Study Of The Value Of DKI In Grading Diagnosis Of Glioma And Its Correlation With Ki-67 Labeling Index

Objective:To investigate the value of magnetic resonance diffusion kurtosis imaging?DKI?in grading diagnosis of glioma and its correlation with Ki-67 labeling index?Ki-67LI?,and to explore whether the tumor parenchymal parameters should be normalized with contralateral normal appearing white matter?NAWMc?when DKI is used to grading diagnosis of glioma.Methods:A total of 54 cases of glioma confirmed by surgery and pathology from December 2015 to October 2018 were collected from the Affiliated Hospital of Zunyi Medical University.There were 31 males and 23 females.The age ranged from 9 to 75years,with an average of 43.7±15.4 years.All patients underwent GE Signa HDxt 3.0T magnetic resonance imaging before surgery,followed by routine plain scan,DKI examination and T₁WI enhanced scan.Use the GE Advantage Work Station 4.6workstation Functool software to map the Region of Interest?ROI?for each parameter map,and obtain the mean kurtosis?MK?,axial kurtosis?AK?,radial kurtosis?RK?,mean diffusivity?MD?,fractional anisotropy?FA?value.The contralateral normal appearing white matter?NAWMc?was used as a reference to obtain standardized normalized mean kurtosis?nMK?,normalized axial kurtosis?nAK?,normalized radial kurtosis?nRK?,normalized mean diffusivity?nMD?,and normalized fractional anisotropy?nFA?values.The expression of Ki-67 was confirmed by immunohistochemistry for all cases.The independent sample t test was used to analyze the difference between each parameter value of high and low grade glioma.One-way analysis of variance and Kruska-wallis test were used to compare the values of parameters of grade II,III and IV glioma.Further,LSD test and Games-Howell test were used to compare the two groups.The receiver operating characteristic?ROC?curve was drawn on the statistically significant parameters.Finally,Spearman correlation was used to analyze the correlation between each parameter and its normalized value and Ki-67 LI.p<0.05 was considered statistically significant.Result:The MK,AK,and RK values of high-grade gliomas?HGGs?were higher than those of low-grade gliomas?LGGs?,and the MD values was lower than those of LGGs,and the difference was statistically significant?p<0.001?;There was no significant difference in FA values between high and low grade gliomas?p=0.117?.After normalization with NAWMc,nMK,nAK,nRK,nMD values were statistically significant between high and low grade gliomas?p<0.05?;There was no significant difference in nFA values between high and low grade gliomas?p=0.05?.MK value has the best grading diagnostic efficiency for high and low grade gliomas,area under the curve?AUC?is 0.965,cut-off value is 0.563,sensitivity and specificity are 90.6%,90.9%,95%CI is 0.920-1.000,p<0.001;For normalized values,nMK has the best diagnostic performance,with an AUC of 0.902 and a cut-off value of 0.626.Sensitivity and specificity are 85%,95.5%,and 95%CI is 0.819-0.985,p<0.001.MK,AK,RK values were statistically different between grade II and grade III,grade III and grade IV gliomas?p<0.05?;The FA value was only statistically different between grade III and IV gliomas?p<0.001?;MD values were only statistically significant between grade II and III gliomas?p=0.033?.After normalization with NAWMc,nMK,nAK,and nRK were only statistically different between grade II and III gliomas?p<0.05?;There was no significant differences in nFA and nMD between grade II and grade III,grade III and grade IV gliomas?p>0.05?.MK values were identified as the highest AUC of grade II and grade III,grade III and grade IV gliomas,0.927,0.906?p value is 0.001,respectively?;After normalization with NAWMc,nRK identified the largest AUC for grade II and III gliomas,which was 0.889?p=0.003?.MK,AK,RK and their normalized values,FA and Ki-67 LI were positively correlated?p<0.05?,among which MK values were the best?r_s were 0.712,p<0.001?;MD and its normalized values were negatively correlated with Ki-67 LI?r_s were-0.399,-0.372,p<0.01,respectively?;nFA had no correlation with Ki-67 LI?r_s=0.268,p=0.05?.Conclusion:DKI can non-invasively evaluate the pathological grade of glioma and predict the expression of Ki-67,MK is an imaging index with good predictive value for grading diagnosis and cell proliferation of glioma.DKI parameters do not necessarily need to be normalized to the NAWMc for classifying glioma.