Functional And Structural MRI Studies On Motor Subtypes And Cognitive Symptom Of Parkinson’s Disease

Parkinson’s disease(PD) is the second most commonneurodegenerative disease. These are the result of a slow and progressive degeneration ofdopaminergic neurons in substantia nigra(SN) and Lewy bodies(LBs) in the somata of the involved nerve cells, which leads to subsequent reduction of dopaminergic input to the striatum and imbalance of dopaminergic and cholinergic system. Major clinical features at presentation include the extrapyramidal motor dysfunction, accompanied by a variety of motor symptoms.The classic signs of PD are resting tremor, muscle stiffness or rigidity, slowness of movement or bradykinesia and postural instability. Patients with PD can be classified as having the tremor-dominant(TD) subtype or the akinetic-rigid(AR) subtype, based on the predominant motor signs. There is clear clinical proof that different subtypes of PD have a different clinical course and prognosis. Several lines of evidence suggest PD tremor may have different underlying pathophysiology processes from those of bradykinesia and rigidity. The neural basis for these disparate manifestations of PD remains unclear.Structural MRI study found decreased gray matter volume in the cerebellum of patients with PD and rest tremor compared with patients with PD without rest tremor. In contrast, voxel-based morphometry showed smaller gray matter and presupplementary motor volume in the PIGD subgroup, particularly in areas involving motor, cognitive, limbic, and associative functions. But brain structure changes tend to appear in the middle-late stage of disease, and early changes more concentrated in the aspect of brain function. Yet the brain function research of PD motor subtype is quite lack, seriously affected further understanding and awareness of neural mechanisms of the PD motor subtype.The purpose of this study was to investigate the changes of brain function of different PD subtypes using resting-state functional magnetic resonance imaging. Using voxel-mirrored homotopic connectivity(VMHC), we examined the homotopic resting-state functional connectivity patterns in akinetic-rigid PD and tremor-dominant PD. To fully characterize the abnormal brain networks in the two subtypes, a recently validated approach, called functional connectivity density(FCD)(with no need for a priori hypothesis) method and seed-based functional connectivity analysis were together performed.Classically, the motor complications are well recognized in PD and the diagnosis of PD is on the base of the existence of a series of motor signs. However,non-motor symptoms, such as cognitive impairment, depression, anxiety or psychotic symptoms are as common as the classical motor symptoms in PD. but the non-motor symptoms are often less well appreciated and not so well treated. Cognitive impairment is known to be a common feature of PD existing in approximately 27% of all non-demented patients; Dementia is also quite common in PD, with 80% of patients developing dementia over the course of their disease. Cognitive impairment and dementia in PD negatively affect quality of life and increase caregiver burden as well as nursing home placement. However, the specific neural mechanisms underlying cognitive impairment and decline are still unclear, making it difficult to develop treatments for cognitive symptoms.Neuropathologic and imaging studies suggest the hippocampus plays a crucial role in cognitive dysfunction of PD patients. So the researches of hippocampal structure become the key point of cognitive dysfunction in PD patients. However, previous studies focused the relationship between the whole hippocampal volume change and the cognitive performance of PD patients. But hippocampus can be divided into different subregions due to cell morphology, and different subfields also play different role in cognitive function. So it is necessary to study hippocampal subregion structure change respectively. Due to lack of longitudinal studies, longitudinal volume change of different hippocampal subfields in PD patients with cognitive decline is unknown. And research for the sensitivity of different subregion structures in disease development and the cognitive decline is still needed.Our goal was to assess the volume of different hippocampal subfields in PD patients with mild cognitive impairment and the longitudinal volume change of different hippocampal subfields in PD patients with cognitive decline using an automatic segmentation method based on the structure of magnetic resonance, and try to discuss neural mechanisms of cognitive impairment in PD from the perspective of anatomical structure.Here are four parts of the study.Methods:1. Each PD subject was diagnosed by an experienced movement disorder specialist according to the UK Parkinson’s Disease Society Brain Bank criteria. Motor performance was assessed using the Hoehn and Yahr scale and the motor section(part III) of the Unified Parkinson’s Disease Rating Scale(UPDRS). The PD subjects were classified as either TD or AR based on the ratio of mean tremor score(TD score) and mean akinetic-rigidity score(AR score). A total of 21 TD-PD patients, 29 AR-PD patients were recruited. An age and sex matched group of 26 healthy older volunteersserved as the normal control group. The images were acquired with a 3.0 T Siemens Tim Trio whole-body MRI system. After a conventional localizer scan and a T2 WI anatomic scan, resting-state functional images were acquired using an