Increased brainstem volume in panic disorder: a voxel-based morphometric study Xenia Protopopescu a,b , Hong Pan a , Oliver Tuescher a , Marylene Cloitre c , Martin Goldstein a , Almut Engelien a , Yihong Yang a , Jack Gorman d , Joseph LeDoux e , Emily Stern a and David Silbersweig a a Functional Neuroimaging Laboratory, Department of Psychiatry, Weill Medical College of Cornell University, New York, New York, USA, b The Rockefeller University Laboratory of Neuroendocrinology, New York, New York, USA, c NYU Child Studies Center, New York University School of Medicine, New York, New York, USA, d Mount Sinai School of Medicine, New York, New York, USA and e Center for Neural Science, New York University, New York, USA Correspondence and requests for reprints to Dr David Silbersweig, Functional Neuroimaging Laboratory, Department of Psychiatry, Box 140, Rm F1302, Weill Medical College of Cornell University, 1300 York Avenue, New York, NY 10021, USA Tel: + 1 212 746 3762; fax: + 1 212 746 5722; e-mail: dasilber@med.cornell.edu Sponsorship: This work was supported by NIMH grant 5-P50MH58911 ‘Center for Neural Systems of Fear and Anxiety,’ the DeWitt Wallace Fund of the New York CommunityTrust, and NIH MSTP grant GM07739 (to X.P.). Received 5 December 2005; revised 10 January 2006; accepted 11 January 2006 Neurocircuitry models of panic disorder have hypothesized that the panic attack itself stems from loci in the brainstem including the ascending reticular system and respiratory and car- diovascular control centers. Voxel-based morphometry with acobian modulation was used to examine gray matter volume changes in 10 panic disorder patients and 23 healthy controls. The panic disorder patients had a relatively increased gray matter volume in the midbrain and rostral pons of the brainstem. Increased ventral hippocampal and decreased regional prefrontal cortex volumes were also noted at a lower signi¢cance threshold. This ¢nding has implications for pathophysiologic models of panic disorder, and provides structural evidence for the role of the brainstem in neurocircuitry models of panic disorder. NeuroReport 17:361^363  c 2006 Lippincott Williams & Wilkins. Keywords: brain structure, brainstem, magnetic resonance imaging, neuroimaging, panic disorder, voxel-based morphometry Introduction Structural magnetic resonance imaging studies in panic disorder have generally been qualitative, or have used manual tracings of regions of interest to perform volumetric analysis. The most consistent structural finding has been decreased temporal lobe volumes [1–3]. One study found panic disorder patients to have smaller amygdalar volumes [4]. A study utilizing voxel-based morphometry (VBM) found that gray matter density of the left parahippocampal gyrus was lower in panic disorder patients [5]. Such findings test for regional differences in concentration of gray matter (per unit volume in native space), rather than gray matter volume [6]. In order to assess regionally specific changes in the amount (volume) of gray matter, volume changes during stereotactic normalization must be incorpor- ated into the analysis by modulating voxel values in the image by the Jacobian determinants derived from the spatial normalization step [7]. Here, we therefore utilized VBM with Jacobian modulation to investigate differences in brain structure between panic disorder patients and controls. Methods Participants consisted of 10 panic disorder patients (mean age¼35.5 years, range¼21–50 years; standard deviation¼9.7; six women, four men) and 23 normal controls (mean age¼28.7 years, range¼22–48 years; standard deviation¼7.5; 11 women, 12 men). Participants gave informed consent before study participation (part of an IRB approved protocol). Participants were right-handed, native English speakers. Panic disorder patients had a primary diagnosis of panic disorder by the Structured Clinical Interview for DSM-IV I (SCID I) [8] criteria. Two participants had agoraphobia. Secondary diagnoses included specific phobia, social pho- bia, generalized anxiety disorder, and major depressive disorder. One participant had past alcohol abuse, and one had past alcohol dependence. Participants (except one) were unmedicated. The disease duration ranged from 1 to 24 years and the severity of the disease varied between 6 and 16 on the total score of the Panic Disorder Severity Scale (PDSS). Image data were acquired with research dedicated GE Signa 3 Tesla magnetic resonance imaging scanners (max gradient strength 40 mT/m, max gradient slew rate 150 T/ m/s) (General Electric Company, Waukesha, Wisconsin, USA). T1-weighted anatomical images were acquired using a spoiled gradient recalled acquisition sequence (TR/ TE¼30/8 ms, flip angle¼451, field of view¼24 cm, slice thickness¼contiguous 1.5 mm, number of averages¼1, BRAIN IMAGING NEUROREPORT 0959-4965  c Lippincott Williams & Wilkins Vol 17 No 4 20 March 2006 361 Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.