Cortical Bone Thickness Predicts the Quantitative Bone Mineral Density of the Proximal Humerus

Introduction: Quantitative bone mineral density (BMD) of the humerus is difficult to determine but relevant for osteoporosis and fracture treatment. Dual-energy X-ray absorptiometry (DXA) of the femur and lumbar spine overestimates the humeral BMD and is not ubiquitously available. Therefore, this study evaluated whether the cortical bone thickness (CBT) of the humerus or DXA of the forearm is able to predict humeral BMD.

Methods: Humeral BMD of 54 upper cadaver extremities (22 pairs, 10 single) (19-90 years) was determined by high-resolution peripheral-quantitative-computed-tomography (HR-pQCT) (volumetric BMD (vBMD)) and DXA (areal BMD (aBMD)) of the proximal humerus and distal forearm. Average and gauge cortical bone thickness (CBTavg/ CBTg) of the humeral diaphysis was determined from standard radiographs (XR) and computed-tomography (CT) and compared to the humeral BMD. Pearson (r) and intraclass-correlation-coefficients (ICC) were used to compare results and rater-reliability.

Results: CBTavg from XR strongly correlated with the humeral BMD (r = 0.78 aBMD (DXA) and r = 0.64 vBMD (HR-pQCT) (p < 0.0001)). The CBTg revealed a weaker correlation (r = 0.57 aBMD and r = 0.43 vBMD). CBT derived from XR strongly correlated to those from the CT (r = 0.82-0.90) and showed an excellent intra- and inter-rater correlation (ICC 0.79-0.92). Distal forearm aBMD correlated well with the humeral aBMD (DXA) (r = 0.77) and paired specimens highly correlated to the contralateral side (humerus r = 0.89, radius r = 0.97).

Conclusions: The CBTavg can reliably be determined from standard radiographs and allows a good prediction of quantitative humeral bone mineral density (aBMD or vBMD) if measurements are not available. Furthermore, the distal forearm or the contralateral humerus can serve as a side to estimate the BMD if the ipsilateral side is impaired.