Functional Magnetic Resonance Imaging Study In Certain Brain Diseases

Part one Altered spontaneous brain activity in patients withtype2diabetes: a resting-state functional magnetic resonanceimaging studyObjectives: To explore functional brain changes in type2diabetesmellitus (T2DM) patients with and without microangiopathy (diabeticretinopathy and/or nephropathy), resting-state functional magneticresonance imaging (rs-fMRI) was used and the regional homogeneity(ReHo), amplitude of low frequency fluctuation (ALFF) and functionalconnectivity (FC) values of the brain were analyzed comprehensively.Furthermore, the relationship between these ReHo, ALFF and FC valuesand their clinical, biochemical, and neuropsychological tests was assessedin order to further clarify the pathogenesis of cognitive impairment inT2DM patients.Materials and methods: Eighty-four subjects were enrolled in thisstudy, including29T2DM patients with microangiopathy (diabeticretinopathy and/or diabetic nephropathy)(T2DM M+, mean age58.2±3.3years, mean disease duration12.9±0.2years),27T2DM patients withoutmicroangiopathy (T2DM M-, mean age59.2±4.5years, mean disease duration11.5±0.2years) and28healthy control subjects (mean age52.2±2.3years). All the subjects were scanned using a GE Signa Hdxt3.0Tscanner (General Electric Medical Systems, USA) and completed clinicalinformation, biochemical and neuropsychological tests as well as diabeticcomplications were also assessed. After the rs-fMRI data werepreprocessed with Data Processing Assistant for rs-fMRI (DPARSF)programs, ReHo and ALFF values for the brain were calculated using theRSET software. Functional connectivity analysis were performed betweenposterior cingulate cortex (PCC) and all the other voxels in the brain with avoxel-wise manner. ReHo, ALFF and FC values for the whole brain of allthe subjects were analyzed using one-way analysis of variance (ANOVA).In case of a significant multivariate F test, post hoc individual tests werecompleted for significance. Group differences were adjusted by Bonferronicorrection. The correlation analysis were conducted between ReHo, ALFF,or FC values of brain regions with significant difference and the results ofclinical, biochemical, or neuropsychological tests by using the Pearson’scorrelation coefficients, p <0.05was considered statistically significant.Results:1Compared with normal control subjects, all T2DM patients showedsignificantly decreased ReHo and ALFF values mainly in visual relatedcortex (bilateral lingual gyrus/calcarine cortex, left superior occipitalgyrus/middle occipital gyrus/cuneus, and bilateral fusiform gyrus), bilateral temporal lobe (superior temporal gyrus/transverse temporalgyrus, and middle temporal gyrus), sensorimotor related cortex(bilateral precentral/postcentral gyrus and left paracentral lobule) andbrain default mode network (DMN)(bilateral anterior cingulate, leftPCC and bilateral cuneus), while increased ReHo and ALFF values inbilateral precuneus/inferior parietal lobe, bilateral superior frontalgyrus/middle frontal gyrus and inferior frontal gyrus, bilateralinsula/thalamus and inferior temporal gyrus.2Compared with M－patients, M+patients showed significantly decreasedReHo values in left inferior occipital gyrus/cuneus/superior occipitalgyrus, left middle occipital gyrus and decreased ALFF values in leftinferior occipital gyrus/middle occipital gyrus.3Functional connectivity analysis demonstrated that both T2DM patientsand normal control subjects had the similar connectivity regions to PCC,mainly located in bilateral PCC, cuneus, precuneus, medial prefrontallobe, part of temporal and occipital lobe, hippocampus, andparahippocampal gyrus. The M+patients showed reduced functionalconnectivity to PCC in several default mode netwoek (DMN) regions,including the left temporal lobe, cingulate gyrus and posterior lobe ofcerebellum, while the M-patients showed reduced functionalconnectivity in the bilateral frontal lobe and left angular gyrus besidescingulate gyrus. Both