Fracture discrimination by combined bone mineral density (BMD) and microarchitectural texture analysis

Area of Science:

• Osteoporosis research
• Medical imaging analysis
• Biomechanical engineering

Background:

• Bone mineral density (BMD) alone has limitations in predicting osteoporotic fractures.
• Bone microarchitecture and texture parameters may offer improved fracture discrimination.
• Trabecular Bone Score (TBS) and mean Hurst parameter (H) are key texture metrics.

Purpose of the Study:

• To combine BMD with microarchitectural texture parameters (spine TBS and calcaneus H) for enhanced osteoporotic fracture detection.
• To evaluate the diagnostic value of BMD, TBS, and H individually and in combination.
• To assess fracture risk in women with normal or osteopenic BMD but low texture parameters.

Main Methods:

• 255 women underwent DXA scans (lumbar spine, total hip, femoral neck) and texture analysis (spine TBS, calcaneus H).
• Multivariate logistic regressions and odds ratios (OR) were used to evaluate fracture association.
• Receiver operating characteristic (ROC) curves assessed the area under the curve (AUC) for diagnostic models.

Main Results:

• Women with fractures were older and had lower BMD, TBS, and H.
• Both TBS and H were significantly associated with fractures, even after adjusting for age and TH-BMD.
• While ROC-AUC did not significantly improve with texture parameters, women with low TBS or H and normal/osteopenic BMD had higher fracture rates.

Conclusions:

• Texture parameters like TBS and H show potential interest in discriminating patients with or without osteoporotic fractures, complementing BMD.
• Further evaluation of the clinical added value of TBS and H alongside other risk factors is warranted.
• Microarchitectural texture analysis offers a promising avenue for improving fracture risk assessment beyond traditional BMD measurements.