| Objective: Based on the techniques of MR Quantitative susceptibility mapping(QSM)and three-dimensional arterial spin labeling(3D-ASL),exploring the change of cerebral iron content and cerebral blood flow in different stages of Alzheimer’s disease(AD)and the correlation between the two quantitative indicators,in order to achieve early diagnosis and prediction for this disease through uderstanding the pathological physiological mechanism of AD patients from the level of brain function.Methods: Twenty-two patients with AD,twenty-two patients with mild cognitive impairment(MCI),twenty-five patients with subjective cognitive decline(SCD),and twenty-five normal control(NC)subjects,were enrolled by the Department of Neurology and Psychology of Guizhou Provincial People’s Hospital in this study from January 1,2018 to December 30,2019.A total of 94 subjects were enrolled,including 40 males and54 females,with an average age of(70.69 ± 7.22)years.A diagnosis of probable AD was based on NIA-AA;A diagnosis of MCI was made according to the Petersen;and A diagnosis of SCD was attained according to the consensus among chinese experts.All data were obtained on a 3.0T MR scanner(Discovery MR 750,GE Healthcare,Milwaukee,WI)equipped with a 32-channel phased-array coil in this study.All participants underwent a 3D T1-weighted imaging acquisition,a QSM acquisition and a 3D-ASL acquisition.Regions of interest(ROIs)were selected in Brainnetome Atlas based on iron deposition and abnormal cerebral blood flow(CBF)in AD,including globus pallidus(GP),putamen(PU),caudate nucleus(CA),hippocampus(HP),thalamus(TH),frontal cortex(FC),parietal cortex(PC),and occipital cortex(OC).And based on QSM-MRI and ASL-MRI Imaging,using the SPM12 software package(http://www.fil.ion.ucl.ac.uk/spm/)implemented for standardized measurements of brain susceptibility values(SUS)and CBF in MATLAB(MathWorks,Natick,MA,USA).The differences analysis of SUS and CBF between each group were performed by covariance analysis,and Bonferroni correction was used for pairwise comparison between groups.Non-parametric test was used for the correlation between SUS and CBF values with gender,and Spearman analysis was used for the correlation between SUS and CBF values with age.The correlation between SUS and CBF values with MMSE and MoCA scores was analyzed by partial correlation analysis,and the correlation between SUS and CBF values was analyzed using the same statistical method.The receiver operating characteristic(ROC)curves of SUS and CBF values were drawn,and the sensitivity,specificity and area under the curve(AUC)were analyzed.Results: Significant differences of CBF and SUS were observed between groups in several ROIs with ASL and QSM(P < 0.05),and then the two groups were compared to show statistically significant differences.Compared with NC group,SUS values of GP,PU and CA in AD,MCI and SCD groups were significantly increased(P < 0.001),and CBF values of most ROIs in AD and MCI groups were significantly decreased,and CBF values of TH in SCD group were significantly increased(P < 0.001).No significant correlation were observed between SUS and CBF values with gender in each group(P > 0.05).In the AD group,the SUS values in HP was negatively correlated with MoCA score(r =-0.688,P = 0.001),and the CBF values in HP was positively correlated with MoCA score(r =0.584,P = 0.007).In the MCI group,CBF values in HP was positively correlated with MMSE score(r = 0.595,P = 0.006)and MoCA score(r = 0.689,P = 0.001).In the progression of AD,SUS and CBF values in PU were negatively correlated.By analyzing the ROC curve,SUS values of the GP,PU and CA can distinguish the NC from the study groups(AD,MCI and SCD)(P ≤ 0.001),and the AUC values in PU are all over 0.95.The CBF values of multiple ROIs can distinguish NC from two study groups(AD and MCI),with the HP being more significant(AUC = 0.927),and then the CA(AUC = 0.911)and PU(AUC = 0.896).The difference of CBF value in TH between NC and SCD group was statistically significant(AUC = 0.838,P < 0.001).SUS values of GP and PU can distinguish SCD from AD,MCI from AD.CBF values in multiple ROIs can distinguish SCD from AD,SCD from MCI(P < 0.05).Among all ROIs,no significant difference were observed between SUS value in SCD and MCI group,and CBF value in MCI and ADgroup(P > 0.05).Conclusion: In the progression of AD,excessive deposition of iron in basal ganglia and extensive reduction of cerebral blood perfusion were observed.QSM detection of putamen iron deposition may become an imaging biomarker for early diagnosis of AD.The increase of putamen iron deposition is accompanied by the decrease of blood perfusion,which may be the interaction between oxidative stress response of free iron and vascular injury.Iron content and blood flow in the hippocampus may be related to the severity of the disease.The iron content and blood flow in specific brain regions are valuable in predicting the diagnosis of AD and the classification of disease course. |
