Differential effects of surface area, gyrification and cortical thickness on voxel based morphometric deficits in schizophrenia
Highlights
► Surface anatomical changes contribute to a proportion of gray matter deficits in VBM. ► The influence of surface anatomy on VBM deficits is regionally variable. ► Thickness, area and gyrification contribute to VBM deficits in schizophrenia.
Introduction
Voxel Based Morphometry (VBM) is a widely used method to quantify gray matter abnormalities in disease states (Ashburner and Friston, 2000). Since its initial application to study structural changes in schizophrenia (Wright et al., 1995), the technique of VBM has evolved significantly. One of the major criticisms of VBM is that the technique is very sensitive to the image registration procedures. Conditions that result in systematic differences in image alignment may produce spurious results (Bookstein, 2001), though the significance of this is refuted (Ashburner and Friston, 2001). This issue is especially relevant for conditions such as schizophrenia, where abnormalities of sulcal patterning around the lateral fissure (Csernansky et al., 2008) can produce group differences in aligning gray matter images for VBM, in particular affecting the perisylvian regions such as insula and the surrounding operculum that are commonly observed to have GMV deficit in schizophrenia (Koutsouleris et al., 2008, Meisenzahl et al., 2008). Recent updates of VBM improve on registration methods (Ashburner, 2007), thus addressing the criticisms raised by Bookstein (2001) to certain extent.
Surface based morphometric (SBM) methods such as Freesurfer circumvent some of the problems associated with VBM by undertaking computations of morphometric properties in the ‘native space’ (Dale et al., 1999). The computationally intensive surface based techniques provide distinct cortical thickness and surface area measures that are shown to be both genetically and phenotypically independent (Panizzon et al., 2009, Winkler et al., 2010). Group differences emerging in a VBM study could be variously attributed to cortical thinning, altered gyrification (cortical folding) or abnormalities in the surface area (Mechelli et al., 2005). Despite the increasing number of studies that employ both VBM and SBM together to study brain structure in disease states (Cerasa et al., 2011, Chee et al., 2011, Lehmann et al., 2011), the relationship between VBM derived gray matter volume (GMV) and SBM derived measures of thickness, gyrification and surface area is unclear.
The importance of structural changes in understanding the pathophysiology of schizophrenia has been well established (McCarley et al., 1999, Shenton et al., 2001). A significant number of VBM investigations of the neuroanatomy of schizophrenia have been published in the last two decades (Ellison-Wright et al., 2008, Glahn et al., 2008, Honea et al., 2005, Leung et al., 2011). Despite the significant spatial heterogeneity of the reported findings, meta-analysis of VBM studies estimates a high likelihood of gray matter deficits in specific regions such as bilateral insula, temporal cortex and anterior cingulate in schizophrenia (Ellison-Wright et al., 2008, Glahn et al., 2008, Honea et al., 2005, Leung et al., 2011). Cortical thickness (Kuperberg et al., 2003) and surface contraction maps (Palaniyappan et al., 2011a) identify regions of structural changes that overlap to some extent with the regions showing GMV reduction in VBM. Focused analyses of the spatial overlap between VBM and SBM in schizophrenia have previously revealed that either thickness or surface area changes were present in clusters that had significant reduction in GMV (Narr et al., 2005, Voets et al., 2008) with some suggestion that cortical folding differences could account for the some of the regional differences. These findings indicate that changes in the anatomical properties of the cortical mantle (thickness, cortical folding or surface area) may underlie the GMV reduction seen in schizophrenia.
In the present study, we investigate the relationship between the GMV and anatomical properties of the cortical mantle in regions showing significant VBM changes in schizophrenia. We hypothesized that SBM measures will mediate the group differences in GMV observed in the VBM analysis. Given the partial overlap in the spatial distribution of VBM and cortical thickness measured using both surface-based (Narr et al., 2005, Voets et al., 2008) and voxel-based methods (Hutton et al., 2009) in previous studies, we expected regional differences in the influence of surface based measures on VBM findings.
Section snippets
Subjects
Data acquired from 98 subjects (57 with schizophrenia and 41 healthy controls) were used in this study (Table 1). Regional Ethics Committees (Nottinghamshire and Derbyshire) approved the study and all participants provided written informed consent. This sample has been described in detail elsewhere in our reports on surface area and gyrification abnormalities (Palaniyappan et al., 2011a, Palaniyappan et al., 2011b). The diagnosis of schizophrenia was made in accordance with the procedure of
Voxel Based Morphometry
The VBM comparison between patients and controls revealed five clusters with significant reduction in GMV (Web Supplement DS1). There were no regions with increased GMV in schizophrenia. As expected, bilateral insula emerged as regions of significant GMV reduction, along with left thalamus, left precuneus and left middle temporal region. Binary masks were derived for right and left insula in addition to left middle temporal and left precuneus clusters. The MNI coordinates of the four cortical
Discussion
Using VBM and SBM on a cross sectional sample with schizophrenia, we have shown that the differences in GMV observed using VBM are partially mediated by surface anatomical properties such as gyrification, surface area and thickness. Reduced gyrification and surface area were observed in three out of four clusters examined, while reduced thickness was observed in two clusters. But the mediating effect of the surface anatomical features on GMV is regionally variable with a large proportion of the
Funding
This work was supported by a New Investigator grant from the University of Nottingham and an Interdisciplinary Research Award from the Nottingham Institute of Neuroscience, University of Nottingham.
