Health Articles

 

0803cns_suppl_rep

August 2003
Volume 8 - Number 8 - Suppl 1
The International Journal of Neuropsychiatric Medicine ACADEMIC SUPPLEMENT REPRINT
Recommendations for the Long-Term
Treatment of Anxiety Disorders
Long-Term Treatment of Obsessive-Compulsive Disorder in Adults
J.H. Greist, B. Bandelow, E. Hollander, D. Marazziti, S.A. Montgomery, D.J. Nutt, A. Okasha, R.P. Swinson, and J. Zohar Long-Term Treatment of Panic Disorder
M.H. Pollack, C. Allgulander, B. Bandelow, G. Cassano, J.H. Greist, E. Hollander, D.J. Nutt, A. Okasha, and R.P. Swinson Long-Term Treatment of Posttraumatic Stress Disorder
D.J. Stein, B. Bandelow, E. Hollander, D.J. Nutt, A. Okasha, M. Pollack, R.P. Swinson, and J. Zohar Long-Term Treatment of Social Phobia
M. van Ameringen, C. Allgulander, B. Bandelow, J.H. Greist, E. Hollander, S.A. Montgomery, D.J. Nutt, A. Okasha, M.H. Pollack, D.J. Stein, and R.P. Swinson Long-Term Treatment of Generalized Anxiety Disorder
C. Allgulander, B. Bandelow, E. Hollander, S.A. Montgomery, D.J. Nutt, A. Okasha, M.H. Pollack, D.J. Stein, and R.P. Swinson Index Medicus/MEDLINE
citation: CNS Spectr
This reprint has been adapted from an academic supplement based on information presented at the World Council of Anxiety Meeting, held September 11, 2000, in Pisa, Italy. Faculty Affiliations and Disclosure Dr. Allgulander is senior lecturer and associate professor of psychiatry at the Karolinska Institutet and head of the Neurotec Department, Division of Psychiatry, at Huddinge University Hospital in Stockholm, Sweden. He is a member of the World Council of Anxiety and a principal investigator for phase II and phase III studies for GlaxoSmithKline, Lundbeck, Pfizer, and Wyeth.
Dr. Bandelow is professor in the Department of Psychiatry and Psychotherapy at the University of Goettingen in Germany. He receives grant/research support from GlaxoSmithKline; is on the scientific advisory board of AstraZeneca, Bayer AG, Bristol-Myers Squibb, Fujisawa, Eli Lilly, GlaxoSmithKline, Lundbeck, Parke-Davis, Pfizer, Sanofi-Synthélabo, and Wyeth; and is on the Speaker's Bureau for Aventis, AstraZeneca, Bayer AG, Boehringer-Ingelheim, Bristol-Myers-Squibb, Cilag, Eli Lilly, GlaxoSmithKline, Janssen, Lundbeck, Meiji-Seiko, Novartis, Organon, Parke-Davis, Pfizer, Pharmacia, Roche, Sanofi-Synthélabo, Solvay, and Wyeth.
Dr. Cassano is professor of psychiatry in the Department of Psychiatry, Neurobiology, Pharmacology, and Biotechnology in the Section of Psychiatry at the University of Pisa in Italy. He reports no financial, academic, or other support for this work.
Dr. Greist is clinical professor of psychiatry at the University of Wisconsin, distinguished senior scientist at the Madison Institute of Medicine, and CEO at Healthcare Technology Systems, Inc., all are in Madison, Wisconsin. He has received grants/research support from Abbott, Forest, GlaxoSmithKline, Janssen, Eli Lilly, Organon, Pfizer, Sanofi-Synthelabo, Solvay, and Wyeth; is a consultant for Bristol-Myers Squibb, GlaxoSmithKline, Eli Lilly, Organon, Pfizer, Sanofi-Synthelabo, Solvay, and Wyeth; and is on the Speaker's Bureau for Bristol-Myers Squibb, Forest, GlaxoSmithKline, Hoffman-La Roche, Eli Lilly, Novartis, Organon, Pfizer, Solvay, and Wyeth. Dr. Hollander is professor of psychiatry and director of the Seaver Center and New York Autism Center of Excellence, the Clinical Psychopharmacology Unit, Compulsive, and the Impulsive and Anxiety Disorders Program, all at Mount Sinai School of Medicine in New York City.
He has received research grants from Abbott, Eli Lilly, Forest, Pfizer, Solvay, and Wyeth; is a consultant for Abbott, Bristol-Myers Squibb, Solvay, and Wyeth; and is a speaker for Abbott, Forest, GlaxoSmithKline, and Wyeth.
Dr. Marazziti is professor of psychiatry at the Department of Psychiatry, Neurobiology, Pharmacology, and Biotechnology in the Section of Psychiatry University of Pisa in Italy. She reports no financial, academic, or other support of this work. Dr. Montgomery is emeritus professor at Imperial College School of Medicine in London, England. He has been consulted by Abbott, Almirall,AstraZeneca, GlaxoSmithKline, Johnson & Johnson, Lilly, Lundbeck, Merck, Merz, Novartis, Organon, Orion, Pfizer, Pierre Fabre, Sanofi, Servier,Solvay, Vela, and Wyeth.
Dr. Nutt is professor of psychopharmacology in the Psychopharmacology Unit at the University of Bristol School of Medical Sciences in the United Kingdom. He is a consultant for Esteve, GlaxoSmithKline, Janssen, Lundbeck, MSD, Servier, Solvay, and Wyeth, and has received honoraria from Astrazeneca, Organon, Pierre Fabre, and Wyeth.
Dr. Okasha is director of the World Health Organization Collaborating Center for Training and Research at the Institute of Psychiatry, Ain Shams University in Cairo, Egypt, and president of the World Psychiatric Association. He reports no financial, academic, or other support for this work.
Dr. Pollack is director of the Center for Anxiety and Traumatic Stress-Related Disorders at Massachusetts General Hospital and associate professor of psy- chiatry at Harvard University School of Medicine, both in Boston. He has received grant/research support from Forest, GlaxoSmithKline, Janssen, Eli Lilly, Pfizer, UCB Pharma, and Wyeth, and is on the Scientific Advisory Boards of Cephalon, GlaxoSmithKline, Janssen, Novartis, Pfizer, UCB Pharma, and Wyeth. Dr. Stein is director of the Medical Research Council Research Unit on Anxiety Disorders at the University of Stellenbosch, Cape Town, in Tygerberg, South Africa, and associate professor at the University of Florida in Gainesville. He has received research grants and/or consultancy honoraria from Astrazeneca, Eli Lilly, GlaxoSmithKline, Lundbeck, Orion, Pfizer, Pharmacia, Roche, Solvay, and Wyeth. The MRC Unit on Anxiety Disorders is supported by each of the pharmaceutical companies with an interest in psychiatry in South Africa.
Dr. Swinson is professor and chair of Psychiatry and Behavioral Neurosciences and Morgan Firestone Chair of Psychiatry at McMaster University, and psychiatrist in chief at St Joseph's HealthCare, in Hamilton, Ontario. He has received honoraria or research grants from Ciba-Geigy, GlaxoSmithKline, Lundbeck, Pfizer, Upjohn, and Wyeth.
Dr. van Ameringen is co-director of the Anxiety Disorders Clinic and associate professor in the Department of Psychiatry and Behavioral Neurosciences at McMaster University in Hamilton, Ontario. He is a consultant for Cephalon, GlaxoSmithKline, Janssen, Novartis, Pfizer, Solvay, and Wyeth; has received research grants from AstraZeneca, Bristol-Myers Squibb, Eli Lilly, GlaxoSmithKline, Janssen, Lundbeck, Pfizer, and Wyeth; and is on the Speaker's Bureau for GlaxoSmithKline and Pfizer.
Dr. Zohar is professor of psychiatry at Chaim Sheba Medical Center in the Division of Psychiatry in Tel-Hashomer, Israel. He reports no financial, academic, or other support for this work. This reprint has been adapted from an academic supplement supported via an unrestricted educational grant from Pfizer Inc. Sponsorship of this review does not imply the sponsor's agreement with the views expressed herein.
Although every effort has been made to ensure that drug doses and other information are presented accurately in this publication, the ultimate respon- sibility rests with the prescribing physician. Neither the publisher, the sponsor, nor the authors can be held responsible for errors or for any conse- quences arising from the use of information contained herein. Readers are strongly urged to consult any relevant primary literature. No claims or endorsements are made for any drug or compound currently under clinical investigation.
In an effort to allow for the widest distribution of these guidelines, the authors have modified the originally printed material to more closely conform to the limitations of product labeling. For many of the drugs discussed herein, initiation at lower doses may increase tolerability and efficacy.
Copyright 2003, MBL Communications, Inc.
333 Hudson St., 7th Floor, New York, NY 10013.
Printed in the USA. All rights reserved, including the right of reproduction, in whole or in part, in any form.
Sertraline is not indicated for the treatment of acute or long-term generalizedanxiety disorder.
Sertraline is indicated for the acute and long-term treatment of depression,social anxiety disorder, panic disorder, and PTSD. Sertraline is also indicated for the treatment of PMDD. Sertraline is not indicated for the treatment of acute or long-term generalizedanxiety disorder.
Sertraline is indicated for the acute and long-term treatment of depression,social anxiety disorder, panic disorder, and PTSD. Sertraline is also indicated for the treatment of PMDD. August 2003
Volume 8 - Number 8 - Suppl 1
The International Journal of Neuropsychiatric Medicine Jerome Engel, Jr., MD, PhD
Yves Lecrubier, MD
Jack M. Gorman, MD
University of California, Los Angeles Hôspital de la Salpêtrière Mount Sinai School of Medicine Mark S. George, MD
Herbert Y. Meltzer, MD
Medical University of South Carolina Vanderbilt University Medical Center ASSOCIATE AND FOUNDING EDITOR
Eric Hollander, MD
Richard B. Lipton, MD
Stuart A. Montgomery, MD
Mount Sinai School of Medicine Albert Einstein College of Medicine St. Mary's Hospital Medical School London, United Kingdom C. Warren Olanow, MD, FRCPC
Charles B. Nemeroff, MD, PhD
Mount Sinai School of Medicine Emory University School of Medicine Joseph Zohar, MD
Chaim Sheba Medical Center Steven George Pavlakis, MD
Humberto Nicolini, MD, PhD
Tel-Hashomer, Israel Maimonides Medical Center National Mexican Institute of Psychiatry Mexico City, Mexico ASSOCIATE INTERNATIONAL EDITORS
Stephen D. Silberstein, MD, FACP
Katharine Phillips, MD
Thomas Jefferson University Brown Medical School Donatella Marazziti, MD
University of Pisa Michael Trimble, MD, FRCP, FRPsych
Harold A. Pincus, MD
National Hospital for Neurology Western Psychiatric Institute & Clinic RAND-University of Pittsburgh Health London, United Kingdom Institute, Pittsburgh, PA Dan J. Stein, MD, PhD
Scott L. Rauch, MD
University of Stellenbosch Margaret Altemus, MD
Massachusetts General Hospital Tygerberg, South Africa Cornell University Medical College Alan F. Schatzberg, MD
Dennis S. Charney, MD
Stanford University School of Medicine Shigeto Yamawaki, MD, PhD
National Institute of Mental Health Hiroshima University School Stephen M. Stahl, MD, PhD
Dwight L. Evans, MD
University of California, San Diego University of Pennsylvania La Jolla, California Norman Sussman, MD, DFAPA
Siegfried Kasper, MD
New York University Medical School BOARD OF ADVISORS
University of Vienna Karen Dineen Wagner, MD, PhD
Mitchell F. Brin, MD
Martin B. Keller, MD
The University of Texas Medical Branch University of California, Irvine Brown Medical School Herman G.M. Westenberg, MD
Jeffrey L. Cummings, MD
Lorrin M. Koran, MD
University Hospital Utrecht University of California, Los Angeles Stanford University School of Medicine Utrecht, The Netherlands Stuart C. Yudofsky, MD
Baylor College of Medicine MBL COMMUNICATIONS Corporate Staff
CEO & PUBLISHER
ASSISTANT ACQUISITIONS EDITOR
CME PROJECTS DIRECTOR
Darren L. Brodeur
Eileen S. Seligman
Chris Van Denburg
Shelley Wong
Lila Moses
Anthony J. Korsak
Adam Bolt
DEPUTY SENIOR EDITOR
José R. Ralat
John Spano
Lawrence Ross, Esq.
NATIONAL ACCOUNT MANAGER
Bressler, Amery, and Ross
Kelly J. Staley
CNS Spectrums' editorial mission is to address relevant neuropsychiatric topics, including the prevalence of comorbid diseases among patients, and original research and reports that emphasize the profound diagnostic and physiologic connections made within the neurologic and psychiatric fields. The journal's goal is to serve as a resource to psychiatrists and neurologists seeking to understand and treat disturbances of cognition, emotion, and behavior as a direct consequence of central nervous system disease, illness, or trauma. Volume 8 – Number 8 (Suppl 1) CNS Spectrums - August 2003 The International Journal of Neuropsychiatric Medicine Table of Contents
CNS Spectrums is an
August 2003
Index Medicus journal Volume 8 - Number 8 - Academic Supplement 1
and is available on MEDLINE Based on Proceedings of the World Council of Anxiety under the citation CNS Spectr.
Meeting Held September 2000, in Pisa, Italy It is also indexed by DIALOG, EMBASE/Excerpta Medica, OVID, Lexis-Nexis, and SilverPlatter. Introduction: WCA Recommendations for the Long-Term
Treatment of Anxiety Disorders
CNS Spectrums is the official journal of the International By Joseph Zohar, MD Neuropsychiatric Association with members in 30 countries.
Long-Term Treatment of Obsessive-Compulsive Disorder
By John H. Greist, MD, Borwin Bandelow, MD, Eric Hollander, MD, (ISSN 1092-8529)is published monthly by Donatella Marazziti, MD, Stuart A. Montgomery, MD, MBL Communications, Inc.
David J. Nutt, MD, Ahmed Okasha, MD, PhD, 333 Hudson Street, 7th Floor Richard P. Swinson, MD, FRCPC, FRCPsych, DPM, New York, NY 10013 and Joseph Zohar, MD One year subscription rates: Long-Term Treatment of Panic Disorder
in-training $75. By Mark H. Pollack, MD, Christer Allgulander, MD, For subscriptions: Borwin Bandelow, MD, Giovanni B. Cassano, MD, or visit our Web site: John H. Greist, MD, Eric Hollander, MD, and David J. Nutt, MD Long-Term Treatment of Posttraumatic Stress Disorder
Send address changes to By Dan J. Stein, MD, PhD, Borwin Bandelow, MD, c/o PPS Medical Marketing Group 264 Passaic Avenue Eric Hollander, MD, David J. Nutt, MD, Ahmed Okasha, MD, PhD, Fairfield, NJ 07004-2595 Mark H. Pollack, MD, Richard P. Swinson, MD, FRCPC, FRCPsych, DPM, Opinions and views expressed by authors and Joseph Zohar, MD are their own and do not necessarily reflect the views of the publisher, MBL Long-Term Treatment of Social Phobia
Communications, Inc., or the editorial By Michael van Ameringen, MD, FRCPC, Christer Allgulander, MD,Borwin Bandelow, MD, John H. Greist, MD, Eric Hollander, MD, CNS Spectrums® is a registered
Stuart A. Montgomery, MD, David J. Nutt, MD, trademark of CNS Spectrums, LLC,
New York, NY.
Ahmed Okasha, MD, PhD, Mark H. Pollack, MD, Permission to reproduce articles in whole or Dan J. Stein, MD, PhD, and Richard P. Swinson, MD part must be obtained in writing from the Long-Term Treatment of Generalized Anxiety Disorder
By Christer Allgulander, MD, Borwin Bandelow, MD,
Eric Hollander, MD, Stuart A. Montgomery, MD,
David J. Nutt, MD, Ahmed Okasha, MD, PhD,
Mark H. Pollack, MD, Dan J. Stein, MD, PhD, and
Copyright 2003 by MBL Richard P. Swinson, MD Communications, Inc. All rights reserved. Printed in the United States.
Volume 8 – Number 8 (Suppl 1) CNS Spectrums - August 2003 WCA Recommendations for the
Long-Term Treatment of Anxiety Disorders
By Joseph Zohar, MD Anxiety disorders represent a spectrum of conditions that by searching MEDLINE as of October 15, 2001. In develop- collectively encompasses more than half of all psychiatric ing these recommendations for the long-term treatment of disorders. In addition to being highly prevalent, anxiety dis- anxiety disorders, the WCA relied on methodologically orders are usually chronic conditions that can substantially sound studies, ie, studies that were at least 24 weeks in dura- impair a patient's quality of life. Affected individuals fre- tion and included a double-blind, randomized, placebo-con- quently present with atypical/subthreshold symptoms and trolled period. No open-label, naturalistic studies or case often exhibit comorbidity, including comorbid psychiatric/ studies were considered for inclusion. medical disorders, thus hindering correct diagnosis. As the In addition to the published literature, studies described recognition and diagnosis of anxiety disorders has improved, in congress abstracts from the following meetings held in the rate of medical office visits in which a diagnosis of anxi- 1999, 2000, and 2001 were included: ety is documented increased 1.4-fold in the United States •American Psychiatric Association Annual Meeting, between 1990 and 1997.1 This higher rate of diagnosis not May 15–20, 1999, Washington, DC only reflects considerable personal suffering, but imposes a •11th World Congress of Psychiatry, August 6–11, 1999, substantial economic burden on society. The annual cost of anxiety disorders in the US was estimated at more than $42 billion in 1990.2 TABLE. LONG-TERM RANDOMIZED CONTROLLED
The last decade has brought about effective new treat- TREATMENT TRIALS OF PHARMACOTHERAPY
ments, both in pharmacotherapy and psychotherapy.
FOR ANXIETY DISORDERS
However, the difficulty of obtaining effective psychotherapy is a major limitation in its use. The majority of clinical trials of pharmacologic agents have focused on the acute treat- ment phase, and therefore less is known about the long-term management of anxiety disorders. (Please see the accompa- nying Table for a listing of drug-specific long-term pharma- cotherapy treatment data). Questions remain regarding the length of time and dose of medication required during the maintenance phase.
Rasmussen et al9 1997 To help close this information gap, the World Council of Anxiety (WCA) has prepared a series of articles based on their meeting held September 11, 2000 in Pisa, Italy. The articles provide a comprehensive, state-of-the-art review of the data on the effectiveness of long-term treatment for anx- Michelson et al14 1999 iety disorders. As few guidelines exist that offer recommen- dations for long-term treatment, this information may assist Lecrubier et al16 the practicing clinician in choosing the most appropriate long-term treatment options for each individual patient. Londborg et al18 2001 The five papers included in this supplement discuss the following anxiety disorders: obsessive-compulsive disorder, panic disorder, posttraumatic stress disorder, social phobia (social anxiety disorder), and generalized anxiety disorder.
Each article covers the following areas: clinical presenta- tion; underlying biological substrates; clinical course, chronicity, and associated disability; diagnosis, including dif- ferential diagnosis; comorbidity and its complexity; acute to long-term treatment options, including pharmacotherapy, Allgulander et al26 2001 psychotherapy, and combination treatment; and at-a-glance Gelenberg et al27 2000 tables of available treatment guidelines and trials.
In these five articles the authors review the available Zohar J. CNS Spectrums. Vol 8, No 8 (suppl 1). 2003.
published literature on anxiety disorders, which was gathered Volume 8 – Number 8 (Suppl 1) CNS Spectrums - August 2003 •12th Congress of the European College of 12. Lepola UM, Wade AG, Leinonen EV, et al. A controlled, prospective, 1-year trial of citalopram in the treatment of panic disorder. Neuropsychopharmacology, September 21–25, 1999, London, United Kingdom 13. Fahy TJ, O'Rourke D, Brophy J, et al. The Galway Study of Panic •American Psychiatric Association Annual Meeting, Disorder. I: Clomipramine and lofepramine in DSM III-R panic disor- May 13–18, 2000, Chicago, Illinois der: a placebo controlled trial. J Affect Disord. 1992;25:63-75.
•22nd Collegium Internationale Neuropsychopharma- 14. Michelson D, Pollack M, Lydiard RB, et al. Continuing treatment of cologicum Congress, July 9–13, 2000, Brussels, Belgium panic disorder after acute response: randomised, placebo-controlled trial •13th Congress of the European College of with fluoxetine. The Fluoxetine Panic Disorder Study Group. Br J Neuropsychopharmacology, September 9–13, 2000, 15. Lydiard RB, Steiner M, Burnham D, et al. Efficacy studies of paroxetine in panic disorder. Psychopharmacol Bull. 1998;34:175-182.
•American Psychiatric Association Annual Meeting, 16. Lecrubier Y, Judge R. Long-term evaluation of paroxetine, May 5–10, 2001, New Orleans, Louisiana clomipramine and placebo in panic disorder. Collaborative Paroxetine •World Congress of Biological Psychiatry Biennial Panic Study Investigators. Acta Psychiatr Scand. 1997;95:153-160.
Meeting, July 1–6, Berlin, Germany 17. Rapaport MH, Wolkow R, Rubin A, Hackett E, Pollack M, Ota KY.
We hope that this supplement contributes to an increased Sertraline treatment in panic disorder: results of a long-term study. Acta awareness and understanding of anxiety disorders.
Psychiatr Scand. 2001;104:289-298.
18. Londborg PD, Hegel MT, Goldstein S, et al. Sertraline treatment of Specifically, by helping clinicians and patients recognize the posttraumatic stress disorder: results of 24 weeks of open-label continua- importance of extended treatment, we hope to improve the tion treatment. J Clin Psychiatry. 2001;62:325-331. long-term management of anxiety disorders. CNS
19. Davidson JR, Rothbaum BO, van der Kolk BA, et al. Multicenter, dou- ble-blind comparison of sertraline and placebo in the treatment of post- traumatic stress disorder. Arch Gen Psychiatry. 2001;58:485-492. 1. Skaer TL, Robison LM, Sclar DA, et al. Anxiety disorders in the USA, 20. Connor KM, Davidson JR, Potts NL, et al. Discontinuation of clon- 1990 to 1997. Trends in complaint, diagnosis, use of pharmacotherapy azepam in the treatment of social phobia. J Clin Psychopharmacol. and diagnosis of comorbid depression. Clin Drug Invest. 2000;20:237-244.
2. Greenberg PE, Sisitsky T, Kessler RC, et al. The economic burden of 21. Hair T, Castrogiovanni P, Domenech J, et al. Short-term efficacy of anxiety disorders in the 1990s. J Clin Psychiatry. 1999;60:427-435.
paroxetine in social anxiety disorder is maintained after long-term treat- 3. Katz RJ, DeVeaugh-Geiss J, Landau P. Clomipramine in obsessive- ment. Int J Neuropsychopharmacol. 2000;3:S227.
compulsive disorder. Biol Psychiatry. 1990;28:401-414.
22. Kumar R, Pitts C, Carpenter D. Response to paroxetine is maintained 4. Tollefson GD, Birkett M, Koran L, et al. Continuation treatment of during continued treatment in patients with social anxiety disorder. Eur OCD: double-blind and open-label experience with fluoxetine. J Clin 23. Walker JR, Van Ameringen MA, Swinson R, et al. Prevention of relapse 5. Romano S, Goodman W, Tamura R, et al. Long-term treatment of in generalized social phobia: results of a 24-week study in responders to 20 obsessive-compulsive disorder after an acute response: a comparison of weeks of sertraline treatment. J Clin Psychopharmacol. 2000;20:636-644.
fluoxetine versus placebo. J Clin Psychopharmacol. 2001;21:46-52.
24. Blomhoff S, Haug TT, Hellstrom K, et al. Randomised control general 6. Cottraux J, Mollard E, Bouvard M, et al. A controlled study of fluvox- practice trial of sertraline, exposure and combined treatment in gener- amine and exposure in obsessive-compulsive disorder. Int Clin alised social phobia. Br J Psychiatry. 2001;179:23-30.
25. Rickels K, Schweizer E, Csanalosi I, et al. Long-term treatment of anxi- 7. Mallya GK, White K, Waternaux C, et al. Short- and long-term treat- ety and risk of withdrawal. Prospective comparison of clorazepate and ment of obsessive-compulsive disorder with Fluvoxamine. Ann Clin buspirone. Arch Gen Psychiatry. 1988;45:444-450.
26. Allgulander C, Hackett D, Salinas E. Venlafaxine extended release 8. Greist JH, Jefferson JW, Kobak KA, et al. A 1 year double-blind place- (ER) in the treatment of generalised anxiety disorder: twenty-four-week bo-controlled fixed dose study of sertraline in the treatment of obses- placebo-controlled dose-ranging study. Br J Psychiatry. 2001;179:15-22. sive-compulsive disorder. Int Clin Psychopharmacol. 1995;10:57-65.
27. Gelenberg AJ, Lydiard RB, Rudolph RL, et al. Efficacy of venlafaxine 9. Rasmussen S, Hackett E, DuBoff E, et al. A 2-year study of sertraline in extended-release capsules in nondepressed outpatients with generalized the treatment of obsessive-compulsive disorder. Int Clin Psychopharmacol. anxiety disorder: A 6-month randomized controlled trial. JAMA. 10. Koran LM, Robinson DG, Hackett E, Rubin A, Wolkow RM. Efficacy of 28. Stocchi F, Nordera G, Jokinen R, Lepola U. Efficacy and tolerability of sertraline in long-term OCD treatment: preliminary results of a multi- paroxetine for the long-term treatment of generalised anxiety disorder center study. Abstract presented at: Annual Meeting of the American (GAD). Abstract presented at: Annual Meeting of the American Psychiatric Association; Washington, DC; May 15-20, 1999. Psychiatric Association; May 5-10, 2001; New Orleans, La.
11. Bergeron R, Hadrava V, Ravindran AV. Sertraline and fluoxetine treat- ment of OCD: a six-month, double-blind, parallel study. Poster present- ed at: 13th Annual Congress of the European College of Neuropsychopharmacology; September 9-13, 2000; Munich, Germany.
Volume 8 – Number 8 (Suppl 1) CNS Spectrums - August 2003 WCA Recommendations for the
Long-Term Treatment of Obsessive-
Compulsive Disorder in Adults
By John H. Greist, MD, Borwin Bandelow, MD, Eric Hollander, MD, Donatella Marazziti, MD, Stuart A. Montgomery, MD, David J. Nutt, MD, FRCP, FRCPsych, FMedSci, Ahmed Okasha, MD, PhD, FRCP, FRCPsych, FACP, Richard P. Swinson, MD, FRCPC, FRCPsych, DPM, and Joseph Zohar, MD paroxetine, sertraline, or citalopram) in preference to •Both psychotherapy and pharmacotherapy are clomipramine, due to the latter's less favorable adverse-event pro- recommended for the treatment of obsessive-compul- file. Further, pharmacotherapy for a minimum of 1–2 years is rec- sive disorder (OCD).
ommended before very gradual withdrawal may be considered. •Selective serotonin reuptake inhibitors (SSRIs) are CNS Spectr 2003;8(suppl 1):7-16 considered first line in the pharmacotherapy of OCD.
•Acute and long-term efficacy has been shown with clomipramine, fluvoxamine, fluoxetine, and sertraline.
Obsessive-compulsive disorder (OCD) is a chronic, debil- •Guidelines recommend maintenance pharmacological itating disorder that causes substantial distress and has a neg- therapy for a minimum of 1–2 years.
ative impact on an individual's relationships and ability to •Despite the availability of SSRIs, many patients remain work. Early accounts of OCD date from the 15th century, treatment-resistant or refractory.
when symptoms resembling those of OCD were reported in a treatise on witchcraft.1 William Shakespeare wrote of Lady Macbeth's compulsive washing, "It is an accustomed action What are the latest psychotherapeutic and pharmacotherapeu- with her, to seem this washing her hands. I have known her tic treatment recommendations for obsessive-compulsive disorder continue with this for a quarter of an hour."2 (OCD)? OCD is a relatively common disorder with a lifetime prevalence of 2% in the general population. It often has an early onset, usually in childhood or adolescence, and frequently Epidemiologic studies have shown that the lifetime preva- becomes chronic and disabling if left untreated. High associated lence of OCD in the general population is between 1% and healthcare utilization and costs, and reduced productivity resulting 3%,3,4 making OCD the sixth most common psychiatric dis- in loss of earning, pose a huge economic burden to OCD patients order in the United States.5 The gender ratio has been found and their families, employers, and society. OCD is characterized to be approximately equal,4,6 although among childhood by the presence of obsessions and compulsions that are time-con- OCD cases, males predominate.6 In this regard, OCD is dif- suming, cause marked distress, or significantly interfere with a ferent from other anxiety disorders, which are more preva- person's functioning. Most patients with OCD experience symp- lent in females than in males. More than 80% of individuals toms throughout their lives and benefit from long-term treatment. date the onset of symptoms before 18 years of age.7 Both psychotherapy and pharmacotherapy are recommended, either alone or in combination, for the treatment of OCD. Cognitive-behavioral therapy is the psychotherapy of choice. OCD is characterized by a range of obsessions and com- Pharmacologic treatment options include the tricyclic antidepres- pulsions that are remarkably heterogeneous but stereotypic sant clomipramine and the selective serotonin reuptake inhibitors across the affected population.6 Obsessions are recurrent or (SSRIs) citalopram, fluoxetine, fluvoxamine, paroxetine, and persistent thoughts, images, or impulses that the patient con- sertraline. These have all shown benefit in acute treatment trials; siders inappropriate and, therefore, struggles to ignore or sup- clomipramine, fluvoxamine, fluoxetine, and sertraline have also press. In an effort to relieve the anxiety caused by the demonstrated benefit in long-term treatment trials (at least presence and intensity of these obsessions, the person feels 24 weeks), and clomipramine, sertraline, and fluvoxamine have compelled to perform specific repetitive behaviors or mental United States Food and Drug Administration approvals for use in acts called rituals.8 The most common obsessions include children and adolescents. Available treatment guidelines recom- contamination; pathological doubt; aggressive, sexual, mend first-line use of an SSRI (ie, fluoxetine, fluvoxamine, or somatic repetitive intrusive thoughts; and the need for These recommendations are based on proceedings from the World Council of Anxiety meeting held September 11, 2000, in Pisa, Italy, and on guidelines and articles published in the medical literature through October 15, 2001.
Please direct all correspondence to: John H. Greist, MD, Healthcare Technology Systems, Inc., 7617 Mineral Point Rd, Suite 300, Madison, WI 53717; Tel: 608-827-2450; Fax: 608-827-2444; E-mail: [email protected].
Volume 8 – Number 8 (Suppl 1) CNS Spectrums - August 2003 Greist JH, Bandelow B, Hollander E, et al
symmetry and precision. The most common compulsions If untreated, the natural course of OCD is usually chronic evoke rituals of cleaning, checking, and counting.6 and often severe enough to have a significant impact on the Most patients will have several obsessions and corresponding lives of patients and their families, friends, employers, and compulsions, although one particular pattern (symptom type) society.20 SSRIs and clomipramine are the mainstay of phar- may predominate.9 In addition, at different points in the course macologic treatment of OCD and have been shown to sig- of their illness, patients report different problems as most promi- nificantly improve symptom severity, quality of life (QOL), nent.9 The patient usually recognizes that the obsessions and and functioning. However, as many as 40% to 60% of compulsions are excessive or unreasonable, which results in sub- patients with OCD may not achieve an adequate response stantial distress. The compulsions of OCD, as opposed to those with SSRI or clomipramine monotherapy, and may require of addiction, are not considered pleasurable. They merely additional/alternative treatment approaches.21 relieve anxiety associated with the obsession, albeit temporarily.
Quality of Life
A survey of 701 OCD patients found that OCD had a sig- nificant impact on QOL measures: 58% reported lower acade- A wealth of information from brain-imaging studies has mic achievement; 64% reported lower career aspirations; and become available that complement research in other areas 40% reported inability to work, with an average loss of 2 years and suggest the existence of an abnormal functioning neural of wages.20,22 Further, 62% of subjects reported having fewer circuitry specific to OCD. Overall, brain-imaging research friends or difficulty maintaining relationships following onset provides evidence that the underlying dysfunction is likely to of OCD, and the majority (92%) experienced lowered self- involve the prefrontal cortex-basal ganglia thalamic circuitry esteem.20 Of note, 13% of respondents attempted suicide sec- rather than any single region of the brain.10 ondary to OCD.20 With treatment, 62% had improved overall QOL and 43% showed improved ability to study or work.22 It appears that serotonin dysfunction is involved in OCD.
Evidence from early neurobiological and brain-imaging studies The vast majority of costs associated with OCD are indi- implicates serotonin (5-HT) receptor dysfunction in the patho- rect, resulting from job loss, absenteeism, and early retire- genesis of the disorder.11 There is also evidence for the possible ment.20 Direct treatment costs are comparatively small in roles of dopaminergic dysfunction,12 other neuropeptide abnor- part because many OCD patients never seek treatment and malities,13 and infective immunological mechanisms.10,14 some who do are incorrectly diagnosed and receive ineffec- The Role of Serotonin
tive treatment. Since OCD frequently manifests during During the 1980s, abnormality of the serotonergic system childhood or adolescence, the potential loss of income over a and, in particular, hypersensitivity of postsynaptic 5-HT lifetime is significant.20 As with other psychiatric disorders, receptors remained the leading hypothesis for the underlying OCD is associated with high healthcare costs in those who pathophysiology of OCD.11 Clinical studies have demon- do seek treatment. A survey by Hollander and colleagues22 strated that antiobsessional activity was a function of sero- found that OCD sufferers who have sought treatment spent tonin reuptake inhibition.15 Further support is derived from $4,000/year on outpatient provider costs and $1,500 on med- studies of markers and biological probes and from behavioral ication (in 1995 dollars). Subjects averaged lifetime hospital- responses to serotonergic challenge: ization costs of $12,500, and one in four were hospitalized for •Treatment response is correlated with decreased OCD, which extrapolates to $5 billion in lifetime hospital 5-hydroxyindolacetic acid (5-HIAA) levels in the cere- costs for the total United States OCD population.20,22 brospinal fluid of OCD patients.16 Furthermore, 28% of this patient population have received •Studies have demonstrated that platelet 5-HT trans- inappropriate treatment, resulting in an additional annual porter dysfunction, serotonin concentrations, and cost of approximately $2.2 billion.20,22 monoamine oxidase activity are correlated with symp- tom severity and response to selective serotonin reup- take inhibitor (SSRI) treatment.17-19 Currently, there are two sets of diagnostic criteria used to •Neuroendocrine challenge tests suggest that 5-HT1 define OCD: the World Health Organization International receptors may be altered in OCD.10 Classification of Diseases, Tenth Revision (ICD-10) criteria23 •Behavioral or physiological responses have been observed and the American Psychiatric Association Diagnostic and following challenge with metachloropheynlpiperazine, a Statistical Manual of Mental Disorders, Fourth Edition (DSM- serotonergic agonist.11 IV) criteria.24 The essential characteristic of OCD is the pres- ence of recurrent obsessive thoughts and compulsive acts, CLINICAL COURSE, CHRONICITY, AND
which are severe and take up many hours of the day or cause significant distress or impairment to the individual. These Most patients with OCD experience persistent symptoms obsessions and compulsions are at some point recognized by throughout their lives and require long-term treatment. the patient as excessive or unreasonable behavior.
Volume 8 – Number 8 (Suppl 1) CNS Spectrums - August 2003 Long-Term Treatment of Obsessive-Compulsive Disorder
ICD-10 classifies OCD as a stand-alone disorder within pharmacological treatment. The total Y-BOCS score gives neurotic, stress-related, and somatoform disorders, whereas the range of severity for patients who have both obsessions DSM-IV classifies OCD as one of a group of anxiety disorders. and compulsions (0–7=subclinical; 8–15=mild; 16–23=mod- erate; 24–31=severe; 32–40=extreme). For those with only Misdiagnosis of Obsessive-Compulsive Disorder
obsessions or compulsions, half of these score ranges corre- Despite high prevalence rates, OCD is still underdiagnosed late reasonably well with these severity descriptors.
and undertreated in both children and adults.10 A survey of 701 OCD patients showed, on average, a 17-year delay between onset of symptoms and commencement of appropri- Acute Treatment Trials
ate treatment.20,22 The average age of onset was determined as 14.5 years, with professional help sought at 25 years of age, A growing body of evidence supports the efficacy of the correct diagnosis made at 30 years of age, and appropriate TCA clomipramine for the treatment of OCD.
treatment given at 31.5 years of age. The reasons for these low Clomipramine, fluvoxamine, and sertraline have received rates of detection and diagnosis may include the following: indications for treatment of OCD in children and adoles- •Patients often conceal their symptoms, which they may cents. Clomipramine was the first agent approved for treat- perceive to be shameful, unique, or perverse.8 ment of OCD by the US Food and Drug Administration. In •Symptoms may overlap with those of other psychiatric two studies, clomipramine treatment for 10 weeks was associ- disorders,20 which can lead to misdiagnosis and inade- ated with significantly greater reductions in OCD symptoms quate and/or inappropriate treatment.
in nondepressed patients (mean Y-BOCS reduction of 38% •Traditionally, there has been a perception among profes- and 44%) compared with placebo (mean Y-BOCS reduction sionals that OCD is a treatment-resistant condition.8 of 3% and 5%).27 Side effects most frequently associated with clomipramine treatment include anticholinergic effects (dry mouth, constipation), postural dizziness, somnolence, weight Various conditions are frequently encountered as comorbid gain,28 and cardiovascular adverse effects (increase in stand- disorders with OCD. The most common comorbidity is major ing heart rate, decrease in standing systolic blood pressure).27 depressive disorder (MDD)25; up to two thirds of OCD patients The efficacy of clomipramine in the treatment of OCD is have a lifetime comorbid major depressive illness.6 After MDD, linked to its potency for inhibition of synaptic serotonin the most prevalent lifetime comorbid diagnoses reported in reuptake. This feature differentiates clomipramine from patients with OCD are: simple phobia (22%), social phobia other available TCAs, which have effects on both noradren- (18%), eating disorder (17%), alcohol dependence (14%), ergic and serotonergic reuptake, but are much less potent panic disorder (12%), and Tourette's syndrome (7%).26 than the SSRIs in serotonergic reuptake blockade. Studies comparing clomipramine with nortriptyline, amitriptyline, imipramine, and desipramine have shown that while these A growing body of evidence supports the efficacious use of TCAs treat depression effectively, they do not appear to the SSRIs fluvoxamine, paroxetine, sertraline, citalopram, have a specific effect on the symptoms of OCD.8,29 and fluoxetine, and the tricyclic antidepressant (TCA) Selective Serotonin Reuptake Inhibitors
clomipramine, for the treatment of OCD. Cognitive-behav- Paroxetine, sertraline, citalopram, clomipramine, fluoxe- ioral therapy (CBT) provides short- and long-term efficacious tine, and fluvoxamine have demonstrated acute efficacy in treatments for OCD without drug side effects. However, OCD treatment trials lasting 4–12 weeks.30,31 Clomipramine because CBT is not widely available, long-term pharma- and the SSRIs fluoxetine, fluvoxamine, paroxetine, and ser- cotherapy is now the usual treatment for OCD. For OCD traline are approved by the FDA for the treatment of OCD.
refractory to CBT and SSRIs, neurosurgical treatment is An indirect comparison of the FDA registration data for sometimes helpful. Efficacy of treatments are typically evalu- clomipramine, paroxetine, fluvoxamine, and sertraline, ated in clinical trials using the following rating instruments: found clomipramine more effective and as well tolerated.32 •Yale-Brown Obsessive-Compulsive Scale (Y-BOCS) Head-to-head comparisons of SSRIs to clomipramine gener- •National Institute of Mental Health Global Obsessive- ally show SSRIs to be equally effective and better tolerated.33- Compulsive Rating Scale (NIMH-OC) 36 Some of these studies have flawed designs that put •Behavioral Avoidance Test (BAT) clomipramine at a distinct disadvantage. In the study by •Maudsley Obsessive-Compulsive Inventory (MOC) Bisserbe and colleagues,33 for example, clomipramine was •Clinical Global Impression Scales: Severity (CGI-S) started at double the recommended starting dose and pro- and Improvement (CGI-I) duced more early dropouts than found in other studies using •Patient's Global Impression Scale (PGI) and the recommended dose.
