use of the standard AP view could under- or overestimate spinal fracture risk in about 30% of women screened for osteoporosis. In conclusion, our results indicate that BMD measurement in lateral view is feasible with a standard DEXA instrument. This mode of scanning, besides overcoming artefacts due to osteoarthritis of the posterior arch and aortic calcifications, appears to provide a greater sensitivity for assessing bone mass loss of the vertebral body than the standard anteroposterior scan.