| Purpose:The BOLD imaging is a commonly used functional magnetic resonance image(fMRI)technique,which relies on regional differences in MR signals between oxygenated and deoxygenated blood to delineate regional activity.These findings will allow for a more comprehensive understanding of neuropathologieal mechanisms of brain changes,which however,should be interpreted cautiously,for spontaneous thoughts and random uncontrolled cognitive processing is difficult to avoid completely during the scanning.As a structure MRI technique,DTI is sensitive to the subtle WM changes that could not be detected by conventional structural MRI,that is,DTI is meaningful in the early stage as well as subsequent clinical intervention of metal disorder diseases.It is believed that changes in specific WM tracts can lead to disruption in information transfer between brain regions resulting in deficits in cognitive functioning.In addition,WM fiber reconstructions allow us to localize and quantify directly anatomical connections or pathways of a fiber tract of interest.The most widely and sensitive used anisotropy measure to value random water movement in tissue is FA,which represents a normalized ratio of diffusion directionality and is considered as a direct marker of white matter integrity.However,conventional diffusion indices(e.g.,FA)required a given diffusion tensor(DT)model(3D anisotropic Gaussian diffusion)that is vulnerable to errors when assuming water diffusion in complex structures,such as in regions where fibers cross.Furthermore,FA only reflect diffusion properties within the voxel,which is thought to be failed to uncover any WM differences,i.e.,inter-voxel measurements of different sensitivity of WM changes should be put into use to reveal more information in mental disorder diseases.To date,a newly developed inter-voxel metric referred to as local diffusion homogeneity(LDH),has been applied to clinical practice and proved to be complementary to the conventional diffusion-MRI markers.This parameter is a model-free inter-voxel index that measures the similarity of the full diffusion profile of water molecules.Specifically,the diffusivity along each diffusion-weighted gradient direction is firstly computed in the original diffusion-weighted image(DWI),then a series of diffusivity that contains the total number of sampled gradient directions can be extracted for each voxel.Kendall's coefficient concordance(KCC)is used to quantify the overall diffusivity similarity of the series within a given voxel and its nearest neighborhood However,inter-voxel metric(i.e.,local diffusion homogeneity,LDH)with different sensitivities to specific WM microstructural properties was neglected.We aimed to investigate WM changes in Type 2 Diabetes Mellitus(T2DM)using both FA and LDH,and systemically compare the two metrics in possible biomarker detection.Meanwhile,the disruption of white matter network measured by parameters of small-world networks such as mean clustering coefficient and global efficiency in T2DM is related to slowing of information processing speed compared to controls.We supposed that the white matter network had already disturbed in early stage of T2DM when there was no sigh of cognitive impairments,which would be helpful in prediagnosis and intervention to prevent or postpone the prossece of cognition decline.Methods:Institutional review board approval and written informed consents from all participants were obtained.T2DM population were selected from hospitalized patients from endocrinology department of the hospital,and healthy controls were from volunteers over the same period.T2DM was diagnosed using fasting blood glucose>7.0 mmol/L ontwo separate occasions,or 2-hour blood glucose level>11.1 mmol/L during a 75-g oral glucose tolerance test.All participants received detailed neurological examination(e.g.,12 pairs of cranial nerves,body movement)by experienced neurologists to make sure no significant cognitive complaints or positive neurological symptoms.General clinical measurements and demographic characteristics for each subjects were collected,including biological tests,a chest X-ray,an electrocardiogram,BMI(weight(kg)/height(meter)2),education level,BP,and course of disease(for T2DM population only that measured from the date when first diagnosed T2DM to the date of MR I scanning).Education level was defined by the highest level of education completed.BP of healthy controls was measured after examinees had rested for at least 5 min;and BP of the T2DM group were reviewed from the medical records.Exclusion criteria for both groups were as follows:impaired glucose tolerance or impaired fasting glucose,serious eye diseases,any sign of cognitive impairment or positive neurological symptoms;any history of neurologic abnormality,serious head injury(with loss of consciousness>5min),severe hypoglycemia or hyperlipemia,left handedness or mixed-hander,BMI larger than 30 