Acknowledgments
We are grateful to the volunteers who participated in this study and would like to acknowledge Pavan Mallikarjun and Verghese Joseph for clinical recruitment. We would like to thank Thomas White, Kathrin Doege, Antonio Napolitano, Kay Head, Dawn-Marie Walker and Dorothee Auer for assisting the data acquisition.
References (51)
- et al.
Comparing surface-based and volume-based analyses of functional neuroimaging data in patients with schizophrenia
NeuroImage
(2008) A fast diffeomorphic image registration algorithm
NeuroImage
(2007)- et al.
Voxel-based morphometry—the methods
NeuroImage
(2000) - et al.
Why voxel-based morphometry should be used
NeuroImage
(2001) “Voxel-based morphometry” should not be used with imperfectly registered images
NeuroImage
(2001)- et al.
Dysbindin C-A-T haplotype is associated with thicker medial orbitofrontal cortex in healthy population
NeuroImage
(2011) - et al.
Symmetric abnormalities in sulcal patterning in schizophrenia
NeuroImage
(2008) - et al.
Cortical surface-based analysis: I. Segmentation and surface reconstruction
NeuroImage
(1999) - et al.
Cortical surface-based analysis: ii: inflation, flattening, and a surface-based coordinate system
NeuroImage
(1999) - et al.
Meta-analysis of gray matter anomalies in schizophrenia: application of anatomic likelihood estimation and network analysis
Biol. Psychiatry
(2008)
Accurate and robust brain image alignment using boundary-based registration
NeuroImage
A comparison between voxel-based cortical thickness and voxel-based morphometry in normal aging
NeuroImage
Gyral and sulcal cortical thinning in adolescents with first episode early-onset psychosis
Biol. Psychiatry
Structural correlates of psychopathological symptom dimensions in schizophrenia: a voxel-based morphometric study
NeuroImage
Cortical thickness and voxel-based morphometry in posterior cortical atrophy and typical Alzheimer's disease
Neurobiol. Aging
MRI anatomy of schizophrenia
Biol. Psychiatry
Structural brain alterations at different stages of schizophrenia: a voxel-based morphometric study
Schizophr. Res.
Regional contraction of brain surface area involves three large-scale networks in schizophrenia
Schizophr. Res.
Folding of the prefrontal cortex in schizophrenia: regional differences in gyrification
Biol. Psychiatry
Cortical thickness is influenced by regionally specific genetic factors
Biol. Psychiatry
A review of MRI findings in schizophrenia
Schizophr. Res.
Automated anatomical labeling of activations in SPM using a macroscopic anatomical parcellation of the MNI MRI single-subject brain
NeuroImage
Evidence for abnormalities of cortical development in adolescent-onset schizophrenia
NeuroImage
Heritability of volume, surface area and thickness for anatomically defined cortical brain regions estimated in a large extended pedigree
NeuroImage
Cortical thickness or grey matter volume? The importance of selecting the phenotype for imaging genetics studies
NeuroImage
Cited by (107)
Brain structural analysis in patients with post-traumatic anosmia: Voxel-based and surface-based morphometry
2023, Journal of NeuroradiologyIncreased cortical surface area but not altered cortical thickness or gyrification in bipolar disorder following stabilisation from a first episode of mania
2023, Progress in Neuro-Psychopharmacology and Biological PsychiatryCitation Excerpt :To explore this, we re-ran the surface area analysis excluding intracranial volume as a covariate, but no significant clusters were identified at all using this method. Differences in the statistical packages employed to analyse the data might also offer some explanation, as there is evidence that only a proportion of variance in VBM derived-grey matter volume (used in the previous analyses) is explained by the Freesurfer derived surface-based morphology measures used here (Palaniyappan and Liddle, 2012). Indeed, similar findings of an absence of regional overlap in Freesurfer or VBM-derived brain measures have been documented in other psychiatric and neurological illnesses, where it has been argued that grey matter volume effects may not be apparent since the combined effects of thickness and surface area may occur in opposite directions (Gerrits et al., 2016; Palaniyappan and Liddle, 2012; Palaniyappan et al., 2011).
Exploring the heterogeneous morphometric data in essential tremor with probabilistic modelling
2023, NeuroImage: ClinicalStructural brain imaging abnormalities correlate with positive symptom in schizophrenia
2022, Neuroscience LettersDissociations in cortical thickness and surface area in non-comorbid never-treated patients with social anxiety disorder
2020, EBioMedicineCitation Excerpt :Third, GMV is a complicated parameter, driven by many factors such as CT, CSA, and cortical folding (gyrification) [24], so group differences in GMV may be not necessarily explained by those in CT or CSA; a cluster with altered GMV may show no differences in CT or CSA [39]. Fourth, VBM and SBM differ in terms of methodology and biology, and SBM is more sensitive to abnormalities of gray matter structure in some circumstances, and differences in results from two methods have been reported [106,107]. Turning now to sampling, medications and comorbidity can confound results.