Follow-up MRI study of the insular cortex in first-episode psychosis and chronic schizophrenia

Abstract

Morphologic abnormalities of the insular cortex have been described in psychotic disorders such as schizophrenia, but it remains unknown whether these abnormalities develop progressively over the course of the illness. In the current study, longitudinal magnetic resonance imaging data were obtained from 23 patients with first-episode psychosis (FEP), 11 patients with chronic schizophrenia, and 26 healthy controls. The volumes of the short (anterior) and long (posterior) insular cortices were measured on baseline and follow-up (between 1 and 4 years later) scans and were compared across groups. In cross-sectional comparison at baseline, the FEP and chronic schizophrenia patients had significantly smaller short insular cortex than did controls. In longitudinal comparison, the FEP patients showed significant gray matter reduction of the insular cortex over time (− 4.3%/2.0 years) compared with controls (0.3%/2.2 years) without significant subregional effects, but there was no difference between chronic schizophrenia patients (− 1.7%/2.4 years) and controls. The gray matter loss of the left insular cortex over time in FEP patients was correlated with the severity of positive and negative symptoms at follow-up. These findings indicate that patients with psychotic disorders have smaller gray matter volume of the insular cortex especially for its anterior portion (short insula) at first expression of overt psychosis, but also exhibit a regional progressive pathological process of the insular cortex during the early phase after the onset, which seems to reflect the subsequent symptomatology.

Introduction

Structural brain abnormalities in schizophrenia have already developed by the onset of psychosis (Vita et al., 2006), suggesting a neurodevelopmental pathology (Weinberger, 1987). Recent longitudinal magnetic resonance imaging (MRI) studies in first-episode schizophrenia have demonstrated progressive ventricular expansion (Cahn et al., 2002, DeLisi et al., 1997, Nakamura et al., 2007, Puri et al., 2005) or volume reduction in frontal and temporal regions (Bachmann et al., 2004, Gur et al., 1998, Ho et al., 2003, Kasai et al., 2003a, Kasai et al., 2003b, Nakamura et al., 2007) in the initial years subsequent to the onset, possibly reflecting a pathological process in ‘late neurodevelopment’ (Pantelis et al., 2005, Pantelis et al., 2007). The few longitudinal studies that directly compare progressive brain changes in first-episode and chronic schizophrenia (Gur et al., 1998, Pantelis et al., 2008) have suggested that such changes are nonlinear and most prominent at the earliest phase of the illness.

The anatomical pattern of progressive brain changes in schizophrenia remains largely unknown. One voxel-based morphometric (VBM) study (Farrow et al., 2005) demonstrated that first-episode schizophrenia patients exhibit progressive gray matter reduction in lateral fronto-temporal and left cingulate regions, but a subsequent study by the same group using a modified methodology (Whitford et al., 2006) showed progressive changes predominantly in parietal cortex. Our recent study in first-episode schizophrenia (Sun et al., in press) based on a cortical pattern matching technique, which allows sensitive assessment of regional progressive changes throughout the lateral cortical surface, found increased brain surface contraction mainly in the dorsal prefrontal cortex. However, this approach cannot examine the cortical regions in deep sulci such as the insular cortex.

Neuroimaging investigations have shown that the pathological process in schizophrenia predominantly affects the fronto-temporolimbic-paralimbic regions, including insular cortex bilaterally (Glahn et al., 2008). Gray matter reduction of the insular cortex, which plays crucial roles in emotional and various cognitive functions as a component of the ‘limbic integration cortex’ (Augustine, 1996), has been repeatedly described in schizophrenia (Crespo-Facorro et al., 2000, Kasai et al., 2003c, Kim et al., 2003, Makris et al., 2006, Saze et al., 2007, Takahashi et al., 2004, Takahashi et al., 2005), although its pattern of topographically specific localization [i.e., sulcally defined and functionally different short (anterior) versus long (posterior) insular cortex (Augustine, 1996, Türe et al., 1999)] is still unclear. Gray matter reduction or dysfunction of the insula has been implicated in manifesting psychotic symptoms (Crespo-Facorro et al., 2000, Shapleske et al., 2002, Shergill et al., 2000) and cognitive impairments (Crespo-Facorro et al., 2001a, Crespo-Facorro et al., 2001b, Curtis et al., 1998). An inverse correlation between insular cortex volume and illness duration in schizophrenia (Takahashi et al., 2004, Takahashi et al., 2005) suggests a regional progressive pathological process in the course of the illness. Negative insular findings in the above-mentioned longitudinal studies based on statistical imaging techniques (Farrow et al., 2005, Whitford et al., 2006) might be due to lower sensitivity compared with manual region of interest (ROI) methods (Giuliani et al., 2005).

This study aimed to examine the progressive gray matter changes of the insular subregions in psychotic disorders using ROI analysis of longitudinal MRI data in both first-episode psychosis (FEP) and chronic schizophrenia patients compared with healthy controls. Based on previous studies, we predicted that both patient groups would show progressive insular cortex atrophy, but its degree would be greater in the FEP patients.