RESUMEN
OBJECTIVES: To explore the influencing factors of bone cement leakage in the paravertebral vein after vertebroplasty for the treatment of osteoporotic vertebral compression fractures (OVCFs) and to determine the correlation between the puncture-side bone cement/vertebral body volume ratio and bone cement leakage in the paravertebral vein. METHODS: This was a retrospective analysis of 495 patients (585 vertebral bodies) with OVCFs treated from August 2018 to May 2021 in our hospital. The patients' postoperative CT data were imported into Mimics software, and the three-dimensional(3D) reconstruction function was used to calculate the bone cement volume (BCV), puncture-side bone cement volume (PSBCV), and vertebral body volume (VBV); the bone cement/vertebral body volume ratio (BCV/VCV%) and puncture-side bone cement/vertebral body volume ratio (PSBCV/VCV%) were additionally calculated. Sex, Age, Body mass index(BMI), Bone density, BCV, PSBCV, VBV, BCV/VCV%, and PSBCV/VCV were compared between the leakage group and the non-leakage group. Logistic regression analysis was used to assess the correlations between the factors that statistically significantly differed between the two groups and the presence of leakage in the paravertebral veins. A receiver operating characteristic (ROC) curve was used to determine the diagnostic value of the PSBCV/VCV% and to obtain the optional cut-off value. RESULTS: A total of 102 males and 393 females with an average age of 72.89 (52 ~ 93) years were included in our study. There were 57 cases of cement leakage (59 vertebral bodies) in the paravertebral vein. There were 438 patients (526 vertebral bodies) without paravertebral cement leakage. Univariate analysis showed that the differences in sex, bone density, PSBCV, and PSBCV/VCV% between the two groups were statistically significant (P < 0.05). Logistic regression analysis showed that there were correlations between sex, bone density, and PSBCV/VCV% and the presence of paravertebral cement leakage (P < 0.05). The ROC curve showed that the area under the curve of the PSBCV/VCV% for the diagnosis of cement leakage in the paravertebral vein was greater than 0.65, and P < 0.05, indicating a diagnostic value. The best cut-off point for the diagnosis of paravertebral cement leakage with the PSBCV/VCV% was 13.68%, with a sensitivity of 84.7% and specificity of 37.8%. CONCLUSION: Sex, bone density, and PSBCV/VCV% are risk factors for cement leakage in the paravertebral veins after vertebroplasty for the treatment of OVCFs; the PSBCV/VCV% is strongly associated with paravertebral venous leakage, and the optimal PSBCV/VCV% is 13.68%. When the PSBCV/VCV% exceeds the optimal value, the risk of cement leakage in the paravertebral vein becomes significantly increased.