A functional MRI comparison of patients with obsessive–compulsive disorder and normal controls during a Chinese character Stroop task

Abstract

Recent functional neuroimaging and neuropsychological studies have suggested that abnormal activity in the anterior cingulate cortex (ACC) might cause an action-monitoring dysfunction in obsessive–compulsive disorder (OCD). To identify the relationship between brain dysfunction and cognitive dysfunction, we examined regional brain changes in OCD with functional magnetic resonance imaging (fMRI) during the performance of a cognitive task. Participants comprised 24 patients with OCD and 14 normal controls. First, we compared the cognitive function in the two groups as assessed by several neuropsychological tests. Then we used fMRI to explore brain correlates of their performance during the Chinese character version of the Stroop test, a task that is strongly related to action-monitoring function. The two groups did not differ on the neuropsychological tests. Both groups also showed similar activation pattern on fMRI. The patients, however, showed weaker activation than the normal controls in the ACC and the right caudate nucleus.

Introduction

Recent functional neuroimaging studies of obsessive–compulsive disorder (OCD) have suggested that the frontal cortex and subcortical structures may play a role in the pathophysiology of the disorder. Several studies of OCD using positron emission tomography (PET) have identified abnormally high functional activity in local regions such as the orbitofrontal cortex (OFC), the caudate nucleus and the anterior cingulate cortex (ACC) (Baxter et al., 1987, Swedo et al., 1989). Studies using single photon emission computed tomography (SPECT) have also suggested dysfunction of the OFC and the caudate nucleus in OCD (Machlin et al., 1991, Busatto et al., 2000). Each study reported increased and/or decreased activity in local regions such as the ACC, the OFC and the caudate nucleus. It has been postulated that a circuit involving the cortico/limbic–basal ganglionic–thalamic systems mediates the symptomatic expression of OCD (Saxena et al., 1998). This circuit might form a positive feedback loop and be excessively active. There is, however, no consensus on how this excessive activity relates to the symptoms of OCD.

Many studies have reported neuropsychological dysfunction in OCD. The results of these neuropsychological studies, however, have not been consistent. Several studies reported that patients with OCD were impaired on measures of executive function (Flor-Henry et al., 1979, Head et al., 1989, Savage et al., 1999). Other studies found nonverbal memory deficits (Christensen et al., 1992, Dirson et al., 1995). Deficits in visuospatial skill, selective attention and working memory have also been reported. One study found no evidence of neuropsychological abnormality in OCD (Abbruzzese et al., 1995). The complexity of neuropsychological functions may make it difficult to identify specific impairments in OCD.

Although the relationship between the brain and neuropsychological dysfunctions of OCD is an important one (Savage, 1998), few studies have examined this relationship with functional imaging approaches. A PET study (Martinot et al., 1990) found that glucose metabolic rates in the frontal cortices of patients with OCD were negatively correlated with subscores of the Stroop test. A SPECT study (Lucey et al., 1997) reported that perfusion in the left caudate and the left inferior frontal cortex of patients with OCD was positively correlated with the number of errors on the Wisconsin Card Sorting Test. Recently, a PET study found that glucose metabolic rate in the prefrontal cortex and putamen was correlated with performance on the Trail-Making Test, Part B, and the Rey-Osterrieth Complex Figure Test (Kwon et al., 2003). In all three of these studies, neuropsychological testing and functional neuroimaging were administered separately, with some time lag, not simultaneously. These studies, therefore, indirectly examined the relationship between the brain and neuropsychological dysfunctions of OCD. A few studies examined the relationship directly. A functional magnetic resonance imaging (fMRI) study (Pujol et al., 1999) reported significantly stronger activation of the left frontal cortex in patients with OCD compared with normal controls during performance of the Word Generation Test. Another fMRI study (Ursu et al., 2003) reported increased ACC activation in OCD during high conflict trials of the continuous performance task as analyzed with region-of-interest (ROI) methods. There is, however, no consensus on the relationship between brain dysfunctions and neuropsychological dysfunctions of OCD.

Recently, many researchers have focused on abnormal activity in the ACC as an important region underlying the symptomatology of OCD. Several previous neuroimaging studies reported hyperactivation of the ACC in OCD (Baxter et al., 1987, Swedo et al., 1989, Rauch et al., 1994). Recent neuropsychological studies have shown that the ACC might play a substantial role in action-monitoring, the ability to process competing types of information. Two studies using both neuropsychological and neuroimaging methods (Gehring et al., 2000, Ursu et al., 2003) indicated that the ACC might be involved in action-monitoring function.

On the basis of the previous studies, we decided to investigate the relationship of obsessive-compulsive (OC) symptoms to the action-monitoring function of the ACC. Our objective was to explore the relationship between regional brain function and neuropsychological function, especially action-monitoring function in OCD. We first compared the neuropsychological performance of patients with OCD with that of normal controls. Then we used fMRI to assess areas of activation during performance of the Stroop test, and compared activation patterns in the patients with those of the normal controls.