Longitudinal Associations Between Changes in Bone Mechanical Strength and Fracture Risk Estimated by FRAC.

Abstract

Introduction: Monitoring bone health for osteoporosis is typically based on measuring areal bone mineral density (aBMD). However, widely used fracture risk prediction tools are primarily driven by clinical risk factors and show limited sensitivity to underlying bone changes over time. This study evaluated the ability of the new Microarchitecture Fracture Risk Assessment Calculator (FRAC) to detect longitudinal changes in fracture risk in relation to bone quality.

Methods: Our study cohort included 601 participants (70.2% female) from a longitudinal population study. HR-pQCT scans of the distal radius and tibia were acquired at two visits, 3-10 years apart. The FRAC 5-year risk of major osteoporotic fracture (MOF) was calculated at both time points. Participants were divided into quartiles based on the absolute change in tibia bone strength between study visits to assess the model's sensitivity to changes in bone fragility. Differences between quartiles were assessed using a Mann-Whitney U test and the standardized response mean (SRM). Additionally, changes in fracture risk by decade, were analyzed to investigate age- and sex-specific trends in fracture risk.

Results: The average age of participants was 53.8 ± 15.4 years, with an average follow-up of 6.8 ± 1.8 years. The greatest absolute annualized changes in FRAC risk occurred in individuals with the largest differences in bone strength (SRM = 0.73-0.78), while the least change was observed in individuals with minimal changes (SRM = 0.07-0.21). Age- and sex-specific trends aligned with previously established patterns of bone aging, showing the greatest annualized changes in fracture risk in menopausal females (40-60 years) and older adults (70+ years).

Conclusion: We demonstrated FRAC is sensitive to changes in fracture risk driven by declines in bone quality in aging adults. These results suggest FRAC is well suited for tracking fracture prediction longitudinally and has potential to monitor osteoporosis disease progression and treatment response.