•Obsessive-Compulsive Inventory Long-Term Treatment Trials
The 10 item Y-BOCS is the most widely used rating OCD is a chronic condition that can cause significant dis- scale for measuring changes in OCD symptoms during ability and often requires long-term management.37 Volume 8 – Number 8 (Suppl 1) CNS Spectrums - August 2003 Greist JH, Bandelow B, Hollander E, et al
Symptoms are likely to recur within a few weeks of discon- although the difference was not significant. Kaplan-Meier tinuation of therapy.38 Because long-term maintenance treat- 1-year relapse rates were 20.6% for fluoxetine and 31.6% for ment is frequently necessary, it is recommended that placebo (P=.137). Single-dose comparisons showed that practitioners prescribe agents that have established acute patients who continued treatment with fluoxetine 60 mg/day and long-term efficacy and a good tolerability profile, such as had significantly lower rates of relapse than those who were SSRIs. Several randomized, controlled trials (24 weeks) have switched to placebo (P=.041). Fluoxetine was well tolerated been reported for each of the pharmacologic agents approved over the 52-week study period.
for OCD treatment (Table).39-47 These treatments will be dis- cussed in order of their approval to market in the US.
The efficacy of long-term treatment with fluvoxamine in OCD has been studied in two trials.42,43 The first study Katz and colleagues39 assessed the efficacy of involved 60 outpatients diagnosed with OCD. A secondary clomipramine in OCD in a randomized controlled trial last- diagnosis of MDD was allowed if it had been preceded by ing >1 year. Patients (N=263) were randomized to receive obsessive-compulsive symptoms. Patients were randomized double-blind clomipramine (100–300 mg/day) or placebo for to receive fluvoxamine (up to 300 mg/day) with antiexpo- 10 weeks. Responders to therapy (n=124) were then entered sure therapy (F), fluvoxamine with exposure therapy (Fe), into a double-blind extension period for an additional or placebo with exposure therapy (Pe) for 24 weeks.42 52 weeks. In the initial 10-week phase of the study, patients Fe and Pe were double-blind while F was single-blind.
receiving clomipramine had clinically and statistically signif- Comparison of Fe with Pe tested the effectiveness of flu- icant reductions in global severity of OCD. Efficacy of voxamine while holding exposure constant. All three clomipramine was maintained in the extension period. At groups showed improvement in rituals and depression from the end of the extension trial, more than one-half of those week 0 to week 24, with a numeric superiority for com- patients who received clomipramine had benefited to such a bined treatment at week 24. Follow-up at 48 weeks (weeks degree that their OCD symptoms (as assessed by NIMH-OC 24 through 48 being drug-free) showed no between-group and CGI) no longer caused a significant interference in their difference in rituals or depression.
lives. A high incidence of medical problems (adverse events In the second study, 39 OCD patients were randomized to or concomitant illness) was noted in the clomipramine- receive double-blind fluvoxamine (50–300 mg/day) or treated group, with 22.7% of clomipramine-treated patients placebo for 10 weeks.43 Patients who completed the 10-week versus 0% of placebo patients discontinuing treatment due study period were then eligible to enter an open-label exten- to adverse events during the extension study.
sion phase of active fluvoxamine treatment for 8–52 weeks.
Among the 28 completers of the 10-week study, mean Following a 13-week double-blind, placebo-controlled improvement in Y-BOCS was significantly greater in the flu- trial of three fixed doses of fluoxetine (20, 40, and voxamine group than in the placebo group (33%±30% versus 60 mg/day) in 274 OCD subjects,40 treatment responders 5%±18%, respectively; P=.025). Twenty-one patients (n=76) were continued on blinded treatment, while acute entered an open-label extension for 8–52 weeks, of which fixed-dose nonresponders (n=198) began an open-label trial nine completed the total 52 weeks of treatment. Patients who on their maximally tolerated dose (up to 80 mg/day) for an previously received active drug treatment during the initial additional 24 weeks.48 In the blinded extension study, at the phase were continued on the same dose, while patients who end of the treatment period all three doses of fluoxetine were previously received placebo were switched to fluvoxamine associated with further Y-BOCS improvement from baseline.
(titrated up to a maximum of 300 mg/day). Fifty-seven per- Mean decreases from baseline to endpoint within the treat- cent (n=12) of patients responded or maintained response to ment group were statistically significant for the 60-mg dose fluvoxamine in the open-label extension. However, seven of only (P=.022). In the open-label study, subjects benefited nine patients reported recurrence of OCD symptoms within a from dose titration, with two thirds achieving a clinical few days to weeks following discontinuation of fluvoxamine.
response during the subsequent 24 weeks and more robust The most common side effects during long-term treatment results seen at either 60- or 80-mg/day doses. Fluoxetine was were sedation, tiredness, and anorgasmia.
well tolerated in both extension studies: 6% of subjects treated with fluoxetine in the responder extension discontin- Sertraline has been extensively studied in the long-term ued due to adverse events.
treatment (up to 2 years) of OCD. Greist and colleagues44 A further study assessed the efficacy and safety of fluoxe- studied the efficacy of sertraline in a 1-year placebo-con- tine versus placebo in preventing relapse of OCD during trolled randomized trial. The study began with a 12-week 1 year of treatment.42 Patients (N=130) were treated with phase of fixed-dose sertraline treatment (50, 100, or single-blind fluoxetine (20, 40, or 60 mg/day) for 20 weeks.
200 mg/day) or placebo, after which treatment responders Responders (n=71) were then randomized to continue dou- (N=118, including placebo patients) were offered an addi- ble-blind treatment with fluoxetine or placebo. Patients who tional 40 weeks of double-blind treatment at previously received fluoxetine had numerically lower relapse rates over assigned doses. At the end of the study, the pooled sertra- 52 weeks compared with those who received placebo, line group exhibited significantly greater improvement Volume 8 – Number 8 (Suppl 1) CNS Spectrums - August 2003 Long-Term Treatment of Obsessive-Compulsive Disorder
than the placebo group on all efficacy measures (Y-BOCS, NIMH-OC, CGI-S, and MOC). Furthermore, pairwise There are no published data for the long-term treatment analysis revealed a significant effect on all three investiga- of OCD with paroxetine. A study was presented at the tor-rated scales for patients receiving 50 or 200 mg/day of American College of Neuropsychopharmacology meeting in sertraline; in the 100-mg/day group there was a significant 2000, but citations are not permitted. effect on the NIMH-OC. Sertraline was well tolerated; during the initial phase, 10% of patients withdrew from To date, only short-term, open-label trials of citalopram in the study due adverse events (compared with 6% of OCD patients who did not respond to one or more SSRIs placebo-treated patients), and only an additional 4% of have been reported.30,31 patients stopped treatment due to adverse events (versus 5% for placebo) during the 40-week continuation phase.
At 1 year of randomized treatment, responders were then There is insufficient systematic research on the use of offered another year of sertraline treatment pharmacologic augmentation for treatment-refractory OCD (50–200 mg/day) in an open-label extension study.45 The patients. Two placebo-controlled trials of lithium augmenta- 38 patients who completed a full 2 years of treatment with tion of SSRIs and three placebo-controlled trials of buspirone sertraline exhibited a mean improvement of 15.6 points on augmentation of SSRIs found lithium and buspirone to be no the Y-BOCS. Patients exhibited significant improvement more efficacious than placebo. Greater efficacy was obtained in OCD symptoms during open sertraline treatment as by adding risperidone to an SSRI or clomipramine in patients assessed by Y-BOCS, NIMH-OC, and CGI-S (P<.05).
who failed to respond to monotherapy in an open trial.21 This Long-term sertraline treatment did not appear to be asso- result was corroborated in a double-blind, placebo-controlled ciated with the emergence, increased incidence, trial.46 In another trial, response to risperidone augmentation or increased severity of adverse events, or with clinically appeared to be influenced by symptom subtypes and comor- significant abnormalities in laboratory tests, vital signs, bid conditions: patients with horrific mental imagery had the strongest and fastest response, and patients with comorbid In the sertraline relapse-prevention study, adults with psychotic disorders improved gradually over 2–3 weeks.51 OCD who had responded to 52 weeks of open-label treat- Risperidone was associated with a high rate of adverse events ment with sertraline (50–200 mg/day) were randomized to that led to discontinuation in 24% of patients.21,52 Preliminary 28 weeks of double-blind, placebo-controlled treatment results of a double-blind, placebo-controlled crossover trial of (N=232).49 Significantly fewer discontinuations due to risperidone (1 mg/day) versus haloperidol (2 mg/day) aug- relapse or insufficient clinical response occurred in sertra- mentation in SSRI-refractory OCD patients for 9 weeks sug- line-treated patients (9%) compared with placebo-treated gest that patients show more clinical benefit with risperidone patients (24%) during the placebo-controlled 28-week phase and improved performance on cognitive tasks.53 Most patients (P=.004). At the end of the randomized study, sertraline was were unable to tolerate haloperidol or experienced consider- found to be significantly superior to placebo in preventing an able side effects.
acute exacerbation of OCD (rates of exacerbation measured The addition of sertraline after 6 months of failed as 12% versus 35%, respectively; P=.001) and in maintain- clomipramine treatment has been shown to improve patient ing improvements in Y-BOCS, NIMH-OC, CGI-S, and response considerably.54 A significant decrease in mean CGI-I scores.47 Sertraline was also significantly superior to global Y-BOCS score was observed in patients treated with placebo in maintaining improvements in QOL, as measured clomipramine and sertraline. Although increasing the dose by the patient-rated Quality of Life Enjoyment and of clomipramine was as effective as the addition of sertraline Satisfaction Questionnaire. Sertraline was well tolerated; to the lower dose, there were increased side effects leading to 13.3% of subjects discontinued the study due to adverse higher rates of discontinuation.
events during the open-label 52-week phase, while 5% of sertraline subjects and 11% of placebo subjects discontinued Treatment of Comorbid Disorders
for these reasons during the 28-week double-blind phase.47 Depression and OCD
Few studies have directly compared different SSRIs in the In the only double-blind study of sufficient size to assess treatment of OCD. In a 24-week study comparing sertraline the efficacy of antidepressants in the treatment of OCD with (50–200 mg/day) and fluoxetine (20–80 mg/day) for the concurrent MDD, sertraline-treated patients were signifi- treatment of OCD,50 equivalent and significant improve- cantly better at endpoint on measures of OCD and MDD ments (P<.001) in Y-BOCS and NIMH-OC scores were symptoms compared with those treated with desipramine observed for both sertraline and fluoxetine. However, (a norepinephrine reuptake inhibitor) after 12 weeks of patients treated with sertraline were 42% more likely to have treatment.29 Sertraline-treated patients were also signifi- achieved a response by week 12 than those treated with flu- cantly more likely to achieve a robust improvement in OCD oxetine. Moreover, sertraline treatment was associated with symptoms (40% reduction in baseline Y-BOCS). More a higher percentage of patients in remission (CGI <2 and patients receiving desipramine than sertraline discontinued Y-BOCS <11) at weeks 12 (P=.047) and 24 (P=.075).
treatment due to adverse events.
Volume 8 – Number 8 (Suppl 1) CNS Spectrums - August 2003 Greist JH, Bandelow B, Hollander E, et al
Tic Disorder and OCD
reports in this area suggest that this group of medications (ie, For OCD accompanied by a tic disorder, small doses of atypical neuroleptics) may be useful.
pimozide or haloperidol, in addition to the serotonergic drug, are associated with a higher therapeutic response.55 Schizophrenia and OCD
If a patient has both schizophrenia and OCD, the addi- CBT provides short- and long-term efficacious treatment tion of antiobsessive treatment to the ongoing antipsychotic for OCD and avoids drug side effects. However, because therapy may be useful.56 This combination has been associ- CBT is not widely available, long-term pharmacotherapy is ated with a somewhat better outcome in some patients who now the usual treatment for OCD. The refinement (first are otherwise difficult to treat. The role of mixed dopaminer- described by Meyer in 196657) of behavioral therapies such as gic and serotonergic blockers, such as risperidone, in this exposure and ritual prevention therapy (the field has subset of patients has not been studied systematically. The adopted "ritual prevention" as a clearer description of what is pharmacologic profile of these medications and some open done rather than "response prevention") in the early 1970s TABLE. LONG-TERM TREATMENT TRIALS OF PHARMACOTHERAPY FOR OCD
Katz et al (1990)39 263 entered, 124 continued RCT ➞ RCT extension RCT fixed dose ➞ RCT 76 responders continued RCT; continuation or switch to 198 switched to OL dose titration OL dose titration Romano et al (2001)41 130 entered OL; OL flexible dose ➞ RCT 71 responders randomized exposure therapy; antiexposure therapy Mallya et al (1992)43 21 continued OL treatment Greist et al (1995)44 325 entered; 118 continued RCT ➞ RCT continuation 59 entered; continuation of ➞ OL extension prior Greist et al (1995)44 Koran et al (1999)47 649 entered; 224 randomized OCD=obsessive-compulsive disorder; RCT=randomized controlled trial; wks=weeks; OL=off-label; NIMH-OC=National Institute of Mental Health Global Obsessive-Compulsive rating scale; PGI=Patient's Global Impression Scale; HDRS=Hamilton Depression Rating Scale; Y-BOCS=Yale-Brown Obsessive Compulsive Scale; CGI-S=Clinical Global Impression-Severity; AEs=adverse events; CGI-I=Clinical Global Impression-Improvement; BAT=Behavioral Avoidance Test; Volume 8 – Number 8 (Suppl 1) CNS Spectrums - August 2003 Long-Term Treatment of Obsessive-Compulsive Disorder
resulted in the first empirically validated treatment for person to a situation that provokes the ritual while asking OCD.58 Until that time, OCD was believed to be refractory the person to forgo ritualizing. Behavioral therapy may be of to psychotherapy and carried a poor prognosis.58 Treatments great benefit to the patient, resulting in the maintenance of such as psychodynamic therapy, which focused on the mean- a clinical response after therapy is discontinued.60 ing of obsessions and compulsions, had not been successful.
The reported success of behavioral treatments in signifi- The efficacy of exposure and ritual prevention, which focus cantly reducing OCD symptoms is 20% to 60% in eligible on compulsive behaviors as treatment targets, has been well patients.61 As with medication treatment, success depends, in documented over the past 2 decades.58 part, on the clinical severity of the problem and whether the patient has a comorbid Axis I disorder and/or personality dis- order. Patients with comorbid severe depression seldom Behavioral therapy (ie, exposure in vivo and ritual pre- respond to behavioral therapy alone.60 The major obstacle to vention in particular), if tolerable to patients, is one of the successful outcomes with behavioral therapy is poor compli- most effective therapies for OCD.59 It involves exposing the ance or adherence. About 15% of patients find the prospect Main Efficacy Variables Discontinuations due to medical problems in extension: 22.7% clomipramine; 0% placebo 5% of responders in RCT extension discontinued due to adverse events No significant difference in No clinically significant differences relapse rate versus placebo overall in AEs, vital signs, or laboratory (only significant for patients on values between fluoxetine and a 60-mg/day dose at start of RCT) placebo in 52-wk phase BAT target ritual score: time, =Placebo+exposure 7% of fluvoxamine-treated discomfort, duration per day patients withdrew due to drug- related side effects 57% of patients improved Most common adverse events were sedation, tiredness, anorgasmia No statistically significant 40 wks RCT continuation continuation study) differences between sertraline and placebo groups in incidence of vital sign or laboratory abnormalities during RCT continuation 52 wks OL extension Mean improvement of Sertraline was well tolerated; 15.6 points on Y-BOCS adverse events decreased during for patients completing second year of treatment 2 years of treatment No statistically significant difference in incidence of 52 wks OL sertraline+ No statistically significant difference in incidence of adverse events, laboratory or vital sign abnormalities MOC=Maudsley Obsessive-Compulsive Inventory; Q-LES-Q=Quality of Life Enjoyment and Satisfaction Questionnaire.
Greist JH, Bandelow B, Hollander E, et al. CNS Spectrums. Vol 8, No 8 (suppl 1). 2003.
Volume 8 – Number 8 (Suppl 1) CNS Spectrums - August 2003 Greist JH, Bandelow B, Hollander E, et al
of exposure and ritual prevention too frightening and refuse significant. Most clinicians believe that a treatment it outright.60 Approximately 10% of patients who try behav- approach combining exposure and ritual prevention and ioral therapy find that it engenders so much anxiety that pharmacotherapy is the best course of action, although the they discontinue treatment.60 However, the 75% adherence limited availability of trained behavioral therapists makes rate with behavior therapy is at least as high as that reported this treatment difficult to administer on a large scale.58 with pharmacotherapy, which should be continued for 1–2 years to reduce the risk of relapse that occurs rapidly OVERVIEW OF AVAILABLE
with shorter durations of treatment. Patients with severe depression rarely respond to behavioral therapy alone and Expert Consensus Guidelines for the treatment of OCD need direct intervention, such as antidepressants or electro- were published in 1997.67 The recommended initial treat- convulsive therapy.60 ment of choice is either CBT alone or in combination with Relapse is reported in about 30% of behavioral therapy an SSRI or clomipramine. The likelihood that medication cases.61 However, a maintenance program, following an will be included in the recommendation depends on the intensive program of exposure and ritual prevention and severity of the OCD and the age of the patient. In mild cognitive therapy (CT), has proven effective in preventing OCD, CBT alone is the initial choice, as it is for younger relapse for up to 2 years posttreatment in a small population patients. Combination treatment was rated best by experts in of patients.62 Patients maintained their gains for the 2-year terms of efficacy, speed, durability, tolerability, and accept- follow-up period on measures of anxiety associated with ability, suggesting that overall it may be the most successful avoidance, obsessions, compulsions, and anxiety. Further, treatment approach for the majority of patients.
patients were able to effectively manage relapses without In terms of psychotherapy, experts consider the combina- additional therapist intervention.61 tion of exposure and ritual prevention and CBT as the opti- Several meta-analyses have suggested behavioral therapy mal behavioral therapy for OCD. Regarding to be at least as effective as pharmacotherapy.63-65 A single pharmacotherapy, both SSRIs and clomipramine are the rec- meta-analysis by Kobak and colleagues58 found no significant ommended treatment options, although the guidelines note difference between exposure and ritual prevention therapy that SSRIs are associated with fewer adverse events than and pharmacotherapy with SSRIs or clomipramine.
clomipramine. If there is no response to first-line treatment In a recent long-term, follow-up study of the effect of with either an SSRI or CBT alone, a combination of both behavioral therapy with pharmacologic treatment in patients treatment strategies is the preferred option. If there is no with OCD, Alonso and colleagues66 observed greater, but not response to combination therapy, switching SSRIs and con- statistically significant, reductions in both Y-BOCS global tinuing CBT is recommended. In patients with a partial (44%±27%) and obsessions subscale (41%±27%) scores in response to combination therapy, switching SSRIs, providing patients who completed behavioral therapy. However, these more CBT, or possibly augmenting treatment with another patients showed significantly greater improvement in the medication is recommended.
compulsions subscale (P=.01).
Once patients have responded to acute-phase treatment for OCD, a maintenance phase of continued pharmacother- apy with monthly follow-up visits for a minimum of CT has also been found to be effective in treating patients 3–6 months is recommended. It is highly recommended that with OCD. However, in research and clinical use, cognitive pharmacotherapy be continued for extended periods of time.
and behavioral techniques are almost always used in combi- One or two years are often needed, and much longer periods nation as CBT.59 The general strategies of CBT are: consider are usually required most of the time. Booster CBT sessions the intrusive obsessional thoughts as stimuli; identify the dis- may help reduce the risk of relapse while medication is being tressing thoughts; challenge these automatic thoughts; and titrated downward and possibly withdrawn. Long-term or change these thoughts to nondistressing thoughts, while lifelong prophylactic maintenance medication is recom- making sure the patient does not avoid the obsessive mended after two to four severe relapses, or three to four thoughts as such avoidance is a form of antiexposure which has been shown to worsen OCD. The underlying dysfunc- tional assumptions are also identified and challenged in ther- apy sessions and the patient continues these exercises as OCD is a common disorder that has a profound effect homework assignments.65 on patient QOL and is associated with significant impair- ment in social and occupational functioning. Both psy- COMBINING PSYCHOTHERAPY AND
chotherapy (ie, CBT) and pharmacotherapy, separately or in combination, are recommended to treat the symptoms A meta-analysis by Kobak and colleagues58 found that of OCD. Pharmacotherapies that target serotonergic treatment with exposure and response prevention and an mechanisms-including clomipramine and the SSRIs ser- SSRI or clomipramine had a greater effect than either treat- traline, citalopram, fluoxetine, fluvoxamine, and paroxe- ment alone, but the differences were not statistically tine are recommended first-line treatment for OCD. The Volume 8 – Number 8 (Suppl 1) CNS Spectrums - August 2003 Long-Term Treatment of Obsessive-Compulsive Disorder
SSRIs fluoxetine, fluvoxamine, and sertraline, and the 17. Flament MF, Rapoport JL, Murphy DL, Berg CJ, Lake CR. Biochemical changes during clomipramine treatment of childhood obsessive-com- TCA clomipramine, have demonstrated efficacy in pub- pulsive disorder. Arch Gen Psychiatry. 1987;44:219-225.
lished long-term treatment trials (6 months). Current 18. Marazziti D, Hollander E, Lensi P, Ravagli S, Cassano GB. Peripheral guidelines recommend that pharmacotherapy of OCD be markers of serotonin and dopamine function in obsessive-compulsive continued for a minimum of 1–2 years in treatment- disorder. Psychiatry Res. 1992;42:41-51.
responsive individuals. In the event of relapse due to treat- 19. Marazziti D, Pfanner C, Palego L, et al. Changes in platelet markers of ment withdrawal, lifelong prophylactic maintenance obsessive-compulsive patients during a double-blind trial of fluvoxam- medication may be necessary. Further studies are required ine versus clomipramine. Pharmacopsychiatry. 1997;30:245-249.
20. Hollander E, Kwon JH, Stein DJ, Broatch J, Rowland CT, Himelein to determine the optimal duration of long-term pharma- CA. Obsessive-compulsive and spectrum disorders: overview and quali- cotherapy for OCD. Finally, despite the number of avail- ty of life issues. J Clin Psychiatry. 1996;57(suppl 8):3-6.
able treatments, many patients remain treatment-resistant 21. Saxena S, Wang D, Bystritsky A, Baxter LR Jr. Risperidone augmenta- and treatment-refractory. Additional research is needed to tion of SRI treatment for refractory obsessive-compulsive disorder. J Clin identify better treatment and management strategies for 22. Hollander E, Broatch J, Himelein C. The economic burden of OCD [abstract]. Curr Opin Psychiatry. 1999;12(suppl 1):116.
23. The International Classification of Diseases. 10th revision. Geneva, Switzerland: World Health Organization; 1992.
1. Kramer H, Sprenger J. Malleus Maleficarum. London, England: Pushkin 24. Diagnostic and Statistical Manual of Mental Disorders. 4th ed.
Press; 1951.
Washington, DC: American Psychiatric Association; 1994.
2. Shakespeare W. MacBeth. Act 5, scene 1, lines 32-34. 25. Black DW, Goldstein RB, Noyes R Jr, Blum N. Psychiatric disorders in 3. Robins LN, Helzer JE, Weissman MM, et al. Lifetime prevalence of spe- relatives of probands with obsessive-compulsive disorder and co-morbid cific psychiatric disorders in three sites. Arch Gen Psychiatry. major depression or generalized anxiety. Psychiatr Genet. 1995;5:37-41.
26. Pigott TA, L'Heureux F, Dubbert B, Bernstein S, Murphy DL. Obsessive 4. Weissman MM, Bland RC, Canino GJ, et al. The cross national epidemi- compulsive disorder: comorbid conditions. J Clin Psychiatry. ology of obsessive compulsive disorder. The Cross National Collaborative Group. J Clin Psychiatry. 1994;55(suppl):5-10.
27. Clomipramine in the treatment of patients with obsessive-compulsive 5. Kessler RC, McGonagle KA, Zhao S, et al. Lifetime and 12-month disorder. The Clomipramine Collaborative Study Group. Arch Gen prevalence of DSM-III-R psychiatric disorders in the United States.
Results from the National Comorbidity Survey. Arch Gen Psychiatry. 28. Lydiard RB, Brawman-Mintzer O, Ballenger JC. Recent developments in the psychopharmacology of anxiety disorders. J Consult Clin Psychol. 6. Rasmussan SA, Baer L, Shera D. Previous SRI treatment and efficacy of sertraline for OCD: combined analysis of 4 multicenter trials. [abstract].
29. Hoehn-Saric R, Ninan P, Black DW, et al. Multicenter double-blind com- Abstract presented at: Annual Meeting of the American Psychiatric parison of sertraline and desipramine for concurrent obsessive-compulsive Association; May 17-22, 1997; San Diego, CA.
and major depressive disorders. Arch Gen Psychiatry. 2000;57:76-82.
7. Riddle M. Obsessive-compulsive disorder in children and adolescents. Br 30. Pato MT. Beyond depression: citalopram for obsessive-compulsive disor- J Psychiatry. 1998;(suppl):91-96.
der. Int Clin Psychopharmacol. 1999;14(suppl 2):S19-S26.
8. Fineberg N. Refining treatment approaches in obsessive-compulsive dis- 31. Marazziti D, Dell'Osso L, Gemignani A, et al. Citalopram in refractory order. Int Clin Psychopharmacol. 1996;11(suppl 5):13-22.
obsessive-compulsive disorder: an open study. Int Clin Psychopharmacol. 9. Sasson Y, Zohar J, Chopra M, Lustig M, Iancu I, Hendler T.
Epidemiology of obsessive-compulsive disorder: a world view. J Clin 32. Greist JH, Jefferson JW, Kobak KA, Katzelnick DJ, Serlin RC. Efficacy Psychiatry. 1997;58(suppl 12):7-10.
and tolerability of serotonin transport inhibitors in obsessive-compul- 10. Rauch SL, Whalen PJ, Curran T, et al. Probing striato-thalamic func- sive disorder. A meta-analysis. Arch Gen Psychiatry. 1995;52:53-60.
tion in obsessive-compulsive disorder and Tourette syndrome using neu- 33. Bisserbe JC, Lane RM, Flament MF, et al. A double-blind comparison of roimaging methods. Adv Neurol. 2001;85:207-224. sertraline and clomipramine in outpatients with obsessive-compulsive 11. Zohar J, Mueller EA, Insel TR, Zohar-Kadouch RC, Murphy DL.
disorder. Eur Psychiatry. 1997;12:82-93.
Serotonergic responsivity in obsessive-compulsive disorder. Comparison 34. Freeman CP, Trimble MR, Deakin JF, Stokes TM, Ashford JJ.
of patients and healthy controls. Arch Gen Psychiatry. 1987;44:946-951.
Fluvoxamine versus clomipramine in the treatment of obsessive com- 12. McDougle CJ, Goodman WK, Price LH, et al. Neuroleptic addition in pulsive disorder: a multicenter, randomized, double-blind, parallel group fluvoxamine-refractory obsessive-compulsive disorder. Am J Psychiatry. comparison. J Clin Psychiatry. 1994;55:301-305.
35. Zohar J, Judge R. Paroxetine versus clomipramine in the treatment of 13. McDougle CJ, Barr LC, Goodman WK, Price LH. Possible role of neu- obsessive-compulsive disorder. OCD Paroxetine Study Investigators. Br ropeptides in obsessive compulsive disorder. Psychoneuroendocrinology. J Psychiatry. 1996;169:468-474.
36. Milanfranchi A, Ravagli S, Lensi P, Marazziti D, Cassano GB. A double- 14. Leonard HL, Swedo SE, Garvey M, et al. Postinfectious and other forms blind study of fluvoxamine and clomipramine in the treatment of obses- of obsessive-compulsive disorder. Child Adolesc Psychiatr Clin N Am. sive-compulsive disorder. Int Clin Psychopharmacol. 1997;12:131-136.
37. Montgomery SA. Long-term management of obsessive-compulsive dis- 15. Zohar J, Zohar-Kadouch RC, Kindler S. Current concepts in the phar- order. Int Clin Psychopharmacol. 1996;11(suppl 5):23-29.
macological treatment of obsessive-compulsive disorder. Drugs. 38. Thoren P, Asberg M, Cronholm B, Jornestedt L, Traskman L.
Clomipramine treatment of obsessive-compulsive disorder. I. A con- 16. Thoren P, Asberg M, Bertilsson L, Mellstrom B, Sjoqvist F, Traskman L.
trolled clinical trial. Arch Gen Psychiatry. 1980;37:1281-1285.
Clomipramine treatment of obsessive-compulsive disorder. II.
39. Katz RJ, DeVeaugh-Geiss J, Landau P. Clomipramine in obsessive-com- Biochemical aspects. Arch Gen Psychiatry. 1980;37:1289-1294.
pulsive disorder. Biol Psychiatry. 1990;28:401-414.
Volume 8 – Number 8 (Suppl 1) CNS Spectrums - August 2003 Greist JH, Bandelow B, Hollander E, et al
40. Tollefson GD, Rampey AH Jr, Potvin JH, et al. A multicenter investiga- 54. Ravizza L, Barzega G, Bellino S, Bogetto F, Maina G. Drug treatment of tion of fixed-dose fluoxetine in the treatment of obsessive-compulsive obsessive-compulsive disorder (OCD): long-term trial with disorder. Arch Gen Psychiatry. 1994;51:559-567.
clomipramine and selective serotonin reuptake inhibitors (SSRIs).
41. Romano S, Goodman W, Tamura R, Gonzales J. Long-term treatment of obsessive-compulsive disorder after an acute response: a comparison of 55. McDougle CJ, Goodman WK, Leckman JF, Lee NC, Heninger GR, fluoxetine versus placebo. J Clin Psychopharmacol. 2001;21:46-52.
Price LH. Haloperidol addition in fluvoxamine-refractory obsessive- 42. Cottraux J, Mollard E, Bouvard M, et al. A controlled study of fluvox- compulsive disorder: a double-blind, placebo-controlled study in amine and exposure in obsessive-compulsive disorder. Int Clin patients with and without tics. Arch Gen Psychiatry. 1994;51:302-308.
56. Sasson Y, Bermanzohn P, Zohar J. Treatment of obsessive-compulsive 43. Mallya GK, White K, Waternaux C, et al. Short- and long-term treat- syndromes in schizophrenia. CNS Spectr. 1997;2:34-35.
ment of obsessive-compulsive disorder with fluvoxamine. Ann Clin 57. Meyer V. Modification of expectations in cases with obsessional rituals.
Psychiatry. 1992;4:77-80. Behav Res Ther. 1966;4:273-280. 44. Greist JH, Jefferson JW, Kobak KA, et al. A 1 year double-blind place- 58. Kobak KA, Greist JH, Jefferson JW, Katzelnick DJ, Henk HJ. Behavioral bo-controlled fixed dose study of sertraline in the treatment of obses- versus pharmacological treatments of obsessive compulsive disorder: a sive-compulsive disorder. Int Clin Psychopharmacol. 1995;10:57-65.
meta-analysis. Psychopharmacology (Berl). 1998;136:205-216.
45. Rasmussen S, Hackett E, DuBoff E, et al. A 2-year study of sertraline in 59. Neziroglu F, Hsia C, Yaryura-Tobias JA. Behavioral, cognitive, and fam- the treatment of obsessive-compulsive disorder. Int Clin ily therapy for obsessive-compulsive and related disorders. Psychiatr Clin North Am. 2000;23:657-670.
46. McDougle CJ, Epperson CN, Pelton GH, Wasylink S, Price LH. A dou- 60. Greist JH. Behavior therapy for obsessive compulsive disorder. J Clin ble-blind, placebo-controlled study of risperidone addition in serotonin reuptake inhibitor-refractory obsessive-compulsive disorder. Arch Gen 61. O'Connor K, Todorov C, Robillard S, Borgeat F, Brault M. Cognitive- behaviour therapy and medication in the treatment of obsessive-com- 47. Goodman WK, Londborg PD, Lydiard RB. Safety of sertraline in long- pulsive disorder: a controlled study. Can J Psychiatry. 1999;44:64-71.
term OCD treatment: preliminary results of a multicenter study.
62. McKay D. A maintenance program for obsessive-compulsive disorder Abstract presented at: Annual Meeting of the American Psychiatric using exposure with response prevention: 2-year follow-up. Behav Res Association; Washington, DC; May 15-20, 1999. 48. Tollefson GD, Birkett M, Koran L, Genduso L. Continuation treatment 63. Christensen H, Hadzi-Pavlovic D, Andrews G, Mattick R. Behavior ther- of OCD: double-blind and open-label experience with fluoxetine. J Clin apy and tricyclic medication in the treatment of obsessive-compulsive dis- order: a quantitative review. J Consult Clin Psychol. 1987;55:701-711.
49. Koran LM, Hackett E, Rubin A, Wolkow R and Robinson D. Efficacy of 64. Cox BJ, Swinson RP, Morrison B, Lee PS. Clomipramine, fluoxetine, sertraline in the long-term treatment of obsessive-compulsive disorder.
and behavior therapy in the treatment of obsessive-compulsive disorder: Am J Psychiatry. 2002; 159:88-95.
a meta-analysis. J Behav Ther Exp Psychiatry. 1993;24:149-153.
50. Bergeron R, Ravindran AV, Chaput Y, et al. Sertraline and fluoxetine 65. van Oppen P, de Haan E, van Balkom AJ, Spinhoven P, Hoogduin K, treatment of obsessive-compulsive disorder: results of a double-blind, 6- van Dyck R. Cognitive therapy and exposure in vivo in the treatment of month treatment study. J Clin Psychopharmacol. 2002;22:148-154.
obsessive compulsive disorder. Behav Res Ther. 1995;33:379-390.
51. Pfanner C, Marazziti D, Dell'Osso L, et al. Risperidone augmentation in 66. Alonso P, Menchon JM, Pifarre J, et al. Long-term follow-up and predic- refractory obsessive-compulsive disorder: an open-label study. Int Clin tors of clinical outcome in obsessive-compulsive patients treated with Psychopharmacol. 2000;15:297-301. serotonin reuptake inhibitors and behavioral therapy. J Clin Psychiatry. 52. McDougle CJ, Goodman WK. Combination pharmacological treatment 2001; 62:535-540.
strategies. In: Hollander E, Stein D, eds. OCD: Diagnosis, Etiology and 67. Treatment of obsessive-compulsive disorder. The Expert Consensus Treatments. New York, NY: Marcel Dekker; 1999:203-223.
Panel for obsessive-compulsive disorder. J Clin Psychiatry. 1997;58(suppl 53. Li X, Jackson WT, Baxter LR. Risperidone vs. haloperidol augmentation in SRI refractory OCD: clinical and neuro-cognitive effects. Abstract presented at: 38th Annual Meeting of the American College Neuropsychopharmacology; December 12-16, 1999; Acapulco, Mexico; Volume 8 – Number 8 (Suppl 1) CNS Spectrums - August 2003 WCA Recommendations for the
Long-Term Treatment of Panic Disorder
By Mark H. Pollack, MD, Christer Allgulander, MD, Borwin Bandelow, MD, Giovanni B. Cassano, MD, John H. Greist, MD, Eric Hollander, MD, David J. Nutt, MD, FRCP, FRCPsych, FMedSci, Ahmed Okasha, MD, PhD, FRCP, FRCPsych, FACP, and Richard P. Swinson, MD, FRCPC, FRCPsych, DPM (DSM-III) in 1980.1 Panic attacks are accompanied by dis- •Selective serotonin reuptake inhibitors are currently the tressing anticipatory anxiety and often agoraphobia. This first-choice drugs in the treatment of panic disorder with disorder shows a chronic course associated with distress and or without agoraphobia.
disability. Patients with panic disorder report impairment in •Paroxetine, sertraline, citalopram, clomipramine, alprazo- social, marital, and vocational functioning.2 Panic disorder, lam, fluoxetine, and clonazepam have an approved indica- which is frequently comorbid with other psychiatric disor- tion for panic disorder in Europe and/or the United States.
ders, is associated with increased medical resource utilization, •Long-term pharmacologic treatment of panic disorder is premature mortality, and reduction in quality of life.2 The safe and effective in accruing continued improvement, dramatic presentation of panic disorder and the efficacy of maintaining benefit, and preventing relapse.
both psychosocial and pharmacologic treatments have made •Cognitive-behavioral therapy, alone or in combination with it the focus of extensive treatment and nosologic research.3 drug therapy, is an effective treatment for panic disorder.
Panic disorder is characterized by recurrent panic attacks, What are the symptoms of panic disorder and how is the disorder along with fearful anticipation of panic or the frightening most effectively treated? One of the most commonly encountered anx- consequences or implications of these attacks.2,4 Panic attacks, iety disorders in the primary care setting, panic disorder is a chronic presenting as sudden-onset episodes of intense fear or anxiety and debilitating illness. The core symptoms are recurrent panic attacks accompanied by rapidly peaking symptoms of cognitive and coupled with anticipatory anxiety and phobic avoidance, which autonomic arousal,2,4-6 occur at seemingly random intervals7 together impair the patient's professional, social, and familial function- and can occur during sleep.8 Panic attacks can also occur in ing. Patients with panic disorder have medically unexplained symp- an attenuated form called limited-symptom episodes.
toms that lead to overutilization of healthcare services. Panic disorder Agoraphobia is a frequent complication of panic disorder.4 is often comorbid with agoraphobia and major depression, and The anticipation of panic attacks often leads to agoraphobic patients may be at increased risk of cardiovascular disease and, possi- symptoms: anxiety about panic disorder symptoms appearing bly, suicide. Research into the optimal treatment of this disorder has in situations where attacks previously occurred, or from been undertaken in the past 2 decades, and numerous randomized, which escape or getting help in the event of an attack may be controlled trials have been published. Selective serotonin reuptake difficult. Epidemiologic surveys suggest that approximately inhibitors have emerged as the most favorable treatment, as they have one third of patients with panic disorder develop comorbid a beneficial side-effect profile, are relatively safe (even if taken in over- agoraphobia,9,10 in clinical samples this proportion is about dose), and do not produce physical dependency. High-potency benzo- 80%.11 Panic disorder in the absence of agoraphobia is some- diazepines, reversible monoamine oxidase inhibitors, and tricyclic times referred to as uncomplicated panic disorder.
antidepressants have also shown antipanic efficacy. In addition, cogni- The following five principal symptom domains of panic tive-behavioral therapy has demonstrated efficacy in the acute and disorder have been identified: panic attacks, including lim- long-term treatment of panic disorder. An integrated treatment ited-symptom episodes; anticipatory anxiety; panic-related approach that combines pharmacotherapy with cognitive-behavioral phobias (including agoraphobia and body-sensation pho- therapy may provide the best treatment. Long-term efficacy and ease bias); well-being/overall severity of illness; and functional of use are important considerations in treatment selection, as mainte- disability (including work, social, and family functioning).
nance treatment is recommended for at least 12–24 months, and in some cases, indefinitely. Symptoms of Panic Attacks
CNS Spectr 2003;8(suppl 1):17-30 The symptoms of a panic attack include heart palpita- tions, pounding heart, or accelerated heart rate; sweating; trembling or shaking; sensations of shortness of breath or of Panic disorder was first defined by the Diagnostic and being smothered; feeling of choking; chest pain or discom- Statistical Manual of Mental Disorders, Third Edition fort; nausea or abdominal distress; feeling dizzy, unsteady, These recommendations are based on proceedings from the World Council of Anxiety meeting held September 11, 2000, in Pisa, Italy, and on guidelines and articles published in the medical literature through October 15, 2001.
