the-eye.eu of Psychiatric Dru… · CONTENTS Preface xv Acknowledgements xvii Disclaimer xix 1...

HANDBOOK OF PSYCHIATRICDRUGS

Handbook of Psychiatric Drugs Jeffrey A. Lieberman and Allan Tasman© 2006 John Wiley & Sons, Ltd. ISBN: 0-470-02821-1

HANDBOOK OFPSYCHIATRIC DRUGS

Jeffrey A. LiebermanProfessor and Chair

Department of PsychiatryColumbia University, College of Physicians and

Surgeons, and Director, New York State PsychiatricInstitute, New York, USA

Allan TasmanProfessor and Chair

Department of Psychiatry and Behavioral SciencesUniversity of Louisville School of Medicine, Louisville,

Kentucky, USA

Copyright © 2006 John Wiley & Sons Ltd, The Atrium, Southern Gate,Chichester, West Sussex PO19 8SQ, England

Telephone (+44) 1243 779777

Email (for orders and customer service enquiries): [email protected] our Home Page on www.wiley.com

All Rights Reserved. No part of this publication may be reproduced, stored in aretrieval system or transmitted in any form or by any means, electronic, mechanical,photocopying, recording, scanning or otherwise, except under the terms of theCopyright, Designs and Patents Act 1988 or under the terms of a licence issued bythe Copyright Licensing Agency Ltd, 90 Tottenham Court Road, London W1T4LP, UK, without the permission in writing of the Publisher. Requests to thePublisher should be addressed to the Permissions Department, John Wiley & SonsLtd, The Atrium, Southern Gate, Chichester, West Sussex PO19 8SQ, England,or emailed to [email protected], or faxed to �+44� 1243 770620.

Designations used by companies to distinguish their products are often claimed astrademarks. All brand names and product names used in this book are trade names,service marks, trademarks or registered trademarks of their respective owners. ThePublisher is not associated with any product or vendor mentioned in this book.

This publication is designed to provide accurate and authoritative information inregard to the subject matter covered. It is sold on the understanding that the Publisheris not engaged in rendering professional services. If professional advice or otherexpert assistance is required, the services of a competent professional should be sought.

Other Wiley Editorial Offices

John Wiley & Sons Inc., 111 River Street, Hoboken, NJ 07030, USA

Jossey-Bass, 989 Market Street, San Francisco, CA 94103-1741, USA

Wiley-VCH Verlag GmbH, Boschstr. 12, D-69469 Weinheim, Germany

John Wiley & Sons Australia Ltd, 42 McDougall Street, Milton, Queensland 4064,Australia

John Wiley & Sons (Asia) Pte Ltd, 2 Clementi Loop #02-01, Jin Xing Distripark,Singapore 129809

John Wiley & Sons Canada Ltd, 22 Worcester Road, Etobicoke, Ontario, CanadaM9W 1L1

Wiley also publishes its books in a variety of electronic formats. Some content thatappears in print may not be available in electronic books.

Library of Congress Cataloging in Publication Data

Lieberman, Jeffrey A., 1948–Handbook of psychiatric drugs / Jeffrey A. Lieberman, Allan Tasman.

p. cm.Includes bibliographical references and index.ISBN-13: 978-0-470-02821-6 (pbk. : alk. paper)ISBN-10: 0-470-02821-1 (pbk. : alk. paper)1. Psychotropic drugs—Handbooks, manuals, etc. 2. Psychopharmacology—Handbooks, manuals, etc. I. Tasman, Allan, 1947– II. Title.RM315.L54 2006615′.788—dc22 2006004739

British Library Cataloguing in Publication Data

A catalogue record for this book is available from the British Library

ISBN-13 978-0-470-02821-6ISBN-10 0-470-02821-1

Typeset in 9.5/11.5pt Times by Integra Software Services Pvt. Ltd, Pondicherry, IndiaPrinted and bound in Great Britain by Clays Ltd, Bungay, Suffolk NR35 1EDThis book is printed on acid-free paper responsibly manufactured from sustainableforestry in which at least two trees are planted for each one used for paperproduction.

To our patients for their courage and for their assistance inthe search for treatments for mental illness.

To my family in gratitude for their love, support andpatience. Jeffrey Lieberman

With love and thanks to my family, and especially my fatherGoodman Tasman, for your inspiration and support.

Allan Tasman

CONTENTS

Preface xvAcknowledgements xviiDisclaimer xix

1 Antipsychotic Drugs 1Introduction 1Pharmacology 1

Chemistry 2Mechanism of Action 3Pharmacokinetics 3

Indications for Use of Antipsychotic Drugs 4Drug Selection and Initiation of Treatment 7

Drug Selection for the Treatment ofSchizophrenia 7

Effects of Antipsychotic Agents on Symptomsof Schizophrenia 15

Drug Selection for the Treatment of BipolarDisorder 19

Drug Selection for the Treatment of MajorDepression With Psychotic Features 19

Drug Selection for the Treatment of DelusionalDisorder 20

Drug Selection for the Treatment of Delirium 20Drug Selection for the Treatment of Psychosis

and Agitation Associated with Dementia 21Drug Selection for the Treatment of Mental

Retardation and Developmental Disorders 21Drug Selection for Huntington’s Disease and

Tourette’s Disorder 22Substance-induced Psychoses 22

Adverse Effects of Antipsychotics 23Acute Extrapyramidal Side Effects (Dystonia,

Parkinsonism, Akathisia) 25

viii CONTENTS

Tardive Dyskinesia and Other TardiveSyndromes 26

Neuroleptic Malignant Syndrome 29Endocrine and Sexual Effects 30Metabolic Effects 31Cardiovascular Effects 32Gastrointestinal Effects 34Hepatic Effects 34Hematological Effects 34Other Side Effects 35

Drug Interactions and Antipsychotic Agents 36Antipsychotic Medications and Pregnancy 38

Additional Reading 39

2 Antidepressants 41Introduction 41Pharmacology 41

Mechanisms of Action 42Pharmacokinetics 44

Indications for Use of Antidepressants 49Panic Disorder (PD) 50Obsessive–Compulsive Disorder (OCD) 50Generalized Anxiety Disorder (GAD) 50Social Phobias and Posttraumatic Stress

Disorder (PTSD) 50Bulimia 51Anorexia Nervosa 51Body Dysmorphic Disorder (BDD) 51Premenstrual Dysphoria (PMDD) 51Childhood Disorders 52Other Psychiatric Disorders 52Other Medical Conditions 53

Drug Selection and Initiation of Treatment forMajor Depression 55

Special Considerations in the Selection of anAntidepressant 56

Initiation of Treatment 58Therapeutic Drug Monitoring 58Early, or Pre-response, Period 62Response, or Acute Treatment, Period 63

CONTENTS ix

Treatment of Partially Responsive andNonresponsive Patients 63

Antidepressant Augmentation 63Changing to a New Agent 66Continuation and Maintenance Periods, and

Discontinuation 67Side Effects 69

Anticholinergic Effects 70Autonomic Effects 70Neurologic Effects 72Weight Gain 73Gastrointestinal Symptoms 73Sexual Dysfunction 73Cardiac Conduction 74Suicide Risk 74Allergic and Hematologic Effects 75Other Effects 75Central Serotonergic Syndrome 75Monoamine Oxidase Inhibitors 75

Drug Interactions 76Tricyclic Antidepressants 76Pharmacokinetic Effects 77Monoamine Oxidase Inhibitors 78Serotonin Reuptake Inhibitors 78Other Second-Generation Antidepressants 79Third-Generation Antidepressants 79

Summary 79Recommendations for the Use of Antidepressants 80

Additional Reading 80

3 Mood Stabilizers 83Introduction 83Pharmacology 85

Chemistry 85Mechanism of Action 85Pharmacokinetics 87

Indications 90Drug Selection and Initiation of Treatment 90

Acute Mania 90Drug Selection 90

x CONTENTS

Relative Efficacy of Different Agents 92Treatment Initiation and Dose Titration 94Acute Depression 99Drug Selection 99Treatment Initiation and Dose Titration 102Breakthrough Episodes 103Maintenance 104Rapid Cycling 110Costs 110

Adverse Effects 111Drug Interactions 119

Pregnancy 124Summary 126Additional Reading 127

4 Anxiolytic Drugs 129Introduction 129A General Approach to Using Medication with

Anxious Patients 130Pharmacology 131

Antidepressants 131Benzodiazepines 132Buspirone 136Beta-Blocker Medications 136Anticonvulsants 139Antipsychotics 139

Indications for Use 139Antidepressants 139Benzodiazepines 145Buspirone 146Beta-blocker Medications 146Anticonvulsants 147Antipsychotics 147Augmentation/Adjuvant Treatments 148

Drug Selection, Dose, and Initiation of Treatment 148Side Effects 150Drug Interactions 151Contraindications and Special Precautions 152Summary 153Additional Reading 154

CONTENTS xi

5 Sedative–Hypnotic Agents 155Introduction 155

Diagnosis 155Treatment Options 156Non-prescription Agents 157Prescription Medications 158

Pharmacology 159Benzodiazepines 159Chloral Hydrate 159Zolpidem 160Zaleplon 160Eszopiclone 160Ramelteon 161

Drug Selection 161Other Prescription Hypnotics 164

Treatment Implementation 166Adverse Effects 167

Benzodiazepines 167Chloral Hydrate 168Zolpidem 168Zaleplon 169Eszopiclone 169Ramelteon 169

Drug Interactions and Special Precautions 170Summary 172Additional Reading 172

6 Psychostimulants 175Introduction 175Pharmacology 175

Chemistry 175Mechanism of Action 176Pharmacokinetics 177Indications and Contraindications 180

Drug Selection 180Treatment Initiation and Dose Titration 183

Treatment Evaluation 185Maintenance Treatment 187Treatment Resistance 187Effects of Treatment on Symptoms 188

xii CONTENTS

ADHD in Adults 189Alternative Preparations to the Standard

Stimulant Medications 189Narcolepsy 190Other Indications 193Costs 193

Adverse Effects 193Drug Interactions 196Summary 197Additional Reading 198

7 Cognitive Enhancers and Treatments forAlzheimer’s Disease 199Introduction 199Pharmacology 199

Mechanism of Action 199Pharmacokinetics 203

Indications 205Drug Selection 205Treatment Initiation 206

Maintenance Treatment 207Treatment Evaluation 208

Adverse Effects 209Cholinesterase Inhibitors 209

Drug Interactions 211Summary 212Additional Reading 213

8 Drugs for Treating Substance Abuse Disorders 215Introduction 215Syndromes Associated with Intoxication 216

Alcohol Intoxication 216Sedative–Hypnotic Intoxication 217Opiate Intoxication 218Cocaine and Amphetamine Intoxication 219Intoxication by LSD, Mescaline, MDMA

(‘Ecstasy’), and Psilocybin 220Phencyclidine Intoxication 221

Drug Treatment of Withdrawal Syndromes 222Alcohol Withdrawal 223

CONTENTS xiii

Withdrawal from Sedative–Hypnotics 230Withdrawal from Opiates 233Management of Withdrawal in Patients with

Multiple Dependencies 237Agents to Aid Relapse Prevention 237

Medications for Alcohol Dependence 238Medications for Cocaine Dependence 241Medications for Opiate Dependence 242Special Considerations 247Drug Treatments for Nicotine Dependence 247

Pharmacotherapies for Substance Abusers withAdditional Psychiatric Illness 248

Pharmacotherapy for Specific PsychiatricDisorders 249

Drug Interactions in Chemical Dependency 251Additional Reading 252

Index 255

PREFACE

There is little question that this is the most exciting timein history to be involved in the treatment of patients withpsychiatric disorders. The explosive growth in our knowledgebase in all areas of the field, particularly in neuroscience,has revolutionized both our understanding of the nature ofpsychiatric illnesses and our ability to provide effective treat-ments. As molecular genetics and pharmacology, neurochem-istry, and new drug discovery techniques continue to advance,physicians are faced with the need to assimilate an ever-changing body of knowledge. In particular, the change in phar-macotherapy for psychiatric illnesses continues at a dizzyingpace.

We have been very gratified by the extremely positive inter-national response to our major textbook, Psychiatry, SecondEdition (Tasman, A, Kay, J, Lieberman, JA, Wiley, 2003).We believe, however, that busy clinicians need a quick refer-ence guide to the most up-to-date information on prescribingmedications for psychiatric illnesses. This book, the Hand-book of Psychiatric Drugs, is based on the outstanding chap-ters on pharmacotherapy in Psychiatry, Second Edition. Thematerial has been condensed, updated to just months beforepublication, and organized by specific classes of medications.To enhance the daily utility of this handbook, we chose aformat that emphasized ease of use, and ensured that eachchapter follows a specific template of topics, including thepharmacology, mechanism of action and pharmacokinetics,indications and methods of prescribing, side effects and druginteractions, and descriptions of each specific drug within theclass. We have aimed to include prescription medications incommon use anywhere in the world. Further, each chapter wasthen reviewed by highly respected psychiatrists with pharma-cotherapy expertise in that particular class of medication.

xvi PREFACE

We feel confident that you will find this book to be one ofyour most useful sources of information regarding your dailyclinical practice.

Jeffrey A. Lieberman, M.D.New York, NY

Allan Tasman, M.D.Louisville, KY

ACKNOWLEDGEMENTS

We would like to gratefully acknowledge the authors of thosechapters in Psychiatry Second Edition from which material inthis book was adapted.

Anri Aoba Antipsychotic DrugsRobert J. Boland AntidepressantsW. Wolfgang Fleischhacker Antipsychotic DrugsMarlene P. Freeman Mood StabilizersAlan J. Gelenberg Mood StabilizersLaurence L. Greenhill StimulantsJeffrey Halperin StimulantsMartin B. Keller AntidepressantsRachel E. Maddux Anxiolytic DrugsJohn March StimulantsStephen R. Marder Antipsychotic DrugsJohn Misiaszek Mood StabilizersSeiya Miyamoto Antipsychotic DrugsScott E. Moseman Mood StabilizersMark H. Rapaport Anxiolytic DrugsLon S. Schneider Cognitive Enhancers and

Treatments for Alzheimer’sDisease

Richard I. Shader Sedative–Hypnotic AgentsErin Shockey StimulantsDouglas A. Songer Sedative–Hypnotic AgentsPierre N. Tariot Cognitive Enhancers and

Treatments for Alzheimer’sDisease

We also acknowledge Jeffrey Selzer, author of the chapter‘Drugs for Treating Substance Abuse in Psychiatric Drugs’.

xviii ACKNOWLEDGEMENTS

We would also like to thank our colleagues from the College ofPhysicians and Surgeons of Columbia University, New York,who contributed to the final editing process:

D.P. DevanandLaurence GreenhillDavid J. HellersteinEdward V. NunesLazlo PappDavid J. PrintzZafar Sharif

DISCLAIMER

Notice

This book discusses new research, treatments, and drug ther-apies in the field of psychiatry. Readers are advised to checkthe product information currently provided by the manufac-turer of each drug to be administered to verify the recom-mended dose, the method and duration of administration, andthe contraindications. It is the responsibility of the treatingphysician, relying on experience and knowledge of the patient,to determine dosages and the best treatment for the patient.Neither the publisher nor the editors assume any responsibilityfor any injury and/or damage to persons or property.

The Publisher

1ANTIPSYCHOTIC DRUGS

INTRODUCTIONInformation about antipsychotic drugs and developments inthe treatment of psychosis is rapidly expanding. The advent ofnewersecond-generationantipsychotics in thewakeofclozapinerepresents the first significant advances in the pharmacologictreatment of schizophrenia and related psychotic disorders,and second-generation antipsychotics have become first-choiceagents for acute and maintenance therapy for these illnesses.

There is growing evidence that most of the new medi-cations can offer advantages over conventional neuroleptics;these include fewer extrapyramidal symptoms, lower risk oftardive dyskinesia, reduced cognitive impairment, and possibleimprovement in negative symptoms. Treatment successes havecontributed to the increased use of newer antipsychotic agentsand have also allowed psychiatrists to expand clinical expec-tations. In addition, these second-generation drugs are beingused increasingly for various conditions beyond schizophrenia,as happened with the conventional antipsychotics.

PHARMACOLOGYThe first-generation antipsychotic agents are equally effec-tive in the treatment of psychotic symptoms of schizophrenia,although they vary in potency and their propensity to inducevarious side effects. All have a high affinity for dopamine


2 HANDBOOK OF PSYCHIATRIC DRUGS

D2 receptors. In addition, all produce extra-pyramidal symp-toms (EPS), including parkinsonism, dystonia, akathisia, areassociated with a substantial risk of tardive dyskinesia (TD),and increase serum prolactin concentration in the usual clin-ical dose range. First-generation agents are usually classi-fied into three groups: phenothiazines, butyrophenones (e.g.,haloperidol), and others (e.g., thiothixene, molindone, andloxapine), based on their structure.

Second-generation antipsychotic drugs are characterized bythe following criteria:

• few or no EPS; significant reduction in tardive dyskinesialiability compared to first-generation antipsychotics

• expanded spectrum of therapeutic efficacy• less prolactin elevation

There is a continuum of typical and atypical effects in atypicaldrugs rather than several dichotomous groups.

ChemistryAntipsychotic drugs bind to numerous neurotransmitterreceptorsubtypes, includingthoseofdopamine,norepinephrine,epinephrine, acetylcholine, serotonin, and histamine. They actto antagonize the endogenous ligands at these receptors. Boththerapeutic and extrapyramidal side effects can be attributed tothe antagonism of dopamine at D2 receptors, with actions at theother neuroreceptors associated with various other side effects.

The typical antipsychotics have been described as being ofhigh (e.g., haloperidol), low (e.g., chlorpromazine), and mid(e.g., loxapine) potency on the basis of their degree ofaffinity for D2 receptors and their therapeutic dose range.Atypical antipsychotics are characterized by generally loweraffinities for D2 receptors and relatively greater affinitiesfor serotonin (5-hydroxytryptamine) 5-HT2A receptors inparticular, but also for noradrenergic receptors (�1 and�2�, muscarinic acetylcholine receptors, histamine, and otherdopamine (DA) subtype receptors. Aripiprazole is currentlythe only antipsychotic that acts as a partial agonist at D2 and5-HT1A receptors as well as an antagonist at the 5-HT2A andD2 receptors, and these properties are believed to accountfor its therapeutic effects. It has no appreciable activity at

ANTIPSYCHOTIC DRUGS 3

muscarinic receptors and modest affinity for alpha-1 adrenergicand histamine H1 receptors.

Mechanism of ActionThe therapeutic actions of antipsychotic drugs are generallyattributed to antagonism of DA receptors, particularly the D2

subtype. Atypical antipsychotics, with their lower D2 receptoraffinities and broader spectrum of pharmacologic properties,also antagonize 5-HT2A receptors, giving possible therapeuticadvantages and a superior motor side effect profile. At this pointit is unclear what clinical effects 5-HT2A antagonism confers,other than mitigating the adverse effect of striatal D2 antag-onism, and propensity to cause EPS. The low EPS liabilityof aripiprazole is at least in part related to its partial agonistactivity at the D2 receptor. Its D2 antagonist activity is broadlycomparable with that of haloperidol and chlorpromazine, but itclearly has weaker cataleptogenic activity. Furthermore, chronictreatment with aripiprazole is associated with much less upregu-lation of striatal D2 receptors compared with haloperidol.

What has been established is that as a consequence of theirdifferent pharmacologic profile, the atypical drugs have a muchwider separation of the dose-response curves of therapeuticantipsychotic action and extrapyramidal side effects.

PharmacokineticsAntipsychotic agents are rapidly absorbed from the gastroin-testinal tract and undergo extensive first pass metabolism.They are highly lipophilic, which results in ready transportacross the blood-brain barrier. Antipsychotics are metabolizedby the cytochrome P450 enzyme system. The isozyme systemspredominantly involved are CYP2D6, CYP1A2, CYP3A4, andCYP2C19, and medications that inhibit or compete for thesesubstrates can increase antipsychotic blood levels. After under-going various degrees of metabolism, antipsychotic drugs andtheir metabolites are glucuronidated in the liver and excretedby the kidney in the urine or in feces.

The average plasma half-life of the antipsychotics as afamily is approximately 20 to 24 hours, allowing for once-daily dosing. Aripiprazole and its active metabolite dehydro-aripiprazole have exceptionally long half-lives of 75 and 94


hours respectively and steady state concentrations are achievedafter 14 days. Some drugs have shorter half-lives (e.g., queti-apine: 6 to 12 hours; ziprasidone: 4 to 10 hours), which suggeststwice-daily administration. However, with repeated dosingthe pharmacodynamic effects may extend beyond the periodsuggested by pharmacokinetic parameters, allowing the consol-idation of dosing to once daily.

Among the second-generation antipsychotics, olanzapine(5–10 mg initial dose) and ziprasidone (10–20 mg initial dose)are available in a parenteral form for acute use in agitatedpatients, giving the benefits of a more rapid onset of actionand the ability to bypass the extensive first-pass metabolismthat these agents undergo.

Several of the antipsychotic drugs (three in the UnitedStates: haloperidol and fluphenazine decanoate, and Risperi-done – Risperdal Consta) are available in long-acting injectablepreparations for intramuscular administration. This allows forless fluctuation in plasma level compared to oral formula-tions, bypasses first-pass metabolism, and can improve patientcompliance.

Recommended dosages for second-generation antipsy-chotic agents are shown in Table 1-1.

INDICATIONS FOR USE OF ANTIPSYCHOTIC DRUGSAntipsychotic agents are effective for treating nearly everymedical and psychiatric condition where psychotic symptomsor aggression are present. They are currently used routinely inthe management of psychosis and/or agitation associated with:

• Schizophrenia and Schizoaffective Disorder• Acute manic and mixed episodes of bipolar disorder• Major depression with psychosis• Delusional disorder• Delirium• Dementia• Mental retardation• Developmental disorders (e.g., Autism)• Huntington’s disease• Tourette’s syndrome• Substance-induced psychoses (psychostimulants, phencycli-

dine, levodopa, steroids)

�TA

BLE

1-1.

Ant

ipsy

chot

icDr

ugs,

Dose

s,Fo

rms,

and

Cost

s

DRUG

CLA

SS

THER

APE

UTIC

DOSE

RAN

GE:

USUA

LO

RAL

DAIL

YDO

SE(m

g/d)

THER

APE

UTIC

EQUI

VALE

NT

ORA

LDO

SE(m

g/d)

FORM

SH

ALF

-LIF

E(h

)

Fir

st-g

ener

atio

nan

tips

ycho

tics

Per

phen

azin

e8–

6410

Tab

lets

(2,4

,8,1

6m

g)9

Con

cent

rate

(16

mg/

5m

l)In

ject

able

solu

tion

(5m

g/m

l)C

hlor

prom

azin

e20

0–1,

000

100

Tab

lets

(10,

25,5

0,10

0,20

0m

g)23

–37

Con

cent

rate

(30,

100

mg/

ml)

Gel

Cap

s(3

0m

g)Sp

ansu

les

(30,

75,1

50m

g)In

ject

able

solu

tion

(25

mg/

ml)

Rec

tal

supp

osit

orie

s(2

5m

g)H

alop

erid

ol2–

202

Tab

lets

(0.5

,1,2

,5,1

0,20

mg)

24C

once

ntra

te(2

mg/

ml)

Inje

ctab

leso

luti

on(5

mg/

ml)

Seco

nd-g

ener

atio

nan

tips

ycho

tics

Ris

peri

done

2–8

1–2

Tab

lets

(0.2

5,0.

5,1,

2,3,

4m

g)P

aren

t3

Con

cent

rate

(1m

g/m

l)M

-tab

s(0

.5,

1,2

mg)

rapi

dly

disi

nteg

rati

ngta

blet

sR

ispe

rdal

Con

sta

long

-act

ing

inje

ctab

le(2

5,37

.5,5

0m

g)q

2w

k)

Met

abol

ite

24

�TA

BLE

1-1.

(Con

tinue

d)

DRUG

CLA

SS

THER

APE

UTIC

DOSE

RAN

GE:

USUA

LO

RAL

DAIL

YDO

SE(m

g/d)

THER

APE

UTIC

EQUI

VALE

NT

ORA

LDO

SE(m

g/d)

FORM

SH

ALF

-LIF

E(h

)

Zip

rasi

done

80–2

0020

Cap

sule

s(2

0,40

,60,

80m

g)In

ject

able

solu

tion

20m

g/m

l4–

10

Clo

zapi

ne25

–600

50T

able

ts(2

5,10

0m

g)8–

12O

lanz

apin

e5–

302.

5–5

Tab

lets

(2.5

,5,7

.5,1

0,15

,20

mg)

Zyd

is(5

,10,

15,

20m

g)ra

pidl

ydi

sint

egra

ting

tabl

ets

Inje

ctab

leso

luti

on5

mg/

ml

27

Que

tiap

ine

150–

800

50–1

00T

able

ts(2

5,50

,100

,200

,300

,400

mg)

7A

ripi

praz

ole

10–3

05–

10T

able

ts(5

,10,

15,2

0,30

mg)

75–9

6O

ral

solu

tion

1m

g/m

lL

ong-

acti

ngpr

epar

atio

nsF

luph

enaz

ine

deca

noat

e12

.5–5

0q

2–4

wk

12.5

–50

Inje

ctab

leso

luti

on(2

5m

g/m

l)2–

3w

k

Hal

oper

idol

deca

noat

e50

–300

q3–

4w

k50

–300

Inje

ctab

leso

luti

on(1

00m

g/m

l)3

wk

Ris

perd

alC

onst

a25

–50

mg

q2

wk

25,3

7.5,

50m

gM

icro

sphe

res

infi

xed

dose

vial

s;2

ml

3–6

days

Ada

pted

and

repr

inte

dfr

omL

iebe

rman

JAan

dM

ende

low

itz

AJ

(200

0).

Ant

ipsy

chot

icD

rugs

,D

oses

,F

orm

s,an

dC

osts

.In

Psy

chia

tric

Dru

gs,L

iebe

rman

JAan

dT

asm

anA

(eds

.)W

BSa

unde

rs,p

p.1–

43.©

2000

,wit

hpe

rmis

sion

from

Els

evie

r.


The development of second-generation antipsychotics has beena major clinical advance for the treatment of schizophrenia.At present, these drugs are used as the first-line treatmentfor schizophrenia, and are being used increasingly for variousconditions beyond schizophrenia. The low incidence of EPSand TD associated with second-generation agents is highlybeneficial in several neuropsychiatric conditions.

DRUG SELECTION AND INITIATION OF TREATMENTDrug Selection for the Treatment of SchizophreniaIn choosing an antipsychotic agent for the treatment of anepisode of schizophrenia, the clinician should be guided by thepatient’s history. The following tests should be performed:

• a complete physical examination including weight• a review of symptoms• routine blood chemistry including fasting blood glucose and

lipid profile• complete blood cell count• electrocardiogram

Once any medical or neurologic causes of the symptoms havebeen ruled out, the clinician should consider the followingquestions:

• Is this a psychotic episode consistent with schizophrenia?• Have affective syndromes such as mania and depression with

psychotic features been ruled out?• Is this the patient’s first episode?• Is there a history of prior antipsychotic treatment response?• How well was the antipsychotic tolerated?• Were there prominent side effects?• Did the patient ever previously have EPS or TD?• Does the patient have a history of neuroleptic malignant

syndrome?• What symptoms (positive or negative) are predominant in

the episode?


• Is there a history of non-compliance, and, if so, has thepatient had a trial of depot preparations?

• If there have been antipsychotic trials that have failed, werethe trials of an adequate dose and of an adequate duration?

• Does the patient meet the criteria for treatment resistance?• Does the patient have a history of cardiac conduction delay?• Does the patient have any prior history of blood dyscrasias?• Does the patient have normal hepatic and renal functions?• Is the patient medically debilitated?

TREATMENT INITIATION AND DOSE TITRATIONOnce the questions above are answered, the clinician shouldselect an antipsychotic medication and initiate the treat-ment trial. The algorithm from the practice guidelines forschizophrenia developed by the American Psychiatric Associ-ation may be useful for drug selection (Figure 1-1).

Selection of an agent in emergency settings for the manage-ment of the gross agitation, excitement, and violent behaviorassociated with psychosis might be based on clinical symptoms,differences in efficacy or side effects of candidate drugs, or,more pragmatically, the formulation of a drug as it affects routeof administration, onset, and duration.

There is now a considerable amount of clinical experiencewith atypical antipsychotics in acute emergency situations, andthey have come to replace first-generation antipsychotics asthe agents of choice when treatment is initiated. Olanzapineand ziprasidone are available in short-acting formulations forintramuscular administration; risperidone and aripiprazole areavailable as oral solutions; and olanzapine and risperidone areavailable as rapidly disintegrating oral tablets. These variousformulations provide tremendous flexibility to the clinician inchoosing the optimal medication and method of administrationbased on clinical considerations.

The use of oral second-generation antipsychotics andbenzodiazepines in combination is the most common medica-tion strategy in psychiatric emergency settings. It is best toavoid combining antipsychotics in favor of sequential trialsof monotherapy with different antipsychotics. Because mostpatients with schizophrenia will require long-term treatmentwith antipsychotics it is imperative that, in addition to short-term treatment goals of control of behavioral dysregulation


Acute Phase

Yes

Yes

Yes

No

No

No

Good responsewithout intolerable

side effects?


side effects?


side effects?

For intolerable side effects:choose a different medicationfrom Group 1 or 2 (refer toTables 2 and 3).


For inadequate therapeuticresponse: choose a differentmedication from Group 1, 2,or 3 (refer to Table 3).

Stabilization orMaintenance Phase


For treatment nonadherence:consider a different medicationfrom Group 4.

Choose medication basedon clinical circumstancesfrom following (refer toTables 3 and 4):

Group 1: First-generation agentsGroup 2: Risperidone, olanzapine, quetiapine, ziprasidone, aripiprazoleGroup 3: ClozapineGroup 4: Long-acting injectable antipsychotic agents

For inadequate therapeuticresponse: choose a differentmedication from Group 1, 2, or 3.For persistent psychotic symptons,clozapine should be given strongconsideration. Consider ECT forpatients with persistent severepsychosis, catatonia, and/or suicidalideation or behavior for whom priortreatments including clozapine havefailed.

Continue acute-phase medication treatment. Consider maintenance ECT forpatients who have responded to an acute course of ECT and whose symptomscannot be controlled with medication maintenance therapy alone.

For residual or intercurrentpositive, negative, cognitive,or mood symptoms:consider a different medicationfrom Group 2 or 3 or appropri-ate adjunctive medication.

FIGURE 1-1. Somatic Treatment of Schizophrenia. Also refer toTable 1-1 and the following: First episode–Group(G) 2; Persis-tent suicidal ideation or behavior–G3; Persistent hostility andaggressive behavior–G3; Tardive dyskinesia–G2 all group 2 drugsmay not be equal in their lower or no tardive dyskinesialiability and G3; History of sensitivity to extrapyramidal sideeffects–G2 except higher doses of risperidone; History of sensi-tivity to prolactin elevation–G2 except risperidone; History ofsensitivity to weight gain, hyperglycemia, or hyperlipidemia–G2ziprasidone or aripiprazole; Repeated nonadherence to phar-macological treatment–G4 (taken from the Practice Guidelinesfor the Treatment of Patients with Bipolar Disorder, SecondEdition from the American Psychiatric Association Practice Guide-lines for the Treatment of Psychiatric Disorders Compendium,Copyright 2004).


and psychotic symptom resolution, clinicians consider long-term side effect profiles of the available treatment options withthe goal of minimizing long-term side effect burden. Ziprasi-done and aripiprazole may be specifically indicated in patientswho are intolerant of the side effects that can occur in greaterfrequency with some of the second-generation drugs, such asweight gain and alterations in glucose and lipid metabolism, asthey do not produce these effects to any significant degree.

One can safely escalate the dose of second-generationantipsychotics more rapidly than is usual in outpatient settingsto achieve target doses typically utilized for the treatment ofschizophrenia. Once behavioral control is achieved, benzo-diazepines should be discontinued and the patient shouldbe maintained on the atypical antipsychotic alone. The doserecommendations for second-generation antipsychotic drugsare summarized in Table 1-2.

The anti-aggressive characteristics of clozapine are wellestablished in chronically psychotic patients; however cloza-pine initiation is contraindicated in the psychiatric emergencysetting because of its serious potential side effects, includingseizures and agranulocytosis.

On the rare occasion that a physician elects to use a first-generation antipsychotic during the first few days of treat-ment, “rapid neuroleptization” should be avoided as there is noevidence for increased efficacy and risk of side effects is greaterat higher doses. If a low-potency agent is chosen, the recom-mendation is to begin with a low dose such as chlorpromazine,50 mg twice a day, and to titrate slowly so that the difficultiesassociated with orthostatic hypotension can be reduced. If ahigh-potency agent such as haloperidol is chosen, the recom-mendation is also to begin a course of prophylactic antiparkin-sonism medication such as benztropine, 1 mg twice a day, todecrease the incidence of EPS side effects. The initial dose ofantipsychotic should be titrated to between 5 and 10 haloperidolequivalents or 200 to 400 chlorpromazine equivalents. At thistime, the only groups of patients in which the first-generationantipsychotics are clearly preferable are those who have a historyof good response to these agents with minimal side effects.

After initiating treatment and titrating to a standard doseof antipsychotic, the clinician, patient, and family must wait forthe antipsychotic to take effect. Because patients are acutely

�TA

BLE

1-2.

Reco

mm

ende

dDo

sage

sfo

rSe

cond

-Gen

erat

ion

Ant

ipsy

chot

icA

gent

s

AVE

RAG

EDO

SERA

NG

E(m

g/da

y)H

ALF

-LIF

E(h

r)(M

EAN

)ST

ART

ING

DOSE

(TO

TAL

mg/

day)

FIRS

TEP

ISO

DERE

CURR

ENT

EPIS

ODE

AVE

RAG

EM

AIN

TEN

AN

CEDO

SE(m

g/da

y)RO

UTES

OF

ADM

INIS

TRA

TIO

N

Clo

zapi

ne10

–105

(16)

25–5

015

0–30

040

0–60

040

0O

ral

Ris

peri

done

3–24

(15)

1–2

2–4

3–6

3–6

Ora

l,de

pot

Ola

nzap

ine

20–7

0(3

0)5–

1010

–20

15–3

010

–20

Ora

l,IM

Que

tiap

ine

4–10

(7)

50–1

0030

0–60

050

0–80

040

0–80

0O

ral

Zip

rasi

done

4–10

40–8

080

–120

120–

200

120–

160

Ora

l,IM

Zot

epin

e12

–30

(15)

50–1

0075

–150

150–

450

75–3

00O

ral

Am

isul

prid

e8–

20(1

2)50

–100

50–3

0040

0–80

040

0–80

0O

ral

Ari

pipr

azol

e(7

5–96

)10

–15

10–3

015

–30

15–3

0O

ral

So

urc

e:A

dapt

edfr

omM

cEvo

yJP

,Sch

eifl

erP

L,F

ranc

esA

(199

9)T

heex

pert

cons

ensu

sgu

idel

ine

seri

es:T

reat

men

tof

schi

zoph

reni

a.J

Cli

nP

sych

iatr

60,1

–80;

Bur

nsM

J(2

001)

The

phar

mac

olog

yan

dto

xico

logy

ofat

ypic

alan

tips

ycho

tic

agen

ts.J

Cli

nT

oxi

col

39,1

–14;

Wor

rel

JA,

Mar

ken

PA

,Bec

kman

SE,e

tal

.(20

00)

Aty

pica

lan

tips

ycho

tic

agen

ts:A

crit

ical

revi

ew.A

mJ

Hea

lth

Sys

tP

har

m57

,238

–255

.


ill, there is often a tremendous temptation and external pres-sure to increase the dose of the antipsychotic in the hope thatthe patient’s condition will improve more rapidly. Despite thishope, there is little, if any, clinical evidence that a higher doseof antipsychotic is in any way advantageous; in fact, it will onlyincrease the likelihood of side effects. During this difficult timeit may be necessary to add sedating medications such as short-acting benzodiazepines (e.g., lorazepam) to help the patientmaintain control until the antipsychotic has started to work.

TREATMENT EVALUATION AND DRUG SWITCHINGIf the first-episode patient has failed to respond to a 6-weektrial of an antipsychotic, the clinician should evaluate possiblenon-compliance with medication, the likelihood of a partialresponseoracompletenonresponse to treatment. If therewasnoresponse, a change to a second antipsychotic from a new family isrecommended. If one of the newer agents was not the clinician’sfirst choice, it should be used at this point in the decision tree.

If a patient has already discontinued use of a medication,then the new treatment is selected and initiated as describedabove. However, if the patient is undergoing maintenance drugtreatment and drugs are to be electively switched in the hopeof achieving a better therapeutic response or alleviating drugside effects, then the goal is to switch medications withoutdestabilizing the patient.

