Set-shifting related basal ganglia deformation as a novel familial marker of obsessive-compulsive disorder

Abstract

The symptoms of obsessive-compulsive disorder (OCD) are suggestive of cognitive rigidity, and previous work identified impaired flexible responding on set-shifting tasks in such patients. The basal ganglia are central to habit learning, and are thought to be abnormal in OCD, contributing to inflexible, rigid habitual patterns of behaviour. Here, we demonstrated that increased cognitive inflexibility indexed by poor performance on the set-shifting task correlated with putamen morphology, and that patients and their asymptomatic relatives had common curvature abnormalities within this same structure. The association between the structure of the putamen and the ED errors was found to be significantly familial in OCD proband-relative pairs. The data implicate changes in basal ganglia structure linked to cognitive inflexibility as a familial marker of OCD. This may reflect a pre-disposing heightened propensity towards habitual response patterns and deficits in goal-directed planning.

Introduction

Obsessive-Compulsive Disorder (OCD) is a debilitating neuropsychiatric condition characterized by repetitive intrusive thoughts and/or repetitive behaviours undertaken according to rigid rules (or in response to obsessions) (1). Neurobiological models of OCD emphasise dysfunction of the basal ganglia, implicated in generating habitual behaviour (2, 3). In a meta-analysis of neuroimaging studies, structural abnormalities of the basal ganglia were identified in OCD patients versus controls, coupled with abnormal functional activation in this region during tests of executive functions, including cognitive flexibility (4). Other meta-analysis did not find macroscopic volumetric basal ganglia abnormalities in OCD (5), but inconsistencies may reflect variable methodological quality of the included data studies. Whereas most studies focused on total volumes of whole subcortical structures, advances in imaging pipelines have made it possible to examine detailed, localised changes in the surface morphology (i.e. shape) of basal ganglia structures (6). This is a promising approach to identify familial markers of OCD, because morphological basal ganglia changes were previously found in OCD patients and their relatives, when conventional volumetric measures were insensitive (6). Furthermore, understanding of the basal ganglia is also relevant for trans-diagnostic approaches, because their dysfunction is implicated in other less-studied obsessive-compulsive related disorders (7).

The basal ganglia play a prominent role in habit learning and regulation (i.e. in cognitive flexibility) (7). One common laboratory-assessed form of cognitive flexibility is set-shifting and it refers to the ability to switch attention from one important aspect of the environment to another during a task, in response to changing reinforcement contingencies (8). Set-shifting is classically measured by an ‘Extra-Dimensional’ (‘ED’) shift in such tasks, whereby participants inhibit attention away from a previously relevant aspect of the environment, onto a different aspect of the environment that was previously irrelevant. For more detailed discussion of the cognitive neuroscience of ED shifting and its relationship to habit and loss of top down flexibility, see (9, 10). Abnormalities in this cognitive function have been extensively implicated in the pathophysiology of OCD (8). Impaired set-shifting performance on the Intra-Dimensional/Extra-Dimensional (IDED) task has been confirmed in a meta-analysis of OCD studies with medium-large effect size, and has also been found to extend to clinically asymptomatic first-degree relatives of patients, thereby constituting a candidate familial marker (10).

Given that basal ganglia morphological changes and impaired set-shifting have separately been implicated as candidate familial markers for OCD in prior work, this study sought to demonstrate a link between the two. We hypothesised (1) that basal ganglia shape would relate to set-shifting performance overall; and (2) that morphological abnormalities would be found in OCD patients and relatives, versus controls, in the identified set-shifting related brain regions.

Methods

Subjects

The pooled dataset included participants from two studies, where participants underwent conventional imaging measures, reported previously (11, 12) (technical details of the scans are provided in the Supplement). The first sample comprised 32 pairs of unaffected first-degree relatives and OCD patient probands, and 32 age/gender matched healthy controls (7). The second sample comprised 44 OCD patients and 43 age/gender matched healthy controls (8).

Participants were screened for mental disorders using extended clinical interview by a psychiatrist, supplemented with the Mini International Neuropsychiatric Inventory (MINI). OCD severity was quantified using the Yale-Brown Obsessive Compulsive Scale (Y-BOCS), and Intelligence Quotient (IQ) was measured using the National Adult Reading Test (NART). OCD patients were excluded if they had any comorbidities, including depression. Controls and first-degree relatives of patients were excluded if they had any history of mental disorders. Controls were excluded if they reported any known history of OCD in first-degree family members. History of serious head injury, substance abuse, epilepsy or MRI contraindications, was exclusionary across all groups.

The study was approved by research ethics committee (Addenbrooke’s NHS Trust Local Research Ethics Committee, Cambridge, UK; and Local Research Ethical Committee at the University of Cambridge, Cambridge, UK). All participants provided written informed consent after reading information sheets about the study and having the opportunity to ask questions.

Behavioural testing

Prior to scanning, all participants undertook the computerised Intra-Dimensional/Extra-Dimensional Set Shift task (IDED) from the Cambridge Neuropsychological Test Automated Battery (CANTAB). Two stimuli were displayed on-screen for each trial, and participants attempted to learn underlying rules governing which stimulus was correct, based on computerised feedback. Set-shifting performance on the task, as indexed by errors on the crucial extra-dimensional attentional shift stage. Of the different stages of the task (i.e., learning, intra-dimensional set shifting) extra-dimensional attentional shift stage was previously found to be selectively impaired in OCD, in meta-analysis (10). ED shifting is widely regarded as the archetypal form of set-shifting on such tasks. A more detailed task description is provided in the Supplement.

Data Analysis

To address the two study hypotheses, our overall a priori analytic approach was: 1) to identify morphological regions of basal ganglia structures related to set-shifting performance; and then 2) to examine whether or not morphology in these set-shifting related areas was abnormal in OCD patients and their relatives versus controls.

