Abstract: BACKGROUND: The current research on bone cement leakage in the paravertebral vein mainly focuses on the puncture site, vertebral body blood vessel distribution and bone cement/vertebral body volume ratio. However, there have been no relevant reports about the effect of puncture-side bone cement/vertebral body volume ratio on cement leakage in the paravertebral vein in and outside China. 
OBJECTIVE: To explore the influencing factors of bone cement paravertebral vein leakage in the treatment of osteoporotic compression fractures by vertebroplasty, and to verify the correlation between puncture-side bone cement/vertebral body volume ratio and bone cement paravertebral vein leakage. 
METHODS: Data of 357 patients (425 vertebral bodies) with osteoporotic vertebral compression fractures treated from August 2018 to April 2020 in Zigong Fourth People’s Hospital were retrospectively analyzed. Postoperative X-ray and CT data were observed to determine whether the vertebral body has paravertebral vein leakage. Mimics software was used to import the patient’s postoperative CT data into the software. The 3D reconstruction function was applied to calculate the bone cement/vertebral body volume ratio, puncture-side bone cement/vertebral body volume ratio. The sex, age, body mass index, bone density, bone cement volume, bone cement volume of the puncture side, vertebral volume, bone cement/vertebral volume ratio, puncture-side bone cement/vertebral volume ratio were compared between the leaked group and the non-leaked group. Logistic regression analysis was used to compare the correlation between the factors with statistical significance between the two groups and the leakage of paravertebral veins. A ROC curve was drawn for the diagnosis of paravertebral vein leakage with statistically significant factors. 
RESULTS AND CONCLUSION: (1) There were 42 cases of paravertebral vein leakage (43 vertebral bodies). There were 315 patients (382 vertebral bodies) without paravertebral vein leakage. (2) Univariate analysis showed that the differences in sex, bone density, bone cement volume of the puncture side, and puncture-side bone cement/vertebral volume ratio of the leaked and non-leaked groups were statistically significant (P < 0.05). (3) Logistic regression analysis showed that there was a correlation between sex, puncture-side bone cement volume, puncture-side bone cement/vertebral body volume ratio and paravertebral vein leakage (P < 0.05). (4) The ROC curve showed that the area of the punctured bone cement/vertebral body volume ratio for the diagnosis of paravertebral vein leakage was greater than 0.6, and P < 0.05, which had certain diagnostic value. The best cut-off point for the diagnosis of paravertebral vein leakage with puncture side bone cement/vertebral volume ratio was 13.33%, sensitivity was 81.4%, and specificity was 26.2%. (5) The sex, bone density, puncture-side bone cement volume, puncture-side bone cement/vertebral body volume ratio are risk factors for vertebroplasty treatment of paravertebral vein leakage in osteoporotic vertebral compression fractures, of which puncture-side bone cement/vertebral body volume ratio with paravertebral venous leakage is closely related, and the optimal puncture-side bone cement/vertebral body volume ratio is 13.33%. When the puncture-side bone cement/vertebral body volume ratio exceeds the optimal value, the risk of paravertebral vein leakage is significantly increased.

Key words: vertebroplasty, paravertebral veins leakage of bone cement, puncture side, bone cement volume, vertebral body volume