Effects of behavioral therapy or pharmacotherapy on brain glucose metabolism in subjects with obsessive–compulsive disorder as assessed by brain FDG PET

Abstract

This prospective study investigated the effect of pharmacotherapy (PT) and cognitive behavioral therapy (CBT) on cerebral glucose metabolism in adults with obsessive–compulsive disorder (OCD). Dynamic positron emission tomography (PET) of the brain with F-18-fluorodeoxyglucose (FDG) was performed before and after treatment in 16 subjects diagnosed for OCD for at least 2 years (PT: n = 7). Pre-to-post-treatment change of scaled local metabolic rate of glucose (SLMRGlc) was assessed separately in therapy responders and non-responders. Correlation was tested between SLMRGlc change and change of OCD, depression, or anxiety symptoms. SLMRGlc increased in the right caudate after successful therapy. The increase tended to correlate with the improvement of OCD symptom severity. The finding of increased local caudate activity after successful therapy is in contrast to most previous studies. Possible explanations include effects of therapy on concomitant depression symptoms and/or the large proportion of early-onset OCD in the present sample.

Introduction

Obsessive–compulsive disorder (OCD) is a psychiatric illness characterized by recurrent unwanted and intrusive thoughts (obsessions) and repetitive behavior (compulsions) often as an attempt to neutralize anxiety or distress caused by the obsessions (Goodman et al., 1989a, Goodman et al., 1989b). Examples of obsessions are the fear of contamination through contact with other people or objects, the feeling of having done something wrong, such as having forgotten to turn off the stove or lock the door, worrying about one's health, often in conjunction with the fear that something bad might happen. Common compulsions include excessive hand washing many times during the day, repetitive checking (e.g. whether doors are locked or the stove is turned off), prayers for protection, and counting rituals. Although these recurrent thoughts and actions are mostly perceived as absurd, unreasonable and sometimes threatening by the subjects with OCD, it seems almost impossible to control them (Goodman et al., 1989a, Goodman et al., 1989b). Attempts to resist these obsessions and compulsions typically result in anxiety and distress (Hamilton, 1967). OCD symptoms are debilitating because they restrict the normal thought processes and can occupy most of the daily activities with the consequence of social isolation. Although the condition was formerly thought to be rare, epidemiologic studies indicate a 12-month prevalence of about 0.6% for OCD (Crino et al., 2005).

Yet, the pathophysiology of OCD remains controversial. In vivo imaging techniques might be useful in this context (Friedlander and Desrocher, 2006). Positron emission tomography (PET) with the glucose analogue ¹⁸F-fluorodeoxyglucose (FDG), for example, allows assessment of the local metabolic rate of glucose (LMRGlc) throughout the brain which is correlated with brain activity (Kadekaro et al., 1985). Therefore, FDG PET appears an appropriate tool for the investigation of putative functional correlates in the brain to the symptoms in OCD.

A number of FDG PET studies have pointed to frontal cortical regions and the basal ganglia as being relevant for the pathophysiology of OCD. These studies rather consistently have shown (relative) hypermetabolism, i.e. (relatively) increased LMRGlc, in the orbitofrontal cortex (Baxter et al., 1987, Baxter et al., 1988, Nordahl et al., 1989, Swedo et al., 1989, Sawle et al., 1991), the anterior cingulate gyrus (Swedo et al., 1989, Perani et al., 1995), and the head of the caudate nucleus (Baxter et al., 1987, Baxter et al., 1988, Benkelfat et al., 1990). However, (relative) hypometabolism in the prefrontal lateral cortex in OCD subjects has also been reported (Martinot et al., 1990).

Therapies that have proved to be effective in alleviating the symptoms of OCD include pharmacotherapy (PT) with serotonin reuptake inhibitors (SRIs) and cognitive behavioral therapy (CBT) in which the patient undergoes a systematic exposure and response prevention (Fornaro et al., 2009).

Brain FDG PET before and after treatment has been used to assess alterations of LMRGlc induced by these therapy forms in OCD subjects. The most consistently reported findings after successful PT are decrease of (relative) LMRGlc in the orbitofrontal cortex (Benkelfat et al., 1990, Swedo et al., 1992, Saxena et al., 1999, Saxena et al., 2002, Kang et al., 2003), in the anterior frontal gyrus (Swedo et al., 1992, Perani et al., 1995, Saxena et al., 2002), and in the caudate (Benkelfat et al., 1990, Baxter et al., 1992, Saxena et al., 1999, Hansen et al., 2002, Saxena et al., 2002). Successful CBT resulted in a decrease of (relative) LMRGlc in the caudate (Baxter et al., 1992, Schwartz et al., 1996). Thus, OCD therapy is thought to ameliorate OCD symptoms by decreasing functional activity along orbitofrontal–basal ganglia-thalamo-cortical circuits (Baxter et al., 1996). However, Martinot and colleagues did not find a significant effect of successful treatment on absolute whole cortex LMRGlc and relative values in the prefrontal lateral cortex (Martinot et al., 1990). Baxter and colleagues reported a significant increase of the (relative) LMRGlc in the caudate after successful PT (Baxter et al., 1987).

In addition, there is still no general agreement about the role of pre-treatment patterns of glucose metabolism for the prediction of response to therapy in OCD. Lower relative glucose metabolism in the orbitofrontal cortex might be associated with greater improvement in OCD symptoms in case of PT (Swedo et al., 1989, Brody et al., 1998, Saxena et al., 1999), whereas higher relative glucose metabolism in the orbitofrontal cortex might be associated with greater improvement in OCD symptoms in case of CBT (Brody et al., 1998). This difference between PT and CBT appears somewhat unexpected.

The primary aim of the present prospective study was to investigate the effect of both PT and CBT on cerebral glucose metabolism as assessed by FDG PET in an independent sample of adults with OCD. Based on the previous FDG PET findings, the hypothesis put to test was that successful therapy results in a decrease of the relative LMRGlc in the orbitofrontal cortex and in the caudate nucleus. In addition to the region-based analysis of these a priori selected brain regions, exploratory voxel-based Statistical Parametric Mapping (SPM) of the whole brain was performed to test whether there might be other brain regions in which relative LMRGlc is changed by successful OCD treatment.

The secondary aim of the present study was to test whether pre-treatment glucose metabolism in the orbitofrontal cortex or in the caudate might be predictive of response to therapy.