The MRI Study Of Brain~1H-MRS, Functional Connectivity And Gary Matter Structure In Patients With Post Stroke Depression

Background and ObjectiveStroke is one of the highest of the three diseases worldwide elderly population mortality, most of the survivors left behind neurological deficits. Varying degrees of neurological defects will lead to the loss of activities of daily living in patients with complete or partial. Post stroke depression(PSD) is an ever-present affeetive disorder after stroke having charaeteristic manifestations of depressing emotion, activity decrease, and sluggish ideation. Depression is a common complication associated with stroke, having a prevalence of20%-79%, with a pooled estimate of40-50%of stroke patients affected.This difference may be related to the study survey instruments, time after stroke assessment and diagnosis of different standards. PSD patients have decreasing of inrerest to participate in activities and medication compliance, also extend recovery time, resulting in Not only affect the prognosis of neurological deficit, prolonged hospitalization and increased medical costs, but also increase mortality. PSD has the characteristics of high incidence, high recurrence, high disability, high suicide. In view of the high incidence of PSD and its harmfulness, With reality, urgency and great social and economic benefit to accelerate the research on PSD, In recent years, PSD research is rapidly becoming one of the research hot topic at home and abroad. Pathogenesis of PSD is not yet explicit. It may be involved the psychosocial and biological mechanisms. The biological mechanisms that stroke lead to lesions of key parts of mood regulating circuit causing monoamine neurotransmitter dysfunction. Such as Vogel et al thinked that PSD is related to the change of neurobiology after brain damage, Because noradrenergic and5-hydroxytryptamine(5-HT) neuronal cell bodies located in the brain stem, thalamus and basal ganglia through its axon to reach the frontal cortex. When lesions involving the above position, resulted in norepinephrine and5-HT content decreased and leads to depression. The psychosocial Mechanisms that influence of family, social, physiological factors of post stroke lead to physiological and psychological imbalance causing reactive depression. Of course, like other psychiatric disorders, most scholars believe that PSD is a multifactorial disease, may be the common effect of biological factors and psychological factors in the pathogenesis of PSD.The diagnosis of PSD mainly depends on the clinical recognition and scale assessment, it is obviously subjective and the lack of effective objective diagnostic method, such as biochemical indexes, imaging. So it is easy to misdiagnosis or missed diagnosis, especially when the patients with aphasia, indifference or cognitive impairment, unable to complete the questionnaire or scale assessment, which directly influences the diagnosis and treatment of PSD. If PSD can be early recognition and diagnosis, active treatment of the original diseases, supplemented by psychological counseling and drug treatment and rehabilitation training. It is helpful to promote the recovery of neural function, improve symptoms and the quality of life of patients.Studies on primary depression has been found the abnormals of brain structures and functions in specific brain regions in patients with major depression by functional MRI (fMRI), magnetic resonance spectroscopy (MRS) and voxel-based morphomctry (VBM), and are the main two neural pathways involving depression emotion processing:1imbic-cortical-Thalamic loop, including the amygdala, Lateral nucleus of the hypothalamus and the ventral frontal cortex; limbic-cortical-striatal-pallidal-thalamic loop, involving the amygdala, cortex ventral and part of the structure of the above pathway. There are also some reports that Correlation between the course of depression and severity of clinical symptoms and functional or structural brain abnormalities, and effective treatment can improve these abnormal. We can understand the biochemical changes in brain metabolism by MRS, it is also contribute to the early identification of depression. VBM is mainly used to display and analysis the main structure of the brain gray matter, it is a convenient, safe and effective non-invasive tool, and can be used to study the pathological state in patients with depression. fMRI has the advantages of non-invasive, non-radiation and higher spatial resolution, we can understand the physiological and pathological brain activity in patients with depression under basal state by fMRI. So these imaging techniques have great potential in clinical studies of depression, it also will play an important role in the pathogenesis of depression, early diagnosis and therapeutic evaluation.But the study of MRS, fMRI and VBM work at present with regard to depression are mainly concerned primary psychiatric diagnoses, while the study about PSD is rarely reported. Thus, we designed a study to examine brain structure and brain function in patients with PSD in the combination of VBM, MRS and fMRI and automatic ROI methods. Our aim was to determine whether patients with PSD present abnormalities in the brain structure and function,and analysis the relationship between brain structure and brain function changes and HDRS score, Duration of depression.MethodsSubjects and groups Overall33first-time ischemic stroke patients(19females and14males) were recruited from among the inpatients treated at the Department of Rehabilitation Medicine at Zhujiang Hospital from2012June to2013May. Clinical assessments included Mini-Mental State Examination (MMSE),17-item Hamilton Depression Rating Scale (HDRS) and bathel index. The33patients