Please direct all correspondence to: Mark H. Pollack, MD, Center for Anxiety and Traumatic Stress Related Disorders, Massachusetts General Hospital, WAC-812, 15 Parkman St, Boston, MA 02114. Tel: 617-724-0844; Fax: 617-726-7541; E-mail: [email protected].
Volume 8 – Number 8 (Suppl 1) CNS Spectrums - August 2003 M.H. Pollack, C. Allgulander, B. Bandelow, et al.
light-headed, or faint; derealization (feelings of unreality) or complicated pattern of comorbid conditions. It is therefore depersonalization (feeling detached from oneself); fear of los- important to determine whether panic attacks are not better ing control or of going crazy; fear of dying; paresthesia accounted for by another anxiety disorder, such as social (numbness or tingling sensations); and chills or hot flashes.5 anxiety disorder (SAD), which occurs on exposure to feared Data suggest the existence of a prodromal phase of panic social situations; specific phobia, which is limited to a single disorder in some patients, ie, a time interval between the situation, eg, elevators; obsessive-compulsive disorder onset of prodromal symptoms and the onset of the character- (OCD), posttraumatic stress disorder (PTSD); or generalized istic manifestations of the fully developed illness.8 anxiety disorder (GAD). In addition, panic attacks may In addition to anticipatory anxiety, most patients with occur in the context of depressive and bipolar disorders.
panic disorder present with somatic symptoms, especially cardiac, gastrointestinal, or neurologic complaints.12 General population surveys have estimated the lifetime prevalence of panic disorder to be between 1% and 4%.3,10,20 Due to the presence of nonspecific somatic symptomatol- A World Health Organization survey of 15 worldwide sites ogy, when establishing the diagnosis of panic disorder, physi- estimated the average current prevalence at 1.1%,21 and the cians must consider other medical conditions, mental Epidemiologic Catchment Area study identified panic disor- disorders, and physiologic causes that may mimic panic der in 1.4% to 8% of primary care patients.12 Lifetime rates of attacks. Misdiagnosis of patients with panic disorder is fre- uncomplicated panic disorder are considerably lower than quent and is associated with considerable economic and rates of panic disorder complicated by comorbid psychiatric social impact.13 Many of the symptoms of panic disorder are conditions.22 Recurring panic attacks not meeting full diag- also cardinal features of cardiovascular diseases.14 Patients nostic criteria for panic disorder also have a relatively high with panic disorder may initially present to emergency care lifetime prevalence (3.6% in a study of 18,000 adults in the complaining of acute chest pain or other cardiovascular United States), and subsyndromal panic is associated with symptoms. Such patients then undergo costly, invasive car- significant morbidity and disability as well.23 The prevalence diac testing.14 At least one third of the 20% to 30% of of panic disorder is greatest in the 25- to 44-year-old age patients who undergo angiography to evaluate chest pain are group.3 It is more prevalent in women than in men, by a fac- found to have normal coronary arteries.15 tor of at least 2-fold.3,10,20,24,25 Although many patients with panic disorder are misdiag- In clinical samples of patients with panic disorder, four of nosed as cardiac cases, there is some evidence that panic dis- five typically also have agoraphobia,11 whereas in samples of order may be linked with cardiovascular disease.14 The the general population, the proportion tends to be lower. In prevalence of panic disorder in both cardiology outpatients the US National Comorbidity Survey, 50% of survey respon- and patients with documented cardiovascular disease ranges dents with panic disorder reported no symptoms of agora- from 10% to 50%.14 Men with panic disorder have been phobia,20 and in another, smaller community study in the shown to have an excess number of deaths due to circulatory US, the ratio was closer to 1:3.9,23 system disease.16 The risk for stroke in individuals with a life- time diagnosis of panic disorder could be higher than that for individuals with other psychiatric disorders or with no other The course of panic disorder is not uniform,26 but typically psychiatric disorder.16 Similarly, patients with panic disorder follows a chronic, recurrent, and usually debilitating course, appear to be at greater risk for hypertension and myocardial with relapses after remission.2,27 The mean age of onset tends infarction than individuals with no other psychiatric disor- to be in early to middle adulthood ( 20–30 years of age), der.17,18 In addition, there are preliminary data linking panic though it can date back to childhood.2,10,17,28 Patients with disorder to reduced heart rate variability, which may predis- early-onset panic disorder have a longer duration of illness pose persons with panic disorder to the development of than later-onset patients and are more likely to have other malignant arrhythmias.14 Panic disorder also shares a number of symptoms with The course of illness of panic disorder appears to be differ- asthma, such as dyspnea, choking, sensations of being smoth- ent for men than it is for women. Men and women are ered, and chest pain. Respiratory factors of this type play an equally likely to experience remission of panic disorder dur- important role in panic disorder, and the majority of panic- ing naturalistic follow-up, but in a 5-year, prospective, fol- disorder patients experience breathing-related problems.
low-up study, the rate of symptom recurrence in women was This makes diagnosing panic disorder in the presence of nearly double the rate in men.30 asthma a challenge. More than one in five individuals with Follow-up studies of up to 20 years' duration have asthma report having experienced panic attacks.19 demonstrated low rates of remission and poor long-term outcome, suggesting that full remission of panic disorder Other Anxiety and Mood Disorders
might occur in only 10% to 35% of patients who have Panic attacks may be experienced in several anxiety disor- undergone clinical treatment.3,9,20,24,30-32 However, patient ders, either as a response to specific triggers or as part of a samples derived from treatment settings may constitute a Volume 8 – Number 8 (Suppl 1) CNS Spectrums - August 2003 Long-Term Treatment of Panic Disorder
sample biased toward the more severely ill, who may have suicide. Lecrubier and Ustun21 showed that, among patients waited many years before receiving specialized clinical treat- with comorbid panic disorder and depression, 43% had a his- ment.23 A comparison of 1-month and lifetime prevalence tory of suicide attempts.
data in epidemiologic studies suggests that the remission rate in general population samples may be greater. These studies show that at least 60% of those with panic disorder Panic disorder complicated by other psychiatric conditions at some period of their lives did not experience symptoms in is more common than panic disorder alone.21 Patients with the month before being interviewed.3,20,23,24 uncomplicated panic disorder represent fewer than one third Long duration of illness and agoraphobia at baseline, and of all patients with panic disorder,22 and 25% of patients with not the severity and frequency of panic attacks, appear to panic disorder complicated by other psychiatric conditions are predict an unfavorable course.9,30,32 At 1 year, patients with likely to develop an additional first-onset disorder within 1 uncomplicated panic disorder were more than twice as likely year of follow-up.22 The presence of comorbidity results in to achieve full remission than patients with panic disorder more severe anxiety and depressive symptoms, higher rate of complicated by agoraphobia.27 suicide attempts, higher frequency of other comorbid condi- tions, and poorer treatment response and compliance than in SOCIAL AND HEALTH CONSEQUENCES
patients with panic disorder alone.21 Comorbidity is also asso- Panic disorder is associated with pervasive social and ciated with worse outcomes on selected measures of sympto- health consequences similar to or greater than those associ- matic and functional impairment.40 ated with major depression.13,33 The often chronic course of There is an association between panic disorder and panic disorder is associated with quality-of-life impair- depression. Between one third and two thirds of patients ment,34 as well as subjective poor physical and emotional with lifetime panic disorder have or have had at least health, substance abuse, increased likelihood of suicide one major depressive episode.23,28,35,40,41 Patients with agora- attempts, lower educational achievement,20 higher likeli- phobia have rates of depression comparable to patients with hood of unemployment and low work productivity,34 uncomplicated panic disorder.28 Patients with a longer dura- impaired social and marital functioning, and financial tion of illness, low self-reported assertiveness, early-onset dependency that cannot be attributable solely to comorbid- depression, melancholic depression, or family histories of ity with other psychiatric disorders.33 anxiety or depression are more likely to experience recurrent Patients with panic disorder are particularly high users of depression.28 Rates of panic disorder are also elevated in healthcare services12 and visit more specialists than patients patients with bipolar disorder.41,42 with other anxiety disorders.13 Studies suggest that approxi- In clinical populations, nearly one in three patients with mately one third of patients with panic disorder visit three or panic disorder also meets the diagnostic criteria for social more healthcare specialists per year.13,35 Almost one fifth of anxiety disorder.43 This comorbidity is associated with an patients with panic disorder had visited a general hospital increased likelihood of developing major depression.28 Other emergency department and one in ten had been hospitalized anxiety disorders associated with panic disorder include at some point for anxiety complaints.35 More than 40% of OCD, PTSD, and GAD; personality disorders and alcohol or panic disorder patients use both psychiatric and medical ser- other drug dependence or abuse also frequently complicate vices, compared with 4% of the general population.33 Nearly the clinical picture.
30% of patients with panic disorder have been assessed by a Although comorbidity of any type may occur in either cardiologist or a neurologist or both.35 They also see otolaryn- men or women with panic disorder, there is a tendency gologists, obstetricians-gynecologists, and urologists more fre- toward higher comorbidity rates in women for specific pho- quently than subjects with other psychiatric disorders.13 bia, GAD, manic-depressive episodes, and dysthymia. Men have a tendency for higher comorbidity rates for SAD and hypochondria, and are significantly more likely to have a Panic disorder and panic attacks are associated with an ele- history of alcohol dependence or abuse.24 vated risk of suicidal ideation and suicide attempts. The life- time rate of suicide attempts in patients with panic disorder is 20%,16,22,36,37 which is similar to that of subjects with major Both heritable factors and stressful life events, particu- depression.22 The comorbid conditions of depression, alcohol larly in early childhood, are conducive to onset of panic abuse, and personality disorders appear to be the main risk fac- disorder.7 Panic disorder is familial; it has its basis in an tors associated with risk of death from suicide attempts.22,38 unusually sensitive fear network, with the central nucleus In a clinical sample of 100 outpatients with panic disor- of the amygdala playing a significant role.7 Disrupted emo- der, Lepine and colleagues39 found that 42% had a history of tional attachments with significant caregivers during suicide attempts. Patients who had attempted suicide were childhood have been identified as a potential risk factor.7 significantly more likely to have suffered from a major The onset of panic disorder may also be precipitated by depressive episode and alcohol or other substance abuse exposure to drugs (eg, cocaine, marijuana, or ampheta- in their lifetime, compared with those who did not attempt mines) or alcohol use or withdrawal.44 Volume 8 – Number 8 (Suppl 1) CNS Spectrums - August 2003 M.H. Pollack, C. Allgulander, B. Bandelow, et al.
The pathophysiology of panic disorder and the neurobio- logic basis of panic attacks have been the focus of much Imipramine is the most widely investigated TCA in the research. Some of the proposed hypotheses of neurobiologic treatment of panic disorder, although it is not approved in dysfunctions involve the classic neurotransmitter systems, the US for this indication. Several trials have compared such as norepinephrine, γ-aminobutyric acid, serotonin, and imipramine with other therapeutic approaches in the acute the peptide cholecystokinin.7,8,45-50 treatment of panic disorder (eg, fluvoxamine,58 alprazo- lam,11,59-63 and cognitive-behavioral therapy [CBT]64).
Generally, these trials reported more side effects with An extensive range of clinician-based and self-report rat- imipramine than the comparator, including significant ing scales have been used in selecting treatment options for effects on a number of cardiovascular variables.59-61 panic disorder. Currently, the most frequently used measures Clomipramine has been evaluated in the treatment of are the panic diary (panic attack frequency), Clinical Global panic disorder in short-term (6–12 weeks), placebo-con- Impression (CGI) scale,51 Hamilton Rating Scale for trolled studies.56,65-72 These studies have consistently shown Anxiety (HAM-A),52 Hamilton Rating Scale for Depression clomipramine to be significantly more effective than placebo (HAM-D),53 Marks-Sheehan Phobia Scale,54 and the Panic in treating panic disorder (assessed using a variety of mea- Associated Symptom Scale.55 sures). Clomipramine has also been compared with Several recently developed scales of panic disorder severity imipramine in several trials. In one, 59 patients with panic assess all five of the key domains that reduce quality of life in a disorder (Diagnostic and Statistical Manual of Mental Disorders, single measure. Examples include the Panic Disorder Severity Third Edition-Revised [DSM-III-R])73 were randomly Scale (PDSS)4 and the Panic and Agoraphobia (P & A) scale.56 assigned to either clomipramine or imipramine at compara- ble, flexible dosages for 10 weeks at the beginning of a 2-year, nonblinded trial.74 By 10 weeks, both drugs were equally effec- The main objectives of treatment are to reduce the num- tive in blocking panic attacks, alleviating phobic avoidance, ber and intensity of panic attacks, reduce anticipatory anxi- and reducing nonspecific aspects of anxiety. However, during ety, and treat any underlying depression or other psychiatric the first 2 weeks, clomipramine was significantly more effec- comorbidities associated with panic disorder. The long-term tive than imipramine in its antipanic and antiphobic effects.74 goal is sustained full remission.4 Both pharmacologic and psy- Additionally, acute treatment with clomipramine (mean chosocial therapies are effective.
dose 109 mg/day) has been shown to be significantly superior to imipramine (mean dose 124 mg/day), as assessed by reduc- tion in full panic attack frequency, total panic attack fre- In 1962, Klein and Fink57 described various patterns of quency, and anxiety between attacks.72 No significant behavioral response to imipramine in a group of 180 inpa- differences in efficacy were observed between clomipramine tients predominantly diagnosed with schizophrenia or affec- (100 mg/day) and lofepramine (140 mg/day) in the 6-week, tive disorders, one of which was a reduction in episodic placebo-controlled, acute phase of the Galway Study of anxiety response, including the cessation of panic attacks, in Panic Disorder.66 a subset of 14 patients. This initial observation led to evalua- TCAs were the first antidepressants widely used for panic tion of a number of agents to treat panic disorder, and tri- disorder, and were considered first-line treatment for many cyclic antidepressants (TCAs), selective serotonin reuptake years.75 Although TCAs are effective in treating panic disor- inhibitors (SSRIs), and high-potency benzodiazepines der, they have disadvantages: they are associated with anti- (BZDs) have been found to be effective. Monoamine oxidase cholinergic and cardiovascular adverse effects that may affect inhibitors (MAOIs) and reversible inhibitors of monoamine patient compliance, and they are toxic in overdose. Because oxidase type A (RIMAs) have also shown efficacy in the of the greater tolerability and comparable efficacy of newer treatment of panic disorder.
classes of antidepressants, the TCAs are now generally There are advantages and disadvantages to each of these reserved for second- or third-line use.
classes of medications, and the choice of agent is determined Selective Serotonin Reuptake Inhibitors
by several factors, including side-effect profile, cost, comor- The SSRIs, either alone or in combination with BZDs, bidities, history of past response or failure, and patient prefer- are the treatment of first choice for panic disorder with most ence. Patients with panic disorder are disproportionately clinicians worldwide.75,76 They lack the serious side effects of sensitive to the side effects of medications and typically TCAs and MAOIs and the dependence problems associated require treatment initiation at lower doses than those used with BZDs.77 The efficacy of SSRIs in the treatment of panic for patients with depression, although maintenance doses disorder has been well established in clinical trials.
can be similar and may sometimes be higher.58 Paroxetine was the first SSRI to receive US Food and Drug Administration approval for use in panic disorder, and Acute Treatment Trials
acute-treatment trials have demonstrated its antipanic effi- Most research on the treatment of panic disorder has been cacy. In a placebo-controlled study of paroxetine in the treat- conducted as relatively brief efficacy trials of 6–12 weeks.
ment of panic disorder, evaluating fixed doses of 10, 20, or Volume 8 – Number 8 (Suppl 1) CNS Spectrums - August 2003 Long-Term Treatment of Panic Disorder
40 mg/day over 10 weeks, the minimum dose of paroxetine tremor, diarrhea, dry mouth, and dyspepsia.78 There were no that showed superiority over placebo was 40 mg/day. The 10- dose-related adverse events. Discontinuation rates were simi- and 20-mg doses showed a tendency toward superiority over lar for sertraline and placebo.
placebo but did not reach statistical significance. Adverse In a randomized, double-blind, parallel-group, flexible- effects that were significant versus placebo included decreased dose comparison of sertraline and placebo that began with a libido and abnormal ejaculation, tremor, diarrhea, dry mouth, 2-week, single-blind placebo lead-in, 168 patients were ran- and dyspepsia.78 In a 12-week, randomized, double-blind trial, domly assigned to either sertraline or placebo for 10 weeks.
Oehrberg and colleagues79 randomized 120 patients to treat- Sertraline was initially given at 25 mg/day for 1 week fol- ment with paroxetine (20, 40, or 60 mg/day) or placebo; stan- lowed by titration to 50–200 mg/day; mean dose at endpoint dardized CBT was given to all patients. Patients who received was 126 mg/day.83 Sertraline was significantly more effective paroxetine with CBT showed significantly greater improve- than placebo in decreasing the number of full and limited- ment in panic attack frequency than those in the placebo symptom panic attacks: the mean number of attacks per group, and a greater percentage had a reduced number of week decreased 88% in the sertraline group versus 53% in panic attacks (1 or 0) in the final 3-week interval (36% ver- the placebo group. The sertraline group also had higher sus 16%, respectively, at week 12; P=.024).
scores on the Quality of Life Enjoyment and Satisfaction Paroxetine 20–60 mg was compared with clomipramine Questionnaire, patient global evaluation, CGI-Severity, and 50–150 mg over 12 weeks in a placebo-controlled study in CGI-Improvement. Sertraline was well tolerated, with only 367 patients with DSM-III-R–defined panic disorder.68 9% of subjects terminating treatment due to adverse effects.
At the study endpoint (weeks 7–9, when at least 70% of the In another study, 176 patients participated in a multiple- patients remained in each treatment group), paroxetine was site trial using identical protocols with a flexible-dose significantly more effective than clomipramine (P=.041) or design.84 After 2 weeks of single-blind placebo treatment, placebo (P=.004), based on the number of patients achiev- subjects were randomized to 10 weeks of double-blind, flexi- ing no full panic attacks. A second short-term, placebo-con- ble-dose treatment with sertraline (50–200 mg/day) or trolled study assessed the relative efficacy of paroxetine placebo. The sertraline group exhibited significantly greater (20–60 mg/day), clomipramine (50–150 mg/day), and CBT improvement at endpoint (P=.01) than the placebo group in in 131 randomly assigned patients with DSM-III-R–defined panic attack frequency. The active treatment group also real- panic disorder.69 Paroxetine (mean dose 38.6 mg/day) was ized significant improvement on the CGI-Improvement significantly more effective than placebo in reducing panic (P=.01) and CGI-Severity (P=.009), PDSS (P=.03), high attacks (P=.01), agoraphobic complaints, anxiety, depres- end-state function assessment (P=.03), patient global evalu- sion, and social dysfunction (P<.005). With the exception of ation rating (P=.01), and quality-of-life scores (P=.003).
scores on panic frequency, paroxetine also showed superiority Adverse events were mild or moderate in both the sertraline over CBT on all rating scales (P<.05), but efficacy measures and placebo groups.
were not significantly different from those of clomipramine, Finally, in a pooled analysis of four double-blind, placebo- with the exception of the CGI-Severity score, on which controlled studies (n=664) (two flexible-dose and two fixed- scores in the paroxetine group were higher (P<.001).
dose), sertraline was effective in treating panic disorder even A recent 10-week, placebo-controlled pilot study evaluated in the presence of baseline clinical variables that have been the effects of combining paroxetine (10–50 mg/day) with a associated with poor response, such as high panic severity, very brief form of CBT in 33 patients with Diagnostic and presence of agoraphobia, greater chronicity, and female gen- Statistical Manual of Mental Disorders, Fourth Edition (DSM- der. Endpoint reduction in panic attack frequency was simi- IV)80–defined panic disorder.81 At week 10, the proportion of lar across all comparisons (78% to 82%, with and without a panic-free patients was significantly higher in the paroxetine- poor prognostic variable) and also when three or more high- than in the placebo-treated group (80% versus 25%; P<.007), risk variables were combined.85 as was the proportion of patients who rated themselves as very Fluvoxamine was the first SSRI to be evaluated in a dou- much improved (60% versus 13%; P<.017).
ble-blind study in patients with panic disorder.75 In the Another SSRI marketed in the US for panic disorder is 8-week study, 50 patients were randomized to fluvoxamine sertraline. In the first controlled study of sertraline in treat- (mean dose 207 mg/day) or placebo.86 Fluvoxamine was sig- ing panic disorder, 178 subjects were randomly assigned to nificantly more effective than placebo in the parameters of 12 weeks of sertraline (50, 100, or 200 mg/day) or placebo panic attack frequency (from week 3) and anxiety, depres- after a 2-week, single-blind, placebo lead-in.82 In the pooled sion, and disability (from week 6; assessed on the Clinical sertraline group (n=127), panic attacks decreased 65% from Anxiety Scale, the Montgomery-Asberg Depression Rating baseline versus 39% in the placebo group (n=44). All three Scale [MADRS], and the Sheehan Disability Scale). Other sertraline groups demonstrated similar efficacy, although the placebo-controlled trials with fluvoxamine have demon- reduction rate was greater among the lower-dose sertraline strated similar results.58,87-89 groups (71%, 83%, and 42% for 50 mg, 100 mg, and 200 mg, Fluvoxamine 150 mg/day was also compared with maproti- respectively). Adverse effects that were significant versus line 150 mg/day, a noradrenergic reuptake inhibitor, in a 6-week placebo included decreased libido and abnormal ejaculation, study in 44 patients with panic disorder.48 The fluvoxamine Volume 8 – Number 8 (Suppl 1) CNS Spectrums - August 2003 M.H. Pollack, C. Allgulander, B. Bandelow, et al.
group had significant reductions in panic attack frequency, placebo, citalopram, or clomipramine (60–90 mg/day).
avoidance behavior, and depressive symptomatology, while Citalopram at doses of 20, 30, 40, or 60 mg/day was signifi- maprotiline had virtually no effect on anxiety symptoms cantly more effective than placebo in treating panic disorder, (assessed with HAM-A and State-Trait Anxiety Inventory).
in both physicians' and patients' Global Improvement Scales, The therapeutic properties of fluvoxamine were apparent begin- as well as MADRS and HAM-A total scores (P<.005).
ning at week 4. Another randomized, placebo-controlled trial However, citalopram at 10–15 mg/day was no more effective compared fluvoxamine and imipramine over 8 weeks in than placebo. No significant differences in treatment effect 54 patients with DSM-III-R–defined panic disorder.58 between citalopram and clomipramine were reported.
Fluvoxamine (mean dose at week 8:147 mg/day) was associated with significantly fewer panic attacks at week 8 than either High-potency BZDs have been shown to be effective in imipramine or placebo (P<.01). However, no statistically signifi- rapidly decreasing panic symptoms in a number of trials.
cant differences in Clinical Anxiety Scale ratings among groups However, a drawback of long-term BZD use is the possibil- were noted at study endpoint.
ity of dependence and associated withdrawal symptomatol- Fluvoxamine has also been compared with CBT in ogy on discontinuation. One study showed that 70% of patients with moderate-to-severe panic disorder,87 and was patients who had panic disorder and were treated with found to be superior to CBT and to placebo on most out- alprazolam had a discontinuation syndrome, which usually come measures (P<.05), including number of panic attacks, involved anxiety and agitation, when they attempted to global improvement, and depression (assessed by Clinical decrease the dosage.92 The antidepressants have been Anxiety Scale, CGI, and MADRS). Additionally, fluvoxam- shown to be more effective than the BZDs in treating con- ine produced improvement earlier than CBT. A subsequent comitant depressive symptomatology and at least as effec- 96-patient, randomized, 4-arm trial (fluvoxamine+exposure, tive in improving anxiety, agoraphobic avoidance, and placebo+exposure, CBT+exposure, and exposure alone) sug- overall improvement.93 gested that the combination of fluvoxamine with exposure Alprazolam is the most widely investigated BZD in the was significantly superior to all other treatments in diminish- treatment of panic disorder. Phase I of the Cross-National ing agoraphobic avoidance, the primary endpoint (assessed Collaborative Panic Study (n=526) was an 8-week, placebo- using an agoraphobia composite) over a 3-month period.89 controlled trial of flexible-dose alprazolam (2–10 mg/day; No differences in number of panic attacks were observed mean dose at week 8 was 5.7 mg/day).94 Alprazolam was sig- among the groups, probably due to the relatively low number nificantly more effective than placebo in reducing panic of panic attacks at baseline.
attack frequency, phobic fears, avoidance behavior, anxiety, A 10-week, randomized, placebo-controlled trial com- and secondary disability (assessed by a range of measures).
pared fluoxetine, 10 and 20 mg/day, and placebo in Some of these patients (n=109) subsequently underwent 243 patients with DSM-III-R–defined panic disorder.90 a closely monitored placebo-controlled tapered reduction of Fluoxetine, 10 mg/day, had a statistically significantly greater study medication over 4 weeks, with further observation for reduction in total panic attack frequency than placebo another 2 weeks.95 The alprazolam-treated group had signifi- (P=.006), and 20 mg/day had a statistically significantly cant relapse between the first and last week of taper, sug- greater improvement than placebo in a range of symptom gested by deterioration in measures of total panic attacks, domains, including anxiety phobia and depression.
freedom from panic attacks, anxiety on the HAM-A Scale, A more recent, placebo-controlled trial in 25 patients overall phobic state, and the Physician's Global Scale. By the with panic disorder who had failed to respond to an 8-week last posttaper observation week, there were no significant trial with fluoxetine 20 mg/day suggests that pindolol 2.5 mg differences between the two groups.
TID has an augmenting effect on fluoxetine in patients with A wider dosage range of alprazolam was investigated in a treatment-resistant panic disorder.91 Pindolol is a β-blocker 6-week, flexible-dose study that randomized 94 patients with with intrinsic sympathomimetic activity. It was examined as defined panic disorder to either alprazolam 2 mg, alprazolam a potential augmenting agent for refractory panic because it 6 mg, or placebo.96 Both active-treatment groups improved also blocks presynaptic 5-HT1A and 5-HT1B autoreceptors, significantly more than the placebo group on most outcome potentially increasing synaptic serotonin. Patients were ran- measures. The pattern of treatment response across measures domized to receive either pindolol or placebo in addition to suggested a dose effect, although only a few statistically sig- continued treatment with fluoxetine 20 mg/day for 4 weeks; nificant differences between the 2- and 6-mg group were those randomized to fluoxetine+pindolol had significant reported (eg, frequency of anticipatory episodes, P<.03; improvements on a number of rating scales, including Work and Social Disability rating, P<.01).
HAM-A, HAM-D, and CGI, compared with the fluoxe- Another study randomized 154 patients to 8 weeks of tine+placebo group.
either alprazolam+exposure therapy, alprazolam+relaxation Currently, only one double-blind, placebo-controlled (a psychologic placebo), placebo+exposure therapy, or study of citalopram in panic disorder has been published.70 placebo+relaxation (double placebo), followed by complete A total of 475 patients (with DSM-III-R criteria for panic dis- taper of medication over weeks 8–16.97 At week 8, patients order) were randomized to 8 weeks of treatment with either were taking a mean of 4.8 mg/day alprazolam. Although Volume 8 – Number 8 (Suppl 1) CNS Spectrums - August 2003 Long-Term Treatment of Panic Disorder
alprazolam was associated with significantly greater improve- long-term (≥6 months) efficacy trials (Table), generally with ments than placebo in assessments of phobic fear and avoid- small numbers of patients. These studies have included ance, work/social disability, and global improvement (P<.05) examination of second-year maintenance and discontinua- at week 8, after withdrawal of treatment, the alprazolam tion of imipramine,119 and comparisons of imipramine with effect disappeared on every measure. During taper and treat- alprazolam26,61,120-122 and CBT.64,123 ment-free follow-up at 6 months, therapeutic gains after Long-term treatment of panic disorder with clomipramine alprazolam were lost, while gains after exposure were main- was investigated in the Galway Study of Panic Disorder, in tained. The findings of this and other studies led investiga- which 57 patients initially randomized to a 6-week, placebo- tors to conclude that relapse is a problem regardless of when controlled trial of clomipramine versus lofepramine were fol- alprazolam (or any other BZD) is stopped.97 lowed for 6 months of open-label treatment.66 Patients Several short-term (≤9 weeks), placebo-controlled studies originally randomized to placebo were alternately reassigned of clonazepam (at doses of 0.5–4 mg/day) have been to either lofepramine or clomipramine for this phase of the reported.98-101 These studies consistently showed clinically and trial. Data collected at weeks 8, 12, and 24 indicated that statistically significant superiority of clonazepam over placebo patients initially assigned to clomipramine and lofepramine in panic disorder as determined by various measures. As with continued to improve and that the course of improvement alprazolam, the gradual tapering of clonazepam (in decrements was similar. Patients initially assigned to placebo and then of 0.25–0.5 mg/day every 3 days) was associated with some randomized to either medication continued to show clinical worsening, particularly in number of panic attacks, but improvement on major efficacy measures and no significant patients did not revert to their baseline condition.99 differences across medications were reported.
Results of a small (n=72), randomized, placebo-con- In a 3-year study, Lotufo-Neto and colleagues124 also inves- trolled, head-to-head, 6-week comparison of clonazepam and tigated recurrence and relapse rates following remission with alprazolam showed comparable efficacy.102 Both clonazepam clomipramine. Of the 81 patients treated for panic disorder and alprazolam were superior to placebo for the treatment of with or without agoraphobia, 70% achieved full remission.
panic disorder, as reflected by changes in panic attack fre- Eighty-one percent of this sample had medication tapered quency, phobic distress, social and work disability, and global off. From this group, 37% experienced relapse immediately, assessments of severity of illness and improvement. No sig- 43% experienced relapse over the course of 3 years, and nificant differences between the two agents were observed.
19% continued to be symptom-free.
In a number of short-term, comparative studies, other Selective Serotonin Reuptake Inhibitors
BZDs have shown significant antipanic effects. Diazepam The long-term efficacy of fluvoxamine in panic disorder (mean dose=44 mg/day),103,104 adinazolam (mean dose=95.5 was investigated in a naturalistic follow-up study of patients mg/day), and lorazepam (mean dose 6 to 7.5 mg/day)105-107 all originally enrolled to a randomized 4-arm trial (fluvoxam- had similar overall efficacy to alprazolam and may provide ine+exposure, placebo+exposure, CBT+exposure, and effective treatment alternatives, although they are not cur- exposure alone).125 Of the 76 patients who completed treat- rently approved in the US for this indication.
ment in the original trial,89 71 (93%) were evaluated 2 years Monoamine Oxidase Inhibitors
later. The treatment effect in the fluvoxamine+exposure MAOIs have not been systematically studied in panic dis- group was maintained, but was no longer superior due to fur- order,108 although at least one study predating the current ther improvements in the other groups. Most patients diagnostic nomenclature and likely including a group of (77%) had received additional treatment during the follow- panic patients is consistent with a therapeutic effect.
up. Forty-four percent of patients who had received fluvox- Reversible Inhibitors of Monoamine Oxidase A
amine in the original study were still using it, although A few randomized, controlled trials have investigated the generally at a much lower dose (≤50 mg/day), and >50% of efficacy of RIMAs in panic disorder and report variable placebo-treated patients had received treatment with flu- results.109-113 RIMAs are distinguished from the older MAOIs voxamine at some point.
by their selectivity and reversibility. The most widely studied Paroxetine has been studied in both relapse prevention and RIMAs are moclobemide and brofaromine. Brofaromine is long-term maintenance. Patients with panic disorder who had not available in the US and moclobemide is not approved received fixed-dose paroxetine (10, 20, or 40 mg/day) in a for use in the US, although it is available in Canada and 22-week, double-blind, placebo-controlled trial were immedi- other countries.
ately rerandomized to continue paroxetine or switch to placebo for an additional 12 weeks.114 Relapse was defined as Long-Term Treatment Trials
the return of panic attacks to a frequency equal to or greater Several studies were conducted on the long-terms phar- than baseline, or an increase in the CGI-Severity scale of macotherapy treatment for panic disorder, particularly with ≥2 points. Relapse rates were significantly higher in the group TCAs, SSRIs, and BZDs (Table).114-118 switched to placebo (11 of 37 patients) than in the group remaining on paroxetine (2 of 43 patients).
Although not approved in the US for the treatment of A 36-week paroxetine maintenance study115 included panic disorder, imipramine has been studied in a number of 176 patients with DSM-III-R panic disorder who had Volume 8 – Number 8 (Suppl 1) CNS Spectrums - August 2003 M.H. Pollack, C. Allgulander, B. Bandelow, et al.
completed a 12-week, double-blind, randomized study of CGI-Improvement and CGI-Severity scores. At the end of paroxetine, clomipramine, and placebo.68 The primary end- the 28-week, double-blind phase, sertraline-treated point was the final timepoint at which ≥70% of patients patients were significantly less likely to discontinue treat- remained in the trial, which was at the eighth 3-week period ment due to relapse or insufficient response (12%) than (corresponding to week 24). At this time, the mean reduc- placebo-treated patients (24%).116 tion relative to baseline in panic attack frequency was signif- Sertraline's efficacy in panic disorder has also been icantly greater for paroxetine than for placebo, but did not shown in a study that examined clinical response and sex- differ significantly from clomipramine. The proportion of ual functioning in a pooled analysis of panic-disorder patients free from panic attacks at week 24 did not differ sig- patients.126 Twelve-month data of completers were pooled nificantly among the three groups.
from three studies with a 10-week placebo-controlled Long-term treatment with sertraline in panic disorder phase, followed by 52 weeks of open-label sertraline treat- has been demonstrated to be effective and well tolerated for ment. Sertraline-treated patients showed a marked up to 80 weeks in a large multicenter US study.116 Patients improvement in symptoms of panic disorder. They had a with DSM-III-R–defined panic disorder, who had completed 98% reduction in frequency of panic attacks, and 93% one of three 10-week randomized, double-blind, placebo- achieved remission (CGI-Improvement score of 1 or 2) or controlled studies, were entered into a 52-week, open-label, no full panic attacks after 12 months of treatment. Sexual continuation phase with sertraline (25 mg/day for 1 week, functioning was affected in 22% of patients during the 50 mg/day for week 2, and then optional titration up to acute phase, and improved over time.
200 mg/day for the rest of the continuation phase). Of the Arato and colleagues127 have recently presented results of 398 patients who entered the open continuation phase, a multicenter, double-blind, randomized comparison of ser- 183 (46%) completed it and were rerandomized to an traline (50–100 mg) versus imipramine (100–200 mg) over additional 28 weeks of double-blind, placebo-controlled 26 weeks in 138 patients with concomitant panic disorder treatment.116 Continued improvement was shown on all and major depression. Both treatments were associated with efficacy measures during open sertraline treatment, with a significant reduction in panic attacks and symptoms of mean reductions of 44% in number of panic attacks, 46% depression, which resulted in improved quality of life. There in number of limited-symptom attacks, and 40% in were significantly fewer discontinuations due to adverse amount of time spent worrying. In addition, there was a events in sertraline- versus imipramine-treated patients, and 46% reduction in the PDSS and a 29% reduction in both sertraline was better tolerated overall.
TABLE. LONG-TERM TREATMENT TRIALS OF PHARMACOTHERAPY FOR PANIC DISORDER114-118
Fahy et al (1992)66 Placebo (RCT phase only) Lydiard et al (1998)114 Lecrubier et al (1997)115 Rapaport et al (2001)116 RCT ➞ OL ➞ RCT Michelson et al (1999)117 Lepola et al (1998)118 * Patients rerandomized to active treatment or placebo.
RCT=randomized controlled trial; OL=open-label; wks=weeks; CGI=Clinical Global Impressions; HAM-A=Hamilton Rating Scale for Anxiety; HAM-D=Hamilton Rating Scale for Depression; MADRS=Montgomery-Asberg Depression Rating Scale; CGI-S=Clinical Global Impressions-Severity; Volume 8 – Number 8 (Suppl 1) CNS Spectrums - August 2003 Long-Term Treatment of Panic Disorder
Saiz Ruiz and colleagues128 gave sertraline to patients with Citalopram has also shown efficacy in long-term treat- panic disorder in the clinical setting. After 6 months of ment of panic disorder. Of 475 patients who completed treatment, 89% of the 886 outpatients were free of panic 8 weeks of treatment with either placebo, citalopram attacks; discontinuation due to adverse events was observed (10–15, 20–30, and 40–60 mg/day), or clomipramine in only 3.6% of patients.
(60–90 mg/day),70 279 entered a double-blind continuation In a long-term trial of fluoxetine, 88 patients who had phase, and 179 completed the full 12 months.118,129 All active responded to acute fluoxetine in a 10-week, placebo-con- medications were superior to placebo in preventing relapse trolled trial (CGI-Improvement score of 1 or 2) were subse- in patients with panic disorder.118 Response (defined as no quently randomized to continued fluoxetine or placebo for panic attacks in the week before assessment) was signifi- 24 weeks.117 Patients who continued fluoxetine experienced cantly higher in the citalopram 20–30 mg and 40–60 mg improvement in panic-attack frequency and phobia rating groups than in the placebo group (P=.001 and P=.003, scale score over the extension phase, whereas those switched respectively). The lowest citalopram dose and clomipramine to placebo experienced statistically significant worsening in also demonstrated modest advantages over placebo HAM-A, HAM-D, and Symptom Checklist-90-Revised (P<.05).130 At all dosages, citalopram was more effective than [SCL-90-R] rating scores. Overall relapse rates were low in placebo in controlling phobic symptoms (assessed using the both groups.
Phobia Scale and SCL-90-R phobia-related factors); Roy-Byrne and colleagues129 retrospectively examined a 20–30 mg was generally the most effective dosage.
medical and pharmacy claims database to analyze use of Alleviation of phobic symptoms tended to continue to emergency room and laboratory resources and costs for increase toward the end of treatment.131 120 patients with panic disorder who underwent SSRI treatment. The mean number of emergency room and lab- In addition to the long-term alprazolam/imipramine com- oratory visits and costs during the 6 months after therapy parisons noted above, a long-term comparison of alprazolam initiation were reduced compared with the 6 months prior with clonazepam has also been published.132 Patients origi- to starting treatment. Sertraline reduced emergency room nally randomized to treatment in a placebo-controlled trial visits by 79.5% and costs by 85.2% (P<.05), while fluoxe- comparing these two agents were reevaluated in a follow-up tine reduced visits by 25.0% and costs by 69.5% (P=not study 1.5 years later, when 78% of patients remained on significant [NS]), and paroxetine reduced visits by 8.6% medication (the majority on BZDs). For those patients on and costs by 30.8% (P=NS).
BZDs, there was a significant worsening of global outcome Main Efficacy Variables 30% discontinuation 10% discontinuation 10% OL; 3% 28 wks RCT discontinuation ND=no data; PAF=panic attack frequency; PDSS=Panic Disorder Severity Scale; SCL-90=Symptom Checklist-90-Revised; ND=no data.
