Structural And Functional Characteristics Of The Corpus Callosum In Schizophrenia Using Magnetic Resonance Imaging

Objective:As a severe and complex mental disorder,schizophrenia usually occurs in late adolescence or early adulthood.Prior postmortem studies of schizophrenia found that there were widespread cortical and subcortical abnormalitis in the patients.However,samples in these studies were typically small and methodological factors related to tissue preparation may have limited detection of abnormalities.With the advent of magnetic resonance imaging(MRI),it became feasible to evaluate cerebral structure and function in vivo to study larger more representative patient samples and avoid some of these potential complications.A series of MRI studies have found dysregulated anatomic and functional brain connectivity in schizophrenia,consistent with the view that schizophrenia is a“dysconnection syndrome”.Notably this includes interhemispheric connectivity,which raises interest in the structural integrity of the corpus callosum(CC),the largest commissural fiber in the human brain.For this reason,disturbances in CC have been of considerable interest in investigations of the etiopathology of schizophrenia.Firstly,the CC is known to be altered in patients with chronic schizophrenia.However,its morphologic characteristics are less well studied in treatment-naive first-episode schizophrenia patients,as is the effect of antipsychotic treatment on this structure.Hence,we recruited a large sample of antipsychotic-naive first-episode schizophrenia patients(AN-FES)to explore callosal morphology.We also tested for changes of the CC morphology over the first year of illness in a subgroup of patient available for follow-up at 1-year.Then,our findings in Chapter one suggested that deficits of the CC in first-episode schizophrenia was not affected by antipsychotic treatment,and also unrelated to drug treatment response.However,the long-term impacts of antipsychotic monotherapies on callosal morphology in schizophrenia remain unclear.Hence,this part of the present study sought to explore micro-and macrostructural characteristics of the CC in never-treated patients and those with long-term mono-antipsychotic treatment.Lastly,MRI findings in previous two chapters have demonstrated significant structural abnormalities in schizophrenia.However,it is unclear whether aberrant inter-hemispheric functional connection is related with abnormalities of the CC in the patients with schizophrenia.In addition,many neuroimaging studies confirmed that fibers across different subregions of the CC are responsible for connecting different cortical regions between two cerebral hemispheres.Therefore,this chapter sought to explore correlations of white matter integrity of the CC with inter-hemispheric functional connection in schizophrenia,and to identify whether fibers across different subregions of the CC are responsible for different functional connection.Materials and Methods:(Chapter One)High resolution T1-weighted images and the Positive and Negative Symptom Scale(PANSS)scores were acquired from one hundred and sixty AN-FES and 155 healthy controls(HCs)before treatment initiation.Among the patients,fortyfour were available for follow-up studies after one year of antipsychotic treatment,and were divided into good-outcome(n = 31)and poor-outcome subgroups(n = 13)based on whether there was a 50% reduction of total PANSS scores from baseline.Automated registration toolbox(ART)was used to automatically identify the midsagittal plane(MSP)and obtain measurement parameters of the CC morphology of each participant,including total CC area,perimeter,length,thickness,circularity and the area of seven CC subregions delineated by Witelson.For subregional values of the CC at baseline,we applied multivariate analysis of covariance(MANCOVA)to test for overall between-group differences.Univariate ANCOVA were conducted to test for baseline group differences in total area,perimeter,length and circularity.We also examined correlations of clinical symptoms with statistically significant group differences in callosal metrics.For longitudinal analyses,paired t analysis was performed to test for the change in CC.Statistical significances of all analyses were set at P<0.05 two-tailed and p-values were corrected for multiple comparisons using the false discovery rate(FDR)procedure.(Chapter Two)Twenty-three clozapine-treated schizophrenia patients(CT-SCZ),19risperidone-treated schizophrenia patients(RT-SCZ),23 never-treated schizophrenia patients(NT-SCZ),and 35 HCs were recruited.High resolution structural images and DTI data for each participant were obtained via a 3.0 T MR scanner.PANSS was used to estimate the severity of clinical symptoms of schizophrenia patients.Free Surfer was used to preprocess imaging data and extract callosal volumes and FA values of each participant.Then,we applied MANCOVA to test for between-group differences in subregional volumes(or FA values)of the CC.Univariate ANCOVA was conducted to test for group differences in CC total volume.We also examined correlations of clinical symptoms with statistically significant group differences in callosal metrics.Statistical significances of all analyses were set at P<0.05 two-tailed and p-values were corrected for multiple comparisons using the FDR procedure.