Following standard pre-processing pipelines (for details see Supplement), vertex analysis was deployed to investigate the shapes of subcortical structures, using FIRST from the FMRIB Software Library (FSL) (13). The vertex locations of each participant were projected onto the surface normal of the average shape template of the 336 training subjects provided by FSL, and the perpendicular distance from the average surface was calculated. Positive value of the vertex indicated deformation in the outward direction, and negative value of a vertex indicated deformation in the inward direction. ‘Randomise’, a permutation-based non-parametric testing method implemented in FSL (5000 iterations), was used to identify sub-cortical structures whose shape was significantly associated (i.e. correlated with) with ED errors, across all subjects (14). Based on extant models of OCD, and structures implicated in flexibility, we focused on caudate, putamen, globus pallidus, and nucleus accumbens (2, 3). Multiple comparison correction at p<0.05 using Threshold-Free Cluster Enhancement (TFCE) was used.

Group differences in demographic data were explored using analysis of variance (ANOVA) or chi-square tests as appropriate. Group differences in sub-cortical clusters associated with ED performance were explored using non-parametric Kruskal-Wallis tests to identify overall group differences; and Wilcoxon tests to compare groups pair-wise if overall effects of groups were found. Correlations (Spearman’s r) were examined between patient-relative pairs in terms of curvature in the identified clusters, and between randomly permuted pairs of subjects, to identify whether the ED related structural changes were significantly familial. Statistical significance was defined as p<0.05 uncorrected.

Results

The groups did not differ significantly from each other in terms of age, gender, and estimated verbal IQ (all p>0.10; Supplementary Table 1). 51 (67.1%) of patients in the OCD group were receiving psychotropic medication. The mean (Standard Deviation, SD) total Y-BOCS scores in the OCD patients was 21.89 (5.30).

Two significant brain clusters were identified in which curvature was significantly associated with the number of ED errors on the IDED task across all study participants (Figure 1, Left Panel). One cluster (overall p=0.0380) was in the left putamen, maximal at [X=-32, Y=-8, Z=-3], and of extent 363 voxels. The other cluster (overall p=0.0252) was in the left pallidum, maximal at [X=-21, Y=-15, Z=-2], and of extent 216 voxels. Scanner type did not significantly impact curvature in these clusters (each p>0.50 by Wilcoxon test comparing cluster parameters between sites).

Figure 1. Left Panel: Basal ganglia regions in which structural curvature was significantly associated with ED shift performance. Blue indicates regions in which ED errors were associated with greater inward curvature; red indicates regions in which ED errors were associated with greater outward curvature. Green indicates the remainder of the given structure. Right Panel: Violin plots showing curvature distributions, in each study group, for the clusters significantly associated with ED performance. ** p<0.01 significant difference between groups by Wilcoxon test. Only group differences in the left putamen cluster were significant.

Groups differed significantly in curvature in the left putamen cluster (Kruskal-Wallis test = 10.001, p=0.007), but not significantly in the left pallidum cluster (Kruskal-Wallis test = 2.308, p=0.315). As shown in Figure 1 (Right Panel), OCD patients had significant inward curvature deformation in the left putamen cluster versus controls (p=0.008), as did their relatives (p=0.008). The group difference in the left putamen cluster was also significant in each study dataset considered separately: Study 1, p=0.0270; Study 2, p=0.0353. Patients did not differ significantly from relatives on this measure (p=0.3661).

Mean curvature in the ED-associated putamen cluster correlated significantly between OCD patient-relative pairs (r²=0.261, p<0.001), but not randomly permuted pairs (p>0.10; bootstrap N=2500, 95% confidence interval for the null model p was 0.37 – 0.39).

Patients with OCD who were receiving psychotropic medication did not differ from patients not receiving such medication, in terms of curvature in the putamen cluster (Wilcoxon test, p=0.78). Curvature in the putamen cluster did not significantly correlate with YBOCS total scores in the OCD patients (p=0.233).

As expected, and previously reported, OCD patients and their relatives were selectively impaired on the ED stage of the IDED task versus controls (Supplementary Results).

Discussion

By conducting an analysis of cognitive function and basal ganglia morphology from magnetic resonance imaging (MRI) data, this study (1) showed that set-shifting performance (ED shifting) correlated with putamen morphology; and (2) demonstrated that OCD patients and their asymptomatic first-degree relatives had common abnormalities in this set-shifting related region. Furthermore, the association between the structure of the putamen and the ED errors was found to be significantly familial in OCD proband-relative pairs. These structural brain changes in OCD patients and their relatives reflect localised (mm level) changes in the shape of specific basal ganglia structures.

The implication of these findings is that putamen shape relates to set-shifting performance, and that deformation of this structure constitutes a related candidate familial marker for OCD. The findings may reflect clustering of different groups in separate areas of the bi-dimensional (i.e. ED shift vs brain structure) variable space. Previous translational research indicate that the putamen plays a cardinal role in learning, adapting stimulus-response associations, and in habitual behaviour (3, 15). ED errors in OCD were previously found to be associated with reduced frontal connectivity between the dorsal striatum and frontal cortical regions (12). Thus, predilection towards habitual response patterns and impaired goal-directed control, linked to morphological changes of the basal ganglia, appear to confer familial risk for OCD. The morphological changes do not appear to be diagnostic markers, nor a consequence of symptoms themselves, nor secondary to psychotropic medication. Structural abnormalities of the basal ganglia may reflect altered neurodevelopmental trajectories associated with the risk of OCD, but to address this would require longitudinal research.

Data Availability

The data are not publicly available as participants did not consent to that.