Medication changes should be performed by a concurrentslow tapering of the initial antipsychotic while the secondantipsychotic is being slowly titrated. The specific rate of cross-titration depends on the dose of the old medication and therelative stability of the patient. In general, the higher the doseand the more unstable the patient, the longer and more gradualthe cross-titration schedule. Although this varies, a rule ofthumb is to cross-titrate by yoked increments and decrementsof 25% every 2 to 5 days. Adjunctive medications should beadjusted or tapered accordingly.

If the patient is judged to be a partial responder to theantipsychotic trial, then the clinician may consider the additionof an agent for augmentation. At this point, it is again impor-tant for the clinician to re-evaluate the presence of affectivesymptoms. If there is significant depressive symtomatology in


the clinical presentation, then the addition of an antidepressantmay be warranted. If the presence of mood symptoms is consis-tent with a manic episode, the addition of a mood stabilizer suchas lithium as an augmentation strategy may be clinically useful.

Treatment During the Resolving Phase If a particularantipsychotic medication has improved the acute symptoms itshould be continued at the same dose for the next six monthsbefore a lower maintenance dose is considered for continuedtreatment. Rapid dose reduction or discontinuation of themedications during the resolving phase may result in relativelyrapid relapse. It is essential to continue to assess side effectspresent in the acute phase and modify treatment to minimizetheir negative impact, and to re-evaluate the necessity of anyadjunctive therapies used in the acute phase.

If the decision has been made to switch to a long-actingdepot antipsychotic agent, this can often be achieved during thisphase. This is also a good time to educate the patient and familyregarding the course and outcome of schizophrenia, as wellas factors that influence the outcome such as drug compliance.

MAINTENANCE TREATMENTThe goals of treatment during the stable or maintenancephase are to maintain symptom remission, to prevent psychoticrelapse, to implement a plan for rehabilitation, and to improvethe patient’s quality of life.

Current guidelines recommend that first-episode patientsshould be treated for one to two years; however, 75% ofpatients will experience relapses after their treatment is discon-tinued. Patients who have had multiple episodes should receiveat least five years of maintenance therapy. Patients with severeor dangerous episodes should probably be treated indefinitely.

Gradual dose reduction to identify the minimum effectivedose for the patient can be attempted in this phase, althoughrelapse rates are excessively high when doses are reduced toabout 10% of the acute dose.

Antipsychotics have been proven to be effective in reducingthe risk of psychotic relapse in maintenance therapy forschizophrenia. In the stable phase of illness, antipsychoticscan reduce the risk of relapse to less than 30% per year.


Without maintenance treatment, 60–70% of patients relapsewithin one year and almost 90% relapse within two years.These results indicate that antipsychotic medications are effec-tive in preventing relapse in most stabilized patients. Thereis also strong evidence that patients who relapsed while onantipsychotic medications had episodes that were less severethan patients not on antipsychotic drugs.

As atypical drugs have fewer EPS side effects, patients onthese compounds may be less likely to be non-compliant withtreatment and may thereby decrease their risk of relapse. Ithas also been suggested that in addition to fewer side effects,atypical drugs may have inherently greater prophylactic effi-cacy than typical drugs and therefore be better for patientswith an increased risk of relapse.

TREATING TREATMENT RESISTANCETreatment resistance is generally defined as a failure of twoprior drug trials of 4–6 weeks duration. Although most defini-tions of treatment resistance focus on the persistence of posi-tive symptoms, there is growing awareness of the problemsof persistent negative symptoms and cognitive impairments,which may have an important impact on level of functioning,psychosocial integration, and quality of life.

Approximately 10 to 15% of patients with first-episodeschizophrenia are resistant to drug treatment, and between25% and 50% of long-term patients will have severe, persistentsymptoms including psychosis.

Only clozapine has consistently demonstrated efficacyfor psychotic symptoms in well-defined treatment refractorypatients; the mechanism responsible for this therapeutic advan-tage remains uncertain. Serum levels of 350 �g/mL or greaterhave been associated with maximal likelihood of response.Depending on the type of residual symptom, augmentationstrategies includeaddinganotherantipsychotic,anticonvulsants,benzodiazepines, and cholinergic agonists may prove useful.

Since the approval of clozapine, attention has shifted toa greater focus on the use of other second-generation antipsy-chotics for managing treatment resistance in schizophrenia.Both olanzapine (15–25 mg/day) and clozapine (200–600mg/day) were shown to be similarly effective in reducing


overall psychotic symptoms in treatment-resistant patients clin-ically eligible for treatment with clozapine. Some preliminaryreports suggest that higher doses of olanzapine may be moreeffective; however, dosage issues of olanzapine have not yetbeen adequately addressed in more controlled conditions. Thereare also several recent reports of beneficial effects of quetiapinein treatment-resistant patients with schizophrenia.

Given the risk of agranulocytosis, the burden of sideeffects, and the requirement of white blood cell monitoring,the second-generation agents (risperidone, olanzapine, queti-apine, ziprasidone, and aripiprazole) should be tried in almostall patients before proceeding to clozapine. Many cliniciansexpress the impression that certain patients do respond pref-erentially to a single agent of this class.

SCHIZOAFFECTIVE DISORDER AND VIOLENT PATIENTSAmong the specific therapeutic effects claimed for atypicaldrugs is their ability to alleviate mood symptoms associatedwith the psychotic disorder. Although this has not been defini-tively proved, preliminary results indicate that mood symp-toms may selectively abate with atypical drugs. This evidencesuggests that patients with schizoaffective disorder, residualmood symptoms (e.g., postpsychotic depression), a history of,or current, suicidal behavior, and violent behavior may benefitmost from treatment with an atypical drug as compared to aconventional antipsychotic agent.

ADJUNCTIVE TREATMENTSFor patients who are unresponsive to antipsychotic agents,including clozapine, and for patients who are responsive buthave substantial residual symptoms, the question is what furtheroptions exist. Adjunctive medications as indicated in the algo-rithm (other than electroconvulsive therapy) have been usedextensively but without any empiric data to demonstrate theirefficacy. These adjuncts include anticonvulsants, lithium, antide-pressants, benzodiazepines, and cholinesterase inhibitors.

Effects of Antipsychotic Agents on Symptoms ofSchizophreniaThe clinical profile of second-generation antipsychotic agents onthe symptoms of schizophrenia are summarized in Table 1-3.

�TA

BLE

1-3.

Clin

ical

Prof

ileof

Seco

nd-G

ener

atio

nA

ntip

sych

otic

Drug

Effic

acy

DRUG

CLO

ZAPI

NE

RISP

ERID

ON

EO

LAN

ZAPI

NE

QUE

TIA

PIN

EZI

PRA

SIDO

NE

SERT

INDO

LEA

MIS

ULPR

IDE

ARI

PIPR

AZO

LEIL

OPE

RIDO

NE

Clin

ical

effe

ctP

sych

otic

sym

ptom

s++

+++

+++

+++

++++

+++

+++

+++

+

Neg

ativ

esy

mpt

oms

++

++

+++

++++

++

Cog

niti

vesy

mpt

oms

++++

+++

??

?++

?

Moo

dsy

mpt

oms

+++

++++

+++

+++

++++

++?

Ref

ract

ory

sym

ptom

s++

+++

++++

??

++?

?

+to

+++,

wea

kly

tost

rong

lyac

tive

;?,q

uest

iona

ble

toun

know

nac

tivi

ty.

So

urc

e:A

dapt

edfr

omD

awki

nsK

,Lie

berm

anJA

,Leb

owit

zB

D,e

tal

.(19

99)

Ant

ipsy

chot

ics:

Pas

tan

dfu

ture

.Nat

iona

lIn

stit

ute

ofM

enta

lH

ealt

hD

ivis

ion

ofSe

rvic

esan

dIn

terv

enti

onR

esea

rch

Wor

ksho

p(J

uly

14,1

998)

.Sch

izo

ph

rB

ull

25,3

95–4

04,w

ith

perm

issi

onfr

omO

xfor

dU

nive

rsit

yP

ress

.


Positive Symptoms Antipsychotic agents have a specificeffect on positive symptoms of schizophrenia including halluci-nations, delusions, and thought disorder. Approximately 30%of patients with acutely exacerbated psychotic symptoms havelittle or no response to conventional antipsychotics, and upto 50% of patients have only a partial response to medica-tion. Although the proportion of patients who improve andthe magnitude of therapeutic effects vary greatly, second-generation antipsychotics appear to be at least as effective forpsychotic symptoms as conventional drugs.

Negative Symptoms Studies of the early course of illnesshave shown that about 70% of schizophrenics develop primarynegative symptoms such as affective blunting, emotional with-drawal, poverty of speech, anhedonia, and apathy, before theonset of positive symptoms. Negative symptoms may representcore features of the illness, and may be associated with pooroutcome and prolonged hospitalization for patients.

Negative symptoms can be divided into three componentsthat are usually difficult to distinguish:

• primary or deficit – enduring negative symptoms• primary – non-enduring negative symptoms• secondary negative symptoms that may be associated with

psychotic symptoms, EPS, depression, and environmentaldeprivation

Conventional antipsychotics are generally less effective againstnegative than positive symptoms of schizophrenia; thus, theefficacy of second-generation antipsychotics on negative symp-toms compared with that of first-generation drugs has receivedmuch attention. Second-generation agents such as cloza-pine, olanzapine and risperidone demonstrate significantlygreater efficacy than conventional agents in reducing negativesymptoms.

However, there is a continuing debate as to whether theseeffects are related to a reduction in EPS or to a direct effecton primary negative symptoms. Moreover, the effect sizesof improvement on negative symptoms for second-generationagents are usually moderate to small in comparison with


placebo or conventional agents. Path analyses have suggestedthat both risperidone and olanzapine exert direct effects on(primary) negative symptoms independent of differences inpsychotic, depressive, or extrapyramidal symptoms. A collabo-rative working group concluded that second-generation drugsare superior in terms of the “totality” of negative symptoms,but their impact on specific components is still under investi-gation. This and other clinical questions will become clear asthe new agents are tested in clinical trials.

Cognitive Symptoms Cognitive impairment appears to be anintegral characteristic of schizophrenia and may be evident inup to 60% of patients. Measurable deficits are prominent intasks involving attention, verbal fluency, memory, and exec-utive function. A wide range of cognitive deficits are usuallypresent at the time of the first psychotic episode and remainrelatively stable or only slowly progressive during the courseof the illness, independent of psychotic symptoms. Cognitivedeficits are particularly prominent in patients meeting criteriafor the deficit syndrome and in patients with TD. They aremore strongly related to social and vocational functioning thanpsychotic symptoms and may influence the quality of life ofpatients. Thus, targeting cognitive impairments appears to bea major focus of the treatment of schizophrenia.

Conventional neuroleptics produce small and inconsistenteffects on cognitive functioning. In general, clozapine, risperi-done, and olanzapine have demonstrated superior efficacycompared to first-generation antipsychotics on tests of verbalfluency, digit–symbol substitution, fine motor function, andexecutive function. Measures of learning and memory wereleast affected by second-generation agents. Because these testsall measure performance during a timed trial, enhanced perfor-mance with second-generation drugs could result, in part,from reduced parkinsonian side effects. Preliminary evidencesuggests that risperidone may be more effective for visual andworking memory than clozapine.

Mood Symptoms and Suicidal Behavior Depressive symp-toms frequently occur in the context of psychotic symptomsor intercurrently between psychotic episodes. Antidepressant


medication used adjunctively to antipsychotic drugs is gener-ally indicated and effective. Atypical antipsychotics have beenreported to have selective benefits against mood symptoms inschizophrenia, both manic and depressive.

Suicidal behavior presents a particular problem in patientswith schizophrenia. Clozapine is approved for use in suicidalpatients with schizophrenia on the basis of results in theInterSePT study. This study found that clozapine treatmentproduced a lower rate of suicidal behavior than the compar-ison treatment of olanzapine in patients with active or historiesof suicidal behavior.

Drug Selection for the Treatment of BipolarDisorderAntipsychotic agents are effective in treating acute manicepisodes; these agents are believed to possess antimanic qual-ities in addition to their antipsychotic properties. One benefitis their rapid onset compared to mood stabilizers; as a result,these agents are often used preferentially or combined untilthe mood stabilizer has reached its therapeutic effectiveness.All the second-generation antipsychotics (except clozapine)are now approved for the treatment of acute manic episodein the United States, as well as the acute treatment of mixedepisodes (except clozapine and quetiapine). Olanzapine andaripiprazole are also approved for maintenance treatment.

A concern with the use of antipsychotics in this popula-tion is the potential for TD. As a result, it is recommendedthat atypical antipsychotics be used with this populationwhen clinically indicated but that attempts be made to treatthis population with mood-stabilizing agents by themselves ifpossible.

Drug Selection for the Treatment of MajorDepression With Psychotic FeaturesClear psychotic symptoms, such as delusions or hallucinations,are observed in approximately 25% of patients with majordepressive disorder. These symptoms often respond poorly toantidepressants when they are administered alone, and usuallyrequire the use of adjunctive antipsychotic agents.


Treatment can be initiated simultaneously, though manyclinicians prefer to start the antipsychotic dose first and thenadd the antidepressant to the regimen. Though there arelimited data on the adequate dose of antipsychotic for thisgroup, most clinicians would recommend 5 to 10 haloperidolequivalents. This group of unipolar depressed patients may beat the highest risk of TD; thus the antipsychotic dosage shouldbe tapered and then discontinued when the patient’s psychoticsymptoms remit.

Drug Selection for the Treatment of DelusionalDisorderDelusional disorder differs from other psychotic disorders interms of family history and age distribution. In addition, ithas displayed a difference in treatment response; as a generalrule, patients with delusional disorder do not respond wellto antipsychotic agents. Some uncontrolled clinical data havesuggested that these patients may do better with drugs fromthe diphenylbutylpiperidine class (e.g., pimozide). However,the majority of this group of patients are untreated and do notseek psychiatric help. There is only very limited experiencewith atypical drugs in this population.

Drug Selection for the Treatment of DeliriumAntipsychotics are effective in treating the psychotic symp-toms and agitation associated with deliria of various etiologies.In treating a delirium, high-potency agents are preferable tolow-potency agents because low-potency agents usually havemore anticholinergic properties and cardiovascular side effects,which can adversely affect a delirious patient. Antipsychoticdrugs are commonly given parenterally when used for thisindication, including by intravenous routes.

Risperidone is also relatively free from anticholinergic sideeffects and has a favorable side-effect profile in relation tothe production of EPS. The newer agents olanzapine andquetiapine are now being utilized for these conditions. Theparenteral forms of the atypical drugs should be particularlyuseful for this indication.


Drug Selection for the Treatment of Psychosis andAgitation Associated with DementiaAntipsychotics are effective in treating the psychotic symptomsthat are often associated with dementias. Additionally, theyhave been demonstrated to have anti-aggressive and calmingeffects against dysregulated behavior and affect. Althoughmany patients with dementia have agitation and behavioraldisturbances that clearly require the use of antipsychotic drugs,these drugs should be used judiciously.

The atypical drugs offer several potential advantages overtypical drugs in treating dementia. They produce fewer EPSand less TD, side effects to which elderly patients are highlysusceptible. They also may have broader efficacy against theconstellation of pathologic symptoms and behaviors (e.g.,mood symptoms, hostility) that occur in dementia. To date,extensive placebo-controlled trials have been conducted withrisperidone, olanzapine and aripiprazole. Other atypical drugsmust be systematically evaluated to determine their efficacy forthis disorder and to determine how well their antiadrenergicand anticholinergic properties are tolerated.

Elderly patients with dementia who are treated with anantipsychotic agent are at increased risk of death. Analyses ofclinical trials with a modal duration of 10 weeks suggest thatthe excess risk is 1.6–1.7 compared with placebo. Most of thedeaths appear to be due to cardiovascular events (includingheart failure and sudden death) or infection. Because theseevents have been associated with antipsychotics regardless oftheir chemical structure, this is probably a class effect associ-ated with their pharmacological activity and the risk appliesalso to both atypical and older antipsychotics and to drugsthat were not included in these trials. The FDA has remindedprescribers that these agents are not approved for the treat-ment of dementia in older people.

Drug Selection for the Treatment of MentalRetardation and Developmental DisordersPatients with mental retardation are another patient popu-lation with psychotic symptoms and behavioral disturbances


about whom there has been controversy. As with patientswith dementia, a number of these patients have documentedpsychosis, and for these patients the use of antipsychotics isclearly indicated. There are other patients, however, whoseprimary symptoms are those of behavioral dyscontrol. Inthis group it is possible that the risks of antipsychotics mayoutweigh their benefits, especially in long-term treatment.Again, atypical drugs may be advantageous because of theirlower EPS and TD liabilities, to which these patients are highlysusceptible.

Drug Selection for Huntington’s Disease andTourette’s DisorderAntipsychotics have been shown to be effective in the manage-ment of Huntington’s disease and Tourette’s disorder. InHuntington’s, various antipsychotics have been used to helpcontrol the agitation and chorea as well as the psychoticsymptoms associated with the disorder. In Tourette’s disorder,the antipsychotics used most extensively to manage patients’vocalizations and tics include haloperidol and pimozide. Mostrecently, risperidone has shown promise in this patient popula-tion. Beyond risperidone, there is little experience with atypicaldrugs in these disorders. The exception to this is clozapine,which was found to have little therapeutic benefit in eithercondition.

On the basis of this evidence, the typical drugs might bethe preferred therapeutic option in these disorders; nonstriatalweak D2 agents would be less likely to be effective.

Substance-induced PsychosesPSYCHOSTIMULANT- AND PHENCYCLIDINE-INDUCED PSYCOSESSubstance intoxication resulting in psychotic symptoms maybe treated effectively with antipsychotic drugs. This is partic-ularly the case with intoxication from psychostimulants suchas amphetamine, methamphetamine, and cocaine and fromphencyclidine. Previously the clinical approach was not to usetypical neuroleptics for fear of exacerbating the patient’s condi-tion with side effects but, rather, to employ benzodiazepines


and a calm low-stimulus environment, unless the conditionpersisted for days or well beyond the period of the toxin’selimination. With the availability of the atypical drugs andtheir ability to alter the effects of NMDA receptor antagonistsas well as psychostimulants, the use of these agents shouldbe evaluated. At present, however, only limited data on theirefficacy in these conditions are available.

LEVODOPA-INDUCED AND STEROID-INDUCED PSYCHOSISAntipsychotics are an integral part of the treatmentof medication-induced psychotic syndromes. The psychosisinduced by levodopa in the treatment of Parkinson’s diseasepresents unique clinical dilemmas. Treatment of the symp-toms with first-generation antipsychotic agents will by defini-tion worsen the Parkinson’s symptoms. The clinician is oftencaught between attempts to reduce the patient’s severe para-noid state and attempts to keep the patient from becomingmore immobile from worsening rigidity and akinesia. Recently,case reports utilizing clozapine and risperidone in this popula-tion have shown encouraging results.

Steroid-induced psychotic symptoms have proved tobe somewhat more complicated: psychotic symptoms may beprolonged, requiring the use of antipsychotics, and at thesame time, despite the emergence of psychosis, some patientsmay still require steroid treatment for their medical condition.There have been a few case reports in which patients knownto become psychotic during a steroid course were pretreatedwith antipsychotics or with a mood stabilizer, with goodresults.

ADVERSE EFFECTS OF ANTIPSYCHOTICSAntipsychotics as a group have a wide range of potential sideeffects corresponding to their pharmacologic properties. Atyp-ical and typical drugs vary markedly in their side-effect profiles;clozapine has the most complicated and potentially serious sideeffects. The side effects for representative high-, mid-, and low-potency typical drugs and individual atypical drugs are shownin Table 1-4.

�TA

BLE

1-4.

Side

-Effe

ctPr

ofile

ofSe

cond

-Gen

erat

ion

Ant

ipsy

chot

icDr

ugs

DRUG

CON

VEN

TIO

NA

LA

GEN

TSCL

OZA

PIN

ERI

SPER

IDO

NE

OLA

NZA

PIN

EQ

UETI

API

NE

ZIPR

ASI

DON

ESE

RTIN

DOLE

AM

ISUL

PRID

EA

RIPI

PRA

ZOLE

ILO

PERI

DON

E

Side

effe

ctE

PSa

+++

0++

+0

+0

+++

+T

D++

+0

+++

0+

0to

++

++

NM

S++

++

+?

++

+?

?P

rola

ctin

elev

atio

n++

+0

+++

0to

+0

0to

+0

to+

++0

0to

+W

eigh

tga

in+

to++

+++

+++

++

0+

+0

?P

rolo

nged

QT

a+

to++

+0

+0

+++

+++

+0

0H

ypot

ensi

ona

+to

++++

++

++++

++

0+

+Si

nus

tach

ycar

diaa

+to

+++

+++

+++

+++

+0

0+

Ant

icho

liner

gic

effe

ctsa

+to

+++

+++

0++

+0

00

00

Hep

atic

tran

sam

init

is+

to++

+++

+++

++

+0

+

Agr

anul

o-cy

tosi

s0

to+

++0

00

00

00

0Se

dati

on+

to++

+++

++

++++

++

++

++

Seiz

ures

a0

to+

+++

00

to+

0to

+0

to+

0to

+0

to+

0to

+0

to+

EP

S,ex

trap

yram

idal

side

effe

cts;

TD

,tar

dive

dysk

ines

ia;N

MS,

neur

olep

tic

mal

igna

ntsy

ndro

me.

+to

+++,

acti

veto

stro

ngly

acti

ve;0

,min

imal

tono

ne;?

,que

stio

nabl

eto

unkn

own

acti

vity

.a

Dos

ede

pend

ent.

Ada

pted

and

mod

ifie

dw

ith

perm

issi

onfr

omD

awki

nsK

,Lie

berm

anJA

,Leb

owit

zB

D,e

tal

.(19

99)

Ant

ipsy

chot

ics:

Pas

tan

dfu

ture

.Nat

iona

lIns

titu

teof

Men

tal

Hea

lth

Div

isio

nof

Serv

ices

and

Inte

rven

tion

Res

earc

hW

orks

hop

(Jul

y14

,199

8).S

chiz

op

hr

Bu

ll25

,395

–404

;Bur

nsM

J(2

001)

The

phar

mac

olog

yan

dto

xico

logy

ofat

ypic

alan

tips

ycho

tic

agen

ts.J

Cli

nT

oxi

col

39,1

–14.


Acute Extrapyramidal Side Effects (Dystonia,Parkinsonism, Akathisia)Antipsychotic-induced EPS occur both acutely and afterchronic treatment. All antipsychotic medications are capableof producing EPS. In general, first-generation antipsychoticsare more likely to cause EPS than second-generation antipsy-chotics when the drugs are used at usual therapeutic doses.Among second-generation drugs, clozapine and quetiapinehave been shown to carry minimal to no risk for EPS withinthe therapeutic dosage range. Risperidone can produce dose-related EPS �≥6 mg/day�. With the exception of akathisia, theincidence of EPS with olanzapine, aripiprazole, and ziprasi-done is not appreciably different from that with placebo. Therelative liability of the individual second-generation agents toproduce EPS will become apparent only when they have beendirectly compared with each other in prospective clinical trials.

Commonly occurring acute EPS include akathisia, dystonia,and parkinsonism, with each having a characteristic time ofonset. This group of acute EPS develops relatively soon afterthe initiation of antipsychotic medications and remits soonafter the drugs are discontinued. These movement disordersare dose-dependent and reversible.

Dystonias tend to be sudden in onset, the most dramaticform of acute EPS, and extremely distressing to patients.They present as sustained muscle contraction with contorting,twisting, or abnormal postures affecting mainly the muscula-ture of the head and neck but sometimes the trunk and lowerextremities. Dystonic reactions usually occur within the firstfew days of therapy. Laryngeal dystonias are the most serious,and are potentially fatal. Risk factors for acute dystoniasinclude a history of prior dystonias, young age, male gender,use of high-potency neuroleptic agents such as haloperidolor fluphenazine, high dose of medication, and parenteraladministration.

Medication-induced parkinsonism is characterized by thesymptoms of idiopathic parkinsonism, including rigidity,tremor, akinesia, and bradykinesia. Risk factors include olderage, higher dose, a history of parkinsonism, and underlyingdamage in the basal ganglia.


Patients with akinesia or bradykinesia suffer from slowmovement, apathy, and with difficulty with spontaneity ofspeech and initiating movement. These symptoms need tobe distinguished from negative symptoms of schizophrenia,depressive symptoms, and catatonia.

Akathisia, the most common EPS of conventional antipsy-chotics agents, is characterized by both the subjective andobjective somatic restlessness. Patients with akathisia mayusually experience an inner tension, discomfort, irritation,anxiety, or irresistible urge to move various parts of theirbodies. Akathisia appears objectively as psychomotor agita-tion, such as continuous pacing, rocking from foot to foot, orthe inability to sit still. Akathisia is typically witnessed in afew hours to days after medication administration. This sideeffect can be seen in up to 20 to 25% of patients treated withconventional agents. Akathisia is frequently cited as a reasonfor poor drug compliance, since it is often extremely distressingto patients. It can also result in dysphoria and aggressive orsuicidal behavior.

The treatment of acute EPS depends on the specificside effect. Dystonia can be quickly and successfully treatedwith an intramuscular injection of an anticholinergic (i.e.,benztropine, diphenhydramine). The initial treatment ofparkinsonian side effects is lowering the dose of antipsy-chotic. If an adequate response is not achieved, adding ananticholinergic, or amantadine (a weak dopamine agonist),may be efficacious. If symptoms persist, switching to ansecond-generation antipsychotic or a low-potency conventionalantipsychotic should be considered. Akathisia is less respon-sive to treatment than are other acute EPS. The first step ofthe treatment of akathisia is lowering the antipsychotic dose.The next step is individual trials of beta-adrenergic blockers(i.e., propranolol), and benzodiazepines (i.e., lorazepam andclonazepam).

Tardive Dyskinesia and Other Tardive SyndromesTD is a repetitive, involuntary, hyperkinetic movementdisorder caused by sustained exposure to antipsychoticmedication. TD is characterized by choreiform movements,


tics and grimaces of the oro-facial muscles, and dyskinesiaof distal limbs, often the paraspinal muscles, and occasion-ally the diaphragm. Younger patients with TD tend to exhibitslower athetoid movements of the trunk, extremities, andneck. In addition to the more frequently observed oro-facialand choreoathetoid signs of TD, tardive dystonias (sustainedabnormal postures or positions) and tardive akathisia (persis-tent subjective and/or objective signs of restlessness) havebeen described. The abnormal movements of TD are usuallyincreased with emotional arousal and are absent when the indi-vidual is asleep. According to the diagnostic criteria proposedby Schooler and Kane, the movements should be present for atleast four weeks, and exposure to antipsychotic drugs shouldhave totaled at least three months. The onset of the abnormalmovements should occur either while the patient is receivingan antipsychotic agent or within a few weeks of discontinuingthe offending agent.

Prevalence surveys indicate that mild forms of TD occur inapproximately 20% of patients who receive chronic treatmentwith conventional antipsychotic medication. A major prospec-tive research demonstrates that the cumulative incidence ofTD is 5% in the first year, 10% the second year, 15% thethird year, and 19% the fourth year in a patient who receivesa typical neuroleptic. Prevalence rates of TD may exceed 50%in high-risk groups, such as the elderly. The reported preva-lence of tardive dystonia is around 1.5 to 4%. Among the mostsignificant predictors of TD are older age, female gender, pres-ence of EPS, diabetes mellitus, affective disorders, and certainparameters of neuroleptic exposure such as dose and durationof therapy.

All first-generation antipsychotic agents are associatedwith a risk of TD. Studies of newer antipsychotics suggestthat TD liability is much lower with the second-generationagents, and clozapine is associated with a substantially lowerrisk for development of TD than other antipsychotic medi-cations. A double-blind, random assignment study of 1,714patients found a 0.52% long-term risk of TD with olanza-pine treatment, as compared to a 7.45% risk with haloperidol.Another published double-blind randomized study showed a


significantly lower risk of TD in olanzapine-treated patients(1%) than haloperidol-treated schizophrenic patients (4.6%).In a double-blind, randomized 1-year study comparing risperi-done to haloperidol, the rate of TD was 0.6% in the risperidonegroup and 2.7% in the haloperidol group. Among a sample ofgeriatric patients, a lower incidence of TD in patients treatedwith risperidone than in those treated with haloperidol, at leastover a 9-month period, has been reported. The rate of TDwith risperidone has been reported to be low (0.6%) for dosescurrently used (2–8 mg/day). The incidence of TD with queti-apine is preliminarily reportedly low or virtually non-existent,although this remains to be demonstrated prospectively. Therisk of TD with ziprasidone and aripiprazole is not known butis expected to be similarly low.

For most patients, TD does not appear to be progressiveor irreversible. The onset of TD often tends to be insidiouswith a fluctuating course. With time, TD will either stabilizeor improve even if the antipsychotic medication is continued,although there are reports of TD worsening during continueddrug therapy. After discontinuation of antipsychotic medica-tion, a significant proportion of patients with TD will haveremission of symptoms, especially if the TD is of recent onsetor the patient is young. Unfortunately, withdrawal of antipsy-chotic agents is seldom an option for patients with seriouspsychosis.

The American Psychiatric Association Task Force on TDhas issued a report in which a number of recommendationswere made for preventing and managing TD. These include(1) establishing objective evidence that antipsychotic medica-tions are effective for an individual; (2) using the lowest effec-tive dose of antipsychotic drugs; (3) prescribing cautiously forchildren, elderly patients, and patients with mood disorders;(4) examining patients on a regular basis for evidence of TD;(5) considering alternatives to antipsychotic drugs, obtaininginformed consent, and also considering a reduction in dosagewhen TD is diagnosed; and (6) considering a number of optionsif the TD worsens, including discontinuing the antipsychoticmedication, switching to a different drug, or considering a trialof clozapine.


Although a large number of agents have been studied fortheir therapeutic effects on TD, there is no definitive drugtreatment for it. Second-generation antipsychotics, in partic-ular clozapine, have been used in clinical practice to treat TD,but there have been no adequately controlled trials to dateto support this practice. It has been suggested that second-generation antipsychotics should be used as first-line treatmentfor patients who have TD or are at risk for TD. Guidelines fortreating TD recommend using second-generation agents formild TD symptoms, and clozapine or a newer agent for moresevere symptoms.

Neuroleptic Malignant SyndromeNeuroleptic malignant syndrome (NMS) is characterized bythe triad of rigidity, hyperthermia, and autonomic instability inassociation with the use of an antipsychotic medication. NMSis often associated with elevation of creatine kinase (greaterthan 300 U/mL), leukocytosis (greater than 15,000 mm3), andchange in level of consciousness. NMS can be of sudden andunpredictable onset, usually occurring early in the course ofantipsychotic treatment, and can be fatal in 5 to 20% ofuntreated cases.

The incidence of NMS varies from 0.02 to 3.23%, reflectingdifferences in criteria. Prevalence rates are unknown, butare estimated to vary from 1 to 2% of patients treatedwith antipsychotic medication. The relative risk of second-generation antipsychotics for NMS is likely to be lower,but conclusive data are not yet available. NMS has beenreported with clozapine, risperidone, olanzapine, and queti-apine. Proposed risk factors include prior episode of NMS,younger age, male gender, physical illness, dehydration, useof high-potency antipsychotics, rapid dose titration, use ofparenteral (IM) preparations, and pre-existing neurologicaldisability.

If NMS is suspected, the offending antipsychotic agentshould be discontinued and supportive and symptomatic treat-ment started. Both dantrolene and dopamine agonists suchas bromocriptine have also been used in the treatment ofNMS. These agents, however, have not shown greater efficacycompared to supportive treatment.


The usual course of treatment is between 5 and 10 days.Long acting depot preparations will prolong recovery time.After several weeks of recovery, treatment may be cautiouslyresumed with a different antipsychotic medication with gradu-ally increased doses.

Endocrine and Sexual EffectsAll standard antipsychotic drugs elevate serum prolactin levelsby blocking the tonic inhibitory actions of dopamine onlactotrophic cells in the pituitary. Among second-generationantipsychotics, risperidone and amisulpride can produce dose-dependent hyperprolactinemia to a greater extent than first-generation antipsychotics, whereas clozapine, olanzapine,ziprasidone, and quetiapine do not cause a sustained eleva-tion of prolactin above normal levels. Aripiprazole, being apartial DA agonist, produces no elevation of prolactin andeven suppresses prolactin levels slightly. Hyperprolactinemiain women can lead to menstrual disturbances, including anovu-latory cycles and infertility, menses with abnormal lutealphases, or frank amenorrhea and hypoestrogenemia. Womenhave also reported decreased libido and anorgasmia, andthere are reports of increased long-term risk of osteoporosisalthough this is controversial. Antipsychotic-induced gyneco-mastia has been reported in 3% of women and 6% of men.Galactorrhea occurs in 2.7% of men and 10 to 50% of women.The major effects of hyperprolactinemia in men are loss oflibido, impotence, hypospermatogenesis, and erectile or ejac-ulatory disturbances. Although amisulpride and risperidonecause significant elevations of prolactin levels, a number ofstudies have found only a small incidence of sexual dysfunc-tions in patients treated with these drugs. This may be due tothe fact that these reports have relied on spontaneous reportingof sexual side effects. In a drug monitoring study, in which sideeffects profiles of haloperidol and clozapine were investigated,a significantly higher frequency of sexual disturbances havebeen found. Using a side effect rating scale, it was shown thatthe prevalence of these adverse events was high during thefirst weeks of the study. However, side effects usually remittedspontaneously despite continuous treatment in the majority ofpatients.


Metabolic EffectsVarious degrees of weight gain have been recognized as acommon problem with conventional antipsychotic medica-tions. Weight gain is an important issue in the managementof patients, because this adverse effect may be associatedwith non-compliance and certain medical illnesses, such asdiabetes mellitus, cardiovascular disease, certain cancers, andosteoarthritis.

Differences have been discovered among second-generation antipsychotics with respect to their ability to induceweight gain (Table 1-4). A recent meta-analysis, which esti-mates the weight change after 10 weeks of treatment at astandard dose, demonstrated that mean increases were 4.45 kgfor clozapine, 4.15 kg for olanzapine, 2.10 kg for risperidone,and 0.04 kg for ziprasidone. The long-term risk of weight gainwith quetiapine appears to be less than that with olanzapineand clozapine. Short-term weight gain (2.16 kg over 10 weeks)with quetiapine appears comparable to risperidone. Ziprasi-done has been associated with minimal weight gain, whichcould distinguish it among other second-generation antipsy-chotics. Similarly, aripiprazole appears to cause little or noweight gain. During long-term treatment, clozapine and olan-zapine have the largest effects on weight gain; risperidoneproduces intermediate weight gain; quetiapine and ziprasidoneproduce the least weight gain. Weight gain does not appear tobe dose-dependent, tends to plateau between 6 and 12 monthsafter initiation of treatment, and is mainly due to an increase inbody fat. The mechanism by which weight gain occurs duringtreatment with antipsychotics is poorly understood, but thebroader receptor affinities of the agents and their antagonismof histamine H1 and serotonin 5-HT2C receptors have beenimplicated. There is currently no standard approach to themanagement of weight gain induced by antipsychotic medi-cation. Patient education prior to initiating treatment shouldbe provided, and regular exercise should be encouraged in allpatients receiving antipsychotic medication. Switching to othersecond-generation antipsychotics with fewer propensities forproducing weight gain may be the most efficient way to dealwith antipsychotic-induced weight gain.


Abnormalities in peripheral glucose regulation anddiabetes mellitus (DM) occur more commonly in schizophrenicpatients compared with the general population. There isgrowing concern with metabolic disturbances associated withantipsychotic use, including hyperglycemia, hyperlipidemia,exacerbation of existing type 1 and 2 DM, new-onset type 2DM, and diabetic ketoacidosis. A number of case reports haveimplicated both clozapine and olanzapine in the emergence ofnon-insulin-dependent (type 2) DM and diabetic ketoacidosis.There are fewer reports describing an association between DMand quetiapine or risperidone, but these drugs do appear tohave this side effect potential albeit to a lesser degree. Incontrast there are many fewer reports for ziprasidone and arip-iprazole which suggests they may have less or no metabolic sideeffect liability. The limited reporting for ziprasidone may berelated to the relatively limited use of the agent at the presenttime. Although no clear mechanism of action of the second-generation agents has been established, significant weightgain or antagonism of specific serotonin receptor subtypesmay contribute to the development of these abnormali-ties. Physicians employing second-generation agents shouldroutinely monitor weight, fasting blood glucose, and lipidprofiles.