(Chapter Three)One hundred and ninety-six antipsychotic-naive schizophrenia patients(AN-SCZs)and 214 HCs were recruited.Imging data of each participant,including high resolution T1 image,and resting-state functional MRI(rs-f MRI)and DTI data,was obtained via a 3.0 T MR scanner.PANSS was conduted to estimate the severity of clinical symptoms of schizophrenia patients.Graph Theoretical Network Analysis(GRETNA)was used to preprocess rs-f MRI data,yielding a 264x264 functional connection matrix for each participant.Free Surfer was used to preprocess all structrual images and DTI data,and to obtain FA values of the CC five subregions.Then,we applied MANCOVA to test for between-group differences in FA values of the CC five subregions.Two-sample t test was used to test for group differences in functional connection between bilateral hemicerebrums.Statistical significances of all analyses were set at P<0.05 two-tailed.Finally,we examined correlations of FA values of the CC five subregions in AN-SCZs with statistically significant group differences in inter-hemispheric functional connections using sparse canonical correlation analysis(s CCA).Results:(Chapter One)Compared with HCs,AN-FES patients showed a significant reduction of thickness in the posterior midbody of the CC.This deficit was correlated with severity of negative symptoms.The averaged thickness values of the other six subregions were not significantly different.In addition,ANCOVAs showed no significant difference in the total area or seven subregional areas of the CC in the MSP between AN-FES and HCs.There was also no significant group difference in other metrics including perimeter,length and circularity between the two groups.After one year of antipsychotic treatment,there was no significant change in CC morphology in schizophrenia patients.There was also no significant difference of CC morphology between good-outcome and poor-outcome subgroups at baseline,or at 1-year followup.(Chapter Two)There were significant deficits in the total and subregional CC volume,and white matter integrity in NT-SCZ in comparison with healthy subjects.RT-SCZ showed significantly increased FA values in the anterior CC region in comparison with NT-SCZ,while there were no significant differences in FA values of the anterior CC region between CT-SCZ and NT-SCZ.Meanwhile,only RT-SCZ showed significantly increased volume in the mid-anterior CC region.Moreover,the volume of the mid-anterior CC region was significantly smaller in CT-SCZ compared to HCs.No significant correlations of clinical symptoms with callosal metrics were observed.(Chapter Three)Compared with HCs,AN-SCZs showed significantlly reduced FA values in the ccentral and posterior regions of the CC.The patients also showed significantly dysregulated functional connections between bilateral hemicerebrums.All five s CCA modes were significantly correlated(P<0.05),and canonical correlation coefficients in the third and fifth clinical mode were 0.77 and 0.76 respectively.It was suggested that there was a significant relationship between alterations of white matter integrity of the CC and functional connections between bilateral hemicerebrums in the patients with schizophrenia.Conclusion:The findings in Chapter One suggest a stable deficit localized to the posterior midbody of the CC in the early course of illness that is related to negative symptom manifestation.This alteration was not affected by antipsychotic treatment,and was also unrelated to drug treatment response.Then,our findings in Chapter Two demonstrate that chronic exposure to antipsychotic medications may have an impact on brain structure of schizophrenia patients,especially in those with risperidone treatment.Lastly,the findings in Chapter Three suggest significant correlations between brain structural and functional connectivity in human.It also demonstrates that alterations of white matter integrity in different subregions of the CC may cause distinct patterns of dysregulated functional connection between bilateral hemicerebrums in schizophrenia.In a word,these results in this study reveal dysregulated develomental trajectory of the CC in schizophrenia,and the interaction of antipsychotics with callosal structures in different stages of illness.suggesting the importance of persistent antipsychotic therapy for schizophrenia.It also notes that structural connectivity is a prerequisite for functional connectivity between two cerebral hemispheres,especially the CC.