Risk factors for different cement distribution indexes for refracture of operated vertebrae and adjacent vertebrae after percutaneous vertebroplasty in patients with thoracolumbar compression fracture

Abstract

Background: Although most studies suggest that unilateral percutaneous vertebroplasty (PVP) increases the probability of secondary adjacent vertebral compression fractures, controversy remains the associated risk factors.

Objective: To explore the risk factors of bone cement distribution index (DI) for surgical and adjacent vertebrae after PVP.

Methods: According to the inclusion criteria, 329 patients who underwent PVP in the Spinal Surgery Department of the Sixth Affiliated Hospital of Xinjiang Medical University between January 2018 and May 2024 were then retrospectively screened, and the patients were divided into two groups based on whether refracture occurred during the follow-up period. The two groups were (1) fracture group (n = 47) and (2) no fracture group (n = 282). The following variables were reviewed in both groups: age, gender, body mass index (BMI), PVP segment, cement leakage, bone density, T, cement dose, smoking, drinking, hypertension, type 2 diabetes mellitus, COPD, cerebral hemorrhage, coronary heart disease, cement DI type, chronic kidney history and previous fracture (caused by osteoporosis). These factors were univariate analyzed and replaced P < 0.05 with binary Logistic analysis to explore the factors associated with postoperative vertebral recompression fracture.

Results: A total of 329 patients were included in this study to compare the parameters between the fracture and nonfracture groups. The results of univariate analysis showed that postoperative vertebral refracture was related to age, cement leakage, bone mineral density T value, history of hypertension, history of type 2 diabetes mellitus, the index type of bone cement distribution, and chronic kidney disease, and the difference was statistically significant (P < 0.05). By binary Logistic regression, Age (OR = 1.103, 95%CI:1.041-1.179, P = 0.002), leakage of bone cement (OR = 11.790,95%CI:4.942-30.637,P < 0.001) MD T value (OR = 5.716, 95%CI: 1.575-28.548, P = 0.016), history of hypertension (OR = 2.721, 95%CI:1.155-6.612, P = 0.023), history of chronic kidney disease (OR = 6.360, 95%CI:1.435-30.833, P = 0.017), type I bone cement DI [OR = 13.636, 95%CI: 3.126-98.477, P = 0.002] and type II cement DI [OR = 8.590, 95%CI:1.890-62.651, P = 0.012] was a risk factor for refracture of the operated and adjacent vertebrae.

Conclusion: The results showed that age, cement leakage, bone mineral density T value, history of hypertension, and cement DI type were risk factors for refracture of the operated vertebrae and adjacent vertebrae. The surgeon conducts a comprehensive evaluation of patients before surgery, which can more accurately estimate the probability of fracture again, and then provide a reference for the formulation of personalized treatment plan, so as to reduce the risk of fracture again in the future.

Keywords: Distribution index; Osteoporotic vertebral compression fracture; Percutaneous vertebroplasty; Risk factors; The adjacent vertebral body fracture; Vertebral body recompression.

Fig. 1

Type of bone cement DIx. The type of cement DI is categorized into Grade I (cemented mass filling, cement area less than 1/2 of the frontal and lateral X-ray vertebral body area), Grade II (cemented mass filling, cement area higher than 1/2 of the frontal and lateral X-ray vertebral body area), Grade III (spongy filling, cement area less than 1/2 of the frontal vertebral body area and more than 1/2 of the vertebral body area in lateral X-ray), Grade IV (cemented mass filling, two intervals, cement area higher than 1/2 of the frontal and lateral X-ray vertebral body area), and Grade V (cemented spongy filling, cement area higher than 1/2 of the frontal and lateral X-ray vertebral body area). ), Grade IV (cement mass filling, two intervals, cement area is more than 1/2 of the vertebral body area in frontal and lateral X-ray), Grade V (cement sponge filling, cement area is more than 1/2 of the vertebral body area in frontal and lateral X-ray)

Fig. 2

Flow chart of patient assignment.

Fig. 3

Forest plot of univariate and multivariate analysis of refracture occurring after PVP

Fig. 4

ROC curve predicting patient refracture after PVP. The curves of age (AUC=0.642), bone cement leakage (AUC=0.773), bone density (AUC=0.633), hypertension (AUC=0.654), chronic kidney disease (AUC=0.722), and Bone cement DI (AUC=0.665), and the curves were above the reference line, indicating that both indexes are risk factors for recurrent fractures of the operated vertebrae and neighboring vertebrae after PVP

Fig. 5

Multivariate analysis of refracture after PVP Forest plot. Bone cement DI type I and bone cement leakag were the most significant risk factors with ORs of 13.636 and 11.790, respectively, indicating that these two underlying conditions significantly increased the risk of events. Variables such as bone cement DI type II and chronic kidney disease also showed higher risk, especially bone cement DI II

Fig. 6

Postoperative x-ray sections of the thoracic and lumbar spine showed bone cement distribution

Fig. 7

Prediction accuracy of the refracture model after PVP. The vertical axis of the coordinate system is the TPR (True Positive Rate/Hit Rate/Recall Rate) with a maximum value of 1, and the horizontal axis is the FPR (False Positive Rate/False Positive Rate) with a maximum value of 1. Where the further the distance between the ROC curve and the baseline, the better the predictive performance of the model