Logistic regression analysis on risk factors of augmented vertebra recompression after percutaneous vertebral augmentation

To explore the high-risk factors of augmented vertebra recompression after percutaneous vertebral augmentation (PVA) in the treatment of osteoporotic vertebral compression fracture (OVCF) and analyze the correlation between these factors and augmented vertebra recompression after PVA. A retrospective analysis was conducted on 353 patients who received PVA for a single-segment osteoporotic vertebral compression fracture from January 2017 to December 2018 in our department according to the inclusion criteria. All cases meeting the inclusion and exclusion criteria were divided into two groups: 82 patients in the recompression group and 175 patients in the non-compression group. The following covariates were reviewed: age, gender, body mass index (BMI), injured vertebral segment, bone mineral density (BMD) during follow-up, intravertebral cleft (IVC) before operation, selection of surgical methods, unilateral or bilateral puncture, volume of bone cement injected, postoperative leakage of bone cement, distribution of bone cement, contact between the bone cement and the upper or lower endplates, and anterior height of injured vertebrae before operation, after surgery, and at the last follow-up. Univariate analysis was performed on these factors, and the statistically significant factors were substituted into the logistic regression model to analyze their correlation with the augmented vertebra recompression after PVA. A total of 257 patients from 353 patients were included in this study. The follow-up time was 12–24 months, with an average of 13.5 ± 0.9 months. All the operations were successfully completed, and the pain of patients was relieved obviously after PVA. Univariate analysis showed that in the early stage after PVA, the augmented vertebra recompression was correlated with BMD, surgical methods, volume of bone cement injected, preoperative IVC, contact between bone cement and the upper or lower endplates, and recovery of anterior column height. The difference was statistically significant (P < 0.05). Among them, multiple factors logistic regression elucidated that more injected cement (P < 0.001, OR = 0.558) and high BMD (P = 0.028, OR = 0.583) were negatively correlated with the augmented vertebra recompression after PVA, which meant protective factors (B < 0). Preoperative IVC (P < 0.001, OR = 3.252) and bone cement not in contact with upper or lower endplates (P = 0.006, OR = 2.504) were risk factors for the augmented vertebra recompression after PVA. The augmented vertebra recompression after PVP was significantly less than that of PKP (P = 0.007, OR = 0.337). The augmented vertebra recompression after PVA is due to the interaction of various factors, such as surgical methods, volume of bone cement injected, osteoporosis, preoperative IVC, and whether the bone cement is in contact with the upper or lower endplates.