Widespread decreased grey and white matter in paediatric obsessive-compulsive disorder (OCD): A voxel-based morphometric MRI study
Introduction
Obsessive-compulsive disorder (OCD) is one of the most disabling anxiety disorders marked by recurrent, intrusive and distressing thoughts (obsessions) and/or repetitive behaviours (compulsions) (American Psychiatric Association, 2000). It is common in children and adolescents with a point prevalence of 2–3% (Rapoport et al., 2000). Further, it has clear continuity into adulthood, with approximately 80% of adult OCD sufferers experiencing the onset of their OCD symptoms before the age of 18 years McGuire et al. (2012).
The emergence and development of imaging techniques have greatly contributed to the understanding of neuropsychiatric disorders such as OCD. Several excellent reviews of the imaging literature in adult OCD have appeared (Saxena and Rauch, 2000, Friedlander and Desrocher, 2006, Fontenelle et al., 2009, Kwon et al., 2009). Most neuroimaging studies in OCD have targeted brain regions identified as abnormal, specifically fronto–striatal–thalamic neural networks, involving the prefrontal regions, together with the cingulate cortex. This is consistent with current aetiological models of adult OCD that implicate abnormal feedback loops within large-scale neurocognitive networks, specifically ‘limbic’ or ‘affective’ cortico–striato–thalamic circuits, including the orbitofrontal cortex (OFC), that have been hypothesised to play a key role in pathophysiology in OCD (Graybiel and Rauch, 2000, Saxena et al., 2001). Further, functional neuroimaging studies have implicated a relative increase in the activity of key subcortical and cortical brain regions that may aid our understanding of OCD. For instance, Insel and Winslow (1992) proposed that there may be increased activity in the caudate nucleus that inhibits globus pallidus outflow from the basal ganglia, which in turn increases thalamic activity, orbitofrontal cortical activity and caudate nucleus activity via the cingulate gyrus.
In contrast, few have comprehensively reviewed imaging studies in paediatric OCD (Huyser et al., 2009). Many brain morphology studies in paediatric OCD are based on volumetric region-of-interest (ROI) studies. Compared with age-matched typically developing participants, paediatric OCD patients have higher grey matter volumes in the anterior cingulate cortex (ACC) (Rosenberg and Keshavan, 1998, Szeszko et al., 2004, Szeszko et al., 2008) and in the OFC (Szeszko et al., 2008, Macmaster et al., 2010). The basal ganglia and thalamus have also been investigated with highly variable results: among the findings are a lower volume of the putamen but not of the caudate nucleus (Rosenberg et al., 1997b), lower volume of the globus pallidus but not of the putamen or caudate nucleus (Szeszko et al., 2004), and higher volume of the thalamus (Gilbert et al., 2000).
Recently, two voxel-based meta-analyses have been conducted examining grey matter differences in patients with OCD based on voxel-based morphometry (VBM) (Radua and Mataix-Cols, 2009, Rotge et al., 2010). Due to the limited VBM studies in the OCD literature, these meta-analyses have employed mixed samples from adult and paediatric studies. Lower grey matter volume has been found in people with OCD in regions such as the frontal eye fields (FEF), dorsolateral prefrontal cortex (DLPFC), medial part of the anterior prefrontal cortex (aPFC) (Rotge et al., 2010) and bilateral dorsal medial frontal/anterior circulate gyri (Radua and Mataix-Cols, 2009). Higher grey matter volumes have been found in people with OCD in basal ganglia regions such as the putamen and caudate (Radua and Mataix-Cols, 2009, Rotge et al., 2010), and the lateral part of the orbitofrontal cortex (OFC). One study found that there was no significant difference between adults and children with OCD (Rotge et al., 2010), while the other did not conduct such an analysis due to too few extant paediatric studies (Radua and Mataix-Cols, 2009).
In these few paediatric VBM studies, the results have been inconsistent with published ROI-based studies and between the different VBM studies. For example, either no difference (Szeszko et al., 2008) or lower grey matter volume in the ACC (Carmona et al., 2007, Gilbert et al., 2008) and lower grey matter volume in bilateral frontal regions (Carmona et al., 2007) have been found in paediatric OCD patients compared with typically developing participants. With regards to the basal ganglia and thalamus, studies have found higher volumes in the caudate nuclei and the right globus pallidus in adolescents with OCD (Zarei et al., 2011), higher volume in the right putamen (Gilbert et al., 2008, Zarei et al., 2011) and bilateral putamen (Szeszko et al., 2008), no thalamic difference (Carmona et al., 2007, Gilbert et al., 2008, Szeszko et al., 2008) or no basal ganglia differences (Carmona et al., 2007). In addition to those predicted regions, decreased grey matter volume bilaterally has been noted in the parietal lobes (Lazaro et al., 2009, Zarei et al., 2011).
Far fewer studies have examined differences in white matter in OCD and most findings are from adult samples: for example, significantly lower total white matter volume in female OCD patients (Jenike et al., 1996) along with decreased bilateral orbitofrontal volume (Szeszko et al., 1999) in OCD patients, although the latter study did not determine if the reduction was specific to grey or white matter due to methodological limitations. Recent white matter VBM studies in paediatric OCD suggest decreased white matter volume in bilateral frontal (Carmona et al., 2007) and right parietal regions (Carmona et al., 2007, Lazaro et al., 2009).