echo-planar-imaging(EPI) sequence. Preprocessing was performed using the Data Processing Assistant for Resting-State fMRI(DPARSF) with statistical parametric mapping(SPM8). The VMHC computation was performed using DPARSF software. Voxel-based comparisons of entire VMHC maps were conducted using REST software. Statistical tests across groups were performed using a voxel-based, one-way analysis of covariance(ANCOVA), with age, gender, and education level as covariates, followed by post hoc, two-sample t-tests. Monte Carlo simulations conducted using the AFNI Alpha Sim program were used to correct for multiple comparisons. A corrected significance level of 0.05 was for the ANCOVA analysis, and a combined P < 0.05 and cluster size > 1,728 mm3 for post hoc, two-sample t-tests. Pearson correlation coefficients between the altered VMHC values and each clinical characteristic of the PD patients were computed.2. Each PD subject was diagnosed by an experienced movement disorder specialist according to the UK Parkinson’s Disease Society Brain Bank criteria. Motor performance was assessed using the Hoehn and Yahr scale and the motor section(part III) of the Unified Parkinson’s Disease Rating Scale(UPDRS). The PD subjects were classified as either TD or AR based on the ratio of mean tremor score(TD score) and mean akinetic-rigidity score(AR score). A total of 25 TD-PD patients, 25 AR-PD patients were recruited. An age and sex matched group of 26 healthy older volunteersserved as the normal control group. The images were acquired with a 3.0 T Siemens Tim Trio whole-body MRI system. After a conventional localizer scan and a T2 anatomic scan, resting-state functional images were acquired using an echo-planar-imaging(EPI) sequence. Data preprocessing and the individual voxel-wise FCD maps were carried out using Neuroscience Information Toolbox.The brain sites identified by previous FCD group comparison between the two subtypes were used as seeds in subsequent FC analyses. To determine the effect of group, a voxel-based one-way analysis of covariance(ANCOVA) was used with age, gender, and education level as covariates both in FCDand FC analyses using REST software.The mean FCD and FC values were extracted to performpost hoc analyses with least significant difference(LSD)-t test. The AFNI Alpha Sim program was used to correct for multiple comparisons. For thesevoxel-wise statistical analyses across groups, we used a unified statistical threshold of p<0.05(Alpha Sim-corrected). And then spearman rank correlation analyses were performed to investigate the relationships between the clinical variables and the altered functional brain properties. ROC curves were computed using the mean global FCD differences between the TD patients and the AR patients that were obtained from the previous FCD statistical tests.3. Each PD subject was diagnosed by an experienced movement disorder specialist according to the UK Parkinson’s Disease Society Brain Bank criteria. Motor performance was assessed using the Hoehn and Yahr scale and the motor section(part III) of the Unified Parkinson’s Disease Rating Scale(UPDRS). Forty-nine patients with PD(23 PD patients with normal cognition and 26 PD patients with MCI) from the movement disorders outpatient clinic were included. In addition, 27 age- and gender-matched healthy controls were recruited from among local individuals who volunteered to participate in scientific studies. Global cognitive status was assessed with the Mini Mental State Examination(MMSE) and Montreal Cognitive Assessment(MoCA). A diagnosis of MCI in the patients with PD was made according to recently proposed consensus criteria. Based on a previously described procedure, total scores on the MMSE and Mo CA were divided into four cognitive-specific domains. The images were acquired with a 3.0 T Siemens Tim Trio whole-body MRI system. After a conventional localizer scan and a T2 anatomic scan, three-dimensional T1-weighted anatomical images were acquired using a volumetric three-dimensional magnetization-prepared rapid gradient–echo(MP-RAGE) sequence. The whole hippocampus was initially segmented by completing the Free Surfer image analysis pipeline(Martinos Center for Biomedical Imaging, Boston, Massachusetts). The FreeSurfer definition of the hippocampal subfields includes the dentate gyrus(CA4-DG), the Ammon’s horn subfields(CA1 and CA2-3), the fimbria, the hippocampal fissure, the subiculum and the presubiculum. Between-group differences in total hippocampal and subfield volumes were evaluated with analysis of covariance(ANCOVA) while controlling for age, sex, education, and ICV, followed by pairwise post hoc Bonferroni tests. The total and subdomain scores were used in the correlation analyses between cognitive performance and total hippocampal or subfield volumes.4. Twenty-eight PD patients were rescanned after two years(mean 899 days, SD 36). Sixteen patients were classified as having global cognitive decline and twelve patients without global cognitive decline. Motor performance was assessed using the Hoehn and Yahr scale and the motor section(part III) of the Unified Parkinson’s Disease Rating Scale(UPDRS). Global cognitive status was assessed with the Mini Mental State Examination(MMSE) and Montreal Cognitive Assessment(Mo CA). The images were acquired with a 3.0 T Siemens Tim Trio whole-body MRI system. After a conventional localizer scan and a T2 anatomic scan, three-dimensional T1-weighted anatomical images were acquired using a volumetric three-dimensional magnetization-prepared rapid