Cerebellar grey matter deficits in first-episode schizophrenia mapped using cortical pattern matching
Introduction
Cerebellar dysfunction in schizophrenia has gained substantial attention in recent years (e.g., Andreasen and Pierson, 2008, Picard et al., 2008, for a review) and led to the notion of “cognitive dysmetria”—a syndrome characterised by neurological soft signs, impaired coordination, abnormal posture and proprioception, impaired eye-blink conditioning, vestibulo-ocular dysadaptation, and poor performance on procedural learning tasks. Post mortem micro- and macroscopic investigations identified the vermis region as most affected in schizophrenia. In their early studies, Weinberger and colleagues (1980) reported anterior vermal atrophy accompanied by Purkinje and granule cell loss, thus resulting in thinning of the granular and molecular layers (Martin and Albers, 1995, for review). These early post mortem findings are consistent with reports derived from in vivo brain imaging research confirming vermal atrophy by computerised tomography (e.g., Weinberger et al., 1979, Heath et al., 1979, Lippmann et al., 1982, Dewan et al., 1983) and magnetic resonance imaging (MRI; e.g., Rossi et al., 1993, Nopoulos et al., 1999, Volz et al., 2000, Ichimiya et al., 2001, Loeber et al., 2001, Joyal et al., 2004, Okugawa et al., 2002, Okugawa et al., 2003, Okugawa et al., 2007). More recent reports also suggest an altered proportional relationship of vermal grey to white matter in schizophrenia (e.g., Lawyer et al., 2009).
Meta-analysis of voxel-based morphometry data by Honea and colleagues (2005) conducted on 15 studies with a total of 390 patients and 364 healthy volunteers confirmed predominantly left-hemispheric cerebellar atrophy in schizophrenia. However, two studies did not include the cerebellum and only four studies independently reported significant cerebellar atrophy. Posterior cerebellar atrophy and increased grey matter in the medial cerebellum and culmen was detected in a sample of 169 patients from the National Institute of Mental Health Genetic Study of Schizophrenia when comparing patient MRI data with data from 212 healthy volunteers by optimised voxel-based morphometry (Honea et al., 2008). Left cerebellar atrophy was also reported for 213 unaffected siblings, although this finding was not confirmed by intra-class correlation analysis conducted on 116 sibling pairs.
MRI studies have also investigated structural/functional relationships and, for instance, have shown that cerebellar atrophy is linked to impaired motor sequencing (Venkatasubramanian et al., 2008). Other research has linked cerebellar dysfunction to neurological soft signs, abnormal posture and gait in schizophrenia (Andreasen and Pierson, 2008, for review). Moreover, functional brain imaging research has also provided some evidence for a cerebellar contribution to neurocognitive impairment in schizophrenia (see Picard et al., 2008, for review).
Clinical lesion studies (Schmahmann et al., 2008) suggest that abnormal vermal-fastigial function can contribute to delusion symptoms, impaired control of attention and affect, and social dysfunction (including autistic spectrum symptoms and impaired Theory of Mind performance) via interactions with the anterior thalamic nuclei, hippocampus, septum, amygdala, ventral tegmental area, periaqueductal grey and mamillary bodies relevant for memory and emotion. This spectrum of putatively cerebellar psychopathology is largely consistent with some of the defining symptoms of schizophrenia, such as psychosis, inattention, flat affect, and social withdrawal.
When investigating associations of neurocognitive function with cerebral structure, Segarra and colleagues (2008) reported a correlation of vermal grey matter atrophy with working memory deficits. By contrast, increased vermal white matter volumes have been linked to poor verbal fluency performance in schizophrenia (Lee et al., 2007). Smaller volumes of the posterior and superior vermis were also found to correlate with impaired cognition in drug-naïve first-episode schizophrenia patients (Okugawa et al., 2007) while positive symptom ratings were reported to correlate with vermal white matter volumes in early-onset schizophrenia (Yoshihara et al., 2008). Moreover, cerebellar atrophy has also been detected in individuals followed longitudinally from the prodromal phase of the illness (Pantelis et al., 2003, Borgwardt et al., 2008). However, in contrast to general cerebral atrophy, progression of vermal atrophy has not been confirmed by repeated MRI in childhood-onset schizophrenia patients (Keller et al., 2003). Together, these findings support the notion of clinically relevant cerebellar neuropathology in schizophrenia that is already present in the emerging phase of illness and, to some extent, also appears to be present in unaffected biological relatives (Honea et al., 2008).