Cardiovascular EffectsOrthostatic hypotension is usually seen with low-potencyconventional antipsychotic agents (e.g., chlorpromazine orthioridazine) and clozapine through alpha-l-adrenergic antag-onism. Among the first-line second-generation antipsychotics,quetiapine has the greatest potential for inducing orthostasisalthough all agents have this potential. Orthostasis is mostlikely to occur during the first few days after initiation oftreatment, or when increasing the dose of medications; mostpatients develop tolerance to it in the following four to sixweeks. Elderly patients are particularly vulnerable to this sideeffect and it may predispose them to falls and increase theincidence of serious injuries or fractures. A gradual upwardtitration of dosage may help to reduce the risk of hypotension,and patients should be advised to change posture slowly.


Tachycardia may occur as a result of the anticholinergiceffects of antipsychotic medications on vagal inhibition, orsecondary to orthostatic hypotension. Clozapine produces themost pronounced tachycardia; approximately 25% of patientswill have a sinus tachycardia with an increase of about 10 to 15beats per minute. Although quetiapine has virtually no cholin-ergic activity, tachycardia is a possible side effect, perhapssecondary to its adrenergic effects on blood pressure. Mostpatients will develop tolerance to this side effect over time.If tachycardia is sustained or becomes symptomatic, an elec-trocardiogram (ECG) should be obtained. Low doses of aperipherally acting beta-blocker such as atenolol can be usefulto treat medication-induced tachycardia without hypotension.

ECG changes are observed with many antipsychotic agents.Chlorpromazine may cause prolongation of the QT and PRintervals, ST depression, and T-wave flattening or inversion,and thioridazine may cause QT and T-wave changes. Theseeffects rarely cause clinically relevant symptoms within thera-peutic dose ranges.

Ziprasidone has very specific recommendations in thepackage insert as to the types of patients it should not beused in. Needless to say, antipsychotics that lead to QTc-prolongation must not be combined with other drugs thathave similar effects. The effect of many but not all antipsy-chotic drugs on the QT interval appears to be dose-related.Several antipsychotic drugs have infrequently been associatedwith malignant arrhythmias such as torsade de pointes. To date,torsade de pointes has not been reported following therapeuticdoses or overdose with ziprasidone or other second-generationantipsychotics.

Sudden unexplained deaths have been rarely reported withtherapeutic doses of antipsychotic drugs, and such deaths couldresult from cardiac arrhythmias in the absence of anotherexplanation. There is, however, currently no evidence thatantipsychotic drugs are associated with an increased prevalenceof sudden deaths due to cardiac events, although a numberof case reports and case series concerning death followingcardiomyopathy, potentially induced by clozapine are a matterof concern.


Gastrointestinal EffectsThe anticholinergic effects of antipsychotic medications caninduce dry mouth and constipation as well as tachycardia,urinary retention, and blurring of vision. These adverse effectsare relatively commonly encountered with low-potency first-generation antipsychotics and may be dose related. In cases ofmore serious gastrointestinal adverse events, such as paralyticileus, which has been reported following treatment with cloza-pine, medication must be discontinued immediately and rele-vant medical or surgical interventions may become necessary.Anticholinergic manifestations are common with poisoningfrom clozapine and olanzapine.

Hepatic EffectsAsymptomatic mild, transient, and reversible elevations ofliver enzyme levels occur infrequently with both first-and second-generation antipsychotic drugs. These abnor-malities usually occur during the first three months oftreatment, are idiosyncratic, and seldom a serious concern.Rarely, symptomatic hepatotoxicity (cholestatic or hepatitic)may be associated with second-generation antipsychotics; inthese cases, the offending medication should be discon-tinued. Recovery occurs in up to 75% of patients withintwo months; 90% recover within one year. Patientstaking antipsychotics who have nausea, fever, abdom-inal pain, and rash should have their liver functionevaluated to exclude hepatotoxicity. Since antipsychotic-induced jaundice is infrequent, other etiologies should beruled out before the cause is judged to be antipsychotictreatment.

Hematological EffectsAntipsychotic medications may cause blood dyscrasias,including neutropenia, leukopenia, leukocytosis, throm-bopenia, and agranulocytosis. Leukopenia, usually transient,commonly occurs early in treatment, and resolves sponta-neously. Chlorpromazine has been associated with benignleukopenia, which occurs in up to 10% of patients. This


phenomenon is even more common following clozapineadministration.

Agranulocytosis (granulocyte count less than 500/mm3)is a fatal side effect of antipsychotic drugs. The risk ofagranulocytosis with clozapine is 1% and is greatest early intreatment, usually within the first 8 to 12 weeks of treatment.It tends to occur slightly more often in women, the elderly,and young patients (less than 21 years old). Agranulocy-tosis from clozapine is usually reversible if the drug is with-drawn immediately. Olanzapine is not associated with severeagranulocytosis. Despite these encouraging studies, there area number of case studies reporting agranulocytosis duringtreatment with olanzapine (and quetiapine) in patients whohad suffered this adverse event during previous clozapineexposure.

Before initiating treatment with clozapine, patients in theUSA must be registered in a program that ensures that theyreceive weekly monitoring of their white blood cell (WBC)count during the first six months of treatment. Current guide-lines require weekly monitoring for one month after the termi-nation of clozapine treatment. Guidelines on the use of cloza-pine vary between different countries.

Other Side EffectsSedation is the single most common side effect among low-potency conventional antipsychotics, as well as clozapine,zotepine, and quetiapine. Although sedation is often benefi-cial at the beginning of treatment to calm down an anxious oraggressive patient, it usually impairs functioning during long-term treatment. Most patients usually develop tolerance overtime, or it may be possible to minimize sedation by dose reduc-tion or by shifting most of the medication to night to reducedaytime sleepiness.

Antipsychotic medications can lower the seizure thresholdto some degree. Seizure is more common with low-potencyfirst-generation antipsychotics and clozapine. Clozapine isassociated with dose-related increase in seizures. For example,doses of clozapine below 300 mg/day have been found to havea seizure rate of about 1%, doses between 300 and 600 mg/day


have a seizure rate of 2.7%, and doses above 600 mg/day havea rate of 4.4%. Strategies to reduce the risk for seizures includeslower dose titration, a lower dose, and the addition of ananticonvulsant agent (i.e., valproic acid).

DRUG INTERACTIONS AND ANTIPSYCHOTICAGENTSMost antipsychotics are metabolized by hepatic microsomaloxidases (cytochrome P450 system); the major isoenzymesystems involved are CYP1A2, CYP2C19, CYP2D6, andCYP3A4. Induction or inhibition of these enzymes by otherdrugs may occasionally produce clinically important druginteractions. Table 1-5 summarizes clinically significant phar-macokinetic drug interactions involving second-generationantipsychotic drugs. SSRI’s, particularly fluoxetine and parox-etine, can increase plasma concentrations of antipsychoticmedications by inhibiting CYP2D6 and decreasing the clear-ance of antipsychotics, possibly leading to toxicity. Conven-tional antipsychotic drug clearance can be decreased by 50%with concurrent administration of certain heterocyclic antide-pressants, beta-blockers, some antibiotic/antifungal agents,and cimetidine. Clozapine toxicity has occurred followingco-administration with the CYP1A2 inhibitors cimetidine,erythromycin, and fluvoxamine.

Quetiapine, ziprasidone, and aripiprazole toxicity can becaused by inhibitors of CYP3A4 such as erythromycin, fluox-etine, nefazadone, and protease inhibitors.

In contrast, drugs such as carbamazepine, phenobarbital,and phenytoin can reduce plasma concentrations of antipsy-chotic drugs by increasing the metabolism of the antipsychoticagent. For example, carbamazepine, commonly combined withantipsychotic medications, can reduce the plasma concentra-tion of haloperidol by 50%. Anticonvulsants, however, maynot have a significant effect on the metabolic clearance of olan-zapine or risperidone as they are not substantially metabolizedthrough CYP3A4. Cigarette smoking increases drug clearancefor many antipsychotic drugs, including clozapine and olan-zapine. The clearance rate of clozapine and olanzapine areincreased by 20 to 50%.


� TABLE 1-5. Pharmacokinetic Drug Interactions Involving Second-Generation Antipsychotic Agents

DRUG AND CYTOCHROMEP-450 ISOENZYME(S) INHIBITORS INDUCERS

Clozapine1A2 Fluoroquinolones,

fluvoxamineSmoking, PAHsa

3A4 Erythromycin,ketoconazole, ritonavir,sertraline,b cimetidine

Rifampin,carbamazepine,phenytoin,barbiturates

2D6 Ritonavir, quinidine,risperidone,b

fluoxetine,b sertralineb

None

Risperidone2D6 Paroxetine, fluoxetine Rifampin,

carbamazepine,phenytoin,barbiturates

Olanzapine1A2 Fluvoxamine Smoking, PAHs,

carbamazepine2D6 None Phenytoin

Quetiapine3A4 Ketoconazole,

erythromycinRifampin,


Ziprasidone3A4 Ketoconazole,



None None

Aripiprazole3A4 Ketoconazole,



2D6 Paroxetine, fluoxetine None

aPAHs, polycyclic aromatic hydrocarbons.bCase reports of mild to moderate elevations in serum concentration.

Adapted from Worrel JA, Marken PA, Beckman SE, et al. (2000) Atypical antipsy-chotic agents: A critical review. Am J Health Syst Pharm 57, 238–255.


Other common interactions are:

• Antacids: can decrease the absorption of the antipsychoticagent from the gut.

• Antipsychotics: can antagonize the effects of dopamineagonists or levodopa when these drugs are used to treatparkinsonism

• Antipsychotic agents: may also enhance the effects of CNSdepressants such as analgesics, anxiolytics, and hypnotics.

• Doses of pre-anesthetic medication or general anestheticsmay need to be reduced.

Antipsychotic Medications and PregnancyMost antipsychotic agents readily cross the placenta and aresecreted in breast milk to some degree. There is little data todemonstrate whether prenatal exposure to antipsychotic agentsis linked to spontaneous abortion, congenital malformations,carcinogenesis, intrauterine growth retardation, or behavioralteratogenicity.

Fetal exposure over the course of pregnancy may affectdevelopment of the dopamine system; the benefits of control-ling psychotic symptoms during pregnancy versus the possiblerisks to the mother and the fetus of withdrawing treatmentand the risks to the fetus of continuing treatment must beconsidered.

There is no evidence that second-generation antipsychoticsare teratogenic in humans. However, if possible, use of antipsy-chotic medication should be avoided during the first trimester,especially between weeks six and ten, unless the patient’spsychosis places the mother and/or her fetus at significant risk.Antipsychotic medications may be relatively safe during thesecond and third trimesters of pregnancy.

If a first-generation antipsychotic is used, high-potencyagents appear to be preferable for first-line management, dueto a lower propensity to cause orthostasis. Low doses shouldbe given, administration of antipsychotic medication should beas brief as possible, and the medication should be discontinuedfive to ten days before delivery to minimize the chances ofthe newborn experiencing EPS. This notion, however, hasbeen re-evaluated, since discontinuation of medication before


delivery may put the mother at risk for decompensation. Anti-cholinergic agents should also be avoided during pregnancy,especially for the first trimester. Since antipsychotics are alsosecreted into breast milk, the infants should not be breast-fedif the mother resumes antipsychotic medication postpartum.

ADDITIONAL READING

1. Lieberman JA, Stroup TS, McEvoy JP, Swartz MS, RosenheckRA, Perkins DO, Keefe RS, Davis SM, Davis CE, Lebowitz BD,Severe J, Hsiao JK: Effectiveness of antipsychotic drugs in patientswith chronic schizophrenia. New England Journal of Medicine 2005;353(12):1209–1223

2. Miyamoto S, Duncan GE, Marx CE, Lieberman JA: Treatmentsfor schizophrenia: a critical review of pharmacology and mecha-nisms of action of antipsychotic drugs. Molecular Psychiatry 2005;10(1):79–104

3. Lehman AF, Lieberman JA, Dixon LB, McGlashan TH, MillerAL, Perkins DO, Kreyenbuhl J; American Psychiatric Associa-tion; Steering Committee on Practice Guidelines: Practice guidelinefor the treatment of patients with schizophrenia, second edition.American Journal of Psychiatry 2004; 161(2 Suppl):1–56

4. Miyamoto S, Aoba A, Duncan GE, Lieberman JA: Acute Pharma-cological Treatment of Schizophrenia. In: Schizophrenia, SecondEdition. Blackwell Publishing, Oxford, 2002

5. Miyamoto S, Duncan GE, Goff DC, Lieberman JA: Therapeutics ofSchizophrenia. In: Neuropsychopharmacology The Fifth Generationof Progress. Lippincott Williams & Wilkins, Philadelphia, pp. 775–807, 2002

6. Schooler NR, Kane JM: Research diagnosis for tardive dyskinesia.Arch. Gen. Psychiat. 1982; 39:486–487.

2ANTIDEPRESSANTS

INTRODUCTIONThe use of antidepressants in the treatment of depressionremains the best-understood use of these medications; however,there is a growing list of other indications for antidepressants,including panic disorder (PD), obsessive–compulsive disorder(OCD), bulimia and posttraumatic stress disorder (PTSD).Many of these illnesses respond best to combination treat-ment modalities that include medication and various forms ofpsychotherapy.

PHARMACOLOGYThe first antidepressant was the monoamine oxidase inhibitor(MAOI), iproniazid, initially licensed as an antituberculosisdrug. This was soon followed by the tricyclic antidepressant(TCA), imipramine. Although both were developed in theearly 1950s, they represented different paths of discovery:iproniazid was the result of clinical and laboratory collabora-tion, whereas imipramine’s introduction was largely based onclinical observation.

Within the past 25 years, similarly efficacious tricyclic andheterocyclic antidepressants have been developed but withvarying side-effect profiles. A new direction in antidepressantpharmacology began in 1987 with the discovery and



marketing of fluoxetine, the first antidepressant selective forserotonin reuptake blockade. Beyond selective serotonin reup-take inhibitors (SSRIs), the latest offerings in antidepressantpharmacology include agents that act at multiple neurotrans-mitter levels.

In evaluations of the many antidepressants available, thefocus has generally been on the agent’s side-effect profileand ease of administration. Despite many efforts in this area,there is no conclusive evidence to demonstrate the clinicalsuperiority of any one group of agents.

Mechanisms of ActionAll known antidepressants affect monoamine neurotransmis-sion.

MAOIs inhibit the activity of monoamine oxidase, resultingin a decrease in the breakdown of dopamine, serotonin, andnorepinephrine in the synapse, thus increasing the amountof these neurotransmitters available for release and synaptictransmission. MAOIs are highly effective antidepressants andanxiolytics but are rarely used due to dietary restrictionsthat must be adhered to in order to avoid tyramine-inducedhypertensive crises.

TCAs block presynaptic reuptake of norepinephrine andserotonin to various degrees. Several are related by metabolism,thus the tertiary amines amitriptyline and imipramine aremetabolized to the secondary amines nortriptyline anddesipramine respectively. All act through reuptake inhibitionand are generally selective for the norepinephrine transporter;several, however, have equal or greater affinity for the serotonintransporter. The TCAs were the drugs of choice for depressionthrough the 1980s. Though effective, their nonselective actionson cholinergic, histaminergic, and presynaptic adrenergic recep-tors resulted in a number of side effects.

SSRIs were first introduced in the late 1980s, and rapidlyeclipsed the TCAs as the drugs of choice for depression. Thereare subtle differences between compounds, mainly in terms oftheir half-life, their potency for reuptake inhibition, and theiraffinity for some other receptors. As SSRIs more selectivelyaffect reuptake, with few effects on the adrenergic, histamin-ergic, and cholinergic systems, side effects have been reduced.

ANTIDEPRESSANTS 43

The selective norepinephrine reuptake inhibitors (NRIs)such as reboxetine (not available in the US) or atomoxetineshare a similar mechanism with the SSRIs, but act on thenorepinephrine transporter and have little affinity for theserotonin transporter.

Several other second-generation agents, sometimesreferred to as “atypical antidepressants”, were introducedduring the same period as the SSRIs. These include bupro-pion, which seems to exert primarily a dopaminergic effect,and trazodone, which is structurally related to the TCAs buthas a primary serotonergic mechanism.

Of the serotonin receptors, 5-hydroxytryptamine type1A �5-HT1A� appears most related to the therapeutic effects ofantidepressants, and this receptor influences norepinephrine,dopamine, acetylcholine, neuropeptides, other serotoninreceptors, and probably beta-receptor downregulation.

The so-called third-generation antidepressants includeserotonin–norepinephrine reuptake inhibitors (SNRIs –venlafaxine, milnacipran and duloxetine), mixed serotoninantagonist/reuptake inhibitors (nefazodone and trazodone)and the mixed serotonin/noradrenaline antagonist mirtaza-pine. SNRIs have equal affinity for the norepinephrine andserotonin transporter. Though, like several of the TCAs,venlafaxine has multiple receptor effects, it is relatively freeof the anticholinergic and antihistaminic side effects that arecommon with the TCAs.

Mixed serotonin antagonist/reuptake inhibitors such asnefazodone have multiple mechanisms of action, with bothserotonin (as well as norepinephrine) transporter inhibi-tion as well as antagonism of 5-HT2A and alpha-1-receptors.Trazodone is similar; however, its effects are somewhat lessspecific.

The mixed serotonin/noradrenaline antagonist mirtazapineis unique in that it appears to work primarily through specificreceptor blockade of the alpha-2-autoreceptors on presynapticnoradrenergic neurons, enhancing noradrenergic output. Thisclass may exert a similar effect toward autoreceptors onserotonin neurons. Antagonism of 5-HT2 and 5-HT3 recep-tors may also concentrate the effect of serotonin on 5-HT1A


receptors. The antidepressants available in the US, their class,and relative costs are listed in Table 2-1.

PharmacokineticsThe pharmacokinetics of TCAs are complex, as reflected inthe diversity of half-lives (10 to 40 hours). TCAs are primarilyabsorbed in the small intestine, reaching peak plasma levelstwo to six hours after oral administration. Absorption can beaffected by changes in gut motility. The drugs are extensivelymetabolized in the liver on first pass through the portal system.They are lipophilic, have a large volume of distribution, andare highly protein-bound (85–95%). TCAs are metabolized toactive metabolites in the liver by hepatic microsomal enzymesand excreted by the kidneys. The rate of metabolism can varygenetically; 7 to 9% of the white population are slow hydrox-ylators. There is a large range of elimination half-lives.

TCAs as a class have a relatively narrow therapeutic index;there is a significant risk of toxicity with blood levels ofonly two to six times the therapeutic level. A 1-week supplycan be fatal in overdose, as blood concentrations of greaterthan 1000 ng/dl are correlated with prolongation of the QRScomplex and arrhythmias. TCAs are commonly ingested agentsby which patients successfully commit suicide by overdose.

MAOIs are also well absorbed from the gastrointestinaltract. Their short half-life of one to two hours is not particularlyrelevant as they bind irreversibly with MAO. Thus, the activityof these drugs depends less on pharmacokinetics and more onthe synthesis of new MAO to restore normal enzyme activity.This synthesis requires approximately two weeks. This class ofdrugs is little used due to their potentially dangerous interac-tions with sympathomimetics and foods containing tyramine.

Each of the six SSRI agents (fluoxetine, paroxetine,sertraline, fluvoxamine, citalopram and its S-enantiomerescitalopram) selectively block the reuptake of serotoninpresynaptically, though each drug differs structurally from theothers. As a result, these agents differ in their pharmacokineticprofiles. Many SSRIs are, like the TCAs, highly protein boundthough the proportions of citalopram and escitalopram thatare protein-bound are 80 and 56% respectively. In contrast,

�TA

BLE

2-1.

Ant

idep

ress

ants

avai

labl

ein

the

Unite

dSt

ates

CLA

SSN

AM

EG

ENER

ICH

OW

SUPP

LIED

RECO

MM

ENDE

DDO

SEPE

RDA

Y∗

PRIC

EIN

DEX:

(DO

LLA

RCO

STO

FA

VERA

GE

DOSE

/d)∗∗

TC

As

Am

itri

ptyl

ine

YT

able

ts(m

g):1

0,25

,50,

75,1

00,1

5015

0m

g0�

26O

ral

solu

tion

:10

mg/

5mL

IM:1

0m

g/m

LIm

ipra

min

eY

Tab

lets

(mg)

:10,

25,5

0C

apsu

les

(mg)

:75,

100,

125,

150

150

mg

3�65

Nor

trip

tylin

eY

Ora

lso

luti

on:1

0m

g/5

mL

Cap

sule

s(m

g):1

0,25

,50,

7575

mg

2�66

Des

ipra

min

eY

Tab

lets

(mg)

:10,

25,5

0,75

,100

,150

150

mg

2�18

Am

oxap

ine

YT

able

ts(m

g):2

5,50

,100

,150

300

mg

5�00

Dox

epin

YO

ral

solu

tion

:10

mg/

mL

Cap

sule

s(m

g):1

0,25

,50,

75,1

00,1

5015

0m

g1�

66

Pro

trip

tylin

eN

Tab

lets

(mg)

:5,1

045

mg

7�99

Tri

mip

ram

ine

NT

able

ts(m

g):2

5,50

,100

150

mg

5�00

Clo

mip

ram

ine

YT

able

ts(m

g):2

5,50

,75

150

mg

3�03

Map

roti

line

YT

able

ts(m

g):2

5,50

,75

150

mg

1�86

MA

OIs

Isoc

arbo

xazi

dN

Tab

lets

:10

mg

30m

g2�

45P

hene

lzin

eN

Tab

lets

:15

mg

60m

g2�

64T

rany

lcyp

rom

ine

NT

able

ts:1

0m

g40

mg

3�94

�TA

BLE

2-1.

(Con

tinue

d)

CLA

SSN

AM

EG

ENER

ICH

OW

SUPP

LIED

RECO

MM

ENDE

DDO

SEPE

RDA

Y∗

PRIC

EIN

DEX:

(DO

LLA

RCO

STO

FA

VERA

GE

DOSE

/d)∗∗

SSR

IsF

luox

etin

eY

Ora

lso

luti

on:2

0m

g/5m

LT

able

ts(m

g):1

0,20

,40

Del

ayed

rele

ase

caps

ules

:90

mg

(“P

roza

cW

eekl

y”)

60m

g1�

10

Sert

ralin

eN

Ora

lso

luti

on:2

0m

g/m

LT

able

ts(m

g):2

5,50

,100

100

mg

3�02

Par

oxet

ine

YO

ral

solu

tion

:10

mg/

5mL

Tab

lets

(mg)

:10,

20,3

0,40

40m

g2�

97

Par

oxet

ine

cont

rolle

dre

leas

eC

ontr

olle

dre

leas

eta

blet

s(m

g):1

2.5,

25,3

7.5

40m

g3�

39F

luvo

xam

ine

YT

able

ts(m

g):2

5,50

,100

200

mg

5�25

Cit

alop

ram

YO

ral

solu

tion

:10

mg/

5m

LT

able

ts(m

g):1

0,20

,40

40m

g2�

65

Esc

ital

opra

mN

Ora

lso

luti

on:1

mg/

1m

LT

able

ts(m

g):5

,10,

2020

mg

2�65

Aty

pica

lB

upro

pion

YT

able

ts(m

g):7

5,10

030

0m

g2�

88B

upro

pion

sust

aine

dre

leas

eY

Sust

aine

dre

leas

e(m

g):1

00,1

5030

0m

g3�

87B

upro

pion

exte

nded

rele

ase

NE

xten

ded

rele

ase

(mg)

:150

,300

300

mg

4�84

Tra

zodo

neY

Tab

lets

(mg)

:50,

100,

150,

300

100

mg

0�22

SNR

IV

enla

faxi

neN

Tab

lets

(mg)

:25,

37.5

,50,

75,1

0022

5m

g4�

30E

xten

ded

rele

ase

(mg)

:37.

5,75

,150

225

mg

3�77

Dul

oxet

ine

NT

able

ts(m

g):2

0,30

,60

60m

g3�

64

Mix

edse

roto

nin

anta

goni

st/

reup

take

inhi

bito

rs

Nef

azod

one

YT

able

ts(m

g):5

0,10

0,15

0,20

0,25

030

0m

g3�

14

Mix

edse

roto

nin/

nora

dren

alin

ean

tago

nist

s

Mir

taza

pine

YT

able

ts(m

g):1

5,30

,45

Dis

solv

able

tabl

ets

(mg)

:15,

30,4

545

mg

0�81

∗ Rec

omm

ende

ddo

sede

rive

dfr

omL

exi-

com

pon

line.

∗ The

pric

ein

dex

isa

roug

hca

lcul

atio

n,as

sum

ing

anav

erag

ere

com

men

ded

dose

for

the

trea

tmen

tof

maj

orde

pres

sion

,and

assu

min

gth

atth

em

ost

effi

cien

tdo

sing

regi

men

,an

dle

ast

expe

nsiv

eta

blet

choi

ce,

ispr

escr

ibed

.W

hen

avai

labl

e,ge

neri

cpr

ices

are

used

.P

rice

sw

ere

deri

ved

from

2006

Car

dina

lH

ealt

hw

hole

sale

pric

egu

ide

prov

ided

byth

eP

harm

acy

Dep

artm

ent

ofth

eN

ewY

ork

Stat

eIn

stit

ute.


however, they have varying half-lives ranging from approxi-mately 24 hours to several days.

All SSRIs are well absorbed and not generally affected byfood administration, though sertraline is an exception to thisrule – its blood level may be increased by food. All have largevolumes of distribution and are extensively protein-bound.They are metabolized by hepatic microsomal enzymes and arepotent inhibitors of these enzymes. The only serotonin reup-take inhibitor with an active metabolite is fluoxetine, whosemetabolite norfluoxetine has a half-life of 7 to 15 days. Thus,it may take several months to achieve steady state with thisagent. This is considerably longer than citalopram (half-life 1.5days) or sertraline and paroxetine (half-lives 24 hours).

There is no correlation between half-life and time to onset.Drugs with shorter half-lives have an advantage in cases whererapid elimination is desired (e.g., in the case of an allergicreaction). Drugs with a longer half-life may also have advan-tages: fluoxetine, for example, has been successfully given ina once-weekly dosing during the continuation phase of treat-ment and a once-weekly formulation of this drug is currentlyavailable. All serotonin reuptake inhibitors are eliminated inthe urine as inactive metabolites. Both fluoxetine and paroxe-tine are capable of inhibiting their own clearance at clinicallyrelevant doses. As such, they have nonlinear pharmacokinetics:changes in dose can produce proportionately large plasmalevels.

Citalopram and its S-enantiomer escitalopram are the mostrecent SSRIs to be introduced in the Untied States; The Foodand Drug Administration (FDA) approved citalopram in 1998and escitalopram in 2002. Of the available SSRIs, these are themost selective for serotonin receptor blockade; escitalopram is100 times more potent than the R-enantiomer.

As with most other antidepressants, bupropion under-goes extensive first pass metabolism in the liver. The parentcompound has a half-life of 10 to 12 hours, but has threeactive metabolites. One, threohydrobupropion, has a half-life of35 hours and is relatively free in plasma (it is only 50% protein-bound). There is considerable individual variability in the levelsof bupropion and its metabolites. Trazodone has a relativelyshort half-life of three to nine hours; as a result of this and its

ANTIDEPRESSANTS 49

apparent lack of active metabolites plasma levels of trazodonecan be quite variable, requiring divided dosing.

Venlafaxine has a short half-life (4 hours); however, it isavailable in an extended release formulation that allows once-daily dosing. It appears to have a dual effect, in which atlower doses it primarily acts on the serotonin transporter, andclinically significant norepinephrine reuptake inhibition is notseen until higher doses are used (150 mg/day and above).

Nefazodone has relatively low bioavailability, and a shorthalf-life (2–8 hours), and thus it is usually given in twice-dailydoses. Mirtazapine has a half-life of 13 to 34 hours.

Duloxetine is the second selective inhibitor of nore-pinephrine and serotonin to be introduced in the United States.Compared with venlafaxine it has relatively greater effect on5-HT reuptake in vitro but the clinical significance of thisdifference is unclear. The half-life of duloxetine is 8–17 hours(mean 12 hours).

The SSRIs are also, to varying degrees, potent inhibitors ofthe P450 hepatic enzyme system. The result is increased bloodlevels of concomitant agents that are also metabolized by thisenzymatic system. Switching from an SSRI to another antide-pressant group illustrates the clinical relevance. For example,when a patient changes from fluoxetine to venlafaxine, theinhibition of the P450 2D6 isozyme will reduce venlafaxinemetabolism for several weeks. The resulting increase in thevenlafaxine blood level can hasten the development of sideeffects. The aware clinician can avoid this by starting with alower than usual initial dose and titrating upward slowly.

INDICATIONS FOR USE OF ANTIDEPRESSANTSAll antidepressants are indicated for the treatment of acutemajor depressive episodes; there is also evidence for their usein the prevention and relapse and recurrence. In addition, anumber of more minor forms of depression may also respondto antidepressant medication, including dysthymic disorder,minor depression, and recurrent brief depression.

All antidepressants appear to treat more than depressivedisorders. Particularly consistent have been data showing theirutility for anxiety disorders; they have begun to supplant the


sedative/hypnotics for these conditions. Many other psychiatricand medical disorders have all been successfully treated withthese agents.

Panic Disorder (PD)SSRIs are considered the first-line treatment for anxiety disor-ders, with good evidence of efficacy in PD. Patients are begunat low doses (e.g., 5 mg of fluoxetine), and increased slowlyto an effective dose to minimize potential side effects. Thereis also strong evidence for the use of TCAs and moderateevidences for the use of MAOIs.

Obsessive–Compulsive Disorder (OCD)SSRIs show strong evidence of effectiveness in the treat-ment of OCD and are considered the first-line treatmentoption in this disorder; all agents appear equally effective. Therecommended starting dose is the same as that for depres-sion, although higher doses (60–80 mg of fluoxetine) may berequired for adequate response. There is evidence for slowbut continued improvement in symptoms for many monthsafter initiation of treatment; medication should therefore betrialed for up to four months if the patient shows a partialresponse. There is also strong evidence for the use of the TCAclomipramide in this disorder. Augmentation with lithium maybe beneficial in partial responders.

Generalized Anxiety Disorder (GAD)Several agents have shown efficacy in this disorder. Goodevidence exists for the use of SSRIs, venlafaxine, nefazodoneand mirtazapine. Both TCAs and MAOIs have also shownmoderate effectiveness; MAOIs also show efficacy in otheranxiety disorders.

Social Phobias and Posttraumatic Stress Disorder(PTSD)SSRIs and venlafaxine have shown efficacy in social phobia(or social anxiety disorder); this condition may also be at least

ANTIDEPRESSANTS 51

partially responsive to TCA medications. SSRIs have showneffectiveness in PTSD.

BulimiaSSRIs are commonly used in the treatment of bulimia nervosa.As with OCD, there appears to be a dose–response effect, withlarger doses often required.

There is strong evidence for the use of TCAs, either aloneor in combination with cognitive therapy, to decrease thebinging and purging of bulimic patients. MAOIs have alsobeen reported successful in the treatment of bulimia, althoughthe many dietary restrictions have made physicians reluctantto prescribe these agents.

Anorexia NervosaThe American Psychiatric Association Practice Guidelinesfor Eating Disorders states that medications should not beused routinely for anorexia nervosa. They should be consid-ered after weight gain and for persistent depression. A fewpublished controlled studies show unimpressive results, andbupropion is contraindicated because of elevated seizure riskin patients with eating disorders.

Body Dysmorphic Disorder (BDD)SSRIs have also shown efficacy in the treatment of BDD.It may be at least partially responsive to TCA medications,particularly those selective for serotonin. Moderate evidenceexists for the use of MAOIs.

Premenstrual Dysphoria (PMDD)PMDD is a chronic cyclical disorder in which serotoninergicfunction is reduced during the luteal phase. The treatment ofchoice is a SSRI; either sertraline or modified-release parox-etine, taken daily or only during the luteal phase. Treatmentwith sertraline should be initiated at a dose of 50 mg/dayand if necessary increased in increments of 50 mg/day at eachmenstrual cycle to a maximum of 150 mg/day if taken every day


or 100 mg/day if taken during the luteal phase only. The initialdose of modified-release paroxetine is 12.5 mg/day increasingto 25 mg/day after one week if necessary.

Intermittent dosing is usually as effective as continuousadministration. Beneficial effects are seen within one to twodays and response rates for improvements in mood and phys-ical symptoms are about 50 to 60%. Evidence of long-termefficacy is lacking. Specialists recommend continuing treatmentuntil the menopause as there is evidence that stopping treat-ment precipitates recurrence. Nevertheless, a trial period offtreatment may be worthwhile after two years if supported bythe patient.

Childhood DisordersTCAs, especially imipramine, are indicated for the treatmentof enuresis. Anxiety and phobias in children (such as sepa-ration anxiety or school phobia and ADHD) are responsiveto TCAs.

Uses for SSRIs in children include repetitive-type abnor-malities associated with autism and mental retardation, ADHD(as an adjunct to methylphenidate), and chronic enuresis.Bupropion has been used successfully in the treatment ofADHD in both children and adults. It may, however, exac-erbate tics in attention-deficit patients with concomitantTourette’s syndrome.

Other Psychiatric DisordersOther indications for SSRIs may include depersonalizationdisorder, obsessive jealousy, pathological gambling, Tourette’ssyndrome, hypochondriasis, and both paraphiliac and nonpara-philiac sexual disorders.

SSRIs have shown effectiveness in a number of morecomplex behavioral disorders such as obesity (particularlyfluoxetine in higher doses), binge eating (sertraline), substanceabuse, and to alleviate certain aggressive behaviors such asimpulsivity and uncontrolled anger in adults, children, and thedemented elderly.

Trazodone is frequently used as a sedative in the elderly; asit can cause or worsen orthostatic hypotension, blood pressure

ANTIDEPRESSANTS 53

should be monitored when used in this group. In the dementedelderly, trazodone is useful in treating behavioral disordersassociated with dementia.

The sedative effect of trazodone makes it useful in weaningpatients from benzodiazepines and other sedative drugs.

Nefazodone has shown efficacy in treating anxiety associ-ated with major depression. Similarly, mirtazapine has shownefficacy in anxiety symptoms in general.

Other Medical ConditionsMigraine and cluster headaches have been responsive toboth TCAs and SSRIs. Diabetic neuropathy and other painsyndromes such as facial pain, fibrositis, and arthritis have beenresponsive to both TCAs and SSRIs and the SNRI duloxe-tine. TCAs such as protriptylene have shown efficacy for sleepapnea. SSRIs have been used in the treatment of restless legsyndrome.

A full list of indications is shown in Table 2-2.

� TABLE 2-2. Various Uses of Antidepressants

Major DepressionAcute depressionPrevention of relapseOther depressive syndrome

Bipolar depressionAtypical depressionDysthymia

Other Uses

Tricyclic AntidepressantsStrong evidence

Panic disorder (most)Obsessive–compulsive disorder (clomipramine)Bulimia (imipramine, desipramine)Enuresis (imipramine)

Moderate evidenceSeparation anxietyAttention-deficit/hyperactivity disorderPhobiasGeneralized anxiety disorder


� TABLE 2-2. (Continued)

Anorexia nervosaBody dysmorphic disorderMigraine (amitriptyline)Other headachesDiabetic neuropathy, other pain syndromes

(amitriptyline, doxepin)Sleep apnea (protriptyline)Cocaine abuse (desipramine)Tinnitus

Evidence for but rarely used for these disordersPeptic ulcer diseaseArrhythmias

Monoamine Oxidase InhibitorsStrong evidence

Panic disorderBulimia

Moderate evidenceOther anxiety disordersAnorexia nervosaBody dysmorphic disorder

Atypical AgentsTrazodone

InsomniaDementia with agitationMinor sedative/hypnotic withdrawal

BupropionAttention-deficit/hyperactivity disorder

Serotonin Reuptake InhibitorsStrong evidence

Obsessive–compulsive disorder (high-dosefluoxetine, sertraline)Bulimia (fluoxetine)Panic disorder

Moderate evidenceGeneralized anxiety disorderObesity (high-dose fluoxetine)Substance abuseImpulsivity, anger associated with personalitydisorderPain syndromes, including diabetic neuropathy

Preliminary evidenceObsessive jealousy

ANTIDEPRESSANTS 55

Body dysmorphic disorderHypochondriasisBehavioral abnormalities associated with autism and mental

retardationAnger attacks associated with depressionDepersonalization disorderSocial phobiaAttention-deficit/hyperactivity disorder (as an adjunct)Chronic enuresisParaphilic sexual disordersNonparaphilic sexual disorders

Selective Serotonin Noradrenaline Reuptake InhibitorsDuloxetine

Moderate evidenceDiabetic peripheral neuropathic pain

VenlafaxineModerate evidence

Generalized anxiety disorderSocial phobiaPanic disorder

DRUG SELECTION AND INITIATION OF TREATMENTFOR MAJOR DEPRESSIONOn average, all antidepressants are equally effective. Withouta personal or family history of response to a particular agent,side effects are the most influential factors when selectingtreatment.

