Evaluation of Prognosis of Convexity Meningioma by Using MRI Imaging Parameters
Gao Xiuzhong,Wang Chaohu,Kuang Huan,Zhang Shichao,Liu Yi,Pan Jun,Qi Songtao
DOI: https://doi.org/10.3760/cma.j.issn.1001-2346.2016.02.011
2016-01-01
Abstract:Objectives To investigate the correlation of preoperative MRI imaging parameters and pathological grades and prognosis of convexity meningioma and to initially establish the imaging scoring method for predicting of the prognosis and biological behavior of meningiomas.Methods From January 2003 to December 2006,the preoperative imaging parameters of 246 patients underwent resection of convexity meningioma (Simpson grade Ⅰ) at the Department of Neurosurgery,Nanfang Hospital,Southern Medical University were analyzed retrospectively.The preoperative imaging parameters included the signal intensity of diffusion tensor imaging (DTI),T1 enhanced images,subarachnoid layer on T2-weighted images,peritumoral edema on T2-weighted images and tumor shape.The relationship between imaging features of tumors and clinical pathological grades was detected by the univariate analysis.Kaplan-Meier survival analysis and multivariate logistic regression analysis were used to evaluate whether the imaging parameters could predict the occurrence of high-grade meningiomas.According to preoperative imaging parameters,the patients were graded and grouped.Cox regression model was used to analyze the relationship between the imaging scores in patients with convexity meningioma and the overall survival (OS) and progression-free survival (PFS).The credibility of preoperative imaging scores was analyzed by the Kappa coefficient test.Results Univariate analysis results indicated that there were significant differences in diffusion tensor imaging (DWI) signal intensity,T1 enhancement images,arachnoid layer on T2 weighted imaging,peritumoral edema on T2-weighted images,and tumor shape in different grades of convexity meningiomas (all P ＜ 0.05).Multivariate logistic regression analysis indicated that DWI hyperintensity is the most important independent predictor of high-grade meningiomas (P ＜ 0.01,OR,17,95％ CI 5.8-47.6),others were disruption of arachnoid layer (P ＜0.01,OR,14,95％ CI 4.3-42.3),peritumoral edema (P ＜0.01,OR,9,95％ CI 3.1-26.8),T1 heterogeneous enhancement (P ＜0.01,OR,6.1,95％ CI2.2-17.1),and tumor shape (P ＜0.01,OR,6.1,95％ CI2.1-17.5).Compared with WHO pathological grades (Ⅰ,Ⅱ,and Ⅲ),the imaging grades (1,2,and 3) had a certain homogeneity (Kappa value,0.511,P ＜ 0.01).Cox regression model analysis showed that age (P ＜ 0.001,OR,1.039,95％ CI1.017-0.062),WHO pathological grade (P=0.017,OR,3.014,95％ CI1.217-7.465),and preoperative imaging score groups (2-3 group,P ＜ 0.01,OR,6.194,95％ CI 1.956-19.610;4-5 group,P ＜ 0.01,OR,21.658,95％ CI 5.701-82.273) could be used as the independent predictors of affecting the prognosis of patients.Conclusions Preoperative imaging grade has a certain degree of consistency with pathological grade,and can predict the prognosis in patients with convexity meningioma.For patients with pathology indicated WHO grade Ⅰ,but the imaging score 2-3 may need more closefollow-up and more radical resection in case of permission.