Pollack MH, Allgulander C, Bandelow B, et al. CNS Spectrums. Vol 8, No 8 (suppl 1). 2003.
Volume 8 – Number 8 (Suppl 1) CNS Spectrums - August 2003 M.H. Pollack, C. Allgulander, B. Bandelow, et al.
between endpoint and follow-up (P<.05), although the final 63 patients. Many patients sought further treatment for clinical status remained significantly better than the pre- panic during follow-up because of a less-than-adequate treatment score (P<.001). There was no significant differ- response to treatment; nevertheless, additional treatment did ence across groups in the number of patients free from panic not result in further improvement.
at follow-up.
Monoamine Oxidase Inhibitors
No controlled long-term trials of MAOIs in the treatment Regular aerobic exercise (eg, running) has also been eval- of panic disorder have been published.
uated as a nonpharmacologic approach to managing panic Reversible Inhibitors of Monoamine Oxidase A
disorder in a 10-week, placebo-controlled comparison with Long-term efficacy data for RIMAs in panic disorder clomipramine. Exercise was less effective than are limited.109,113 clomipramine, but more effective than placebo, in most pri- mary and secondary outcome measures of efficacy, including the P&A scale, Fear Questionnaire, and CGI scores.56,71 Exposure-based psychotherapy has been compared with medication in at least two controlled trials.89,97 In a placebo- The combination of medication and CBT, a treatment controlled comparison with alprazolam, 156 patients were approach that is reviewed in some detail by Gelder,133 may randomized to 8 weeks of alprazolam+exposure therapy, alpra- prove more effective than either treatment alone in the zolam+relaxation (a psychologic placebo), placebo+exposure management of panic disorder.134 Results from a limited num- therapy, or placebo+relaxation (double placebo), followed by ber of studies support this approach, although more research taper off medication over weeks 8–16.97 All 4 treatment is needed before definitive conclusions can be made about groups improved on panic measures, but on nonpanic mea- the relative benefits of combined treatment.
sures, exposure had twice the effect of alprazolam at the end The addition of brief dynamic psychotherapy (15 weekly of treatment. During taper and treatment-free follow-up at 6 sessions) to treatment with clomipramine significantly months, therapeutic gains after exposure were maintained, reduced the subsequent relapse rate of panic disorder com- while gains after alprazolam were lost.
pared with clomipramine alone (20% versus 75%, respec- A randomized, 3-month, 4-arm trial (exposure+fluvox- tively) during long-term follow-up.135 Earlier trials have amine, exposure+placebo, exposure+CBT, and exposure provided similar evidence for the efficacy of imipramine alone) with 96 patients showed that all four treatments were treatment combined with exposure therapy. A trial con- effective and resulted in a significant decrease in phobic ducted in 62 chronically agoraphobic patients tested clinical avoidance.89 Effectiveness of exposure therapy alone did not measures of global severity, phobia, panic, anxiety, depres- differ significantly from that of exposure+CBT or expo- sion, and behavioral performance before treatment and at sure+placebo, but was inferior to exposure+fluvoxamine.
weeks 4, 8, and 12 of treatment. The combination of imipramine and exposure therapy was as effective as imipramine monotherapy, and more effective than exposure Perhaps the most common nonpharmacologic therapy alone.136 Another early trial in 76 agoraphobic women com- used in the management of panic disorder is CBT, and sev- pared combined imipramine+group exposure in vivo treat- eral studies have confirmed its effectiveness.64,87,88,123 CBT is ment with combined placebo+group exposure in vivo designed to help the patient understand the role of his or her treatment.137 The imipramine+exposure group demonstrated cognitions in the development of panic and to accept a more significantly greater improvements than the placebo+expo- benign interpretation of the bodily sensations.87 The ele- sure group in primary phobia, spontaneous panic, and global ments involved in CBT are diverse and can involve anxiety management skills, cognitive restructuring, and progressive Paroxetine has also been studied in combination with exposure to panic attack triggers. Homework assignments CBT. In a 12-week, randomized, double-blind trial by form an important part of this approach.
Oehrberg and colleagues,78 discussed previously, 120 patients There are numerous studies demonstrating the efficacy of received CBT with either paroxetine or placebo.
CBT. In one of the largest placebo-controlled trials, Barlow Significantly greater improvements were seen with paroxe- and colleagues64 found an equivalent response rate between tine+CBT than with placebo+CBT.
medication and CBT, with a small advantage for imipramine More recently, a 10-week, placebo-controlled pilot by the end of the acute phase (12 weeks). However, at fol- study evaluated the effects of combining paroxetine low-up 6 months after treatment discontinuation, patients (10–50 mg/day) with a very brief form of CBT in 33 patients who had received CBT alone maintained their improvement with DSM-IV panic disorder.80 Patients in both groups significantly better (4% relapse) than those given (ie, paroxetine+CBT and placebo+CBT) improved similarly imipramine (25% relapse), based on PDSS responder crite- and substantially on most measures during the 10 weeks of ria.64 An earlier, smaller study by Brown and Barlow132 exam- acute treatment. However, the proportion of panic-free ined long-term outcome (24-month follow-up) of CBT in patients was significantly higher in the paroxetine-treated Volume 8 – Number 8 (Suppl 1) CNS Spectrums - August 2003 Long-Term Treatment of Panic Disorder
group than in the placebo group (80% versus 25%; P<.007), vide clinicians with a better understanding of management as was the proportion of patients who rated themselves as issues in panic disorder and to give clinical recommenda- very much improved at week 10 (60% versus 13%; P<.017).
tions for treatment.76 The statement is based on a series of According to a recent randomized, placebo-controlled six review articles, each of which focuses on a key area in trial, combining imipramine with CBT also appears to be the clinical management of panic disorder, and on the rel- an effective approach to panic disorder, both acutely and evant scientific literature. The six areas covered in the in the long term.64 This five-arm trial compared CBT only (n=77), imipramine only (n=83), CBT+imipramine (1) Response, remission, and relapse4; (n=65), placebo only (n=24), and CBT+placebo (n=63) (2) Impact of comorbidity on treatment approach138; in patients with panic disorder. During the 12-week acute (3) Differential efficacy of drug treatments93; treatment phase, combined imipramine and CBT treat- (4) Tolerability and safety of drug treatments139; ment resulted in limited benefit over monotherapy. By the (5) Clinical and preclinical mechanisms of drug end of the 6-month maintenance phase, the combined treatment was superior to CBT alone, CBT+placebo, and (6) Long-term treatment.77 imipramine alone on PDSS measures.
The consensus statement recommends SSRIs as the first-choice drug type. It also notes that full remission of OVERVIEW OF AVAILABLE
panic disorder will require a minimum of 9–12 months of treatment and that continued treatment for up to 1 year is Four sets of guidelines for the management of patients effective in maintaining and improving acute response and with panic disorder have been published. The most compre- preventing relapse. In terms of duration, the consensus hensive are those from the American Psychiatric statement recommends that treatment should continue for Association (APA) published in 1998.134 These detailed at least 12–24 months and should be discontinued only if guidelines aim to provide psychiatrists with recommenda- the patient is not currently experiencing a stressful life tions on the overall care of patients with panic disorder, and event and full remission is maintained. Treatment should were developed under the auspices of the Steering be continued in patients with persistent symptoms or a Committee on Practice Guidelines.
history of severe relapse.
The guidelines note that both CBT and medication Roy-Byrne and colleagues140 developed guidelines for have been shown to be effective for panic disorder and also the family physician to improve the recognition and treat- point out that there is no convincing evidence that one ment of panic disorder in primary care. These guidelines modality is superior for all patients or for particular patient briefly discuss diagnosis and the rationale for treatment subpopulations. They recommend that the choice of treat- of panic disorder and provide guidance on appropriate ment should be based on individualized assessment of effi- choices regarding treatment type, dosing, titration, side- cacy, benefit and risks of each modality, costs, and the effect management, maintenance therapy, and referral. patient's personal preferences.
A treatment algorithm assumes an 8-week period for ini- The APA Guidelines also consider optimal length of tial treatment and recommends that patients continue treatment, and note that the acute phase of treatment with pharmacotherapy for at least 1 year. In general, discontin- either CBT or medication generally lasts about 12 weeks. At uation should only be considered if there are no signs this point, the patient should have markedly fewer and less of symptomatology, serious medical illness, or major intense panic attacks than before treatment (ideally no panic attacks), should worry less about panic attacks, and should The Ontario Program for Optimized Therapeutics offers experience minimal or no phobic avoidance. With CBT, the guidelines for the primary care management of anxiety dis- guidelines suggest that the frequency of visits is generally orders in Canada.141 They recommend CBT as first-line decreased after the acute phase of treatment and eventually treatment for anxiety disorders if patient compliance is not discontinued within several months.
an issue. For pharmacotherapy of anxiety disorder with or After successful acute treatment with medication, the without agoraphobia, citalopram, fluoxetine, fluvoxamine, patient should continue to receive drug therapy for a mini- paroxetine, sertraline, and venlafaxine are all recom- mum of 12–18 months. Discontinuation should be mended as first-line treatments. The guidelines suggest ini- attempted only if the patient experiences significant or full tiating therapy at a low dose, with gradual upward titration improvement. Patients who partially or fully relapse follow- until complete symptom remission is achieved. After ing drug discontinuation should resume medication immedi- 12 weeks of treatment, if response is apparent, pharma- ately. They could benefit from prolonged treatment, cotherapy should be maintained for at least 1 year; then if although no specific length of time is given. Longer periods the patient maintains a full remission (no panic attacks), of initial treatment with medication may decrease the risk of the clinician can consider stopping therapy gradually over relapse when medication is stopped.
2–6 months. As relapse is common following discontinua- The International Consensus Group on Depression and tion of medication, relapsing patients should begin taking Anxiety published a consensus statement in 1998 to pro- medication again or be treated with CBT.
Volume 8 – Number 8 (Suppl 1) CNS Spectrums - August 2003 M.H. Pollack, C. Allgulander, B. Bandelow, et al.
attacks in an asthmatic population. Behav Res Ther. 1994;32:411-418.
20. Eaton WW, Kessler RC, Wittchen HU, et al. Panic and panic disorder SSRIs are currently the drugs of first choice in the treat- in the United States. Am J Psychiatry. 1994;151:413-420.
ment of panic disorder with or without agoraphobia, and 21. Lecrubier Y, Ustun TB. Panic and depression: a worldwide primary care paroxetine, sertraline, citalopram, clomipramine, alprazolam, perspective. Int Clin Psychopharmacol. 1998;13(suppl 4):S7-S11.
fluoxetine, and clonazepam are approved treatments for 22. Johnson J, Weissman MM, Klerman GL. Panic disorder, comorbidity, panic disorder in Europe and/or the US. Most randomized, and suicide attempts. Arch Gen Psychiatry. 1990;47:805-808.
23. Klerman GL, Weissman MM, Ouellette R, et al. Panic attacks in the controlled treatment trials of panic disorder are limited to community. Social morbidity and health care utilization. JAMA. ≤12 weeks of treatment. While the number of long-term studies is limited and often confined to naturalistic follow- 24. Kessler RC, McGonagle KA, Zhao S, et al. Lifetime and 12-month up, currently available data suggest that the efficacy of stud- prevalence of DSM-III-R psychiatric disorders in the United States.
ied drugs is maintained with continued treatment.
Results from the National Comorbidity Survey. Arch Gen Psychiatry. Long-term data are available for all the medications that are 25. Wittchen HU, Essau CA, von Zerssen D, et al. Lifetime and six-month approved for the treatment of panic disorder, except for the prevalence of mental disorders in the Munich Follow-Up Study. Eur BZDs, which have only naturalistic long-term follow-up Arch Psychiatry Clin Neurosci. 1992;241:247-258.
data. Available evidence suggests that long-term pharmaco- 26. Katschnig H, Amering M, Stolk JM, et al. Long-term follow-up after a logic treatment of panic disorder is safe and effective in drug trial for panic disorder. Br J Psychiatry. 1995;167:487-494.
accruing continued improvement, maintaining benefit, and 27. Keller MB, Yonkers KA, Warshaw MG, et al. Remission and relapse in subjects with panic disorder and panic with agoraphobia: a prospective preventing relapse. In addition, CBT, alone or in combina- short-interval naturalistic follow-up. J Nerv Ment Dis. 1994;182:290-296.
tion with drug therapy, is also effective for the treatment of 28. Stein MB, Tancer ME, Uhde TW. Major depression in patients with panic disorder. CNS
panic disorder: factors associated with course and recurrence. J Affect 29. Segui J, Marquez M, Garcia L, et al. Differential clinical features of early-onset panic disorder. J Affect Disord. 1999;54:109-117.
Diagnostic and Statistical Manual of Mental Disorders. 3rd ed. Washington, DC: American Psychiatric Association; 1980.
30. Yonkers KA, Zlotnick C, Allsworth J, et al. Is the course of panic disor- 2. Pollack MH, Marzol PC. Panic: course, complications and treatment of der the same in women and men? Am J Psychiatry. 1998;155:596-602.
panic disorder. J Psychopharmacol. 2000;14:S25-S30.
31. Keller MB, Baker LA. The clinical course of panic disorder and depres- 3. Regier DA, Narrow WE, Rae DS. The epidemiology of anxiety disorders: sion. J Clin Psychiatry. 1992;53(suppl):5-8.
the Epidemiologic Catchment Area (ECA) experience. J Psychiatr Res. 32. Warshaw MG, Massion AO, Shea MT, et al. Predictors of remission in patients with panic with and without agoraphobia: prospective 5-year 4. Shear MK, Clark D, Feske U. The road to recovery in panic disorder: follow-up data. J Nerv Ment Dis. 1997;185:517-519.
response, remission, and relapse. J Clin Psychiatry. 1998;59(suppl 8):4-8.
33. Markowitz JS, Weissman MM, Ouellette R, et al. Quality of life in panic 5. Diagnostic and Statistical Manual of Mental Disorders. 4th ed. Washington, disorder. Arch Gen Psychiatry. 1989;46:984-992.
DC: American Psychiatric Association; 1994.
34. Ettigi P, Meyerhoff AS, Chirban JT, et al. The quality of life and 6. Carr RE. Panic disorder and asthma. J Asthma. 1999;36:143-152.
employment in panic disorder. J Nerv Ment Dis. 1997;185:368-372.
7. Gorman JM, Kent JM, Sullivan GM, et al. Neuroanatomical hypothesis 35. Swinson RP, Cox BJ, Woszczyna CB. Use of medical services and treat- of panic disorder, revised. Am J Psychiatry. 2000;157:493-505.
ment for panic disorder with agoraphobia and for social phobia. CMAJ. 8. Fava GA, Mangelli L. Subclinical symptoms of panic disorder: new 1992;147: 878-883 insights into pathophysiology and treatment. Psychother Psychosom. 36. Weissman MM, Klerman GL, Markowitz JS, et al. Suicidal ideation and suicide attempts in panic disorder and attacks. N Engl J Med. 9. Katschnig H, Amering M. The long-term course of panic disorder and its predictors. J Clin Psychopharmacol. 1998;18:6S-11S.
37. Cox BJ, Direnfeld DM, Swinson RP, et al. Suicidal ideation and suicide 10. Weissman MM, Bland RC, Canino GJ, et al. The cross-national epi- attempts in panic disorder and social phobia. Am J Psychiatry. demiology of panic disorder. Arch Gen Psychiatry. 1997;54:305-309.
11. Cross-National Collaborative Panic Study SPI. Drug treatment of panic 38. Henriksson MM, Isometsa ET, Kuoppasalmi KI, et al. Panic disorder in disorder. Comparative efficacy of alprazolam, imipramine, and placebo.
completed suicide. J Clin Psychiatry. 1996;57:275-281.
Br J Psychiatry. 1992;160:191-202.
39. Lepine JP, Chignon JM, Teherani M. Suicide attempts in patients with 12. Katon WJ, Von Korff M, Lin E. Panic disorder: relationship to high panic disorder. Arch Gen Psychiatry. 1993;50:144-149.
medical utilization. Am J Med. 1992;92:7S-11S.
40. Cowley DS, Flick SN, Roy-Byrne PP. Long-term course and outcome in 13. Kennedy BL, Schwab JJ. Utilization of medical specialists by anxiety panic disorder: a naturalistic follow-up study. Anxiety. 1996;2:13-21.
disorder patients. Psychosomatics. 1997;38:109-112.
14. Fleet R, Lavoie K, Beitman BD. Is panic disorder associated with coro- 41. Ball SG, Otto MW, Pollack MH, et al. Predicting prospective episodes nary artery disease? A critical review of the literature. J Psychosom Res. of depression in patients with panic disorder: a longitudinal study. J Consult Clin Psychol. 1994;62:359-365.
15. Mukerji V, Beitman BD, Alpert MA. Chest pain and angiographically 42. McElroy SL, Altshuler LL, Suppes T, et al. Axis I psychiatric comorbid- normal coronary arteries. Implications for treatment. Tex Heart Inst J. ity and its relationship to historical illness variables in 288 patients with bipolar disorder. Am J Psychiatry. 2001;158:420-426.
16. Coryell W, Noyes R, Clancy J. Excess mortality in panic disorder. A 43. Stein MB, Tancer ME, Gelernter CS, et al. Major depression in patients comparison with primary unipolar depression. Arch Gen Psychiatry. with social phobia. Am J Psychiatry. 1990;147:637-639.
44. Aronson TA, Craig TJ. Cocaine precipitation of panic disorder. Am J 17. Weissman MM, Markowitz JS, Ouellette R, et al. Panic disorder and cardiovascular/cerebrovascular problems: results from a community sur- 45. Nutt DJ. Antidepressants in panic disorder: clinical and preclinical vey. Am J Psychiatry. 1990;147:1504-1508.
mechanisms. J Clin Psychiatry. 1998;59(suppl 8):24-28.
18. Davies SJ, Ghahramani P, Jackson PR, et al. Association of panic disor- 46. Bradwejn J, Koszycki D, Shriqui C. Enhanced sensitivity to cholecys- der and panic attacks with hypertension. Am J Med. 1999;107:310-316.
tokinin tetrapeptide in panic disorder. Clinical and behavioral findings.
19. Carr RE, Lehrer PM, Rausch LL, et al. Anxiety sensitivity and panic Arch Gen Psychiatry. 1991;48:603-610.
Volume 8 – Number 8 (Suppl 1) CNS Spectrums - August 2003 Long-Term Treatment of Panic Disorder
47. Klein E, Zohar J, Geraci MF, et al. Anxiogenic effects of m-CPP in 72. Modigh K, Westberg P, Eriksson E. Superiority of clomipramine over patients with panic disorder: comparison to caffeine's anxiogenic imipramine in the treatment of panic disorder: a placebo-controlled effects. Biol Psychiatry. 1991;30:973-984.
trial. J Clin Psychopharmacol. 1992;12:251-261.
48. den Boer JA, Westenberg HG. Effect of a serotonin and noradrenaline 73. Diagnostic and Statistical Manual of Mental Disorders. 3rd ed rev.
uptake inhibitor in panic disorder; a double-blind comparative study with Washington, DC: American Psychiatric Association; 1987.
fluvoxamine and maprotiline. Int Clin Psychopharmacol. 1988;3:59-74.
74. Cassano GB, Petracca A, Perugi G, et al. Clomipramine for panic disor- 49. Norman TR, Gregory MS, Judd FK, et al. Platelet serotonin uptake in der: I. The first 10 weeks of a long-term comparison with imipramine. panic disorder: comparison with normal controls and the effect of treat- J Affect Disord. 1988;14:123-127.
ment. Aust N Z J Psychiatry. 1988;22:390-395.
75. Sheehan DV. Current concepts in the treatment of panic disorder. J Clin 50. Charney DS, Heninger GR, Jatlow PI. Increased anxiogenic effects of Psychiatry. 1999;60(suppl 18):16-21.
caffeine in panic disorders. Arch Gen Psychiatry. 1985;42:233-243.
76. Ballenger JC, Davidson JR, Lecrubier Y, et al. Consensus statement on 51. Guy W. ECDEU Assessment Manual for Psychopharmacology. US Dept panic disorder from the International Consensus Group on Depression Health, Education and Welfare publication (ADM) 76–338. Rockville, and Anxiety. J Clin Psychiatry. 1998;59(suppl 8):47-54.
Md: National Institute of Mental Health; 1976:218-222.
77. Davidson JR. The long-term treatment of panic disorder. J Clin 52. Hamilton M. Diagnosis and rating of anxiety. Br J Psychiatry. 1969;3:76-79.
Psychiatry. 1998;59(suppl 8):17-21.
53. Hamilton M. A rating scale for depression. J Neurol Neurosurg 78. Ballenger JC, Wheadon DE, Steiner M, et al. Double-blind, fixed-dose, placebo-controlled study of paroxetine in the treatment of panic disor- 54. Sheehan DV. The Anxiety Disease. New York, NY: Bantam Books.1983.
der. Am J Psychiatry. 1998;155:36-42.
55. Argyle N, Deltito J, Allerup P, et al. The Panic-Associated Symptom 79. Oehrberg S, Christiansen PE, Behnke K, et al. Paroxetine in the treat- Scale: measuring the severity of panic disorder. Acta Psychiatr Scand. ment of panic disorder. A randomised, double-blind, placebo-controlled study. Br J Psychiatry. 1995;167:374-379.
56. Bandelow B, Broocks A, Pekrun G, et al. The use of the Panic and 80. Diagnostic and Statistical Manual of Mental Disorders. 4th ed.
Agoraphobia Scale (P & A) in a controlled clinical trial.
Washington, DC: American Psychiatric Association; 1994.
81. Stein MB, Ron NG, Walker JR, et al. Do selective serotonin re-uptake 57. Klein DF, Fink M. Psychiatric reaction patterns to imipramine. Am J inhibitors enhance the efficacy of very brief cognitive behavioral thera- py for panic disorder? A pilot study. Psychiatry Res. 2000;94:191-200.
58. Bakish D, Hooper CL, Filteau MJ, et al. A double-blind placebo-con- 82. Londborg PD, Wolkow R, Smith WT, et al. Sertraline in the treatment trolled trial comparing fluvoxamine and imipramine in the treatment of of panic disorder. A multi-site, double-blind, placebo-controlled, fixed- panic disorder with or without agoraphobia. Psychopharmacol Bull. dose investigation. Br J Psychiatry. 1998;173:54-60.
83. Pohl RB, Wolkow RM, Clary CM. Sertraline in the treatment of panic 59. Rizley R, Kahn RJ, McNair DM, et al. A comparison of alprazolam and disorder: a double-blind multicenter trial. Am J Psychiatry. imipramine in the treatment of agoraphobia and panic disorder.
60. Taylor CB, Hayward C, King R, et al. Cardiovascular and symptomatic 84. Pollack MH, Otto MW, Worthington JJ, et al. Sertraline in the treat- reduction effects of alprazolam and imipramine in patients with panic ment of panic disorder: a flexible-dose multicenter trial. Arch Gen disorder: results of a double-blind, placebo-controlled trial. J Clin 85. Pollack MH, Rapaport MH, Clary CM, Mardekian J, Wolkow R.
61. Schweizer E, Rickels K, Weiss S, et al. Maintenance drug treatment of Sertraline treatment of panic disorder: response in patients at risk for panic disorder. I. Results of a prospective, placebo-controlled compari- poor outcome. J Clin Psychiatry. 2000;61:922-927.
son of alprazolam and imipramine. Arch Gen Psychiatry. 1993;50:51-60.
86. Hoehn-Saric R, McLeod DR, Hipsley PA. Effect of fluvoxamine on 62. Andersch S, Rosenberg NK, Kullingsjo H, et al. Efficacy and safety of alpra- panic disorder. J Clin Psychopharmacol. 1993;13:321-326.
zolam, imipramine and placebo in treating panic disorder. A Scandinavian 87. Black DW, Wesner R, Bowers W, et al. A comparison of fluvoxamine, multicenter study. Acta Psychiatr Scand Suppl. 1991;365:18-27.
cognitive therapy, and placebo in the treatment of panic disorder. Arch 63. Leon CA, De Arango MV, Arevalo W, et al. Comparison of the effect of Gen Psychiatry. 1993;50:44-50.
alprazolam, imipramine and placebo in the treatment of panic disorders 88. Sharp DM, Power KG, Simpson RJ, et al. Global measures of outcome in Cali, Colombia. Acta Psiquiatr Psicol Am Lat. 1990;36:59-72.
in a controlled comparison of pharmacological and psychological treat- 64. Barlow DH, Gorman JM, Shear MK, et al. Cognitive-behavioral thera- ment of panic disorder and agoraphobia in primary care. Br J Gen Pract. py, imipramine, or their combination for panic disorder: a randomized controlled trial. JAMA. 2000;283:2529-2536.
89. de Beurs E, van Balkom AJ, Lange A, et al. Treatment of panic disorder 65. Johnston DG, Troyer IE, Whitsett SF. Clomipramine treatment of ago- with agoraphobia: comparison of fluvoxamine, placebo, and psychologi- raphobic women. An eight-week controlled trial. Arch Gen Psychiatry. cal panic management combined with exposure and of exposure in vivo alone. Am J Psychiatry. 1995;152:683-691.
66. Fahy TJ, O'Rourke D, Brophy J, et al. The Galway Study of Panic 90. Michelson D, Lydiard RB, Pollack MH, et al. Outcome assessment and Disorder. I: clomipramine and lofepramine in DSM III-R panic disorder: clinical improvement in panic disorder: evidence from a randomized a placebo controlled trial. J Affect Disord. 1992;2563-2575.
controlled trial of fluoxetine and placebo. The Fluoxetine Panic 67. Hoffart A, Due-Madsen J, Lande B, et al. Clomipramine in the treat- Disorder Study Group. Am J Psychiatry. 1998;155:1570-1577.
ment of agoraphobic inpatients resistant to behavioral therapy. J Clin 91. Hirschmann S, Dannon PN, Iancu I, et al. Pindolol augmentation in patients with treatment-resistant panic disorder: a double-blind, place- 68. Lecrubier Y, Bakker A, Dunbar G, et al. A comparison of paroxetine, bo-controlled trial. J Clin Psychopharmacol. 2000;20:556-559 clomipramine and placebo in the treatment of panic disorder.
92. Rickels K, Schweizer E, Weiss S, et al. Maintenance drug treatment for Collaborative Paroxetine Panic Study Investigators. Acta Psychiatr panic disorder. II. Short- and long-term outcome after drug taper. Arch Gen Psychiatry. 1993;50:61-68.
69. Bakker A, van Dyck R, Spinhoven P, et al. Paroxetine, clomipramine, 93. den Boer JA. Pharmacotherapy of panic disorder: differential efficacy and cognitive therapy in the treatment of panic disorder. J Clin from a clinical viewpoint. J Clin Psychiatry. 1998;59(suppl 8):30-36.
94. Ballenger JC, Burrows GD, DuPont RL, et al. Alprazolam in panic dis- 70. Wade AG, Lepola U, Koponen HJ, et al. The effect of citalopram in order and agoraphobia: results from a multicenter trial. I. Efficacy in panic disorder. Br J Psychiatry. 1997;170:549-553.
short-term treatment. Arch Gen Psychiatry. 1988;45:413-422.
71. Broocks A, Bandelow B, Pekrun G, et al. Comparison of aerobic exer- 95. Pecknold JC, Swinson RP, Kuch K, et al. Alprazolam in panic disorder cise, clomipramine, and placebo in the treatment of panic disorder. Am and agoraphobia: results from a multicenter trial. III. Discontinuation J Psychiatry. 1998;155:603-609.
effects. Arch Gen Psychiatry. 1988;45:429-436.
Volume 8 – Number 8 (Suppl 1) CNS Spectrums - August 2003 M.H. Pollack, C. Allgulander, B. Bandelow, et al.
96. Lydiard RB, Lesser IM, Ballenger JC, et al. A fixed-dose study of alpra- 119. Mavissakalian MR, Perel JM. 2nd year maintenance and discontinua- zolam 2 mg, alprazolam 6 mg, and placebo in panic disorder. J Clin tion of imipramine in panic disorder with agoraphobia. Ann Clin 97. Marks IM, Swinson RP, Basoglu M, et al. Alprazolam and exposure 120. Lepola UM, Rimon RH, Riekkinen PJ. Three-year follow-up of alone and combined in panic disorder with agoraphobia. A controlled patients with panic disorder after short-term treatment with alprazo- study in London and Toronto. Br J Psychiatry. 1993;162:776-787.
lam and imipramine. Int Clin Psychopharmacol. 1993;8:115-118.
98. Beauclair L, Fontaine R, Annable L, et al. Clonazepam in the treatment 121. Lepola U, Koponen H, Leinonen E. A naturalistic 6-year follow-up of panic disorder: a double-blind, placebo-controlled trial investigating study of patients with panic disorder. Acta Psychiatr Scand. the correlation between clonazepam concentrations in plasma and clin- ical response. J Clin Psychopharmacol. 1994;14:111-118.
122. Curtis GC, Massana J, Udina C, et al. Maintenance drug therapy of 99. Rosenbaum JF, Moroz G, Bowden CL. Clonazepam in the treatment of panic disorder. J Psychiatr Res. 1993;27(suppl 1):127-142.
panic disorder with or without agoraphobia: a dose-response study of 123. Clark DM, Salkovskis PM, Hackmann A, et al. A comparison of cog- efficacy, safety, and discontinuance. Clonazepam Panic Disorder Dose- nitive therapy, applied relaxation and imipramine in the treatment of Response Study Group. J Clin Psychopharmacol. 1997;17:390-400.
panic disorder. Br J Psychiatry. 1994;164:759-769.
100. Moroz G, Rosenbaum JF. Efficacy, safety, and gradual discontinuation 124. Lotufo-Neto F, Bernik M, Ramos RT, et al. A dose-finding and contin- of clonazepam in panic disorder: a placebo-controlled, multicenter uation study of clomipramine in panic disorder. J Psychopharmacology. study using optimized dosages. J Clin Psychiatry. 1999;60:604-612.
101. Valenca AM, Nardi AE, Nascimento I, et al. Double-blind clonazepam 125. de Beurs E, van Balkom AJ, van Dyck R, et al. Long-term outcome of vs placebo in panic disorder treatment. Arq Neuropsiquiatr. pharmacological and psychological treatment for panic disorder with agoraphobia: a 2-year naturalistic follow-up. Acta Psychiatr Scand. 102. Tesar GE, Rosenbaum JF, Pollack MH, et al. Double-blind, placebo- controlled comparison of clonazepam and alprazolam for panic disor- 126. Pohl R, Holland P, Chung H, Clary C. Effectiveness of sertraline in der. J Clin Psychiatry. 1991;52:69-76.
long-term treatment of panic disorder: treatment response, quality of 103. Noyes R, Burrows GD, Reich JH, et al. Diazepam versus alprazolam for life, and sexual functioning. Poster presented at: the American the treatment of panic disorder. J Clin Psychiatry. 1996;57:349-355.
Psychiatric Association Annual Meeting, 2001; New Orleans, LA. 104. Dunner DL, Ishiki D, Avery DH, et al. Effect of alprazolam and 127. Arato M, Lepola U, Austin C. Sertraline vs imipramine treatment of comor- diazepam on anxiety and panic attacks in panic disorder: a controlled bid panic disorder and major depression. Poster presented at: the American study. J Clin Psychiatry. 1986;47:458-460.
Psychiatric Association annual meeting,; May 2001; New Orleans, LA.
105. Charney DS, Woods SW. Benzodiazepine treatment of panic disorder: 128. Saiz Ruiz J, Gomez Beneyto M, Gutierrez Casares JR, et al. Sertraline a comparison of alprazolam and lorazepam. J Clin Psychiatry. in the treatment of panic disorder: a naturalistic study. Poster present- ed at: the American Psychiatric Association Annual Meeting; May 106. Schweizer E, Fox I, Case G, et al. Lorazepam vs. alprazolam in the treatment of panic disorder. 2001; New Orleans, LA. 107. Schweizer E, Pohl R, Balon R, et al. Lorazepam vs. alprazolam in the 129. Roy-Byrne PP, Clary CM, Miceli RJ, Colucci SV, Xu Y, Grudzinski treatment of panic disorder. Pharmacopsychiatry. 1990;23:90-93.
AN. The effect of selective serotonin reuptake inhibitor treatment of 108. Sheehan DV, Ballenger J, Jacobsen G. Treatment of endogenous anxi- panic disorder on emergency room and laboratory resource utilization.
ety with phobic, hysterical, and hypochondriacal symptoms. Arch Gen J Clin Psychiatry. 2001;62:678-682.
130. Leinonen E, Lepola U, Koponen H, et al. Citalopram controls phobic 109. van Vliet IM, Westenberg HG, den Boer JA. MAO inhibitors in symptoms in patients with panic disorder: randomized controlled trial.
panic disorder: clinical effects of treatment with brofaromine. A dou- J Psychiatry Neurosci. 2000;25:25-32.
ble blind placebo controlled study. Psychopharmacology (Berl).
131. Pollack MH, Otto MW, Tesar GE, et al. Long-term outcome after acute treatment with alprazolam or clonazepam for panic disorder. 110. van Vliet IM, den Boer JA, Westenberg HG, et al. A double-blind J Clin Psychopharmacol. 1993;13:257-263.
comparative study of brofaromine and fluvoxamine in outpatients 132. Brown TA, Barlow DH. Long-term outcome in cognitive-behavioral with panic disorder. J Clin Psychopharmacol. 1996;16:299-306.
treatment of panic disorder: clinical predictors and alternative strate- 111. Bakish D, Saxena BM, Bowen R, et al. Reversible monoamine oxidase-A gies for assessment. J Consult Clin Psychol. 1995;63:754-765.
inhibitors in panic disorder. Clin Neuropharmacol. 1993;16(suppl 2):S77-S82.
133. Gelder MG. Combined pharmacotherapy and cognitive behavior 112. Loerch B, Graf-Morgenstern M, Hautzinger M, et al. Randomised therapy in the treatment of panic disorder. J Clin Psychopharmacol. placebo-controlled trial of moclobemide, cognitive-behavioural ther- apy and their combination in panic disorder with agoraphobia. Br J 134. American Psychiatric Association. Practice guidelines for the treat- ment of patients with panic disorder. Am J Psychiatry. 1998;155:1-34.
113. Tiller JW, Bouwer C, Behnke K. Moclobemide and fluoxetine for 135. Wiborg IM, Dahl AA. Does brief dynamic psychotherapy reduce the panic disorder. International Panic Disorder Study Group. Eur Arch relapse rate of panic disorder? Arch Gen Psychiatry. 1996;53:689-694.
Psychiatry Clin Neurosci. 1999;249(suppl 1):S7-S10.
136. Mavissakalian M, Michelson L. Agoraphobia: relative and combined 114. Lydiard RB, Steiner M, Burnham D, et al. Efficacy studies of paroxe- effectiveness of therapist-assisted in vivo exposure and imipramine. tine in panic disorder. Psychopharmacol Bull. 1998;34:175-182.
J Clin Psychiatry. 1986;47:117-122.
115. Lecrubier Y, Judge R. Long-term evaluation of paroxetine, 137. Zitrin CM, Klein DF, Woerner MG. Treatment of agoraphobia with clomipramine and placebo in panic disorder. Collaborative group exposure in vivo and imipramine. Arch Gen Psychiatry. Paroxetine Panic Study Investigators. Acta Psychiatr Scand. 138. Lecrubier Y. The impact of comorbidity on the treatment of panic dis- 116. Rapaport M, Walkow R, Rubin A, et al. Sertraline treatment of panic order. J Clin Psychiatry. 1998;59(suppl 8):11-14.
disorder: results of a long-term study. Acta Psychiatr Scand. 139. Baldwin DS, Birtwistle J. The side effect burden associated with drug treatment of panic disorder. J Clin Psychiatry. 1998;5(suppl 8):39-44.
117. Michelson D, Pollack M, Lydiard RB, et al. Continuing treatment of 140. Roy-Byrne P, Stein M, Bystrisky A, et al. Pharmacotherapy of panic panic disorder after acute response: randomised, placebo-controlled disorder: proposed guidelines for the family physician. J Am Board Fam trial with fluoxetine. The Fluoxetine Panic Disorder Study Group. Br J Psychiatry. 1999;174:213-218.
141. Anxiety Review Panel. Evans M, Bradwejn J, Dunn L, eds. Guidelines 118. Lepola UM, Wade AG, Leinonen EV, et al. A controlled, prospective, for the treatment of anxiety disorders in primary care. Toronto, Canada: 1-year trial of citalopram in the treatment of panic disorder. J Clin Queen's Printer of Ontario; 2000. Available at: www.opot.org Accessed November 10, 2001.
Volume 8 – Number 8 (Suppl 1) CNS Spectrums - August 2003 WCA Recommendations for the Long-Term
Treatment of Posttraumatic Stress Disorder
By Dan J. Stein, MD, PhD, Borwin Bandelow, MD, Eric Hollander, MD, David J. Nutt, MD, FRCP, FRCPsych, FMedSci, Ahmed Okasha, MD, PhD, FRCP, FRCPsych, FACP, Mark H. Pollack, MD, Richard P. Swinson, MD, FRCPC, FRCPsych, DPM, and Joseph Zohar, MD in the 17th century, Samuel Pepys recorded his flashbacks of • Posttraumatic stress disorder (PTSD) is a commonly the great fire of London. However, it was not until after the unrecognized, chronic condition associated with signif- Vietnam War that PTSD received widespread recognition as icant disability, increased societal costs, and reduced an independent disorder, and only in 1980 was PTSD included in the official nomenclature.1 • Psychotherapy and pharmacotherapy, both separately PTSD represents a pathological response after trauma.
and in combination, are useful in the treatment of PTSD. The disorder was originally thought to be primarily associ- • While the selective serotonin reuptake inhibitors ser- ated with war and combat, but studies conducted during traline, paroxetine, and fluoxetine have shown efficacy the past several decades have revealed that PTSD can also in acute treatment trials, sertraline is the only one for arise after experiencing or witnessing a variety of severe which long-term efficacy has been demonstrated.
traumatic events that involve actual or threatened death • Current guidelines recommend that pharmacotherapy or serious injury to self or others. Such events include be continued in chronic PTSD for a minimum of interpersonal violence (eg, physical/sexual abuse, physical 12–24 months and that maintenance cognitive-behav- assault, kidnapping, torture, military combat, terrorist ioral therapy sessions be provided as needed. attacks), man-made (eg, motor vehicle accidents), or nat- ural (eg, fire) accidents or disasters.
Posttraumatic stress disorder (PTSD) is a common and dis- EPIDEMIOLOGY OF POSTTRAUMATIC
abling condition. In addition to combat-related PTSD, the disorder occurs in civilians exposed to severe traumatic events, with the com- PTSD is one of the most common anxiety disorders, munity prevalence rate for the combined populations reaching as affecting 8% to 12% of the general population at some time high as 12%. If left untreated, PTSD may continue for years after during their lives.2,3 While estimates for lifetime prevalence the stressor event, resulting in severe functional and emotional of exposure to a traumatic event vary widely—from 39% to impairment and a dramatic reduction in quality of life, with negative 89%—PTSD develops in only 10% to 20% of people economic consequences for both the sufferer and society as a whole. exposed to trauma.2-4 The likelihood of developing PTSD Although PTSD is often overlooked, diagnosis is relatively straight- following exposure to a traumatic event depends on a num- forward once a triggering stressor event and the triad of persistent ber of factors, including: symptoms—reexperiencing the traumatic event, avoiding stimuli •Type and severity of trauma. Childhood abuse, rape, associated with the trauma, and hyperarousal—have been identi- assaultive violence, and sudden unexpected death of a fied. However, comorbid conditions of anxiety and depression fre- loved one are associated with high rates of PTSD in the quently hamper accurate diagnosis. Treatment for PTSD includes community.3 The National Comorbidity Survey (NCS) psychotherapy and pharmacotherapy. The latter includes selective of the United States general population found that rape serotonin reuptake inhibitors (SSRIs), tricyclic antidepressants, and was associated with the highest likelihood of developing monoamine oxidase inhibitors. Only SSRIs have been proven effec- PTSD among both men and women.2 Other traumas tive and safe in long-term randomized controlled trials. Current that the NCS associated with a high probability of guidelines from the Expert Consensus Panel for PTSD recommend developing PTSD included combat exposure, childhood treatment of chronic PTSD for a minimum of 12–24 months. abuse, sexual molestation, physical attack, and being CNS Spectr 2003;8(suppl 1):31-39 threatened with a weapon.
