Discordantly high spinal bone mineral density values in patients with adult lumbar scoliosis

Study Design. Retrospective study. Objective. The purposes of this study were: 1) to investigate the validity of bone mineral density measurements with DEXA in patients with adult lumbar scoliosis and 2) to investigate the association between osteoporosis and adult lumbar scoliosis. Summary of Background Data. Osteoporosis and lumbar degenerative scoliosis are phenomena encountered with increased frequency in aging, often concurrently. It has been suggested that scoliosis predisposes to osteoporosis, but degenerative scoliosis could falsely elevate spinal bone mineral density measurements. Methods. The feasibility of measuring Cobb's angle in DEXA scans was established in 48 surgical candidates with standing anteroposterior lumbar radiographs and supine DEXA scans. Charts and radiographs of 454 consecutive adult patients evaluated at an osteoporosis center were reviewed thereafter. The association between age, lumbar curve, and various bone density measurements was investigated. Bone density measurements between nonscoliotic and scoliotic patients with and without a history of adolescent scoliosis were compared. Results. Cobb's angle on DEXA scan was measured with an error of 4 and correlated highly with the plain lumbar radiographs. The prevalence of scoliosis was 9.47% in this cohort of patients. Advancing age was associated with an increase in osteoporosis in both scoliotic and nonscoliotic patients. Scoliotic patients demonstrated increased spinal bone mineral density (BMD) measurements compared with nonscoliotic patients, resulting in discrepancies between hip and spine BMD values. This discrepancy correlated with aging and curve magnitude (up to 30% for curves of 43). Scoliotic patients demonstrated significantly lower hip BMD values than nonscoliotic. Curve magnitude did not correlate with severity of osteoporosis. Conclusions. Cobb's angle measurements on DEXA scans are reliable and comparable to conventional radiographs. Spinal BMD values are less valuable for monitoring osteoporosis than hip values in scoliotic patients; an increasing discrepancy with age was noted. Scoliotic patients exhibited discordantly high spinal BMD values, despite significant hip osteoporosis. The discrepancy correlated with aging and curve magnitude. Scoliosis was common among the osteoporotic population (9.47%). Lumbar scoliosis is a useful clinical marker for osteoporosis, irrespective of scoliosis history and magnitude. Viable alternatives for osteoporosis evaluation of adult patients with lumbar scoliosis are hip DEXA values, in conjunction with other BMD measurements.