Both longitudinal and cross-sectional factors should beconsidered when selecting an antidepressant for majordepression:

• What were the course, duration, and severity of any previousepisodes of depression?

• Is there a history of antidepressant response?• How well was the antidepressant tolerated?• Is there a family member with a history of antidepressant

response; if so, to which medication?• If there is a history of antidepressant failure, were the trials

of an adequate dose and duration?• Are melancholic, atypical, or psychotic features present?


• Does the patient have a history of sensitivity to anticholin-ergic, histaminic, �-adrenergic, serotonergic, or noradren-ergic side effects?

• Does the patient have suicidal features or a history of impul-sivity that may increase the risk for potential overdose?

• Does the patient have a history of a cardiac conduction delayor recent myocardial infarction (which would contraindicatethe use of TCAs)?

• Does the patient have a history of symptoms suggestive ofmania (or more minor variants of manic episodes)? If so,avoid TCAs and MAOIs

• Does the patient presently take or need sympathomimetics(which would contraindicate the use of MAOIs)?

• If an antidepressant trial has failed, was the patient a partialresponder or a nonresponder?

• Were any augmentation strategies employed?

Special Considerations in the Selection of anAntidepressantThe following factors should be considered when selecting anantidepressant agent.

Gender Women may have slower gastrointestinal absorptionthan men due to less gastric acid and slower gastric emptying.A woman’s volume of distribution differs as well, given herincreased ratio of adipose tissue to lean body mass. Waterretention associated with the menstrual cycle may also affectthe volume of distribution. Oral contraceptives can alter thehepatic metabolism of TCAs.

Ethnic and Racial Factors African-Americans may be morelikely than those of European descent to be slow metabo-lizers of antidepressant due to possible genetic differences inmetabolic enzyme expression; a growing number of studiessuggest that African-Americans will have higher plasma levelsper dose of antidepressant. This has been primarily demon-strated with TCAs; data with SSRIs are lacking. Some datasuggest that Asians are slower to metabolize nortriptyline thanother groups.

ANTIDEPRESSANTS 57

The greatest concern regarding race and ethnicity issuggested in a study of prescribing practices in WestchesterCounty (New York) mental health clinics, which found thatminorities were less likely than nonminorities to be offeredantidepressant treatment, independent of diagnosis.

Age Is the patient elderly? If so, SSRIs are indicated as half-lives and steady-state concentrations of these agents are onlyminimally affected by age (although paroxetine may be anexception to this).

Comorbidities Patients with renal impairment may requiredose adjustment for many antidepressants; fluoxetine andsertraline appear to be the exception. Liver disease canincrease levels of TCAs – blood levels should be monitored.SSRIs are extensively metabolized in the liver and lower dosesshould be given in hepatic-impaired patients. Nefazodone isnot indicated due to association with hepatic injury.

Pregnant women and nursing mothers Clinicians mustconsider whether the potential benefits of treatment justifythe potential risk to the fetus. The risks of untreated depres-sion are well-described, and the risks of medication treatmentduring pregnancy and lactation are less well-known. In general,studies have not shown increases in fetal malformations fromantidepressant exposure. Reports of withdrawal syndromesamong neonates have led some experts to recommend taperingand discontinuing antidepressants 10 to 14 days before themother’s due date. Antidepressants are excreted in varyingdegrees in breast milk, so breastfeeding in antidepressant-treated women should be done with caution.

Predictors of poor response In assessing the likelihood ofsuccess, the following predictors of poor response havebeen identified: neurovegetative symptoms such as hyper-somnia, hyperphagia, psychomotor agitation, anxiety and irri-tability, personality features (such as neurotic, hypochondri-acal, and hysterical traits), multiple prior episodes, delusions,and psychomotor agitation.


Initiation of TreatmentStarting doses, titration schemes, and target doses forcommonly-used TCAs, MAOIs, SSRIs, and other agents areshown in Table 2-3. Once medication is initiated, it should begradually increased to therapeutic levels by titration.

TCAs are usually started at a relatively low dose; theselow doses are preferred in elderly patients. In the frail elderly,further dose reductions may be needed −50% or less of theusual starting dose.

The usual treatment plan with SSRIs is to start a patient atthe lowest effective dose and increase as indicated by clinicalresponse. An increased response in usually seen with increaseddose, however the dropout rate due to side effects alsoincreases. For children, adolescents, the elderly, and patientswho find medications generally difficult to tolerate, 50%reductions in the starting doses shown in Table 2-3 are accept-able. During titration, SSRIs may temporarily increase anxiety,insomnia, or both. These side effects can be alleviated by dosereduction, short-term use of benzodiazepines for anxiety, andthe addition of either hypnotics or trazodone for sleep.

With both TCAs and SSRIs there is a significant delaybetween initiation of medication and response; there is noreason to believe that increasing the dose beyond the thera-peutic range hastens response.

Bupropion (immediate and sustained release forms) andtrazodone require divided doses. In the elderly, a usual startingdose of bupropion is 100 mg/day of the sustained release prepa-ration, then increased to 100 mg b.i.d. Single daily dosing ispossible using the extended release form of bupropion.

Therapeutic Drug MonitoringAlthough blood levels are available for many antidepressants,those for imipramine, desipramine, and nortriptyline have beenbest established. Imipramine and desipramine appear to havea curvilinear dose–response curve with an optimal range of150 to 300 ng/mL. Nortriptyline appears to have a therapeuticwindow in the range of 50 to 150 ng/mL, usually reached bydoses ranging from 50–100 mg/day. These blood levels are

�TA

BLE

2-3.

Star

ting

Dose

,Titr

atio

nSt

eps,

and

Targ

etDo

ses

for

Com

mon

Ant

idep

ress

ants

STA

RTIN

GDO

SETI

TRA

TIO

NST

EPS

TARG

ETDO

SEDI

SCO

NTI

NUA

TIO

N/T

APE

RIN

G

TC

As

Am

itry

ptyl

ine,

imip

ram

ine,

doxe

pin,

amox

apin

e,cl

omip

ram

ine,

desi

pram

ine

50–7

5m

g/da

y↑o

f25

mg/

day

ever

y2–

3da

ysor

wee

kly

↑of

75m

g/da

y10

0–15

0m

g/da

yby

2nd

wee

k,15

0–20

0m

g/da

yby

3rd−

4thw

eeks

(300

mg/

dfo

ram

oxap

ine)

Tap

eral

lT

CA

sat

ara

teof

25–5

0m

g/da

yev

ery

2–3

days

Nor

trip

tylin

e25

–50

mg/

day

↑by

50m

g/w

eek

inyo

ung

pts,

25m

g/w

eek

inel

derl

y50

–100

mg/

day

Pro

trip

tylin

e10

–15

mg/

day

↑by

5–10

mg/

wee

k60

mg/

day

MA

OIs

Phe

nelz

ine

30m

g/da

y↑

by15

mg

afte

r3

days

,the

nw

eekl

y↑o

f15

mg/

day

45–9

0m

g/da

y

Tra

nylc

ypro

mid

e20

mg/

day

↑by

10m

gaf

ter

3da

ys,t

hen

daily

↑of

10m

gaf

ter

1w

eek

30–6

0m

g/da

y

Isoc

arbo

xazi

d20

mg/

day

↑by

10m

gaf

ter

3da

ys30

mg/

day

�TA

BLE

2-3.

(Con

tinue

d)

STA

RTIN

GDO

SETI

TRA

TIO

NST

EPS

TARG

ETDO

SEDI

SCO

NTI

NUA

TIO

N/T

APE

RIN

G

SSR

IsF

luox

etin

e10

–20

mg/

day

↑in

10–2

0m

gin

crem

ents

60m

g/da

y80

mg/

dfo

rO

CD

and

bulim

iane

rvos

a

Abr

upt

disc

onti

nuat

ion

acce

ptab

ledu

eto

long

half

-lif

eP

arox

etin

e10

–20

mg/

day

↑in

10–2

0m

gin

crem

ents

40m

g/da

yP

arox

etin

eco

ntro

lled

rele

ase

12.5

–25

mg/

day

↑in

12.5

–25

mg

incr

emen

ts50

mg/

day

Tap

erth

ese

SSR

Isov

er7–

10da

ys

Cit

alop

ram

10–2

0m

g/da

y↑i

n10

–20

mg

incr

emen

ts40

mg/

day

Sert

ralin

e25

–50

mg/

day

↑in

25–5

0m

gin

crem

ents

50–1

50m

g/da

yfo

rde

pres

sion

,15

0–20

0m

g/da

yfo

rO

CD

and

bulim

iane

rvos

aF

luvo

xam

ine

50m

g/da

yat

bedt

ime

↑in

50m

gin

crem

ents

100–

200

mg/

day

Esc

ital

opra

m10

mg/

day

↑in

10m

gin

crem

ents

20m

g/da

y

‘Aty

pica

l’an

tide

pres

sant

sT

razo

done

50–1

00m

g/da

y(r

equi

res

divi

ded

dose

s)↑i

n50

mg

incr

emen

ts15

0–30

0m

g/da

y(d

ose

not

toex

ceed

400

mg/

day

outp

atie

nt,

600

mg/

day

inpa

tien

t)B

upro

pion

sust

aine

dre

leas

e

100

mg/

day

for

seve

ral

days

then

b.i.d

.↑b

y10

0–15

0m

gin

crem

ents

300

mg/

day

(max

450

mg/

day)

Bup

ropi

onex

tend

edre

leas

e

150

mg/

day

3rdG

ener

atio

nV

enla

faxi

neex

tend

edre

leas

e37

.5m

g/da

y↑

to75

mg/

day

wit

hin

4–7

days

,the

nby

75m

g/da

yev

ery

4da

ys

225

mg/

day

(max

300–

450

inso

me

pati

ents

)

Dul

oxet

ine

20m

gb.

i.d.

↑by

20m

gin

crem

ents

20m

gor

30m

gtw

ice

daily

or60

mg

once

daily

Mir

taza

pine

15m

g/da

y↑b

y15

mg

incr

emen

ts15

–45

mg/

day

(max

60m

g/da

y)N

efaz

odon

e10

0m

gb.

i.d.

↑by

100–

200

mg/

wee

k30

0–60

0m

g/da

y


nominal, as some patients do respond above or below theseranges, and blood level monitoring should not be a substitutefor clinical observation.

Drug levels have not been well established for theMAOIs and the serotonin reuptake inhibitors. For the lattercompounds there are numerous speculations about possibledose–response curves, including linear and “therapeuticwindow” models.

Early, or Pre-response, PeriodPatients should initially be followed weekly to judge theirresponse to treatment and manage the side effects ofthe various medications; given realistic time and economicconstraints, such contacts may at times be by telephone.

Side effects are a primary reason for treatment non-adherence, and patients may be more inclined at this earlyphase to simply discontinue medication rather than to firstdiscuss it with their physician. It is inevitable that patients willexperience some side effects from their medications. Patientsare able to tolerate side effects better when they are framed in apositive light—as a sign that the drug is present in their system.

Strategies for reduction of side effects include:

• SSRIs, through scored tablets or liquid preparations, canbe begun at 25–50% of normal starting dose in patientsparticularly sensitive to side effects. Panic disorder patientsoften fit into this category

• TCAs, although often given in a single-daily dose, could bedivided across the day to minimize dose-related side effects

• Taking medications with food may decrease nausea• Reminding patients that sedating medications should be

taken in the evening, and activating ones in the morning• Flexibility in time of dosing: a significant percentage of

patients on SSRIs experience sedation rather than insomnia,warranting a change to evening dosing

Should intolerable side effects warrant medication change,select a medication in the same class as the first but with adifferent side effect profile; patients can have variable reac-tions to different SSRIs despite their apparent similarity. Forinstance, paroxetine is more sedative, and sertraline may causemore gastrointestinal distress than other SSRIs.

ANTIDEPRESSANTS 63

Response, or Acute Treatment, PeriodThis period overlaps with the initial phase of treatment andcontinues until response is achieved, usually two to fourmonths. The goal during this phase is to control the presentsymptoms of depression. It is important to differentiatebetween partial and complete response; complete responseimplies total recovery from all symptoms of depression,whereas a partial response is usually defined as a reduction insymptoms.

The time to response varies between patients. Few patientsshow a significant response before two weeks. The usual rangefor response is three to four weeks; however, it can take sixweeks or longer. For patients who complete a satisfactory treat-ment regimen, the response rate for antidepressants is about60 to 70%, although some of these responses will be partial.Response rates may be as high as 80% with antidepressantswhen an adequate dose is given for an adequate time.

TREATMENT OF PARTIALLY RESPONSIVE ANDNONRESPONSIVE PATIENTSThere is little benefit in making treatment changes before threeweeks, other than to mitigate side effects. Changes in treatmentstrategy should be considered after the physician is satisfiedthat the patient has been treated with an adequate dosage ofthe antidepressant for an adequate time, with current medica-tions increased to the limit of side effect tolerance.

In patients showing an inadequate response after a reason-able time, the decision is either to continue with the samemedication and augment with an additional agent, or to switchmedications altogether, depending on whether the patient hasshown any response to the current strategy. Partial respondersmay be more likely to benefit from treatment augmentations,whereas patients who show no response or worsen during treat-ment warrant a new agent.

Antidepressant AugmentationTypical augmentation strategies shown in Table 2-4 include theaddition of lithium carbonate, thyroid hormone, a stimulant,


an atypical antipsychotic, or by the use of antidepressantcombinations.

LITHIUM AUGMENTATIONLithium has been used successfully with most antidepressants,including TCAs, serotonin reuptake inhibitors, and bupropion,with response rates as high as 65%. The blood level of lithiumnecessary for adjunctive use has not been well established. It isprobably best to start at a low dose (300 mg b.i.d) and increaseto a therapeutic blood level (0.8 to 1.2 mEq/L) if there is noresponse. It may take three to six weeks for augmenting effect.

If augmentation is successful, it should be continuedthroughout the acute phase of treatment. This strategy can belimited by side effects and lithium’s narrow therapeutic index.It is associated with hypothyroidism, tremor, increased thirst,increased urination, nausea, weight gain, and acne. Patientsshould be cautioned to avoid dehydration, which can precipi-tate toxic lithium blood levels.

THYROID HORMONE AUGMENTATIONStarting dose is 25 �g/day of triiodothyronine, which can beincreased to 50 �g/day in a week if there is no response.The trial should continue for at least three weeks. Reportedresponse rates for thyroid augmentation are lower than thosefor lithium augmentation (25%). As with lithium augmenta-tion, if there is a positive response, it occurs relatively early.

STIMULANT AUGMENTATIONMethylphenidate and dextroamphetamine at dosages between5 and 20 mg have been used for antidepressant augmentation,but there are few systematic data regarding the proper dose orlength of treatment for this potential use.

ANTIDEPRESSANT DRUG AUGMENTATIONSecond-generation antipsychotic drugs are increasingly beingused adjunctively with antidepressants for residual ortreatment-resistant symptoms. APDs are generally used withintheir therapeutic range.

�TA

BLE

2-4.

Aug

men

tatio

nSt

rate

gies

AG

ENT

DOSI

NG

STRA

TEG

YLE

NG

THO

FTR

IAL

REPO

RTED

RESP

ON

SERA

TECO

MM

ENTS

Lit

hium

carb

onat

eSt

art

at30

0m

gb.

i.d.,

incr

ease

toth

erap

euti

cbl

ood

leve

l(0

.8–1

.2m

Eq/

L)

3–6

wks

As

high

as65

%B

est

docu

men

ted

stra

tegy

;has

been

com

bine

dw

ith

mos

tag

ents

Tri

iodo

thyr

onin

eSt

art

at25

�g/

d,m

ayin

crea

seto

50�

g/d

At

leas

t3

wks

At

25%

Equ

alto

lithi

umin

one

plac

ebo

cont

rolle

dtr

ial

Stim

ulan

ts(m

ethy

lphe

nida

tede

xtro

amph

etam

ine)

††

†F

ewsy

stem

atic

data

Aty

pica

lan

tips

ycho

tics

Beg

inw

ith

low

dose

(e.g

.,ol

anza

pine

2.5

mg/

day,

arip

ipra

zole

5m

g/da

y)gr

adua

llyin

crea

seto

full

ther

apeu

tic

dose

ifne

eded

††

Mai

nly

open

tria

ls;c

ontr

olle

dst

udie

sin

prog

ress

Com

bine

dan

tide

pres

sant

ther

apy

May

need

low

erdo

ses

than

usua

l(d

ueto

enzy

me

inhi

biti

on)

†

††

Mai

nly

open

tria

ls;c

ontr

olle

dst

udie

sin

prog

ress

Psy

chot

hera

pyN

/AV

arie

sby

ther

apy

Var

ies

byth

erap

yG

ood

data

for

both

cogn

itiv

e–be

havi

oral

ther

apy

and

inte

rper

sona

lth

erap

y

†In

adeq

uate

data

.


COMBINED ANTIDEPRESSANT THERAPYOpen trials have supported the use of combined therapy ofa TCA and a serotonin reuptake inhibitor in patients forwhom either class alone has failed. When antidepressants arecombined, it is important to remember that the serotonin reup-take inhibitors can potentiate TCA levels, and this should bemonitored carefully. MAOIs have also been used in combi-nation with TCAs, although this should be monitored closelygiven the risk of potential toxic interactions. Given the risk ofa serotonin syndrome, MAOIs should not be combined withserotonin reuptake inhibitors. Bupropion is frequently addedto an SSRI to enhance efficacy or to alleviate side effects suchas decreased libido, fatigue, and sedation.

NONPHARMACOLOGICAL APPROACHESA number of nonpharmacological augmentation options exist,particularly the concomitant use of psychotherapy. The combi-nation of psychotherapy and medication may offer benefitsthat either therapy alone cannot offer, including additional effi-cacy as well as prevention of relapse. This has been shown tosome degree for both cognitive–behavioral therapy and inter-personal therapy.

Changing to a New AgentPatient showing no response or whose condition deterioratesduring therapy should trial an alternative single agent. Recentstudies support the belief that it is best to switch to an agentof a different class, and approximately 50% of patients unre-sponsive to a first trial respond to an antidepressant of adifferent class.

When the switch involves an MAOI, sufficient time must begiven for medication clearance. Although seldom used, MAOIsmay be very effective in patients not responsive to other classesof antidepressants. Generally, 10 to 14 days is required forclearance of TCAs and MAOIs. Fluoxetine requires a muchlonger period—6 weeks—whereas sertraline and paroxetinerequire about two weeks when switching to an MAOI. Anotheralternative is changing from an SSRI medication to an SNRI.There is some evidence that SNRIs may lead to higher ratesof response and remission than SSRIs.

ANTIDEPRESSANTS 67

ECT ECT and other forms of stimulant therapies (e.g., vagalnerve stimulation) should be considered for patients who arenon-responsive to pharmaco- and psychotherapies.

Continuation and Maintenance Periods, andDiscontinuationCONTINUATION PERIODThis period usually lasts five to eight months after the endof the acute treatment period. The goal at this phase is theprevention of relapse. There is a high risk of relapse if treat-ment is discontinued after the acute treatment phase, withrates of 15% after six months and 22% after 12 months. Thetwo best predictors of relapse are a high number of previousdepressive episodes (greater than three predicted relapses) andunderlying dysthymic disorder.

Once a patient has responded to medication, treatmentshould be continued for a minimum of four to six months fromthe point of initial response. This period should be lengthenedfor the patient with a history of longer depressive episodes.

In the past, it was suggested that, on achievement ofeuthymia, doses could be reduced. However, it is more likelythat levels similar to those needed at the acute stage of treat-ment will be required throughout the continuation period.

MAINTENANCE PERIODThe goal of the maintenance period is to prevent the recur-rence of depression. There are a number of reasons to considerlong-term prophylactic therapy for depression rather thanmedication withdrawal:

1. Depression is a lifelong disease, with recurrence being thenorm rather than the exception

2. As the number of acute episodes increases, the risk of futureepisodes increases as well, and the interval between episodesshortens

3. Each subsequent episode carries a higher morbidity anddisability

4. There is a fear that treatment response may decrease withan increasing number of depressive episodes


Several factors influence the decision to maintain long-term prophylaxis for depression, including the seriousness ofprevious episodes, the severity of impairment caused by suchepisodes, the degree of response to previous treatments, andthe ability of the patient to tolerate the drug. Central in thedecision process is the concept of recurrent depression: thatsome patients are more likely than others to have a recurrenceof the disease. Three previous episodes of depression makerecurrent depression likely.

The best predictors of the likelihood of recurrence appearto be older age of onset and number of episodes. Long-termmaintenance is the treatment of choice for the following groupsof patients:

1. 50 years old or more at the time of the first depressiveepisode

2. 40 years old or more at first episode and have had at leastone subsequent recurrence

3. Anyone who has had more than three episodes

The recommended length of maintenance therapy varies fromfive years of treatment to indefinite continuation. Recentstudies with both acute major depression and recurrent depres-sion have shown significantly higher relapse rates 2–3 yearspost-discontinuation compared with patients on maintenancetreatment.

Regarding choice of agent, there are no rigorous studiescomparing different antidepressants during the maintenanceperiod. It is usually assumed that the same agent used in theacute and continuation period will also be the preferred onein the maintenance period.

Equally important in preventing recurrence of depressionis the problem of maintaining adherence to medication longafter the acute episode has resolved. Proper education andsupport will help with compliance. Toleration of side effects isimportant; patients are more likely to comply with agents thathave more favorable side effect profiles. SSRIs or bupropion(sustained or extended release forms) are generally the best-tolerated antidepressants.

Although lower doses for prophylaxis have been recom-mended, there are few data to support this contention. Even

ANTIDEPRESSANTS 69

though lower doses may increase compliance, full doses shouldbe used until new information indicates otherwise.

DISCONTINUATION OF TREATMENTBefore discontinuing treatment with an antidepressant it isimportant to remember that depression is often a lifelongdisease with a chronic course. One should always weighthe benefits of discontinuation against the risks of recurrentdepression. Patients with a single episode of acute depressionand who have an onset before age 50 are the best candidatesfor discontinuation.

For TCAs, the usual strategy is to taper the dose at a rate of25 to 50 mg/day every 2 to 3 days. Too rapid a decrease in dosemay produce symptoms of cholinergic ‘rebound’ or supersen-sitivity (nausea, vomiting, cramps), other signs of autonomichyperactivity (diaphoresis, anxiety, agitation, headache), andinsomnia as early as 48 hours or as late as 2 weeks after discon-tinuation. These early symptoms may be mistaken for relapseof depression.

MAOIs may also have a withdrawal syndrome on abruptcessation of treatment, including symptoms of psychosis;however, this is rarer than that seen with the TCAs.

Recommendations for discontinuing SSRIs depend on theparticular drug. Fluoxetine has a long half-life and abruptwithdrawal should be permissible. Shorter half-life drugs, suchas sertraline, citalopram, escitalopram, and paroxetine, mayrequire a 7–10-day taper. The withdrawal syndrome withshorter-acting agents includes symptoms of fatigue, insomnia,abdominal distress, and influenza-like symptoms. The samemay be true for venlafaxine and duloxetine.

A first episode of depression has a high risk of recur-rence and the risk is higher in patients who have only apartial response to treatment. After discontinuation the goalis to enable early intervention if symptoms recur. The patientshould be educated to recognize the symptoms of depressionand seek help at an early stage.

SIDE EFFECTSOnce the choice of an antidepressant has been made, themain goal is to maximize therapeutic effects and minimize side


effects. Good preparation and reassurance of the patient isessential. Even relatively benign side effects are a major causeof treatment nonadherence; drop-out rates ranging from 7 to44% have been reported in various studies of TCAs, and from7 to 23% in studies of serotonin reuptake inhibitors. Commonside effects for TCAs and SSRIs are shown in Table 2-5.

Anticholinergic EffectsCommon side effects include dry mouth, constipation, urinaryretention, blurred vision, and less commonly narrow-angleglaucoma. These effects are usually dose-related and worse inpatients with pre-existing defects.

Dry mouth can be alleviated by saliva substitutes or sugar-free hard candy, over-the-counter bulk-forming laxatives cantreat the constipation caused by some agents, and the cholin-ergic agonist bethanechol has been used to treat anticholin-ergic side effects, particularly urinary hesitancy and blurredvision. The usual adult dosage is 10 to 50 mg t.i.d. or q.i.d.Pilocarpine eye drops can also be used to treat blurred vision.SSRIs in general are free from these side effects, as is bupro-prion, although duloxetine and paroxetine may cause mild drymouth and constipation.

Other effects include tachycardia, impaired memory andcognition (in severe cases leading to delirium), and exacerba-tion of texisting tardive dyskinesia.

Rarely, in combination with other anticholinergic drugs,TCAs can precipitate a central anticholinergic delirium andconfusion. Patients with psychotic depression may be at higherrisk as their treatment regimen may include a TCA, a low-potency antipsychotic, and an antiparkinsonism agent. If thissyndrome is suspected, a physostigmine challenge can be diag-nostic and may produce a dramatic reversal. Abrupt with-drawal of TCAs can put patients at risk of cholinergic rebound.This syndrome includes gastrointestinal distress, anxiety, andautonomic instability. Whenever possible, TCA doses shouldfirst be slowly tapered and then discontinued.

Autonomic EffectsThe most frequent and treatment-limiting side effect ofTCAs is orthostatic hypotension. The geriatric population is

ANTIDEPRESSANTS 71

� TABLE 2-5. SSRI, SNRI and TCA Side Effects

Side Effects of Serotonin BlockadeJitteriness, activationInsomniaAnorexia, weight lossNausea, vomiting, diarrheaSexual dysfunctionExtrapyramidal-like side effectsIncrease in suicidal ideas

Signs and Symptoms of Central Serotonergic SyndromeConfusion, disorientationHeadacheAutonomic instabilityLethargyRestlessnessAbdominal cramps, diarrheaTremor, myoclonic jerks

Side Effects of Norepinephrine BlockadeTremorsJitterinessTachycardiaDiaphoresisAugmentation of pressor effects of sympathomimeticsSexual dysfunction

Anticholinergic Side EffectsDry mouthConstipation (rarely—paralytic ileus)Urinary hesitancy (rarely—dystonic bladder)Blurred visionSinus tachycardiaMemory impairmentWorsening of narrow-angle glaucoma

Anticholinergic Delirium“Mad as a hatter”: confusion, disorientation, visual hallucinations“Hot as a hare”: hyperpyrexia“Blind as a bat”: loss of visual accommodation“Red as a beet”: peripheral vasodilation“Dry as a bone”: drying of mucous membranes

From Mendelowitz AJ, Dawkins K and Lieberman JA (2000). Antide-pressants. In Psychiatric Drugs, Lieberman JA and Tasman A (eds.) WBSaunders, pp. 44–77. © 2000, with permission from Elsevier.


especially vulnerable to falls; patients should be educated inhow to change position safely. Newer TCAs such as nortripty-line and doxepin have a lower risk of this side effect. Trazodonehas significant �1-adrenergic blockade and as a result alsoinduces significant orthostatic hypotension.

Both SSRIs and TCAs may also cause sweating, palpita-tions, and increased blood pressure. Venlafaxine and duloxe-tine also cause sweating and tremor, and slightly increase bloodpressure and heart rate; increases in blood pressure are mostnotable in doses of venlafaxine above 300 mg/day.

Neurologic EffectsSeveral TCAs may lower the threshold for epileptic seizures,although as a class they have relatively low risk for inducingseizures. Initial doses of TCAs should be reduced in epilepticpatients. Some SSRIs may increase the frequency of seizuresat high dose, particularly fluoxetine at doses of 100 mg/day ormore. Buproprion is the agent most likely to reduce the seizurethreshold. In doses below 450 mg/d, the incidence of seizuresis less than 0.5%. To avoid inducing a seizure, one should givebupropion in divided doses, with a maximum dose of 450 mg/d(400 mg/d for the extended release formulation). Seizure riskis likely to be lower with sustained or extended release formsof buproprion.

Myoclonic twitches and tremor of the tongue and upperlimbs can occur in 5–10% of patients on SSRIs and lessfrequently with TCAs; this may respond to dose reduction orthe addition of a low-dose beta blocker such as propranolol.Anxiety, nervousness, insomnia, and sedation may also occurwith SSRI agents; these symptoms may respond to a dosereduction or to the short-term addition of a low-dose benzo-diazepine. Insomnia is often already part of the depression;the addition of a low-dose of a sedating antidepressant such astrazodone, or a sedative–hypnotic, may allieviate symptoms inthe short term. It is more common with high doses of fluoxe-tine. Headache may also occur with fluoxetine.

More rarely, paresthesia, ataxia, and speech blockade mayoccur with both TCAs and SSRIs.

There is some question as to whether SSRIs caninduce extrapyramidal-like symptoms, specifically akathisia. It

ANTIDEPRESSANTS 73

remains unclear whether this is actually an extrapyramidaleffect or part of the agitation and anxiety that these agents canproduce. The TCA amoxapine can cause extrapyramidal sideeffects including acute dystonic reactions, parkinsonism symp-toms, akathisia, tardive dyskinesia, and neuroleptic malignantsyndrome. If akathisia is suspected, dosage reduction may berequired.

Weight GainBoth TCAs and SSRIs cause weight gain; up to 30% of patientson SSRIs experience this side effect, with increases of 8 kgor more. This is less of a concern during the acute phase oftreatment, when insomnia and anorexia are often present, butthey can be treatment limiting when long-term use is indicated.Medication switches to less weight-increasing medication maybe helpful.

Gastrointestinal SymptomsNausea, vomiting, anorexia, and diarrhea are frequentcomplaints with SSRIs. They are often dose and titrationdependent and can be lessened by dose reduction, a slowerinitial titration, and having the patient take medications withfood. Anorexia usually occurs only early in treatment and is notpersistent. Side effects and treatment discontinuation duringthe early days of treatment associated with paroxetine may bereduced by prescribing the modified-release formulation PaxilCR. Compared with immediate-release tablets, this formula-tion significantly reduces the incidence of nausea during thefirst week of therapy. The most frequent adverse effect asso-ciated with both venlafaxine and duloxetine is nausea.

Sexual DysfunctionEvery class of antidepressants is associated with some degreeof sexual dysfunction. Difficulties can include problems withlibido, impotence, ejaculatory dysfunction, and anorgasmy.The incidence of sexual dysfunction is as high as 30% withSSRIs. As with other side effects, some of these may be symp-toms of the depression itself, and they may predate the initi-ation of the antidepressant trial. Treatment can include dose


reduction, skipping of a dose of an SSRI with a short half-life(e.g., sertraline), or changing to an antidepressant of a differentclass. Anorgasmia may be treated by the addition of yohimbineor amantadine or by the use of bethanechol or cyproheptadineone to two hours before sexual activity. Decreased libido maybe treated by the addition of the antidepressant bupropion; ingeneral, this agent has the best profile with respect to sexualdysfunction. Trazodone has been associated with priapism inrare cases; this must be treated urgently to avoid long-termimpairment.

Cardiac ConductionTCAs affect cardiac conduction; as a result, they have antiar-rhythmic properties and can slow cardiac conduction. Theseeffects are responsible for the cardiotoxicity of TCAs andtheir danger in overdose. TCAs do not affect cardiac output,however they should not be used in patients with a preex-isting conduction delay greater than a first-degree block or inpatients immediately after myocardial infarction.

Suicide RiskSSRIs have a very wide therapeutic index and are not lethal inoverdose. This has made them especially popular in the treat-ment of patients with a history of impulsivity. The possibilitythat treatment with any type of antidepressant may increasethe risk of suicide in adults is currently under active review.It is clinically prudent to limit the amount prescribed duringthe initiation of therapy or when patients are only partiallytreated. The risk of suicide attempts during these intervals ishigh. More than a 1-week supply of a TCA is potentially lethalin an overdose attempt.

Pooled analysis of clinical trials involving children andadolescents suggests that antidepressants are associated withsuicidal thinking or behavior in about 4% of patientscompared with 2% receiving placebo during the early monthsof treatment.

All patients treated with an antidepressant (adults as wellas children and adolescents) should therefore be closely moni-tored during the first few months. Treatment may need to

ANTIDEPRESSANTS 75

be adjusted or discontinued if depression continues to worsenor suicidal ideation emerges, especially if the symptoms aresevere, abrupt in onset, or not part of the patient’s presentingsymptoms. Patients should be warned of this possibility andprovided with information about the risk and advice on whatprecautions to take. They should be warned to report newor increased signs of irritability, agitation, or suicidality to ahealthcare professional.

Allergic and Hematologic EffectsTCAs are associated with exanthematous rashes in 4–5% ofpatients, while SSRIs have been known to cause a variety ofrashes in a similar percentage of patients.

Other EffectsTCA’s can cause gynecomastia and amenorrhea in somepatients, while SSRIs can cause increased prolactin levelscausing mammoplasia and galactorrhea in both men andwomen. SSRIs have also been known to reduce blood glucoselevels, though this is rare. Nefazodone has been associatedwith rare cases of life-threatening liver failure, and has beenremoved from the market in many European countries. As aresult, nefazodone should not normally be considered for usebefore other antidepressants.

Central Serotonergic SyndromeAny agent that increases serotonergic function can put patientsat risk of a central serotonergic syndrome. This syndrome mostcommonly affects patients on multiple serotonergic drugs, andinvolves multiple systems. The syndrome can in rare casesprogress to rhabdomyolysis, acidosis, respiratory compromise,disseminated intravascular coagulation, and cardiac collapse.

Monoamine Oxidase InhibitorsThe declining popularity of MAOIs in the United States isprimarily the result of the need for dietary restrictions andthe potential for serious side effects. Use of MAOIs must be


limited to patients who are compliant with a tyramine-free dietand do not use any medications that contain sympathomimeticamines. A hypertensive crisis is a potentially fatal, though rare,complication of these drugs. The crisis is usually preceded bya prodrome that includes increased blood pressure, headache,stiff neck, and vomiting. Treatment needs to be immediate andmay include the administration of phentolamine, 5 mg i.v.

DRUG INTERACTIONSThe most common drug interaction with antidepressants istheir influence on the metabolism of other drugs. Antide-pressants are metabolized through catabolic enzymes locatedin a variety of places, but primarily in the smooth endo-plasmic reticulum of hepatocytes: the cytochrome P450 (CYP)enzymes. Many other drugs are also metabolized throughsimilar pathways: it is estimated that about half of all drugsprescribed depend on CYP for their metabolism.

Of these enzymes, most drugs are metabolized by theenzymes CYP3A4 (50%) and CYP2D6 (20%). Other clinicallyimportant CYP enzymes include CYP2C9, and CYP2C19 andCYP1A2, the latter which is found in the brain and may affectCNS distribution of antidepressants. A list of the enzymes andsome drugs commonly affected by them is found in Table 2-6.

Tricyclic AntidepressantsAs with any combination therapy, the side effects describedpreviously can be additive with other similar drugs. Most prob-lematical are the anticholinergic effects of the TCAs. Suchcholinergic blockade is a property shared by many other medi-cations, including numerous over-the-counter preparations.The general sedative properties of these medications can alsoaugment any soporific. The slowing of cardiac conduction canalso potentiate other medications that produce similar effects,such as type IA antiarrhythmics and anticholinergic medica-tions. Adrenergic receptor blockade can worsen the orthostatichypotension caused by other medications, including vasodila-tors and low-potency antipsychotic medications.

ANTIDEPRESSANTS 77

� TABLE 2-6. Cytochrome P-450 Isoenzymes and Common Medi-cations Inhibiting or Inducing Them

INHIBITORS INDUCERS

1A2 Cimetidine, fluoroquinolones,fluvoxamine, ticlopidine

Tobacco

2C19 Fluoxetine, fluvoxamine,ketoconazole, lansoprazole,omeprazole, ticlopidine

2C9 Amiodarone, fluconazole, isoniazid,ticlopidine

Rifampin, secobarbital

2D6 Amiodarone, chlorpheniramine,cimetidine, clomipramine,fluoxetine, haloperidol,methadone, mibefradil,paroxetine, quinidine, ritonavir

3A4,5,7 HIV protease inhibitors:(indinavirn, elfinavir, ritonavir,saquinavir, amiodarone, but notazithromycin) cimetidine,clarithromycin, erythromycin,fluoxetine, fluvoxamine,grapefruit juice, itraconazole,ketoconazole, mibefradil,nefazodone, troleandomycin

Carbamazepine,phenobarbital,phenytoin,rifabutin, rifampin,St. John’s wort,troglitazone

Pharmacokinetic EffectsAbsorption of TCAs can be inhibited by cholestyramine,which therefore must be given at different time intervalsthan the antidepressants. Specific substances reported toincrease TCA levels include fluoxetine, antipsychotic medica-tions, methylphenidate, and cimetidine. Methylphenidate hasbeen combined with desipramine to treat attention deficitsand depression in children. The combination therapy had ahigher incidence of ECG changes (particularly higher ventric-ular rates), nausea, dry mouth, and tremor. Enzyme “inducers”that can lower tricyclic agent levels include phenobarbital andcarbamazepine. The nicotine from cigarettes can also induceenzyme activity.

