Related Experiment Video
Updated: May 23, 2025
Author Spotlight: Improving Percutaneous Vertebroplasty Surgical Accuracy and Efficiency Through Advanced Puncture Techniques
Published on: August 9, 2024
CT-based vertebral three-dimensional Hounsfield unit can predict the new vertebral fracture after percutaneous
Rongkun Chang1,2, Feng Chang3, Yang You1,2
1People's Hospital of Ningxia Hui Autonomous Region, Ningxia Medical University, No. 301, Zhengyuan North Street, Jinfeng District, Yinchuan City, 750001, Ningxia Hui Autonomous Region, China.
Background:
Vertebral Hounsfield unit (HU) were regarded as a new way to predict fragility fracture. However, HU values were measured in a single plane, which is not accurate for the entire vertebral body. This study aimed to create a new CT-based metric for assessing bone mineral density, three-dimensional Hounsfield unit value (3D-HU), and to evaluate its effect in independently predicting new vertebral fracture (NVF) after percutaneous vertebral augmentation (PVA) in postmenopausal women.
Methods:
This study reviewed female patients with osteoporotic vertebral compression fracture (OVCF) who were treated at our hospital. Patients were divided into NVF and control groups according to whether they had NVF. 3D-HU of the L1-4 vertebrae was measured using preoperative computed tomography (CT) scanning of the lumbar spine. Demographics, procedure-related data, and radiological data were collected. Pearson correlation test was used to determine the correlation between 3D-HU and BMD T-score. The independent risk factors of NVF were determined by multivariate logistic regression analyses. Receiver operating characteristic curve (ROC) was used to evaluate the predictive performance of 3D-HU.
Results:
This study involved 349 postmenopausal women who were treated with PVA between January 2017 and August 2022. Among them, 61 people suffered the NVF following PVA. The mean 3D-HU was 40.64 ± 22.43 in the NVF group and 79.93 ± 25.69 in the without NVF group (p < 0.001). Multivariate analysis showed that lower 3D-HU (OR = 0.927; 95%CI = 0.906-0.945; p < 0.001) was the only independent predictor of NVF following PVA. The predictive accuracy of 3D-HU was 87.7%, which was higher than that of the HU value (82.3%), and it was highly positively correlated with BMD T-score (r = 0.628, p < 0.001).
Conclusions:
Lower 3D-HU was significantly associated with NVF following PVA in postmenopausal women. In addition, vertebral 3D-HU had better predictive power than HU values. 3D-HU assessment prior to PVA may provide insight into a patient' s risk for NVF.
