Three-Dimensional Hounsfield Units Measurement of Pedicle Screw Trajectory for Predicating Screw Loosening in Lumbar Fusion Surgery

Abstract

Purpose: Dual-energy X-ray absorptiometry (DXA) is commonly used for evaluation of bone mineral density before spinal surgery, but frequently leads to overestimation in degenerative spinal diseases due to osteoproliferation factors. We introduce a novel method to compare the predictive ability of Hounsfield Units (HU) and DXA methods to predict screw loosening after lumbar interbody fusion surgery in degenerative spinal diseases by measuring HU of pedicle screw trajectory on computed tomography (CT) images preoperatively. Patients and Methods: This retrospective study was conducted on patients who underwent posterior lumbar fusion surgery for degenerative diseases. CT HUs measurement was performed using medical imaging software, including the cancellous region on cross-sections of the vertebral body and three-dimensional pedicle screw trajectory. Receiver operating characteristic (ROC) curve analyses were performed for the risk of pedicle screw loosening in association with the Hounsfield scale and preoperative BMD, and the area under the curve (AUC) and the cutoff values were calculated. Results: A total of 90 patients were enrolled and were divided into loosening (n = 33, 36.7%) and non-loosening groups (n = 57, 63.3%). No significant differences in age, gender, length of fixation and preoperative BMD were found between both groups. The loosening group showed lower CT HU values in the vertebral body and screw trajectory than the non-loosening group. Screw trajectory HU (ST-HU) exhibited a higher AUC value than vertebral body HU (B-HU). The cutoff values of B-HU and ST-HU were 160 and 110 HUs, respectively. Conclusion: Three-dimensional pedicle screw trajectory HU values yields a stronger predictive value than vertebral body HU values and BMD and may provide more guidance for surgery. The risk of screw loosening is significantly increased at ST-HU <110 or B-HU <160 at L5 segment.