Background/Purpose: Bone mineral density (BMD) of the spine or hip are predictors of osteoporotic fracture, but individuals who sustain “osteoporosis-related” fractures frequently have normal BMD or osteopenia at the hip and spine. In addition, BMD of the spine and, to a lesser extent, of the hip, is commonly elevated in older adults due to degenerative changes. This is not observed with BMD of the radius 1/3. However, the predictive value of radius BMD for fracture is not well established. Thus, our aim was to investigate the applicability of radius 1/3 BMD as a predictor of fractures in community-dwelling elderly people whose fracture risk is known to be increased (Domiciano DS et al. Osteoporos Int. 2021;32(4):747-757; Domiciano DS et al. Osteoporos Int. 2014;25(12):2805-15).

Methods: 707 older adults (≥65 years) were evaluated at baseline and after a mean follow-up of 4.3 ± 0.8 years. Clinical questionnaire, BMD, and laboratory tests were performed at baseline. Non-vertebral fracture (hip, proximal humerus, rib, forearm) was determined during the follow-up. New vertebral fracture was defined by radiograph. Multivariate Poisson regression models were used to verify whether radius BMD was an independent risk factor for fracture, alone and in combination with spine and/or hip.

Results: 449 women (72.9±4.8 years) and 258 men (72.3±4.7 years) were included in the study. The agreement in the diagnostic classification (normal, osteopenia and osteoporosis) between spine/hip and radius 1/3 was 53.1% in women and 39.1% in men. 245 (56.3%) women and 73 (28.9%) men were osteoporotic by spine/hip, whereas 21 (4.8%) women and 6 (2.4%) men had osteoporosis only at radius, constituting an 8.6% increase in the number of women and 8.2% in the number of men defined as osteoporotic (n=266, 59% and n=79; 31%, respectively). In the multivariate regression, age (OR 1.06, 95% CI 1.02-1.10, p=0.005), estimated filtration glomerular rate (OR 1.02, 95%CI 1.00-1.03, p=0.008 ), falls (≥2/year) (OR 1.53, 95% CI 1.03-2.28, p=0.036), femoral neck BMD (RR 1.27, 95% CI 1.00-1 .59, p=0.048), total hip BMD (RR 1.43, 95% CI 1.12-1.79, p=0.003) and radius 1/3 BMD (RR 1.27, 95% CI 1.06-1.49, p=0.006, per each 1 SD decrease) were independent predictors of fracture in women. Radius 1/3 BMD measurement enhanced fracture risk estimates only when combined with spine (RR 1.41, 95% CI 1.04-1, 87, p=0.028). In regression models of combination of radius + hip or radius + spine/hip, BMD of radius 1/3 became non-significant and only hip BMD remained predictor of fracture [(RR 1.37, 95% CI 1.00-1,85, p=0.049) and (RR 1.72, 95% CI 1.06-2.78, p=0.028), respectively]. In men, fitting a model of risk factors was prevented by the limited number of events in male sample.

Conclusion: Although the reclassification of BMD values according to radius analysis resulted in only a small increase in osteoporosis diagnosis, radius 1/3 BMD assessment led to a better fracture prediction when combined with spine but not hip BMD. These findings suggest that, in elderly subjects, radius BMD measurement is clinically valid when the hip BMD cannot be assessed (for instance, in cases of bilateral total hip replacement or severe degenerative changes in the hips).

« Back to ACR Convergence 2022

ACR Meeting Abstracts - https://acrabstracts.org/abstract/bone-mineral-density-of-radius-1-3-led-to-a-better-fracture-prediction-when-combined-with-spine-but-not-hip-bmd-results-from-the-sao-paulo-ageing-health-spah-study/