The Application Research Of 3D-ASL Combined With Multi-b-value DWI In The Preoperative Grading Of Brain Glioma

Objective : To explore the clinical application value of three dimensional arterial spin labeling(3D-ASL) and multi b value(b=1000s/mm~2 ?b=2000s/mm~2? b=3000s/mm~2)diffusion weighted imaging(DWI) in the preoperative grading diagnosis of brain gliomas.Methods : In this study,we retrospectively collected 41 cases of pathologically confirmed brain glioma patients admitted to the neurosurgery department of Chenzhou First People's Hospital affiliated to Nanhua University from March 2018 to January 2019.According to different postoperative pathological results,patients were divided into low-grade glioma group(5 cases of WHO ?,14 cases of WHO ?)and high-grade glioma group(7 cases of WHO ?,15 cases of WHO ?).All patients were preoperatively examined by 3.0T MRI for conventional MRI,three kinds of b value of DWI,3D-ASL and enhanced MRI,through workstations on the original data post-processing,the cerebral blood flow(CBF) value,apparent diffusion coefficient(ADC1000,ADC2000,ADC3000) value,and CBF value and ADC value of the contralateral normal parenchyma area were measured in the tumor parenchyma,the relative cerebral blood flow(rCBF) value and relative apparent diffusion coefficient(rADC) value were obtained by dividing them separately.The above imaging data were compared with the pathological results of the tumor,SPSS23.0 statistical software was used for statistical analysis of the data.Results:1.The CBF value and rCBF value of high-grade glioma group were significantly higher than those of low-grade glioma group,and the difference between the two groups was statistically significant(P< 0.05).The pathological grade of glioma was highly positively correlated with CBF value and rCBF value of tumor parenchyma,and the correlation was higher with rCBF value(r=0.818,P<0.05).2.The ADC values and rADC values of the high-grade glioma obtained by the b-value DWI of the three groups were all lower than those of the low-grade glioma group,and the differences were statistically significant(P<0.05).3.With the increase of b value,the ADC value of the tumors in both high or low-grade gliomas decreased gradually,and the difference was statistically significant by anova(P<0.05).4.The correlation analysis showed that the grade of glioma was highly negatively correlated with the ADC value and rADC value of the three types of b values,and the higher the b value was,the better the correlation was.When b=3000s/mm(rADC3000,r=-0.823,P<0.05),the highest correlation was found.5.According to ROC curve analysis,among the ADC values and rADC values obtained by the three groups of b value DWI,the maximum area under the curve of rADC3000 was 0.976.When the rADC3000 threshold is 1.281,the sensitivity and specificity of differentiating high and low grade gliomas are 94.7% and 90.9% respectively;The area under the curve of rCBF(AUC=0.974)obtained by 3D-ASL was larger than that of CBF(AUC=0.950),when the threshold of rCBF was 1.846,the sensitivity and specificity of differentiating high and low grade gliomas were 89.5% and 95.5% respectively;When the two technologies were combined,the area under the curve of rCBF+rADC3000 was 0.986,and the sensitivity and specificity to distinguish high and low grade gliomaswere 94.7% and 95.5% respectively.6.The accuracy rate of conventional magnetic resonance in distinguishing high or low grade gliomas was 78.0%,and that by combining ASL and multi-b value DWI was 92.7%.The difference between the two methods was statistically significant by paired-sample chi-square test(P<0.05).Conclusion:Both 3D-ASL and DWI examination can better diagnose the preoperative grading of glioma;Among the three kinds of b-value DWI examinations,the rADC value of high b-value(b = 3000s/mm) DWI has the highest application value for preoperative grading diagnosis of glioma;Combined application of 3D-ASL and high b-value DWI can improve the accuracy of preoperative grading diagnosis of glioma,and both of them can be used as routine MRI examination methods for preoperative grading diagnosis of glioma for clinical application and promotion.