Application Of Quantitative Parameters Of DKI And Dynamic Contrast-enhanced MRI In The Differential Diagnosis Of Parotid Tumor

Objective:To study the application of quantitative parameters of DKI and dynamic contrast-enhanced MRI in the differential diagnosis of parotid tumor. Method:Totally 47 patients with different pathological types of parotid tumors were enrolled from May 2014 to June 2015. All patients received routine MRI and DKI, DCE-MRI examinations with Siemens Aera 1.5T superconductive MRI scanner. After the scanning, the original images of DKI and DCE-MR were transferred for post-processing in a commercial workstation. The region of interest(ROI) was plotted around the tumor. The DKI quantitative parameters of MK,RK,KA in the lesion of the tumor through the DKE commercial software were collected, and the the original images of DCE-MR were transferred to Tissue 4D(a kind of Siemens Medical Systems).The pharmacokinetic modeling was based on the Tofts two-compartment model that allowed to analyze the Semi quantitative parameter of time intensity curve(TIC) and to calculate quantitative parameters of Ktrans,Ve,Kep. Meanwhile, the receiver operating characteristic(ROC) curves and the area under the curve(area under curve, AUC) were used to analyze the diagnostic efficiency of the DKI quantitative parameters of MK,RK,KA and the DCE-MRI quantitative parameters of Ktrans,Kep,Ve on differential diagnosis of parotid tumor, to determine the best diagnostic threshold value by according to the most Youden index. P<0.05 was considered statistically significant. All data were analyzed by SPSS18.0 statistical software. Result:(1) Among the 47 patients, there were 37 cases of benign parotid tumor and 10 cases of parotid malignant tumor. Among the 37 cases of benign parotid tumor, 28 cases showed hyperintense in T2 WI, 10 cases involved deep and shallow of parotid gland, 20 cases with varying degrees of cystic change, the same signs occured in the 10 cases of malignant parotid tumor were 7cases,5cases,6cases, there were no significant difference between the groups(P=0.965、0.317、0.918).(2) Among the 37 cases of benign parotid tumor, the cases of type A, B and D curves were 20,the cases of type C curve were 17. Among the 10 cases of malignant parotid tumor, the cases of type A, B and D curves were 2,the cases of type C curve were 8.The type A, B and D curves as the criteria of benign parotid tumor, the type C curve as the criteria of malignant parotid tumor, the difference between the group of benign and malignant parotid tumor was no statistically significant(P=0.119).(3)The MK,RK,KA values of benign parotid tumor(be exclusive of adenolymphoma) were lower than malignant parotid tumor, the difference between the groups was statistical significance(P<0.05). The MK,RK,KA values of mix tumor were obviously lower than adenolymphoma and malignant parotid tumor, the difference between the groups was statistically significant(P<0.05). The MK,RK,KA values all have high diagnostic efficiency in the differential diagnosis of mix tumor and adenolymphoma or mix tumor and malignant parotid tumor. The KA value of adenolymphoma was higher than malignant parotid tumor, the difference between the groups was statistically significant(P<0.05).(4)The Ktrans, Kep values of benign parotid tumor(be exclusive of adenolymphoma)were lower than malignant parotid tumor, while the value of Ve was higher than malignant parotid tumor,the difference between the groups was statistically significant(P<0.05).The Ktrans、Kep values of mix tumor were lower than that of malignant parotid tumor and adenolymphoma,the difference between the groups was statistically significant(P<0.05). While the Ve values of mix tumor was higher than malignant parotid tumor and adenolymphoma, the difference between the groups was statistically significant(P<0.05). The Ktrans、Ve、Kep values all have high diagnostic efficiency in the differential diagnosis of mix tumor and malignant parotid tumor or mix tumor and adenolymphoma. The Ktrans, Kep values of adenolymphoma were higher than malignant parotid tumor, the difference between the groups was statistically significantt. Conclusion:(1) The imaging features of conventional MRI and the type of TIC curve had certain overlap in the different pathological types of parotid tumors.(2) Quantitative parameters of DKI and DCE-MRI are helpful to the differential diagnosis of different pathological types of parotid tumors, especially for the differential diagnosis of mix tumor and malignant parotid tumor or mix tumor and adenolymphoma which were lack of MRI feature.