gradient–echo(MP-RAGE) sequence. For longitudinal analyses, we used FreeSurfer’s longitudinal stream to process two serial MRIs from study participants to provide accurate estimates of subtle changes over time. The FreeSurfer definition of the hippocampal subfields includes the dentate gyrus(CA4-DG), the Ammon’s horn subfields(CA1 and CA2-3), the fimbria, the hippocampal fissure, the subiculum and the presubiculum. Repeated measures ANOVA was performed on in hippocampal subfield volume in follow-up groups(PD without cognitive decline and PD with cognitive decline) with group as a between-subjects factor, and time(baseline, follow-up) as a within-subjects repeated measure. The MMSE and MoCA scores were used in the correlation analyses between cognitive performance and total hippocampal or subfield volumes.Results:1. Compared with the control subjects, the TD-PD patients exhibited significant decreases in VMHC in the bilateral posterior lobe of the cerebellum; in contrast, the AR-PD patients showed lower VMHC values in the precentral gyrus. There was a significant negative correlation between tremor scores and VMHC values for the bilateral posterior lobe of the cerebellum only in TD-PD group(r=-0.55, p=0.01).2. 1). TD vs. AR. Specifically, three clusters, which included left inferior frontal gyrus, right middle frontal gyrus and right superior frontal gyrus, showed decreased global FCD in the TD patients. One cluster that included the cerebellum anterior lobe showed increased global FCD in the TD patients. 2). TD vs. HC. Compared to HCs, the TD patients had a significantly increased global FCD in bilateral cerebellum posterior lobe, fusiform gyrus and rectus gyrus. 3). AR vs. HC. Compared to the HCs, the AR patients exhibited significantly increased global FCD in the left cerebellum posterior lobe, the right middle temporal gyrus, the bilateral middle frontal gyrus and the left orbital inferior frontal gyrus. For the seed at the left cerebellum anterior lobe, TDs showed significantly increased FCs in the left ventral and dorsal postcentral gyrus, the left superior frontal gyrus, the right insula and orbit inferior frontal gyrus compared to ARs. In the PD group(TD and AR patients), a significant negative correlation was identified between the H-Y scale score and the FCD value of the cerebellum anterior lobe. Furthermore, in the PD group, a significant positive correlation was identified between TD score and the FCD value of the cerebellum anterior lobe, while a negative correlation was observed between TD score and the FCD values of three regions(the inferior frontal gyrus, middle frontal gyrus and superior frontal gyrus). The results of the ROC analysis indicated that the combination of the four FCD indexes had the highest power to discriminate the AR patients from the TD patients, with an AUC of 0.96(95% CI=0.91–1.00). At a cutoff of 0.23, where the sensitivity and specificity curves intersected, the sensitivity and specificity was 95.2% and 80.9% respectively and classification accuracy was 88.0%.3. Total bilateral hippocampal volume was significantly decreased in the PD-MCI patients compared to both the PD patients(left: P = 0.02, right: P = 0.01) and the controls(left: P = 0.007, right: P = 0.002), after adjusting for post hoc Bonferroni corrections(Table 3, 4 Fig. 2). The hippocampal subfields showed distinct patterns of between-group differences(Table 3, 4 Fig. 2). The left CA1(P = 0.03) and CA4-DG(P = 0.03) regions as well as the bilateral CA2-3(left: P = 0.03, right: P = 0.02) and subiculum(left: P = 0.008, right: P = 0.002) regions were smaller in the PD-MCI patients than in the controls. The right CA4-DG volume was significant decreased in the PD-MCI patients compared to the PD patients(P = 0.01) and controls(P = 0.005). In contrast, there were no statistically significant between-group differences in the volume of the left CA1 or of the bilateral presubiculum, fimbria and fissure subfields. The correlation analyses showed that MMSE total score, MMSE memory score and MoCA memory scores correlated with bilateral total hippocampal volume in the PD-MCI group but not in the other groups.Conclusion:1. Using VMHC, we found that functional coordination between homotopic brain regions is impaired in AR-PD and TD-PD patients. This study provides evidence of both cerebellum-related connectivity deficits in TD-PD. The finding that VMHC values and tremor scores were significantly correlated suggests that VMHC measurements may be of potential clinical relevance in TD-PD.2. Using FCD and seed-based FC approaches we found that the functional integration in cerebellum and frontal lobe are altered in both subtypes of PD. They also provided new perspective for the classification of PD subtypes, based on a comprehensive view of network disorganization in the two subtypes of PD.3. Using structural MRI we demonstrated the selective regional vulnerability of the hippocampus in the PD-MCI. These results corroborate neuropathological findings and add novel information about the involvement of the hippocampus in the cognitive dysfunction of PD.4. Using structural MRI we found longitudinal volume change characteristics of different hippocampal subfields in PD patients with cognitive decline with the progress of the disease course. And the CA2-3 and subiculum subfield appeared sensitive in PD with cognitive decline longitudinally, suggesting that the reduction in the volume of CA2-3 and the subiculum could represent a presymptomatic indicator for the development of dementia in PD.