Simple recalibration multipliers could improve FRAX estimation of vertebral fractures in patients who underwent VFA: JAMA

A new study published in the Journal of American Medical Association showed that simple multipliers can significantly increase the agreement between observed and anticipated fracture risk and regain FRAX calibration in people with fractures diagnosed by vertebral fracture assessment (VFA).

The most popular and reliable fracture risk prediction tool in the world is FRAX. Dual-energy x-ray absorptiometry (DXA) vertebral fracture evaluation can be used to detect vertebral fractures, which are a sign of later osteoporotic fractures. This work was set to evaluate the FRAX calibration and provide a straightforward technique for enhancing the FRAX-predicted fracture likelihood when there is a VFA-identified fracture present.

All participants in the Manitoba Bone Mineral Density Registry who had a VFA performed between March 31, 2010, and March 31, 2018, had their DXA and VFA data examined in this prognostic analysis. These people were randomized to either the validation cohort or the development cohort. Expert readers coded VFAs as either positive (≥1 vertebral fractures found) or negative using a modified algorithm-based qualitative technique. The period of statistical analysis was August 7, 2022, through May 22, 2023.

The provincial data on connected population-based health care was utilized to determine incident fractures and deaths. To forecast the 10-year observed risk of fracture, cumulative incidence curves for hip fracture and major osteoporotic fractures (MOF) were created by taking competing mortality into account. With and without VFA data, the observed probability was contrasted with the fracture probability predicted by FRAX, and the FRAX was recalculated using the resulting multipliers.

The entire group of 11,766 people was divided into 2 groups at random where the validation cohort and the development cohort. FRAX was well calibrated in subgroups with negative VFA findings during a mean (SD) observation duration of 3.8 (2.3) years, with the longest observation being 7.5 years.

The 10-year FRAX-predicted MOF likelihood for people with a positive VFA test but no history of clinical fractures was 23.4% with VFA information and 16.3% without. Based on the observed 10-year risks of 26.9% and 11.2%, the recalibration multipliers for MOF and hip fracture were 1.15 and 1.31, respectively.

The 10-year FRAX-predicted probabilities for MOF and hip fracture were 25.0% and 9.3%, respectively, among those with a positive VFA test and a history of clinical fracture. Overall, the findings of this predictive analysis indicate that in individuals with fractures diagnosed by VFA, FRAX understated fracture risk.

Source:

Ye, C., Leslie, W. D., Morin, S. N., Lix, L. M., McCloskey, E. V., Johansson, H., Harvey, N. C., Lorentzon, M., & Kanis, J. A. (2023). Adjusting FRAX Estimates of Fracture Probability Based on a Positive Vertebral Fracture Assessment. In JAMA Network Open (Vol. 6, Issue 8, p. e2329253). American Medical Association (AMA). https://doi.org/10.1001/jamanetworkopen.2023.29253