| [Objective]To evaluate the diagnostic value of diffusion weighted imaging(DWI) and susceptibility weighted imaging (SWI) in the grading of cerebral neuroepithelial tumors by analyzing the imaging features of cerebral neuroepithelial tumors on DWI and SWI.[Materials and methods]1.Clinical data71patients(46males and25females,2-80years old, mean age=32years) with cerebral neuroepithelial tumors were examined with routine MR sequences, DWI and SWI. The tumours were evaluated as WHO grade Ⅰ in7cases, grade Ⅱ in26cases, grade Ⅲ in8cases and grade Ⅳ in30cases, according to the2007WHO Classfication of Tumors of the Central Nervous System. All the cerebral neuroepithelial tumors were divided into low grade group (WHO grade Ⅰ and Ⅱ) and high grade group (WHO grade Ⅲ and Ⅳ) according to their biological behavior.2.Instrument and scanning methodAll the examinations were performed on GE Signa EXCITE HD3.0T MR scanner using an8-channel phase array head coil. All the patients underwent conventional axial MR T1-weighted imaging(T1WI), T2-weighted imaging(T2WI), fluid-attenuated inversion recovery(FLAIR), and contrast enhanced T1-weighted imaging(CE-T1WI).Diffusion weighted imaging(DWI) was performed with single shot spin echo-echo planar imaging sequence(SE-EPI), with TR=6200ms, TE=86.8ms, FOV240mm×240mm, matrix192×192,5.0mm thickness with1.5mm gap, b=0s/mm2and1000s/mm2.Susceptibility weighted imaging (SWI) was performed with high resolution3D FLASH sequence, with TR=43ms, TE=26ms,FA20°, FOV220mm X220mm, matrix448×384,2.0mm thickness with0mm gap.3.Data processing and analysis(1)DWI:The ADC maps and values were calculated on an ADW4.3workstation. The apparent diffusion coefficient(ADC) values and the exponent diffusion coefficient (EDC) values were measured at the tumorous solid portion, the peritumoral edematous region, the peritumoral white matter area and the corresponding opposite white matter. The ADC values and EDC values in our study represent averaged ADC and EDC of three regions of interest(ROI), each ROI was about30to50mm2. The relative apparent diffusion coefficient(rADC) values and the relative exponent diffusion coefficient(rEDC) values of the tumorous core, peritumoral edematous region and peritumoral white matter were calculated.(2)SWI:Phase images and magnitude images were obtained by using the SWI postprocessing software of the MR system and created a minimum-intensity projection settings(MinPs) by applying a minimum intensity projection to several adjacent magnitude images. The small vessels of the tumors were counted layer by layer. The ability of showing small vessels between SWI and conventional imaging techniques was compared. The signal intensity, distribution of venous structures and bleeding condition of the tumors were observed respectively on SWI. The blood products detected by SWI were measured in each slice and the total hemorrhage volume(cm3) within the tumor were calculated. The number of small vessels and the sectional volume of hemorrhage were compared within the two groups of cerebral neural epithelial tumors.4.Statistics analysis:SPSS13.0software package was applied to statistical analysis in this study. The measurement data were recorded as (mean零standard deviation). The statistically significant difference was set at P<0.05.DWI:Independent-Samples T Test was performed to compare the ADC values, EDC values, rADC values and rEDC values of the all measure areas between the two groups of cerebral neuroepithelial tumors. The ADC values, EDC values, rADC values and rEDC values of the tumorous solid portion, the peritumoral edematous and the peritumoral white matter area of the tumors were compared with general linear model repeated measures. If P<0.05,use LSD method for Post hoc Multiple comparisons.SWI:The differences between routine MR sequences(T1WI、T2WI、FLAIR、 CE-T1WI) and SWI in small vessels were analyzed by Wilcoxon test. Chi-square test was performed to analyze the differences of the signal intensity, distribution of venous structures and bleeding condition of the tumors between the two groups of cerebral neuroepithelial tumors. Mann-Whitney test for independent two samples was used to analyze the difference between high and low grade neuroepithelial tumors in the terms of small vessels and bleeding displayed in SWI. Spearman correlation analysis were performed to analyze the relationship of the hemorrhage volume and the number of the small vessels, and so as to the tumor grading.[Results] (1)ADC values and rADC values of tumorous solid portion in high grade neuroepithelial tumors were significantly lower than that in low grade neuroepithelial tumors(P<0.05). EDC values and rEDC values of tumourous solid portion in high grade neuroepithelial tumors were significantly higher than that in low grade neuroepithelial tumors(P<0.05).(2)There were no differences in ADC values, EDC values, rADC values and rEDC values of the peritumoral edematous and the peritumoral white matter area of the tumors between low and high grade groups (P>0.05).(3)Differences of ADC values, EDC values, rADC values and rEDC values between the tumorous solid portion and the tumorous peritumoral edematous region at low grade group are not significant(P>0.05), whereas significant at high grade group(P<0.05).(4)There were significant differences between SWI(7.59±4.41) and conventional MR sequences(1.59±1.86) in displaying small vessels(Z=-6.857, P<0.05), SWI was better than conventional MR sequences.(5)Most of the low grade neuroepithelial tumors showed hypersignal intensity, while most of the high grade neuroepithelial tumors displayed iso-or mild low signal intensity on SWI. There was statistical significance in the signal intensity between low and high grade groups(χ2=50.784, P<0.05).(6) The venous structures within low grade neuroepithelial tumors were sparse, but were abundant in high grade group. There was statistical significance in the distribution of venous structures between low and high grade groups (χ2=35.440, P<0.05).(7)The incidence of hemorrhage in high grade tumors(78.9%) was higher than that in low grade tumors(48.5%). There was statistical significance in the bleeding incidence between low and high grade groups(χ2=7.184, P<0.05). (8)There were statistical significances in the volume of blood products and the number of small vessels detected on SWI between low and high grade groups(P<0.05). Low grade tumors had less hemorrhage and small vessels.(9)The hemorrhage volume and the number of small vessels had postive correlation, both two index showed postive correlation with the tumor grading.[Conclusions](1)Combining DWI and SWI with conventional MR imaging can increase the accuracy of pre-operative imaging grading of cerebral neuroepithelial tumors.(2)The ADC values, EDC values, rADC values and rEDC values of the tumorous solid portion are valuable in the grading of the cerebral neuroepithelial tumors.(3)There were no differences in ADC values, EDC values, rADC values and rEDC values of the peritumoral edematous and the peritumoral white matter area of the tumors between low and high grade groups. The differences between the low and the high grade neuroepithelial tumors in peritumoral edematous and the peritumoral white matter region are still need to study and research.(4)The differences of ADC values, EDC values, rADC values and rEDC values between the tumorous solid portion and the tumorous peritumoral edematous region are significant at high grade group. DWIs are useful in determining the peritumoral neoplastic cell infiltration at high grade group.(5)SWI is more sensitive compare to conventional imaging techniques in showing small vessels in cerebral tumors, and SWI has better display effect compare to conventional sequences.(6)The signal intensity, distribution of venous structures and bleeding incidence of the tumors were significant different between low grade and high grade neuroepithelial tumors. The amount of small vessels and cerebral bleeds on SWI may help physician to facilitate the tumor grading in clinical practice. In addition, the hemorrhage volume and the number of the small vessels had postive correlation, both two index showed postive correlation with the tumor grading. |
PDF Full Text Request