Inter-Rater Reliability of Hounsfield Units as a Measure of Bone Density: Applications in the Treatment of Thoracolumbar Fractures

Abstract

The study sought to investigate the reliability of computed tomography (CT)-derived Hounsfield unit (HU) measurements and ascertain the correlation between HU with quantitative CT (qCT)-derived bone mineral density (BMD) in cases of traumatic thoracolumbar fracture, based on native CT scans. This study is a retrospective cross-sectional analysis of data sets from patients who received native CT scans and bone mineral density measurements (qCT) of the same vertebral body. Two different CT scanner models were used. The inter-rater reliability of 4 raters, which measured HU in native CT scans, was calculated using intraclass correlation coefficient for absolute agreement (ICC(3,1)). For the correlation between HU and qCT values, respectively the prediction of qCT based on HU, linear regression was used. Bland-Altman plots were used for visual comparison of predicted and measured qCT values. In total 305 data sets were analyzed. CT scanner model was found to have no significant impact on HU (P= 0.125). The inter-rater reliability for HU measurements from native CT scans was ICC(3,1)=0.932 (95%confidence interval 0.919-0.943, P < 0.001). The linear regression showed significant correlation of HU and qCT values for each rater (P < 0.001). The equation for qCT prediction with averaged coefficient and constant is qCT= 0.8 HU+ 5. In the Bland-Altman plots no bias of predicted qCT values could be found, but a trend to overestimate predicted higher qCT values and underestimate lower qCT values, respectively. HU measurement shows very high inter-rater reliability. The HU values correlate closely with qCT BMD values. In summary, it seems that HU measurement is a suitable tool to readily and accurately assess bone quality without further scans or effort in cases of thoracolumbar spinal trauma.