were divided into2groups:"PSD" group consisted of13stroke patients meet the Diagnostic and Statistical Manual of Mental Disorders(DSM-IV) diagnostic criteria for depressive episodes. and an HDRS score>17points,"non-PSD" group included20stroke patients meet the DSM-IV diagnostic criteria for depressive episodes or have an HDRS score<7. After all subjects were gave written informed consent, MRI scan were obtained ordering to rsfMRI, MRS and VBM. Region of interest and data preprocessing1. MRS We selected the normal tissue symmetric parts of the same volume of the prefrontal cortex as ROI, the peak area of NAA, Cho and Cr were measured and the ratios of NAA/Cr and Cho/Cr were determined. For numerical data, according to the result of a test for normality, Two-sample t-test or the Mann-Whitney U test, and bivariate correlation analysis was used for statistical analysis.2. rsfMRI We selected the pregenual anterior cingulate cortex (pgACC), Bilateral amygdala and thalamus as regions of interest (ROI). Two sample t-test was used to compared the Z-score between two groups.3. VBM We selected the bilateral hippocampus and anterior cingulate cortex (ACC) as ROI. Age, gender and Total Intracranial Volume were included as nuisance covariates. Statistical significance of group differences in each region was set at P<0.002(AlphaSim corrected, derived from an uncorrected voxel level threshold of P<0.001and cluster size>1026mm3), For the PSD patients, each patient’s GMV/GMC values were extracted for each cluster of volume/concentration difference obtained by the procedures for partial correlation analyses, correlating extracted gray matter values with the HDRS score and duration of depression, age and gender was included as nuisance covariates. The correlative relationship was considered to be significant at P<0.05.ResultsAll patients finished MRI scan. Data could be analyzed finally including13PSD patients (7females and6males) and20non-PSD patients (12females and8males) in MRS,13PSD patients (7females and6males) and18non-PSD patients (11females and7males, two patients were excluded due to bad image quality) in rsfMRI,13PSD patients (7females and6males) and18non-PSD patients (11females and7males, two patients were excluded due to bad image quality) in VBM. Data of13PSD patients and20age and gender-matched non-PSD patients were analyzed in MRS part,and13PSD patients and18age and gender-matched non-PSD patients were analyzed in rsfMRI and VBM part.1. there were no significant differences between subjects with or without depression with regard to age, MMSE score, barthel index, gender distribution(Fisher exact test) and years of education in three parts. The 17-item HDRS scores were significantly difference between subjects with or without depression.2. MRS Compared with the non-PSD patients, the PSD patients had significantly lower NAA/Cr ratio in left prefrontal lobe (t=-3.396, p=0.002) and significantly higher Cho/Cr ratio in bilateral prefrontal lobe(left t=4.17, p<0.001; right t=2.288, p=0.29). No statistical difference of NAA/Cr ratio was found between two groups in right prefrontal lobe. In PSD group, we also analysis the correlation between HAMD score and NAA/Cr and Cho/Cr ratio in the prefrontal cortex, Significant Positive correlation existed between Hamilton Depression Scale (HAMD) scores and NAA/Cr values in the left in left prefrontal lobe in PSD group.3. rsfMRI Compared with the non-PSD patients, the PSD patients showed significantly decreased functional connectivity in the pregenual anterior cingulate cortex and left amygdala and right thalamus (P<0.05). No statistical difference of functional connectivity in the pregenual anterior cingulate cortex and right amygdala, left thalamus (P>0.05).4. VBM Compared with subjects without depression, patients with PSD demonstrated increased GMC in the left ACC (MNI coordinates:x=-1.5, y=45, z=6,673voxels, t=-15.44, P<0.002,AlphaSim corrected) and the right ACC (MNI coordinates:x=9, y=34.5, z=4.5,405voxels, t=8.72, P<0.002,AlphaSim corrected), Significant GMC differences between groups in the bilateral hippocampal gyrus were not observed. However, reduced GMV was observed in the bilateral ACC (MNI coordinates:x=4.5, y=31.5, z=-3,1207voxels, t=13.66, P<0.002,AlphaSim corrected) and the right hippocampal gyrus (MNI coordinates:x=37.5, y=-9, z=22.5,310voxels, t=8.72, P<0.002,AlphaSim corrected). There were no GMV increases between groups at the same level of significance. In PSD group, we did not found regional gray matter changes significantly associate with HDRS score and Duration of depression (P>0.05).Conclusion1. NAA and Cho abnormal of pre-frontal were found in PSD patients, the NAA and Cho abnormalities may play an important role in the pathogenesis of PSD.2. Findings showed significant positive correlation in left prefrontal between NAA/Cr ratio and HAMD score in PSD patients, suggesting that it has a certain value in reflecting the severity of PSD.3. rsfMRI results suggest that abnormal functional connectivity within mood regulating circuit existed in patients with post stroke depression during resting state, it contribute to Early diagnosis, treatment and further understanding of the pathogenesis of the PSD.4. Patients with PSD demonstrated increased GMC in the bilateral ACC and reduced GMV in the bilateral ACC right hippocampal gyrus are similar to Primary depression.5. we did not found regional gray matter changes significantly associate with HDRS score and Duration of depression. The inclusion of patients with mild to moderate severity of illness, the relatively short duration of depression and the lack of major depression, may result in the absence of a significant correlation between regional volume/concentration and the HDRS score.