Does adult lumbar scoliosis affect spine T-score accuracy in osteoporosis diagnosis?-a cross-sectional study of adults aged 50 years and older.

Abstract

Background: Dual-energy X-ray absorptiometry (DEXA) is the gold standard for assessing bone mineral density (BMD) and diagnosing osteoporosis (T-score ≤-2.5). Lumbar spine T-scores are often used when hip measurements are unavailable or unreliable. However, degenerative spinal changes can artifactually elevate lumbar spine BMD, leading to discordance between spine and hip T-scores and potential misclassification of osteoporosis. The impact of adult degenerative scoliosis on this discordance remains poorly defined. This study evaluates the association between lumbar scoliosis and spine-hip T-score discordance in adults ≥50 years.

Methods: A retrospective cross-sectional study included consecutive adults ≥50 years who underwent DEXA scans over a 4-year period at a university hospital. Patients with secondary causes of osteoporosis were excluded. Participants were grouped by presence or absence of scoliosis. Spine and hip T-scores were recorded and classified as no, minor, or major discordance. No discordance indicated no grade difference between sites. Minor discordance referred to a one-grade difference (normal vs. osteopenia or osteopenia vs. osteoporosis). Major discordance referred to a two-grade difference (normal vs. osteoporosis).

Results: A total of 6,969 patients were included; 91.6% were female with a mean age of 60.6 years. Scoliosis was present in 637 patients (9.1%), with a mean Cobb angle of 17° (range, 10-66°). Spine T-scores classified more patients as normal (43.0% vs. 31.1%) and fewer as osteopenic (41.5% vs. 48.9%) or osteoporotic (15.5% vs. 19.1%) compared with hip T-scores. Patients with scoliosis were older (67 vs. 60 years, P<0.001) and more likely female (94% vs. 91%, P<0.001). Spine-hip discordance was more frequent in scoliosis patients, with minor discordance in 44.1% vs. 39.0% and major discordance in 5.0% vs. 2.8% of non-scoliosis patients.

Conclusions: Lumbar scoliosis is associated with increased spine-hip T-score discordance, limiting the reliability of lumbar spine measurements in osteoporosis assessment. These effects are more pronounced in older adults and females. Clinicians should interpret spine T-scores cautiously in the presence of scoliosis and prioritise hip measurements when assessing bone health or planning interventions where bone quality is critical.