The Application Of MR Diffusion Kurtosis Imaging In Glioma Grading

Background and purposeGliomas are most common primary brain tumors in adults,which accounts about 40~50% of cases,with an incidence of 4 to 7 per 100000 annual.Gliomas with characteristics of high incidence and high recurrence rate,high mortality and low cure rate.Accurate glioma grading is essential for the choice of surgical approach,the development of radiotherapy and chemotherapy regimens,and the assessment of the patient's prognosis.At present,Magnetic Resonance Imaging become the preferred imaging method for clinical glioma grading because of its advantage of noninvasive,nonradiative,and high soft tissue resolution.Diffusion weighted imaging is widely used,for its unique advantage in the tumor diagnosis.DWI plays an irreplaceable role in the diagnosis and differential diagnosis of glioma.With the development of MR diffusion technology,from the development of diffusion weighted imaging to the current diffusion tensor imaging(DTI),diffusion kurtosis imaging(DKI)and other technical models.These newly developed technical models,with their respective characteristics in the glioma diagnosis and research process to play a role,and become the new research hotspot.The aim of this study was to evaluate the value of diffusion kurtosis imaging(DKI)in the diagnosis of glioma grading.Materials and methodsThis study was conducted during the November 2015 to September 2016.150 cases have been collected,and the number of patient was proved to be glioma by surgical pathology is 73.All of patients were scanned by SIMENS Prisima 3.0 T MR scanner,64-channel head and neck joint coil have been used.MR conventional sequence and DKI sequence have been scanned,30 min has been cost.During the scanning of MRI,patients were informed to keep body stationary and relax.DKI was scanned using a spin-echo echo-planar imaging sequence and the parameters were: FOV = 220×220 mm2,slice thickness = 5 mm,slices = 20,TR/TE = 3500/78 ms.DKE post-processing software was used to calculated diffusion parameters — fractional anisotropy(FA),mean kurtosis(MK)and mean diffusivity(MD)— were calculated in the solid part of glioma.Regions of interest(ROIs)were manually drawn around the solid part of the tumor.The MD value,FA value and MK value of each grade glioma were compared by one-way ANOVA and LSD test using SPSS(21.0,IBM)statistical software.Draw the ROC curve,calculate the area under the curve(AUC),and predict the threshold for the valid parameter.The histogram was used to analyze the distribution of sex and age in gliomas at all levels.And the correlation between each grade glioma and each group was analyzed by Spearman rank correlation.Results73 glioma patients account all participants about 48.67%.The age range of glioma patients was 14-75 years old,and the average age of them was 48.6±13.4 years old,and high incidence of age from 46.5 to 51.5 years old,which of them include 41 male and 32 female.This group of gliomas in 58.9% of the tumor involving more than two lobes,69.3% of patients had symptoms of cranial hypertension,59.1% of patients with cranial hypertension had more than two brain lobes were involved.Patients with epilepsy symptoms,whose tumor were mainly located in the temporal lobe,frontal and parietal lobes.People with sensory disturbances,whose tumor are mainly located in basal ganglia,thalamus and frontal lobe.There were 19.2%(14/73)of glioma patients with not only one clinical symptom.When the gliomas were divided into three groups at grade II,? and ?,there was no significant difference of MD and FA in the glioma group(P> 0.05).But MK value can indentify ?(n=21),?(n=13)and ?(n=39)grade glioma.MK is significant higher in ? grade glioma than that in ?grade glioma(P <0.001).Taking 0.587 as cut-off value for classified?and ? grade glioma,the sensitibity was 1.0 and specificity was 95.2%,and the AUC was 0.996.When it comes to ? and ? grade glioma,MK was significant higher in ?grade glioma than ? grade glioma,and when cut-off value was 0.838,the sensitibity was 92.3% and specificity was 92.3%,and the AUC was 0.953.The statistical correlation analysis proved,there was a certain correlation between age and grade(R=0.463,P<0.001),and correlation between age and MK value(R=0.367,P=0.001),and correlation between age and FA value(R=-0.231,P= 0.049).There was no significant differences between male and female in FA value,MD value and MK value(P>0.05).ConclusionsThe incidence of glioma in this group has gender differences and age differences.MK value is a helpful parameter for grading glioma.