Guanethidine is contraindicated with TCAs, as it relies onneuronal reuptake for its antihypertensive effect. Clonidine,


a presynaptic alpha-2-receptor noradrenergic agonist, is alsocontraindicated, as it works in an antithetical fashion totricyclic medications.

Monoamine Oxidase InhibitorsAs with the dietary proscriptions, any medication that increasestyramine can precipitate a hypertensive crisis; such medicationsinclude numerous over-the-counter preparations for coughs,colds, and allergies. The same rule applies to sympathomimeticdrugs (such as epinephrine and amphetamines or cocaine) anddopaminergic drugs such as anti-parkinsonian medications.

The combination of MAOIs and narcotics, particularlymeperidine, may cause a fatal interaction. The reaction canvary from symptoms of agitation and hyperpyrexia to cardio-vascular collapse, coma, and death. A similar reaction has alsobeen reported when propoxyphene, diphenoxylate hydrochlo-ride, and atropine are used with MAOIs. The combination ofMAOIs with agents such as SSRIs can cause the serotoninsyndrome.

Serotonin Reuptake InhibitorsSerotonin reuptake inhibitors are potent inhibitors of theCYP2D6 pathway and can slow the metabolism of anydrug that is also metabolized by that pathway. Such drugsinclude TCAs, carbamazepine, phenothiazines, butyrophe-nones, opiates, diazepam, alprazolam, verapamil, diltiazem,cimetidine, and bupropion. Paroxetine appears to be the mostpotent inhibitor of this metabolic pathway, with fluoxetinealso showing high potency. Sertraline is a somewhat lesspotent inhibitor. These pharmacokinetic interactions are bestmanaged with dosage adjustment. Fluoxetine, for example, canbe safely used with tricyclic medications if TCA blood levelsand, possibly, electrocardiograms are monitored. Although itbinds more weakly to the CYP2D6, sertraline has also beenreported to raise the level of TCAs significantly. In the case ofbupropion, this relative increase in the blood level can increasethe risk for seizures.

Particular caution should be used when a patient usingmultiple medications starts a serotonin reuptake inhibitor, as

ANTIDEPRESSANTS 79

the interactions with other drugs can cause dangerous increasesin levels. For example, in the cardiac patient, levels of warfarinshould be monitored as fluoxetine has been reported to raisethese levels. Several case reports exist of increased antiar-rhythmic levels after introduction of fluoxetine, which resultedin potential serious bradyarrhythmias.

Fluoxetine has also been reported to raise lithium levels.The mechanism for this is not clear, as lithium is primarilyexcreted through the kidneys.

Other Second-Generation AntidepressantsFew reports exist of interactions with other drugs andtrazodone, although trazodone may increase levels of digoxin,phenytoin, and possibly warfarin. Bupropion causes fewdrug–drug interactions. The main interactions reported haveoccurred when bupropion is combined with another dopamin-ergic agent. For example, when bupropion was used withl-dopa, the combination caused excitement, restlessness,nausea, vomiting, and tremor.

Third-Generation AntidepressantsVenlafaxine does not substantially inhibit the CYP enzyme,and is not highly protein-bound, thus it tends to have few clin-ically significant drug–drug interactions. Duloxetine is metab-olized by 1A2 and 2D6 P450 isoenzymes, and may increaseplasma levels of other antidepressants, antipsychotics, andType 1C antiarrhythmics, such as flecainide (Tambocor). Nefa-zodone is highly protein-bound, and has several active metabo-lites. It is also a strong inhibitor of CYP3A4, and affectsother drugs also metabolized by that pathway; however, it haslittle affinity for the CYP2D6 enzyme. Mirtazapine is highlyprotein-bound as well, but appears to only weakly affect thecytochrome enzymes.

SUMMARYThere remain a number of important limitations regarding thepharmacotherapy of depression. Some of these limitations canbe addressed by continued progress on current research. Other


limitations await truly novel research into the mechanism ofdepression.

Recommendations for the Use of AntidepressantsAllowing for the many limitations, we can generalize fromavailable data to make the following recommendations:

1. All patients with acute major depression should be consid-ered reasonable candidates for pharmacotherapy.

2. There is adequate evidence to make the same recommenda-tion for other forms of depression. This is particularly truefor dysthymia, and may be true for other minor forms ofdepression as well.

3. There is good evidence for the use of antidepressantsfor non-mood disorders as well, particularly the anxietydisorders.

4. There remains no strong evidence from choosing one medi-cation over another, and treatment recommendations shouldbe made on the basis of tolerability and, if appropriate, cost.

5. Extended treatment should be recommended both forpatients with chronic depression and recurrent depression.

6. Maintenance treatment should consist of the same dose ofantidepressant as used to achieve acute phase remission.

7. The exact length of maintenance treatment is not known.The decision for indefinite treatment should be a risks–benefit decision, made with the informed consent of thepatient, who is entitled to be informed of the limitations inour knowledge.

ADDITIONAL READING

1. Weissman, MM. Treatment of Depression: Bridging the 21st Century.American Psychiatric Press, Washington D.C., 2001

2. Potokar J, Thase M. Advances in the Management and Treatmentof Depression. Taylor & Francis, London, 2003

3. Lieberman JA, Greenhouse J, Hamer RM, Krishnan KR, NemeroffCB, Sheehan DV, Thase ME, Keller MB: Comparing the effectsof antidepressants: consensus guidelines for evaluating quantitativereviews of antidepressant efficacy. Neuropsychopharmacology 2005;30(3):445–460

ANTIDEPRESSANTS 81

4. Wong IC, Besag FM, Santosh PJ, Murray ML. Use of selectiveserotonin reuptake inhibitors in children and adolescents. DrugSafety 2004; 27:991–1000

5. Practice guidelines for the treatment of patients with major depres-sive disorder, Second Edition. In: American Psychiatric AssociationPractice Guidelines for the Treatment of Psychiatric Disorders,Compendium 2004, American Psychiatric Publishing, ArlingtonVA, 2004

3MOOD STABILIZERS

INTRODUCTIONAlthough a wide variety of medications are employed in thetreatment of bipolar disorder and related conditions, the essen-tial ingredient in any treatment regimen is one or more moodstabilizers. By definition, these are medications which caneffect at least one phase of the illness without worseningany other phases. Currently accepted mood stabilizers fallinto three families: (1) lithium, (2) a subset of anti-epilepticdrugs (anticonvulsants), and (3) all of the second-generationor atypical antipsychotics. Agents currently approved for thetreatment of bipolar disorder are listed in Table 3-1.

Lithium was approved by the US Food and Drug Admin-istration (FDA) in 1970, is the medication with the longesthistory of use for bipolar disorders, and remains a primaryagent for the acute and maintenance treatment of the disorder.However, the rate of non-response to lithium is between 20and 40%, it is less effective for bipolar depression than formood elevation, and it possesses a relatively narrow thera-peutic index. In addition, lithium is most effective in patientswith classic symptom patterns (e.g., pure euphoric mania) andearly in the course of illness. Patients with dysphoric and mixedepisodes, rapid cyclers, patients with a history of neurologicdisease, and those with comorbid substance abuse may be



� TABLE 3-1. Current FDA Indications

MEDICATION ACUTE INDICATION MAINTENANCE INDICATION

Lithium Mania ManiaDivalproex ManiaCarbamazepine Mania∗

Lamotrigine Mania or Depression,Monotherapy

Olanzapine ManiaMania, Adjunct toLi and Valproate

Olanzapine &Fluoxetine Comb.

Depression

Quetiapine Mania, MonotherapyMania, Adjunct toLi and Valproate

Risperidone Mania, MonotherapyMania, Adjunct toLi and Valproate

Ziprasidone Mania, MonotherapyAripiprazole Mania, Monotherapy Mania, Monotherapy

∗ Extended release formulation.

less likely to respond to lithium than some other mood stab-ilizers. Therefore, although lithium remains a mainstay ofmanagement, newer agents such as anticonvulsants and atyp-ical antipsychotics play an increasingly important role.

Long-term prospective studies have shown that, evenwith conscientious medication treatment, patients with bipolardisorder are still symptomatic approximately half of thetime. Despite increasing numbers of medications approvedfor the treatment of bipolar disorder, we still often strugglewith incomplete efficacy, significant side effects, and theneed to combine medications to maximize response. Residualsymptoms and breakthrough episodes all too often leadto suffering, dysfunction, and loss of life. However, whenthe pharmacologic management of the bipolar patient isapproached in an organized and systematic way, basedupon updated knowledge of the strengths and limitations ofspecific agents, and when comorbidities are addressed and

MOOD STABILIZERS 85

psychosocial supports bolstered, one can often see tremendousimprovements.

PHARMACOLOGYChemistryLithium is clearly the most structurally simple psychotropicagent, with therapeutic activity residing in the lithium ion, Li+.The anticonvulsants are chemically unrelated agents that arenot classified by their chemistry, except for closely relatedagents such as carbamazepine and oxcarbazepine, or differentsalts such as valproate and divalproex. The chemistry ofantipsychotic agents is summarized in Chapter 1.

Mechanism of ActionThe mechanism of action of anti-manic drugs is poorlyunderstood. Whereas alterations in monoamine systems aregenerally believed to be critical for antidepressant response, itis not clear which neurotransmitter systems are involved in themood stabilization. Indeed, few significant changes in neuro-transmitter levels have been measured following treatmentwith mood stabilizers. Rather, a range of clinical and preclin-ical studies strongly suggests that the critical site of actionfor many mood stabilizers occurs at the level of intracellularsecond-messenger systems.

LITHIUMLithium is a monovalent cation and appears to exert its clinicaleffect by directly interacting with intracellular targets. At ther-apeutic serum concentrations, lithium is a direct competitorof magnesium at several important regulatory enzymes,including inositol-monophosphatase (IMPase), which catalyzesthe rate limiting step in the phosphoinositol signaling cascade.According to the inositol depletion hypothesis, inhibition ofIMPase by lithium reduces myoinositol and phosphoinositidephosphate (PIP-2), thereby reducing G protein mediatedsignaling, intracellular calcium mobilization and protein kinase


C (PKC) activation. Ultimately, these cascades produce wide-ranging effects on gene transcription, ion channel functionand even neuronal structure and synaptic function. It hasbeen postulated that lithium shifts the level of activationin these complex and integrated second messenger systems,providing neurons (and the circuits in which they act) withgreater stability and resilience. Magnetic resonance spec-troscopy studies have recently demonstrated that five days oflithium treatment significantly reduces frontal cortex myoinos-itol levels. In support of this system being therapeuticallyrelevant, two PKC inhibitors, tamoxifen and riluzole, havebeen shown to be effective in treating bipolar mania anddepression, respectively. Another enzyme directly inhibitedby lithium is glycogen synthase kinase-3 (GSK-3), which hasbeen implicated in a wide range of normal and pathologicalprocesses. GSK-3 influences transcription factors and regula-tors of cellular metabolism, impacting upon neuronal resilienceand survival. Specific GSK-3 inhibitors are being activelystudied for potential therapeutic use in Alzheimer’s disease,diabetes and other disorders.

ANTICONVULSANTSThe mechanisms by which some of the anticonvulsants stabi-lize mood are unknown. Valproate acts at sodium channels,at several important steps in GABA metabolism, and onthe activity of histone deacetylase (which influences genetranscription). It has also be shown in preclinical studiesto produce changes in the phosphoinositol pathway similarto those caused by lithium, including a reduction of PKCactivity at clinically relevant serum levels.

Carbamazepine exerts a range of effects on neurotransmittersystems, increasing acetylcholine in the striatum, and decreasingdopamine and GABA turnover, and norepinephrine release.It is not clear which, if any, of these effects are relevant to itstherapeutic benefit for bipolar disorder. There has been lessstudy of carbamazepine’s effect on second messenger systems,but it appears to reduce phosphoinositol signaling and theactivities of adenylate cyclase and guanylate cyclase.

The anticonvulsant activity of lamotrigine is hypothesizedto result from blockade of voltage-sensitive sodium channels,

MOOD STABILIZERS 87

which inhibits the release of the presynaptic excitatoryamino acids aspartate and glutamate. Lamotrigine also blockshigh-voltage activated N- and P-type calcium channels andinhibits serotonin reuptake. To date, the influence of lamot-rigine on intracellular signaling cascades is unknown.

ANTIPSYCHOTICSThe pharmacology of antipsychotic drugs is described inChapter 1. Their mechanism of action in the treatmentof bipolar disorder is uncertain but may involve the sameeffects on serotonin and dopamine systems that are believedto underlie the effectiveness of these medications forschizophrenia. Atypical antipsychotics are preferred to olderagents due to a reduced risk of tardive dyskinesia, a generallymore benign side effect profile, and evidence of some efficacyin the treatment of bipolar depression.

PharmacokineticsThe pharmacokinetic properties of mood stabilizers aresummarized in Table 3-2.

� TABLE 3-2. Pharmacokinetics

GENERIC NAME TRADE NAMEPEAK PLASMA

LEVELS (h) HALF-LIFE (h)

Carbamazepine Tegretol and others 1.5–6 25–65 or 8–29Gabapentin Neurontin 3 5–7Lamotrigine Lamictal 1–5 26Lithium Eskalith and others 1–4 18–36Topiramate Topamax 2 21Valproate Depakote and others 2–4 9–16Aripiprazole Ability 3–5 75∗

Olanzapine Zyprexa 6 27Quetiapine Seroquel 1.5 7Risperidone Risperdal 1 3∗∗

Ziprasidone Geodon 6–8 4–10

∗ t1/2 of active metabolite is 94 hours.∗∗ t1/2 of active metabolite is 24 hours.


LITHIUMLithium is available as both immediate release lithium carbonateas well as in two slow-release formulations. The latter aregenerally better tolerated, produce lower peak levels for a givenarea under the curve, and allow for once daily administration.In any form, lithium has the following properties:

• Rapid and complete absorption after oral administration.• Low protein binding.• Absence of liver metabolism.• Peak plasma levels achieved within 1.5 to 2 hours for stan-

dard preparations or 4 to 4.5 hours for slow-release forms.• Plasma half-life of 17 to 36 hours.• 95% drug excretion by the kidneys, with excretion propor-

tionate to plasma concentrations.

Because lithium is filtered through the proximal tubules,factors that decrease glomerular filtration rates will decreaselithium clearance. Sodium also is filtered through the proximaltubules, so a decrease in plasma sodium can increase lithiumreabsorption and lead to increased plasma lithium levels.Conversely, an increase in plasma lithium levels can cause anincrease in sodium excretion, depleting plasma sodium.

ANTICONVULSANTSValproate is available as valproic acid or divalproex sodium,a compound containing equal parts valproic acid and sodiumvalproate. Divalproex is better tolerated than valproic acid, hasbeen studied more extensively, and is more frequently used.Divalproex is available in immediate, delayed, and extendedrelease forms. Valproate is characterized by the following:

• Rapid absorbtion after oral administration, reaching peakplasma levels within 2 to 4 hours of ingestion.

• Bioavailability unaffected by food, though absorption maybe delayed.

• Rapid distribution and high (90%) plasma protein binding.• Half-life between 9 and 16 hours, with 1 to 4 days needed

to attain steady state.

Carbamazepine is available in an immediate release andthree extended release forms. It is metabolized by the

MOOD STABILIZERS 89

hepatic cytochrome P450 2D6 system, which it also robustlyinduces. As a result of this autoinduction, the rate ofmetabolism of carbamazepine (and other P450 substrates)usually increases over the first several weeks of treatment(see ‘Drug Interactions’ on page 119). Initial steady state maybe attained within 4 to 5 days, but autoinduction may delayfinal steady state until 3 to 4 weeks after treatment initiation.This calls for added vigilance on the part of the clinician, as thelevel of carbamazepine must be monitored and its dose oftenmust be raised during this early phase of treatment. Additionalpharmacokinetic properties of carbamazepine include:

• Occurrence of peak plasma levels within 4 to 6 hours afteringestion of the solid dosage form.

• Bioavailability of 85%, which may be reduced when the drugis taken with meals.

• Plasma protein binding of 80%.• A half-life (after 3–4 weeks of treatment) ranging from 5 to

26 hours.• Production of an active metabolite, 10, 11-epoxide, which has

a half-life of about 6 hours. This metabolite, also known asoxcarbazepine, has been utilized as an anticonvulsant itself,and is undergoing clinical testing as a mood stabilizer.

Lamotrigine is primarily metabolized by glucuronic acid conju-gation. Enzyme induces, such as carbamazepine, increase lamot-rigine metabolism whereas competitors, such as valproate,markedly decrease its metabolism (see ‘Drug Interactions’ onpage 119). Other pharmacokinetic parameters include:

• Complete absorbtion after oral administration.• Bioavailability of 98%, which is unaffected by food.• Peak plasma concentrations occurring between 1 and 2 hours

after administration, but delayed to up to 5 hours in patientsalso taking valproate.

• Protein binding of approximately 55%, making clinicallysignificant interactions with other protein-bound drugsunlikely.

• Half-life of 12–27 hours, increasing to 59 hours in patientsalso taking valproate.

• Occurrence of steady-state levels in 3 to 10 days.


ANTIPSYCHOTICSThe pharmacokinetic characteristics of atypical antipsychoticsare described in Chapter 1.

INDICATIONSThe principle indications in the treatment of bipolardisorder are:

• Acute mania and mixed mania• Acute depression• Maintenance therapy• Rapid cycling

DRUG SELECTION AND INITIATION OF TREATMENTAcute ManiaThe goal of treatment should be to suppress completely allsymptoms of mania and return the patient to his or hermental status quo ante. Mood, thinking, and behavior shouldnormalize, though different symptom clusters may improve atdifferent rates. For economic or social reasons, it is sometimesnecessary to discharge inpatients with partially remitted maniaor hypomania. In this setting, it is imperative to follow suchpatients closely (perhaps using partial hospitalization or inten-sive outpatient programs) until total remission of symptomshas occurred. In particular, it is important to monitor, andperhaps provide prophylactic pharmacotherapy, for the rapidslide into acute depression that frequently follows mania.

Drug SelectionLITHIUMFor many patients with hypomania, lithium by itself can inducea total remission. Even in acute mania, response rates in clin-ical trials using lithium monotherapy can be as high 80–90%. Ineveryday clinical practice, however, the use of additional moodstabilizers (anticonvulsants or atypical antipsychotics) may benecessary for remission. In addition, manic psychosis callsfor the use of adjunctive antipsychotics and in many patientsuse of benzodiazepines may be very helpful for symptomatic

MOOD STABILIZERS 91

treatment of agitation and anxiety. Certain clinical features aresomewhat predictive of lower lithium response rates. Theseinclude:

• Rapid cycling course• Mixed mania• Greater number of cumulative episodes (> 10 lifetime)• Comorbid anxiety disorders• Comorbid substance abuse

ANTICONVULSANTSExperts usually rank lithium as the treatment of choice fora patient with classic mania, but divalproex is an acceptablefirst-line alternative. It may be used singly in patients whocannot tolerate lithium. For patients who do not respond tolithium, there are no secure data on whether divalproex shouldbe added as an adjunct or substituted, but many psychiatristswould choose the former in a patient who appears to respondat least partially to lithium and the latter in patients for whomlithium seems to afford no benefit. Carbamazepine has alsobeen used in bipolar patients non-responsive to or intolerant oflithium. This use is based upon studies using active compara-tors (including lithium) and demonstrating comparable overallefficacy. Despite being used for decades in this way, it isonly recently that FDA approval was sought and obtainedfor an acute antimanic indication for a slow release formof carbamazepine. Older long prospective studies comparingcarbamazepine to lithium suggested a superiority of the anti-convulsant in patients with mixed or dysphoric mania, rapidcycling, or significant comordity. As such, it appears to share asimilar profile to divalproex. However, it has a rather differentside effect profile, having less risk of weight gain, tremor, orhair loss, but greater risk of neurological side effects, such asataxia, vertigo, and diplopia. Two other liabilities associatedwith carbamazepine include an increased risk of agranulocy-tosis and a robust induction of hepatic enzymes. The latteroften results in more frequent blood level monitor and the needto compensate for dropping carbamazepine levels. In addition,the dosing of other medications must often be adjusted afteraddition of carbamazepine.


ATYPICAL ANTIPSYCHOTICSWith the exception of clozapine, all other atypical antipsy-chotic drugs have received approval by the FDA for useas monotherapy treatments for acute mania. Clozapine hasbeen shown to be effective in patients with treatment-resistantsymptoms but other atypicals with fewer serious adverseeffects are preferred for routine use. Placebo-controlledmonotherapy trials have demonstrated efficacy for olanza-pine at 10 or 15 mg/day, risperidone at 3–6 mg/day, quetiapineat 600 mg/day, ziprasidone at mean doses of either 112 or130 mg/day, and aripiprazole at 10–15 mg/day. The presence orabsence of psychosis has not been shown to influence antipsy-chotic efficacy in mania.

Relative Efficacy of Different AgentsThere are relatively few head to head comparisons of antimanicagents and the results can be summarized as follows:

• Olanzapine and valproate have been compared in twodouble blind studies, which yielded equivalent efficacy inone study and a modest superiority of olanzapine in theother. Tolerability favored valproate in both studies.

• Valproate and carbamazepine may have greater efficacythan lithium in patients with mixed mania, rapid cycling, orhigh levels of comorbidity.

• Quetiapine has performed comparably to lithium incontrolled trials.

• Risperidone, quetiapine, and aripiprazole have all performedsimilarly to haloperidol in active comparator trials.

Acute mania is often a clinic emergency and is generallytreated in an inpatient setting. Rapid mood stabilization andthe reduction of agitation take priority over concerns aboutmedication side effects. Polypharmacy is the rule in this setting,with atypical antipsychotic agents frequently added to lithiumor an anticonvulsant. In support of this approach, recent studieshave demonstrated improved antimanic response with theaddition to lithium or valproate of olanzapine, risperidone, orquetiapine. Generally, these combinations are quite safe, but

MOOD STABILIZERS 93

caution should be used if combining lithium with older antipsy-chotics as it may increase the risk of extrapyramidal effectsand possibly, of neurotoxicity. In addition, the combination ofthioridazine and lithium is contraindicated due to an increasedrisk of ventricular arrhythmias.

The American Psychiatric Association guidelines for bipolardisorder have recommend lithium or valproate plus an atyp-ical antipsychotic as the first-line therapy for acute mania.Less severely ill patients may be treated with a single anti-manic agent (lithium, valproate, or an atypical antipsychotic;this recommendation was published prior to the FDA approvalof extended release carbamazepine). The Expert ConsensusGuideline Series published in 2004 provides similar first-linerecommendations but also lists lithium alone as a first-line treat-ment for euphoric mania or for non-rapid cycling hypomania.Valproate is preferred to lithium if either psychosis or rapidcycling is present. Short-term adjunctive treatment with a benzo-diazepine is endorsed as an augmentation for agitation andanxiety, and for providing needed sedation. If an antidepressantappears to be linked to switching or cycle acceleration in a manicpatient, it should be tapered and discontinued. Bear in mind,however, that manic episodes are often followed by a period ofsevere depression, and also that abrupt antidepressant discontin-uation may increase the risk of a sudden manic shift. The optionsfor treatment of acute mania are listed in Table 3-3.

ANXIOLYTICSAmong current anxiolytic agents, benzodiazepines are usuallyselected as adjuncts to treat acute mania because of theirsafety and efficacy. Although some have claimed specificantimanic efficacy for clonazepam or other benzodiazepines,most psychiatrists are more impressed with their benefitsas adjunctive sedatives than with more specific antimanicactivity. Lorazepam is often the benzodiazepine selected forthis indication because it yields predictable blood levels whenadministered intramuscularly. Benzodiazepines have a widemargin of safety and can be safely administered in even veryhigh doses, suppressing potentially dangerous excitement andallowing patients much needed sleep. When used togetherwith an antipsychotic agent, benzodiazepines counteract the


� TABLE 3-3. Treatments for Acute Mania

TREATMENT ADVANTAGES DISADVANTAGES

Lithium Efficacy: 70–80% Side effects: sedation,weight gain, tremorhair loss, polyuriaLow therapeuticindex

Valproate Comparable efficacy to LiMay be better in mixedstates

Side effects: sedation,weight gain, tremor,hair loss

Carbamazepine Comparable efficacy to LiMay be better in mixedstates

Side effects: neurologic(ataxia, vertigo,diplopia)

AtypicalAntipsychotics

Comparable efficacy to Li Side effects, differ byindividual agent. Riskof tardive dyskinesia

TypicalAntipsychotics

Comparable efficacy to LiRapid onset of action

May worsen bipolardepression. Greaterrisk of tardivedyskinesia

ElectroconvulsiveTherapy

Efficacy ∼80%� Safe forpatients unable to takemedication

More difficult toadminister. Long-termeffectiveness unclear

Anxiolytics Good as adjunctivesedatives, wide marginof safety

Probably not specificallyantimanic. Potentialabuse and dependence

antipsychotic agent’s tendency to provoke extrapyramidalreactions and seizures. For lorazepam, 1 to 2 mg can be admin-istered by mouth or intramuscularly as frequently as hourly.

Treatment Initiation and Dose TitrationIt is important to carry out a range of tests before embarkingon treatment to identify patients at increased risk of adverseeffects (Table 3-4).

LITHIUMPretreatment testing for lithium includes a complete bloodcount, electrolyte determinations, renal panel, and thyroidfunction tests. An electrocardiogram should also be obtained.

MOOD STABILIZERS 95

� TABLE 3-4. Pretreatment Tests

MEDICATION TESTING

Any Medication • Comprehensive medical history• Physical examination including vital signs and

weight• Pregnancy test, if applicable

Lithium • Complete blood count• Electrolytes• Renal panel (blood urea nitrogen, creatinine, and

routine urinalysis)• Thyroid panel plus thyroid-stimulating hormone• Electrocardiogram

Valproate • Complete blood count• Liver function tests

Carbamazepine • Complete blood count• Liver function tests• Renal panel• Urinalysis

AtypicalAntipsychotics

• Blood glucose• Blood lipids

Lithium dosage may be based on a plasma concentrationsampled 12 hours after the last dose, or the drug may begradually titrated to a dose that is tolerated and within therange usually considered “therapeutic.” As with any drug,approximately five half-lives must elapse for steady state tobe achieved; for an average adult, this takes about five daysfor lithium (longer in the elderly or in patients with impairedrenal function). To treat acute mania, plasma concentrationsshould typically be greater than 0.8 mEq/L but to avoid toxi-city the level should not exceed 1.5 mEq/L. It is important toknow what other medications a patient may be taking becausemany drugs interact with lithium and can lead to increasedor decreased lithium levels and possibly to adverse effects(see ‘Drug Interactions’ on page 119).

To reach therapeutic levels rapidly in healthy youngerpatients with normal renal and cardiac function, the psychia-trist may prescribe 300 mg of lithium carbonate four times dailyfrom the outset, sampling the first plasma level after five days(or sooner should toxic signs become apparent). Thereafter,


the dose should be adjusted to achieve a 12-hour plasmaconcentration between 0.8 and 1.3 mEq/L at steady state. Bycontrast, in a patient with mild hypomanic symptoms it may bewiser to begin with a lower lithium dose, such as 300 mg b.i.d.,taking longer to achieve therapeutic levels but minimizing sideeffects that could trouble the patient and promote medica-tion non-adherence. Once steady state has been achieved attherapeutic concentrations and the patient is clinically stable,lithium can be administered to most patients in a once-dailydose, usually at bedtime. Not only is this schedule easier toremember, but it tends to decrease such common side effectsas tremor and polyuria.

ANTICONVULSANTSBefore initiating treatment with valproate or divalproex, thepsychiatrist should obtain a comprehensive medical history andensure that a physical examination has been performed, withparticular attention to suggestions of liver disease or bleedingabnormalities (see Table 3-4). Baseline liver and hematologicalfunctions are measured before treatment, every one to fourweeks for the first six months, and then once every three tosix months. Mild elevations in transaminase levels are notuncommon and should be monitored, but tend to be mild andself-limited, not requiring discontinuation of the medication.Evidence of abnormalities in hemostasis or coagulation, suchas hemorrhage or increased bruising, should prompt a reduc-tion of dosage or withdrawal of therapy. The drug should bediscontinued immediately in the presence of significant hepaticdysfunction.

In healthy adults, divalproex is started at 750 mg/day individed doses and then adjusted to achieve a 12-hour serumvalproate concentration between 50 and 125 �g/mL. The timeof dosing is determined by possible side effects (particularlysedation), and once-a-day dosing can be employed if toler-ated. The use of slow-release formulations reduces side effects(particularly gastrointestinal) and facilitates the use of oncedaily dosing. As with lithium, the antimanic response tovalproate typically occurs after one to two weeks. It may bepossible to obtain a faster response with oral loading of dival-proex at 20 mg/kg/day for the first five days. This produces

MOOD STABILIZERS 97

serum concentrations greater than 50 mg/L after two days oftreatment and is generally well tolerated with few side effects.

Carbamazepine is typically started at 200 to 400 mg/dayin three or four divided doses, and increased to 800 to1000 mg/day by the end of the first week. As with divalproex,slow release formulations may improve tolerability and facili-tate use of once daily dosing. If clinical improvement is insuf-ficient by the end of the second week, and the patient has nothad intolerable side effects to the drug, carbamazepine maybe increased to as high as 1600 mg/day. As there is little datarelating carbamazepine blood levels and psychiatric efficacy,the anti-epileptic blood level range (4 to 15 ng/mL) is gener-ally utilized. If carbamazepine is combined with lithium orantipsychotics, lower doses and blood levels of carbamazepinemay suffice. If divalproex and carbamazepine are adminis-tered simultaneously, blood levels of each should be monitoredcarefully because of the complex pharmacokinetic interactionsbetween the two agents.

ANTIPSYCHOTICSAlthough in the 1970s it was commonplace to use ultra-high doses of antipsychotics in attempts to suppress psychoticsymptoms (sometimes in excess of 100 mg of haloperidoldaily), this is seldom necessary and may be counterproduc-tive: it can increase the risk of acute cardiovascular compli-cations, seizures, acute dystonia, and possibly neurolepticmalignant syndrome. However, in emergent situations, intra-muscular (IM) administration of an antipsychotic can helpreduce agitation more rapidly than lithium or an anticon-vulsant. IM formulations exist for several first-generationantipsychotics and but only for one atypical antipsychotics,ziprasidone.

Dosing recommendations for atypical antipsychotics arelisted in Table 3-5.

Recently, significant concern has been raised about therisks of metabolic complications associated with the useof atypical antipsychotics. Weight gain, hyperlipidemia, anddiabetes (including abrupt and often catastrophic new onsetdiabetes) have been observed to occur at relatively high rates.Although the FDA has mandated a black box warning for


� TABLE 3-5. Atypicals in the Treatment of Acute Mania

GENERICNAME

BRANDNAME

DOSE ONFIRST DAY

ACUTE DOSEADJUSTMENT

MAINTENANCEDOSE

Aripiprazole Ability 15−30 mg q.i.d. − 15−30 mg q.i.d.Olanzapine Zyprexa 10−15 q.i.d. increments of

5 mg per day5−20 mg q.i.d.

Quetiapine Seroquel 50 mg b.i.d. 200 mg b.i.d. byday 4 inincrements of50 mg b.i.d.

200−400 mgb.i.d.

Risperidone Risperdal 2−3 mg q.i.d. increments of1 mg per day

1−6 mg q.i.d.

ZiprasidoneGeodon

(po)40 mg b.i.d. 60−80 mg b.i.d.

on day 240−80 mg b.i.d.

Geodon(im)

10 mg q2hr

the entire class of atypical antipsychotics, the risk does appearto be greatest with clozapine and olanzapine. The ConsensusDevelopment Conference on Antipsychotic Drugs and Obesityand Diabetes has made recommendations for screening priorto treatment with an atypical antipsychotic and for monitoringduring treatment. Before starting the medication, weigh poten-tial benefits against metabolic and medical risks based uponthe following assessments:

• Personal and family history of diabetes, obesity, hyperlipi-demia, hypertension, and cardiovascular disease

• Baseline height and weight (which should be used to calcu-late the body mass index, BMI) and waist circumference(measured at the height of the umbilicus)

• Blood pressure• Fasting plasma glucose• Fasting lipid profile

The presence of metabolic abnormalities or significantpersonal or family history might prompt preferential use ofatypical antipsychotics with lower risk (ziprasidone or aripipra-zole) or intermediate risk (risperidone or quetiapine) relativeto higher risk (olanzapine and clozapine).

MOOD STABILIZERS 99

Follow up monitoring of patients on atypical antipsychoticsshould be conducted as follows:

• Review personal and family history annually.• Measure weight and BMI at 4, 8, and 12 weeks, then

quarterly.• Measure waist circumference annually.• Measure blood pressure and fasting plasma glucose at 12

weeks, then annually.• Measure fasting lipids at 12 weeks, then every five years.

Acute DepressionLong-term prospective studies have demonstrated that depres-sion is much more frequent and protracted than mood eleva-tion in bipolar disorder, particularly for bipolar II disorder,and accounts for a much greater proportion of time spentill. The depressed phase of bipolar illness is also the phasemost associated with suicide attempts and residual depres-sive symptoms are most predictive of functional impairmentsin partially remitted bipolar patients. Unfortunately, depres-sion occurring in bipolar disorder is also the syndrome of theillness for which we have the least effective and safe treat-ments. Traditional antidepressants often promote increasedmood instability, cycle acceleration, and mood switching fromdepression to mania or hypomania. Tricyclic antidepressantsare the most likely to trigger instability, but no antidepressanthas been shown to be completely safe in this regard. In light ofthis fact, as well as the robust antidepressant effects of somenewer mood stabilizers, traditional antidepressants should notbe the first-line treatment for bipolar depression. Treatmentsfor acute depression in bipolar disorder are shown in Table 3-6.

Drug SelectionCurrent guidelines by the American Psychiatric Associationlist lithium or lamotrigine as first-line monotherapy for acutebipolar depression. Antidepressant monotherapy is not recom-mended except in combination with lithium in severely illpatients. The recommendation for lamotrigine use as a first-linetreatment derives from two sources. The first is a large double


� TABLE 3-6. Treatments for Acute Depression in Bipolar Disorder


Mood stabilizers Provide maintenanceand antimanic efficacy

Antidepressantefficacy relativelymodest for mostmood stabilizers

If use of antidepressantsbecomes necessary,should reduce the riskof switching or cycleacceleration

Side effects may beproblematic:sedation, weightgain/metabolic,tremor, etc.

Lamotrigine Significantly moreefficacious than mostother moodstabilizers.

Antimanic efficacy isrelatively modest, somay not providecomprehensivetreatment asmonotherapy forbipolar I patients

No evidence ofinduction of switchingor rapid cycling

Generally well toleratedQuetiapine Recent data suggest

comparable efficacyto lamotrigine

Side effects may beproblematic:sedation, weightgain/metabolic

May be efficacious fordepression at doseslower than thoserequire for mania,improving tolerability

Risk of tardivedyskinesia withlong-termtreatment

At higher doses may beable to providebalanced and strongantidepressant andantimanic efficacy

Antidepressants Probably efficacious inthe acute setting

Can trigger switchinto mania andincrease cycling

ECT Acutely efficacious Cognitive side effectsSafe for patients unable

to take medication(e.g., pregnancy)

Need for anesthesiaMaintenance after acute

treatment problematic

MOOD STABILIZERS 101

blind, placebo controlled trial of lamotrigine monotherapy(50 mg vs. 200 mg final dose, after appropriate titration).The response rate with the 200 mg final dose was 56%, whichwas significantly greater than that of placebo using several clin-ical measures. Though the study was not powered to fully testthis, the results of the study suggested that the 200 mg finaldose was more effective than 50 mg final dose. The secondline of evidence supporting lamotrigine for bipolar depressioncomes from a larger prospective maintenance study of remittedbipolar I and II patients, comparing lamotrigine monotherapyat 200 mg with lithium monotherapy and with placebo. Inthis study, lamotrigine was superior to placebo for prophy-laxis of depressive or elevated relapses, but more robustlyso for depression. In contrast, lithium was more effectivein prevented elevation. Furthermore, unlike with traditionalantidepressants, there is no clear evidence of lamotrigineprovoking cycling or switching.

