Computational protocol: Brain structure abnormalities in young women who presented conduct disorder in childhood/adolescence

Similar protocols

Protocol publication

[…] Independent sample t tests and Fischer’s exact tests were conducted, using the Statistical Package for Social Sciences (SPSS, Version 21) to compare groups on sociodemographic and clinical characteristics. Differences in GMV between women with CD and HW were examined using a general linear model. In order to detect group differences, statistical parametric maps were computed on a voxel-by-voxel basis for GMV, and independent two-sample t tests were used. In order to avoid possible edge effects between different tissue types, we excluded all voxels with gray matter values less than 0.1 (absolute threshold masking). Total intracranial volume (TIV) was calculated as the sum of gray matter, white matter, and cerebrospinal fluid volumes, as estimated by the MATLAB get_totals script (http://www.cs.ucl.ac.uk/staff/g.ridgway/vbm/get_totals.m). First, we conducted exploratory between-group GMV whole-brain comparisons controlling for total intracranial volume (TIV). The uncorrected statistical threshold for voxels in the exploratory whole-brain analysis was set at p < .001. The significance threshold for whole-brain analyses was set at p < .05, using family-wise error (FWE) correction for multiple testing at voxel/cluster level. In order to correct for the nonisotropic smoothness of the data, nonstationary cluster correction was applied (Hayasaka, Phan, Liberzon, Worsley, & Nichols, ). Second, we used the small volume correction (SVC) approach to restrict our analysis to the predefined ROIs. ROIs were created for each hemisphere, using the Mars bar (MRC Cognition and Brain Sciences Unit, Cambridge, UK) and atlas for automated anatomical labeling (Tzourio-Mazoyer et al., ). The ROI for ACC is limited by the paracingulate sulcus rostrally and the white matter of the corpus callosum caudally. The hippocampus included the dentate gyrus, the uncus, and the hippocampus proper. We combined the mask of superior, medial, middle and inferior subregions of the OFC to get one mask for OFC. The anterior insula was defined by restricting the insula mask to the region anterior to the anterior commissure plane (i.e., y > 0). Clusters were reported as significant if they survived FWE correction for multiple comparisons, significance level of p < .05, within the ROIs.Two sets of analyses were conducted in attempt to disentangle neural correlates of CD and those of comorbid disorders and maltreatment. Step 1 aimed to determine whether observed group differences would remain after comparisons were adjusted for past comorbid disorders and maltreatment. Four models were computed. All models adjusted for TIV, age, and IQ. In addition, Model 1 adjusted for past alcohol and drug dependence, Model 2 adjusted for past anxiety and depression disorders, Model 3 for history of physical and sexual abuse, and Model 4 for all comorbid disorders and maltreatment. Since few of the participants presented comorbid disorders in adulthood when undergoing the brain scan, Step 2 aimed to determine whether differences between those with prior CD and HW would remain after adjusting for current alcohol and drug use, anxiety and depression symptoms, and all continuous measures. Three models were computed, again adjusting for TIV, age, and IQ. Model 1 adjusted for current alcohol and drug use, Model 2 for current anxiety and depression symptoms, and Model 3 for all current measures. PCL Facet 2 scores were low. However, to determine their association with GMV, regression analyses were computed to estimate the association with whole brain and ROIs after adjusting for TIV, age, and IQ. […]

Pipeline specifications

Software tools SPSS, AAL
Applications Miscellaneous, Magnetic resonance imaging
Organisms Homo sapiens
Diseases Brain Diseases, Hamartoma Syndrome, Multiple, Substance-Related Disorders, Genetic Diseases, Inborn
Chemicals Ethanol