Cognitive performance is related to cortical grey matter volumes in early stages of schizophrenia: a population-based study of first-episode psychosis.

Background: Neuropsychological deficits have been reported in association with first-episode psychosis (FEP). Reductions in grey matter (GM) volumes have been documented in FEP subjects compared to healthy controls. However, the possible inter-relationship between the findings of those two lines of research has been scarcely investigated.
Objective: To investigate the relationship between neuropsychological deficits and GM volume abnormalities in a population-based sample of FEP patients compared to healthy controls from the same geographical area.
Methods: FEP patients (n=88) and control subjects (n=86) were evaluated by neuropsychological assessment (Controlled Oral Word Association Test, forward and backward digit span tests) and magnetic resonance imaging using voxel-based morphometry.
Results: Single-group analyses showed that prefrontal and temporo-parietal GM volumes correlated significantly (p<0.05, corrected) with cognitive performance in FEP patients. A similar pattern of direct correlations between neocortical GM volumes and cognitive impairment was seen in the schizophrenia subgroup (n=48). In the control group, cognitive performance was directly correlated with GM volume in the right dorsal anterior cingulate cortex and inversely correlated with parahippocampal gyral volumes bilaterally. Interaction analyses with "group status" as a predictor variable showed significantly greater positive correlation within the left inferior prefrontal cortex (BA46) in the FEP group relative to controls, and significantly greater negative correlation within the left parahippocampal gyrus in the control group relative to FEP patients.
Conclusion: Our results indicate that cognitive deficits are directly related to brain volume abnormalities in frontal and temporo-parietal cortices in FEP subjects, most specifically in inferior portions of the dorsolateral prefrontal cortex.