•Gender. PTSD appears to be twice as common in women as in men,2,3,5 even when controlling for the type of trauma The nonmedical literature has long documented symp- exposure.3 PTSD also lasts longer in women (median dura- toms of posttraumatic stress disorder (PTSD). For example, tion, 48.1 months versus 12.0 months in men)3 These recommendations are based on proceedings from the World Council of Anxiety meeting held September 11, 2000, in Pisa, Italy, and on guidelines and articles published in the medical literature through October 15, 2001.
Please direct all correspondence to: Dan J. Stein, MD, PhD, University of Stellenbosch, PO Box 19063, Tygerberg 7505, South Africa. Tel: 27-21-938- 9228; Fax: 27-21-933-5790; E-mail: [email protected].
Volume 8 – Number 8 (Suppl 1) CNS Spectrums - August 2003 D.J. Stein, B. Bandelow, E. Hollander, et al
•Factors surrounding the trauma. Factors preceding trauma provides a useful basis for future research on interventions to (eg, pre-existing depression or anxiety, family psychiatric enhance fear extinction and decrease symptoms.
history, early separation from parents4), factors during In addition to these structural and functional brain the trauma (eg, dissociation), and factors operating after changes, neuroendocrinological alterations have been the trauma (eg, lack of social support and additional life associated with PTSD8 and it has been suggested that the stress6), have an effect on development of PTSD. disorder is characterized by decreased serum cortisol due to increased hypothalamic-pituitary-adrenal (HPA) axis neg- ative feedback.8,15 This line of argument suggests that the Symptoms of PTSD are grouped into three clusters nature of HPA dysfunction seen in PTSD differs signifi- according to Diagnostic and Statistical Manual of Mental cantly from that seen in depression, supporting the speci- Disorders, Fourth Edition (DSM-IV) criteria7: ficity of the neurobiology of PTSD. There has been (1) Re-experiencing. This may involve persistent re- interest in the use of cortisol release factor antagonists for experiencing of the traumatic event, with flash- mood and anxiety disorders, and future work in this area backs, intrusive thoughts, recurrent distressing may ultimately lead to novel interventions for the treat- recollections and dreams; having the feeling that ment of PTSD.
the traumatic event is recurring; and psychological Possible evidence of serotonergic dysregulation in distress and physiological reactivity when exposed PTSD includes frequent symptoms of aggression, impulsiv- to reminders.
ity, depression, suicidality, and the demonstrated clinical (2) Avoidance/numbing. Sufferers seek to avoid stimuli asso- efficacy of selective serotonin reuptake inhibitors ciated with the traumatic event including thoughts, (SSRIs).16 Possible evidence of adrenergic and noradrener- conversations, people, and places. There may be gic dysregulation includes exaggerated startle response, inability to recall aspects of the traumatic event, exaggerated increases in heart rate and blood pressure on markedly reduced interest in participating in once-typ- exposure to reminders of the traumatic event, and elevated ical activities, feelings of detachment or emotional 24-hour urine catecholamine excretion.17 Psychotropic numbing, or a feeling of reduced life expectancy.
medications with prominent actions on noradrenaline that (3) Hyperarousal. Persistent symptoms of hyperarousal are have been used for the treatment of PTSD include cloni- among the first symptoms experienced in PTSD.
dine, propranolol, prazosin, tricyclic antidepressants These may include sleep disturbance, hypervigilance, (TCAs), and monoamine oxidase inhibitors (MAOIs).
exaggerated startle response, irritability or outbursts of The dopamine system may also play an important role in anger, and concentration difficulties.
PTSD, and the role of new-generation antipsychotics in PTSD is receiving increased attention.
UNDERLYING BIOLOGICAL ETIOLOGY
There is growing evidence that PTSD is not merely a nor- CLINICAL COURSE, CHRONICITY, AND
mal response to an abnormal event, but is a medical disorder mediated by psychobiological dysfunction occurring after Once PTSD develops, it is typically chronic (lasting trauma. Specific neuroanatomical, biochemical, and psycho- >3 months) and recurrent. In the NCS, 25% of untreated logical abnormalities have been found in patients with patients recovered over the course of 12 months and 40% PTSD, and this work has provided a rationale for using par- recovered by 24 months; however, 50% still met criteria for ticular interventions in the treatment of this disorder.
PTSD 6 years after the event.2 Although there has been some variability in findings across If left untreated, PTSD can result in: studies, a useful basis has been laid from which future work •Significant functional and emotional impairment •Dramatic reduction in the patient's quality of life Neuroimaging techniques, for example, have shown sev- •Predisposition to other psychiatric and physical illnesses eral important changes in the brains of PTSD victims com- •Significantly increased likelihood of suicide attempts, pared with controls. These include possible reductions in hospitalization, and alcohol dependence or abuse2,18-20 hippocampal volume, and functional changes in the amyg- •Higher healthcare utilization and costs dala, hippocampus, anterior paralimbic regions, and Broca's Patients with PTSD exhibit varying degrees of func- area.12-15 Whether the structural hippocampal changes tional impairment.21 Nevertheless, assessments of quality observed in PTSD patients predispose them to PTSD or if of life and functioning have shown that, in general, PTSD they are a direct effect of the disorder is still under investiga- significantly affects an individual's well-being and quality tion. However, the clinical relevance of this work is sup- of life, particularly physical health and functioning, and ported by data showing that reduced hippocampal volume is role functioning at home, work, or school.22 PTSD is more associated with severity of the traumatic exposure, symp- strongly associated with suicidal behavior than are most toms, and cognitive problems. Similarly, the finding that other anxiety disorders. Among individuals with PTSD, neuronal circuits involved in fear conditioning overlap with the reported rate of attempted suicide is 19%, which is those highlighted in functional imaging studies of PTSD comparable with that seen in depression.23 Volume 8 – Number 8 (Suppl 1) CNS Spectrums - August 2003 Long-Term Treatment of Posttraumatic Stress Disorder
With regard to the significant cost burden to both the •Comorbid major depression—increased likelihood is individual and society in terms of lost earnings and associ- 7-fold for men and 4-fold for women.
ated healthcare costs, general population research in the •Generalized anxiety disorder—increased likelihood is US estimates that 38% of people with PTSD are in treat- 6-fold for men and 3-fold for women.
ment in a given year, which is comparable to rates among •Panic disorder—increased likelihood is 4-fold for men people with major depression (36%) but higher than those and 3-fold for women.
seen in other anxiety disorders (23%) or substance use dis- In addition, individuals with PTSD are 8 (women) and 14 (men) times more likely to have three or more psychiatric In 1998, the cost of anxiety disorders in the US was esti- disorders than those without PTSD.2 mated at $63 billion. PTSD and panic disorder were associ- ated with the highest rates of service use.25 Furthermore, PTSD is associated with an estimated 3.6 days of work The goals of PTSD treatment are to reduce core symp- impairment per month, translating into an annual produc- toms, comorbidity, and disability; improve quality of life; and tivity loss in the US in excess of $3 billion.24 prevent recurrent episodes.
Current treatment approaches involve psychotherapy, pharmacotherapy, or a combination of both. Successful treat- Currently, there are two sets of diagnostic criteria used to ment for PTSD should demonstrate effects across the spec- define PTSD: (1) the World Health Organization trum of PTSD symptom clusters. A number of rating International Classification of Diseases, Tenth Revision (ICD- instruments are used to assess the severity of PTSD symp- 10) criteria26 and (2) the American Psychiatric Association toms and determine the efficacy of treatments in clinical DSM-IV criteria.7 While not identical, the ICD-10 and trials, including: DSM-IV criteria are generally in agreement. For example, •Clinician-Administered PTSD Scale, Part 2 (CAPS-2)28 ICD-10 concurs with DSM-IV27 that the essential feature of •Clinical Global Impression (CGI) scales: CGI–Severity PTSD is the development of characteristic symptoms follow- (CGI-S) and Improvement (CGI-I)29 ing exposure to an extreme traumatic stressor, with the sub- •Impact of Event Scale–Revised (IES)30 ject's response involving intense fear, helplessness, or horror.
•Davidson Trauma Scale (DTS)31 Both sets of diagnostic criteria identify exposure to a trau- •Treatment Outcome PTSD Scale (TOP-8)32 matic event followed by the PTSD symptoms of re-experi- •Mississippi Scale for Combat-Related PTSD encing, avoidance, and hyperarousal.
DSM-IV criteria specify that the duration of illness must exceed 1 month before a diagnosis can be given. In addition, diagnosis requires that the symptoms cause clinically signifi- SSRIs have been recommended as the first-line medica- cant distress or impairment in social, occupational, or other tions of choice for the treatment of PTSD.35,36 Sertraline, important areas of functioning. Further, DSM-IV defines paroxetine, and fluoxetine have demonstrated effectiveness PTSD as either chronic or acute, depending on whether and tolerability in randomized, controlled acute treatment tri- symptoms last for >3 months or <3 months, respectively.
als. Sertraline, however, is the only treatment for which long- Despite their similarities, the ICD-10 criteria for PTSD term efficacy (up to 1 year) has been demonstrated. Other are less rigid than those of DSM-IV in the following ways26: pharmacologic agents, including TCAs and MAOIs, have •Symptoms are not required to cause clinically signifi- been studied in trials of PTSD patients. Early studies of TCAs cant distress.
suggested their efficacy in the treatment of PTSD in combat •Symptoms indicating numbing of responsiveness are veterans, but these studies were small and did not use clini- not emphasized.
cian-rated PTSD scales.37,38 MAOIs may be effective, but side •Only one symptom of avoidance needs to be present effects, potentially dangerous interactions, and the need for compared to a minimum of three avoidance symptoms dietary restrictions may limit their use in clinical practice.39 in DSM-IV. In some studies of combat-related PTSD, medication has •There is no differentiation between acute and chronic not proven effective.40 The following fundamental differ- ences between combated-related and civilian PTSD may lead to differences in clinical trial data41: •Differential nature of the traumatic stress in combat ver- PTSD is associated with an increased risk of comorbid sus civilian settings.
anxiety disorders, depression, and substance abuse. The risk •Differences in the clinical presentation of PTSD of developing comorbid disorders appears to be related to the (eg, markedly greater baseline severity in veteran study severity of trauma and consequent complexity of the PTSD participants compared with civilian subjects).
reaction.24 Population studies have shown that >80% of peo- •Increased longitudinal comorbidity of combat-related ple with PTSD also have a history of at least one other psy- PTSD (eg, increased occurrence of alcoholism, sub- chiatric disorder, including24: stance abuse, medical comorbidity).
Volume 8 – Number 8 (Suppl 1) CNS Spectrums - August 2003 D.J. Stein, B. Bandelow, E. Hollander, et al
•Gender-specific neurobiological factors that may render controlled trials are needed to confirm these results within SSRIs more effective in women than in men.
the military veteran population.
•Psychosocial context of the illness in the military vet- The short-term efficacy of fluoxetine has been demon- eran setting (eg, ongoing monetary support for combat- strated in three randomized controlled trials (RCTs) of related PTSD diagnosis).
5 and 12 weeks' duration. In a preliminary 5-week study of 64 randomized subjects (men and women; military veter- Acute Treatment Trials
ans and civilians), 47 subjects completed the trial.40 Selective Serotonin Reuptake Inhibitors and
Among completers, CAPS-2 score was significantly Serotonin Norepinephrine Reuptake Inhibitors
reduced from baseline for fluoxetine-treated patients Sertraline was the first drug approved by the US Food and (average dose, 40 mg/day) versus placebo. Improvements Drug Administration for the treatment of PTSD, and was were predominantly seen in the numbing and hyperarousal subsequently approved in many other countries. Sertraline symptom clusters. Civilian trauma patients demonstrated has demonstrated acute efficacy in the treatment of PTSD in significant improvement versus placebo, whereas those two large multicenter trials, as well as in smaller studies in with combat-related trauma did not.
specific PTSD patient populations.41-45 A further 12-week study compared fluoxetine with Davidson and colleagues42 randomized 208 patients placebo in 53 civilians with PTSD.46 A statistically signifi- (predominantly civilians) to receive either sertraline cant number of patients in the fluoxetine group compared (50–200 mg/day) or placebo for 12 weeks. All patients had a with the placebo group (85% versus 62%, respectively, baseline total severity score ≥50 on CAPS-2. The primary P<.06) were classified as responders at week 12 weeks based efficacy measures were total scores on CAPS-2, IES, and on the Duke Global Rating (DUKE) criterion of 1 or 2 CGI. Sertraline was superior in significantly reducing total (much or very much improved). Using the more stringent CAPS-2 (P=.003), and IES (P=.02) scores, and improving criterion of a DUKE score of 1 (very much improved), CGI-S, CGI-I (P<.001), and DTS (a secondary outcome response rates were lower (59% for fluoxetine and 19% for parameter; P=.002) scores. An intent-to-treat (ITT) end- point analysis demonstrated a 60% responder rate for sertra- A recent double-blind, placebo-controlled study of fluox- line versus 38% for placebo (P=.004). Sertraline was well etine for the treatment of PTSD was conducted mainly in tolerated, and reported adverse events were consistent in fre- areas affected by war.47 Patients were randomized to 12 weeks quency and type with the previously established safety profile.
of treatment with either fluoxetine (n=226; 20–80 mg/day) In a second trial, Brady and colleagues43 randomized or placebo (n=75). Compared with placebo, fluoxetine was 187 patients (mostly civilians) to receive either sertraline associated with a significantly greater reduction in baseline (50–200 mg/day) or placebo for 12 weeks. As in the TOP-8 score (P=.006), as well as significantly higher Davidson study,42 all patients had a baseline total severity response rates (P=.02) and significantly greater improve- score ≥50 on CAPS-2; the primary efficacy measures were ment on most secondary measures, including CAPS-2 total CAPS-2, IES, and CGI scores. The researchers found (P=.021), the Hamilton Rating Scale for Anxiety (HAM- that sertraline was superior to placebo in significantly reduc- A) (P=.001), and the Montgomery-Asberg Depression ing total CAPS-2 (P<.05) starting from week 2, and improv- Rating Scale (P<.001). Fluoxetine treatment significantly ing CGI-I, CGI-S (P≤.02) and DTS (P=.003) scores.
reduced PTSD symptoms compared to placebo. There were Sertraline was also associated with significantly greater no significant differences in the number of patients reporting improvements (P=.004) in quality-of-life scores compared adverse events between groups.
with placebo. In addition, sertraline treatment was associ- Paroxetine is approved in the US and several other ated with a similar magnitude of improvement in the avoid- countries for the treatment of PTSD. A recent fixed-dose, ance/numbing, arousal, and re-experiencing/intrusion placebo-controlled trial examined the efficacy and safety clusters, with improvements in the first two being signifi- of paroxetine in the treatment of 551 patients with cantly greater with sertraline than with placebo. Sertraline chronic PTSD (DSM-IV and CAPS-2 score ≥50).48 was well tolerated, with insomnia being the only adverse Patients were randomly assigned to placebo (n=186), event reported significantly more often with sertraline than paroxetine 20 mg/day (n=183), or paroxetine 40 mg/day (n=182) for 12 weeks. Paroxetine-treated patients in both In a 10-week, double-blind, pilot study, 42 Israeli mili- dosage groups showed significantly greater improvements tary veterans were randomized to receive either a flexible compared with placebo-treated patients in both CAPS-2 dose of sertraline (50–200 mg/day) or placebo.41 Responder total score and rate of response for global improvement on rates for completers were 53% for sertraline and 20% for the CGI scale. Paroxetine treatment also led to significant placebo based on CGI-I criteria (P=.057), and 41% for improvement over placebo on all three PTSD symptom sertraline and 20% for placebo based on combined CGI-I clusters (re-experiencing, avoidance/numbing, hyper- and CAPS-2 reduction criteria (P=.28). Sertraline treat- arousal), social and occupational impairment, and comor- ment was well tolerated, with a 13% discontinuation rate bid depression. Both doses of paroxetine were well due to adverse events. Additional adequately powered tolerated in this study.48 Volume 8 – Number 8 (Suppl 1) CNS Spectrums - August 2003 Long-Term Treatment of Posttraumatic Stress Disorder
Ruggiero and colleagues49 conducted a flexible-dose trial Other Agents
of paroxetine for the treatment of chronic PTSD. In this Buspirone, a 5-HT1A partial agonist, has shown limited double-blind, placebo-controlled study, 307 subjects were benefit in treating hyperarousal symptoms of PTSD in an randomly assigned to receive paroxetine (20–50 mg/day) or open-label study. Several benzodiazepines have been stud- placebo for 12 weeks. Primary efficacy variables included ied in small-scale, short-term clinical trials. However, the change from baseline to endpoint in CAPS-2 total score and lack of demonstrated efficacy and difficulties associated the proportion of responders on CGI-I. Other outcome mea- with discontinuation of treatment limit the use of these sures included change from baseline in CAPS-2; total scores drugs in PTSD.54-56 on TOP-8, DTS, and the Sheehan Disability Scale; and the Mood stabilizers may be particularly effective as single or proportion of patients achieving response and remission.
adjunctive medication for the explosive behavior common PTSD symptoms were significantly more reduced on both in PTSD.57 The anticonvulsants carbamazepine and sodium primary and secondary outcome measures in the paroxetine valproate have shown positive results in small, open-label group (n=151) than in the placebo group (n=156). CAPS-2 trials in combat veterans. Another anticonvulsant, lamotrig- total score was significantly more improved in the paroxetine ine (up to 500 mg/day), has been examined in a small group compared with the placebo group from week 8 (P<.05) (N=15), 12-week, double-blind, placebo-controlled pilot on, and significantly more paroxetine-treated patients study of civilians with PTSD.58 The response rate in patients achieved response (P<.001) by week 12. Treatment was well receiving lamotrigine was twice as high compared with tolerated, with the frequency and type of adverse events cor- patients receiving placebo (50% versus 25%, respectively), responding to the known safety profile of paroxetine.
as assessed by the DUKE. Lamotrigine demonstrated greater To date, only open-label studies of citalopram, fluvoxam- efficacy for the re-experiencing and avoidance/numbing ine, nefazodone, and venlafaxine have been reported.
symptom clusters. Further large, double-blind RCTs are required to confirm the usefulness of lamotrigine as a primary Early studies of TCAs suggested efficacy in the treatment or adjunctive treatment for patients with PTSD.
of PTSD in combat veterans.37,38 The effectiveness of TCAs, however, may be offset by the significant incidence of Long-Term Treatment Trials
adverse events, risk for overdose, and poor compliance rates.
Very few trials have investigated whether long-term An 8-week study of amitriptyline versus placebo in war vet- pharmacologic treatment of PTSD is efficacious and safe.
erans (N=46) found amitriptyline (mean dose=169 mg/day) Only sertraline has been studied for up to 1 year, in both superior to placebo in completers (n=33) based on IES, continuation and relapse-prevention design studies CGI, Hamilton Rating Scale for Depression, and HAM-A (Table59,60). In the continuation trial, 249 outpatients with scores.37 There are no published studies showing efficacy in PTSD who completed one of two 12-week, randomized, trials >12 weeks.
double-blind studies of sertraline and placebo were eligible Monoamine Oxidase Inhibitors
to enter a 24-week open-label trial of sertraline (50–200 Phenelzine has demonstrated efficacy in the treatment of mg/day).59 Significant improvement with sertraline was PTSD in a small 8-week study comparing the effects of observed both in patients who had received sertraline and phenelzine (n=19) with the TCA imipramine (n=23) and in those who had received placebo in the 12-week feeder placebo (n=18).50 PTSD symptoms, as assessed by improve- studies. At endpoint, patients in the 24-week open-label ment in IES score from baseline, were significantly reduced study showed significant and substantial improvement in the phenelzine (44%) and imipramine (25%) treatment (P<.05) from baseline on CAPS-2, CGI-S, CGI-I, and groups, while there was some reduction with placebo (5%).
IES. The data indicated that 92% of patients who The intrusion (but not avoidance) subscale of the IES responded to acute-phase treatment with sertraline sus- showed significant improvement; initial mild to moderate tained their initial response. Furthermore, 54% of patients depressive symptoms did not significantly improve. At end- who failed to respond to acute treatment with sertraline point, CGI-I showed similar improvement with both were converted to responders in the continuation phase.63 imipramine and phenelzine (65% and 68%, respectively), Sertraline was well tolerated, with discontinuation due to while 28% of the placebo group was rated as improved.
adverse events observed in only 8.6% of patients.
Two RCTs have studied the effects of the selective In a relapse-prevention design study, 96 outpatients reversible MAOI brofaromine on PTSD. A 12-week study, who had completed and responded to 24 weeks of open- comprised mostly of combat veterans, did not demonstrate label continuation treatment with sertraline59 were ran- any significant differences between brofaromine and placebo domized to receive sertraline (50–200 mg/day) or placebo on PTSD-specific measures (CAPS-2),51 although it did for an additional 28 weeks.60,61 Kaplan-Meier analyses were show greater efficacy on the CGI-I.52 In a 14-week European used to estimate time to relapse, rates of relapse or discon- study, the drug effect was more robust on the civilian popula- tinuation due to clinical deterioration, and acute exacer- tion CGI-I.57 However, the manufacturer terminated further bations. Patients who continued sertraline treatment for a clinical development of brofaromine based on the results of further 28 weeks maintained their improvements. The depression trials, and it is no longer available.52 proportion of patients who relapsed on sertraline (5%) was Volume 8 – Number 8 (Suppl 1) CNS Spectrums - August 2003 D.J. Stein, B. Bandelow, E. Hollander, et al
significantly lower (P=.02) than the proportion of patients long-term supportive therapies are recommended for pro- who relapsed on placebo (26%): patients receiving longed, complex, and intractable PTSD.23 placebo were 6.4 times as likely to experience a relapse as patients receiving sertraline. Patients receiving placebo also relapsed or discontinued due to clinical worsening sig- Exposure therapy is designed to help the patient confront nificantly earlier than patients receiving sertraline. Mean certain situations, people, objects, memories, or emotions changes in secondary endpoints on CAPS-2, CGI-I, CGI- that have become associated with the stressor and now S, and IES were significantly different between the two evoke an unrealistically intense fear. This can be done either groups. Sertraline's ability to sustain improvements was by repeated emotional recounting of the traumatic experi- comparable across all three core PTSD symptom clusters.
ences until they no longer provoke high stress, or by con- Sertraline was well tolerated, with no treatment-emergent, frontations with the actual triggers that are now safe but are treatment-related adverse events observed at a rate ≥10%.
being actively avoided by the patient.
Changes in quality of life and psychosocial functioning Several studies have investigated the effects of exposure across 64 weeks (12-week double-blind acute phase fol- therapy in veterans from the Vietnam War with PTSD.63 lowed by a 24-week open-label continuation phase and a These studies indicated that exposure therapy was effective 28-week double-blind relapse-prevention phase) of sertra- in reducing PTSD symptoms in this population, but thera- line treatment (50–200 mg/day) were also examined.62 peutic effects were modest.64 In a study by Foa and Patients were assessed using the Quality of Life Enjoyment colleagues65 comparing the effect of exposure therapy with and Satisfaction Questionnaire (Q-LES-Q), and the stress inoculation therapy in female assault victims, exposure Medical Outcomes Study 36-Item Short Form. At the end therapy, stress inoculation therapy, and a combination of the of 12 weeks, 58% of sertraline responders had achieved a two all reduced the severity of PTSD and depression to a Q-LES-Q total score within 10% of community norms.
similar extent while a wait-list control group saw no Continuation treatment for 24 weeks led to an additional improvement. These effects were maintained at 6- and 20% improvement in quality of life and functional mea- 12-month follow-up. In the more rigorous ITT analysis, sures. Overall, sertraline treatment resulted in sustained however, exposure therapy was superior to the other treat- and progressive improvement in quality of life and func- ments in improving posttreatment anxiety and global social tional measures during >12 months of treatment.
adjustment, and had larger effect sizes on PTSD severity, depression, and anxiety.
Psychotherapy is recommended as a first-line treatment for mild and sometimes moderate PTSD and in combina- Anxiety management (or stress inoculation training) tion with pharmacotherapy in more severe cases.35 teaches a set of skills that help the patient cope with stress, Cognitive-behavioral therapy (CBT), including exposure including relaxation training, breath training, positive therapy, anxiety management, and cognitive therapy, thinking, assertiveness training, and negative-thought focuses on the traumatic event and is effective in the man- stopping. As described in the preceding section on expo- agement of PTSD.23 CBT is generally short term, averag- sure therapy, Foa and colleagues65 found that anxiety man- ing 8–12 sessions held once or twice a week.36 To alleviate agement was as effective as exposure therapy at reducing such symptom subsets as anger and interpersonal problems, the severity of the symptoms of PTSD, although results in TABLE. LONG-TERM TREATMENT TRIALS OF PHARMACOTHERAPY FOR PTSD
Londborg et al (2001)59 12-wk RCT acute phase 24-wk OL continuation phase Davidson et al (2000)60 12-wk RCT acute phase 24-wk OL continuation phase 28-wk RCT relapse-prevention PTSD= posttraumatic stress disorder; RCT=randomized controlled trial; OL=open-label; wk=week; CAPS-2=the Clinician-Administered PTSD Scale, Part 2; CGI=Clinical Global Impression–Improvement; CGI-S=Clinical Global Impression–Severity; IES=Impact of Event Scale–Revised.
Volume 8 – Number 8 (Suppl 1) CNS Spectrums - August 2003 Long-Term Treatment of Posttraumatic Stress Disorder
the intent-to-treat sample indicated that exposure therapy In terms of pharmacotherapy, SSRIs are the first-line choice was more effective.
regardless of the type of symptom that is most prominent. If the patient responds only partially, treatment may be aug- mented with the addition of another medication or combined Cognitive therapy aims to help modify unrealistic with psychotherapy. The most highly recommended adjunc- assumptions, beliefs, and automatic thoughts that lead to dis- tive medication is a mood stabilizer. For patients not respond- turbing emotions. The goal of cognitive therapy is to teach ing to the maximum tolerated dose of the initial treatment, patients to identify their own particular illogical and unreal- switching to a different SSRI or other medication, or different istic thought processes and adopt more realistic responses psychotherapy, is recommended. Regarding treatment dura- that will generate more balanced emotions.
tion, continuing medication for at least 12 months is recom- In one study of victims of mixed traumas with PTSD, mended before dose tapering can be considered. In chronic 10 weekly sessions of prolonged exposure were compared PTSD with residual symptoms, treatment for at least 24 with cognitive restructuring, a combination of the two months is recommended before considering treatment with- therapies, and relaxation control.66 Exposure therapy and cognitive restructuring singly or combined improved symptoms of PTSD markedly, but the two therapies were no more effective when combined. At 6-month follow-up, PTSD is a commonly unrecognized, chronic condition the group that received exposure therapy alone seemed to that has a profound effect on the quality of life of the indi- maintain its gains to a greater extent than the group that vidual sufferer. In addition, PTSD is associated with signifi- received cognitive therapy alone.
cant disability, increased healthcare utilization and costs, and reduced productivity. Psychotherapy and pharma- OVERVIEW OF AVAILABLE
cotherapy, both separately and in combination, are recom- mended to treat the symptoms of PTSD. The SSRIs Relatively few guidelines on the treatment of PTSD have sertraline, paroxetine, and fluoxetine have demonstrated been published.35,36 Practice Guidelines from the their efficacy in the treatment of PTSD in randomized International Society for Traumatic Stress Studies recom- acute treatment trials. Sertraline is the only SSRI for which mend SSRIs as first-line treatment for PTSD in civilians36; long-term efficacy has been demonstrated in both continu- evidence for their use in veterans is difficult to interpret due ation and relapse-prevention trials. Despite the lack of to the severity and chronicity of PTSD in the veteran long-term RCTs, current guidelines from the Expert cohorts tested thus far. In addition, the guidelines strongly Consensus Panel for PTSD recommend that pharma- recommend the use of some form of exposure therapy in the cotherapy be continued in chronic PTSD for a minimum of treatment of PTSD.
12–24 months and that for CBT, "booster sessions" should In 1999, over 100 psycho- and pharmacotherapy experts be provided as needed.35 Further research is needed to opti- were surveyed, resulting in publication of the Expert mize the treatment of the sub-groups of PTSD patients who Consensus Guidelines for the treatment of PTSD.35 These do not respond to initial treatment with an SSRI or CBT.
guidelines recommend psychotherapy as a first-line treatment Additional long-term RCTs are also required to provide for mild PTSD or a combination of psycho- and pharma- further clarification of the most appropriate treatment cotherapy for patients with more severe or chronic problems.
strategies for patients suffering from PTSD. CNS
92% of acute-phase responders maintained response 8.6% of patients discontinued due to adverse events during continuation phase adverse events: upper respiratory tract infection, 54% of patients who failed to respond to acute phase (24.2%), headache (22.7%), ejaculatory failure in treatment with sertraline converted to responders males (17.6%), insomnia (17.2%), diarrhea (16.4%), nausea (2.5%), malaise (10.9%), fatigue (10.9%), dry mouth (10.9%), dizziness (10.2%) No treatment-emergent, treatment-related adverse events at a rate ≥10% for sertraline-treated patients all above plus time to and rate of relapse Stein DJ, Bandelow B, Hollander E, et al. CNS Spectrums. 2003. Vol 8, No 8 (suppl1). 2003. Volume 8 – Number 8 (Suppl 1) CNS Spectrums - August 2003 D.J. Stein, B. Bandelow, E. Hollander, et al
26. The International Classification of Diseases. 10th revision. Geneva, Switzerland: World Health Organization; 1992.
1. Diagnostic and Statistical Manual of Mental Disorders. 3rd ed. Washington, 27. Peters L, Slade T, Andrews G. A comparison of ICD-10 and DSM-IV cri- DC: American Psychiatric Association; 1980.
teria for posttraumatic stress disorder. J Trauma Stress. 1999;12:335-343.
2. Kessler RC, Sonnega A, Bromet E, Hughes M, Nelson CB. Posttraumatic 28. Blake DD, Weathers FW, Nagy LM, et al. The development of a stress disorder in the National Comorbidity Survey. Arch Gen Psychiatry. Clinician-Administered PTSD Scale. J Trauma Stress. 1995;8:75-90.
29. Guy W. Clinical Global Impressions. NCDEU Assessment Manual. 3. Breslau N, Kessler RC, Chilcoat HD, Schultz LR, Davis GC, Andreski P.
Psychopharmacology. 1976:217-222. Trauma and posttraumatic stress disorder in the community: the 1996 30. Weiss, D, Marmar, C. The Impact of Event Scale–Revised. In: J Wilson, Detroit Area Survey of Trauma. Arch Gen Psychiatry. 1998;55:626-632.
T Keane, eds. Assessing Psychological Trauma and PTSD. New York, NY: 4. Breslau N, Davis GC, Andreski P, Peterson E. Traumatic events and post- Guildford; 1997.
traumatic stress disorder in an urban population of young adults. Arch 31. Davidson JR, Book SW, Colket JT, et al. Assessment of a new self-rating Gen Psychiatry. 1991;48:216-222.
scale for post-traumatic stress disorder. Psychol Med. 1997;27:153-160.
5. Breslau N, Davis GC, Andreski P, Peterson EL, Schultz LR. Sex differ- 32. Davidson JR, Colket JT. The eight-item treatment-outcome post-trau- ences in posttraumatic stress disorder. Arch Gen Psychiatry. matic stress disorder scale: a brief measure to assess treatment outcome in post-traumatic stress disorder. Int Clin Psychopharmacol. 1997;12:41-45.
6. Brewin CR, Andrews B, Valentine JD. Meta-analysis of risk factors for 33. Keane TM, Caddell JM, Taylor KL. Mississippi Scale for Combat- posttraumatic stress disorder in trauma-exposed adults. J Consult Clin Related Posttraumatic Stress Disorder: three studies in reliability and validity. J Consult Clin Psychol. 1988;56:85-90.
7. Diagnostic and Statistical Manual of Mental Disorders. 4th ed. Washington, 34. Sloan P, Arsenault L, Hilsenroth M, Harvill L. Use of the Mississippi DC: American Psychiatric Association; 1994.
scale for combat-related PTSD in detecting war-related, non-combat 8. Yehuda R. Linking the neuroendocrinology of post-traumatic stress disor- stress symptomatology. J Clin Psychol. 1995;51:799-801.
der with recent neuroanatomic findings. Semin Clin Neuropsychiatry.
35. The Expert Consensus Guideline Series. Treatment of posttraumatic stress disorder. The Expert Consensus Panels for PTSD. J Clin 9. Post-traumatic Stress Disorder: Diagnosis, Management and Treatment. Nutt DJ, Psychiatry. 1999;60(suppl 16):3-76.
Davidson J, Zohar J, eds. London, England: Martin Dunitz Publishers; 2000.
36. Foa EB, Keane TM, Friedman MJ. Effective Treatments for PTSD: 10. Nutt DJ. The psychobiology of posttraumatic stress disorder. J Clin Practice Guidelines from the International Society for Traumatic Stress Studies. New York, NY: Guildford Press; 2000.
11. Yehuda R. Post-traumatic stress disorder. N Engl J Med. 2002;346:108-114.
37. Davidson J, Kudler H, Smith R, et al. Treatment of posttraumatic stress 12. Pitman RK, Shin LM, Rauch SL. Investigating the pathogenesis of post- disorder with amitriptyline and placebo. Arch Gen Psychiatry. traumatic stress disorder with neuroimaging. J Clin Psychiatry. 38. Stein DJ, Zungu-Dirwayi N, van Der Linden GJ, Seedat S.
13. Pitman RK, Shin LM, Rauch SL. Investigating the pathogenesis of post- Pharmacotherapy for posttraumatic stress disorder. Cochrane Database traumatic stress disorders with neuroimaging. J Clin Psychiatry.
Syst Rev. 2000:CD002795.
39. Davidson JRT, Connor KM. Management of posttraumatic stress disor- 14. Hamner MB, Lorberbaum JP, George MS. Potential role of the anterior der: diagnostic and therapeutic issues. J Clin Psychiatry. 1999;60(suppl cingulate cortex in PTSD: review and hypothesis. Depress Anxiety. 40. van der Kolk BA, Dreyfuss D, Michaels M, et al. Fluoxetine in posttrau- 15. Pitman RK. Overview of biological themes in PTSD. Ann N Y Acad Sci. matic stress disorder. J Clin Psychiatry. 1994;55:517-522.
41. Zohar J. Double-blind, placebo-controlled pilot study of sertraline in 16. Southwick SM, Paige S, Morgan CA 3rd, Bremner JD, Krystal JH, military veterans with posttraumatic stress disorder. J Clin Charney DS. Neurotransmitter alterations in PTSD: catecholamines and serotonin. Semin Clin Neuropsychiatry. 1999;4:242-248.
42. Davidson JR, Rothbaum BO, van der Kolk BA, Sikes CR, Farfel GM.
17. Southwick SM, Bremner JD, Rasmusson A, Morgan CA 3rd, Arnsten A, Multicenter, double-blind comparison of sertraline and placebo in the Charney DS. Role of norepinephrine in the pathophysiology and treat- treatment of posttraumatic stress disorder. Arch Gen Psychiatry. ment of posttraumatic stress disorder. Biol Psychiatry. 1999;46:1192-1204.
18. Warshaw MG, Fierman E, Pratt L, et al. Quality of life and dissociation 43. Brady K, Pearlstein T, Asnis GM, et al. Efficacy and safety of sertraline in anxiety disorder patients with histories of trauma or PTSD. Am J treatment of posttraumatic stress disorder: a randomized controlled trial.
19. Brady KT. Posttraumatic stress disorder and comorbidity: recognizing 44. Rothbaum BO, Ninan PT, Thomas L. Sertraline in the treatment of the many faces of PTSD. J Clin Psychiatry. 1997;58(suppl 9):12-15.
rape victims with posttraumatic stress disorder. J Trauma Stress. 20. Solomon SD, Davidson JR. Trauma: prevalence, impairment, service use, and cost. J Clin Psychiatry. 1997;58(suppl 9):5-11.
45. Brady KT, Sonne SC, Roberts JM. Sertraline treatment of comorbid 21. Friedman MJ. PTSD diagnosis and treatment for mental health clini- posttraumatic stress disorder and alcohol dependence. J Clin Psychiatry. cians. Community Ment Health J. 1996;32:173-189.
22. Zatzick DF, Marmar CR, Weiss DS, et al. Posttraumatic stress disorder and 46. Connor KM, Sutherland SM, Tupler LA, Malik ML, Davidson JR.
functioning and quality of life outcomes in a nationally representative Fluoxetine in post-traumatic stress disorder. Randomised, double-blind sample of male Vietnam veterans. Am J Psychiatry. 1997;154:1690-1695.
study. Br J Psychiatry. 1999;175:17-22.
23. Ballenger JC, Davidson JR, Lecrubier Y, et al. Consensus statement on 47. Martenyi F, Brown E, Zhang H, Prakash A, Koke S. Fluoxetine versus posttraumatic stress disorder from the International Consensus Group placebo in posttraumatic stress disorder. Poster presented at: 154th on Depression and Anxiety. J Clin Psychiatry. 2000;61(suppl 5):60-66.
Annual Meeting of the American Psychiatric Association; New 24. Kessler RC. Posttraumatic stress disorder: the burden to the individual Orleans, La; May 5-10, 2001.
and to society. J Clin Psychiatry. 2000;61(suppl 5):4-12.
48. Marshall RD, Beebe KL, Oldham M, Zaninelli R. Efficacy and safety of 25. Greenberg PE, Sisitsky T, Kessler RC, et al. The economic burden of paroxetine treatment for chronic PTSD: a fixed-dose, placebo-con- anxiety disorders in the 1990s. J Clin Psychiatry. 1999;60:427-435.
trolled study. Am J Psychiatry. 2001;158:1982-1988.
Volume 8 – Number 8 (Suppl 1) CNS Spectrums - August 2003 Long-Term Treatment of Posttraumatic Stress Disorder
49. Ruggiero L, Pitts CD, Dillingham K, Zaninelli R. A flexible-dose study 59. Londborg PD, Hegel MT, Goldstein S, et al. Sertraline treatment of of paroxetine in the treatment of PTSD. Poster presented at: 154th posttraumatic stress disorder: results of 24 weeks of open-label continua- Annual Meeting of the American Psychiatric Association; May 5-10, tion treatment. J Clin Psychiatry. 2001;62:325-331.
2001; New Orleans, La.
60. Davidson JR, Londborg P, Pearlstein T, Rothbaum B, Brady K, Farfel G.
50. Kosten TR, Frank JB, Dan E, McDougle CJ, Giller EL Jr.
Sertraline and post-traumatic stress disorder: results of 24 weeks of Pharmacotherapy for posttraumatic stress disorder using phenelzine or open-label treatment followed by a 28-week discontinuation study.
imipramine. J Nerv Ment Dis. 1991;179:366-370.
