Bone mineral density and geometry parameters determined in vitro from dual-energy digital radiography images in the assessment of bone maximal load of reindeer femora

Abstract

Background

Dual-energy digital radiography (DEDR) has been shown to be a potential method to determine bone mineral density (BMD) and predict maximal load with similar accuracy as standard bone densitometry using DXA (dual-energy X-ray absorptiometry). In addition to bone density, bone geometry has also been shown to have effect on bone fragility and fracture risk.

Purpose

To examine the combination of BMD and geometry parameters, as determined from a DEDR experiment, to predict bone maximal load.

Material and Methods

Reindeer femora (n = 47) were imaged at two energies (79 kV_p and 100 kV_p) using a clinical digital radiography system. BMD was determined in four regions from these images using the DXA calculation principle. Various geometrical parameters were determined from the 79 kV_p image. Femora were mechanically tested using axial loading configuration. Pearson correlation coefficients were determined between geometrical parameters and BMDs or maximal load. Multiple stepwise linear regression analysis was used to find the best combination to predict bone maximal load.

Results

From the geometrical parameters, femoral shaft diameter (FSD) and femoral neck axis length (FNAL) correlated best with the maximal load (r = 0.629 and r = 0.446, P < 0.01, respectively). The best combination of parameters to predict bone fragility was BMD at Ward's triangle, FSD and FNAL (r = 0.787, P < 0.05), whereas the correlation coefficient between BMD at Ward's triangle and maximal load was 0.653 (P < 0.05).

Conclusion

The combination of DEDR-based BMD and geometrical parameters predicts reindeer bone maximal load with reasonable accuracy and the combined analysis improves the prediction of maximal load compared to BMD prediction only.