Osteoporosis Assessment Using Bone Density Measurement in Hounsfield Units in the Femoral Native CT Cross-Section: A Comparison with Computed Tomography X-Ray Absorptiometry of the Hip.

Background/

Objectives

Bone mineral density (BMD) loss leads to osteoporosis, significantly increasing fracture risk in both the axial and peripheral skeleton. The extent to which it is possible to estimate the degree of osteoporosis in the hip by determining the density in Hounsfield Unit (HU) measurements derived from computed tomography (CT) scans and to calculate quantitative BMD and T values from the HU values should be examined.

Methods

A total of 240 patients (mean age: 64.9 ± 13.1 years, BMI: 26.8 ± 6.8 kg/m2) underwent CT-based BMD assessments using CTXA-Hip. Subregions of the proximal femur, including the femoral head, femoral neck, and intertrochanteric region, were analyzed for cancellous density in HUs using circular and irregular region-of-interest (ROI) measurements. Correlations between HU values and DEXA-equivalent BMD (mg/cm2) and T values were computed. Predictive power for osteoporosis was evaluated using ROC curve analysis.

Results

Cancellous bone density in the proximal femur showed a significant decline with increasing age and decreasing BMI (p < 0.05). The median BMD for the entire hip was 0.684 mg/cm2, and the median HU value for the proximal femur was 123.15. Strong correlations were observed between HU values and BMD (R2 = 0.904, p < 0.001) and T values (R2 = 0.911, p < 0.001). A T value of -2.5 corresponded to an HU value of 95.79 in the entire femur. ROC analysis demonstrated high sensitivity (0.92) and specificity (0.93) for HU-based osteoporosis prediction.

Conclusions

HU measurements provide a reliable method for estimating BMD and T values in the proximal femur, offering a valuable diagnostic tool for osteoporosis. The highest predictive accuracy was achieved when using an irregular ROI from the entire proximal femoral region.