Font Size: a A A

Correlation Of FA And Metabolite Concentrations Measured Using ~1H-MRS In Healthy Cerebral White Matter And Brain Tumors

Posted on:2016-06-06Degree:MasterType:Thesis
Country:ChinaCandidate:S N ChengFull Text:PDF
GTID:2284330461486255Subject:Medical imaging and nuclear medicine
Abstract/Summary:PDF Full Text Request
Part 1Correlation of fractional anisotropy and metabolite concentrations measured using 1H-MRS of cerebral white matter in healthy adultsObjectiveTo investigate the relationship between FA and metabolite concentrations of bilateral medium corona radiata in healthy adults using DTI and’H-MRS technique. Combining DTI with 1H-MRS reveals the relationship between structure and metabolic characteristics of white matter.Materials and MethodsThe study was approved by the Review Board of Shandong Medical Imaging Institute. All subjects had been given informed consent before the enrollment. Studies were performed in 43 (23 men and 20 women) healthy volunteers, aged from 18 to 66 (man,18~61 years, mean age 42.6years, women 21-66 years, mean age 45.3 years). All volunteers were divided into two groups:junior group (n=22, mean age=31.7 years), senior group (n=21, mean age=56.6 years).All participants underwent DTI and 1H-MRS using 3.0T MR system (Siemens, Skyra, Germany), equipped with sixteen-channel phased array head coil.4~6 voxels were selected from each side and the inclusion criteria included:(1) Voxel was covered with pure white matter and without gray matter, (2) Voxel was not contaminated by cerebrospinal fluid, (3) Voxel was located in the center of the volume of interest (VOI). The area under the peaks was calculated by the software as a measure of the metabolite concentration. The concentrations of N-acetylaspartate (NAA,2.02ppm), choline (Cho,3.22ppm), creatine (Cr,3.02ppm) and the ratio of NAA/Cr were measured in each selected voxel. The size (10mm×10mm) of the cross section of a region of interest (ROI) on FA maps was equal to the size of the cross section of corresponding spectroscopic voxel. The ROIs were manually drawn on the FA map of each individual subjects and then calculated the mean of FA value among all voxels in each ROI.The correlations between FA and age were analyzed by Pearson’s correlation coefficients in total group, junior group and senior group. The correlations between FA and metabolite concentrations were estimated using partial correlation test controlling for the age related bias.0<r<0.4, mild correlation,0.4≤r<0.7, medium correlation,0.7≤r≤1, strong correlation. Values were significant difference at p<0.05.ResultsThere were no significant abnormal signals in all 43 healthy volunteers. The total number of samples is 502,257 in junior group and 245 in senior group respectively. The FA and NAA shows significant negative relationship with age among total 502 samples (r=-0.198, p<0.001, r=-0.199, p<0.001). The correlation is stronger in senior group than junior group (r=-0.222, p<0.001, r=-0.423, p<0.001). The correlation between age and Cho, Cr are mild (r=0.199, p<0.001, r=0.333, p<0.001).In 502 samples, the partial correlation of FA with the corresponding NAA is positive (r=0.432, p<0.001). It is stronger in senior group than junior’s (r=0.393, p<0.001, r=0.454, p<0.001). The partial correlation coefficient of FA with Cho is mild in senior group (r=-0.147, p<0.01). The positive correlation of FA with the corresponding NAA/Cr was significant in total 502 samples (r=0.144, p<0.01) but was not significant in the senior group. There is no significant correlation between FA and Cho/Cr.ConclusionThere is a significant relationship between FA and NAA concentration, which is closely related to the structure and metabolic characteristics of neurons and glia. The combination of DTI and 1H-MRS is capable to provide a better understanding of relationship between changes in white matter structure and metabolic characteristics.Part 2Correlation of FA and metabolite concentrations measured using 1H-MRS in brain tumorsObjectiveTo evaluate FA and metabolite concentrations and their correlation in brain tumors using DTI and 1H-MRS technique. Combining DTI with 1H-MRS reveals the microstructure and metabolic characteristics of tumors.Materials and MethodsThe study was approved by the Review Board of Shandong Medical Imaging Institute. All subjects had been given informed consent before the enrollment. Studies were performed in 33 patients with brain tumors (25 gliomas,8 metastatic tumors) confirmed by clinical and pathology. All patients underwent DTI and 1H-MRS using 3.0T MR system (Siemens, Skyra, Germany), equipped with sixteen-channel phased array head coil.1-7 voxels were selected from tumors, edema, and normal white matter. The concentrations of N-acetylaspartate (NAA,2.02ppm), choline (Cho, 3.22ppm), creatine (Cr,3.02ppm) and the ratio of NAA/Cr, Cho/Cr were measured in each selected voxel. The size (10×10 mm) of the cross section of a region of interest (ROI) on FA maps was equal to the size of the cross section of corresponding 1 H-MRS voxel. Then calculated the mean FA value among all voxels in each ROI.The significant differences between tumors, edema, and normal white matter in gliomas and metastatic tumors were assessed by independent-samples t tests. The correlations between FA and 1H-MRS metabolite (NAA, Cho, Cr) were analyzed by Pearson’s correlation coefficients in gliomas and metastatic tumors. Values were significant difference at p<0.05.ResultsGliomas group:The NAA, Cho, Cr, NAA/Cr, Cho/Cr show significant difference between tumor and edema except FA. Each parameters shows no significant difference between edema and normal white matter. The FA shows morderate correlation with NAA (r=0.451, p<0.001) and mild correlation with NAA/Cr (r=0.191,p<0.05) in tumors. The correlation between FA and NAA, Cho, Cho/Cr were significant (r=0.394, p<0.01; r=0.419, p<0.01; r=0.307, p<0.05)in edema. The FA shows mild correlation with NAA and NAA/Cr (r=0.374, p<0.001; r=0.225, p<0.05) in normal white matter.Metastatic tumors group:The FA, NAA, Cho, Cho/Cr show significant difference between tumor and edema except Cr and NAA/Cr. The FA, NAA, Cho, Cr, Cho/Cr show significant difference between edema and normal white matter except NAA/Cr. The FA shows no significant correlation with other parameters in tumors. The correlation between FA and NAA, Cho and Cr were significant (r=0.571, p<0.01; r=0.612, p<0.01; r=0.484, p<0.05) in edema. The FA shows mild correlation with NAA and NAA/Cr (r=0.366, p<0.05; r=0.338, p<0.05) in normal white matter.ConclusionThe combination of DTI and 1H-MRS is capable to provide a better understanding of boundary, infiltration degree, abnormal of white matter. It can also reveals the microstructure and metabolic characteristics of tumors.
Keywords/Search Tags:Diffusion tensor imaging (DTI), magnetic resonance spectroscopy (MRS), white matter, brain tumor
PDF Full Text Request
Related items