| PusposeGliomas are the most common primary tumors in human brain. We treat different grade of gliomas with different therapies. Several studies have shown that a common feature of gliomas is an increased Cho and an increased Lip/Lac level,a decreased NAA level, but optimum methods for standardization of in vivo proton spectra in grading gliomas are also a matter of debate and there are less reports about that in our country. The purpose of our study is to evalute proton MR spec-troscopic pattern of the solid positions in various histologic grades gliomas and to report characteristic spectral patterns that maybe helpful in grading gliomas.Materials and MethodsWe performed PROBE SI examinations and evaluated spectra in 10 nomal adults(5 male and 5 female,mean age 39. 4 years) and 23 patients, 16 - 66 years old ( mean age 43. 3 years, 16male and 7female)who had primary intracranial gliomas. All patients underwentsurgery. Histological classification of gliomas was made according to criteria of WHO and tumors were grouped into three teams : benign team( grade 1 and 2 ) , anaplastic team ( grade 3 ) , malignant team ( grade 4). All examination were performed using PRESS Probe SI before surgery. MRI and PROBE SI multivoxel MRS were performed with GE singna CV 1. 5T whole body MR unit and a standard head coil. Before spectroscopy, we obtained conventional MRI using a clinical protocol that include SET1 WI( 440/9.0ms) , FSE T, WI ( 5407 9. Oms) ,T2WI(4400/99. 6ms) and contrast enhanced MRI to guide us in locating ROI for Probe SI studies. The size of ROI was as large as posible to include all tumor body and nomal brain. Probe SI provides automated shimming , water suppression, data processing and online spectrum display. The spectra measurements were the post - handling and we obtained the spectra data such as NAA/R, Cho/R,CR/R , Lip ( Lac )/R, NAA/Cho and Cho/Cr. Means , standard deviations based on the sampling were calculated and t - test were made.RESULTSAll 23 patients performed surgery and histologic classification of gliomas was made by two experienced neurohistologist ( benign team 6 subjects;anaplastic team 8 subjects;malignant team 9 subjects). The spectra of tumors and the control spectra of nomal team were obtained. We got three definite signals in each group, they are NAA, Cho and Cr. We observed decreased NAA and increased Cho in gliomas, we observed significant Lip/Lac signal in five patients of malignant glioma, on the margin of the tumor adjancent to the necrosis. In nomal team, Cho/R 1.99+0.146,Cr/R 1, NAA/R 1.776+0. 083, Lip(Lac)/R0.096±0.024,NAA/Cho 1.530 ±0. 166,Cho/Cr 1.215 ±0. 167; in benign team Cho/R 1. 314 ± 0. 285, Cr/rl. 008 ± 0. 265, NAA/R1.343 ±0.. 228 ,Lip( Lac)/R0. 103 ±0. 023 ,NAA/Cho 0. 819 ±0.140,Cho/Cr 1.533 ±0.285;anaplastic team Cho/R 2.368 ±0.829,Cr/R0.984±0.283,NAA/RO. 861 ±0. 210, Lip (Lac)/R 0.132 ± 0. 058, NAA/Cho 0. 381 ± 0. 060, Cho/Cr 2. 863 ± 0. 998; malignant team Cho/R 2.026 ± 0. 685 , Cr/R 0. 594 ± 0.299, NAA/R 0 .467 ±0. 180,Lip(Lac)/R0.673 ±0. 387 ,NAA/ChoO. 228 ± 0. 076,Cho/Cr 3. 906 ± 1. 843. A significant difference exists between NAA and NAA/Cho ratios for nomal team and gliomas of benign team, anaplastic team and malignant team ( p < 0. 05 ) ; a signicant difference also exists between the Cho/Cr ratios and Cho for nomal team , low grade gliomas and high grade gliomas, but between anaplastic team and malignant team there is no significant difference ( p < 0.05). There is no reliable correlation between the Cr, Lip/Lac and the grade of gliomas.DiscussionMR imaging has become an important tool in the location and diagnosis of gliomas, but when the grading of gliomas is concerned, its a-bility is limited. Proton MR sectroscopy has made it possible to obtain functional and metabolic information about different type of brain gliomas non - invasively. Unfortunately, conventional MR spectroscopy examinations are time consuming and operater dependent and involve farily complex manual postprocessing work. Some inve...
|
PDF Full Text Request