Both schizophrenia and bipolar disorder have been associated with progressive changes in grey matter (GM) volume. However, the temporal trajectories of these changes are poorly understood. The aim of this study was to assess longitudinal changes in grey matter volume subsequent to the first episode of schizophrenia and of affective psychoses. Adolescent patients with a first episode psychosis (n = 26) were scanned twice using magnetic resonance imaging, at first presentation and after a 3-year follow-up period. An age-matched group of healthy volunteers (n = 17) was scanned at the same time points. Within-group and between-group changes in regional grey matter volume were examined using voxel-based morphometry. There were significant group by time interactions (pFDRcorr < 0.05) in the frontal, temporal, parietal, cerebellar cortex, and in the thalamus, mainly reflecting longitudinal reductions in the controls but not in the patients. Subdivision of the patient group revealed that there were similar longitudinal reductions in patients with affective psychoses as in the controls but no volumetric changes in patients with schizophrenia. Psychosis with onset in adolescence or early adulthood may be associated with a delay or a loss of longitudinal reductions in regional grey matter volume that normally occur at this stage of development. These changes may be specific to schizophrenia.
Normal brain development is characterised by a preadolescent increase in grey matter (GM) volume, with GM volume in frontal and parietal regions maximal at 12 years and temporal GM volume peaking at 16 years of age (Giedd et al., 1999). This is followed by a loss of GM in the frontal and parietal cortices that continues through adolescence and into adulthood (Giedd et al., 1999). Data from volumetric neuroimaging studies suggest that brain development may be disrupted in both schizophrenia and affective psychoses (Hirayasu et al., 1998, Pantelis et al., 2005, Moorhead et al., 2007 and Koo et al., 2008). In adolescents with childhood-onset schizophrenia, an accelerated rate of GM loss over a 2-year period has been reported in the frontal, cingulate, parietal and temporal cortex (Rapoport et al., 1999, Farrow et al., 2005 and Thompson et al., 2001). In adolescents with a first episode of affective psychosis, longitudinal reductions in the inferior temporal and anterior cingulate gyri over the first 2 years of illness have been reported (Farrow et al., 2005). On the other hand, longitudinal increases in global cortical GM volume in first episode affective psychosis have been reported in association with mood stabilizer treatment (Nakamura et al., 2007).
While these findings suggest that there may be longitudinal volumetric changes in psychotic disorders, the extent to which the results depend on the type of disorder (schizophreniform or affective), the stage of the illness, and the effects of treatment is still unclear. We sought to address these issues in the present study. Our first aim was to examine, using voxel-based morphometry (VBM), longitudinal changes in GM volume in the first 3 years after the onset of psychosis in adolescent and young adult patients. The second objective was to compare these data in patients with schizophrenia and affective psychoses. The first hypothesis was that the onset of psychosis in adolescence or early adulthood would be associated with an alteration in the normal pattern of longitudinal changes in regional GM volume. Our second hypothesis was that this alteration would be more evident in patients with schizophrenia than in those with an affective psychosis (Kasai et al., 2003 and Farrow et al., 2005).
