| Part I Characterization of 1H-MRS in the peritumoral region of supretentorial gliomasObjective:To investigate the metabolism changes in different peritumoral regions of various grades of gliomas for the assessment of the usefulness of 1H-MRS in evaluating the peritumoral region of supretentorial gliomas. Methods:1H-MRS examination is preoperatively performed in 51 patients with pathologically proved gliomas that include 17 low-grade gliomas (LGGs) and 34 high-grade gliomas (HGGs). The metabolite ratios of Cho/Cr, NAA/Cho, NAA/Cr are measured from spectral maps in the tumor solid portion, as well as in proximal peritumoral, distal peritumoral, and contralateral normal brain regions. The ratios of Cho/nCr, Cho/nCho, NAA/nCr, NAA/nNAA are calculated and normalized by the metabolite values in the contralateral normal brain region. SPSS is employed to analyze the differences in the metabolite ratios between LGGs and HGGs. Cutoff values of those ratios in the proximal peritumoral region for distinguishing WHO Ⅱ gliomas from WHO IV gliomas are determined by receiver operating characteristic (ROC) analysis. Results: For HGGs, the proximal peritumoral Cho/Cr、NAA/Cho、Cho/nCho、Cho/nCr、 NAA/nCr、NAA/nNAA ratios are statistically different from that at the distal peritumoral region (P<0.05); Only the distal peritumoral NAA/Cr is significantly different from that at the contralateral normal brain region (P<0.01). For LGGs, no significant differences are found in the proximal and distal peritumoral regions in the comparison with the contralateral region. The proximal peritumoral Cho/Cr, NAA/Cho, NAA/nCr, NAA/nNAA ratios in HGGs are also statistically different from those of LGGs, but no significant differences are found in the distal peritumoral regions of the two groups. ROC analysis demonstrated cutoff values of 1.012,1.219, 0.723 for proximal peritumoral NAA/Cho, NAA/nCr, NAA/nNAA ratios may provide high sensitivity and specificity for the discrimination between WHO II and WHO IV gliomas. Conclusion:The different metabolite ratios in the different peritumoral regions of various grads of gliomas suggest that 1H-MRS may be used for glioma grading and detecting infiltration of tumor cells in the peritumoral region.Part II Characterization of DTI quantitative metrics in the peritumoral region of supretentorial gliomasObjective:To investigate DTI quantitative metrics in different peritumoral regions of various grades of gliomas for assessing the usefulness of DTI in evaluating the peritumoral region of supretentorial gliomas.Methods:DTI examination is preoperatively performed in 69 patients with pathologically proved gliomas that include 23 low-grade gliomas (LGGs) and 46 high-grade gliomas (HGGs).The ADC and FA values are measured from the corresponding maps in the tumor solid portion, as well as in proximal peritumoral, distal peritumoral, and contralateral normal brain regions.The relative ADC and FA ratios are calculated and normalized by the corresponding values in the contralateral normal brain region. We compare FA and ADC values/ratios of peritumoral regions between LGGs and HGGs and investigate the relationships between the relative ADC and FA ratios of tumors, using t test and the Spearman correlation analysis. Results:For HGGs, the proximal peritumoral ADC and FA values/ratios are statistically different from that at the distal peritumoral region (P<0.05); For LGGs, no significant differences are found. The proximal and distal peritumoral ADC and FA values/ratios in HGGs are also statistically different from that of LGGs. Strong correlation is found between the relative ADC and FA ratios of tumors. Conclusions:The statistically different ADC and FA ratios in the different peritumoral regions of low-and high-grade gliomas suggest that DTI quantitative metrics, complemental to conventional MRI and 1H-MRS, may be useful for glioma grading and detecting infiltration of tumor cells in the peritumoral region. Part Ⅲ Characterization of T2*-PWI in the peritumoral region of supretentorial gliomasObjective:To investigate PWI in different peritumoral regions of various grades of gliomas for assessing the usefulness of PWI in evaluating the peritumoral region of supretentorial gliomas. Methods:PWI examination is preoperatively performed in 52 patients with pathologically proved gliomas that include 16 low-grade gliomas (LGGs) and 36 high-grade gliomas (HGGs). The rCBV and rCBF values are measured from the corresponding maps in the tumor solid portion, as well as in proximal peritumoral, distal peritumoral, and contralateral normal brain regions. The relative rCBV and rCBF ratios are calculated and normalized by the corresponding values in the contralateral normal brain region.We compare rCBV and rCBF values/ratios of peritumoral regions between LGGs and HGGs and investigate the relationship between the relative rCBV and rCBF ratios of tumors using t test and the Spearman correlation analysis. Results:For HGGs, the proximal peritumoral rCBV and rCBF values/ratios are statistically different from those at the distal peritumoral region (P<0.05); For LGGs, no differences are found between those two regions. The proximal peritumoral rCBV and rCBF values/ratios in HGGs are also statistically different from those of LGGs. Strong correlation is found between the relative rCBV and rCBF ratios of tumors. Conclusions:The statistically different rCBV and rCBF ratios in the different peritumoral regions of low-and high-grade gliomas suggest that PWI quantitative metrics may be useful for glioma grading and detecting infiltration of tumor cells in the peritumoral region and that it may be complemental to conventional MRI and 1H-MRS.Part IV Multiparametrics of fMRI in supretentorial gliomas:correlation with histopathology based on stereotactic biopsiesObjective:To evaluate the relationship between the multiparametrics acquired from fMRI (1H-MRS、DTI、PWI) and the histopathological features in different biopsy sites. Methods:Six patients with pathologically proved gliomas (two WHO Ⅱ gliomas and four gliomblastomas) receive the scanning of 1H-MRS, DTI, PWI and 3D anatomical MR navigation sequences. Biopsy trajectories are determined by both neuroradiologist and neurosurgery. The ROIs of spectral maps involved by the biopsy trajectories are marked and coregistered on the 3D anatomical MR imaging for intraoperative navigation. Biopsies are performed by one neurosurgery according to the marked 3D anatomical MR imaging by using a stereotactic needle tracked by a navigation system. All parametrics were measured and recorded from spectral maps, ADC, FA, rCBV maps in the biopsy sites and the corresponding contralateral normal brain regions. The correlation of those multiparametrics from biopsy sites with the histopathological finding (cell density, microvessel density, and Ki-67 labelling index) are examined by the Spearman correlation analysis and principal component analysis. Results:The correlation analysis demonstrates that Cho/Cr, Cho/nCr, rADC ratios correlate well with cell density, microvessel density, and Ki-67 labelling index (P<0.05); NAA/Cho, NAA/Cr, Cho/nCho, NAA/nCr, NAA/nNAA,rCBV ratios correlate well with cell density and microvessel density (P<0.05); but rFA ratios do not appear to be correlated with those the histopathological markers (P>0.05). Using principal component analysis and multivariate linear regression, it is found that the first principal component (PC1) and the second principal component (PC2) are significantly correlated with the histopathological findings. Conclusions:The significant correlations between the MR imagings and the histopathological features demonstrate that multiparametric functional MR assessment of gliomas based on 1H-MRS, DTI, PWI, especially PC1 and PC2, may be useful for assessing the extent and the degree of pathologic changes (i.e., tumor cell infiltration). |
PDF Full Text Request