A Fracture Risk Assessment Tool for High Resolution Peripheral Quantitative Computed Tomography

Abstract

Most fracture risk assessment tools use clinical risk factors combined with bone mineral density (BMD) to improve assessment of osteoporosis; however, stratifying fracture risk remains challenging. This study developed a fracture risk assessment tool that uses information about volumetric bone density and three-dimensional structure, obtained using high-resolution peripheral quantitative compute tomography (HR-pQCT), to provide an alternative approach for patient-specific assessment of fracture risk. Using an international prospective cohort of older adults (n = 6802) we developed a tool to predict osteoporotic fracture risk, called μFRAC. The model was constructed using random survival forests, and input predictors included HR-pQCT parameters summarizing BMD and microarchitecture alongside clinical risk factors (sex, age, height, weight, and prior adulthood fracture) and femoral neck areal BMD (FN aBMD). The performance of μFRAC was compared to the Fracture Risk Assessment Tool (FRAX) and a reference model built using FN aBMD and clinical covariates. μFRAC was predictive of osteoporotic fracture (c-index = 0.673, p < 0.001), modestly outperforming FRAX and FN aBMD models (c-index = 0.617 and 0.636, respectively). Removal of FN aBMD and all clinical risk factors, except age, from μFRAC did not significantly impact its performance when estimating 5-year and 10-year fracture risk. The performance of μFRAC improved when only major osteoporotic fractures were considered (c-index = 0.733, p < 0.001). We developed a personalized fracture risk assessment tool based on HR-pQCT that may provide an alternative approach to current clinical methods by leveraging direct measures of bone density and structure. © 2023 The Authors. Journal of Bone and Mineral Research published by Wiley Periodicals LLC on behalf of American Society for Bone and Mineral Research (ASBMR).

Keywords: BONE MICROARCHITECTURE; FRACTURE PREDICTION; HIGH RESOLUTION PERIPHERAL QUANTITATIVE COMPUTED TOMOGRAPHY; MACHINE LEARNING; OSTEOPOROSIS; RISK ASSESSMENT.

Figure 1:

The μFRAC 5-year predicted risk of any osteoporotic fracture in males (left) and females (right) relative to age. The smoothed line and shaded region represent the fitted LOWESS (locally weighted scatterplot smoothing) curve and 95% confidence intervals for the relationship between 5-year osteoporotic fracture risk and age.

Figure 2:

Variable importance (VIMP) for the complete and simplified μFRAC models to predict any osteoporotic fracture. The boxplots represent the distribution of VIMP score for each parameter across the 100 bootstrap iterations, and parameters are sorted from highest to lowest VIMP within each model. The dashed line represents zero VIMP, and the parameters coloured in red have a VIMP above the overall mean of the complete μFRAC model (0.012), illustrated for comparison purposes. Abbreviations: FN aBMD is femoral neck areal bone mineral density; Tt.BMD, Ct.BMD, and Tb.BMD are total, cortical, and trabecular bone mineral density respectively; Tb.N and Tb.1/N.SD are trabecular number and inhomogeneity respectively; Ct.Th^d is derived cortical thickness; and Tt.Ar, Ct.Ar, and Tb.Ar are total, cortical, and trabecular respectively.

Figure 3:

Time-dependent receiver operator characteristic (ROC) curves for prediction of any osteoporotic fracture risk by μFRAC, FRAX, and the reference model built using femoral neck areal bone mineral density, sex, age, height, weight, and previous fracture as input parameters. The dashed line represents an area under the curve (AUC) of 0.50. *All models were significantly different (p<0.05) from one another after post-hoc comparisons except for μFRAC and μFRAC′ (p>0.68).

Figure 4:

Time-dependent receiver operator characteristic (ROC) curves for prediction of major osteoporotic fracture risk by μFRAC, FRAX, and the reference model built using femoral neck areal bone mineral density, sex, age, height, weight, and previous fracture as input parameters. The dashed line represents an area under the curve (AUC) of 0.50. *All models were significantly different (p<0.05) from one another after post-hoc comparisons except for μFRAC, μFRAC′, and μFRAC″ (p>0.05).