Recently, quetiapine has emerged as a second moodstabilizer with potentially robust antidepressant efficacy.A large multicenter placebo controlled study (There are 2studies: BOLDER studies) compared quetiapine monotherapyat two doses (300 mg and 600 mg) in bipolar I and II patientsin an acute major depression. Response and remission ratesfor both active groups were identical (58% and 36%, respec-tively) and significantly better than for the placebo group.The effect size for quetiapine in this study was comparableto that of lamotrigine for bipolar depression and significantlygreater than that of olanzapine when used as a monotherapyfor bipolar depression.

Apart from lamotrigine and quetiapine, there are few dataavailable to stratify the other mood stabilizers in terms ofantidepressant efficacy. Lithium, carbamazepine, and olanza-pine appear to be effective for bipolar depression at a muchlower rate than lamotrigine or quetiapine, and risperidone,ziprasidone, and aripiprazole have insufficient data to evaluate.Nevertheless, there is reason to consider lithium seriously inpatients with persistent, frequent, or severe bipolar depression.Lithium is the only pharmacologic intervention that exhibitsa specific anti-suicide effect, to some degree independent ofits mood efficacy within patients. Lithium treatment reduces


the suicide rate in bipolar patients approximately seven-fold.However, in the same samples, discontinuation of lithiumresults in a rapid return to previous (or even, transiently,to higher) suicide rates. Moreover, when tapered quickly orabruptly discontinued, a twenty-fold increase in suicides andattempts occurs in the first year. Comparable anti-suicideefficacy in bipolar patients has not been demonstrated for anti-convulsants or antipsychotics, though confirmation of lithium’suniqueness will require more studies of other agents. In addi-tion, both lithium and anticonvulsants can have a positiveimpact on impulsivity and aggression, two important suiciderisk factors.

Probably all available antidepressant drugs can be effec-tive in treating bipolar depression but we lack sufficientdata to stratify them in terms of efficacy. There is someevidence that MAO inhibitors may be somewhat more effec-tive than other classes but they may also be more likely thanother classes of antidepressants (except for TCAs) to provokemood instability. Of course, antidpressants should never beprescribed for bipolar patients in the absence of adequatemood stabilizers, which reduce the risk of switching or cycleacceleration. If mania occurs, the antidepressant medicationshould be discontinued immediately. It is not clear how longantidepressants should be continued following resolution ofan acute major depression. The Expert Consensus Guidelinesrecommended continuing antidepressants for 18–30 weeks innon-rapid cycling bipolar I patients following resolution ofdepression and for 9–17 weeks in rapid cycling patients.

Treatment Initiation and Dose TitrationDosing guideline for lithium should parallel those for hypo-manic patients, trading slower titration and the risk of delayedresponse to minimize side effects and risk of toxicity. However,the target blood level should still be set at 0.8 or greater. Dosingguidelines for lamotrigine are particularly important to follow,as a more accelerated rate of titration significantly increases therisk of Stevens-Johnson Syndrome (SJS), an immune-mediatedsystemic reaction which often present with a severe rash and


� TABLE 3-7. Lamotrigine Dosing

NO WITHANTICONVULSANTS VALPROATE WITH CBZ

Weeks 1 & 2 25 mg/d 12.5 mg/d 50 mg/dWeeks 3 & 4 50 mg/d 25 mg/d 50 mg b.i.d.Week 5+ Add 50–100 mg every

1–2 weeksAdd 25–50 mg

every 1–2weeks

Add 100 mg/devery 1–2weeks

Maintenance 100–400 mg 100–200 mg/d 250 mg b.i.d.

has a 10–15% mortality rate (see ‘Adverse Effects’ on page111). The dosing of lamotrigine is also complicated by phar-macokinetic interactions (Table 3-7). Specifically, clearanceof lamotrigine is reduced by valproate, causing an approxi-mate doubling of blood levels for a given dose and requiring,therefore, that doses at each point in the titration behalved compared to the lamotrigine monotherapy titration.Conversely, lamotrigine levels are cut roughly in half by P450enzyme inducing medications such as carbamazepine. Thispermits an increase in the dose at any point in the titrationcompared to monotherapy dosing.

Quetiapine dosing for depression can proceed more slowlythan for mania. Since 300 mg was identical in efficacy to600 mg in the primary study (two Studies – BOLDER I andBOLDER II) informing the use of quetiapine for depression,it is prudent to have a lower target dose and provide suffi-cient time during the titration to allow for responses at lowerdoses. A slower titration and lower target might also reduceside effects and the potential for metabolic complications.

Breakthrough EpisodesThe first step in managing a breakthrough episode of eithermania or depression is to check medication adherence,evaluate other potential precipitants of relapse (e.g., seasonalchanges, which are common in bipolar disorder, onset of peri-menopause, severe stressors, recurrent substance abuse, etc.)and evaluate medication doses and blood levels. Patientson lithium should have thyroid function tests checked as


well, since hypothyroidism secondary to lithium treatment candestabilize mood. Some data also suggest that levels of freeT4 in the euthyroid but low normal range prolong episodesof bipolar depression and reduce the likelihood of successfultreatment.

In parallel with addressing any clear precipitants, the medi-cation regimen must be thoughtfully and deliberately adjustedto both provide treatment for the acute episode and alsoto improve the level of prophylactic efficacy. Primary moodstabilizer doses should be increased as tolerated, particularlyfor relapse into elevation, and use of additional medicationsconsidered (e.g., adding lamotrigine for relapse into depres-sion). If a pattern of increased mood instability and increasedcycling becomes apparent, then serious consideration shouldbe given to discontinuing any antidepressant in the regimen,even if the most recent relapse was into a depressed state.Often the best treatment for rapid cycling is not the additionof any medication but the discontinuation of an antidepres-sant. Consistent use by the patient of a daily prospective moodrating instrument, such as the Life Chart, can be very helpfulfor discerning an increase in instability.

MaintenanceNowhere is the potential gap between the results of clinicaltrials and clinical experience more apparent than in mainte-nance treatment. For example, valproate is one of the mostcommonly used maintenance treatments for bipolar disorderand most psychiatrists agree that it is clearly quite effectivein this role. The Expert Consensus Guidelines also endorsevalproate as a first-line maintenance treatment. This is truedespite the absence of a large scale study clearly demonstratingthis robust effect. Fortunately recent years have witnessed asignificant increase in maintenance studies and some of thesedata can now inform treatment selection.

The risk of relapse is particularly high in the six monthsfollowing an acute episode. Psychosocial interventions shouldbe offered in addition to drug treatments, which are summa-rized in Table 3-8.


� TABLE 3-8. Advantages and Disadvantages of Specific Mainte-nance Treatments.


Lithium Most robust dataset andclearly very effectivefor prevention ofmania

Side effects and lowtherapeutic index

Specific anti-suicide effect Lethality in overdoseRisk of renal damage

Valproate Possibly comparable tolithium

Weight gain and otherside effects

May be more effective forpatients with rapidcycling, mixed episodesand comorbidity

Teratogenicity, forfemale patients inreproductive years

Olanzapine Possibly comparable tolithium for preventionof mania

Weight gain andmetabolic risks

Tardive dyskinesia riskLamotrigine Probably more effective

than other moodstabilizers fordepression

Probably somewhatless effective forprevention of mania

Generally bettertolerated than othermood stabilizers

Carbamazepine Profile comparable tovalproate

More maintenancedata needed

Less risk of weight gainTypical

antipsychoticsLikely good for

prophylaxis of maniaMay worsen

depressionHigh risk of tardive

dyskinesiaAntidepressants May prevent recurrence

of depressionMay precipitate mania

and provoke rapidcycling

ECT Acutely effective for bothmania and depression

Not adequatelystudied

Maintenance of acuteresponse may bepoor

Psychotherapy As adjunct, increasesmedication compliance,overall functioning

Not efficacious alone


LITHIUMFor bipolar maintenance, lithium has the longest and largestdatabase of clinical experience and by far the greatest numberof patients studied in rigorous clinical trials. In randomizedtrials, the relapse rate in patients treated with lithium is about37% compared with 79% with placebo. Predictors of poorresponse to lithium are listed in Table 3-9.

Most patients starting lithium maintenance are alreadytaking lithium after an acute episode. A medical history shouldalso have been obtained that included questions about: pastmedical and family history of renal, thyroid, cardiac, andcentral nervous system disorders; other drugs a patient maybe taking (including prescription, over-the-counter, and illicit);the use of such common substances as caffeine, nicotine, andalcohol; and special diets or diet supplements. They shouldalso have had baseline medical tests, including assessmentof thyroid function (thyroid panel plus thyroid-stimulatinghormone) and renal function (blood urea nitrogen, creatinine,and routine urinalysis), a complete blood count, electrolytedeterminations, an electrocardiogram, and a physical exami-nation (see Table 3-4).

If a patient who is to start lithium maintenance therapy isnot already taking lithium, a slower titration than that usedfor acute mania can be utilized. A physically healthy, average-size adult may be started with 300 mg of lithium carbonatetwice daily, or 600 mg at night if a slow release formulation isemployed. An elderly, ill, or slightly built individual can beginwith as little as 300 mg per day. Because it takes about five days

� TABLE 3-9. Predictors of Poor Responseto Lithium Prophylaxis

Rapid or continuous cyclingMixed states or dysphoric maniaAlcohol or drug abuseNon-compliance with treatmentCycle pattern of depression–mania–euthymiaPersonality disturbanceHistory of poor interepisode functioningPoor social support systemThree or more prior episodes


to achieve a steady state (longer in the elderly and those withrenal impairment), a 12 hour trough lithium level should bedrawn at approximately that interval. Dose adjustments canbe made at intervals of five or more days, with repeat levels asneeded to obtain a therapeutic level. Levels between 0.8 and1.0 mEq/L afford threefold greater protection against recur-rent episodes than a range of 0.4 to 0.6 mEq/L. Furthermore,patients in the higher range are less likely to experience subsyn-dromal symptoms (hypomania or minor depression) and, ifsuch symptoms do appear, are less likely to go on to a fullepisode. However, higher blood levels are associated withmore side effects and the risk of non-compliance. Sometimeseducation and reassurance are sufficient to keep patients athigher levels. In addition, remedies are available to treat someof the more aggravating side effects (e.g., beta blockers fortremor). Side effects may diminish with a decrease in lithiumlevel, but both psychiatrist and patient should be aware thatthis may decrease the level of protection against mood swings.

There have been no systematic studies of the lithium level–clinical response relationship in elderly patients, but becauseolder people are more sensitive to side effects, it may beprudent to attempt to maintain elderly bipolar patients at lowerplasma lithium concentrations. It should also be kept in mindthat a given dose will frequently produce higher blood levelsas a person ages (likely due to gradual reduction in renal func-tion), so dose reduction may be necessary over time.

Monitoring of patients on maintenance lithium treatmentfocuses on three elements: blood level, renal function, andthyroid function. The following is a reasonable regimen formonitoring over time:

• Lithium level every 3 months.• Electrolytes, blood urea nitrogen, and creatinine every six

months, or sooner if the lithium level rises acutely andwithout any other explanation.

• TSH, and free and total T4 every six months, or soonerif the patient’s mood becomes more unstable or symptomssuggestive of hypothyroidism occur.

• An annual electrocardiogram for men over 40 and womenover 50 years of age.


If lithium needs to be discontinued for any reason, it ispreferable to taper the medication over six or more weeks.Rapid discontinuation increases the risk of relapse and mayfurther increase the risk of suicidal actions.

It can be very difficult to determine whether a symptom,such as a minor drop in mood or several days of reducedsleep, represents a transient fluctuation or is a harbinger of amajor affective episode. Daily mood charting and close contactwith patient can help in this determination and also providethe opportunity to rapidly respond to an incipient relapse.Mild hypomanic symptoms may be more likely to be predic-tive of a manic relapse than depressive symptoms predictiveof a recurrence of major depression. As described above, opti-mizing current medications and adding adjunctive treatments(e.g., benzodiazepine in a patient who feels slightly agitatedand is not sleeping well) may be sufficient to prevent a moredramatic worsening in mood.

ANTICONVULSANTSValproate is an important alternative to lithium as a main-tenance treatment for bipolar disorder. It may be prefer-able for patients with rapid cycling, a history of dysphoricor mixed mania, comorbid substance abuse or anxiety disor-ders, or organic brain disease. Carbamazepine has a some-what greater medical risk, in terms of bone marrow suppres-sion, but may be better tolerated than valproate for somepatients. In clinical trials, approximately 60% of patients havea moderate or marked response to carbamazepine comparedwith 22% with placebo. In comparisons with lithium, carba-mazepine was associated with equivalent or greater rates ofimprovement. Although no head to head comparisons existfor valproate and carbamazepine in maintenance treatment,other clinical data and experience suggest that they are compa-rably effective and work well in an overlapping subset ofpatients.

When valproate is used for maintenance therapy, it is bestto start with a low dose, such as 250 to 500 mg daily, buildingthe dose gradually to maintain a plasma level between 50 and125 �g/mL. Treatment with carbamazepine should be initi-ated at a low dose, 100 mg/day, increasing in increments of


100 mg/day every four to five days. Dosing is usually two to fourtimes daily for the immediate release formulations. Althougha correlation between blood level and clinical response has yetto be established, most psychiatrists are guided by the thera-peutic range in patients with epilepsy: usually 4 to 15 �g/mL.Most patients taking carbamazepine for bipolar disorder aremaintained with doses between 400 and 1800 mg/day. Becauseof enzyme induction, it is often necessary to increase the doseafter two to three weeks of treatment to maintain the sameblood level.

As described above, in the section on treatment of bipolardepression, two large maintenance trials evaluated lamot-rigine relative to lithium and placebo in remitted bipolar Iand II patients. Lamotrigine was more robust in preventingrelapse into depression and lithium was more robust inpreventing relapse into mania, but both agents has some, albeitlower, efficacy in preventing the other phase (i.e., lithium fordepression and lamotrigine for mania). As a broad mainte-nance regimen, it is certainly worth considering combininga predominantly antidepressant mood stabilizer, such aslamotrigine with a predominantly antimanic one, such aslithium.

ANTIPSYCHOTICSMore data are gradually emerging to support the efficacy ofatypical antipsychotics as maintenance treatments for bipolardisorder. Olanzapine has the largest dataset to date andappears to have a profile similar to lithium, valproate, andcarbamazepine, being more effective for prevention of maniathan for prevention of depression. It is too early to judge theother atypical antipsychotics in terms of relative efficacy forprevention of mania and depression.

One major drawback to the use of atypical antipsychoticsfor maintenance is the cumulative risk for tardive dyskinesia.Unfortunately, at this time there are insufficient data to clearlyweigh that risk or to stratify specific agents. However, assumingcomparable efficacy, one might argue that lithium and the anti-convulsants should be considered ahead of atypical antipsy-chotics as maintenance treatments.


Rapid CyclingPatients with rapid-cycling bipolar disorder—defined as fouror more affective episodes in one year, with or without anintervening period of euthymia—tend to be less responsiveto lithium treatment. Indeed, the initial description of rapidcycling arose from studies of lithium failure during mainte-nance treatment. Whether rapid cycling is a natural progressionof the illness or a separate disorder has yet to be deter-mined, but there are certain risk factors for rapid cycling.Those include gender (women comprise 70% of rapid cyclingpatients), use of antidepressants, and current or past thyroiddisease.

Apart from the reduced effectiveness of lithium in rapidcycling, there are relatively few data to guide medicationselection for the treatment of rapid cycling. More data supportthe use of anticonvulsants in this population, but atypicalantipychotics may also be superior to lithium in this setting.One important intervention to consider is discontinuation ofany antidepressant medications, even in the setting of a depres-sive relapse. As with any breakthrough episode, rapid cyclingshould be approached by evaluating mood stabilizer adher-ence, dosage, and blood levels. Thyroid function should bechecked and any hypothyroidism addressed.

CostsThe costs of the most widely used drugs as of December 2005are summarized in Table 3-10.

� TABLE 3-10. Current medication costs. THIS IS HIGHLY VARIABLEAND EVER-CHANGING

MEDICATION FORMULATION QUANTITY PRICE ($)

Depakene 250 mg caps 100 207�01Lithium Carbonate 150 mg caps 100 15�54

300 mg caps 100 17�77600 mg caps 100 42�30

Lithium CR∗ 450 mg tabs 180 81�97Valproic Acid 250 mg caps 100 29�97


Divalproex sodium 125 mg EC tabs 180 110�29250 mg EC tabs 180 204�61500 mg EC tabs 180 359�96250 mg 24 hr tabs 90 112�84500 mg 24 hr tabs 90 186�28

Carbamazepine 200 mg tabs 180 31�99Tegretol XR 200 mg 12 hr tabs 180 118�24

400 mg 12 hr tabs 180 149�89Trileptal 150 mg tabs 180 206�96

300 mg tabs 180 343�96600 mg tabs 180 683�48

Zyprexa 2.5 mg tab 100 522�195 mg tab 100 616�667.5 mg tab 100 746�6610 mg tab 100 955�5415 mg tab 100 1244�4320 mg tab 100 1717�7425 mg tab 100 174�28

Quetiapine 100 mg tab 100 295�86200 mg tab 100 556�08300 mg tab 100 744�58

Risperidone 0.25 mg tab 100 284�420.5 mg tab 100 312�741 mg tab 100 315�522 mg tab 100 534�543 mg tab 100 555�524 mg tab 100 824�99

Ziprasidone 20 mg cap 100 433�3240 mg cap 100 440�3160 mg cap 100 479�5080 mg cap 100 498�29

Ability 5 mg tab 100 974�3410 mg tab 100 1011�1515 mg tab 100 971�8120 mg tab 100 1345�0730 mg tab 100 1345�07

∗ CR = controlled release.

ADVERSE EFFECTSLITHIUMOne major limitation of lithium is its narrow therapeutic index;plasma concentrations only slightly higher than the therapeuticrange can have severe side effects. Levels of 2.0 mEq/L and


above, particularly if prolonged, can cause central nervoussystem impairment, renal shutdown, coma, permanent braininjury, and death. Management of severe lithium intoxication(Table 3-11) may require hemodialysis and medical intensivecare to maintain fluid and electrolyte balance, prevent furtherabsorption of the drug, and maximize the rate of elimination.

To avoid this, patients and their families need to be instructedcarefully on a number of matters related to lithium levels. Theyshould be alerted to the early signs of intoxication—such asincreased tremor, confusion, and ataxia—and directed to stop

� TABLE 3-11. Management of Serious Lithium Toxicity

1. Rapidly assess (including clinical signs and symptoms, serum lithiumlevels, electrolytes, and electrocardiogram), monitor vital signs, andmake an accurate diagnosis.

2. Discontinue lithium therapy.3. Support vital functions and monitor cardiac status.4. Limit absorption.

a. If patient is alert, provide an emetic.b. If patient is obtunded, intubate and suction nasogastrically

(prolonged suction may be helpful because lithium levels in gastricfluid may remain high for days).

5. Prevent infection in comatose patients by body rotation andpulmonary toilet.

6. When lithium has reached nontoxic levels, vigorously hydrate (ideally5 to 6 L/d); monitor and balance the electrolytes.

7. In moderately severe cases:

a. Implement osmotic diuresis with urea, 20 g given intravenouslytwo to five times per day, or mannitol, 50 to 100 g/d, givenintravenously.

b. Increase lithium clearance with aminophylline, 0.5 g up to every 6hours, and alkalinize the urine with intravenously administeredsodium lactate.

c. Ensure adequate intake of sodium chloride to promote excretion oflithium.

8. Implement hemodialysis in the most severe cases. These arecharacterized by:

a. Serum lithium levels between 2 and 4 mEq/L with severe clinicalsigns and symptoms (particularly decreasing urinary output anddeepening CNS depression).

b. Serum lithium levels greater than 4 mEq/L.


taking further lithium and contact their psychiatrist promptly. Alithium level should be obtained as soon as possible that sameday and the symptoms followed closely for the next several days.The psychiatrist should also be contacted if the patient developsnew gastrointestinal symptoms, especially vomiting or diarrhea.These symptoms may reflect an elevated lithium level or may,even if caused by an unrelated factor (such as gastrointestinalinfection), trigger an elevation in the lithium level as a conse-quence of fluid and electrolyte changes. Other causes of fluidloss with insufficient replacement, such as excessive sweatingdue to heat exposure or fever, should also prompt a tempo-rary discontinuation of lithium treatment until fluid balanceis restored. Concomitant use of other medications may alsoincrease levels of lithium and lead to lithium intoxication (See‘Drug Interactions’).

The most common adverse effects from lithium are(Table 3-12):

• nausea• vomiting• diarrhea• postural tremor• polydipsia• polyuria

If troublesome, these can usually be mitigated by altering thelithium timing or formulation or by adding a remedy. Specificinterventions include the following:

• Beta blocking drugs generally dramatically reduce orcompletely eliminate lithium tremor. Propranolol is oftenused and must be dosed b.i.d. or t.i.d. to cover the entire day(unless a long acting form of propranolol is used). Treatmentis initiated at 10 mg b.i.d. to t.i.d. and increased as toleratedand needed. Doses above 160 mg are not usually required.

• Diuretics for polydipsia or polyuria. Diuretics may paradox-ically decrease urine volume, but because some (such asthiazides) can also raise the serum lithium concentration,lithium level must be carefully monitored and the lithium doseadjusted downward appropriately. Potassium levels shouldalso be monitored and supplementation may be necessary.


� TABLE 3-12. Adverse Effects: Lithium, Valproate, andCarbamazepine

LITHIUMCommon Rare RareDermatitis Ataxia HyperparathyroidismFatigueGastrointestinal

upsetHeadacheHypothyroidismMemory and

concentrationdifficulties

Muscle weaknessPolydipsiaPolyuriaTremorWeight gain

Autonomic slowing ofbladder and bowelfunction

Cardiovascularcomplications

Diabetes mellitusDysarthriaEdemaElevated WBC

countExtrapyramidal

reactionsGoiterHypercalcemia

HyperthyroidismMetallic tasteNystagmusOrganic brain

syndromeParathyroid hyperplasia

and parathyroidadenomas

Tearing, itching,burning, or blurringof the eyes

TinnitusVertigoVisual distortion

VALPROATEAlopecia Acute pancreatitis HepatotoxicityGastrointestinal

upsetSedationTremorWeight lossWeight gain

Anemia (includingmacrocytic with orwithout folatedeficiency)

AsterixisAtaxiaBone marrow

suppressionBreast enlargementComaDermatitisDiplopia, DizzinessDysarthriaEdema of the

extremitiesEncephalopathy with

feverEnuresisEosinophilia, Erythema

multiformeGalactorrheaHallucinationsHeadache

HypofibrinogenemiaHyponatremiaIncoordinationIrreguar mensesLeukopeniaMacrocytosisNystagmusParotid gland swellingPhotosensitivityPorphyria, acute

intermittentPruritusRelative lymphocytosisSecondary amenorrheaStevens-Johnson

syndromeSIADHThrombocytopeniaThyroid function testabnormalities


CARBAMAZEPINECommon (Diminish in Time orWith Temporary Reduction inDose)AtaxiaBlurred visionDiplopiaDizzinessDrowsinessFatigueHeadacheNausea

Less CommonCardiovascular complicationsGastrointestinal upsetHyponatremiaSkin reactions (if severe, may

require discontinuation ofcarbamazepine)

UncommonCognitive impairmentChillsGenitourinary effectsFever

HepatitisIncreased intraocular pressureJaundice, cholestatic and

hepatocellularLiver function abnormalitiesRenal damage leading to

oliguria and hypertensionSIADHTransient leukopenia

(carbamazepine may becontinued unless infectiondevelops)

Water intoxication

RareAgranulocytosisAplastic anemiaLupus erythematosus-like

syndromePulmonary hypersensitivity

SIADH, Syndrome of inappropriate secretion of antidiuretic hormone.

• Gastrointestinal problems may be alleviated by takinglithium with food or switching to a slow release preparation(particularly if gastric distress is the problem).

• Many skin reactions have been described in association withlithium therapy, including atopic dermatitis, acne, psoriasis,and hair loss. These can usually be treated by standarddermatological means but occasionally are severe enough toforce the discontinuation of therapy.

• Electrocardiographic effects of lithium are also usuallybenign and tolerable. Rarely, however, an effect such asslowing of the sinus node can lead to severe bradycardia andsyncope. If continued lithium therapy is a high priority, use ofa pacemaker to provide cardiac stability could be considered.

• Through the dysregulation of calcium metabolism, lithiummay cause hypercalcemia and hyperparathyroidism, which in


turn may lead to osteopenia, osteoporosis, bone resorption,and hypermagnesemia.

• Weight gain is another common adverse effect troublesomein long-term lithium treatment. Its mechanism is unknown.Patients should avoid caloric beverages, restrict food calories,and increase physical activity. Unlike the weight gain asso-ciated with medications such as olanzapine, lithium inducedweight gain is dose-related. Therefore, reduction of the doseshould be considered if the clinical situation allows it.

• Other occasional side effects of lithium are edema and ametallic taste in the mouth.

ANTICONVULSANTSAdverse effects that appear early in the course of valproatetherapy are usually mild and transient, and tend to resolvein time. Gastrointestinal upset is probably the most commoncomplaint in patients taking valproate and tends to be less ofa problem with the enteric-coated divalproex sodium prepa-ration. The administration of a histamine H2 antagonistsuch as famotidine (Pepcid) may alleviate persistent gastroin-testinal problems (the metabolism of valproate is inhibited bycimetidine).

• Other common complaints include tremor, sedation,increased appetite and weight, and alopecia.

• Weight gain is even more of a problem when other drugs areadministered that also promote weight gain, such as lithium,antipsychotics, and other anticonvulsants.

• Less common are ataxia, rashes, and hematological dysfunc-tion, such as thrombocytopenia and platelet dysfunction.

• Platelet count usually recovers with a dosage decrease, butthe occurrence of thrombocytopenia or leukopenia maynecessitate the discontinuation of valproate.

Serum hepatic transaminase elevations are common, doserelated, and usually self-limiting and benign. Fatal hepatotox-icity is extremely rare, is usually restricted to young children,and usually develops within the first six months of valproatetherapy. Hepatic function should be monitored periodicallyduring treatment as previously described.


Other serious problems include pancreatitis and teratoge-nesis. There is also a concern that polycystic ovary syndrome,possibly associated with weight gain, may be a risk for youngwomen who take valproate. If at any point during adminis-tration the side effects of valproate become intolerable, thepsychiatrist may need to discontinue it and try one of theother treatments described in this section as an alternative. Ifvalproate is tolerated but not totally effective, the psychiatristmight use one of the other treatments as an adjunct.

Adverse reactions to carbamazepine (Table 3-12) are morelikely if the dose of carbamazepine is built up rapidly:

• The most common effects in the first couple of weeksare drowsiness, dizziness, ataxia, diplopia, nausea, blurredvision, and fatigue.

• These tend to diminish in time or to respond to a temporaryreduction in dose.

• Less common reactions include gastrointestinal upset,hyponatremia, and a variety of skin reactions, some ofwhich are severe enough to require discontinuation of carba-mazepine.

• Carbamazepine can be teratogenic and should be avoidedduring pregnancy.

• About 10% of patients experience transient leukopenia, butunless infection develops, carbamazepine may be continued.

• More serious hematopoietic reactions, including aplasticanemia and agranulocytosis, are rare.

Frequent blood monitoring, previously recommended, isgenerally considered unnecessary after the first few monthsof therapy. Instead, patients and their families should beinstructed to contact the psychiatrist immediately if petechiae,pallor, weakness, fever, or infection occurs.

Carbamazepine has an antidiuretic hormone-like effect,which may help counteract the opposite effect (which producespolyuria) of lithium when the two drugs are used together.By itself, carbamazepine can cause hyponatremia. Gradualbuildup of carbamazepine levels and smaller, more frequentdosing can help to minimize side effects, particularly in patientsconcomitantly taking other drugs, such as lithium.


Recently oxcarbazepine, the 10-keto analog of carba-mazepine, has become available as an alternative to carba-mazepine for the treatment of seizure disorders. Oxcarbazepinehas less protein binding at 40%, and therefore less possibilityfor drug–drug interactions with highly protein bound drugs.

The only contraindication to lamotrigine use is hypersensi-tivity to the drug, though there is a box warning about derma-tologic events, particularly rashes.

• These rashes generally occur in approximately 10% ofpatients after two to eight weeks and are usually macular,papular, or erythematous in nature.

• Of those who develop a rash, one in 1,000 adults can proceedto a Stephens–Johnson type syndrome.

• Because it is impossible to distinguish which rashes willdevelop into this serious condition, it is advisable to discon-tinue treatment at any sign of drug-induced rash.

Otherwise the most frequently encountered side effectsinclude dizziness, ataxia, somnolence, headache, blurred vision,nausea, vomiting, and diplopia. To reduce the risk of rash andother side effects lamotrigine must be started at a low dose andtitrated slowly, especially if combined with valproate therapy.

ANTIPSYCHOTICSAlthough clozapine is free from most of the extrapyramidalside effects (EPS) associated with typical antipsychotics andmay not cause tardive movement disorders, it does carry therisk of serious adverse reactions, including agranulocytosis,seizures, and cardiorespiratory complications. It should be usedfor bipolar patients only after other options have failed. Manyexperienced clinicians believe clozapine is superior to tradi-tional antipsychotics in the treatment of bipolar disorder, butdefinitive, controlled data are lacking.

Olanzapine has structural and pharmacologic similarities toclozapine but is generally less toxic. It shares with clozapine alow incidence of EPS and of hyperprolactinemia, as comparedto typical agents.

• Adverse effects during the short-term use of olanza-pine include constipation, weight gain, akathisia, drymouth, tremor, increased appetite, orthostatic hypotension,and tachycardia.


• The most common side effect is somnolence, which is whyolanzapine is usually given in one q.h.s. dose.

• In clinical trials, there was a 0.9% incidence of seizures, andalthough there were confounding factors, olanzapine shouldbe used cautiously in patients with known seizure disordersor lowered seizure thresholds.

• Because of alanine aminotransferase elevations in somepatients, it is also advised to use olanzapine cautiously inpatients with known or suspected hepatic impairment.

As mentioned earlier, maintenance therapy with atypicalantipsychotics requires that careful attention must be paid topatient weight and blood lipid and glucose levels. This is partic-ularly true for clozapine and olanzapine but may also be trueto a lesser degree for the other atypical agents.

• Weight gain on olanzapine may be as great as 5–7 kgthough this plateaus in under a year. More modest weightgain occurs less frequently with quetiapine and risperi-done. Severe weight gain occurs in a minority of patientsand, unfortunately, there are currently no means to predictwhether a specific patient will experience significant weightgain. However, weight gain with atypical antipsychoticsmay be more clearly associated with a dramatic incrase inappetite, in food consumption, and in carbohydrate craving.It may be possible to evaluate patients for these featuresand be prepared to see alternative treatments if they occur.

• Atypical antipsychotics are also associated with hyperlipi-demia and with development of abnormalities in glycemiccontrol (ranging from relative insulin resistence to dramaticand life-threatening new onset diabetes). Recommendationsfor monitoring metabolic parameters were discussed earlier.

• Some pharmacologic interventions which may reduceantipsychotic induced weight gain include addition of topi-ramate or of histamine-2 receptor blockers.

DRUG INTERACTIONSLITHIUMThe narrow therapeutic index of lithium means thatmany pharmacokinetic and pharmacodynamic interactions are


potentially clinically significant, and several are associated witha high risk of serious toxicity (Table 3-13).

ANTICONVULSANTSThe anticonvulsants have (predominantly pharmacokinetic)drug interactions due to the importance of hepatic metabolismfor their clearance (Table 3-13).

• Valproate is metabolized by the hepatic cytochrome P4502D6 system.

• Unlike carbamazepine, it does not induce its ownmetabolism or hepatic metabolism in general, but it doesappear to inhibit the degradation of other drugs metabolizedin the liver.

• Valproate inhibits drug oxidation and may increase serumlevels of concomitantly administered drugs that are oxida-tively metabolized, such as phenobarbital, phenytoin, andtricyclic antidepressants.

• Coadministration of carbamazepine, or other microsomalenzyme-inducing drugs, will decrease plasma levels ofvalproate, and drugs that inhibit the P450 system, suchas selective serotonin reuptake inhibitors, can increasethem.

• The coadministration of other highly protein bound drugs,such as aspirin, can increase free valproate blood levels andprecipitate toxic effects.

• Valproate increases the plasma concentrations of phenobar-bital, lamotrigine, and carbamazepine, decreases those ofphenytoin and ethosuximide, and may augment the sedativeaction of diazepam.

• Carbamazepine, phenobarbital, primidone, and phenytoinall decrease plasma levels of valproate, and salicylates canincrease them.

Lamotrigine is metabolized predominately by glucuronicacid conjugation in the liver, with the conjugate and theremaining 10% of the unmetabolized drug excreted in the urine.

• Clearance is markedly increased with the administration ofother drugs that induce hepatic enzymes—including pheny-toin, carbamazepine, and phenobarbital.


� TABLE 3-13. Drug Interactions with Lithium, Valproate, andCarbamazepine

LITHIUMIncrease Lithium

LevelsDecrease Lithium

LevelsIncrease Adverse

ReactionsAngiotensin-

converting enzyme(ACE) inhibitors

Alprazolam (Xanax)Amiloride (Midamor)Antipsychotic agentsEthacrynic acid

(Edecrin)Fluoxetine (Prozac)Ibuprofen (Motrin

and others)Indapamide (Lozol)Indomethacin

(Indocin)Mefenamic acid

(Ponstel)Naproxen (Naprosyn)Nonsteroidal

anti-inflammatorydrugs (NSAIDs)

Phenylbutazone(Butazolidin andothers)

Some antibioticsSpironolactone

(Aldactone andothers)

Sulindac (Clinoril)Thiazide diureticsTriamterene

(Dyrenium andothers)

Zomepirac (no longeravailable)

Aminophylline(Aminophyllinand others)

CaffeineCarbonic

anhydraseinhibitors

Dyphylline(Lufyllin, Dilor)

LaxativesOsmotic diureticsOxtriphylline

(Choledyl)Theobromine

diuretic(Athenol andothers)

Theophylline(Tedral andothers)

Atracurium (Tracrium):prolonged neuromuscularblocking effects

Carbamazepine (Tegretoland others): antithyroideffects

Chlorpromazine (Thorazineand others):extrapyramidalsymptoms, delirium,cerebellar functionimpairment

Clozapine (Clozaril):neurotoxicity

Diltiazem (Cardizem):neurotoxicity

Electroconvulsive therapy:confusion

Fluoxetine (Prozac): lithiumtoxicity

Fluvoxamine (Luvox):seizures

Haloperidol (Haldol):neurotoxicity

Hydroxyzine (Atarax,Vistaril, and others):cardiovascular toxicity

Iodine: antithyroid effectsMethyldopa (Aldomet and

others): hypertension,toxic symptoms atnormal blood levels

Metronidazole (Flagyl andothers): lithium toxicity

Neuroleptics:extrapyramidalsymptoms,somnambulism,neurotoxicity



Pancuronium (Pavulon):prolonged neuromuscularblocking effects

Succinylcholine (Anectineand others): prolongedneuromuscular blockingeffects

Verapamil (Calan andothers): neurotoxicity

VALPROATEIncrease Clearance of Valproate Valproate Decreases Levels OfCarbamazepine (Tegretol)Mefloquine (Lariam)PhenobarbitalPhenytoinRifampin (Rifadin and others)

Increase Levels of ValproateAspirinHighly protein-bound drugs

Valproate Increases Serum Levels OfBarbituratesDiazepam (Valium and others)Ethosuximide (Zarontin)PhenobarbitalPhenytoinPrimidoneTolbutamide (Orinase)Tricyclic antidepressantsZidovudine (Retrovir)

CARBAMAZEPINEIncrease Levels of CarbamazepineCimetidine (Tagamet)DiltiazemErythromycinFluoxetine (Prozac)Fluvoxamine (Luvox)Isoniazid (Nydrazid and others)Propoxyphene (Darvon and others)Valproic acidVerapamil

Decrease Levels of CarbamazepinePhenobarbitalPrimidonePhenytoin (Dilantin and others)

CarbamazepineLamotrigine (Lamictal)

Valproate Increases CNSDepressant Effects Of

AlcoholBupropionClozapine (Clozaril)HaloperidolLoxapine (Loxitane)Maprotiline (Ludiomil)MolindoneMonoamine oxidase inhibitorsPhenothiazinesPimozideThioxanthenesTricyclic antidepressants

Increased Risk of Hemorrhage WithAspirinAnti-inflammatory analgesicsSulfinpyrazone (Anturane)

Carbamazepine Decreases Levels OfAntipsychoticsBenzodiazepines (except clonazepam)CorticosteroidsHormonal contraceptivesThyroid hormoneTricyclic antidepressants

OtherLithium and carbamazepine:

may increase neurotoxicity


• Adding lamotrigine to carbamazepine can decrease steady-state concentrations of lamotrigine by approximately 40%.

