Predictive ability of novel volumetric and geometric indices derived from dual-energy X-ray absorptiometric images of the proximal femur for hip fracture compared with conventional areal bone mineral density: the Japanese Population-based Osteoporosis (JPOS) Cohort Study

3D-DXA Based Finite Element Modelling for Femur Strength Prediction: Evaluation Against QCT

Osteoporosis is characterised by the loss of bone density resulting in an increased risk of fragility fractures. The clinical gold standard for diagnosing osteoporosis is based on the areal bone mineral density (aBMD) used as a surrogate for bone strength, in combination with clinical risk factors. Finite element (FE) analyses based on quantitative computed tomography (QCT) have been shown to estimate bone strength better than aBMD. However, their application in the osteoporosis clinics is limited due to exposure of patients to increased X-rays radiation dose. Statistical modelling methods (3D-DXA) enabling the estimation of 3D femur shape and volumetric bone density from dual energy X-ray absorptiometry (DXA) scan have been shown to improve osteoporosis management. The current study used 3D-DXA based FE analyses to estimate femur strength from the routine clinical DXA scans and compared its results against 151 QCT based FE analyses, in a clinical cohort of 157 subjects. The linear regression between the femur strength predicted by QCT-FE and 3D-DXA-FE models correlated highly (coefficient of determination R2 = 0.86) with a root mean square error (RMSE) of 397 N. In conclusion, the current study presented a 3D-DXA-FE modelling tool providing accurate femur strength estimates noninvasively, compared to QCT-FE models.

Abaloparatide Effects on Cortical Volumetric BMD and Estimated Strength Indices of Hip Subregions by 3D-DXA in Women With Postmenopausal Osteoporosis

In ACTIVE, abaloparatide increased areal BMD (aBMD) of the hip and femoral neck vs teriparatide and placebo in women with osteoporosis. Previously, 3D-processing of dual X-ray absorptiometry (DXA) scans of a subgroup of ACTIVE subjects showed similar increases in trabecular volumetric BMD (Tb.vBMD) and greater increases in cortical vBMD (Ct.vBMD) of the total hip with abaloparatide vs teriparatide. The current analyses from this subgroup describe 2D- and 3D-DXA data for hip subregions. Randomly selected subjects from ACTIVE (n = 250/treatment group) who received 18 mo of placebo, abaloparatide 80 µg, or open-label teriparatide 20 µg by daily subcutaneous injection underwent hip DXA at baseline, and mo 6 and 18 of treatment. Areal BMD of the femoral neck, trochanter, and femoral shaft was determined using standard 2D-DXA and 3D-SHAPER software to retrospectively evaluate changes from baseline in volumetric parameters of these 3 hip subregions, including trabecular and cortical segmentation. Changes in biomechanical parameters cross-sectional moment of inertia (CSMI), section modulus (Z), and buckling ratio were also evaluated. Femoral neck, trochanter, and shaft aBMD increased in the abaloparatide and teriparatide groups at mo 6 and 18 vs placebo, with greater increases for abaloparatide vs teriparatide at the femoral neck at mo 6 and the shaft at mo 6 and 18. All 3 subregions showed similar significant increases in Tb.vBMD with abaloparatide and teriparatide vs placebo, whereas Ct.vBMD of all 3 subregions showed greater increases after 18 mo of abaloparatide vs teriparatide. Biomechanical parameters improved in all subregions with abaloparatide and teriparatide vs placebo, with greater improvements in CSMI and Z of the femoral neck and lower shaft after 6 and 18 mo of abaloparatide vs teriparatide. Differential femoral neck and shaft Ct.vBMD responses may explain the greater increases in CSMI and Z of those subregions with abaloparatide vs teriparatide.

Differences in Hip Geometry Between Female Subjects With and Without Acute Hip Fracture: A Cross-Sectional Case-Control Study

BACKGROUND AND PURPOSE

Although it is widely recognized that hip BMD is reduced in patients with hip fracture, the differences in geometrical parameters such as cortical volume and thickness between subjects with and without hip fracture are less well known.

MATERIALS AND METHODS

Five hundred and sixty two community-dwelling elderly women with hip CT scans were included in this cross-sectional study, of whom 236 had an acute hip fracture. 326 age matched women without hip fracture served as controls. MIAF-Femur software was used for the measurement of the intact contralateral femur in patients with hip fracture and the left femur of the controls. Integral and cortical volumes (Vols) of the total hip (TH), femoral head (FH), femoral neck (FN), trochanter (TR) and intertrochanter (IT) were analyzed. In the FH and FN the volumes were further subdivided into superior anterior (SA) and posterior (SP) as well as inferior anterior (IA) and posterior (IP) quadrants. Cortical thickness (CortThick) was determined for all sub volumes of interest (VOIs) listed above.

RESULTS

The average age of the control and fracture groups was 71.7 and 72.0 years, respectively. The fracture patients had significantly lower CortThick and Vol of all VOIs except for TRVol. In the fracture patients, cortical thickness and volume at the FN were significantly lower in all quadrants except for cortical volume of quadrant SA (p= 0.635). Hip fracture patients had smaller integral FN volume and cross-sectional area (CSA) before and after adjustment of age, height and weight. With respect to hip fracture discrimination, cortical volume performed poorer than cortical thickness across the whole proximal femur. The ratio of Cort/TrabMass (RCTM), a measure of the internal distribution of bone, performed better than cortical thickness in discriminating hip fracture risk. The highest area under curve (AUC) value of 0.805 was obtained for the model that included THCortThick, FHVol, THRCTM and FNCSA.

CONCLUSION

There were substantial differences in total and cortical volume as well as cortical thickness between fractured and unfractured women across the proximal femur. A combination of geometric variables resulted in similar discrimination power for hip fracture risk as aBMD.