Structural abnormalities in frontal, temporal, and limbic regions and interconnecting white matter tracts in schizophrenic patients with prominent negative symptoms

Objective: Imaging studies of schizophrenia have repeatedly demonstrated gl obal abnormalities of cerebral and ventricular volumes. However, pathologic al changes at more local levels of brain organization have not yet been so clearly characterized because of the few brain regions of interest heretofo re included in morphometric analyses as well as heterogeneity of patient sa mples. Method: Dual echo magnetic resonance imaging (MRI) data were acquired at 1. 5 T from 27 right-handed patients who met DSM-IV criteria for schizophrenia with enduring negative symptoms and from 27 healthy comparison subjects. B etween-group differences in gray and white matter volume were estimated at each intracerebral voxel after registration of the images in standard space . The relationship between clinical symptom scores and brain structure was also examined within the patient group. Spatial statistics and permutation tests were used for inference. Results: Significant deficits of gray matter volume in the patient group we re found at three main locations: 1) the left superior temporal gyrus and i nsular cortex, 2) the left medial temporal lobe (including the parahippocam pal gyrus and hippocampus), and 3) the anterior cingulate and med ia I fron tal gyri. The volume of these th ree regions combined was 14% lower in the patients relative to the comparison subjects. White matter deficits were fo und in similar locations in the left temporal lobe and extended into the le ft frontal lobe. The patient group showed a relative excess of gray matter volume in the basal ganglia. Within the patient group, basal ganglia gray m atter volume was positively correlated with positive symptom scores. Conclusions: Anatomical abnormalities in these schizophrenic patients with marked negative symptoms were most evident in left hemispheric neocortical and limbic regions and related white matter tracts. These data are compatib le with models that depict schizophrenia as a supraregional disorder of mul tiple, distributed brain regions and the axonal connections between them.