Panic attack symptom dimensions and their relationship to illness characteristics in panic disorder Alicia E. Meuret a, * , Kamila S. White b , Thomas Ritz a , Walton T. Roth c , Stefan G. Hofmann d , Timothy A. Brown d a Department of Psychology, Southern Methodist University, 6424 Hilltop Lane, Dallas, TX 75205, USA b Department of Psychology, University of Missouri-Saint Louis, One University Boulevard, St. Louis, MO 63121, USA c Department of Psychiatry and Behavioral Sciences, Stanford School of Medicine Stanford, and the VAPA HCS, 3801 Miranda Avenue, Palo Alto, CA 94304, USA d Center for Anxiety and Related Disorders at Boston University, 648 Beacon Street, Boston, MA 02215, USA Received 3 June 2005; received in revised form 12 September 2005 Abstract Subtyping panic disorder by predominant symptom constellations, such as cognitive or respiratory, has been done for some time, but criteria have varied considerably between studies. We sought to identify statistically symptom dimensions from intensity ratings of 13 DSM-IV panic symptoms in 343 panic patients interviewed with the Anxiety Disorders Interview Schedule for DSM-IV Lifetime Version. We then explored the relation of symptom dimensions to selected illness characteristics. Ratings were submitted to explor- atory maximum likelihood factor analysis with a Promax rotation. A three-factor solution was found to account best for the variance. Symptoms loading highest on the first factor were palpitations, shortness of breath, choking, chest pain, and numbness, which define a cardio-respiratory type (with fear of dying). Symptoms loading highest on the second factor were sweating, trembling, nausea, chills/ hot flashes, and dizziness, which defines a mixed somatic subtype. Symptoms loading highest on the third factor were feeling of unre- ality, fear of going crazy, and fear of losing control, which defines a cognitive subtype. Subscales based on these factors showed mod- erate intercorrelations. In a series of hierarchical multiple regression analyses, the cardio-respiratory subscale was a strong predictor of panic severity, frequency of panic attacks, and agoraphobic avoidance, while the cognitive subscale mostly predicted worry due to panic. In addition, patients with comorbid asthma had higher scores on the cardio-respiratory subscale. We conclude that partly inde- pendent panic symptom dimensions can be identified that have different implications for severity and control of panic disorder. Ó 2005 Elsevier Ltd. All rights reserved. Keywords: Panic disorder; Panic attack; Symptom dimensions; Respiratory subtype; Agoraphobia 1. Introduction Considerable heterogeneity exists in the clinical pre- sentation of panic disorder. DSM-IV criteria require intense distress or fear and 4 out of 13 symptoms to be present for a diagnosis of panic. However, these symptoms seem to have several sources: the autonomic nervous system (e.g., pounding heart, sweating), the respiratory system (shortness of breath, chest tightness), and the central nervous system/cognitive processing (e.g., depersonalization, fear of losing control, fear of dying). A substantial number of symptoms are sugges- tive of hypocapnia, such as lightheadedness or dizziness, paresthesias, numbness, and tingling sensations. This has fueled discussions on the role of hyperventilation 0022-3956/$ - see front matter Ó 2005 Elsevier Ltd. All rights reserved. doi:10.1016/j.jpsychires.2005.09.006 * Corresponding author. Tel.: +1 214 768 3422; fax: +1 214 768 3910. E-mail address: ameuret@smu.edu (A.E. Meuret). www.elsevier.com/locate/jpsychires Journal of Psychiatric Research 40 (2006) 520–527 J OURNAL OF P SYCHIATRIC RESEARCH