The Effect Of Ventricular Opening During Surgical Operation Of Glioblastoma On Brain Tumor Patients:An Observational Clinical Study

Object:Ventricular opening during glioblastoma(GBM)resection is controversial.There is insufficient evidence that open ventricles have an adverse effect on their prognosis.We retrospectively evaluated the effect of ventriculectomy on overall survival(OS),hydrocephalus,cerebrospinal fluid leakage,and postoperative complications in patients undergoing GBM surgery,further analyzed their relationship with prognosis,and provided medical evidence for clinical practice,and to provide initial support for GBM surgical patients to choose a more scientific surgical treatment.Methods:Retrospective analyzing of patients who underwent primary GBM resection between Jan 2010 and Mar 2017 were made.And the corresponding MRI and CT imaging data and clinical data such as survival time(OS),hydrocephalus or cerebrospinal fluid leakage,postponing follow-up treatment,postoperative central nervous system complications,postoperative hospital stay,etc were extracted.And the influencing factors were:age at diagnosis,excision range,tumor size,Karnofsky function score,postoperative adjuvant therapy and so on.We use Excel software to input and organize data,establish database,and use SPSS 15.0 software to do statistical analysis.This study used univariate and multivariate analysis separately to discuss the relationship between ventricular opening and prognosis in GBM resection.Survival analysis in univariate analysis using Kaplan-Meier method,the two groups were compared using the?~2 test(two-sided test,test level?=0.05).Multivariate analysis of survival analysis using multivariate Cox regression method,multivariate analysis of dependent variables for binary variables using Logistic regression,and dependent variables for continuous variables using multiple linear regression methods.First,we use the Kaplan-Meier method and the?~2 test to screen out the possible influencing factors among the various characteristics of the study population.Second,the results of the screening and the existing reports were compared to determine the age,resection range,tumor size,Karnofsky score as a covariate and included in the multivariate Cox regression model of the"open ventricular end-OS relationship"for correction analysis and RR,95%CI and P values were calculated.Then,whether or not the ventricle was opened,whether hydrocephalus and cerebrospinal fluid leakage occurred,and the possible influencing factors were included.Univariate and multivariate logistic regression analysis was performed,and OR,95%CI,and P value were calculated.Results:519 eligible cases were included in this study,of which 212 patients had open ventricles and 307 had intact ventricles(no ventricles were open).OS in the two groups was 12.1 months(95%CI 10.9 to 15.5)and 15.7 months(95%CI 11.1 to20.8),but there was no significant difference between the two groups(P=0.152).Multivariate Cox regression Shown(n=134),ventricles open(RR=1.04,P=0.692)and extent of tumor erosion to the ventricles(mild VS no:RR=0.94,P=0.139;extensive vs no:RR=1.20,P=0.360)were not independent risk factors of OS,but older than 55 years old(RR=3.74,P<0.001),KPS<70(RR=2.41,P=0.013),tumor volume>32cm~3(RR=1.63,P=0.005)and subtotal cut(RR=1.39,P=0.036)were independent risk factors for OS.Univariate logistic regression analysis revealed that the extent of resection,tumor size,lateral lesions,and extent of erosion of the ventricles were all related to ventricles(P<0.05).Intraoperative opening of ventricles was a risk factor for hydrocephalus.The results were statistically significant(OR=21.7,95%CI 5.62~83.8 P=0.006).The other two statistically significant risk factors were larger tumor volumes(OR=7.60,95%CI 2.07 to 27.9 P=0.014)and tumor eroded ventricles(mild VS no:OR=12.7,P=0.041;extensive VS no:OR=15.8,P=0.028).Ventricular opening was the only significant risk factor for cerebrospinal fluid leakage(OR=5.29,95%CI 2.04~13.7,P=0.021),but the incidence of neurological deficits was not significantly changed(P=0.227).Multivariate logistic regression analysis showed that only subtotal resection(OR=5.35,95%CI 2.16 to 13.3,P<0.001)and adhesion to the ventricles(mild vs no:OR=28.1,95%CI 17.0 to 46.4,P<0.001;extensive vs no:OR=52.8,95%CI 26.4~105.6,P<0.001)was an independent risk factor for ventricular opening.Open ventricle,large tumor size and erosion of the ventricles were not independent risk factors for postoperative hydrocephalus(P>0.05).Conclusion:1.Kaplan-Meier analysis showed that there was no statistical difference in median OS between the ventricle open group and the unopened ventricles,but the extent of surgical resection was an independent predictor of OS in GBM patients,thus clinically achieving greater tumor resection rate(R0 resection),open ventricle could be considered.2.Surgical excision range,tumor size,lesion side and the tumor erosion of the ventricle--these factors are related with the opening of the ventricle,reminder that whether or not to open the ventricle in the clinic,the size of the tumor,the lateral side of the lesion,and the erosion of the ventricle should be considered.3.Ventricular opening is not independent risk factor of postoperative hydrocephalus,neurological deficits,prolonged hospital stay,and delayed adjuvant treatment,reminder that ventricle open surgery is safe,which does not increase the incidence of surgical complications such as hydrocephalus,neurological deficit,does not affect postoperative adjuvant therapy in patients with GBM surgery.4.Ventricular opening is an independent risk factor for postoperative cerebrospinal fluid leakage.Prompt ventricle opening may increase the incidence of postoperative cerebrospinal fluid leakage,which should be paid attention to in clinical practice.