No abstract

OBJECTIVE. Our purpose wasto evaluate therelationships betweentheratioof maxi mum relative cerebral blood volume (rCBV) (rCBV ratio = rCBV[tumor] I rCBV[contralat eral white matter]) and histologic and angiographic vascularities of gliomas using the gradient-echo echoplanar MR imaging technique. We also evaluated the usefulness of rCBV maps for grading gliomas. SUBJECTS AND METHODS. We examined 30 patients with histologically verifiedgliomas. Gliomas were classified as glioblastoma, anaplastic glioma with enhancement, ana plastic glioma without enhancement, and low-grade glioma. The maximum rCBV ratio of each glioma was compared with both histologic and angiographic vascularities, and the rela tionship between the maximum rCBV ratios and each type of glioma was established. RESULTS. The maximumrCBV ratiosof thegliomassignificantly correlated with both for low-gradegliomas.The maximumrCBV ratiosofthe glioblastomaswere significantlyhigher than those of the anaplastic gliomas without enhancement (p = .002) and the low-grade gliomas (p < .001). The maximum rCBV ratios of the anaplastic gliomas with enhancement were higher than those of the anaplastic gliomas without enhancement and the low-grade glio mas, but the differences were not statistically significant (p = .08 and p = .03, respectively). CONCLUSION. The resultsof perfusion-sensitive MR imagingwith gradient-echoechoplanar technique correlated with both histologic and angiographic vascularities. [6] reportedthat the degree of vascularity of gliomas observed with a spin echo echoplanar technique correlated well with histologic vascularity and that these re suIts might show the ability of the spin-echo echoplanar technique to detect neovasculariza tion at the capillary level. However, tumor an giogenesis is a complex process, and the existence of larger neovascular structures (such as feeding arteries and draining veins) associated with malignant tumors is not un common. A study documented that some tu mors have average capillary diameters two to three times those of normal tissues [7].The gradient-echo echoplanar technique produces hemodynamic maps by representing theeffectsoftotal bloodvolumefromcapillar ies to large vessels and weighs all vessels ape proximately equally [8]. In addition, this technique, because of its greater sensitivity to

Developments in MRI have made it possible to use diffusion-weighted MRI, perfusion MRI and proton MR spectroscopy (MRS) to study lesions in the brain. We evaluated whether these techniques provide useful, complementary information for grading gliomas, in comparison with conventional MRI. We studied 17 patients with histologically verified gliomas, adding multivoxel proton MRS, echoplanar diffusion and perfusion MRI the a routine MRI examination. The maximum relative cerebral blood volume (CBV), minimum apparent diffusion coefficient (ADC) and metabolic peak area ratios in proton MRS were calculated in solid parts of tumours on the same slice from each imaging data set. The mean minimum ADC of the 13 high-grade gliomas (0.92+/-0.27 x 10(-3) mm(2)/s) was lower than that of the four low-grade gliomas (1.28+/-0.15 x 10(-3) mm(2)/s) ( P<0.05). Means of maximum choline (Cho)/N-acetylaspartate (NAA), Cho/creatine (Cr), Cho/Cr in normal brain (Cr-n) and minimum NAA/Cr ratios were 5.90+/-2.62, 4.73+/-2.22, 2.66+/-0.68 and 0.40+/-0.06, respectively, in the high-grade gliomas, and 1.65+/-1.37, 1.84+/-1.20, 1.61+/-1.29 and 1.65+/-1.61, respectively, in the low-grade gliomas. Significant differences were found on spectroscopy between the high- and low-grade gliomas ( P<0.05). Mean maximum relative CBV in the high-grade gliomas (6.10+/-3.98) was higher than in the low-grade gliomas (1.74+/-0.57) ( P<0.05). Echoplanar diffusion, perfusion MRI and multivoxel proton MRS can offer diagnostic information, not available with conventional MRI, in the assessment of glioma grade.

Our aim was to review the MRI appearances of primary intracranial germ-cell tumours (GCT). We reviewed the MRI studies of 32 patients: 19 with germinomas, five with teratomas, one with an embryonal carcinoma, five with mixed and two with malignant nongerminomatous GCT. Eleven were in the pineal region, 12 suprasellar, five in the both sites, two in the basal ganglia and two in the corpus callosum. Contrast-enhanced images were available for 27 patients. The solid parts of GCT were nearly isointense with grey matter on both T1- and T2-weighted images. In seven patients with nongerminomatous GCT high-signal components were found on T1-weighted images, representing haemorrhage, high-protein fluid or fat. Cystic components were detected in 17 of 27 patients; eight germinomas and all nine nongerminomatous GCT had cysts. The solid components of germinomas enhanced homogeneously in eight cases and heterogeneously in 10, while all nongerminomatous GCT showed heterogeneous enhancement. MRI features tumours can facilitate correct diagnosis of GCT, including histological subtypes.

Although anisotropy measurements were higher in compact than non-compact white matter in all three age groups, the increase in anisotropy was greater in non-compact white matter across each of the three groups. These data suggest that although myelination is initially greater in compact white matter, the change in myelination may be greater in noncompact white matter during the first few years after infancy.