Correlation of quantitative computed tomography derived bone density values with Hounsfield units of a contrast medium computed tomography in 98 thoraco-lumbar vertebral bodies.

INTRODUCTION: Vertebral fractures in patients with bone density reduction are often a major challenge for the surgeon, as reduced bone density can lead to screw loosening. Several options are available to determine bone density preoperatively. In our study, we investigated the correlation of Hounsfield units (HU) of a contrast medium computed tomography (CT) to the bone density values of a quantitative computed tomography (QCT) and computed a formula to estimate bone density values using HU. MATERIALS AND METHODS: In our retrospective data analysis, we examine 98 vertebral bodies from 35 patients who received a contrast medium CT of the spine and a QCT, performed no longer than 1 month apart. The determined HU from the contrast medium CT were compared with the bone density values of the QCT and examined for correlations. Linear logistic regression was used to estimate bone density values base on HU. RESULTS: A strong correlation was found between the HU measured in the CT and the bone density values (r = 0.894, p < 0.001), irrespective of patients' gender. We also found no correlation differences when the HU were measured at different levels. Bland-Altman plot demonstrated good agreement between the two measurements. The following formula was developed to estimate bone density values using HU: QCT-value = 0.71 × HU + 13.82. CONCLUSIONS: Bone density values correlate well to HU measured in contrast medium CT. Using simple formula, the bone density of a contrast medium CT of vertebral bodies can be estimated based on HU without additional examinations and unnecessary costs.

Correlation of Bone Density Values of Quantitative Computed Tomography and Hounsfield Units Measured in Native Computed Tomography in 902 Vertebral Bodies.

OBJECTIVE: Due to the increasing age of patients, the evaluation of bone density is crucial, especially in preparation for spinal surgery. The aim of this study was to determine bone density using a computed tomography (CT) and to correlate Hounsfield units (HU) with bone density values of a quantitative computed tomography (QCT). METHODS: The study is a monocentric, retrospective data analysis. We examined 902 vertebral bodies from a total of 369 patients who received a CT of the thoracolumbar spine in the period from 2015 to 2019 and compared the HU with values of a QCT. A general equation for calculation the QCT values was established. RESULTS: We found a significant correlation between the Hounsfield units and the corresponding QCT-values (r = 0.944, P < 0.001). We also demonstrated that the calculated QCT values are independent of patient sex (P < 0.942). Furthermore, we could not demonstrate differences in the correlation of the 3 measured levels (axial, sagittal, and coronary) to the QCT values. The QCT-values can be calculated on the basis of a native CT of the lumbar spine using the equation: QCT = 17.8 + 0.7 × HU. CONCLUSIONS: The equation allows calculating bone density values without the need for an additional QCT and without further radiation exposure or costs. With this measuring method it is possible to obtain additional information from a computed tomography.