Application Of Apparent Diffusion Coefficient In Glioma Postoperative Recurrence

Objective: To investigate the value of Apparent Diffusion Coefficient(ADC)in the diagnosis of postoperative recurrence of glioma.Materials and Methods: Retrospective analysis of postoperative MRI data of 46 patients with glioma confirmed by pathology in Qingda Affiliated Hospital from February 2016 to December 2017(30 males,16 females,aged 12-70 years old,The mean age of 44 years,according to reoperation pathology or follow-up confirmed the diagnosis was divided into recurrence(29 cases)and no recurrence two groups(17cases).All patients underwent MRI examination by GE HDXT3.0T superconducting magnetic resonance imaging equipment.The normal scan sequence was horizontal axis T1 FLAIR /FSE T2 WI /T2FLAIR;MRI enhanced scan: intravenous injection of0.2 mmol/kg gadopentetate injection(Gd-DTPA)The horizontal axis,coronal and sagittal three plane scans;DWI uses plane echo pulses(EPI)sequence,diffusion gradient of 3 vertical planes,b=0 and 1000s/mm2,matrix 160×160.layer thickness6.5 mm,layer spacing 1 mm,FOV 24 cm.The DWI raw data is generated by the Functool software and the ADC value is measured.The ADCmean value of the solid abnormal signal area and the ADCmean value of the normal parenchymal area of the contralateral mirror are measured respectively,and the relative ADC value is calculated(r ADCmean=patient ADC value/contralateral side)Mirror normal brain parenchymal ADC values).The regions of intrest(ROI)are placed in the real abnormal signal area as much as possible to avoid the area where the cysts become necrotic or hemorrhagic;and the small ROI is measured at the most significant extent of tumor diffusion limitation(ADCmin)and the diffusion limitation is the least.The ADC value of the significant(ADCmax).The ADC value of the selected ROI is directly measured from the post-processed image of the Functool software in the GE AW2.0 workstation.The ADC value of each patient was measured 3 times and averaged for comparison to reduce the error.The SPSS20.0 statistical software was used to analyze the mean and standard deviation of different ADC values in the recurrent group and the non-recurring group.All the measured data were tested to conform to the homogeneity of the normal distribution and the variance,and then independent.The sample t test compared the differences between the different ADC values for the glioma recurrence group and the non-recurrence group.The difference between low-grade glioma(9 cases)and high-grade glioma(20 cases)after recurrence of ADCmean was compared,and the statistical significance level was p<0.05.The ROC curve analysis was used to determine the optimal threshold value and the sensitivity and specificity of different ADC values for the diagnosis of postoperative recurrence.The area under the ROC curve(AUC)was calculated,and the AUC>0.5 was statistically significant.RESULTS:(1)The ADCmean value and r ADCmean value of the recurrent group were lower than those of the non-recurrent group and the r ADCmean value: the ADCmean values of the recurrent group and the non-recurrent group were 1.09±0.23×10-3mm2/s and 1.31±0.17 ×10-3mm2/s respectively,the difference between the two groups was statistically significant(P=0.001,P <0.05).The r ADCmean values of the recurrent group and the non-recurrent group were 1.14±0.34×10-3mm2/s and1.33±0.22×10-3mm2/s,respectively.The difference between the two groups was statistically significant(P=0.035,P<0.05).(2)The difference between ADCmin and ADCmax in the recurrent group and the non-recurrent group was statistically significant: ADCmin of the recurrent group was 1.01±0.15×10-3mm2/s,and ADCmax was 1.16±0.17×10-3mm2/ s,the ADCmin of the non-recurrent group was1.18±0.10×10-3mm2/s,and the ADCmax was 1.32±0.16×10-3mm2/s.The difference between the two groups was statistically significant(P=0.005,P=0.002),P <0.05).(3)ADCmean and r ADCmean values have a good diagnostic value for postoperative recurrence of glioma: ROC curve analysis shows that the optimal cutoff value of ADC value for diagnosis of glioma recurrence is 1.15×10-3mm2/s,sensitivity,The specificity and area under the curve(AUC)were 94.1%,58.6%,and 0.774,respectively.The optimal cut-off value for r ADCmean value in diagnosing glioma recurrence was 1.13×10-3mm2/s,sensitivity,specificity,AUC were 88.2% and 62.1%and 0.706,respectively.(4)ADCmin and ADCmax in the recurrent and non-recurrent groups have certain diagnostic value: ROC curve analysis shows that the optimal cut-off value of ADCmin glioma recurrence is 1.06×10-3mm2/s,sensitivity,specificity,The area under the curve(AUC)was 88.2%,69%,and 0.820,respectively.The optimal cut-off value of ADCmax glioma recurrence was 1.22×10-3mm2/s,sensitivity,specificity,AUC were 76.5% and 62.1% and 0.704,respectively.(5)There was no significant difference in ADCmean between low-grade glioma(9cases)and high-grade glioma(20 cases).The ADCmean was 1.18±0.14×10-3mm2/s and1.05±0.26×10-3mm2/s respectively.statistically no significant difference between the two.(P = 0.10,P > 0.05).Conclusions:(1)Different ROI type selection affects the mean ADC size of the lesion.(2)ADC value has certain value in the differential diagnosis of glioma recurrence or not.The mean values of ADCmin,ADCmean and ADCmax were statistically different between the two groups.Among them,ADCmean had the highest sensitivity and ADCmin had the highest specificity.The area under the curve(AUC)is the largest and the diagnostic efficiency is the highest.(3)The ADCmax of the recurrent group is similar to the ADCmin of the non-recurrent group,so although there is a certain diagnostic value,the sensitivity is low and the diagnostic efficiency is low.(4)ADCmean was not statistically significant after recurrence of low-grade glioma and high-grade glioma.There was no significant difference in ADC value after recurrence regardless of the grade of glioma.