• Adding lamotrigine to valproate, however, can decreasesteady-state levels of valproate by approximately 25%, whilethe steady-state levels of lamotrigine increases approxi-mately twofold. In this case, the starting dose of lamotrigineshould be lowered, and the titration made slowly.

Concomitant administration of drugs that inhibit P450 willincrease plasma levels of carbamazepine. Conversely, drugsthat induce P450 enzymes—such as phenobarbital, phenytoin,or primidone—can decrease carbamazepine levels.

ANTIPSYCHOTICSTypical antipsychotics may interact with many other medica-tions and the reader should consult the product prescribinginformation for further details. However, their use is decliningin favor of the atypicals, the most widely used of which areolanzapine and risperidone. For effects of combining lamot-rigine with other anticonvulsants, see Table 3-14.

• The metabolism of olanzapine and risperidone is induced bycarbamazepine

• Levels of risperidone are increased by fluoxetine andparoxetine

• Levels of olanzapine are increased by fluvoxamine• Neither olanzapine nor risperidone significantly interact with

lithium• Olanzapine does not interact with valproate; risperidone

slightly increases valproate levels

� TABLE 3-14. Effects of Combining Lamotrigine with OtherAnticonvulsants

Effects of Anticonvulsants on Lamotrigine LevelsCarbamazepine decreases levels 40%Oxcarbazepine decreases levels 30%Phenobarbital decreases levels 40%Phenytoin decreases levels 50%Valproate increases levels 100%


PregnancyAll mood stabilizers are potentially teratogenic and the risksand benefits of treatment must be carefully assessed in thecontext of use during pregnancy and breastfeeding.

LITHIUMExposure of a fetus to lithium in the first trimester hasbeen associated with an increased risk of Ebstein’s anomaly,a cardiac defect characterized by a malformation of thetricuspid valve. Data from a voluntary, retrospective source,the Danish Registry of Lithium Babies, provided an estimatedrelative risk for Ebstein’s anomaly (compared to the generalpopulation) of 400. More recently, prospective cohort studiesfound risk ratios of up to 3 for all malformations and up to7.7 for cardiac malformations. This corresponds to a risk ofEbstein’s anomaly of less than 1 in 2000. This could be seenas a tolerable level of risk, particularly for a patient who hasfailed other mood stabilizer treatments and becomes rapidlyvery ill when she is not on lithium.

ANTICONVULSANTSAnticonvulsants as a class are clearly the most dangerous ofthe mood stabilizers to use during pregnancy. Fetal exposurein the first trimester to monotherapy treatment with anti-convulsants increases the risk of major congenital malforma-tions 2–3 fold. When anticonvulsants are combined, the riskincreases further. Valproate produces the greatest increase inrisk of any individual anticonvulsant, being associated withmajor malformations at a rate several fold higher than otherspecific anticonvulsant, including carbamazepine. This excessrisk associated with valproate is dose-related, becoming mostevident at doses greater than 800–1000 mg/day. Among thenewer anticonvulsants, lamotrigine has the largest prospec-tive dataset. Rates of major anomalies were 2.9, 12.5 and2.7% following 1st trimester exposure to lamotrigine alone,lamotrigine with valproate, and lamotrigine with other anti-convulsants, respectively. These data have been interpretedto suggest that lamotrigine has a risk no greater than otheranticonvulsants and less than that of valproate. Similar to


valproate, lamotrigine also produces greater rates of anoma-lies at higher doses. There is insufficient data at presentto evaluate the risk of congenital anomalies following firsttrimester exposure to oxcarbazepine. The specific congenitalabnormalities most closely linked to anticonvulsant exposureare neural tube defects, especially spina bifida. Use of folatesupplementation may reduce this risk, but certainly does noteliminate it. As with congenital anomalies in general, neuraltube defects occur with greatest frequency following valproateexposure (1–5%) relative to carbamazepine exposure(0.5–1%).

ANTIPSYCHOTICSThere are less data available to guide risk/benefit assessmentwhen considering using an atypical antipsychotic in pregnancy,but there is no evidence at present of any clear increase inthe rate of fetal anomalies overall or evidence for the pres-ence of any specific congenital syndrome associated with expo-sure. The typical antipsychotics have a much longer historyof use in pregnancy and are also not associated with anyspecific syndrome. There is some evidence, however, thathigher potency neuroleptics may be safer than lower potencyagents.

GENERAL PRINCIPLESBecause they are, of course, intimately interconnected, boththe developing fetus and the mother must be considered intreating the pregnant women with bipolar disorder. Moodstabilizer discontinuation at any time is associated with anincreased risk of relapse, which may lead to self-destructive orreckless behavior. In addition, we know from many preclin-ical studies that severe stress during pregnancy, in a range ofspecies including primates, can produce long-standing abnor-malities in the hypothalamic-pituitary-adrenal axis and increasethe vulnerability to behaviors suggestive of depression andanxiety. Therefore, the ideal of minimizing exposure to medica-tions during pregnancy may not always be the best option.

It is critical to carefully evaluate the potential risks andbenefits of maintaining or discontinuing mood stabilizer treat-ment during pregnancy for each individual patient, in light


of their need for specific agents and likelihood and speed ofrelapse off of medication. This must be done in a very open andcomprehensive fashion, fully involving the patient and otherimportant individuals, and emphasizing a thorough discussionof both the known risks and the limits of our knowledge.Ideally, options for pharmacologic intervention, if either eleva-tion or depression intercedes, should be thoroughly discussed,in order to allow for the most efficient and informed interven-tion, should one be needed.

Another critical feature to consider is that the windowof opportunity for medication discontinuation is very narrowwith respect to the congenital abnormalities of concern withmood stabilizers. Cardiac development and neural tube closureoccur relatively early, in the first trimester. Therefore, a patientpresenting with one or two missed periods is likely already tobe beyond the stage where medication discontinuation will beable to prevent increased risk. This argues for having a fulldiscussion of risks with reproductive age women and a discus-sion about contraceptive methods being employed. Keep inmind, as well, that low potency oral contraceptives may berendered ineffective by enzyme inducing medications such ascarbamazepine.

Evidence from retrospective studies suggests that there isno protective effect of pregnancy on mood stability. In theimmediate post-partum period, bipolar women are at markedlyelevated risk of relapse. The most thorough study to date founda 90% relapse rate for unmedicated bipolar women in the firsttwo months post-partum. Therefore, regardless of the decisionsmade during pregnancy, aggressive pharmacotherapy shouldbe initiated in the immediate post-partum period.

SUMMARYRecent years have seen a remarkable increase in our under-standing of bipolar disorder and the range of pharmacologicagents with which it can be treated. Lithium and valproateremain the core treatment options for bipolar disorder butare increasingly being supported by additional anticonvulsantand atypical antipsychotic agents. Although not definitivelydemonstrated yet, two relatively new medications, lamotrigine


and quetiapine, hold significant promise as mood stabilizerswith more robust antidepressant efficacy, helping to addressan area of critical need. The increase in treatment optionsis simultaneously promising and challenging. It requires thatclinicians learn to employ newer and older agents in the mostrational manner, with decisions soundly based upon evidenceof efficacy in clinical trials, as well as upon the individualclinical characteristics of each patient with whom we work.Carefully reviewing and documenting treatment response andtolerability is especially important in the unstable setting ofbipolar illness, and can be aided by the use of clinical ratingscales and daily prospective mood charting.

Even the best medication selection is of limited valueby itself. A close, collaborative, and honest working alliancebetween the psychiatrist, the patient and family members, aswell as a commitment to psychoeducation, is important forfostering trust, treatment adherence, and insight. Supportivepsychotherapy techniques employed by a therapist knowledge-able about bipolar illness can also be extremely useful, helpingpatients to learn to properly differentiate normal emotions orreactions from affective symptoms, and promoting a healthylifestyle with good sleep hygiene, reduction of stress, and avoid-ance of drugs and alcohol.

ADDITIONAL READING

1. Keck, PE, Jr., Perlis RH, Otto MW, Carpenter D, Ross R,Docherty JP. The Expert Consensus Guideline Series: Treat-ment of Bipolar Disorder 2004. Postgrad Med Special Report.2004(December): 1–120.

2. American Diabetes Association; American Psychiatric Associa-tion; American Association of Clinical Endocrinologists; NorthAmerican Association for the Study of Obesity. Consensusdevelopment conference on antipsychotic drugs and obesity anddiabetes. J Clin Psychiatry. 2004 Feb; 65(2):267–72.

4ANXIOLYTIC DRUGS

INTRODUCTIONThere has been an exponential increase in the number ofmedications demonstrated to be effective for the treatment ofanxiety and anxiety disorders.

Barbiturates and meprobamate were some of the firstagents shown to be effective in decreasing anxiety, but wereaddictive and often lethal in overdose. In the early 1960s,Klein demonstrated that the tricyclic antidepressant (TCA)imipramine was useful in the treatment of panic disorder. Inthe 1970s, studies showed that monoamine oxidase inhibitors(MAOIs) were also effective in the treatment of certain anxietydisorders, such as social anxiety and anxiety with coexistingdepression.

The introduction of benzodiazepines in the early 1960swas a major advance; they were much safer than the barbi-turates and meprobamate, had a rapid onset of action, and abroad spectrum of efficacy extending from situational anxietyto pathological anxiety disorders. Many different benzodi-azepines, with different absorption times and half-lives, weredeveloped and have been valuable not only for treating anxietysymptoms and anxiety disorders but for treating seizure disor-ders and alcohol withdrawal. Unfortunately, with wide-scaleusage, problems with craving, dependence, and withdrawalwith abrupt discontinuation were noted. The next major class



of agents approved was the azopyrones, of which buspirone isthe most well known. This agent was effective in generalizedanxiety disorder (GAD) but not for other anxiety disorders.

The selective serotonin reuptake inhibitors (SSRIs) asa class have demonstrated effficacy for most anxiety disor-ders. Although these agents have a delayed onset whencontrasted with the benzodiazepines, they have a broader spec-trum of action, no problems with dependence, and much lessof a problem with withdrawal syndromes. Controlled trialshave demonstrated the efficacy and safety of the selectiveserotonin-norepinephrine reuptake inhibitor (SNRI) medica-tion venlafaxine in the treatment of anxiety disorders includingsocial anxiety disorder, generalized anxiety disorder, posttrau-matic stress disorder, panic disorder, and obsessive–compulsivedisorder. Other classes of medications used in the treatment ofanxiety disorders (either as primary treatments or as adjuvants)include anticonvulsants, beta-blockers, and atypical antipsy-chotics.

A GENERAL APPROACH TO USING MEDICATIONWITH ANXIOUS PATIENTSPatients may present to physicians with many differentconcerns related to anxiety. Such patients commonly havethe need for reassurance that their disorder is treatable,and that their physicians truly hear their concerns and willattempt to help them. Taking a complete history is essentialnot only for making an appropriate diagnostic formulationbut for developing a therapeutic alliance. Medical evaluation,including laboratory testing, and assessment of current andpast substance abuse or dependence, are particularly impor-tant in the evaluation of patients presenting with symptoms ofanxiety. The differential diagnosis for such patients includes:

• Adjustment disorders secondary to life stressors• Anxiety symptoms or anxiety disorders secondary to a

medical condition• Anxiety secondary to alcohol or substance abuse, depen-

dence, or withdrawal• Generalized anxiety disorder (GAD)• Panic disorder (PD), with or without agoraphobia

ANXIOLYTIC DRUGS 131

• Social anxiety disorder (e.g., social phobia)• Specific phobia• Posttraumatic stress disorder (PTSD)• Obsessive–compulsive disorder (OCD)

Sharing the diagnostic formulation with the patient isan important intervention that often facilitates the patient’scommitment to the treatment plan. This is crucial since anxiouspatients may be reluctant to take medication. When they dotake medications, they commonly ruminate about medicationside effects. Patients with anxiety symptoms or anxiety disor-ders are likely to have somatic preoccupations and heightenedsomatic sensitivity. A collaborative approach where physiciansand patients form a “team” to monitor both the potentialbenefits and risks of any medication intervention frequentlyenhances adherence. An important rule when initiating phar-macological treatment for patients with anxiety disorders is“to start low and go slow”. Interestingly, although patientswith anxiety disorders frequently require more gradual initialtitration schedules, they often attain maintenance dosages ofantidepressant medications that are greater than the doses usedto treat major depressive disorder.

Along with medication interventions, psychoeducationabout anxiety disorders is often a key part of treatment. Inaddition, psychotherapeutic interventions have been demon-strated to be effective in anxiety disorders, particularlycognitive-behavioral therapy, which may include cognitiverestructuring, relaxation and breathing exercises, and gradedexposure to anxiety-provoking stimuli.

PHARMACOLOGYAntidepressantsIt has long been known that antidepressants are also frequentlyeffective treatments for anxiety disorders. The basic action ofthe majority of the antidepressants is to increase the avail-ability of neurotransmitters in the synaptic cleft. The chemistryand pharmacokinetics of these agents are reviewed fully inChapter 2.


The most widely used antidepressants with anxiolytic prop-erties are the SSRIs. These agents have the broadest spectrumof activity which spans the entire spectrum of DSM-IV-TRanxiety disorders.

The mechanism of action of the TCAs again involves theinhibition of reuptake sites; their disadvantages include theirside effect profile and their potential lethality in overdose.MAOIs can also be effective anxiolytics. The pharmacology ofboth groups of agents are reviewed in Chapter 2.

Other antidepressants with anxiolytic effects have differentmechanisms of action. These include the inhibition of bothserotonin and norepinephrine transporter sites (as seenwith venlafaxine dosed above 150 mg/d, and with duloxetineand clomipramine); antagonism of the presynaptic alpha-2-adrenergic receptors (mirtazepine); and antagonism of postsy-naptic serotonin type-2 receptors (nefazodone).

The relative differences in terms of major pharmacokineticand pharmacodynamic properties are outlined in Table 4-1.

BenzodiazepinesThe benzodiazepines as a class work by increasing the relativeefficiency of the gamma-aminobutyric acid (GABA) receptorwhen stimulated by GABA. They bind to a site located adja-cent to the GABA receptor and cause an allosteric changeto the receptor that facilitates the increased passage of thechloride ions intracellularly when GABA interacts with thereceptor complex. This leads to a relative hyperpolarizationof the neuronal membrane and inhibition of activity in thebrain.

The benzodiazepines as a group have different affinities forGABA receptors; some agents bind to only one of the twotypes of GABA receptors. Both clonezapam and alprazolamwork only on the central GABAA receptor, while diazepambinds to both GABAA and GABAB receptors. The pharma-cokinetic properties of the benzodiazepines are outlined inTable 4-1. The half life of clonazepam is significantly longerthan alprazolam (30–40 hours vs. 6–27 hours respectively); thisis reflected in the longer time to steady-state plasma levels forclonazepam (up to 1 week). The relative pharmacodynamic

�TA

BLE

4-1.

Phar

mac

okin

etic

Prop

ertie

sof

Psyc

hotr

opic

Med

icat

ion

Used

for

the

Trea

tmen

tof

Anx

iety

Diso

rder

s

SSR

Is

Flu

oxet

ine

Flu

voxa

min

eP

arox

etin

eP

arox

etin

eC

RSe

rtra

line

Esc

ital

opra

mC

ital

opra

mT

max

h6–

83–

85.

26–

104.

5–8.

43–

52–

4D

ose-

prop

orti

onal

plas

ma

leve

lN

oN

oN

oN

oY

esY

esY

es

T1/

2(h

)24

–72

15.6

2115

–20

2630

33M

etab

olit

eac

tivi

tyN

orfl

uoxe

tine

(equ

al)

<10

%<

2%<

1%D

esm

ethy

l-se

rtra

line

6–15

%

Non

e<

10%

Met

abol

ite

T1/

24–

16d

––

–62

–104

h50

–60

h–

Stea

dyst

ate

plas

ma

leve

l4–

5w

k∼1

wk

10d

10–1

4d

∼1w

k10

d∼1

wk

Usu

alda

ilydo

sage

rang

e10

–80

mg

100–

300

mg

10–6

0m

g12

.5–7

5m

g50

–200

mg

10–2

0m

g10

–60

mg

Oth

erA

nti

dep

ress

ants

Mir

taza

pine

Nef

azod

one

Ven

lafa

xine

Ven

lafa

xine

XR

Tm

ax

h2

12

5.5

Dos

e-pr

opor

tion

alpl

asm

ale

vel

Yes

No

Yes

Yes

T1/

2(h

)37

(fem

ales

),26

(mal

es)

2–4

5±

2

�TA

BLE

4-1.

(Con

tinue

d)

Oth

erA

nti

dep

ress

ants

Mir

taza

pine

Nef

azod

one

Ven

lafa

xine

Ven

lafa

xine

XR

Met

abol

ite

acti

vity

Neg

ligib

leH

ydro

xyne

fazo

done

O-d

esm

ethy

l-ve

nlaf

axin

eC

-des

met

hyl-

venl

afax

ine

Met

abol

ite

T1/

2(h

)–

1.5–

411

±2

11±

2

Stea

dyst

ate

plas

ma

leve

l5

d4–

5d

3d

3d

Usu

alor

aldo

sage

15–6

0m

g10

0–80

0m

g45

–75

mg

75–2

25m

g

Anti

anxie

ty

Bus

piro

neA

lpra

zola

mA

lpra

zola

mX

RC

lona

zepa

mL

oraz

epam

Tm

ax

h0.

6–1.

51–

25–

111–

41–

1.5

Dos

e-pr

opor

tion

alpl

asm

ale

vel

No

Yes

Yes

Yes

Yes

T1/

2(h

)2–

36.

3–26

.96.

3–26

.930

–40

12–1

5M

etab

olit

eac

tivi

tyU

nim

port

ant

�-h

ydro

xyl-

alpr

azol

am50

%�

-hyd

roxy

l-al

praz

olam

and

4-hy

drox

yl-

alpr

azol

am

No

Uni

mpo

rtan

t

Stea

dyst

ate

plas

ma

leve

l–

3–4

d3–

4d

1w

k4

d

Usu

alor

aldo

sage

15–9

0m

g1–

10m

g1–

10m

g1–

6m

g1–

6m

g

.

Anti

psy

choti

cs

Ola

nzap

ine

Que

tiap

ine

Ris

peri

done

Zip

rasi

done

Ari

pipr

azol

eT

max

h6

1.5

16–

83–

5D

ose-

prop

orti

onal

plas

ma

leve

lY

esY

esY

esY

esY

es

T1/

2(h

)21

–54

621

–30

775

–96

Met

abol

ite

acti

vity

No

No

9-hy

drox

yris

peri

done

Yes

Deh

ydro

-ari

pipr

azol

eSt

eady

stat

epl

asm

ale

vel

4–6

d2

d5–

6d

14d

Usu

alor

aldo

sage

5–10

mg

300–

400

mg

b.i.d

.2–

8m

g20

–80

mg

b.i.d

.10

–30

mg

Anti

convuls

ants

Gab

apen

tin

Val

proi

cac

idP

rega

balin

Tia

gabi

neT

opir

amat

eT

max

h4–

51.

50.

754

Dos

e-pr

opor

tion

alpl

asm

ale

vel

No

No

No

No

Yes

T1/

2(h

)5–

79–

166.

37–

921

Met

abol

ite

acti

vity

No

No

No

No

No

Stea

dyst

ate

plas

ma

leve

l1

d7

d24

–48

h2

d4

d

Usu

alor

aldo

sage

2,40

0m

g/d

750–

2,50

0m

g/d

150–

600

mg

4–32

mg/

d10

0–20

0m

g/d

T1/

2,t

erm

inal

half

-lif

e.T

max,t

ime

ofm

axim

umpl

asm

aco

ncen

trat

ion.


and pharmacokinetic properties of the benzodiazepines arefurther outlined in comparison to the other medications inTable 4-2.

Buspirone

Buspirone is believed to exert its anxiolytic effect by acting asa partial agonist at the 5-HT1A autoreceptor. Stimulation ofthe 5-HT1A autoreceptor causes decreased release of serotonininto the synaptic cleft. However, buspirone also exerts anothereffect through its active metabolite 1-phenyl-piperazine (1-PP), which acts on alpha-2-adrenergic receptors to increasethe firing rate of the locus coeruleus. Some not yet well-characterized combination of these effects may be responsiblefor the anxiolytic effect of buspirone.

It usually takes approximately four weeks for the benefitof buspirone therapy to be detected in patients with GAD. Amajor advantage of buspirone is that it does not cross-reactwith benzodiazepines. The most common side effects associ-ated with buspirone include dizziness, gastrointestinal distress,headache, numbness, and tingling. The pharmacokinetics andaverage daily dosage are described in Table 4-1. The mostcommon pharmacokinetic and pharmacodynamic actions ofbuspirone are described in Table 4-2.

Beta-Blocker Medications

Beta-adrenergic blockers competitively antagonize nore-pinephrine and epinephrine at the beta-adrenergic receptor(Table 4-2). It is thought that the majority of positive effects ofbeta-blockers are due to their peripheral (rather than central)actions. Beta-blockers can decrease many of the peripheralmanifestations of anxiety such as tachycardia, diaphoresis,trembling, and blushing. The advent of more selective beta-blockers that only block the beta-2-adrenergic receptor hasbeen beneficial since blockade of beta-1-adrenergic receptorscan be associated with bronchospasm.

�TA

BLE

4-2.

ASu

mm

ary

ofPh

arm

acol

ogic

Prop

ertie

sof

Med

icat

ions

Com

mon

lyUs

edto

Trea

tA

nxie

ty

ON

SET

OF

ACT

ION

TITR

ATI

ON

ABU

SELI

ABI

LITY

NEE

DFO

RDI

SCO

NTI

NUA

TIO

NTI

TRA

TIO

N

POTE

NTI

AL

FOR

WIT

HDR

AW

AL

SYN

DRO

ME

PRO

BABI

LITY

OF

LETH

ALI

TYIN

OVE

RDO

SE

Sert

ralin

eD

elay

ed(I

n2

wk)

Yes

Ver

ylo

wY

es,

but

not

man

dato

ryV

ery

low

Low

Par

oxet

ine

Del

ayed

(In

2w

k)Y

esV

ery

low

Yes

Mod

erat

eL

owF

luvo

xam

ine

Del

ayed

(In

2w

k)Y

esV

ery

low

Yes

,bu

tno

tm

anda

tory

Ver

ylo

wL

ow

Flu

oxet

ine

Del

ayed

(In

2w

k)Y

esV

ery

low

No

Low

est

Low

Cit

alop

ram

Del

ayed

(In

2w

k)Y

esV

ery

low

Yes

,bu

tno

tm

anda

tory

Ver

ylo

wL

ow

Esc

ital

opra

mD

elay

ed(I

n2

wk)

Som

etim

esV

ery

low

Yes

,pr

obab

lyno

tm

anda

tory

Ver

ylo

wL

ow

Ven

lafa

xine

Del

ayed

(In

2w

k)Y

esV

ery

low

Yes

Mod

erat

eL

owD

ulox

etin

eD

elay

ed(I

n2

wk)

Yes

Ver

ylo

wY

esM

oder

ate

Low

Mir

taze

pine

Del

ayed

Yes

Ver

ylo

wY

esM

oder

ate

Low

Nef

azod

one

Del

ayed

Yes

Ver

ylo

wY

esM

oder

ate

Low

Bup

ropi

onD

elay

edY

esV

ery

low

Yes

Low

Low

TC

AS

Del

ayed

(2w

k)Y

esV

ery

low

Yes

Mod

erat

eM

oder

ate–

high

MA

OIs

Del

ayed

(2w

k)Y

esV

ery

low

Yes

Mod

erat

eM

oder

ate–

high

�TA

BLE

4-2.

(Con

tinue

d)

ON

SET

OF

ACT

ION

TITR

ATI

ON

ABU

SELI

ABI

LITY

NEE

DFO

RDI

SCO

NTI

NUA

TIO

NTI

TRA

TIO

N

POTE

NTI

AL

FOR

WIT

HDR

AW

AL

SYN

DRO

ME

PRO

BABI

LITY

OF

LETH

ALI

TYIN

OVE

RDO

SE

Bus

piro

neD

elay

ed(2

wk)

Yes

Ver

ylo

wY

esL

ow–m

oder

ate

Low

Clo

naza

pam

Rap

idY

esM

oder

ate

Yes

Mod

erat

e–hi

ghL

owA

lpra

zola

mV

ery

rapi

dY

esM

oder

ate

Yes

Hig

hL

owA

lpra

zola

mX

RR

apid

Yes

Mod

erat

eY

esM

oder

ate–

high

Low

Lor

azep

amV

ery

rapi

dY

esM

oder

ate

Yes

Mod

erat

e–hi

ghL

owD

iaze

pam

Rap

idY

esM

oder

ate

Yes

Mod

erat

eL

ow�

-blo

cker

sR

apid

Som

etim

esL

owN

o(a

cute

use)

No

(acu

teus

e)L

ow–m

ediu

mG

abap

enti

nM

oder

ate

(d)

Yes

Low

Yes

Low

–mod

erat

eL

owR

ispe

rido

neR

apid

Pro

babl

yL

owY

esL

owL

owO

lanz

apin

eR

apid

Pro

babl

yL

owY

esL

owL

ow


AnticonvulsantsThe precise mechanism of action of many mood stabilizers arenot yet fully understood. Gabapentin, pregabalin, and viga-batrin increase brain GABA levels or neurotransmission atleast in part by targeting the metabolic pathways of GABA.Tiagabine selectively increases synaptic GABA availability byblocking the reuptake of GABA via transporter inhibition.Evidence exists, to a greater or lesser extent, that all of theseagents possess anxiolytic properties. See Tables 4-1 and 4-2 fora review of some of the more salient pharmacological proper-ties of these agents.

AntipsychoticsConventional or typical antipsychotics are rarely used as adju-vant medication for anxiety disorders due to problems withextrapyramidal side effects and the risk of developing tardivedyskinesia. The newer class of atypical antipsychotic medica-tions appear to have a decreased risk of these side effects andare beginning to be used as adjuvants in patients with treat-ment resistant anxiety disorders.

Although the different atypical antipsychotic medicationshave varying affinities for dopamine Type-2 and serotoninType-2 receptors, this is the common mechanism of action ofthese agents. The atypical antipsychotic medications also differdramatically in terms of their pharmacodynamic properties; afull review of these can be found in Chapter 1 and in Tables 4-1and 4-2.

INDICATIONS FOR USEThe efficacy of various psychotropic drugs for the treatmentof anxiety disorders is summarized in Table 4-3.

AntidepressantsSSRISThe SSRIs are considered the first-line treatment option formost of the anxiety disorders, including generalized anxietydisorder (GAD), social anxiety disorder (SAD), panic disorder

�TA

BLE

4-3.

The

Effic

acy

ofPs

ycho

trop

icM

edic

atio

nsfo

rth

eTr

eatm

ent

ofA

nxie

tyDi

sord

ers

GEN

ERA

LIZE

DA

NXI

ETY

DISO

RDER

SOCI

AL

AN

XIET

YDI

SORD

ERPA

NIC

DISO

RDER

POST

TRA

UMA

TIC

STRE

SSDI

SORD

ERO

BSES

SIVE

–CO

MPU

LSIV

EDI

SORD

ER

Stro

ngev

iden

ceSS

RIs

SSR

IsSS

RIs

SSR

IsSS

RIs

Ven

lafa

xine

Bup

ropi

on-S

RV

enla

faxi

neT

CA

sM

AO

IsC

lom

ipra

min

e

Tra

zodo

neM

AO

IsT

CA

sB

enzo

diaz

epin

esM

AO

IsB

enzo

diaz

epin

esT

CA

sB

uspi

rone

Ben

zodi

azep

ines

Som

eev

iden

ceN

efaz

odon

eM

irta

zapi

neV

enla

faxi

neN

efaz

odon

e

Gab

apen

tin

Mir

taze

pine

Nef

azod

one

Clo

naze

pam

+se

rtra

line

Bus

piro

nead

junc

tto

benz

odia

zepi

neV

alpr

oic

acid

Gab

apen

tin

Tia

gabi

neP

agoc

lone

Ven

lafa

xine

Lam

otri

gine

Val

proa

te

Nef

azod

one

Mir

taza

pine

Clo

nidi

ne

MA

OIs

Ola

nzap

ine

augm

enta

tion

ofSS

RI

Ris

peri

done

augm

enta

tion

ofSS

RI

Ven

lafa

xine

Mir

taza

pine

Not

effe

ctiv

eT

CA

sT

razo

done

Tra

zodo

neB

uspi

rone

Bup

ropi

onT

CA

sP

indo

lol

augm

enta

tion

ofSS

RI

Bus

piro

ne

No

data

MA

OIs

Tra

zodo

neD

ulox

etin

eB

upro

pion

Mir

taza

pine

Bup

ropi

on


(PD), posttraumatic stress disorder (PTSD), and obsessive–compulsive disorder (OCD).

In GAD, paroxetine and sertraline have both shown effi-cacy in both adults and children, with symptom reduction andremission as measured by a variety of scales. Lower doses ofsertraline are normally used for children with GAD and appearto be effective and well tolerated. Escitalopram has also beendemonstrated to be effective in GAD.

SSRIs have emerged as first-line treatment for socialanxiety disorder (SAD), also known as social phobia. Mostof the efficacy data are derived from multicenter, double-blind trials of paroxetine, sertraline, and fluvoxamine. The useof sertraline for the longer-term treatment of social anxietydisorder has been investigated and shown efficacy in alleviatingsymptoms and prevention of relapse. Other SSRIs includingfluoxetine, citalopram, and escitalopram also seem to be effec-tive in the treatment of this disorder.

SSRIs are generally accepted as a first-line treatment forpanic disorder (PD). The major advantage of these agents istheir tolerability and thus longer-term acceptance by patients.At present there is evidence that fluoxetine, sertraline, parox-etine, fluvoxamine, and citalopram are effective in the acutetreatment of panic disorder. Fluoxetine and sertraline havebeen shown to reduce panic attack frequency, global distress,and agoraphobic distress. Maintenance treatment with sertra-line in one study was associated with continued improvementand protected patients from recurrence. Paroxetine has beenreported to have a more rapid onset of action than otherSSRIs in PD, and to show continued improvement over time.Citalopram and escitalopram are effective treatments of panicdisorder, decreasing both panic attacks and phobic symptoms.Fluvoxamine has also shown efficacy for the treatment of panicdisorder with or without agoraphobia.

SSRIs are effective for the treatment of PTSD, decreasingmany of the core symptoms of this disorder. Fluoxetine wasthe first SSRI to be studied in randomized clinical trials, andshowed significant improvement in both civilian and veteranpopulations. Two SSRIs have been approved by the FDAfor the treatment of PTSD, sertraline and paroxetine. In onestudy, doses of sertraline of 50 and 200 mg/day gave sustained


response; patients who responded during the acute phase notonly maintained their response but continued to improve withsix months of continuation treatment. Paroxetine in doses of20 to 50 mg/day has been associated with treatment responseacross various types of trauma and in both men and women.Medication discontinuation has been associated with a signifi-cant risk of relapse and reemergence of the core symptoms ofPTSD. This suggests that some patients will require sustainedSSRI treatment, possibly for years, in order to protect themfrom exacerbation of their symptoms.

Large, well-designed, double-blind, placebo-controlledtrials have demonstrated that the SSRIs fluoxetine, paroxe-tine, fluvoxamine, citalopram, and sertraline are effective acutetreatments for OCD in both adults and children, as measuredby a variety of severity scales. No significant differenceshave been noted between citalopram, paroxetine, and fluvox-amine, although citalopram may show efficacy in patients withtreatment refractory OCD who were previously treated withanother SSRI for at least six months. Sustained improvementhas been shown with sertraline over a period of two yearsof treatment in both adults and children with OCD, and alsoappears to have a role in relapse prevention. DiscontinuingSSRI treatment is associated with an exacerbation of symp-toms and worsening in quality of life. These data suggest thatSSRIs are a first-line pharmacotherapy for both acute andmaintenance treatment of OCD in children and adults.

SNRISIn comparison to the SSRIs, there is less research demon-strating the efficacy of SNRI medications for treatment ofanxiety disorders. In controlled trials, venlafaxine (particularlythe extended release form) has been shown to be effectivein the treatment of GAD, SAD, PTSD, PD, and OCD. Forexample, in one study of GAD, venlafaxine ER in doses as lowas 37.5 mg/day and as high as 225 mg/day was found to be effec-tive in decreasing symptoms of anxiety. Side effects appearedto be mild and tended to decrease in number and intensity overthe course of the studies. Nausea, dry mouth, and somnolencewere the most commonly repeated side effects. Venlafaxine


ER has also been shown to be effective in the short-term treat-ment of generalized social anxiety disorder. Its place relativeto SSRIs in panic disorder is not yet clear; for instance, ina recent large study in panic disorder Venlafaxine ER treat-ment led to greater remission and fewer limited symptompanic attacks but not to a greater percentage of patients freeof full-symptom panic attacks. Venlafaxine has been shownto produce improvement in PTSD symptom severity in somesmall studies.

Comparable studies on the use of duloxetine, a newly-released SNRI, in anxiety disorders have not yet been reported.

TCAS

The tricyclic antidepressants have been demonstrated to haveefficacy for many anxiety disorders, including SAD, PD, PTSD,and OCD, but not social anxiety disorder. However, the TCAshave significant side effects, may lead to more difficulty indosage titration, and are potentially lethal in overdose. Inaddition, many anxiety disorders require long-term medicationtreatment. Since the significant side-effect burden of TCAsoften leads to long-term non-compliance, and hence a greaterchance of relapse, their use has diminished in recent years.

A variety of tricyclic antidepressants (TCAs) have beendemonstrated to be effective treatments for GAD. Similarly, inpanic disorder, numerous TCAs (particularly the serotonergicagents such as imipramine) have been shown to be effectivefor both acute and maintenance treatment of panic attacks. InPTSD, several TCAs (including amitriptyline and imipramine)have shown efficacy in decreasing signs and symptoms of thatdisorder. However, following the advent of SSRIs with theirsignificant safety and tolerability advantages, PTSD studieswith TCAs have not been pursued.

The TCA clomipramine is particularly effective in the treat-ment of OCD. Used in doses ranging from 100 to 250 mg/day,clomipramine has been shown to be as effective as the SSRIsand may, in some instances, be more effective in treatingOCD symptoms. However, it is highly anticholinergic andsedating, unlike the SSRI medications, and more likely to leadto discontinuation.


MAOISMAO inhibitor medications are efficacious in many anxietydisorders, but are not commonly used as first-line treatmentsbecause the need to follow a tyramine-free diet to avoid hyper-tensive crisis makes this class of drugs unappealing for mostpatients. For instance, first-generation MAOIs are effectivein the treatment of social anxiety disorder, and may be moreeffective than the SSRI medications. These agents may beuseful for severely ill patients with PD, particularly thosewith comorbid depressive disorder. A few small double-blindstudies suggest that MAOIs are more effective than placebo inthe treatment of PTSD. MAOIs have shown efficacy in somestudies of OCD, but are not generally used as a primary treat-ment. Reversible MAO inhibitors such as moclobemide andbrofaromine (neither is available in the US) are less likely toinduce hypertensive crisis than first-generation MAOIs. Theyhave also been studied in anxiety disorders including SAD andPD, and have demonstrated efficacy, though perhaps less thanfirst-generation MAOIs.

OTHER ANTIDEPRESSANT AGENTSAnxiety disorders may respond to other classes of antidepres-sants, in general more variably than to the SSRIs.

Bupropion Buproprion-SR has been found to be effective inthe treatment of social anxiety disorder, and to be ineffectivein PD, and also in an open-label study with PTSD.

Mirtazapine Mirtazapine may be effective in the treatment ofpanic disorder and generalized anxiety disorder. Mirtazapinehas also been studied in patients with chronic PTSD; resultsfrom small studies indicate it may also be efficacious as anadjuvant in the treatment of PTSD. In OCD, small studies withmirtazapine (as primary treatment or augmentation of SSRImedication) suggest possible benefit, even an acceleration oftreatment response.

Nefazodone The efficacy of nefazodone has been assessedin open-label studies in PTSD; results suggest that this agentmay decrease the core symptoms of PTSD, improve sleep, anddecrease symptoms of anger. It may be helpful for patientswho are refractory to SSRI treatment.


Trazodone Some older studies also suggest that trazodone isan effective treatment for GAD.

BenzodiazepinesAs a class, benzodiazepines are efficacious for the treatment ofmany anxiety disorders, including panic disorder, social anxietydisorder, generalized anxiety disorder, alcohol withdrawal, andsituational anxiety. Although obsessive–compulsive disorderfalls within the taxonomy of anxiety disorders, benzodiazepinesdo not seem to be particularly effective in treating thesepatients. The BZDs may be contraindicated in the ongoingtreatment of PTSD; though open-label studies have suggestedefficacy, a recent double-blind study found no benefit, and itis believed that BZDs may cause depression in such patientsand potentially worsen the course of the disorder. There areconcerns about dependency and stigma associated with the useof benzodiazepine medications.

