The correlation between self-Hounsfield units and adjacent vertebral fracture after percutaneous vertebral augmentation: a retrospective cohort study

Quant Imaging Med Surg. 2024 Mar 15;14(3):2405-2414. doi: 10.21037/qims-23-1233. Epub 2024 Mar 4.

Abstract

Background: Adjacent vertebral fracture (AVF) represents a prevalent and challenging complication after percutaneous vertebral augmentation (PVA) treatment for osteoporosis vertebral compressive fracture (OVCF). Lower bone mineral density (BMD) and intervertebral leakage are reportedly independent risk factors for AVF. Vertebral Hounsfield units (HU) measured from computed tomography (CT) scans can evaluate bone quality. This study sought to explore the risk factors associated with AVF and analyze the relationship between AVF and the Hounsfield units of adjacent vertebrae (self-HU) following PVA.

Methods: In this retrospective cohort study, we included consecutive OVCF patients who presented to Xuzhou Central Hospital in Jiangsu Province, China from 1 January 2016, to 31 December 2019 for PVA treatment. Clinical and imaging data were collected, and baseline data were recorded. Patients were divided into the AVF group and the no-AVF group based on the presence of AVF during follow-up. Patients in the AVF group were further subdivided into the leakage group and the no-leakage group according to the presence of intervertebral leakage. Age, body mass index (BMI), fracture location, prior fracture, self-HU, and intervertebral leakage were included in univariate logistic regression analysis. Variables with a P value of less than 0.1 were then included in multivariate logistic regression analysis to determine the risk factors for AVF. Kaplan-Meier curves were plotted to assess the effect of intervertebral leakage on AVF using a log-rank test.

Results: A total of 460 patients were included in this study and followed up for an average of 50.9 months (range, 37-83 months). Among them, 82 cases (17.83%) developed AVF and were included in the AVF group. Multivariate logistic regression analysis showed that lower self-HU [odds ratio (OR) =0.972, 95% confidence interval (CI): 0.959-0.985, P<0.001] and intervertebral leakage (OR =2.618, 95% CI: 1.415-4.844, P=0.002) were risk factors for AVF following PVA. In the AVF group, 29 patients (35.37%) with intervertebral leakage were included in the leakage group. Patients in the leakage group had a shorter time to AVF (22.07±13.83 vs. 31.42±18.73, P=0.021) and higher self-HU (78.05±16.41 vs. 64.23±20.49, P=0.002) than those in the no-leakage group. Kaplan-Meier curves showed that the fracture-free time was shorter in the leakage group compared to the no-leakage group (log-rank test, P=0.014).

Conclusions: Lower self-HU and intervertebral leakage are risk factors for AVF, and higher self-HU may lead to AVF when intervertebral leakage is present.

Keywords: Hounsfield units (HU); Osteoporosis vertebral compressive fracture (OVCF); adjacent vertebral fracture (AVF); computed tomography scan (CT scan); percutaneous vertebrae augmentation.