Poster presented at: XIII Congress of the European College of 51. Baker DG, Diamond BI, Gillette G, et al. A double-blind, randomized, Neuropsychopharmacology; September 9-13, 2000; Munich, Germany.
placebo-controlled, multi-center study of brofaromine in the treatment of 61. Davidson J, Pearlstein T, Londborg P, et al. Efficacy of sertraline in pre- post-traumatic stress disorder. Psychopharmacology (Berl). 1995;122:386-389.
venting relapse of posttraumatic stress disorder: results of a 28-week dou- 52. Davidson JR. Biological therapies for posttraumatic stress disorder: an ble-blind, placebo-controlled study. Am J Psychiatry. 2001;158:1974-1981.
overview. J Clin Psychiatry. 1997;58(suppl 9):29-32.
62. Rapaport M, Endicott J, Clary C. Quality of life in long-term sertraline 53. Katz RJ, Lott MH, Arbus P, et al. Pharmacotherapy of post-traumatic treatment for PTSD. Poster presented at: 7th World Congress of stress disorder with a novel psychotropic. Anxiety. 1994;1:169-174.
Biological Psychiatry; July 1-6, 2001; Berlin, Germany.
54. Gelpin E, Bonne O, Peri T, Brandes D, Shalev AY. Treatment of recent 63. Foa EB, Meadows EA. Psychosocial treatments for posttraumatic stress trauma survivors with benzodiazepines: a prospective study. J Clin disorder: a critical review. Annu Rev Psychol. 1997;48:449-480.
64. Keane TM, Fairbank JA, Caddell JM. Implosive (flooding) therapy 55. Viola J, Ditzler T, Batzer W, et al. Pharmacological management of post- reduces symptoms of PTSD in Vietnam combat veterans. Behav Ther. traumatic stress disorder: clinical summary of a five-year retrospective study, 1990-1995. Mil Med. 1997;162:616-619.
65. Foa EB, Dancu CV, Hembree EA, Jaycox LH, Meadows EA, Street GP.
56. Risse SC, Whitters A, Burke J, Chen S, Scurfield RM, Raskind MA.
A comparison of exposure therapy, stress inoculation training, and their Severe withdrawal symptoms after discontinuation of alprazolam in combination for reducing posttraumatic stress disorder in female assault eight patients with combat-induced posttraumatic stress disorder. J Clin victims. J Consult Clin Psychol. 1999;67:194-200.
66. Marks I, Lovell K, Noshirvani H, Livanou M, Thrasher S. Treatment of 57. Yehuda R. Managing anger and aggression in patients with posttraumat- posttraumatic stress disorder by exposure and/or cognitive restructuring: ic stress disorder. J Clin Psychiatry. 1999;60(suppl 15):33-37.
a controlled study. Arch Gen Psychiatry. 1998;55:317-325.
58. Hertzberg MA, Butterfield MI, Feldman ME, et al. A preliminary study of lamotrigine for the treatment of posttraumatic stress disorder. BiolPsychiatry. 1999;45:1226-1229.
Volume 8 – Number 8 (Suppl 1) CNS Spectrums - August 2003 WCA Recommendations for the
Long-Term Treatment of Social Phobia
By Michael Van Ameringen, MD, FRCPC, Christer Allgulander, MD, Borwin Bandelow, MD, John H. Greist, MD, Eric Hollander, MD, Stuart A. Montgomery, MD, David J. Nutt, MD, FRCP, FRCPsych, FMedSci, Ahmed Okasha, MD, PhD, FRCP, FRCPsych, FACP, Mark H. Pollack, MD, Dan J. Stein, MD, PhD, and Richard P. Swinson, MD, FRCPC, FRCPsych, DPM studies of clonazepam are limited but support the drug's efficacy. There is also evidence for the effectiveness of exposure-based • Selective serotonin reuptake inhibitors are the drugs of strategies of cognitive-behavioral therapy, and controlled studies first choice in the treatment of social phobia, paroxe- suggest that the effects of treatment are generally maintained at tine being the only one approved by the Food and long-term follow-up. In light of the chronicity and disability associ- Drug Administration for this indication.
ated with social phobia, as well as the high relapse rate after short- • Sertraline, clonazepam, phenelzine, and moclobemide term therapy, it is recommended that effective treatment be have also been studied in long-term randomized clini- continued for at least 12 months. CNS Spectr 2003;8(suppl 1):40-52 • Cognitive-behavioral therapy has demonstrated efficacy in randomized clinical trials but has limited availability • Current guidelines recommend that pharmacotherapy be continued for a minimum of 12 months after symp- Social phobia (social anxiety disorder) is a chronic and highly prevalent disorder that is often associated with signifi- • More controlled research is needed on the relative and cant psychosocial impairment.1 Although highlighted as a combined short- and long-term efficacy of pharmaco- substantial public health problem by many researchers,2,3 logic and psychologic treatment for social phobia.
social phobia remains widely underrecognized and under- treated.4 Very little had been known about cultural differ- ences in the prevalence of social phobia, but increased What is the best approach for treating patients with social pho- cross-cultural awareness and information regarding preva- bia (social anxiety disorder) over the long term? Social phobia is lence, presentation, and diagnosis of the disorder are now the most common anxiety disorder, with reported prevalence rates becoming available.5 of up to 18.7%. Social phobia is characterized by a marked and Only a small proportion of people with social phobia seek persistent fear of being observed or evaluated by others in social professional treatment.6-9 The reason for this is unclear, but performance or interaction situations and is associated with physi- theories suggest that the fear of social situations extends to cal, cognitive, and behavioral (ie, avoidance) symptoms. The help-seeking situations,10 or that affected individuals do not onset of social phobia typically occurs in childhood or adolescence define their anxiety as an illness.6 Among patients with and the clinical course, if left untreated, is usually chronic, symptoms of social phobia, those with agoraphobia appear to unremitting, and associated with significant functional impair- be the most likely to seek treatment,6 perhaps because symp- ment. Social phobia exhibits a high degree of comorbidity with toms of agoraphobia may be more easily recognized as devia- other psychiatric disorders, including mood disorders, anxiety dis- tions from normal behavior. Interestingly, patients with orders, and substance abuse/dependence. Few people with social social phobia are more likely to be recognized by their pri- phobia seek professional help despite the existence of beneficial mary care physician if they have a comorbid condition such treatment approaches. The efficacy, tolerability, and safety of the as major depression.11 selective serotonin reuptake inhibitors (SSRIs), evidenced in ran- domized clinical trials, support these agents as first-line treatment. CLINICAL PRESENTATION AND DIAGNOSIS
The benzodiazepine clonazepam and certain monoamine oxidase Social phobia is characterized by a marked and persistent inhibitors (representing both reversible and nonreversible fear of being observed or evaluated by others in social perfor- inhibitors) may also be of benefit. Treatment of social phobia may mance or interaction situations.12,13 The individual fears act- need to be continued for several months to consolidate response ing in a way that will cause him or her to be humiliated or and achieve full remission. The SSRIs have shown benefit in long- embarrassed, or exhibit anxiety symptoms, and consequently term treatment trials, while long-term treatment data from clinical seeks to avoid situations where close scrutiny, whether real or These recommendations are based on proceedings from the World Council of Anxiety meeting held September 11, 2000, in Pisa, Italy, and on guidelines and articles published in the medical literature through October 15, 2001.
Please direct all correspondence to: Michael Van Ameringen, MD, FRCPC, Department of Psychiatry & Behavioral Neurosciences, McMaster University, 1200 Main St West, Hamilton, Ontario L8N 3Z5, Canada. Tel: 905-521-2100; Fax: 905-521-2628; E-mail: [email protected]. Volume 8 – Number 8 (Suppl 1) CNS Spectrums - August 2003 Long-Term Treatment of Social Phobia
imagined, might take place. The majority of patients with lic performance (eg, public speaking), when scrutiny by oth- social phobia experience onset in childhood or adoles- ers may be possible.The generalized subtype, found in 50% cence,6,14 a critical time in terms of social and academic of lifetime cases of social phobia, is more severe and com- development. Onset at this impressionable age contributes plex, and is characterized by strong fear and avoidance of to the significant interference of this disorder in the patient's multiple social interactional and performance situations.10 personal, academic, and professional life.
The generalized form of social phobia is usually more func- The key diagnostic criteria for social phobia in both the tionally disabling, familial, and longer lasting than the non- Diagnostic and Statistical Manual of Mental Disorders, Fourth generalized, specific subtype; there is a lesser chance of Edition (DSM-IV)12 and the International Classification of spontaneous recovery with this subtype and it is associated Diseases, Tenth Revision (ICD-10)13 are mostly similar but with a higher risk of comorbidity and impairment.3,22 have the following differences: Findings from a large community survey (N=1,956) indi- (A) The DSM-IV12 requires the presence of significant cate that delineation of the two subtypes appears to be some- interference with the patient's normal routine, occu- what arbitrary1 because a continuum of severity associated pational (or academic) functioning, or social activi- with a greater number of feared situations and greater disabil- ties/relationships, while the ICD-1013 cautions ity is seen with social phobia. There is currently no definitive against the use of lifestyle impairment as a diagnostic threshold demarcating the two subtypes, though a preliminary criterion because of its dependence on the social receiver operating characteristic analysis23 suggests that a cut environment, which may be influenced by culture score of 60 on the Liebowitz Social Anxiety Scale (LSAS) and differ according to the social situation15; may be a useful method for identifying the generalized sub- (B) The ICD-1013 recognizes specific physical symptoms type, with a sensitivity of 73% and a specificity of 74%.
(eg, blushing), while the DSM-IV12 refers to anxiety symptoms that may take the form of a panic attack; (C) The ICD-1013 diagnosis requires one of the following While differentiating social phobia from other anxiety symptoms: persistent fear of social situations, fear of humil- disorders can be difficult, the context of the symptoms pro- iation, or avoidance of social situations, while DSM-IV vides important diagnostic clues. The differentiation of criteria12 stipulate that all three symptoms be present.
social phobia from normal shyness is a qualitative and quan- titative issue related both to the type of fear and avoidance behavior and the level of distress and impairment associated Social phobia is characterized by the presence of physical, with social fears.
cognitive, and behavioral (avoidance) symptoms. The physi- The most difficult differential diagnostic consideration cal symptoms may resemble anxiety attacks and include is the potential confusion with panic disorder with or blushing, sweating, shaking, palpitations, nausea, diarrhea, without agoraphobia.17 Several features are common to and speech block.16,17 Increased heart rate and blood pressure social phobia and panic disorder: individuals with either are also associated with social phobia when subjects are disorder may experience panic attacks, comorbid depres- exposed to their feared stimuli.17 Maladaptive thoughts about sion, distress in social situations, substance abuse, and sui- social situations are important cognitive symptoms.16 cidal ideation.23 However, a diagnosis of panic disorder Significant functional impairments are associated with requires a history of unexpected panic attacks that do not social phobia, particularly in the areas of partner and family occur exclusively in social situations.12 It is also helpful for relationships, education, and career development.1,15,18-20 a clinician to ask patients what they fear in a given situa- People with social phobia have been reported to achieve tion. Generally, patients with social phobia fear humilia- lower-than-average educational levels and have less stable tion and embarrassment, while those with panic disorder employment histories compared with the general popula- fear the potential medical consequences of their panic tion.19,20 Most of these patients have lower socioeconomic attack symptoms.17 In addition, the mean age of onset is status, with >70% falling into the lower half of the popula- younger for social phobia (15.6 years; standard deviation tion in terms of socioeconomic achievement.17 Furthermore, [SD]=9.9) than for panic disorder (29.57 years; suicidal ideation has been shown to occur more frequently in SD=9.5).23,24 Response to deliberate provocation, course, people with social phobia than in the general population.21 and response to treatment, as well as six specific symptoms In one study, an increase in the number of suicide attempts (palpitations, chest pains, tinnitus, blurred vision, was observed in patients with social phobia and a comorbid headaches, fear of dying, dry mouth) also help to distin- condition such as depression.21 guish between the two disorders,24-26 as does the presence of flushing and tremor in some cases.27 Social phobia commonly overlaps with avoidant personal- Two subtypes of social phobia are recognized: a nongener- ity disorder (APD). The DSM-IV indicates that APD should alized (discrete/specific) form and a generalized form,12,16 also be considered as an additional diagnosis in persons with although the ICD-10 does not differentiate between the generalized social phobia; the two diagnostic groups do not two.13 The nongeneralized form usually involves fear of pub- appear to differ qualitatively in any significant way.12,28 Volume 8 – Number 8 (Suppl 1) CNS Spectrums - August 2003 M. Van Ameringen, C. Allgulander, B. Bandelow, et al
Other potential differential diagnoses for social phobia are described in a review by Moutier and Stein28 and include Natural course studies show that the onset of social pho- major depression with social withdrawal, generalized anxiety bia typically occurs in the early teenage years,3,10 although disorder, obsessive-compulsive disorder, body dysmorphic affected individuals who seek treatment frequently wait until disorder, medical conditions such as Parkinson's disease, and later in life to do so.2,9 Social phobia tends to follow a chronic paranoid symptoms in psychoses (the feeling of being course and demonstrates high levels of comorbidity with observed by others).
other anxiety, mood, and substance abuse disorders. It has Differences in the phenomenology of anxiety symptoms been hypothesized that social phobia increases the risk for may also come with age (eg, elective mutism in children or onset of secondary mood disorders.40 Therefore, a relation- adolescents)29,30 and culture (eg, in the Japanese culture, "tai- ship can be expected between primary social phobia and the jin kyofusho," which is a fear of offending or hurting others subsequent onset, course, and severity of secondary disorders.
through one's awkward social behavior or an imagined physi- This, in turn, is a strong predictor of the chronicity of social cal defect).31,32 phobia9,41 The US National Comorbidity Survey (NCS)40 showed that 27% of patients who have lifetime major depression, 29% patients with dysthymia, and 47% patients Estimates of the prevalence of social phobia have with bipolar disorder, also have social phobia. Additionally, recently been refined by large-scale epidemiological stud- 34% of patients with social phobia, compared with about ies of psychiatric disorders, which have found the disorder 15% patients without social phobia, have a mood disorder.
to be common within the community.1,33 Social phobia is The combination of high prevalence, early onset, and a the most prevalent of the anxiety disorders34 and, accord- chronic lifetime course, as well as risks of comorbidity, ing to the US National Comorbidity Survey, is the third underdiagnosis, and low probability of treatment, make most common psychiatric disorder after major depression social phobia important from a public health perspective.2,3,17 and alcohol dependence.35 However, it remains underrec- Social phobia is associated with lost wages, reduced produc- ognized and undertreated.4 Because only a small propor- tivity, increased disability, and lower quality-of-life indices.29 tion of people with social phobia seek treatment,6 it is Individuals with generalized social phobia earn salaries 19% unlikely that those treated are representative of the major- lower than those without the disorder.42 The costs of these ity of people affected by the disorder.1 impairments are not restricted to individual patients but also Previously, the reported prevalence of social phobia affect employers, health insurance providers, and the ranged from 1.9% to 18.7%, as a result of differing diagnostic national economy.43,44 thresholds and variations in the stringency of the definitions of distress and impairment.1,3,6,7,22,36 Most early studies were based on Diagnostic and Statistical Manual of Mental Disorders, Social phobia is poorly understood from a neurobiological Third Edition (DSM-III)37 criteria assessed with the perspective. Although selective serotonin reuptake Diagnostic Interview Schedule (DIS), which covered only inhibitors (SSRIs) are successfully as pharmacologic treat- three social fear situations; estimates of prevalence from ment, it is unknown whether this condition—and therefore these studies are therefore regarded as somewhat conserva- the success of treatment with SSRIs—is associated with tive.1 This underestimation was corrected in the successor to alteration of serotonin (5-HT) neurotransmission.45 the DIS, the Composite International Diagnostic Interview Researchers have recently begun to explore the underlying (CIDI) of the World Health Organization (WHO), which neurobiology of the disorder, employing various approaches, includes a wider range of social situations and therefore including assessment of central neurotransmitter function, yields considerably higher estimates.6,7 response to chemical and neuroendocrine challenges, and In epidemiological samples, there is a higher prevalence functional neuroimaging.44,46-48 Current neurobiological evi- of social phobia in females than in males, but in clinical dence, however, suggests that there are abnormalities in the samples the gender distribution is equal, perhaps due to dopamine and serotonin systems of these patients.49,50 selection and/or recognition bias in clinical settings.3,8 Rates of social phobia are consistently higher in younger- versus older-age cohorts and are inversely associated with Mood and Anxiety Disorders
Individuals with social phobia are at increased risk for Recent cohorts have reported a higher lifetime preva- other psychiatric disorders, with mood disorder (particularly lence of generalized social phobia than in the past.9,38 major depression) being the most common other lifetime Current estimates place the point prevalence of social diagnosis.2,8,45 Epidemiological and clinical samples suggest phobia at 4% to 6% and the lifetime risk between 7% and that more than one third of people with social phobia report 13%.39 People with social and economic advantages (eg, a lifetime mood disorder (34% in the NCS,2 35% in a clini- white, educated, married status) appear to be at greater cal study by Stein and colleagues,51 and as high as 83% in a risk for social phobia; this finding is not explained by study by van Ameringen and colleagues52). Social phobia increased comorbidity with other mental disorders.38 also exhibits a high degree of comorbidity with other anxiety Volume 8 – Number 8 (Suppl 1) CNS Spectrums - August 2003 Long-Term Treatment of Social Phobia
disorders,33 particularly with panic disorder and generalized Paroxetine was the first SSRI to gain approval for the anxiety disorder, with a reported lifetime prevalence of 50% treatment of social phobia (May 1999) and is approved for and 32%, respectively, in one clinical sample.52 this indication in 35 countries, including Canada, Germany, In both clinical and epidemiological samples, social the US, and the United Kingdom. Evidence of paroxetine's phobia occurs prior to any episode of mood or other anxi- effectiveness first arose from open-label studies.62,63 ety disorder in the majority of individuals,2,40,52 although Subsequently, three 12-week, double-blind, placebo-con- this observation should be treated with caution as it is trolled, flexible-dose trials consistently demonstrated that based on retrospective analyses. It suggests, however, that paroxetine (20–50 mg/day) is both effective and well toler- in most individuals there is a window of opportunity to ated in the acute treatment of social phobia.64-66 The first treat social phobia prior to the onset of a secondary mood involved 183 participants, the second included 92 partici- or anxiety disorder.40 pants, and the third included 290 subjects, demonstrating paroxetine's efficacy in a total of 565 participants. Of the Substance Abuse Disorders
183 patients for whom efficacy data were available in a study Individuals affected by social phobia may self-medicate by Stein and colleagues,64 55% of those receiving paroxetine with alcohol or other substances to relieve their anxiety.21 versus 22% of patients taking placebo were responders Substance and alcohol abuse are more common lifetime (based on Clinical Global Impression-Improvement scale diagnoses in these individuals than in the general popula- [CGI-I] score <2 [much or very much improved]) at the end tion.45 In patients with social phobia, the incidence of alco- of 12 weeks of treatment (P=.001). In addition, mean holism has been reported to range from 14% to 40%.53 change from baseline on the LSAS was significantly greater Although the relationship between alcohol dependence and (P<.001) in paroxetine-treated patients (30.5±2.66) than in social phobia is complex, many patients must drink before placebo-treated patients (14.5±2.63). Paroxetine treatment entering social situations. Social phobia has also been found also resulted in significant improvement in five of six sec- to be associated with heavy smoking and nicotine depen- ondary efficacy measures and did not cause any unusual dence in both cross-sectional retrospective and prospective- adverse events. At study endpoint, the mean dose of paroxe- longitudinal analyses.54,55 Additionally, the failure rate of tine was 36.6±12.1 mg. The results of a fourth 12-week study smoking cessation (89%) is higher in individuals with social in which 384 patients were randomized to fixed-doses of phobia than in the general population.55 paroxetine found comparable CGI-I responder rates on the 20- (45%), 40- (47%), and 60-mg dose (43%).67 The higher response rates when flexible doses of paroxetine were Increased awareness of social phobia has led to a delin- employed (in the range of 50% to 55%) suggest that even eation of disease subtypes and research on its treatment.16 though no dose-response group effect was observed, many A number of treatment approaches have been demonstrated individual patients apparently benefit from the ability to to be beneficial for social phobia, including pharmacother- titrate to higher doses.
apy, psychotherapy, or a combination of both.
Sertraline is also FDA approved for the treatment of The rating instruments most commonly used to assess social phobia (February 2003). After an early sertraline treatment effects in published pharmacologic studies of crossover study by Katzelnick and colleagues,61 subsequent social phobia include the LSAS,56 the Brief Social Phobia evidence of sertraline's efficacy in the treatment of general- Scale (BSPS),57,58 and the Fear Questionnaire (FQ)59 ized social phobia was supported by findings from a 20-week, double-blind, placebo-controlled study involving 204 patients.68 At the end of the study, significantly more Although a number of pharmacologic agents have been patients randomized to sertraline (50–200 mg/day) were investigated in the treatment of social phobia, very few are cur- responders based on CGI-I criteria. In addition, significantly rently licensed for this indication and most prescribing is off- greater reductions were reported in the MFQ social phobia label.60 The current body of knowledge on pharmacotherapeutic subscale (MFQ-SP) and BSPS among sertraline-treated intervention is based predominantly on short-term clinical trials patients. It was observed that 53% (71/203) of sertraline- investigating the generalized subtype of social phobia.
treated patients and 29% (20/203) of placebo-treated patients were responders at endpoint. Thirty percent of sertraline- Acute Treatment Trials
treated patients compared with 13% of placebo-treated Selective Serotonin Reuptake Inhibitors
patients showed marked clinical improvement. The sertraline Clinical trials have demonstrated the efficacy, tolerability, group demonstrated mean reductions of 32.6% and 34.3% in and safety of SSRIs in social phobia and consistently support the MFQ-SP and BSPS, respectively, compared with reduc- their use as first-line treatment.16 tions of 18.8% and 18.6% for the placebo group. Overall, ser- In an early double-blind crossover study,61 12 outpatients were traline was well tolerated with no differences in randomly assigned to 10 weeks of sertraline (50–200 mg/day, discontinuation rates compared with placebo-treated patients.
flexible dosing) and 10 weeks of placebo. The results indi- These results were confirmed by findings from a random- cated that sertraline was effective and well tolerated.61 ized, double-blind, placebo-controlled trial of flexible-dose Volume 8 – Number 8 (Suppl 1) CNS Spectrums - August 2003 M. Van Ameringen, C. Allgulander, B. Bandelow, et al
sertraline (50–200 mg) in patients with moderate-to-severe moclobemide versus 13.5% for placebo; P=not significant) generalized social phobia, as evidenced by a baseline LSAS was observed by Schneier and colleagues.83 score of 91, which is the highest among published data seen Phenelzine and moclobemide were compared in a double- in SSRI trials.69 Sertraline was shown to be highly efficacious blind, placebo-controlled, parallel-group trial that enrolled and well tolerated. At last-observation-carried-forward end- 78 patients with social phobia and lasted more than point, sertraline-treated patients showed significantly higher 24 weeks.83 Patients were treated for 8 weeks initially (mean CGI-I response rates than placebo-treated patients (47% dose at end of the first phase was 580.7 mg/day for moclobe- versus 26%; P<.001), as well as greater decreases in mean mide and 67.5 mg/day for phenelzine), after which nonre- LSAS total score (-31 versus -21). At week 12, the mean sponders (as measured by CGI) were withdrawn from the change in LSAS total score was significantly greater in the study. The efficacies of moclobemide and phenelzine were sertraline group than in the placebo group (-34 versus -23; statistically comparable for most outcome measures and sig- P<.001), as was the percentage of CGI-I responders (56% nificantly superior to placebo from week 4 onward.
versus 29%; P<.01). Sertraline was also superior to placebo Moclobemide was tolerated much better than phenelzine on such secondary efficacy measures as the BSPS and and had fewer and milder side effects than monoamine oxi- Hamilton Anxiety and Depression rating scales, and on dase inhibitors (MAOIs).84 quality of life and functional measures, including the Quality Brofaromine is another RIMA that has been studied in of Life, Enjoyment, and Satisfaction Questionnaire and the the treatment of social phobia. In a 12-week, double-blind, Sheehan Disability Inventory.
placebo-controlled trial in 30 patients with Diagnostic and Finally, a third placebo-controlled trial70 examined the effi- Statistical Manual of Mental Disorders, Third Edition-Revised cacy of sertraline (50–150 mg/day for 24 weeks) with or without (DSM-III-R)85 social phobia, brofaromine (150 mg/day) led exposure therapy in 387 patients in the general practice setting.
to "clinically relevant improvement" in 80% of patients.86 This trial confirmed the efficacy of sertraline over placebo Significant improvement was observed in social phobia, pho- (P<.001) in the treatment of generalized social phobia.70 bic avoidance, general anxiety, and interpersonal sensitivity Fluvoxamine (100–300 mg/day) has also been shown to be measures in brofaromine- but not placebo-treated patients.
effective, safe, and generally well tolerated in the short-term In another 12-week, double-blind, placebo-controlled pharmacologic management of social phobia in a number of study of brofaromine 150 mg/day administered to randomized, double-blind, placebo-controlled trials.71,72 In each 77 patients with a DSM-III-R diagnosis of social phobia, trial, there was a significantly higher proportion of responders 78% of brofaromine-treated patients were CGI-I responders in the fluvoxamine group than in the placebo group, with sta- compared with 23% of placebo-treated patients.87 tistically significant effects seen on social phobia and psychoso- Significant improvement with brofaromine treatment com- cial disability rating scales at 12-week trial endpoints.
pared with placebo was also observed on the LSAS and the The efficacy of citalopram in social phobia has not been Montgomery-Asberg Depression Rating Scale. However, well documented. Case series73-75 indicate that citalopram brofaromine has been withdrawn from the worldwide mar- may be effective at doses of 20–40 mg/day, but controlled tri- ket and is currently unavailable.
als are needed to confirm these data. Similarly, findings from Monoamine Oxidase Inhibitors
a number of small, short-term, open-label trials of fluoxetine A number of studies have demonstrated the superior effi- (up to 80 mg/day) have suggested potential efficacy in social cacy of MAOIs (phenelzine,88,89 tranylcypromine90) in treat- phobia,76-78 but no double-blind studies have been published.
ing generalized social phobia. However, because of Reversible Inhibitors of Monoamine Oxidase Type A
significant adverse events and interactions, these agents are Four placebo-controlled studies of moclobemide, a generally reserved for patients who are refractory to other reversible inhibitor of monoamine oxidase type A treatment options.91 (RIMA), achieved varied results (reviewed in van der The efficacy of the irreversible, nonselective MAOI Linden and colleagues79). Moclobemide 600 mg/day has phenelzine (up to 90 mg/day) was first indicated in 1985 been reported to yield a 52% response rate (P=.0022 ver- by results from a small open-label pilot study.92,93 The first sus placebo) in a severely affected subset of patients com- published placebo-controlled study that evaluated pared with response rates of 37% for 300 mg/day phenelzine for social phobia compared pharmacotherapy moclobemide (P=.2016 versus placebo) and 30% for with phenelzine or alprazolam with cognitive-behavioral placebo,80 and has shown superiority versus placebo begin- group therapy (CBGT) in 65 patients.94 All treatments, ning at week 4 in a trial by Versiani and colleagues.81 including pill placebo, were associated with substantial However, these positive effects have not been supported improvements in patients with severe and chronic social by findings from other placebo-controlled studies.82,83 A phobia; self-directed exposure was strongly encouraged large (N=523) multicenter study showed that 35% of sub- among all participants, which probably contributed to this jects taking the highest dose of moclobemide observation. Although all treatments significantly (900 mg/day) and 33% of placebo-treated patients were improved symptoms of social phobia and a trend toward CGI-I responders after 12 weeks of treatment.84 An even greater efficacy with phenelzine was observed, the results smaller clinical effect (CGI-I responder rates of 17.5% for did not favor any one treatment.
Volume 8 – Number 8 (Suppl 1) CNS Spectrums - August 2003 Long-Term Treatment of Social Phobia
Phenelzine 15–90 mg/day was subsequently shown to be social phobia in double-blind, placebo-controlled trials.15 effective in a 12-week placebo-controlled trial enrolling The efficacy of imipramine in social phobia was not sup- 133 patients, which also compared treatment with cognitive- ported by a small, 8-week, open-label study (N=15).101 behavioral therapy (CBT).88 At endpoint, phenelzine was A number of other agents, including venlafaxine102-104 and associated with superior response rates (Social Phobic mirtazapine,105 are in the early stages of evaluation for the Disorders Severity and Change Form) compared with treatment of social phobia.
placebo (65% versus 33%, respectively; P<.005) and greater β-blockers (eg, atenolol), sometimes used in clinical prac- change on dimensional measures (Anxiety Disorders tice for the treatment of social phobia, may be effective for Interview Schedule-Revised, Clinician's Severity Rating, performance anxieties, such as public speaking fears.19 and LSAS). Response to phenelzine was more evident than However, they do not appear to have efficacy in generalized response to CBT after 6 weeks (35% versus 59%, respec- social anxiety disorder and are not indicated for its treat- tively; P<.03) and was superior at 12 weeks on several mea- ment.16,106-107 A small (N=16) placebo-controlled trial did not sures (LSAS social avoidance and social fear), although both reveal any additional benefits from the adjunctive use of pro- treatments were more efficacious than control conditions pranolol in individuals with social phobia who underwent a (pill placebo or placebo attention treatment).
4-week course of social skills training.108 Furthermore, a study Results of a multiphase, placebo-controlled, comparative comparing the efficacy of either phenelzine or the cardiose- trial (N=74) of phenelzine (up to 90 mg/day) and the lective β-adrenergic blocker atenolol with placebo in β-blocker atenolol (up to 100 mg/day) in patients with DSM-III 74 patients with DSM-III social phobia found that while social phobia were described in a series of reports by Liebowitz phenelzine was significantly more effective than atenolol and colleagues.89,95,96 The overall CGI-I responder rate at 8 weeks and placebo, atenolol was not significantly more effective for phenelzine was 64%, superior to both atenolol (30%) and placebo (23%). Patients who responded to treatment at the end Finally, the results of a 12-week, follow-up, double- of 8 weeks were continued on the same dosage in an 8-week blind, placebo-controlled study109 of a small (N=21) open- double-blind maintenance phase. At the end of 16 weeks, label trial did not confirm the previously observed modest phenelzine was still significantly superior to placebo. Patients efficacy of the azaspirone anxiolytic buspirone in the with generalized social phobia constituted 76% of the sample treatment of social phobia.110 and were preferentially responsive to phenelzine.
Anticonvulsants are also being evaluated for the treat- ment of social phobia. The safety and efficacy of gabapentin Although only limited studies have been conducted to was evaluated in a randomized, double-blind, placebo-con- date,97-99 benzodiazepines, such as clonazepam, have been trolled trial in which 69 patients received either gabapentin reported to be effective in the treatment of social phobia.
(900–3,600 mg/day) or placebo for 14 weeks.111 Patients Benzodiazepines are not recommended as first-line agents treated with gabapentin showed a significant reduction in because they do not also treat associated comorbid condi- social phobia symptoms, as assessed by LSAS total score tions, such as major depression and obsessive-compulsive dis- (P<.05), compared with placebo-treated patients. The treat- order, and because of tolerability/dependence issues.
ment effects of gabapentin appeared to be dose-dependent Benzodiazepines may, however, benefit patients who are and no serious adverse events were noted. In an 11-week refractory to other treatments, and they may be useful as aug- double-blind study of pregabalin, a follow-up compound to menting agents.17,91 However, benzodiazepines may be associ- gabapentin that is in development,112 135 patients were ran- ated with long-term dependence.17,91 domized to either low- (150 mg/day; n=42) or high-dose In a 10-week double-blind pilot study enrolling (600 mg/day; n=46) pregabalin or placebo (n=46). Efficacy 75 patients with social phobia, clonazepam (2.4 mg/day was assessed primarily as change in LSAS score from baseline mean maximum dose at endpoint) was reported to yield a to endpoint, and high-dose pregabalin (600 mg/day) signifi- 78% CGI-I responder rate compared with a 20% response cantly improved LSAS score and several secondary efficacy rate with placebo (P=.0001).98 Although unsteadiness and measures between week 1 and week 11.
dizziness were reported, clonazepam was generally well toler- ated. In a 2-year follow-up of 56 patients (75%), subjects ini- Long-term Treatment Trials
tially treated with clonazepam were reported to exhibit Few studies have investigated the efficacy and safety of significantly less severe social phobia scores (LSAS, FQ, and long-term treatment of social phobia with agents that have Sheehan Disability Scale [SDS]) compared with placebo shown short-term efficacy (Table).70,81,99,113-115 subjects, suggesting evidence of long-term benefit.100 Selective Serotonin Reuptake Inhibitors
In the only double-blind, placebo-controlled trial of alpra- Sertraline is the only SSRI to gain approval for the long- zolam,94 the response rate was only 38% compared with 69% term treatment of social phobia (February 2003). The long- for phenelzine, 24% for CBT, and 20% for placebo (P<.09).
term safety and efficacy of sertraline treatment for social phobia was demonstrated in a 24-week, double-blind, Tricyclic antidepressants, shown to be effective in other placebo-controlled relapse prevention study.114 Fifty patients anxiety disorders, have not been extensively investigated in with generalized social phobia who had responded to Volume 8 – Number 8 (Suppl 1) CNS Spectrums - August 2003 M. Van Ameringen, C. Allgulander, B. Bandelow, et al
20 weeks of flexible-dose sertraline (50–200 mg/day) in this Maintained efficacy of paroxetine in the long-term relapse prevention study were then randomly assigned to ser- treatment of social phobia has also been demonstrated in a traline or placebo for another 24 weeks.68 An additional 36-week study which assessed the potential for relapse 15 patients who had responded in the placebo arm of the after medication discontinuation.114 After a 12-week, sin- original trial also continued to receive double-blind placebo gle-blind acute phase, during which 380 patients with treatment in the continuation study. At endpoint, signifi- DSM-IV social phobia were treated with paroxetine cantly fewer (4% [1/25]) patients in the sertraline-continua- (20–50 mg), 323 patients whose CGI-S score had tion group than the placebo-switch group (36% [9/25]) had decreased by at least two points were randomized in dou- relapsed (P=.01). Mean rating scale scores (Clinical Global ble-blind fashion to an additional 24 weeks of treatment Impression-Severity [CGI-S], BSPS, and MFQ-SP) contin- with paroxetine or placebo. At endpoint, the relapse rate ued to improve in the sertraline-continuation group but was significantly lower in the paroxetine group than in the deteriorated in both the placebo-switch and placebo-respon- placebo group (14% versus 39%; P<.001). Paroxetine was der groups. This study suggests that sertraline treatment last- well tolerated for the duration of the study.
ing up to 11 months effectively prevents relapse and the Continuing improvements in patients' symptoms and dis- reemergence of social phobia symptoms. Long-term sertra- abilities have been reported in a long-term, open-label line treatment was well tolerated in this study.
extension of the first randomized, double-blind, placebo- Paroxetine's potential to prevent relapse in social phobia controlled study of paroxetine in social phobia.64,115 Ninety was studied in a 23-week trial by Stein and colleagues.63 The patients who had completed 12 weeks of treatment with researchers treated 36 subjects with open-label paroxetine paroxetine or placebo (Phase 1) received open-label paroxe- (10–50 mg/day) for 11 weeks. Of the 23 responders (CGI), tine treatment for 24 weeks (Phase 2) followed by a 16-week 16 agreed to enter a double-blind placebo-switch phase for double-blind re-randomization phase (Phase 3) with paroxe- an additional 12 weeks. Only one of the eight patients who tine or placebo. The number of paroxetine responders continued on paroxetine relapsed, compared with five of increased from 55% (50/91) at the end of Phase 1 (compared eight patients switched to placebo. Although these results with 23.9% [22/92] in the placebo group [odds ratio 3.88; suggest an advantage for paroxetine, the small sample size 95% confidence interval, 2.81-5.36])59 to 89% (57/64) after does not allow for definite conclusions.
24 weeks of open-label paroxetine treatment (Phase 2).
TABLE. LONG-TERM TREATMENT TRIALS OF PHARMACOTHERAPY FOR SOCIAL PHOBIA
Walker et al (2000)113 Blomhoff et al (2001)70 Hair et al (2000)114 Kumar et al (1999)115 Connor et al (1998)99 Versiani et al (1992)81 RCT=randomized controlled trial; OL=off-label; wk=week; CGI-S=Clinical Global Impression-Severity; CGI-I=Clinical Global Impression-Improvement; MFQ=Marks Fear Questionnaire; MFQ-SP=Marks Fear Questionnaire-Social Phobia subscale; BSPS=Brief Social Phobia Scale; SPS=Social Phobia Scale; LSAS=Liebowitz Social Anxiety Scale; SADS=Social Anxiety and Distress Scale; SDS=Sheehan Disability Scale; MSPS-S=Marks-Sheehan Main Phobia Severity Scale.
Volume 8 – Number 8 (Suppl 1) CNS Spectrums - August 2003 Long-Term Treatment of Social Phobia
Concomitant with this finding was a continuing improve- significance not reported). However, moclobemide was toler- ment in LSAS score and corresponding improvements in ated much better than phenelzine. All patients withdrawn disability. Long-term treatment with paroxetine (up to from active treatment at week 16 had relapsed by week 9 months) was well tolerated.
24 (mean score on all parameters had increased).
Monoamine Oxidase Inhibitors and Reversible
The only other published long-term clinical trial data on Inhibitors of Monoamine Oxidase Type A
MAOIs and RIMAs in social phobia come from open-label Phenelzine and moclobemide were compared in a double- studies of tranylcypromine90 and moclobemide.85 A follow-up blind, placebo-controlled, parallel-group trial that enrolled study of moclobemide examined its efficacy in the long-term 78 patients with social phobia and lasted more than treatment of social phobia. A total of 93 outpatients suffer- 24 weeks.81 Patients were treated initially for 8 weeks (mean ing from severe generalized or circumscribed social phobia doses at end of the first phase: 580.7 mg/day moclobemide were initially treated with moclobemide. Responders (59/93) and 67.5 mg/day phenelzine), after which CGI-I nonrespon- continued treatment for 2 years, entered a 1-month drug-free ders were withdrawn from the study. All other patients period, and then completed another 2-year moclobemide (n=45) continued with the same treatment for an additional treatment period. During the drug-free period, 88% of 8 weeks (mean doses at end of the second phase: patients deteriorated. However, all patients responded again 582.3 mg/day moclobemide and 69.3 mg/day phenelzine), in the next 2-year period. A post-study follow-up at after which placebo responders and relapsers (in any group) 6–24 months found that 63.2% of patients were almost were withdrawn from the study. In the final phase of the asymptomatic or had only mild symptoms.
trial, one half of the responding patients continued on active treatment while the other half were switched to placebo. At Long-term treatment data from clinical studies of clon- week 4 and onward, both phenelzine and moclobemide were azepam are limited to the findings of a small study assessing clinically and statistically more effective than placebo. At whether clonazepam could safely and effectively be adminis- week 16, 82% of moclobemide- and 91% of phenelzine- tered over a period of 11 months.99 In the first 6 months of treated patients were almost asymptomatic (based on the trial, 56 patients received open-label clonazepam treat- changes in CGI-I, SDS, and Social Phobia Scale [SPS] ment (up to 2.5 mg/day), which was associated with signifi- scores) compared with 43% in the placebo group (statistical cant decreases in a number of social and disability assessment Main Efficacy Variables 20-wk RCT and then 4% discontinuation 20-wk re-randomized 12% discontinuation Side effects more frequent and severe in phenelzine group than in other two groups Note: Venlafaxine extended release (XR) has since been approved for use in the United States for the treatment of social phobia.
