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

Area of Science:

• Orthopedics and Bone Health
• Medical Imaging and Diagnostics
• Geriatric Medicine

Background:

• Current osteoporosis fracture risk assessment tools often combine bone mineral density (BMD) with clinical risk factors, yet accurately stratifying risk remains a challenge.
• Existing methods may not fully capture the complex interplay of bone density and structural integrity contributing to fracture susceptibility.

Purpose of the Study:

• To develop and validate a novel fracture risk assessment tool, microFRAC, utilizing high-resolution peripheral quantitative computed tomography (HR-pQCT) data.
• To evaluate microFRAC's performance against established tools like FRAX and BMD-based models for predicting osteoporotic fractures in older adults.

Main Methods:

• Development of the microFRAC model using random survival forests on an international prospective cohort (n=6802) of older adults.
• Input variables included HR-pQCT derived volumetric bone density and microarchitecture parameters, clinical risk factors, and femoral neck areal BMD (FN aBMD).
• Comparative analysis of microFRAC against FRAX and FN aBMD reference models using c-index for predictive performance.

Main Results:

• microFRAC demonstrated predictive accuracy for osteoporotic fractures (c-index=0.673), outperforming FRAX (c-index=0.617) and FN aBMD models (c-index=0.636).
• Removing most clinical risk factors and FN aBMD from microFRAC did not significantly impact its 5- and 10-year fracture risk prediction capabilities.
• The tool's performance improved when focused on major osteoporotic fractures (c-index=0.733).

Conclusions:

• microFRAC offers a personalized fracture risk assessment approach leveraging direct measures of bone density and microarchitecture from HR-pQCT.
• This HR-pQCT-based tool presents a potential advancement over current clinical fracture risk assessment strategies.
• The findings suggest that detailed bone structure analysis may enhance the prediction of osteoporotic fractures.