Femoral neck-parallel reconstruction enhances correlation between CT-based measurements and dual-energy X-ray...

To evaluate whether femoral neck-parallel reconstruction of CT images improves the correlation with dual-energy X-ray absorptiometry for osteoporosis assessment compared with conventional table-parallel reconstruction.

A retrospective analysis was conducted on 174 patients who underwent hip CT and dual-energy X-ray absorptiometry. Two CT image reconstruction methods were compared: femoral neck-parallel and table-parallel methods. CT values were extracted from cancellous bone regions in the femoral neck corresponding to the area measured by DXA. Correlation coefficients between CT values and dual-energy X-ray absorptiometry-derived bone mineral density and T-scores were calculated. A stratified analysis by femoral neck-to-table angle (< 5° vs. ≥ 5°) was performed. Receiver operating characteristic (ROC) curve analysis was performed to compare the ability of each method to discriminate between patients with and without low bone mass (defined as osteoporosis or osteopenia).

CT values from the femoral neck-parallel method showed significantly stronger correlations with bone mineral density (r = 0.74) and T-score (r = 0.77) than those from the table-parallel method (bone mineral density: r = 0.66, T-score: r = 0.71) (p = 0.001 for both). In patients with ≥ 5° misalignment, the femoral neck-parallel method showed significantly higher correlation coefficients than the table-parallel method (bone mineral density, p < 0.001; T-score, p = 0.005). ROC curve analysis revealed that the area under the curve for detecting low bone mass was higher in the femoral neck-parallel method (0.87) than in the table-parallel method (0.83) (p = 0.012).

Femoral neck-parallel reconstruction significantly improves the correlation between CT values and dual-energy X-ray absorptiometry for low bone mass assessment compared to table-parallel reconstruction.