Van Ameringen M, Allgulander C, Bandelow B, et al. CNS Spectrums. Vol 8, No 8 (suppl 1). 2003.
Volume 8 – Number 8 (Suppl 1) CNS Spectrums - August 2003 M. Van Ameringen, C. Allgulander, B. Bandelow, et al
scores. Patients who responded to treatment were subse- avoidance-learning and negative cognitive patterns associ- quently randomized to receive either continuation treatment ated with social phobia by exposing the patient to the feared with clonazepam for another 5 months (n=17) or undergo situation. Various treatment programs, including education discontinuation treatment (using a slow-taper method) with and exposure instructions, have been developed.118 a double-blind placebo substitution (n=19). Relapse rates Many experts consider exposure to feared stimuli, were 0% and 21.1% in the continuation and discontinuation objects, and situations to be an essential component of groups, respectively, and were significant using Kaplan-Meier effective treatment for phobic disorders.119 Although expo- survival analysis with extended time to relapse for continua- sure can be conducted in the patient's imagination or in tion treatment (P<.05). The results of clinical efficacy scales vivo (ie, real-life exposure to the actual situation), studies at endpoint suggested a modest but statistically significant show that in vivo exposure is more effective.119 Once expo- difference in favor of continuing treatment over tapered sure is initiated, cognitive restructuring can provide withdrawal, supportive of benefit for long-term clonazepam patients with a context in which to interpret their experi- treatment in social phobia.
ences during exposure practices.119 A study designed to determine the necessity for the cogni- tive-restructuring component of CBGT compared the ther- A number of psychotherapeutic treatment modalities have apy to an exposure-based treatment without formal cognitive been investigated for the treatment of social phobia, including restructuring. Both methods were superior to a wait-list con- exposure therapy, cognitive therapy, interpersonal psychother- trol, with some advantage of exposure therapy over CBGT.120 apy, group social skills training, and supportive therapy.
Two controlled studies have reported some advantage of The maintenance and treatment-free follow-up phases combining exposure and cognitive therapy over the separate (each lasting 6 months) of the trial by Heimberg and col- treatments.121,122 A more recent trial compared 27 social pho- leagues88 provide a valuable perspective of long-term treat- bia patients receiving CBT (individual cognitive therapy fol- ment outcome with CBGT versus phenelzine.116 There was a lowed by group social skills training) with 28 patients trend for greater relapse with phenelzine than with CBGT receiving supportive therapy (ST).123 Sustained improve- during treatment-free follow-up, despite the fact that ment was observed in both groups at follow-up (up to phenelzine-treated patients showed greater improvement 72 weeks) but overall, CBT was more effective than ST in upon entering the maintenance phase of the study and, improving social phobia measures.
among nonrelapsing completers, greater improvement at As mentioned earlier, Gelernter and colleagues94 com- endpoint. Thus, the authors suggest that CBGT may lead to pared CBGT with phenelzine, alprazolam, and pill placebo a greater likelihood of maintaining response after treatment in 65 patients who were given directions for self-directed has terminated than does phenelzine, perhaps due to the exposure to feared stimuli, and reported a substantial coping skills explicitly provided by CBT. It should be noted, improvement in all treatment groups. However, the sample however, that these data are based on a very small sample size in this study was probably too small (eg, 13 evaluable size (N=28) in the follow-up phase.116 patients in the phenelzine group) to detect differences among treatments. A study by Heimberg and colleagues88 that enrolled 133 patients compared CBGT, educational Surgical sympathectomy has been suggested to relieve supportive group therapy, phenelzine, and pill placebo.
symptoms of social phobia, especially blushing, and might CBGT was found to be superior to pill placebo and educa- conceivably be a viable alternative when more conservative tional group therapy, but slightly inferior to phenelzine on treatments prove to be insufficient after prolonged attempts.
some measures.
A prospective, uncontrolled study enrolling 260 patients with DSM-IV social phobia indicated that a reversible bilat- COMBINED PHARMACOLOGY
eral sympathetic block helps relieve the physical signs of social phobia, such as hand or facial sweating, blushing, and In a long-term study of social phobia, 387 patients meet- rapid heartbeat, as well as reduce other social phobia symp- ing DSM-IV criteria for generalized social phobia in a pri- toms, such as avoidance behavior.14 Patient satisfaction with mary care setting were randomized to sertraline with and the procedure was reported to exceed 90%. However, proper without exposure therapy.70 Patients were randomized to one controlled studies are still needed to verify these results.
of four groups: sertraline 50–150 mg+exposure therapy, ser- traline alone, placebo+exposure therapy, or placebo alone for 24 weeks. Response rates (defined as CGI-S score ≤3, CGI-I The efficacy of CBT in social phobia has been demon- score ≤2, and ≥50% reduction in the self-assessed SPS) in strated in several controlled trials (using waiting lists or other the four groups were 46%, 40%, 33%, and 24%, therapies as controls) and meta-analyses.117 However, in spite respectively.70 Differences in the sertraline-alone and sertra- of its proven efficacy in treating this disorder, CBT is line+exposure groups over the placebo-alone group were sta- not readily available, primarily for reasons of cost and short- tistically significant (P<.001), and exposure therapy alone ages in appropriately trained therapists. CBT targets the was also statistically better than placebo (P=.04). Although Volume 8 – Number 8 (Suppl 1) CNS Spectrums - August 2003 Long-Term Treatment of Social Phobia
no significant difference was observed between patients who received exposure therapy versus those who did not Social phobia is a complex syndrome that often begins (P=.140), the addition of exposure therapy to sertraline during adolescence and has a chronic, unremitting course.
resulted in a trend toward significance (P=.059).
Social phobia is a chronic illness associated with significant A recent article124 provides 24 weeks of additional fol- levels of distress and social and occupational disability. Its low-up evidence to the long-term sertraline versus expo- detrimental effects may be compounded by frequently pre- sure therapy study. This study represents the largest sent comorbid psychiatric conditions, such as depression, reported sample that evaluates the maintenance of treat- substance abuse, and other anxiety disorders. Although ment effect post-discontinuation. The study found an social phobia has been identified as the most prevalent of the advantage for exposure therapy alone compared to sertra- anxiety disorders, it is still poorly recognized in clinical prac- line in terms of response maintenance. Despite the large tice. However, it is receiving more attention with the sample size, interpretation of the results is complicated by increasing availability of effective treatments, particularly the difficulty of conducting such an ambitious long-term with the evidence that is accumulating for the efficacy and treatment and follow-up study in the primary care setting.
safety of SSRIs in long-term treatment. Recognition and Specifically, treatment assignment was not blinded, 19% treatment of social phobia not only benefits the patient's of patients in the exposure therapy group were treated quality of life, but also decreases the social burden that this with SSRIs during follow-up, there was significantly disorder places on society.
greater attrition in the exposure-alone group (n=11) com- Based on available data, SSRIs are the drugs of choice pared to the sertraline-alone group (n=10), and finally, in social phobia. While only paroxetine has a Food and relapse was not formally assessed. For these reasons, Drug Administration-approved indication for social pho- whether the modest advantage of exposure therapy is a bia, other drugs appear to be widely used off-label.
genuine treatment effect is uncertain and awaits confirma- Sertraline, paroxetine, clonazepam, phenelzine, and tion by additional studies.
moclobemide have each been studied in long-term ran- The relative efficacies of clonazepam and CBGT in the domized clinical trials, and moclobemide has an indication treatment of social phobia have also been compared. In a for social phobia in several European countries. In addi- 12-week trial, patients were randomly assigned to clon- tion, although CBT has demonstrated efficacy in random- azepam or CBGT. All patients improved significantly, with ized clinical trials, it has limited availability.
little difference between the two treatment groups.125 Discontinuation of effective short-term therapy fre- quently results in relapse. Consequently, serious considera- OVERVIEW OF AVAILABLE
tion must be given to long-term treatment for a minimum of 12 months. Future controlled trials of long-term treatment Several publications have included guidelines and recom- are needed, including comparisons of pharmacotherapy, psy- mendations for the treatment of social phobia,79,126,127 but the chotherapy, and combination trials, to further clarify the most comprehensive recent treatment guidelines are most appropriate treatment strategy for this illness. CNS
included in a consensus statement by the International Consensus Group on Depression and Anxiety.16 The guide- lines recommend SSRIs as first-line therapy and suggest that 1. Stein MB, Torgrud LJ, Walker JR. Social phobia symptoms, subtypes, and sever- treatment may have to be continued for several months to ity: findings from a community survey. Arch Gen Psychiatry. 2000;57:1046-1052.
consolidate response and achieve full remission. They also 2. Kessler RC, Stang P, Wittchen HU, Stein M, Walters EE. Lifetime co- suggest that if treatment is effective it should be continued morbidities between social phobia and mood disorders in the US National Comorbidity Survey. Psychol Med. 1999;29:555-567.
for a minimum of 12 months. Long-term treatment is indi- 3. Wittchen HU, Stein MB, Kessler RC. Social fears and social phobia in a cated if symptoms are unresolved, the patient has a comorbid community sample of adolescents and young adults: prevalence, risk fac- condition or history of relapse, or there was an early onset of tors and co-morbidity. Psychol Med. 1999;29:309-323.
social phobia.
4. Katzelnick DJ, Greist JH. Social anxiety disorder: an unrecognized prob- The consensus group highlights the fact that there are lem in primary care. J Clin Psychiatry. 2001;62(suppl 1):11-15.
insufficient clinical data on which to base a recommenda- 5. Ballenger JC, Davidson JR, Lecrubier Y, et al. Consensus statement on tion for the management of patients who fail to respond to transcultural issues in depression and anxiety from the International Consensus Group on Depression and Anxiety. J Clin Psychiatry. an SSRI and calls for further research on the effectiveness of augmentation therapy with clonazepam or switching to 6. Magee WJ, Eaton WW, Wittchen HU, McGonagle KA, Kessler RC.
Agoraphobia, simple phobia, and social phobia in the National The guidelines also highlight evidence for the effective- Comorbidity Survey. Arch Gen Psychiatry. 1996;53:159-168.
ness of exposure-based strategies of CBT in social phobia, 7. Stein MB, Walker JR, Forde DR. Setting diagnostic thresholds for social but note that this evidence comes from relatively small trials.
phobia: considerations from a community survey of social anxiety. Am J The consensus group notes the need for comparative studies 8. Essau CA, Conradt J, Petermann F. [Frequency and comorbidity of social of psychosocial treatment and pharmacotherapy, as well as anxiety and social phobia in adolescents. Results of a Bremen adolescent studies evaluating their combined efficacy.
study.] Fortschr Neurol Psychiatr. 1998;66:524-530. German Volume 8 – Number 8 (Suppl 1) CNS Spectrums - August 2003 M. Van Ameringen, C. Allgulander, B. Bandelow, et al
9. Kilts CD, Gross R, Newport J, et al. A functional neuroimaging study of 35. Kessler RC, McGonagle KA, Zhao S, et al. Lifetime and 12-month social anxiety and its treatment. Int J Neuropsychopharmacol. prevalence of DSM-III-R psychiatric disorders in the United States.
Results from the National Comorbidity Survey. Arch Gen Psychiatry. 10. Kessler RC. The epidemiology, natural history, and pharmacoeconomics of social anxiety and PTSD. Eur Neuropsychopharmacol. 2000;10:S185-S186. 36. Wittchen HU, Nelson CB, Lachner G. Prevalence of mental disorders 11. Lecrubier Y, Weiller E. Comorbidities in social phobia. Int Clin and psychosocial impairments in adolescents and young adults. Psychol 12. Diagnostic and Statistical Manual of Mental Disorders. 4th ed.
37. Diagnostic and Statistical Manual of Mental Disorders. 3rd ed. Washington, Washington, DC: American Psychiatric Association; 1994.
DC: American Psychiatric Association; 1980. 13. The International Classification of Diseases. 10th revision. Geneva, 38. Heimberg RG, Stein MB, Hiripi E, Kessler RC. Trends in the preva- Switzerland: World Health Organization; 1992.
lence of social phobia in the United States: a synthetic cohort analysis 14. Burke KC, Burke JD, Regier DA, Rae DS. Age at onset of selected men- of changes over four decades. Eur Psychiatry. 2000;15:29-37.
tal disorders in five community populations. Arch Gen Psychiatry. 39. Wittchen HU, Fehm L. Epidemiology, patterns of comorbidity, and associated disabilities of social phobia. Psychiatr Clin North Am. 15. Lopez-Ibor JJ, Ayuso Gutierrez JL. Social phobia: a debilitating disease that needs treatment. Int Clin Psychopharmacol. 1997;12(suppl 6):S11-S16.
40. Kessler RC. Anxiety, mood, and substance abuse disorders: patterns and 16. Ballenger JC, Davidson JR, Lecrubier Y, et al. Consensus statement on correlates of comorbidity. Eur Neuropsychopharmacol. 1999;9:S142-S144. social anxiety disorder from the International Consensus Group on 41. Stein MB, Chartier MJ, Hazen AL, et al. A direct-interview family Depression and Anxiety. J Clin Psychiatry. 1998;59(suppl 17):54-60.
study of generalized social phobia. Am J Psychiatry. 1998;155:90-97.
17. Brunello N, den Boer JA, Judd LL, et al. Social phobia: diagnosis and 42. Katzelnick DJ, Kobak KA, DeLeire T, et al. Impact of generalized social epidemiology, neurobiology and pharmacology, comorbidity and treat- anxiety disorder in managed care. Am J Psychiatry 2001;158:1999-2007.
ment. J Affect Disord. 2000;60:61-74.
43. Katzelnick DJ, Kobak KA, Helstad CP, et al. The direct and indirect 18. Bech P, Angst J. Quality of life in anxiety and social phobia. Int Clin costs of social phobia in managed care patients. Paper presented at: 19th Annual Conference of the Anxiety Disorders Association of America; 19. Wittchen HU, Beloch E. The impact of social phobia on quality of life.
March 1999; San Diego, Calif.
Int Clin Psychopharmacol. 1996;11(suppl 3):15-23.
44. Westenberg HG. Neurobiology of anxiety disorders: focus on neu- 20. Wittchen HU. Disability associated with social anxiety disorder. Eur roimaging. Eur Neuropsychopharmacol. 2000;10:S189. Neuropsychopharmacol. 1999;9:S313. 45. Ameringen MV, Mancini C, Farvolden P, Oakman J. Drugs in develop- 21. Schneier FR, Johnson J, Hornig CD, Liebowitz MR, Weissman MM.
ment for social anxiety disorder: more to social anxiety than meets the Social phobia. comorbidity and morbidity in an epidemiologic sample.
SSRI. Expert Opin Investig Drugs. 2000;9:2215-2231.
Arch Gen Psychiatry. 1992;49:282-288.
46. Van Ameringen M, Mancini C, Farvolden P, et al. The neurobiology of 22. Kessler RC, Stein MB, Berglund P. Social phobia subtypes in the social phobia: from pharmacotherapy to brain imaging. Curr Psychiatry National Comorbidity Survey. Am J Psychiatry. 1998;155:613-619.
47. Nutt DJ, Bell CJ, Malizia AL. Brain mechanisms of social anxiety disor- 23. Mennin DS, Fresco DM, Heimberg RG, Schneier FR, Davies SO, der. J Clin Psychiatry. 1998;59(suppl 17):4-11.
Liebowitz MR. Screening for social anxiety disorder in the clinical set- 48. van der Linden GJ, van Heerden B, Warwick J, et al. Functional brain ting: using the Liebowitz Social Anxiety Scale. J Anxiety Disord. imaging and pharmacotherapy in social phobia: single photon emission computed tomography before and after treatment with the selective 24. Norton GR, McLeod L, Guertin J, Hewitt PL, Walker JR, Stein MB.
serotonin reuptake inhibitor citalopram. Prog Neuropsychopharmacol Biol Panic disorder or social phobia: which is worse? Behav Res Ther. 49. den Boer JA, Bosker FJ, Slaap BR. Neurobiology of social anxiety disor- 25. Reich J, Noyes R, Yates W. Anxiety symptoms distinguishing social pho- der. Eur Neuropsychopharmacol. 1999;9:S165.
bia from panic and generalized anxiety disorders. J Nerv Ment Dis. 50. Schneier FR, Liebowitz MR, Abi-Dargham A, Zea-Ponce Y, Lin SH, Laruelle M. Low dopamine D(2) receptor binding potential in social 26. Kinrys G, Bostwick JM. Pharmacological dissection of panic disorder phobia. Am J Psychiatry. 2000;157:457-459. and social phobia: using valproic acid as a clinical tool. Int J 51. Stein MB, Tancer ME, Gelernter CS, Vittone BJ, Uhde TW. Major Neuropsychopharmacol. 2000;3:S289-S290. depression in patients with social phobia. Am J Psychiatry. 27. Amies PL, Gelder MG, Shaw PM. Social phobia: a comparative clinical study. Br J Psychiatry. 1983;142:174-179.
52. van Ameringen M, Mancini C, Styan G, Donison D. Relationship of 28. Moutier CY, Stein MB. The history, epidemiology, and differential diag- social phobia with other psychiatric illness. J Affect Disord. 1991;21:93-99.
nosis of social anxiety disorder. J Clin Psychiatry. 1999;60(suppl 9):4-8.
53. Lepine JP, Pelissolo A. Social phobia and alcoholism: a complex rela- 29. Stein DJ. Diagnostic dilemmas in social anxiety disorder and posttrau- tionship. J Affect Disord. 1998;50(suppl 1):S23-S28.
matic stress. Eur Neuropsychopharmacol. 2000;10:S186-187. 54. Wittchen HU. Nicotine consumption in mental disorders: a clinical 30. Black B, Uhde TW. Elective mutism as a variant of social phobia. J Am epidemiological perspective. Eur Neuropsychopharmacol. 2000;10:S119.
Acad Child Adolesc Psychiatry. 1992;31:1090-1094.
55. Nascimento I, Nardi AE, Valenca AM, et al. Cigarette smoking and 31. Kleinknecht RA, Dinnel DL, Kleinknecht EE, Hiruma N, Harada N.
anxiety disorders. Int J Neuropsychopharmacol. 2000;3:S284.
Cultural factors in social anxiety: a comparison of social phobia symp- 56. Liebowitz MR. Social phobia. In: Klein D, ed. Modern Problems of toms and Taijin kyofusho. J Anxiety Disord. 1997;11:157-177.
Pharmacopsychiatry: Anxiety. Basel, Switzerland: Karger; 1987:141-173.
32. Kirmayer LJ. The place of culture in psychiatric nosology: Taijin 57. Davidson JR, Potts NL, Richichi EA, et al. The Brief Social Phobia kyofusho and DSM-III-R. J Nerv Ment Dis. 1991;179:19-28.
Scale. J Clin Psychiatry. 1991;52(suppl):48-51.
33. Stein MB, Chavira DA. Subtypes of social phobia and comorbidity with 58. Davidson JR, Miner CM, de Veaugh-Geiss J, Tupler LA, Colket JT, depression and other anxiety disorders. J Affect Disord. 1998;50(suppl Potts NL. The Brief Social Phobia Scale: a psychometric evaluation.
Psychol Med. 1997;27:161-166.
34. Stein MB. Coming face-to-face with social phobia. Am Fam Physician. 59. Marks IM, Mathews AM. Brief standard self-rating for phobic patients.
Behav Res Ther. 1979;17:263-267.
Volume 8 – Number 8 (Suppl 1) CNS Spectrums - August 2003 Long-Term Treatment of Social Phobia
60. Cattalani M, Faraone SV, Tsuang MT. The year in psychiatry. Praxis 82. Noyes R, Moroz G, Davidson JR, et al. Moclobemide in social phobia: a Post. January 24, 2001.
controlled dose-response trial. J Clin Psychopharmacol. 1997;17:247-254.
61. Katzelnick DJ, Kobak KA, Greist JH, Jefferson JW, mantle JM, Serlin 83. Schneier FR, Goetz D, Campeas R, Fallon B, Marshall R, Liebowitz MR.
RC. Sertraline for social phobia: a double-blind, placebo-controlled Placebo-controlled trial of moclobemide in social phobia. Br J crossover study. Am J Psychiatry. 1995;152:1368-1371.
62. Mancini C, van Ameringen M. Paroxetine in social phobia. J Clin 84. Versiani M, Nardi AE, Mundim FD, Pinto S, Saboya E, Kovacs R. The long-term treatment of social phobia with moclobemide. Int Clin 63. Stein MB, Chartier MJ, Hazen AL, et al. Paroxetine in the treatment of generalized social phobia: open-label treatment and double-blind place- 85. Diagnostic and Statistical Manual of Mental Disorders. 3rd ed rev.
bo-controlled discontinuation. J Clin Psychopharmacol. 1996;16:218-222.
American Psychiatric Association; 1980.
64. Stein MB, Liebowitz MR, Lydiard RB, Pitts CD, Bushnell W, Gergel I.
86. van Vliet IM, den Boer JA, Westenberg HG. Psychopharmacological Paroxetine treatment of generalized social phobia (social anxiety disor- treatment of social phobia: clinical and biochemical effects of bro- der): a randomized controlled trial. JAMA. 1998;280:708-713.
faromine, a selective MAO-A inhibitor. Eur Neuropsychopharmacol 65. Allgulander C. Paroxetine in social anxiety disorder: a randomized placebo-controlled study. Acta Psychiatr Scand. 1999;100:193-198.
87. Fahlen T, Nilsson HL, Borg K, Humble M, Pauli U. Social phobia: the 66. Baldwin D, Bobes J, Stein DJ, Scharwachter I, Faure M. Paroxetine in clinical efficacy and tolerability of the monoamine oxidase-A and sero- social phobia/social anxiety disorder. Randomised, double-blind, place- tonin uptake inhibitor brofaromine: a double-blind placebo controlled bo-controlled study. Paroxetine Study Group. Br J Psychiatry. study. Acta Psychiatr Scand 1995;92:351-358.
88. Heimberg RG, Liebowitz MR, Hope DA, et al. Cognitive behavioral 67. Lydiard RB, Bobes J. Therapeutic advances: paroxetine for the treat- group therapy vs phenelzine therapy for social phobia: 12-week out- ment of social anxiety disorder. Depress Anxiety. 2000;11:99-104.
come. Arch Gen Psychiatry. 1998;55:1133-1141.
68. van Ameringen M, Lane RM, Walker JR, et al. Sertraline treatment of 89. Liebowitz MR, Schneier F, Campeas R, et al. Phenelzine vs atenolol in generalized social phobia: a 20-week, double-blind, placebo-controlled social phobia. A placebo-controlled comparison. Arch Gen Psychiatry. study. Am J Psychiatry. 2001;158:275-281.
69. Liebowitz MR, DeMartinis N, Weihs KL, Chung H, Fayyad R, Clary 90. Versiani M, Mundim FD, Nardi AE, Liebowitz MR. Tranylcypromine in CM. Results from a randomized, double-blind, multicenter trial of ser- social phobia. J Clin Psychopharmacol. 1988;8:279-283.
traline in the treatment of moderate-to-severe social phobia (social anx- 91. Sareen L, Stein M. A review of the epidemiology and approaches to the iety disorder). Poster presented at: 40th Annual Meeting of the treatment of social anxiety disorder. Drugs. 2000;59:497-509.
American College of Neuropsychopharmacology; December 9-13, 2001; 92. Liebowitz MR, Gorman J, Fyer A, et al. Psychopharmacological treat- Waikaloa, Hawaii.
ment of social phobia. Psychopharmacol Bull. 1985;21:610-614.
70. Blomhoff S, Haug TT, Hellstrom K, et al. Randomised control general 93. Liebowitz MR, Fyer AJ, Gorman JM, Campeas R, Levin A. Phenelzine practice trial of sertraline, exposure and combined treatment in gener- in social phobia. alised social phobia. J Clin Psychopharmacol. 1986;6:93-98.
Br J Psychiatry. 2001;179:23-30.
71. van Vliet IM, den Boer JA, Westenberg HG. Psychopharmacological 94. Gelernter CS, Uhde TW, Cimbolic P, et al. Cognitive-behavioral and treatment of social phobia; a double blind placebo controlled study with pharmacological treatments of social phobia: a controlled study. Arch fluvoxamine. Psychopharmacology (Berl). 1994;115:128-134.
Gen Psychiatry. 1991;48:938-945.
72. Stein MB, Fyer AJ, Davidson JR, Pollack MH, Wiita B. Fluvoxamine 95. Liebowitz MR, Gorman JM, Fyer AJ, et al. Pharmacotherapy of social treatment of social phobia (social anxiety disorder): a double-blind, phobia: an interim report of a placebo-controlled comparison of placebo-controlled study. Am J Psychiatry. 1999;156:756-760.
phenelzine and atenolol. J Clin Psychiatry. 1988;49:252-257.
73. Lepola U, Koponen H, Leinonen E. Citalopram in the treatment of social 96. Liebowitz MR, Schneier F, Campeas R, et al. Phenelzine and atenolol in phobia: a report of three cases. Pharmacopsychiatry. 1994;27:186-188.
social phobia. Psychopharmacol Bull. 1990;26:123-125.
74. Varia IM, Cloutier CA, Doraiswamy PM. Treatment of social anxiety 97. Munjack DJ, Baltazar PL, Bohn PB, Cabe DD, Appleton AA.
disorder with citalopram. Prog Neuropsychopharmacol Biol Psychiatry. Clonazepam in the treatment of social phobia: a pilot study. J Clin 75. Bouwer C, Stein DJ. Use of the selective serotonin reuptake inhibitor 98. Davidson JR, Potts N, Richichi E, et al. Treatment of social phobia with citalopram in the treatment of generalized social phobia. J Affect Disord. clonazepam and placebo. J Clin Psychopharmacol. 1993;13:423-428.
99. Connor KM, Davidson JR, Potts NL, et al. Discontinuation of clon- 76. van Ameringen M, Mancini C, Streiner DL. Fluoxetine efficacy in azepam in the treatment of social phobia. J Clin Psychopharmacol. social phobia. J Clin Psychiatry. 1993;54:27-32.
77. Fairbanks JM, Pine DS, Tancer NK, et al. Open fluoxetine treatment of 100. Sutherland SM, Tupler LA, Colket JT, Davidson JR. A 2-year follow- mixed anxiety disorders in children and adolescents. J Child Adolesc up of social phobia. Status after a brief medication trial. J Nerv Ment 78. Pallanti S, Quercioli L, Rossi A, Pazzagli A. The emergence of social 101. Simpson HB, Schneier FR, Campeas RB, et al. Imipramine in the phobia during clozapine treatment and its response to fluoxetine aug- treatment of social phobia. J Clin Psychopharmacol. 1998;18:132-135.
mentation. J Clin Psychiatry. 1999;60:819-823.
102. Altamura AC, Pioli R, Vitto M, Mannu P. Venlafaxine in social pho- 79. van der Linden GJ, Stein DJ, van Balkom AJ. The efficacy of the selec- bia: a study in selective serotonin reuptake inhibitor non-responders.
tive serotonin reuptake inhibitors for social anxiety disorder (social Int Clin Psychopharmacol. 1999;14:239-245.
phobia): a meta-analysis of randomized controlled trials. Int Clin 103. Ninan PT. Use of venlafaxine in other psychiatric disorders. Depress Anxiety. 2000;12(suppl 1):90-94.
80. The International Multicenter Clinical Trial Group on Moclobemide in 104. Van Ameringen M, Mancini C, Oakman JM. Nefazodone in social Social Phobia. Moclobemide in social phobia: a double-blind, placebo-con- phobia. J Clin Psychiatry. 1999;60:96-100.
trolled clinical study. Eur Arch Psychiatry Clin Neurosci. 1997;247:71-80.
105. van Vliet IM, van Veen F, Westenberg HG. Mirtazapine in social anx- 81. Versiani M, Nardi AE, Mundim FD, Alves AB, Liebowitz MR, Amrein iety disorder. Int J Neuropsychopharmacol. 2000;3:S283. R. Pharmacotherapy of social phobia: a controlled study with moclobe- 106. Baldwin D. Comprehensive treatment of social anxiety disorder. Eur mide and phenelzine. Br J Psychiatry. 1992;161:353-360.
Volume 8 – Number 8 (Suppl 1) CNS Spectrums - August 2003 M. Van Ameringen, C. Allgulander, B. Bandelow, et al
107. Turner SM, Beidel DC, Jacob RG. Social phobia: a comparison of 118. Haug TT, Hellstrom K, Blomhoff S, Humble M, Madsbu HP, Wold JE.
behavior therapy and atenolol. J Consult Clin Psychol. 1994;62:350-358.
The treatment of social phobia in general practice. Is exposure therapy 108. Falloon IR, Lloyd GG, Harpin RE. The treatment of social phobia.
feasible? Fam Pract. 2000;17:114-118.
Real-life rehearsal with nonprofessional therapists. J Nerv Ment Dis. 119. Antony MM, Swinson RP. Exposure-based strategies and social skills training. In: Antony MM, Swinson RP, eds. Phobic Disorders and Panic 109. van Vliet IM, den Boer JA, Westenberg HG, Pian KL. Clinical effects in Adults: A Guide to Assessment and Treatment. 1st ed. Washington, of buspirone in social phobia: a double-blind placebo-controlled study.
DC: American Psychological Association; 2000:191-238.
J Clin Psychiatry. 1997;58:164-168.
120. Hope DA, Heimberg RG, Bruch MA. Dismantling cognitive-behav- 110. Schneier FR, Saoud JB, Campeas R, et al. Buspirone in social phobia.
ioral group therapy for social phobia. Behav Res Ther. 1995;33:637-650.
J Clin Psychopharmacol. 1993;13:251-256.
121. Butler G, Cullington A, Munby M, Amies P, Gelder M. Exposure and 111. Pande AC, Davidson JR, Jefferson JW, et al. Treatment of social pho- anxiety management in the treatment of social phobia. J Consult Clin bia with gabapentin: a placebo-controlled study. J Clin 122. Mattick RP, Peters L. Treatment of severe social phobia: effects of 112. Feltner DE, Pollack MH, Davidson JR, et al. A placebo-controlled guided exposure with and without cognitive restructuring. J Consult study of pregabalin treatment of social phobia: outcome and predictors Clin Psychol. 1988;56:251-260.
of response. Eur Neuropsychopharmacol. 2000;10:S344-S345. 123. Cottraux J, Note I, Albuisson E, et al. Cognitive behavior therapy ver- 113. Walker JR, Van Ameringen MA, Swinson R, et al. Prevention of sus supportive therapy in social phobia: a randomized controlled trial.
relapse in generalized social phobia: results of a 24-week study in responders to 20 weeks of sertraline treatment. J Clin Psychopharmacol. 124. Haug TT, Blomhoff S, Hellstrom K, et al. Exposure therapy and sertra- line in social phobia: 1-year follow-up of a randomised controlled trial.
114. Hair T, Castrogiovanni P, Domenech J, et al. Short-term efficacy of Br J Psychiatry. 2003;182:312-318.
paroxetine in social anxiety disorder is maintained after long-term 125. Otto MW, Pollack MH, Gould RA, Worthington JJ 3rd, McArdle ET, treatment. Int J Neuropsychopharmacol. 2000;3:S227. Rosenbaum JF. A comparison of the efficacy of clonazepam and cogni- 115. Kumar R, Pitts C, Carpenter D. Response to paroxetine is maintained tive-behavioral group therapy for the treatment of social phobia. during continued treatment in patients with social anxiety disorder.
J Anxiety Disord 2000;14:345-358.
Eur Neuropsychopharmacol. 1999;9:S312-S313. 126. Dandrifosse AC, Ansseau M. [Guidelines for the treatment of anxiety 116. Liebowitz MR, Heimberg RG, Schneier FR, et al. Cognitive-behav- disorders in adults.] Rev Med Liege. 2000;55:400-408. French ioral group therapy versus phenelzine in social phobia: long-term out- 127. Ballenger JC. Clinical guidelines for establishing remission in patients come. Depress Anxiety. 1999;10:89-98.
with depression and anxiety. J Clin Psychiatry. 1999;60(suppl 22):29-34.
117. Heimberg RG. Current status of psychotherapeutic interventions for social phobia. J Clin Psychiatry. 2001;62(suppl 1):36-42.
Volume 8 – Number 8 (Suppl 1) CNS Spectrums - August 2003 WCA Recommendations for the Long-Term
Treatment of Generalized Anxiety Disorder
By Christer Allgulander, MD, Borwin Bandelow, MD, Eric Hollander, MD, Stuart A. Montgomery, MD, David J. Nutt, MD, FRCP, FRCPsych, FMedSci, Ahmed Okasha, MD, PhD, FRCP, FRCPsych, FACP, Mark H. Pollack, MD, Dan J. Stein, MD, PhD, and Richard P. Swinson, MD, FRCPC, FRCPsych, DPM in other acute trials. Paroxetine is the first of the selective sero- tonin reuptake inhibitors (SSRIs) to receive US approval for the • Both pharmacotherapy and, to a lesser extent, psy- treatment of GAD. Paroxetine demonstrated superiority to chotherapy have shown benefit in the acute treatment placebo in short-term trials, and investigations into the use of other of generalized anxiety disorder (GAD).
SSRIs are ongoing. This suggests that other SSRIs, and serotonin • Buspirone, venlafaxine, and paroxetine are approved and noradrenaline reuptake inhibitors, are likely to be effective in for the treatment of GAD.
the treatment of GAD. Of the psychological therapies, cognitive- • Although benzodiazepines are commonly used for the behavioral therapy (CBT) shows the greatest benefit in treating acute treatment of GAD, their use may be limited due to GAD patients. Treatment gains after a 12-week course of CBT adverse effects, withdrawal symptoms, development of may be maintained for up to 1 year. Currently, no guidelines exist tolerance, and high relapse rates upon discontinuation. for the long-term treatment of GAD. • More research is needed to determine the usefulness of CNS Spectr 2003;8(suppl 1):53-61 the selective serotonin reuptake inhibitors in the treat-ment of GAD, and the best approach for the long-termtreatment of the disorder.
Generalized anxiety disorder (GAD) is a common, dis- abling psychiatric condition characterized by persistent, What are the current recommendations for the long-term treat- excessive worry and anxiety about events of daily life. GAD ment of generalized anxiety disorder (GAD)? GAD is a common is marked by the duration, frequency, and intensity of anxi- disorder with a lifetime prevalence of 4% to 7% in the general ety of a feared event being far out of proportion to the actual population. GAD is characterized by excessive, uncontrollable likelihood or impact of the event.1 The sufferer is unable to worry or anxiety about a number of events or activities that the control the worry, which is accompanied by various symp- individual experiences on more days than not over a 6-month toms, such as restlessness, fatigability, difficulty concentrat- period. Onset of GAD symptoms usually occurs during an indi- ing, muscle tension, irritability, sleep disturbances, and vidual's early twenties; however, high rates of GAD have also gastrointestinal distress. Affected individuals experience sig- been seen in children and adolescents. The clinical course of nificant disability in work and social functioning.
GAD is often chronic, with 40% of patients reporting illness last- GAD is frequently associated or comorbid with other ing >5 years. GAD is associated with pronounced functional mental disorders, such as major depression, dysthymia, panic impairment, resulting in decreased vocational function and disorder, and agoraphobia.2 The mean age of onset of GAD reduced quality of life. Patients with GAD tend to be high users of appears to be in the late teens or early twenties.3 The course outpatient medical care, which contributes significantly to health- of illness is usually chronic, with the severity of symptoms care costs. Currently, benzodiazepines and buspirone are pre- fluctuating over the patient's lifetime.
scribed frequently to treat GAD. Although both show efficacy in The diagnosis of GAD was first introduced in 1980 by the acute treatment trials, few long-term studies have been performed. American Psychiatric Association in the Diagnostic and Benzodiazepines are not recommended for long-term treatment of Statistical Manual of Mental Disorders, Third Edition (DSM- GAD, due to associated development of tolerance, psychomotor III).4 Prior to 1980, GAD and panic disorder were conceptu- impairment, cognitive and memory changes, physical dependence, alized as the core components of anxiety neurosis. The and a withdrawal reaction on discontinuation. The antidepressant recognition that GAD and panic, although co-occurring, are venlafaxine extended-release (XR) has received approval for the sufficiently distinct to be considered independent disorders treatment of GAD in the United States and many other coun- led to their separation in DSM-III.5 tries. Venlafaxine XR has demonstrated efficacy over placebo in Early clinical studies evaluating GAD, as described by two randomized treatment trials of 6 months' duration as well as DSM-III, found that the disorder usually occurred in the These recommendations are based on proceedings from the World Council of Anxiety meeting held September 11, 2000, in Pisa, Italy, and on guidelines and articles published in the medical literature through October 15, 2001.
Please direct all correspondence to: Christer Allgulander, MD, The Neurotec Department, Section of Psychiatry, M57 Huddinge, University Hospital, S-141 86 Huddinge Sweden; Tel: 46-8-585-85797; Fax: 46-8-585-85720; E-mail: [email protected].
Volume 8 – Number 8 (Suppl 1) CNS Spectrums - August 2003 Allgulander C, Bandelow B, Hollander E, et al
presence of another comorbid anxiety or mood disorder.
data were obtained on different patient populations. General Some investigators suggested that GAD might be better con- United States population surveys show that GAD has a life- ceptualized as a prodrome, residual, or severity marker than time prevalence of 4% to 7% and a 1-year prevalence of 1% as an independent disorder.5 This view subsequently became to 4%.2,10,11 Using strict DSM-IV criteria, a national survey of invalid due to changes that were made in the revised criteria more than 4,000 people in Germany found 12-month preva- of Diagnostic and Statistical Manual of Mental Disorders, Third lences of GAD and major depressive disorder (MDD) to be Edition-Revised6 (duration requirement extended from 1 1.5% and 8.3%, respectively.11,12 Higher rates of GAD were month to 6 months) and Diagnostic and Statistical Manual of found in women (2.7%) and the elderly (2.2%).11 In addition, Mental Disorders, Fourth Edition (DSM-IV)7 (change in the GAD has been found to be more common among people of definition of excessive worry and the required number of low socioeconomic status (SES), racial/ethnic minorities, and associated psychological symptoms).5 unmarried persons.5 Diagnosis of GAD can be made using either DSM-IV7 or The critical feature of GAD is excessive anxiety or International Classification of Diseases, Tenth Revision (ICD- worry occurring consistently on most days over a period ≥6 10)8 criteria. By definition, GAD is a chronic disorder (Table months. The individual finds these concerns, which may 19). The essential feature of GAD is excessive anxiety or worry often be related to common events or activities, difficult to (apprehensive expectation) about a number of events or activ- control.1 Worries may be related to job and career, ities, which occurs during a majority of the days within a finances, health of friends and family, safety of children, period ≥6 months.7 The anxiety and worry are accompanied everyday chores, or time management. The extent of by at least three additional symptoms from a list that includes worry is in excess of the actual likelihood or impact of the restlessness, being easily fatigued, difficulty in concentrating, event being targeted.
irritability, muscle tension, and disturbed sleep.7 Three major changes in the diagnostic criteria for GAD were made in the DSM-IV6,7: TABLE 1. DSM-IV CRITERIA FOR THE
•The definition of excessive anxiety and worry was DIAGNOSIS OF GAD9
•A new criterion was added, stating that the worry must Excessive anxiety and worry be difficult to control •The associated symptom criterion was focused, deleting Difficult to control most of the previous autonomic symptoms and leaving only the symptoms that reflect hypervigilance and motor Symptoms occur more days than not for 6 months tension (ie, restlessness and difficulty in concentrating) Significant distress or social impairment At least three ancillary symptoms Due to the frequent overlap of anxiety and somatic symp- Ancillary Symptoms toms across different diagnostic syndromes, a variety of psy- chiatric diagnoses should be considered in patients presenting with GAD symptoms. These include social pho- bia, depression, substance abuse, and personality disorders.