kg/m2,substance(e.g.,alcohol,tobacco,psychoactive drug)abuse,hypertension(the cut-off value that based on 2003 JNC 7 report:a systolic BPN=140 mm Hg or a diastolic BPN=90 mm Hg),hyperlipemia,specific abnormalities finding in conventional MRI scans,or other factors that might affect brain structure and function.Structural three-dimensional T1-wighted images and diffusion tensor image(DTI)were obtained from the T2DM subjects and health controls.The diffusivity along each diffusion-weighted gradient direction was calculated using original DWI,and subsequently,a series of diffusivity of each voxel yield a vector.The KCC was applied to quantify the overall similarity of each diffusivity vector of each voxel to those 26 neighborhoods.Voxel-based comparisons of LDH and FA between the two groups were computed.Statistical analyses were conducted with REST 1.8 toolbox.An independent two-sample t-test using the general linear model was performed to compare the between-group differences of FA and LDH on the WM skeleton mask.For each voxel,Pearson correlation was used to detect the correlations between diffusion matrix(LDH and FA)and clinical measurements(BMI and SBP)across subjects.Age,gender,and education level were considered as covariates during group comparisons and the correction analysis to regress out potential nuisance effects.All the statistical maps were corrected for multiple comparisons using a Monte Carlo simulation(the AlphaSim program).The parameters were:FWHM(full-width half-maximum)=6mm,1000 simulations,edge connection.Voxels level P value<0.01 and cluster-level P value<0.05 with cluster size>71mm3 were considered to show significant difference between the two groups.Based on the automated anatomical labeling atlas(AAL)standard brain template,all brain fibers were reconstructed into 116 cortical and subcortical regions(including the cerebellum)as nodes for network connectivity.Each connection is weighted with the average diffusivity.Each subject thus obtained a weighted 116×116 two-dimensional connection matrix.Using this matrix,the“Graph Theoretical Network Analysis(GRETNA)" is used to calculate the network index of each subject,including the index of local connectivity.1.Aggregation coefficient:the density of connections between neighbors.Determine whether the neighbor is a neighbor or not.2.Local connection efficiency:local connection efficiency of the reaction node;global connection index:average shortest path(the best edge connection on the node)and whole brain connection efficiency(reciprocal of the shortest path).The changes in brain structure networks were compared between the two groups and their relationship with clinical indicators was calculated.Finally,the graphical metric for each subject was normalized to a randomly generated network of comparable values(100 calculations)to assess whether the network had a small-world attribute.The properties of the small-world network are agglomeration coefficients greater than one.The changes in brain structure networks were compared between the two groups and their relationship with clinical indicators was calculated.Statistics:Two-sample t test based on a general linear model to calculate the difference between groups.Pearson's correlation was used to calculate the correlation between clinical indicators and diffusion indicators and small-world network indicators.In all statistics,age,gender,and education process were all covariates,and statistical thresholds were all taken as P<0.05.AlphaSim correction was used for the multiple polygraph corrections of the charts(precursor levels P<0.01 to form clumps,and the mass level P<0.05 was significantly different).Results:All subjects were middle-age.Compared to controls,T2DM patients had significantly higher LDH in pons involving the middle cerebellar peduncle(MCP)and corticospinal tract(CST),the left temporal pole involving uncinated fasciculus(UF)and inferior longitudinal fasciculus(ILF);whereas lower FA in left superior corona radiation.In general,BMI associated WM regions mainly loaded in the bilateral association fibers(UF,ILF.SLF)and fiber tracks that connected to the motor andsomatosensory areas(pre-and post-central gyrus):while significantly correlations between SBP and LDH/FA were more widely detected,including limbic system fibers(mCB and fornix),collect fibers extending to temporal lobe,callosal fibers(sCC and gCC),the right association fibers(UF,ILF,SLF,and IFO)and cerebrum(vermis and right cerebellum crus).Among T2DM subjects,greaterBMI were associated with increased LDH in the right fusiform gyrus,the left inferior parietal lobe and semiovale center,but decreased LDH in the right supra-marginal gyrus and superior radiation,and decreased FA in the left calcarine cortex;greater SBP were associated with increased LDH in widely regions including the right inferior temporal and occipital lobe,right cerebellum_crusl,bilateral cingulum bundle and orbitofrontal area,but decreased LDH in