Amygdala Volume Reductions in Pediatric Patients with Obsessive–Compulsive Disorder Treated with Paroxetine: Preliminary Findings

Abstract

The amygdala is believed to be highly relevant to the pathophysiology of obsessive-compulsive disorder (OCD) given its prominent role in fear conditioning and because it is an important target of the serotonin reuptake inhibitors (SRIs), the pharmacotherapy of choice for OCD. In the present study, we measured in vivo volumetric changes in the amygdala in pediatric patients with OCD following 16 weeks of monotherapy with the selective SRI, paroxetine hydrochloride. Amygdala volumes were computed from contiguous 1.5 mm magnetic resonance (MR) images in 11 psychotropic drug-naive patients with OCD prior to and then following treatment. Eleven healthy pediatric comparison subjects also had baseline and follow-up scans, but none of these subjects received medication. Patients demonstrated significant asymmetry of the amygdala (L4R) prior to pharmacologic intervention in contrast to healthy comparison subjects who showed no asymmetry at the time of their baseline scan. Mixed model analyses using age and total brain volume as time varying covariates indicated that left amygdala volume decreased significantly in patients following treatment. The reduction in left amygdala volume in patients correlated significantly with higher paroxetine dosage at the time of the follow-up scan and total cumulative paroxetine exposure between the scans. No significant changes in either right or left amygdala volume were evident among healthy comparison subjects from the baseline to the follow-up scan. These preliminary findings suggest that abnormal asymmetry of the amygdala may play a role in the pathogenesis of OCD and that paroxetine treatment may be associated with a reduction in amygdala volume.

Structural magnetic ressonance imaging in anxiety disorders: an update of research findings

“…In a study on 23 pediatric patients with a diagnosis of OCD but with no previous treatment, a smaller volume of the globus pallidus and a greater volume of the anterior cingulate cortex were detected compared to 26 healthy controls using semi-automated, ROI-based methods. 56 A greater volume of the pituitary gland was also detected using the manual ROI method in untreated pediatric patients with OCD, being more prominent among boys. 57 A recent study using the ROIbased approach on a sample of never-medicated adults detected a reduced volume of the right and left orbitofrontal cortex and a greater volume of the thalamus, also bilaterally.…”

“…59 Serotoninergic transmission in this brain area has been associated with the modulation of fear and of conditioned anxiety, factors that seem to play an important role in OCD. 59 Other limbic structures have been evaluated in the various morphometric MRI studies of OCD, having been detected inconsistencies in volumetric changes, such as an increase 56 and a reduction of the cingulate cortex, 54,60 an increase of the temporolimbic cortex and of the insula, 54 and a reduction of the insulo-opercular region. 53 Reduction of the white and gray matter in the right parietal cortex was also found, 54,60 an area not previously investigated in structural studies by ROI, but which had already shown lower activity in studies of functional neuroimaging.…”

Review of structural neuroimaging in patients with refractory obsessivecompulsive disorder

“…Abnormalities in the regions of hippocampus and amygdala have been emphasized in studies involving positron emission tomography (PET) or functional magnetic resonance imaging (fMRI), and researchers have commented that these regions may play an important role in the pathophysiology of OCD, with neglected discussion [27,28] . Moreover, in OCD, there is a loss of normal hemispheric asymmetry of the hippocampus-amygdala complex [16] and differences in amygdala volumes [29] .…”

Development and sexual dimorphism of the pituitary gland

“…On the University of Pittsburgh MR scanner a three dimensional SPGR acquisition in the steady state pulse sequence was utilized to obtain 1.5 mm coronal images (TE = 20 msec, TR = 40 msec, acquisition matrix = 256 x 192, FOV = 20 cm, flip angle = 10°). Image acquisition has been described in detail in our previous reports (Gilbert et al, 2000, MacMaster and Kusumakar, 2004a, MacMaster and Kusumakar, 2004b, MacMaster et al, 2006, Nolan et al, 2002, Rosenberg et al, 1997, Szeszko et al, 2004a, Szeszko et al, 2004b. All MRI scans were reviewed to rule out any clinically significant abnormalities.…”