Poor concentration GAD must be distinguished from an anxiety disorder due to a general medical condition—the diagnosis of which is based on history, laboratory findings, or physical examination—or a substance-induced anxiety disorder in Sleep disturbance which a substance is judged to be etiologically related to the anxiety disorder.7 When another Axis I disorder is pre- sent, an additional diagnosis of GAD should be made only Focus of anxiety/worry not another disorder (eg, panic) when the focus of anxiety and worry is unrelated to the Not part of mood disorder, psychotic disorder, or pervasive other disorder, not restricted, for example, to panic attack developmental disorder (panic disorder), being embarrassed in public (social pho- bia), or being contaminated (obsessive-compulsive disor- Not substance related DSM-IV=Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition; GAD=generalized anxiety disorder.
The definition of GAD has changed considerably over the Allgulander C, Bandelow B, Hollander E, et al. CNS Spectrums. Vol 8, No last 2 decades, and as a consequence, most epidemiological 8 (suppl 1). 2003. Volume 8 – Number 8 (Suppl 1) CNS Spectrums - August 2003 Long-Term Treatment of Generalized Anxiety Disorder
CLINICAL COURSE, CHRONICITY, AND
respectively. A study conducted in France showed that GAD results in a substantial loss of productivity even in the The clinical course of GAD is chronic and may be either absence of a comorbid psychiatric disorder.16 constant or fluctuating, with spontaneous improvements and GAD has a pronounced negative effect on quality of life relapses. The mean age of onset appears to be in the late and is associated with significant psychosocial impairment, teens to early twenties,3 although the onset of the disorder although not as much as is observed in patients suffering may be prepubertal. In the Epidemiological Catchment Area from major depression or panic disorder.17 A German (ECA) study, the usual duration of illness was 6–10 years, national survey of over 4,000 subjects found that on quality- with 40% reporting duration of illness >5 years.10 For cur- of-life measures (Short Form-36) for subscales of general rently active cases of GAD, the mean duration of illness to health, mental health, role limitations due to emotional date was reported as 8.5 years.10 health, and vitality, mean scores were significantly lower GAD is a disabling clinical condition that results in sig- among subjects with GAD than among subjects with nificant impairments in terms of work productivity, resulting MDD.11 For patients with comorbid GAD and MDD, mean in days where a sufferer is limited or even completely unable scores on subscales for general health, mental health, social to perform everyday activities, and reductions in quality of functioning, and vitality were significantly lower than in life and well-being.12 The NCS found that the majority of patients with MDD alone.
individuals with GAD reported substantial interference with their lives (49%), high probability of seeking professional COMORBIDITY OF GENERALIZED
help for GAD symptoms (66%), and the taking of medica- tions for relief of GAD symptoms (44%).2,13 Symptoms of anxiety may appear across many psychiatric Medically, 35% of subjects with GAD from the ECA diagnoses. Perhaps as a result of this phenomenon, the diagno- sample reported that their physical health was only fair to sis of GAD is often noted as comorbid with other psychiatric poor.10 This was also reflected in the high utilization of disorders, including posttraumatic stress disorder, OCD, panic healthcare resources by patients with GAD. In one outpa- disorder (or social phobia), and mood disorders. Approximately tient setting, a 20% lifetime prevalence of GAD was found 25% to 30% of patients with GAD present with comorbid among high users of medical care.14 depression, and 20% to 30% of patients with depression meet Few studies have been published concerning the socioe- the diagnostic criteria for GAD.18,19 In a national survey in conomic impact of GAD. In the ECA survey, 38% of sub- Germany of the patients who met the DSM-IV diagnostic cri- jects characterized their emotional health as only fair to teria for 12-month GAD, 41% had comorbid current MDD, poor, 27% reported currently receiving disability payments while 59% had comorbid 12-month MDD.12 Furthermore, it due to their anxiety, and approximately one half of the sub- has been reported that up to 56% of all 12-month GAD cases jects were gainfully employed.10 A diagnosis of GAD was fulfill the criteria for another anxiety disorder.11 Up to 50% of associated with three times the likelihood of working at a patients with GAD may have a coexisting personality disorder low occupational level, and more than twice the likelihood diagnosis, and as many as 80% of those with treatment-resis- of earning an annual salary <$10,000.10 Souêtre and col- tant GAD may have this condition.18 leagues15 have evaluated the direct and indirect costs associ- In a naturalistic sample of slightly less than 1,000 patients ated with GAD and GAD comorbid with other psychiatric observed in the ambulatory care setting and diagnosed with and somatic disorders in 1,000 patients observed in the GAD, 60% were experiencing one or more symptoms of ambulatory care setting. Total direct and indirect costs asso- comorbidity.15 Addictive patterns (including alcoholism) ciated with GAD were estimated at $733/patient for a were found to be the most prevalent type of psychiatric 3-month period. This included costs of hospitalization, use of comorbidity (25%), followed by depressive symptoms (23%) outpatient services, and lost days of work. For patients with and phobic symptoms (16%). For medical comorbidities, GAD comorbid with other psychiatric disorders, the total gastroenterological and gynecological symptoms were the costs rose to $1,208/patient for the 3-month period. Relative most prevalent (15% and 10%, respectively).15 use of medical services varied between the two groups.
Hospitalization accounted for 35% of the total costs and for over 53% of direct healthcare costs in patients with comor- bid GAD, decreasing to 22% and 32%, respectively, in Brain-imaging studies have demonstrated that the patho- patients with no comorbidities.15 GAD was associated with genesis and control of fear and anxiety involve the frontal cor- lost productivity, with patients estimated to lose an average tex, medial temporal lobe, and limbic system.20 Positron of 45 work days/year due to the disorder, increasing to emission tomography data from patients with GAD suggest 57 work days/year for patients with a concomitant comorbid that cerebral metabolism is increased in the right temporal and disorder.15 The prevalence of absenteeism from work was frontal cortex and decreased in the basal ganglia and white higher for patients with comorbidities (34% versus 27%; matter.21 Benzodiazepine receptor density in the left temporal P<.05). Absenteeism from work accounted for 34% and 31% pole was significantly decreased among patients with GAD, in of total costs for patients with and without comorbidities, line with results obtained in patients with panic disorder.20 Volume 8 – Number 8 (Suppl 1) CNS Spectrums - August 2003 Allgulander C, Bandelow B, Hollander E, et al
GAD. The use of tricyclic antidepressants (TCAs) decreased The pathogenesis and control of anxiety appears to from 20% to 15%.26 involve several neurotransmitters including excitatory amino acids (eg, glutamate), inhibitory amino acids (eg, γ- Acute Treatment Trials
aminobutyric acid [GABA]), and monoaminergic neuro- transmitters (eg, serotonin [5-HT], dopamine, and Benzodiazepines have been commonly used as the first- line acute treatment for GAD. Various benzodiazepines have Benzodiazepine anxiolytic medications act mainly by been shown to be equally effective in the short-term treat- potentiating GABA neurotransmission through the ment of GAD (eg, diazepam, clorazepate, alprazolam, GABAA receptor subtype. Increased GABA neurotransmis- lorazepam).28 However, as a consequence of the changing sion suppresses neuronal firing, thus inhibiting or regulating definition of GAD over the past 2 decades, therapeutic data other neurotransmitters in limbic areas, including 5-HT, with benzodiazepines have been collected on different noradrenaline ([NA], norepinephrine), and dopamine.22 patient populations. None of the benzodiazepines has Noradrenergic cell bodies in the locus ceruleus have high demonstrated efficacy in patient populations selected using concentrations of GABA receptors. The beneficial effects of either ICD-10 or DSM-IV criteria.
benzodiazepines may be mediated, in part, through GABA- Benzodiazepines exert their effects by enhancing GABA facilitated reductions in NA activity.22,23 activity through interaction with the benzodiazepine– The role of 5-HT in anxiety is supported by its modula- GABAA receptor complex.29 Treatment with benzodiazepines tory effects on the locus ceruleus and its dense downstream may cause clinically significant sedation, as well as atten- projections into the amygdala.22 Frontal-lobe feedback sys- tional, psychomotor, cognitive, and memory effects.
tems to limbic structures may also be serotonin sensitive.
Tolerance to most of these adverse events has been noted in Norepinephrine turnover is also implicated, and the interac- patients treated with stable doses of benzodiazepines for tions of these two closely linked systems appear to be recipro- longer than 6–8 weeks17; however, it is possible that some cal.22 Buspirone (a 5-HT1A receptor partial agonist), attentional and psychomotor impairment may persist.
venlafaxine, (which acts mainly through serotonin reuptake Although benzodiazepines are effective in treating anxi- inhibition but may also have noradrenaline reuptake inhibi- ety symptoms, their use is limited by several factors related to tion at higher doses), and some of the SSRIs have demon- long-term use.30 Physical dependence with long-term use and strated efficacy in the treatment of GAD.
the withdrawal syndrome that develops upon drug discontin- uation are key concerns. Withdrawal symptoms (eg, nervous- ness, insomnia, restlessness, lethargy, nausea, depression) A number of clinical trials have demonstrated the benefit appear within 6–12 hours and peak within 2–4 days of stop- of anxiolytic medication therapy and some psychotherapies ping a short-half-life benzodiazepine (eg, alprazolam or in the acute treatment of GAD patients. Pharmacologic lorazepam), before subsiding in 2–3 weeks.17 Discontinuation therapies include benzodiazepines, buspirone, and antide- of high-potency, short-half-life benzodiazepines may cause pressants. The psychological therapy that has shown greatest the most severe withdrawal symptoms (eg, nervousness, benefit is cognitive-behavioral therapy (CBT).
insomnia, restlessness, lethargy, nausea, depression). Patients The rating instruments most commonly used to assess with a previous history of alcohol or substance abuse may be treatment effects in published pharmacological studies of more prone to withdrawal effects.31 Studies of the long-term efficacy of benzodiazepines have reported the development •Hamilton Rating Scale for Anxiety (HAM-A)24 of tolerance or loss of effect over time in the treatment of •Clinical Global Impression–Severity (CGI-S) and anxiety.30 Additionally, a high relapse rate (65%) is observed Improvement scales25 in the 6-month period following benzodiazepine discontinu- ation after short-term treatment.30 However, Rickels and col- leagues32 demonstrated that clorazepate efficacy could be In a 1997 survey, 34% of 51 internationally recognized maintained over 6 months.
experts selected benzodiazepines as the first-line therapy for Although observed principally in persons who abuse GAD.26 This percentage had decreased from the 41% mea- other drugs and not in long-term GAD users, benzodi- sured in a survey 5 years earlier, but nonetheless showed that azepines also have the potential for recreational abuse.17 despite other available treatment options, benzodiazepines Seizures can be precipitated by abrupt withdrawal of high remained a popular choice of initial therapy, probably due to doses of benzodiazepines.
their rapid onset and good efficacy.27 Buspirone was the pre- Benzodiazepines also may have additive effects due to ferred first-line treatment option for 15% of responders in pharmacodynamic interactions when used concomitantly both 1992 and 1997, while use of selective serotonin reup- with central nervous system depressants, including alcohol.33 take inhibitors (SSRIs) had increased from 2% to 19%. In 1999 and 2000, venlafaxine extended release (XR) and Azapirones, which act as partial agonists at the 5-HT1A paroxetine were approved in the US for the treatment of receptor level, were developed in an attempt to produce Volume 8 – Number 8 (Suppl 1) CNS Spectrums - August 2003 Long-Term Treatment of Generalized Anxiety Disorder
effective antianxiety agents with no abuse potential, XR formulation of venlafaxine was developed to allow for although only one, buspirone, made it to market.
prolonged duration of absorption and once-daily dosing, Advantages of azapirones include no withdrawal reaction, no with the potential of an improved side-effect profile.40 potentiation of alcohol or other sedative hypnotics, and a The efficacy of venlafaxine XR has been investigated in lack of abuse potential.34 A disadvantage is the relatively five outpatient placebo-controlled studies, which showed a gradual onset of action (2 weeks). Side effects most fre- consistent pattern of improvement in the treatment of quently reported include gastrointestinal symptoms (appetite GAD.41 Four of these studies demonstrated a statistically disturbances and abdominal complaints) and dizziness.
superior improvement compared with placebo.40-44 Buspirone was approved in the US as a treatment for In an 8-week randomized study of 365 GAD patients symptoms of GAD in 1986. In most other countries where without comorbid MDD, venlafaxine XR (75 or buspirone is licensed, it is for the treatment of anxiety but not 150 mg/day) was significantly more effective than placebo at for the treatment of GAD. Buspirone, in addition to its 5- week 8, as assessed by the HAM-A and patient-rated HT1A activity, has an affinity for dopamine D2 receptors.1 Hospital Anxiety and Depression Scale (HADS).42 Buspirone has been shown in some studies to be equally effec- Venlafaxine XR 75 mg/day was superior to buspirone at tive as such benzodiazepines as diazepam, clorazepate, alprazo- weeks 3, 4, and 8, as assessed by CGI-S scores.
lam, and lorazepam in treating GAD, while other studies Discontinuation rates due to adverse events were 22% for failed to show superiority over placebo.35 Although buspirone venlafaxine XR 75 mg/day, 28% for venlafaxine 150 mg/day, has consistently demonstrated a reduction in anxiety symp- 15% for buspirone, and 10% for placebo.42 toms, it has not been shown to be effective in reducing the In contrast to the older TCAs, venlafaxine XR generally severity and frequency of panic attacks.35 Buspirone is also has a more favorable adverse-event profile and greater tolera- more effective than placebo in improving anxiety and depres- bility.43 In short-term treatment with venlafaxine XR, the sive symptoms in GAD patients with concurrent depressive most common adverse events (occurring in ≥10% of patients symptoms.36 Prior benzodiazepine treatment (within 4 weeks and at least twice the frequency found with placebo) are to 5 years) has a negative impact on buspirone treatment out- nausea, somnolence, dry mouth, dizziness, sweating, consti- come, with increased rates of treatment discontinuation, pation, and anorexia.43 adverse-event reports, and reduced response rates.37 Selective Serotonin Reuptake Inhibitors
Since buspirone was evaluated as a treatment for GAD Paroxetine was the first SSRI to receive approval for the using DSM-III criteria, which required only a 1-month treatment of GAD in the US (2001) and is the only SSRI to duration of anxiety symptoms, few data on the efficacy of date for which there are published results of randomized buspirone in subjects with chronic symptoms of anxiety short-term efficacy trials. In an 8-week study, 81 DSM-IV- diagnosed patients with GAD were randomized to receive paroxetine (20 mg/day), imipramine (50–100 mg/day), or Imipramine has shown some efficacy in the treatment of 2´-chlorodesmethyldiazepam, a standard benzodiazepine GAD. An 8-week study of 230 patients with DSM-III GAD (3–6 mg/day).45 Significant improvement in anxiety ratings compared the efficacy of imipramine with diazepam, tra- was observed after 2 weeks of treatment in all three groups.
zodone, and placebo. By Week 3, all three active drugs pro- At study endpoint, the response rates, as assessed by ≥50% duced comparable improvement. However, by Week 8, only improvement in baseline HAM-A total score, were 68% in imipramine, not placebo, had sustained significant improve- the paroxetine group, 72% in the imipramine group, and ment across all efficacy measures, including HAM-A total 55% in the 2´-chlorodesmethyldiazepam group. The most score, CGI-S, the Covi anxiety scale, and the patient-rated frequently reported adverse events for each drug were nausea Hopkins Symptom Checklist (P<.01).38 Significantly more (paroxetine); drowsiness (2´-chlorodesmethyldiazepam); and side effects were elicited by imipramine and trazodone (mean dry mouth, drowsiness, and constipation (imipramine).
maximum daily doses, 143 and 255 mg, respectively) than by Subsequently, three placebo-controlled, 8-week studies diazepam (26 mg/day) or placebo. Imipramine was more further evaluated paroxetine in the treatment of GAD.46,47 A effective in treating the psychic symptoms of anxiety, while large US multicenter, fixed-dose study evaluated the efficacy diazepam appeared to be more effective in alleviating of paroxetine (20 and 40 mg/day) in 566 patients with somatic symptoms. Another 6-week, double-blind, parallel- GAD.48 Both doses of paroxetine showed a statistically sig- design study of 60 patients with GAD found imipramine to nificant reduction from baseline on both HAM-A total score be more effective than alprazolam in attenuating psychic and the patient self-rated Sheehan Disability Scale over symptoms, such as dysphoria and negative anticipatory placebo (P<.001) at week 8.
thinking, but less effective than alprazolam in attenuating The other two studies were flexible-dose studies using somatic symptoms.39 paroxetine 20–50 mg/day.46,47 In the first study, paroxetine- Serotonin and Noradrenaline Reuptake Inhibitors
treated patients experienced a statistically significant In 1999, venlafaxine XR, a serotonin and noradrenaline improvement over placebo-treated patients in HAM-A reuptake inhibitor (SNRI), became the first antidepressant score at weeks 6 and 8 (P=.041 and P=.008, respectively).47 to be approved for the treatment of GAD in the US. The In the second study, paroxetine treatment resulted in a Volume 8 – Number 8 (Suppl 1) CNS Spectrums - August 2003 Allgulander C, Bandelow B, Hollander E, et al
statistically significant reduction in HAM-A score compared taken in single-blind phase) or placebo.51 Paroxetine treat- with placebo in the intent-to-treat observed-case data set ment resulted in significantly fewer relapsed patients com- (P<.05).47 The safety profile was similar to that seen for pared to placebo (10.9% versus 39.9%, respectively), and paroxetine studies in depression and other anxiety disorders.47 time to relapse for patients on placebo was 4.7 times more than for paroxetine-treated patients. The proportion of Long-Term Treatment Trials
patients achieving remission (HAM-A total score ≤7) at the Few trials have studied the long-term (≥6 months) treat- end of the single-blind phase was 42.5% and increased in the ment of GAD (Table 2).32,49-51 Venlafaxine has been studied double-blind phase to 73%, compared with 34% for patients in only two 6-month efficacy trials. The first was a random- on placebo at the end of this phase. During the double-blind ized, double-blind, parallel-group, flexible-dose comparison phase, a significant decrease in HAM-A scores was also of venlafaxine XR (75–225 mg/day) with placebo in noted in paroxetine-treated patients compared to those 251 patients with GAD without MDD.49 Anxiety scores given placebo. An improvement in functional disability was were significantly improved by venlafaxine XR compared observed in patients treated with paroxetine, and overall, with placebo from week 1 or 2 through week 28 on all pri- paroxetine was well tolerated.
mary efficacy measures, including HAM-A total, HAM-A A prospective, double-blind study compared clorazepate psychic anxiety factor, and CGI scale scores. The most com- (a benzodiazepine with a long half-life) with buspirone treat- mon treatment-emergent adverse event was nausea, followed ment in 134 patients with GAD.32 The primary objective of by somnolence and dry mouth.
this study was to assess the results of treatment withdrawal.
In the second study, 544 patients were randomized to Patients were treated with therapeutic doses of clorazepate receive a fixed dose of venlafaxine XR (37.5, 75, or (7.5–60 mg/day) or buspirone (5–40 mg/day) for 6 months 150 mg/day) or placebo.40,50,52 In a last-observation-carried-for- before therapy was abruptly withdrawn. Both treatments ward analysis, all three doses of venlafaxine XR resulted in sta- demonstrated effective control of symptoms during the trial.
tistically significant decreases from baseline on all primary There was a significant increase in symptom severity consis- efficacy variables (HAM-A total, HAM-A psychic anxiety tent with withdrawal reaction (five or more new symptom factor, CGI). Moreover, as assessed by decreases in HAM-A complaints on the Physician Checklist of Withdrawal total scores, the highest dose (150 mg/day) of venlafaxine XR Symptoms) for the clorazepate group (72%) but not the bus- was superior to the lowest dose (37.5 mg/day), from pirone group (9%) (P<.001). There was no evidence of tol- 3–6 months of treatment. Discontinuation of therapy with erance development with either drug during 6 months of venlafaxine XR occurred mainly at the beginning of treatment; treatment; improvement in HAM-A assessed after 1 month the adverse events most commonly leading to discontinuation of treatment was maintained for the next 5 months in both were nausea, headache, dizziness, sweating, and insomnia.52 groups. The most common adverse events reported with Paroxetine has also been assessed for the treatment of clorazepate were drowsiness, fatigue, and lack of coordina- GAD in a 32-week multicenter relapse prevention study, tion; with buspirone, adverse events were lightheadedness, which included an 8-week, single-blind, flexible-dose dizziness, and insomnia. During the course of the study, sig- (20–50 mg/day) treatment period followed by a 24-week nificantly more patients discontinued treatment with bus- double-blind phase with patients randomized to either pirone (45%) than with clorazepate (26%). In practice, most paroxetine (20–50 mg/day, patients maintained at dose clinicians taper benzodiazepines slowly.
TABLE 2. LONG-TERM TREATMENT TRIALS OF PHARMACOTHERAPY FOR GAD
Study Duration (Weeks) Gelenberg, et al (2000)49 Venlafaxine XR Placebo Allgulander, et al (2001)50 Venlafaxine XR Placebo Rickels, et al (1988)32 Stocchi, et al (2001)51 GAD=generalized anxiety disorder; XR=extended release; RCT=randomized controlled trial; HAM-A=Hamilton Rating Scale for Anxiety; CGI=Clinical Global Impression.
Volume 8 – Number 8 (Suppl 1) CNS Spectrums - August 2003 Long-Term Treatment of Generalized Anxiety Disorder
Reviews of randomized clinical trials suggest that CBT is One trial compared the efficacy of 12 sessions of ND ther- better than no treatment, broadly equivalent to medica- apy, AR, or CBT for GAD, although not with placebo con- tion, and more efficacious than other psychotherapies, with trol.59 For this study's purposes, ND therapy was described to clinically significant improvement occurring in 50% to patients as a means of exploring their life experiences in a 70% of cases.53,54 On average, if psychological therapy is relaxed environment with the goal of increasing their self- given, about 60% of patients will show some form of signifi- awareness and understanding of anxiety. Treatment with AR cant improvement at 6-month follow-up, but only 40% will and CBT generated clinically meaningful change on both have recovered in terms of State-Trait Anxiety responder status and end-state functioning more often than Inventory–Trait subscale scores.55 However, this figure may did ND therapy (P<.01). Follow-up indicated that by overestimate the effects of short-term psychotherapy since 12-months postassessment, significantly more ND therapy many patients will likely have received medication or addi- patients received subsequent treatment (61% of ND therapy, tional psychotherapy during the follow-up period.
17% of AR, and 16% of CBT patients).
Analysis of six randomized controlled trials comparing various psychological therapies in treating patients with GAD suggests that CBT is the most effective therapy and A limited number of trials comparing a combination of results in significant improvements for 50% to 60% of pharmacotherapy and psychotherapy with single therapy have treated patients.55 The trials employed various treatment failed to show additional benefit of combined treatment.60 One modalities including behavioral therapy, cognitive therapy 6-week, randomized, double-blind study examined combined (or combination of both), nondirective (ND) therapy, treatment of diazepam or placebo plus weekly psychotherapy.61 applied relaxation (AR), self-control desensitization, ana- A small effect of diazepam was observed at the first-week assess- lytical psychotherapy, or anxiety management training.
ment but not thereafter. A more detailed breakdown revealed Importantly, only one trial, which compared cognitive some efficacy in patients with moderate or severe levels of anx- therapies to behavioral therapies, employed a wait-list iety, but not in those with low levels of anxiety.
control group.
Diazepam treatment in combination with psychotherapy was also examined in 101 patients with GAD, each allocated The aim of CBT is to alleviate the suffering and interfer- to one of five treatment groups—diazepam, placebo, CBT, ence caused by a disorder by helping the patient recognize diazepam plus CBT, or placebo plus CBT.62 After 10 weeks of and alter patterns of distorted thinking and dysfunctional treatment, the CBT-diazepam group showed the greatest behavior. The effect size of CBT over that of the waiting improvements in symptom severity and overall change in list was 1.22 based on HAM-A scores and 2.48 compared to symptoms. However, a 6-month follow-up revealed that sim- pre-treatment within-group scores.56 ilar proportions of patients from each treatment group Some studies suggest that treatment gains following CBT received psychotropic medication after the end of the study.
may be maintained for up to 1 year, whereas response after In another study, 60 patients with GAD were randomized medication discontinuation is attenuated.57 Moreover, in to one of four treatment strategies: buspirone and anxiety other studies CBT has been associated with decreased use of management training; buspirone and ND therapy, in which patients were allowed to choose the topics they would Discontinuation due to adverse event: 24% venlafaxine XR (P=.05) Most common adverse events with venlafaxine XR leading to discontinuation: nausea, headache, dizziness, sweating, insomnia No significant difference between treatments Discontinuation rate: Most common treatment-related adverse events: single- blind phase: nausea (21.5% of patients); double-blind phase: headache (6.9% paroxetine versus 5.2% placebo) Allgulander C, Bandelow B, Hollander E, et al. CNS Spectrums. Vol 8, No 8 (suppl 1). 2003.
Volume 8 – Number 8 (Suppl 1) CNS Spectrums - August 2003 Allgulander C, Bandelow B, Hollander E, et al
discuss; placebo and anxiety management training; and placebo and ND therapy.60 The psychological treatments com- 1. Allgulander C. Generalized anxiety disorder. Clinical characteristics prised seven sessions of 45 minutes each. Buspirone was given and treatment. South African Rev Psychiatry. 2001;1:6-12.
in flexible dosage up to 30 mg/day. Patients exhibited signifi- 2. Wittchen HU, Zhao S, Kessler RC, Eaton WW. DSM-III-R generalized cant improvement on HAM-A and HADS at 4 and 8 weeks.
anxiety disorder in the National Comorbidity Survey. Arch Gen However, there were no differences among treatment groups.
3. Rickels K, Schweizer E. The clinical course and long-term management of It is possible that the small sample size (distributed among four generalized anxiety disorder. J Clin Psychopharmacol. 1990;10:101S-110S.
treatment groups) and high dropout rate for patients receiving 4. Diagnostic and Statistical Manual of Mental Disorders. 3rd ed.
buspirone treatment (12 of 16 dropouts) limited the ability of Washington, DC: American Psychiatric Association; 1980.
this study to detect statistical differences.60 5. Kessler RC, Keller MB, Wittchen HU. The epidemiology of generalized anxiety disorder. Psychiatr Clin North Am. 2001;24:19-39.
OVERVIEW OF AVAILABLE
6. Diagnostic and Statistical Manual of Mental Disorders. 3rd ed. [revised].
Washington, DC: American Psychiatric Association; 1987.
Currently, there are no published guidelines for the long- 7. Diagnostic and Statistical Manual of Mental Disorders. 4th ed.
term treatment of GAD. Schweizer and Rickels Washington, DC: American Psychiatric Association; 1994.
lined strategies for the treatment of GAD in the primary care 8. International Classification of Diseases. 10th revision. Geneva, Switzerland: World Health Organization; 1992.
setting. In this report, the authors suggest that use of a benzo- 9. Lader MH. The nature and duration of treatment for GAD. Acta diazepine is favored for short-term treatment when there is evidence of episodes of panic or when GAD presents with 10. Blazer DG, Hughes D, George LK. Generalized anxiety disorder. In: Robins prominent adrenergic symptoms, even in the absence of panic LN, Regier DA, eds. Psychiatric Disorders in America: The Epidemiologic attacks.63 Buspirone should be favored over benzodiazepines in Catchment Area Study. New York, NY: The Free Press; 1991:180-203.
clinical situations where the following concerns arise63: 11. Carter RM, Wittchen HU, Pfister H, Kessler RC. One-year prevalence •Impairment in psychomotor function, attention, of subthreshold and threshold DSM-IV generalized anxiety disorder in a vigilance, or cognition and memory nationally representative sample. Depress Anxiety. 2001;13:78-88.
•Potentiation of alcohol or sedative effects of other 12. Wittchen HU, Carter RM, Pfister H, Montgomery SA, Kessler RC.
Disabilities and quality of life in pure and comorbid generalized anxiety disorder and major depression in a national survey. Int Clin •Prominent aggression and/or irritability •Abuse potential (eg, diagnosis or history of alcoholism) 13. Kessler RC, McGonagle KA, Zhao S, et al. Lifetime and 12-month •Risk of acute discontinuation prevalence of DSM-III-R psychiatric disorders in the United States.
With the advent of the newly approved antidepressants Results from the National Comorbidity Survey. Arch Gen Psychiatry. venlafaxine XR and paroxetine for GAD, the SSRIs and SNRIs may be preferred for both acute and long-term ther- 14. Katon W, Von Korff M, Lin E, et al. Distressed high utilizers of medical apy for GAD. Because psychiatric comorbidity in patients care: DSM-III-R diagnoses and treatment needs. Gen Hosp Psychiatry. with GAD is common, pharmacotherapies for GAD should 15. Souêtre E, Lozet H, Cimarosti I, et al. Cost of anxiety disorders: impact have demonstrated efficacy in comorbidity.46 CBT has also of comorbidity. J Psychosom Res. 1994;38(suppl 1):151-160.
shown efficacy in several studies, with sustained results up to 16. Lecrubier Y, Weiller E. GAD: current treatment and costs. Eur 1 year after treatment.57 17. Schweizer E. Generalized anxiety disorder. Longitudinal course and pharmacologic treatment. Psychiatr Clin North Am. 1995;18:843-857.
GAD is known to be a chronic, recurring, and distressing 18. Papp LA, Gorman JM. Generalized anxiety disorder. In: Comprehensive disorder that responds to both pharmacological and some Textbook of Psychiatry. 6th ed. Baltimore, Md: Williams and Wilkins; psychotherapeutic interventions. Both pharmacotherapy and, to a lesser extent, some types of psychotherapy have 19. Environmental factors in the etiology of anxiety. Available at: shown benefit in the acute treatment of GAD. Medication http://www.acnp.org/g4/GN401000127/Default.htm. Accessed: May 2003. 20. Tiihonen J, Kuikka J, Rasanen P, et al. Cerebral benzodiazepine receptor treatments which are Food and Drug Administration- binding and distribution in generalized anxiety disorder: a fractal analy- approved for the treatment of GAD include buspirone, ven- sis. Mol Psychiatry. 1997;2:463-471.
lafaxine, and paroxetine. Although benzodiazepines are 21. Wu JC, Buchsbaum MS, Hershey TG, Hazlett E, Sicotte N, Johnson JC.
commonly used for the acute treatment of GAD, their use PET in generalized anxiety disorder. Biol Psychiatry. 1991;29:1181-1199.
may be limited due to adverse effects, withdrawal symptoms, 22. Ninan PT. The functional anatomy, neurochemistry, and pharmacology development of tolerance, and high relapse rates upon dis- of anxiety. J Clin Psychiatry. 1999;60(suppl 22):12-17.
continuation. Few randomized controlled trials of either 23. Nutt DJ. Neurobiological mechanisms in generalized anxiety disorder. J pharmacotherapy or psychotherapy have studies the long- Clin Psychiatry. 2001;62(suppl 11):22-27.
term treatment of GAD and no guidelines currently exist.
24. Hamilton M. The assessment of anxiety by rating. Br J Med Psychol. Further long-term studies, which will allow researchers to 25. Guy W. Clinical Global Impressions. ECDEU Assessment Manual for observe remission rates and thereby give a clearer profile of Psychopharmacology. Rockville, MD: US National Institute of Health, the possible outcomes of treatment, are required. CNS
Psychopharmacology Research Branch; 1976:217-222. Volume 8 – Number 8 (Suppl 1) CNS Spectrums - August 2003 Long-Term Treatment of Generalized Anxiety Disorder
26. Uhlenhuth EH, Balter MB, Ban TA, Yang K. International study of 46. Davidson JR. Pharmacotherapy of generalized anxiety disorder. J Clin expert judgment on therapeutic use of benzodiazepines and other psy- Psychiatry. 2001;62(suppl 11):46-50.
chotherapeutic medications: VI. Trends in recommendations for the 47. Pollack MH, Zaninelli R, Goddard A, et al. Paroxetine in the treatment pharmacotherapy of anxiety disorders, 1992-1997. Depress Anxiety.
of generalized anxiety disorder: results of a placebo-controlled, flexible- dosage trial. J Clin Psychiatry. 2001;62:350-357.
27. Uhlenhuth EH, Balter MB, Ban TA, Yang K. International study of 48. Bellew KM, McCafferty JP, Iyengar M, Zaninelli RM. Paroxetine treat- expert judgement on therapeutic use of benzodiazepines and other psy- ment of GAD: a double-blind placebo-controlled trial. Poster presented chotherapeutic medications: II. Pharmacotherapy of anxiety disorders. J at: Annual Meeting of the American Psychiatric Association; May 13- Affect Disord. 1995;35:153-162.
18, 2000; Chicago, Ill.
28. Brawman-Mintzer O. Pharmacologic treatment of generalized anxiety 49. Gelenberg AJ, Lydiard RB, Rudolph RL, Aguiar L, Haskins JT, Salinas disorder. Psychiatr Clin North Am. 2001;24:119-137.
E. Efficacy of venlafaxine extended-release capsules in nondepressed 29. Nutt DJ, Malizia AL. New insights into the role of the GABA(A)-ben- outpatients with generalized anxiety disorder: A 6-month randomized zodiazepine receptor in psychiatric disorder. Br J Psychiatry. controlled trial. JAMA. 2000;283:3082-3088.
50. Allgulander C, Hackett D, Salinas E. Venlafaxine extended release 30. Mahe V, Balogh A. Long-term pharmacological treatment of generalized (ER) in the treatment of generalised anxiety disorder: twenty-four-week anxiety disorder. Int Clin Psychopharmacol. 2000;15:99-105.
placebo-controlled dose-ranging study. Br J Psychiatry. 2001;179:15-22.
31. Gorman JM, Papp LA. Chronic anxiety: deciding the length of treat- 51. Stocchi F, Nordera G, Jokinen R, Lepola U. Efficacy and tolerability of ment. J Clin Psychiatry. 1990;51(suppl):11-15.
paroxetine for the long-term treatment of generalised anxiety disorder 32. Rickels K, Schweizer E, Csanalosi I, et al. Long-term treatment of anxi- (GAD). Poster presented at: Annual Meeting of the American ety and risk of withdrawal. Prospective comparison of clorazepate and Psychiatric Association; May 5-10, 2001; New Orleans, La.
buspirone. Arch Gen Psychiatry. 1988;45:444-450.
52. Hackett D, Desmet A, Salinas EO. Dose-response efficacy of venlafax- 33. Lydiard RB. An overview of generalized anxiety disorder: disease state— ine XR in GAD. Poster presented at: 11th World Congress of appropriate therapy. Clin Ther. 2000;22(suppl A):A3-19.
Psychiatry; August 6-11, 1999; Hamburg, Germany.
34. Lydiard RB, Brawman-Mintzer O, Ballenger JC. Recent developments 53. Durham RC, Allan T. Psychological treatment of generalized anxiety in the psychopharmacology of anxiety disorders. J Consult Clin Psychol. disorder. A review of the clinical significance of results in outcome stud- ies since 1980. Br J Psychiatry. 1993;163:19-26.
35. Apter JT, Allen LA. Buspirone: future directions. J Clin 54. Borkovec TD, Whisman MA. Psychosocial treatment for generalized anxiety disorder. In: Mavissakalian M, Prien R, eds. Long-term Treatment 36. Sramek JJ, Tansman M, Suri A, et al. Efficacy of buspirone in general- for the Anxiety Disorders. Washington, DC: American Psychiatric Press; ized anxiety disorder with coexisting mild depressive symptoms. J Clin 55. Fisher PL, Durham RC. Recovery rates in generalized anxiety disorder 37. DeMartinis N, Rynn M, Rickels K, Mandos L. Prior benzodiazepine use following psychological therapy: an analysis of clinically significant and buspirone response in the treatment of generalized anxiety disorder.
change in the STAI-T across outcome studies since 1990. Psychol Med. J Clin Psychiatry. 2000;61:91-94.
38. Rickels K, Downing R, Schweizer E, et al. Antidepressants for the treat- 56. Borkovec TD, Ruscio AM. Psychotherapy for generalized anxiety disor- ment of generalized anxiety disorder. A placebo-controlled comparison der. J Clin Psychiatry 2001;62 (suppl 11):37-42.
of imipramine, trazodone, and diazepam. Arch Gen Psychiatry. 57. Gould RA, Otto MW, Pollack MH, et al. Cognitive behavioral and pharmacological treatment of generalized anxiety disorder: a prelimi- 39. Hoehn-Saric R, McLeod DR, Zimmerli WD. Differential effects of nary meta-analysis. Behav Ther. 1997;28:285-305.
alprazolam and imipramine in generalized anxiety disorder: somatic ver- 58. Brown TA, O'Leary TA, Barlow DH. Generalized anxiety disorder. In: sus psychic symptoms. J Clin Psychiatry. 1988;49:293-301.
Barlow DH, ed. Clinical Handbook of Psychological Disorders. 2nd ed.
40. Hackett D. Venlafaxine XR in the treatment of anxiety. Acta Psychiatr New York, NY: Guildford; 1993:137-188.
Scand. 2000;102(suppl 46):30-35.
59. Borkovec TD, Costello E. Efficacy of applied relaxation and cognitive- 41. Kelsey JE. Efficacy, safety, and tolerability of venlafaxine XR in general- behavioral therapy in the treatment of generalized anxiety disorder. ized anxiety disorder. Depress Anxiety. 2000;12(suppl 1):81-84.
J Consult Clin Psychol. 1993;61:611-619.
42. Davidson JR, Dupont RL, Hedges D, Haskins JT. Efficacy, safety, and tol- 60. Lader MH, Bond AJ. Interaction of pharmacological and psychological erability of venlafaxine extended release and buspirone in outpatients treatments of anxiety. Br J Psychiatry. 1998;(suppl 34):42-48.
with generalized anxiety disorder. J Clin Psychiatry. 1999;60:528-535.
61. Shapiro AK, Struening EL, Shapiro E. Diazepam: how much better than 43. Sheehan DV. Venlafaxine extended release (XR) in the treatment of placebo? J Psychiatr Res. 1983;17:51-73.
generalized anxiety disorder. J Clin Psychiatry. 1999;60(suppl 22):23-28.
62. Power KG, Simpson RJ, Swanson V, Wallace LA. Controlled compari- 44. Rickels K, Pollack MH, Sheehan DV, Haskins JT. Efficacy of extended- son of pharmacological and psychological treatment of generalized anx- release venlafaxine in nondepressed outpatients with generalized anxi- iety disorder in primary care. Br J Gen Pract. 1990;40:289-294.
ety disorder. Am J Psychiatry. 2000;157:968-974.
63. Schweizer E, Rickels K. Strategies for treatment of generalized anxiety 45. Rocca P, Fonzo V, Scotta M, Zanalda E, Ravizza L. Paroxetine efficacy in in the primary care setting. J Clin Psychiatry. 1997;58(suppl 3):27-31.
the treatment of generalized anxiety disorder. Acta Psychiatr Scand. Volume 8 – Number 8 (Suppl 1) CNS Spectrums - August 2003 Provided by an unrestricted educational grant from

Source: http://barendspsychology.com.webhosting45.transurl.nl/wp-content/uploads/2015/01/International-journey-of-neuropsychiatric-medicine.pdf