Central tibia marrow density is associated with incident fractures among postmenopausal females.

Abstract

to examine regional tibial marrow density as measured by peripheral quantitative computed tomography (pQCT) and determine its association with incident fragility fractures among postmenopausal females. The Appendicular Muscle and Bone Extension Research Study (AMBERS) is a longitudinal cohort study that recruited postmenopausal females 60-85 yr of age. pQCT distal tibia scans were acquired at baseline (22.5 & 27.5 mm proximal from tibial endplate). Marrow was segmented from trabeculae using the Iterative Threshold-Seeking Algorithm. Central versus peripheral regions were delineated using edge detection and threshold-based concentric peeling, and marrow density was calculated. Self-reported incident fragility fractures across 7 follow-up years were ascertained by X-ray confirmation. Cox proportional hazards models, competing risk (against mortality), and conditional Cox models assessed the relationship between marrow density and time to first fracture. Models were adjusted for age, body mass index, diabetes/hypertension, hypo/hyperthyroidism, antiresorptive usage, diagnosis of osteoarthritis, physical activity, lumbar spine bone mineral density, and trabecular tissue mineral density at the same distal tibia site. Among 312 postmenopausal females, 41(13.1%) experienced an incident fracture over a median follow-up of 83.9(24.8) months. Lower central marrow density at the distal (hazard ratio (HR) per SD: 1.81(1.21, 2.71)) and proximal slice (HR:1.58(1.06, 2.36)) each associated with increased fracture risk. Peripheral marrow density only associated with increased fracture risk (2.65(1.19, 5.92) when comparing intermediate to highest levels. These trends held true after accounting for competing mortality, and time-dependent effects. Our findings suggested that central tibia marrow density as quantified by pQCT is a proxy for marrow composition that reveals important 4 to 7-yr risk of fragility fractures among postmenopausal females. These results highlight the potential of marrow density as a separate risk factor for fractures from bone density, warranting further investigation as a potential contributor to fracture risk pending external validation.