We investigated regional cerebellar grey matter in well-remitted first-episode schizophrenia outpatients with less than 2 years’ duration of illness by applying a novel MRI brain imaging analysing technique. This method has been widely used for mapping abnormalities of the cerebral cortex (Thompson et al., 2004) and is applied here to the cerebellum for the first time.
The analysis methods preserve the three-dimensional lobular information of the cerebellum when generating statistical maps for group comparisons. In other words, data from the same lobules are averaged together across subjects when generating the average maps. Surface-based anatomical landmarks are used to enforce more accurate co-registration of individual lobule anatomy and more precise mapping of group-averaged surface anatomy. This is of particular importance when investigating potential regional grey matter differences in clinical populations where brain pathology can render more common co-registration algorithms less accurate, and where anatomical variance may be greater than in controls (Narr et al., 2001, Thompson et al., 1998).
We predicted that there would be cerebellar grey matter deficits in first-episode schizophrenia that would also affect grey to white matter ratios (Lawyer et al., 2009), with the most pronounced regional grey matter deficit in the vermis area in this early phase of illness (Okugawa et al., 2007).
Section snippets
Methods
Ethics approval for this study was granted by the human research ethics committees of the University of Newcastle and Hunter New England Health. Participants gave written informed consent.
Results
Total cerebellar volume and total grey matter volume of first-episode schizophrenia patients did not differ from healthy control subjects (Table 1). By contrast, first-episode schizophrenia was associated with increased total white matter volume (t = −2.17; P < 0.05) and smaller total grey to white matter ratios (t = 2.91; P < 0.01).
Grey matter deficits in first-episode schizophrenia patients were confirmed by permutation testing at P < 0.005 for the left and at P < 0.003 for the right hemisphere of the
Discussion
In terms of overall volume, the cerebellum was not atrophic in first-episode schizophrenia, but patients exhibited increased total white matter volumes, resulting in smaller global grey to white matter ratios. Regional grey matter deficiencies were also detected. These findings suggest that grey matter loss contributing eventually to global cerebellar atrophy may be partly compensated initially by increased volumes of global white matter and, furthermore, that grey matter deficits are
Acknowledgments
The study was supported by project grants of the National Health & Medical Research Council of Australia (300734), Pfizer Neuroscience, the Hunter Medical Research Institute, and the University of Newcastle. We gratefully acknowledge infrastructure support from the Hunter Medical Research Institute, the Schizophrenia Research Institute, and New South Wales Health and valuable assistance by V. Case, G. Cooper, D. Draganic, R. Fulham, S. Hudson, G. O'Connor. Algorithm development for this study
References (52)
- et al.
The role of the cerebellum in schizophrenia
Biological Psychiatry
(2008) Structural neural networks subserving oculomotor function in first-episode schizophrenia
Biological Psychiatry
(2004)Eye-blink conditioning deficits indicate temporal processing abnormalities in schizophrenia
Schizophrenia Research
(2009)Reductions in frontal, temporal and parietal volume associated with the onset of psychosis
Schizophrenia Research
(2008)Cerebellar morphology in chronic schizophrenic patients: a controlled computed tomography study
Psychiatry Research
(1983)Is gray matter volume an intermediate phenotype for schizophrenia? A voxel-based morphometry study of patients with schizophrenia and their healthy siblings
Biological Psychiatry
(2008)Reduced volume of the cerebellar vermis in neuroleptic naïve schizophrenia
Biological Psychiatry
(2001)- et al.
The neural basis of smooth pursuit eye movements in the rhesus monkey brain
Brain and Cognition
(2008) MRI volumetry of the vermis and the cerebellar hemispheres in men with schizophrenia
Psychiatry Research
(2004)Functional magnetic resonance imaging studies of eye movements in first episode schizophrenia: smooth pursuit, visually guided saccades and the oculomotor delayed response task
Psychiatry Research
(2006)