In addition, the corpus callosum has been noted to be larger (except for the isthmus) in paediatric OCD patients (Rosenberg et al., 1997a), and anomalies in signal intensity, width, and length of the corpus callosum have also been reported in OCD patients in controlled MRI studies [female adult (Jenike et al., 1996); paediatric (Rosenberg et al., 1997a); and adult (MacMaster et al., 1999)]. This is consistent with the corpus callosum being a predicted brain region of interest in OCD, given that primary and association cortices are topographically mapped in the corpus callosum (Raybaud, 2010).
In summary, although morphometric investigations of OCD have been variable, the most consistent findings are volumetric abnormalities (either decrease or increase) involving the ventral prefrontal cortex (which includes the OFC) and the striatum, as well as white matter abnormalities involving the cingulum bundle and the corpus callosum. The contradictory findings are most likely due to factors including the data analysis method (ROI or VBM), and/or different OCD sample selection criteria (age, gender, OCD symptom dimensions, comorbidity, length of illness, medication treatment and/or sample size). Despite these methodological shortcomings, the published data suggest that the parietal region could be a focus for further exploration, along with the ventral prefrontal–striatal regions, if a comprehensive pathophysiology of paediatric OCD is to emerge.
Hence, the current study sought to carefully define medication naïve children with similar OCD symptom dimensions and lengths of illness, without key comorbid conditions such as attention deficit hyperactivity disorder, combined type (ADHD-CT), combined type, major depressive disorder and tic disorders. Voxel-based morphometry was used to compare the brains of the paediatric OCD patients with those of typically developing participants, matched on age, gender, handedness and performance IQ. We hypothesised that compared with typically developing participants, young people with OCD would demonstrate grey and white matter abnormalities in a large-scale neural network of brain regions, consistent with theoretical models of OCD, including the ventral prefrontal and cingulate cortices, the corpus callosum and the parietal cortex.
Section snippets
Participants
Eight young people fulfilling DSM-IV criteria for obsessive-compulsive disorder (4 M:4 F) aged 8–15 years (mean 11.7±2.7 years) were identified at the Royal Children's Hospital, Melbourne, Australia. OCD was defined categorically, using a semi-structured clinical interview with one of both of the participants' parents and the young people themselves, the Anxiety Disorders Interview Schedule for Children (A-DISC; Silverman and Albano, 1996), in addition to using the clinician completed
Overall brain volume
Overall, young people with OCD demonstrated significantly lower ICV (grey matter+white matter+CSF), grey matter and white matter tissue volumes. The ICV was significantly reduced (∼9%) in the OCD group compared with the typically developing group (p=0.02 and d.f.=19). Significant decreases in absolute volumes in the OCD group were found in overall white (47 mL; p=0.03 and d.f.=19) and grey matter volumes (67 mL; p=0.02 and d.f.=19). Regional differences in grey matter and white matter segmented
Discussion
The VBM approach used in this carefully defined group of children and adolescents with OCD and matched typically developing participants yielded the following important findings: the expected brain volume differences were evident in grey matter ventral prefrontal cortex and the striatum, as well as white matter abnormalities involving the cingulum bundle and the corpus callosum. Further, as hypothesised, the parietal region was anomalous in the OCD group. Specifically, relative to typically
Acknowledgements
This work was supported by the Eric Ormond Baker Trust, by the National Health and Medical Research Council, by the Victorian Government's Operational Infrastructure Support Program and by the Royal Children's Hospital staff and patients. T.S. was supported by a NHMRC Career Development Award.
References (59)
A fast diffeomorphic image registration algorithm
NeuroImage
(2007)- et al.
Unified segmentation
NeuroImage
(2005) - et al.
Pediatric OCD structural brain deficits in conflict monitoring circuits: a voxel-based morphometry study
Neuroscience Letters
(2007) - et al.
Neuroimaging studies of obsessive-compulsive disorder in adults and children
Clinical Psychology Review
(2006) - et al.
Neural correlates of symptom dimensions in pediatric obsessive-compulsive disorder: a functional magnetic resonance imaging study
Journal of the American Academy of Child and Adolescent Psychiatry
(2009) - et al.
Gray matter differences between pediatric obsessive-compulsive disorder patients and high-risk siblings: a preliminary voxel-based morphometry study
Neuroscience Letters
(2008) - et al.
Toward a neurobiology of obsessive-compulsive disorder
Neuron
(2000) - et al.
Heterarchical reinforcement-learning model for integration of multiple cortico-striatal loops: fMRI examination in stimulus–action–reward association learning
Neural Networks
(2006) - et al.
Paediatric obsessive-compulsive disorder, a neurodevelopmental disorder? Evidence from neuroimaging
Neuroscience and Biobehavioral Reviews
(2009) - et al.
Neurobiology of obsessive compulsive disorder
Psychiatric clinics of North America
(1992)