the right temporal pole and decreased FA in vermis.Some overlapping regions in BMI or SBP associated WM changes were detected by both LDH and FA.The measures of network integration showed decreased global efficiency and increased path length in patients with T2DM comparedto those in healthy subjects and the measures had a correlation with BMI,SBP,LDL,and the level of hyperglycemia.We suggest that disrupted network integration with widespread disruption of WM fiber integrity represent a manifestation of central neuropathy in diabetes,possibly contributing to mental slowing and cognitive impairments in individuals with diabetes.Conclusion:In this study,voxel-based comparative analysis between LDH and FA,as well as correlation analysis between diffusion metrics and clinical measurements across subjects revealed inter-index subject variability in specific WM regions,suggesting that inter-voxel and intro-voxel have different sensitivities to specific WM microstructural properties;meanwhile,the overlapping regions detected by both LDH and FA suggested that the LDH index was calculated reliably.It was believed that the decreased LDH value that represented disruption of fiber coherence,was associated with the cognitive dysfunctions of special regions,just as changes of FA/MD or ALFF/ReHo were believed to reflect changes of cognition.Increased LDH in the current study indicated the enhancement of coherence of local fibers,which may be due to changes of the fiber meylination,diameter or density differs along each WM tract,which was so called the neuronal plasticity that allowed the human brain to adapt to environmental pressure,physiologic changes,and experiences.The toxic effects of hyperglycemia that lead to a complex pattern of changes in synaptic plasticity in the rat cerebral cortex and hippocampus,as well as the associated cognitive dysfunction have been observed.Higher fluctuation in blood glucose,insulin resistance,overweight,hypertension,or even diabetes education during the long-term duration of disease which might lead to lifestyle changes,could have an effect on the microstructure of WM that required further research.There were none lacunar infarcts in all subjects in the current study,and also the occasionally WMHs found in the two groups was considered as a common finding in the general.However,greater BMI,SBP,and FDG were found in T2DM compared to controls.T2DM together with obesity,hypertension and other risk factors,constitute the metabolic syndrome that share common pathways leading to complex metabolic,inflammatory and microvascular disturbances that may further contribute to WM microstructural damage and cognitive decline.Intro-voxel matric had increased sensitive in detecting WM microstructure changes in T2DM.Increased LDH in pons and left temporal pole could be interpretedas compensatoryrecruitment of cognitive resources to "cover" the potential impairment of feeding circuitry and the right association fibers inmiddle-age T2DM patients without obvious impaired cognition,which were of great clinical significance for proactive treatments to prevent or reduce future cognitive decline.Our second part of the study of white matter networks is consistent with the previous functional and structural network conclusions.Both found that early T2DM had functional impairment and compensatory existence,and also confirmed the association between white matter changes and functional impairment in the first part.And the existence of compensatory mechanisms.Specifically,this study shows that the white matter structure network shows that the local connection efficiency of the bilateral cingulate gyrus,the right sacral pole,and the left inferior temporal gyrus of T2DM is significantly lower than that of the normal group.Compared with the results of the white matter test in Chapter 1,the reduction in the local efficiency of the right bony pole is consistent with the impairment of the right bony pole function in Chapter 1.A reduction in the local efficiency of the left lower quadrant may result in compensatory enhancement of the left bony pole function.Reduction of local cingulate connections in the bilateral cingulate,and the positive association of LDH and FA values with SBP in the bilateral cingulate gyrus shown in Chapter 1 is due to inconsistent conclusions due to compensatory mechanisms,possibly due to positively correlated brain-behavior.The relationship only suggests that blood pressure leads to localized changes in synapses rather than cognitive enhancement.This finding shows that the pathological significance of brain structure and function indicators needs further research to confirm,and also shows the correlation of brain structure,function indexes and clinical measurement values in previous studies,and does not necessarily have attenuation or compensation;at the same time,it also shows that The importance of multimodal magnetic resonance methods. |
PDF Full Text Request