M+and M-patients showed enhanced functional connectivity to PCC in bilateral precuneus, left superior parietal lobuleand superior occipital gyrus, right inferior temporal gyrus and middletemporal gyrus, right fusiform gyrus/lingual gyrus and inferior occipitalgyrus, and bilateral basal ganglia.4The ReHo and ALFF values of lingual gyrus/calcarine cortex of T2DMpatients were inversely correlated with the duration of disease, BMI andTMT-B, and positively correlated with CFT. Furthermore, the ALFFvalues of left superior temporal gyrus were inversely correlated with theduration of disease and BMI in T2DM patients.Conclusion:1The ReHo and ALFF values were decreased or increased in multiplebrain regions of both T2DM patients, suggesting abnormalsynchronization and spontaneous activity of low-frequency BOLD inthose regions, and there may be a functional defect in the brain.2The ReHo and ALFF values of bilateral visual related cortex in M+patients were decreased compared with those of M-patients, suggestingthat the former have worse visual cortex function and visual spatialcognitive impairment than the latter with the left hemisphere of thebrain more serious.3Functional connectivity analysis displayed reduced or enhancedconnectivity regions to PCC in DMN of T2DM patients，suggestingdegraded DMN connection performance and abnormal coordination. 4The correlation between ReHo or ALFF values and duration of disease,BMI, TMT-B and CFT in the visual cortex of T2DM patients suggeststhat ReHo or ALFF values may be used as objective approaches toevaluate potential cognitive impairment in T2DM patients. Part two Differentiation of intracranial tuberculomas andhigh grade gliomas using proton MR spectroscopy anddiffusion-weighted imagingObjectives: The purpose of this study was to determine whether protonMR spectroscopy (1H-MRS) and diffusion-weighted imaging (DWI) can beused to differentiate intracranial tuberculomas from high grade gliomas(HGGs).Materials and methods: A total of41patients (19with intracranialtuberculomas and22with HGGs) were examined in our study.1H-MRSand DWI were performed at a1.5T MR scanner before operation ortreatment.1H-MRS were performed using PRESS sequence (TR/TE:1000ms/144ms). Concentrations of N-acetylaspartate (NAA), creatine (Cr),choline (Cho), and lipid and lactate (LL) in the solid part of each lesionwere expressed as metabolite ratios and were normalized to thecontralateral hemisphere. DWI were performed using single shot SE-EPI sequence with two different b values (0s/mm2and1000s/mm2). Theapparent diffusion coefficient (ADC) was also calculated. The metaboliteratios and ADC values in the solid part of intracranial tuberculomas andHGGs were compared with Wilcoxon rank sum test using the software ofSAS9.1. Diagnostic accuracy was compared using receiver operatingcharacteristic (ROC) analysis.Results: The Cho/Cr、Cho/NAA and rCho ratios of the intracranialtuberculomas (2.42±1.39,1.86±0.95and1.23±0.78, respectively) werelower than those of the HGGs (3.86±2.23,3.96±2.52and1.98±0.53,respectively)(P<0.05). The minimum ADC value of the intracranialtuberculomas ((1.09±0.79)×103mm2/s) was higher than that of theHGGs((0.83±0.67)×103mm2/s)(P<0.05). No significant differenceswere found on the ratios of rLL、rCr and rNAA between the intracranialtuberculomas and HGGs (P>0.05). The sensitivity (93.8%), specificity(93.7%) and positive likelihood ratio (15.0%) of minimum ADC value fordifferentiating intracranial tuberculomas from HGGs were highest. Azvaluearea under the ROC curve of minimum ADC value is the highest (0.99),indicating that discriminatory capability was higher by minimum ADCvalue than by Cho/Cr, Cho/NAA and rCho ratios.Conclusion: The Cho/Cr, Cho/NAA, rCho and minimum ADC value canbe used to differentiate intracranial tuberculomas from HGGs efficientlyand the discriminatory capability was highest by minimum ADC value. These results suggest a promising role for1H-MRS and DWI in thedifferentiation between intracranial tuberculomas and HGGs.