Short acting benzodiazepines, longer acting benzodi-azepines, and even the low potency benzodiazepines haveall been demonstrated to be effective treatments for GAD.However, despite evidence that the anxiolytic effect does nottend to diminish over time and that the majority of patientsdo not abuse benzodiazepines, many physicians are reticent toinitiate long-term BZD treatment for GAD.

The benzodiazepines clonazepam and alprazolam havebeen shown to be efficacious in treating social anxiety disorder.The potential for abuse and drug withdrawal is a particularlyproblematic issue in social anxiety disorder because of the highrate of comorbid substance abuse, and their use must be moni-tored carefully in such patients. Benzodiazepines may be bestsuited for patients with situational and performance anxietyon an as needed basis.

There is strong evidence that two high potency benzo-diazepines, alprazolam and clonazepam, are effective in thetreatment of PD. Alprazolam has been demonstrated in bothshort-term and long-term studies to be effective; however,there has been considerable concern about the risk of depen-dency and also the difficulties discontinuing alprazolam for asignificant minority of patients. The immediate-release form


of this agent has a relatively short half-life, requiring frequentdosing (up to four times a day). Recently an extended-releaseform of alprazolam has been introduced, with once-dailydosing.

Clonazepam at doses of 0.5–4.0 mg/day is effective and welltolerated in PD. Longer-term clonazepam treatment was asso-ciated in one study with continuing improvement and mainte-nance; the maintenance dosage was either constant with theinitial response dose or decreased over time. There were nosignificant adverse effects associated with longer-term clon-azepam treatment, though symptoms may recur upon discon-tinuation. Diazepam and lorazepam have also shown efficacyin treatment of PD.

BuspironeBuspirone has been shown to be particularly effective againstthe psychological symptoms of GAD such as worry, tension,irritability, and apprehension, but appears to be less effec-tive in ameliorating somatic symptoms. The onset of actionof this agent is at least two weeks; it may take three to fourweeks before one sees a truly beneficial effect. Buspirone alsorequires divided doses to be effective.

Buspirone has been shown to be ineffective in the treatmentof OCD, SAD, and PD, though it may have benefit as anadjunct to SSRIs in OCD.

Beta-blocker MedicationsBeta-blockers may be useful for individuals who have situa-tional anxiety or performance anxiety. They generally havenot been effective in treating anxiety disorders such as gener-alized social anxiety disorder, panic disorder, or obsessive–compulsive disorder.

These medicines are most useful for treatment of anxietysyndromes and disorders with somatic symptoms associatedwith increased adrenergic tone. Propranolol in doses of 10to 40 mg/day has been used for performance anxiety, onan as-needed basis. Recent research on PTSD suggests adifferent role for the beta-blocker medications. It appearsthat traumatic memories may be reduced by medications that


prevent presynaptic norepinephrine release (such asalpha-2-adrenergic agonists or opioids), or block postsynapticnorepinephrine receptors, blocking a cycle that is mediatedby noradrenergic hyperactivity in the basolateral amygdala.There are two controlled studies of trauma victims presentingto emergency rooms suggesting that post-trauma treatmentwith propranolol can decrease subsequent PTSD symptoms.

AnticonvulsantsRecent studies have suggested efficacy of the anticonvul-sants gabapentin, pregabalin, vigabatrin (not available in theUS), and tiagabine in anxiety disorders. However, furtherstudies are warranted to determine if these medications shouldbe used as monotherapy or as augmenting agents in indi-viduals who are partially or non-responsive to conventionaltherapy.

Gabapentin at doses of 900–3,600 mg/day has been shownto be effective in the treatment of SAD.

Anticonvulsants such as tiagabine, valproic acid, andgabapentin maybe useful in PD with atypical or treatmentresistant features.

In PTSD, open label studies of valproic acid, carbame-zepine, and topiramate and a small double-blind study oflamotrigine have suggested benefit. Further studies are neededto determine whether anticonvulsant treatments will be bene-ficial in treatment of PTSD, either as a monotherapy or as anadjuvant with antidepressants.

AntipsychoticsThe development of the atypical antipsychotics has led toa reemergence of interest in their possible use either as aprimary or adjuvant treatment for a variety of anxiety disor-ders. Small studies of risperidone and olanzapine augmen-tation in PTSD (also, quetiapine) in veterans showed someefficacy, however to date, the data are not overwhelminglyfavorable for the use of atypical antipsychotic medicationas either monotherapy or augmentation therapy for PTSD.


Currently, atypical antipsychotics are most commonly usedin anxiety disorder treatment for augmentation of othertreatments.

Augmentation/Adjuvant TreatmentsThere is evidence that augmentation strategies can be effectivein several anxiety disorders:

• Patients affected by OCD do not always respond tomonotherapy with SSRIs or clomipramine; risperidone(initiated at 1 mg/day, titrating up to 3 mg/day over 2 weeks)and olanzapine (initiated at 2.5 mg/day and titrated up to5 mg/day) may be useful in augmenting the clinical responsein treatment refractory OCD.

• Buspirone may be used as an adjuvant to SSRI treatmentfor social anxiety disorder when patients exhibit only partialresponse to monotherapy.

• Buspirone may also be used adjunctively with a benzodi-azepine in the treatment of PD.

• In PTSD, topiramate augmentation of SSRIs seems to beparticularly useful in improving sleep, decreasing night-mares, and decreasing intrusive thoughts.

• Co-administration of sertraline (100mg/day) and clon-azepam (0.5 mg/day) for PD has shown some efficacy.

DRUG SELECTION, DOSE, AND INITIATION OFTREATMENTThe major initial choice in treatment of anxiety disorders isthe class of medication. Initiation of treatment, dosage titra-tion, and continuation of treatment as well as discontinuationdepend on medication class (see Table 4-2). Even within amedication class (e.g., antidepressants), side effects may vary(see Table 4-4). Abuse liability, delayed onset of action, andprobability of lethality in overdose may play a role in drugchoice. Since anxiety disorders are often chronic, the long-term tolerability of medication is often another key factorin initial treatment choice. Generally single-agent treatment

�TA

BLE

4-4.

Com

mon

Repo

rted

Side

Effe

cts

ofA

ntid

epre

ssan

tM

edic

atio

ns

SSRI

SNRI

MIR

TAZE

PIN

EN

EFA

ZODO

NE

BUPR

OPI

ON

MA

OI

TCA

Nau

sea

SSR

Isi

deef

fect

sIn

crea

sed

appe

tite

Seda

tion

Tre

mor

Dry

mou

thD

rym

outh

Dec

reas

edm

otiv

atio

nD

rym

outh

Wei

ght

gain

Fat

igue

Pal

pita

tion

sB

lurr

edvi

sion

Blu

rred

visi

onIn

som

nia

Con

stip

atio

nSe

dati

onA

taxi

aN

umbn

ess

Inso

mni

aC

onst

ipat

ion

Fat

igue

Agi

tati

onF

atig

ueD

rym

outh

Diz

zine

ssD

izzi

ness

Wei

ght

gain

Eja

cula

tory

dysf

unct

ion

Ano

rgas

mia

Dec

reas

edlib

ido

Wei

ght

gain

Dia

phor

esis

Incr

ease

dbl

ood

pres

sure

Wei

ght

gain

Agi

tati

onC

onst

ipat

ion

Liv

erfa

ilure

Spac

ines

sA

taxi

aSe

izur

es(i

nhi

ghdo

ses)

Ort

host

atic

hypo

tens

ion

Con

stip

atio

nE

jacu

lato

rydy

sfun

ctio

nA

norg

asm

iaA

nxie

tyH

yper

tens

ive

cris

is

Fat

igue

Seda

tion

Ort

host

atic

hypo

tens

ion

Inso

mni

aH

yper

som

nia

Eja

cula

tory

dysf

unct

ion

Init

ial

jitte

rine

ss/

anxi

ety


is preferred initially, often using SSRI medication. BZD orother anxiolytics may be required during the weeks beforeSSRI medication takes effect, but can often be tapered atthat time. Patients who are not fully treatment-responsive mayrequire switches or augmentation of medication. Duration oftreatment generally ranges from several months (for acutedisorders) to a lifetime (for chronic disorders such as OCD).When medication is discontinued in anxiety disorders, thereshould be a gradual taper over many weeks to avoid reboundanxiety and to assess for the return of anxiety symptoms. Theusual daily dose ranges for a variety of agents is shown inTable 4-1. In general, starting doses of 50% of the initialdose shown in this table are used for the first seven days oftreatment.

Variations for some disorders include the following:

• Sertraline for treatment of GAD is usually initiated at25 mg/day for the first week then titrated to 50 mg/day.

• Venlafaxine XR for the treatment of GAD and socialanxiety disorder is usually initiated at 75 mg/day for the firstweek, increasing to 150 mg/day for the second week. Furthertitration to 225 mg/day may be needed.

• Gabapentin has shown efficacy at doses of up to3,600 mg/day for social anxiety disorder.

• For social anxiety disorder, sertraline is usually initiatedat 25 mg/day for the first week, then escalated to 50–150 mg/day.

• Doses of 20–30 mg/day of citalopram seem to be the mosteffective for the treatment of PD.

• Paroxetine is typically dosed at 20 to 50 mg/day for PTSD.• Clinically effective doses for nefazodone in PTSD appears

to be 400–600 mg/day.

SIDE EFFECTSThe common side effects for antipsychotics, SSRIs, mood stabi-lizers, and benzodiazepines are given in Chapters 1–3 and 5respectively.

Although a few individuals starting SSRI treatment mayhave some initial problems with restlessness and increasedanxiety, data suggest that starting at lower doses such as


25 mg/day of sertraline or 10 mg/day of paroxetine maydecrease the risk of antidepressant “jitteriness.” This maybe particularly helpful for patients with panic disorder, whomay be acutely sensitive to the activating effects of SSRIs.Patients may experience the common side effects of SSRIssuch as headaches, nausea, and diarrhea, however, mostdiminish with time and are well tolerated, particularly if thepatients are informed of the possibility of these transient sideeffects.

The most common side effect of benzodiazepines is drowsi-ness, present in approximately 10% of patients. As a result,patients should be cautioned to be careful when driving orusing machinery whilst on these agents. Residual daytime seda-tion often occurs the day after use of medication. Dizzinessand ataxia can be experienced, which can lead to increasedincidence of falls in the elderly.

Other rarer side effects can include:

• Maculopapular rashes and itchiness• Anterograde amnesia (particularly with high-potency

agents)• Mild cognitive deficits

DRUG INTERACTIONSThe common drug interactions for the antipsychotics, SSRIs,and mood stabilizers are given in Chapters 1–3 respectively.Combinations of drugs commonly used in treatment of anxietydisorders may lead to particular interactions. For instance,buspirone combined with other serotonergic drugs can lead toserotonin syndrome.

The most common interaction of the benzodiazepinesoccurs with other CNS depressants such as alcohol, barbitu-rates, tricyclic and tetracyclic drugs, opioids, antihistamines,and dopamine receptor antagonists. When BZDs are takenconcurrently with other sedative agents such as alcohol,barbiturates or analgesics, marked drowsiness, disinhibitionand respiratory depression can occur. Other interactions areshown in Table 4-5.


� TABLE 4-5. Benzodiazepine Drug Interactions

Decreased Absorption Increased MetabolismAntacids CarbamazepineFood

Increased CNS EffectsAntihistaminesNarcotic analgesicsTricyclic antidepressantsAlcoholBarbiturates

Decreased CNS EffectsMethylxanthines (theophylline,

aminophylline, caffeine)

RifampinCorticosteroidsBarbiturates

Decreased MetabolismCimetidineAzole antifungals (ketoconazole,

miconazole, itraconazole)ErythromycinDisulfiramOral contraceptivesNorfluoxetine*FluvoxamineNefazodoneIsoniazid

∗ Major metabolite of fluoxetine.From Fogelman S and Greenblatt DJ (2000) Anxiolytics. In Psychiatric Drugs,Lieberman JA and Tasman A (eds.) WB Saunders, pp. 128–155. © 2000, withpermission from Elsevier.

CONTRAINDICATIONS AND SPECIAL PRECAUTIONSBenzodiazepines may be teratogenic; their use in pregnancy isthus not advised. Their use in the third trimester may precipi-tate a withdrawal syndrome in neonates; in addition, they aresecreted in breast milk in sufficient concentrations to causedrowsiness, bradycardia, and dyspnea in infants. The use ofSSRIs in pregnancy and post-partum are detailed in Chapter 2.

The elderly and patients with hepatic disease are morelikely to experience adverse effects and toxicity to benzodi-azepines due to their reduced metabolism of these agents.Caution should also be used in prescribing these agents forpatients with COPD and sleep apnea, a history of substanceabuse, cognitive disorders, renal disease, CNS depression,porphyria, and myasthenia gravis.

Problems with tolerance, dependence or withdrawal are notusually experienced with short-term use of BZDs at moderatedoses, although the short-acting agents may cause increasedanxiety the day after taking a single dose. Some patients also


� TABLE 4-6. Benzodiazepine Withdrawal Syndrome

Anxiety Irritability Perceptual disturbancesInsomnia Tinnitus DepressionAnorexia Increased

sensitivity tolight and sounds

Autonomic overactivity

Vertigo Headache SeizuresTremor

From Fogelman S and Greenblatt DJ (2000) Anxiolytics. In Psychiatric Drugs,Lieberman JA and Tasman A (eds.) WB Saunders, pp. 128–155. © 2000, withpermission from Elsevier.

experience a tolerance for the anxiolytic effects of benzodi-azepines and require increasing doses to maintain remissionof symptoms. Use of BZDs with longer half-lives (e.g., clon-azepam, or extended-release lorazepam) may decrease theserisks.

Withdrawal syndrome The appearance of this syndromedepends on the length of time the patient has been takinga benzodiazepine, the dosage, the rate at which the drug istapered, and the half-life of the compound. Abrupt discontin-uation of agents with a short half-life is particularly associatedwith severe withdrawal symptoms, including depression, para-noia, delirium, and seizures. Some of these symptoms may occurin up to 50% of patients treated; however, severe symptomsonly occur in patients taking high doses for long periods. Alpra-zolam seems to be particularly associated with immediate severewithdrawal and should be tapered gradually. The symptomsof benzodiazepine withdrawal syndrome are listed in Table 4-6.

SUMMARYClinicians have a wide array of medications available for thetreatment of anxiety symptoms and anxiety disorders. Thebreadth of treatment options available greatly facilitates ourability to help patients. We have safe and effective treatmentsfor everything from the short-term treatment of anxiety reac-tions (such as an acute adjustment disorder with anxiety),to previously intractable and life-long anxiety disorders like


OCD. However, the most important therapeutic agent wepossess is still sound clinical skills and judgement. Appropriatediagnosis and rapport are the foundations of any pharmaco-logical intervention we make with our patients.

ADDITIONAL READING

1. Nutt DJ, Ballenger JC. Anxiety Disorders. Blackwell Publishing,Malden, MA, 2002

2. Pitman RK, Delahanty DL. Conceptually driven pharmacologicapproaches to acute trauma. CNS Spectr. 2005; 10:99–106

3. McDonough M, Kennedy N. Pharmacological Management ofObsessive-Compulsive Disorder: A Review for Clinicians. HarvardReview of Psychiatry, May 2002; 10:127–137

4. Blanco C, Antia SX, Liebowitz MR. Pharmacotherapy of socialanxiety disorder. Biological Psychiatry 2002; 51:109–120

5. Bandelow B, Ruther E. Treatment-resistant panic disorder. CNSSpectrums 2004; 9:725–739

5SEDATIVE–HYPNOTIC

AGENTS

INTRODUCTIONInsomnia is a significant health care problem. The direct andindirect costs of insomnia are enormous for the entire indus-trialized world. Estimates of prevalence vary from 10 to 50%of the adult population, depending on the diagnostic, duration,and severity criteria used. An aging population, hectic workand personal lifestyles, and an increase in the frequency ofnontraditional work hours (e.g., shift work) all may play a role.

Older adults, women and patients with underlying psychi-atric disorders, such as schizophrenia, anxiety and mood disor-ders, are particularly prone to sleep difficulties. Treatmentof the underlying psychiatric illness often improves sleep,obviating the need for hypnotics. However, some of the verymedications used to control the condition could cause insomniaas well. For example, SSRIs may induce insomnia, whileeffectively treating an underlying depression.

DiagnosisNormal sleep consists of alternating episodes of rapid eye move-ment (REM) sleep and non-REM sleep. Non-REM sleep isfurther divided to 4 stages (stage 1, a brief transition between



wakefulness and sleep; stage 2, accounting for most of the timein non-REM sleep, features spindles and K-complexes on theEEG; stages 3 and 4, also called delta sleep, characterized byhigh amplitude, slow, delta waves on EEG, is thought to corre-spond to restorative sleep). Sleep latency, typically prolongedin insomnia, refers to the time it takes to fall asleep once inbed. In general terms, insomnia refers to the real or perceivedinability tosleep,andpresentsasdifficulty initiatingand/ormain-taining sleep.

Primary insomnia is diagnosed in the absence of any othersleep disorder (e.g., narcolepsy, obstructive sleep apnea), orthose due to medical conditions or to medication or substanceuse. To formally make a diagnosis of primary insomnia, theDSM-IV-TR requires that the sleep disturbance last at least onemonth and that it causes significant interference in emotional,social, or occupational functioning. Polysomnography or actig-raphy are not indicated for the routine evaluation of acute orchronic insomnia, but may be useful when sleep disorders such asobstructive sleep apnea or nocturnal myoclonus are suspected.

The more common secondary insomnia may be due to avariety of conditions such as hyperthyroidism, chronic pain,severe chronic obstructive pulmonary disease (COPD), rest-less legs syndrome, and/or medication (e.g., stimulants) orsubstance (e.g., cocaine, caffeine) use. Insomnia is also verycommon in patients with mood, anxiety, and psychotic disor-ders. Given the frequent co-occurrence, and that the directionof causality between insomnia and other conditions remainsunclear, the 2005 NIH State of the Science Conference State-ment concluded that secondary insomnia would be namedmore appropriately as “comorbid insomnia.”

Although total sleep time could be increased in theelderly, the total amount of time spent in stages 3 and 4sleep is reduced. The most common complaints of olderpatients, frequent awakening and the inability to sustain sleepthroughout the night, results in less restorative sleep.

Treatment OptionsTreatment must be based on careful diagnostic assessment,including interview with a bed partner, review of sleep logs,

SEDATIVE–HYPNOTIC AGENTS 157

and a diligent search for possible causes of secondary insomnia.Severe, chronic insomnia may only respond to medications,but non-pharmacological options should always be exploredfirst. When indicated, hypnotic use is generally recommendedfor short-term use only. However, realistic alternative treat-ments to address the needs of patients suffering from chronicinsomnia are not available. For now, basic sleep hygiene tech-niques, as summarized in Table 5-1, must be part of all treat-ments of insomnia.

Non-prescription AgentsOver-the-counter (OTC) sleep aids most often contain thehistamine �H1� receptor antagonist diphenhydramine or some

� TABLE 5-1. Sleep Hygiene Techniques

1. Do not go to bed when you are not tired.2. Avoid napping during the day, even when you are tired. Especially

avoid early-evening naps.3. Wake up at the same time each day. Do not “catch up” on your

sleep on the weekend.4. Do not drink caffeinated beverages within 6 h of bedtime and

minimize total daily use.5. Avoid heavy meals too close to bedtime, but don’t go to bed hungry,

as this may disrupt sleep. A warm, noncaffeinated beverage and asmall carbohydrate snack may be soothing and enhance drowsiness.

6. Regular exercise in the late afternoon may deepen sleep. Strenuousexercise too close to bedtime (i.e., 4–5 h) may interfere with sleep.

7. If you do not fall asleep within a half-hour or so, get up and read abook or watch television. When you stay in the bed and don’t sleepfor long periods of time, the bed becomes associated with notsleeping.

8. Minimize noise, light, and extremes in temperature during the sleepperiod.

9. Avoid the use of alcohol as a sleep-enhancer. Although it maypromote sleep onset, early morning awakening is quite common andthe sleep is generally nonrestorative in nature.

10. Use progressive muscle relaxation or deep-breathing techniques toenhance relaxation and minimize anxiety and stress.

11. Ensure that pain complaints have been appropriately evaluated andtreated by your physician. Pain interferes with sleep.

12. Do not smoke cigarettes too close to bedtime or if you awaken in thenight. Nicotine is a stimulant.


other sedating antihistamine such as doxylamine. They tend tohave a prolonged duration of action leading to sedation andslowed reaction times the day after their use. Their side effectsare substantial, including urinary retention, blurred vision,orthostatic hypotension, elevated liver enzymes etc. Tachyphy-laxis often develops within several days to 1–2 weeks, limitingtheir utility to short-term insomnia problems. Unless pain isa significant complaint, the common practice of combining ananalgesic such as aspirin or acetaminophen and an antihis-tamine (e.g., Tylenol PM) is no more effective than the use ofan antihistamine alone.

Melatonin is a naturally occurring pineal gland peptidehormone available in OTC formulations. The FDA classifiesit as a nutritive or dietary supplement. As dietary supplementsare not reviewed by the FDA, the strength and purity of mela-tonin cannot be guaranteed. When taken orally, melatoninalters circadian rhythms, lowers core body temperature, andreduces daytime alertness. Melatonin may be particularly effec-tive when the normal circadian cycle is disrupted (e.g., jetlag, shift work). Also unregulated, valerian may alleviate mild,short-term insomnia, but safety and efficacy have not beenestablished.

Prescription MedicationsBenzodiazepines (BZDs) (triazolam, temazepam, estazolam,flurazepam, quazepam), and more recently, a series of non-BZD type hypnotics, or “Z-drugs” (zolpidem, zaleplon, es/zopiclone), have replaced most older hypnotics such as barbi-turates, glutethimide, methaqualone, methyprylone, meproba-mate, and tybamate. The replaced hypnotics were highlyaddicting and, due to their low therapeutic index, were alsodangerous in overdose. In contrast, BZDs, and particularlythe Z-drugs, are characterized by decreased abuse potential,fewer drug-drug interactions, broader therapeutic index, andlower risk in overdose. Daytime sedation and the risk for abuseand physiological dependence are the least likely complica-tions of Z-drug use. BZDs and the Z-drugs are currently themost frequently prescribed hypnotics, but the off-label use(FDA-approved drug use for not FDA-approved indication)


of sedating antidepressants (e.g., trazodone) as sleep aids, iscommon clinical practice. The melatonin agonist ramelteon,the most recently FDA-approved hypnotic in the US, is anagent with unique mechanism of action. The pyrazolopy-rimidine indiplon, not yet available in the US, is also anon-BZD GABA/A modulator with promising side-effectprofile. Other pharmacological options, limited to a few well-defined conditions with many caveats, include anticonvulsants(e.g., gabapentine, valproate, tiagabin) and antipsychotics (e.g.,haloperidol, quetiapine).

PHARMACOLOGYBenzodiazepinesAll BZDs are central nervous system depressants via gamma-aminobutyric acid (GABA) agonism. Due to non-selectiveGABA binding, BZDs also possess anxiolytic, anticonvul-sant, and myorelaxant properties. By modulating the effectsof GABA, BZDs increase the frequency of chloride channelopenings. In contrast, barbiturates and alcohol increase theduration of chloride channel opening. This seemingly minordistinction accounts for the greater safety of BZDs in over-dose, with less likelihood of respiratory depression or coma.When alcohol is mixed with a BZD overdose, the synergismmay result in fatal respiratory depression. Although effectivehypnotics, BZDs significantly alter normal sleep architec-ture; daytime fatigue is probably related to reduced slow-wave, restorative sleep, while the memory problems reportedby BZD users may be due to reduced REM sleep. REMsuppression causes REM rebound upon BZD discontinuation,manifesting as vivid dreams and nightmares.

Chloral HydrateDespite a rapid onset of action and relatively short elimina-tion half-life, chloral hydrate, one of the oldest hypnotics, israrely used as a hypnotic agent today, primarily due to itsnarrow therapeutic index. At dosages between 0.5 and 1.5 g, itis usually effective within 30 minutes. Chloral hydrate is metab-olized in the liver and kidneys, and excreted renally. Chloral


hydrate generally increases sleep stages 2 and 4, decreasesstage 3 sleep, and has no effect on REM sleep.

ZolpidemThe imidazopyridine zolpidem, just like all Z-drugs, is a BZD-like partial GABA-agonist with a non-BZD-like chemicalstructure. It is active at the � receptor, which is a subunit ofthe GABAA receptor. BZDs activate all � receptor subtypes.Zolpidem appears to bind preferentially to �1 receptors.Although this selective binding is not absolute, it may explainthe relative absence of myorelaxant, anxiolytic, and anticon-vulsant effects at hypnotic dose, as well as the preservation ofstages 3 and 4 sleep. Polysomnography indicates that zolpideminduces a sleep pattern similar to that of physiological sleep,and produces little disruption of sleep architecture followingabrupt discontinuation. Zolpidem is rapidly absorbed throughthe gastrointestinal tract, reaching peak concentration from30 minutes to 2 hours after administration. It is metabolized inthe liver to several inactive metabolites and has an eliminationhalf-life of approximately 2.5 hours.

ZaleplonZaleplon, of the pyrazolopyrimidine class, also a non-BZD,binds preferentially to the �1, �2, and �3 subunits of theGABA/A receptor. Peak serum concentration occurs withinone hour of ingestion. Absolute bioavailability is only 30%as it undergoes extensive hepatic first-pass metabolism. Zale-plon is primarily metabolized by aldehyde oxidase and formsa number of inactive metabolites. Its elimination half-life isapproximately one hour, making it an ideal hypnotic wheninsomnia is primarily due to the difficulty of falling asleep.

EszopicloneThe cyclopyrrolone eszopiclone, the stereoselective isomer ofzopiclone, is the most recently (2004) marketed non-BZD“Z-drug” in the US. Due to rapid absorption and a half-lifeof six hours, it is indicated for both sleep initiation and sleep


maintenance. Daytime residual somnolence and memory prob-lems are much less likely with zopiclone, and probably witheszopiclone as well, compared to BZDs. Eszopiclone is theonly hypnotic approved for long-term administration, basedon a large, controlled treatment study that showed no toler-ance to its hypnotic effect after six months of continuous use,followed by 12 months of open extension. Sleep architecture isonly minimally altered; stage 2 and slow-wave sleep increase,REM remains unchanged. Polysomnography shows continuedsleep improvement after drug discontinuation with no signifi-cant rebound insomnia.

RamelteonThis most recently FDA-approved hypnotic in the US, is anovel agent with high selectivity for the melatonin receptorsMT1 and MT2. These receptors, located in the suprachiasmaticnucleus, have been implicated in the regulation of sleepinessand the sleep-wake cycle. Therefore, ramelteon may improvesleep through the endogenous sleep regulating system. Withrapid absorption and a half-time of 90 minutes, ramelteonreduces sleep latency and increases total sleep time, with noresidual sedation. Ramelteon does not appear to alter sleeparchitecture, and given the absence of significant side effects,even at significantly higher than recommended doses, its ther-apeutic margin seems wider compared to most hypnotics. Ifthese reported properties, including efficacy, are confirmedin clinical practice, ramelteon should become the preferredfirst-line choice for the treatment of insomnia.

DRUG SELECTIONIn general, choosing the most appropriate hypnotic should bebased on the type of sleep difficulty being experienced, the ageof the patient, comorbid diagnoses, medical and psychiatrichistory, and concomitant medications used.

BZDs are effective for transient and situational insomnias,with reported improvement in sleep onset, number of awaken-ings at night, total sleep duration, and the quality of sleep. Asdoses, and consequently brain concentrations are increased,


drowsiness and relaxation shift into decreased wakefulness andthen sleep.

Although only five benzodiazepines are marketed ashypnotic agents in the US (Table 5-2), most BZDs inducesleep, provided an appropriate dose is chosen. There is noconvincing evidence that marketed benzodiazepine hypnoticsdiffer in terms of efficacy or safety when they are administeredappropriately. They are similar in their effects on sleep archi-tecture and differ only in onset and duration of action. Tria-zolam has a rapid onset and a short duration of action, makingit appropriate for patients with sleep initiation difficulties,while flurazepam, with a longer onset of action and a muchlonger duration of action, is a better choice for patients withmiddle or terminal insomnia.

Shorter half-life BZDs cause less daytime sedation andresidual cognitive effects, but may be associated with increaseddaytime, particularly morning, anxiety. Rebound insomniaupon abrupt discontinuation is common, especially after usefor several consecutive nights, but can be avoided by gradualtaper. The likelihood of rebound insomnia seems greater inpatients who experience greater hypnotic efficacy. For at leastsome patients, this may reflect the development of physiologicdependence and a withdrawal phenomenon.

Longer half-life benzodiazepines, preferred for middle orterminal insomnia, often cause daytime sedation, decreasedreaction time, and/or impaired coordination. Flurazepam andquazepam, with long half-life active metabolites, often accumu-late with repeated administration, particularly in the elderly.Patients with significant daytime anxiety may benefit from thelonger acting benzodiazepines, as the residual amount of medi-cation left in the morning may still have anxiolytic benefits.BZDs are the preferred hypnotics for patients with insomniaand comorbid anxiety disorders.

BZD hypnotic use by the elderly, likely due to the greaterprevalence of sleep disorders, is much more common thanthat in the general adult population. The elderly, as a group,have slower hepatic metabolism and less efficient renal excre-tion which is often more pronounced among men. In coordi-nation, sedation and confusion can occur when longer-actingBZDs with active metabolites build up over time. Compared

�TA

BLE

5-2.

Hyp

notic

Age

nts

GEN

ERIC

NA

ME

TRA

DEN

AM

EDO

SEST

REN

GTH

S(m

g)H

ALF

-LIF

E(h

)O

NSE

TO

FA

CTIO

NCO

MM

ENTS

Chl

oral

hydr

ate

Noc

tec

500m

g/5m

l8−

10In

term

edia

teA

ctiv

em

etab

olit

etr

ichl

oroe

than

olha

sC

NS

depr

essi

veef

fect

Tri

azol

amH

alci

on0.

125,

0.25

1�5−

5�5

Inte

rmed

iate

May

caus

ean

tero

grad

eam

nesi

aT

emaz

epam

Res

tori

l7.

5,15

,30

9−12

Slow

No

acti

vem

etab

olit

esE

staz

olam

Pro

Som

1,2

10−2

4In

term

edia

teA

ctiv

em

etab

olit

eha

sw

eak

seda

ting

prop

erti

esas

wel

lF

lura

zepa

mD

alm

ane

15,3

030

−100

Fas

tA

ctiv

em

etab

olit

ede

salk

ylfl

uraz

epam

accu

mul

ates

and

caus

esad

diti

vese

dati

onQ

uaze

pam

Dor

al7.

5,15

25−4

0In

term

edia

teSa

me

acti

vem

etab

olit

eas

flur

azep

amE

szop

iclo

neL

unes

ta1,

2,3

6F

ast

Zal

eplo

nSo

nata

5,10

1F

ast

Zol

pide

mA

mbi

en5,

101�

4−4�

5F

ast

Ram

elte

onR

ozer

em8

1−2�

6Sl

owU

niqu

eM

OA

mel

aton

inre

cept

orsu

btyp

eag

onis

t


to short-acting BZDs, much higher rates of falls and hip frac-tures have been observed in older patients taking long-actingBZDs; this risk increases in direct proportion to the dailydose. A generally safe clinical strategy when managing late-lifeinsomnia is to halve the starting dose and titrate up slowlyas tolerated following the longstanding advice “start low, goslow.” Avoiding long-acting benzodiazepines appears prudentas well.

At present, conclusive evidence does not support clinicallysignificant differences among the three available Z-drugs.Consequently choosing one over the others remains some-what arbitrary. Quick onset of action and short eliminationhalf-life may make zaleplon more appropriate for patients withdifficulty in initiating sleep only. If maintaining sleep is alsoproblematic, zaleplon’s short half-life permits middle of thenight or early morning administration without significant next-day sedation as well. Zolpidem and eszopiclone seem compa-rable in efficacy and in onset and duration of action. Both arelikely to address difficulties in initiating and maintaining sleep.Compared to BZDs, residual daytime fatigue and somnolenceare less pronounced during continuous use; rebound insomniais rare upon discontinuation.

Ramelteon appears to least alter the intrinsic sleep archi-tecture and therefore, if efficacy is established using activecomparators, it may become the “ideal” hypnotic.

Other Prescription HypnoticsThese agents have not been approved by the FDA forinsomnia, but due to their sedating properties, they are consid-ered useful alternative hypnotics in cases of treatment resis-tance, intolerable side effects, drug-drug interactions, certaincomorbid conditions, current or past history of drug use etc.

At present, the most frequently used, non-FDA-approvedhypnotic in the US is the triazolopyridine trazodone, marketedas an antidepressant. It is rarely used alone as an antidepres-sant because of its strong sedating qualities and the need forb.i.d. administration. Tolerance to its sedating effects developsonly rarely with long-term use, making it an excellent optionfor those with chronic insomnia. In spite of a paucity of data in


support of efficacy, trazodone is also the most popular choiceto counter SSRI-induced insomnia. Hypnotic dosages varymarkedly, from 25 to 300 mg, depending on individual suscep-tibility to its sedating effects. Trazodone increases slow wavesleep and total sleep time and does not appear to affect REMsleep. Its elimination half-life of between six and nine hoursrenders trazodone likely to cause daytime drowsiness. Usingthe lowest effective dosage and/or taking it in late eveningrather than at bedtime may minimize this effect. Side effectsalso include hypotension and constipation, but priapism isexceedingly rare, occurring in less than one in 40,000 cases.Since anticholinergic side effects are not common, trazodonehas some advantages over TCAs.

Small doses of sedating TCAs (e.g., amitriptyline, doxepin)are sometimes used as hypnotics. Patients with chronic painmay particularly benefit from this treatment strategy; TCAs arefrequently used to potentiate the benefits of traditional painmedications. They are potent inhibitors of REM sleep, andtheir discontinuation is often associated with REM rebound.Anticholinergic and adrenergic side effects and toxicity in over-dose must be weighed against potential benefits. Mirtazapine isa newer sedating antidepressant that targets both the noradren-ergic and serotonergic systems. It may be a useful alternativeto the tricyclic antidepressants as it is less anticholinergic andless toxic in overdose. Mirtazapine has strong antihistaminicactivity that may cause significant weight gain with chronic use.

Hydroxyzine hydrochloride and hydroxyzine pamoate aresedating H1 receptor antagonists that are used occasionally ashypnotic agents. Next-day sedation is a common problem withantihistamines as they have a relatively long elimination half-life. They are highly anticholinergic and may cause hypoten-sion; they are not recommended as first-line agents for thetreatment of insomnia.

Barbiturates, butabarbital, phenobarbital, and secobarbital,are still prescribed for insomnia, but mostly outside the US.Barbiturates are potentially lethal in overdose due to respi-ratory depression; they are highly addicting and their abruptdiscontinuation may result in potentially fatal withdrawalreaction. Their use has generally been replaced by safer agentswith much better side effect profiles.


Certain sedating anticonvulsants are reasonable third andfourth line sleep aids or add-on hypnotics for patients who donot respond or become tolerant to standard hypnotics. Bipolarpatients suffering from significant insomnia may benefitfrom valproate, when BZD-type hypnotics are contraindi-cated, gabapentine or tiagabine might be good alternatives.Secondary insomnia due to restless legs syndrome may prefer-entially respond to dopaminergic drugs, such as pramipexoleor levodopa. As a last resort, agitated elderly patients withdementia may preferentially respond to low dose sedatingantipsychotics such as haloperidol or quetiapine. Although thenew generation atypical antipsychotics are less likely to inducetardive dyskinesia, other complications like diabetes, obesity,and dyslipidemia suggest that the risks of using antipsychoticsas hypnotics remain substantial.

TREATMENT IMPLEMENTATIONIt is always advisable to provide patients with information ongood sleep hygiene practices (see Table 5-1). Although behav-ioral techniques such as stimulus control, progressive musclerelaxation, biofeedback, and sleep-restriction seem to comple-ment pharmacotherapy, cognitive-behavioral therapy (CBT)alone may have better long-term outcome, compared to thecombination of CBT and medications. In contrast, added CBTseems to facilitate the tapering of hypnotics.

When a hypnotic agent is prescribed, patients should beadvised to take the medication on an empty stomach and withample fluids (e.g., a full glass of water) to promote rapid onsetof effect. For patients prone to nocturia, fluid intake should belimited during the hours before bedtime.

The clinician should always caution patients regardingpossible side effects:

• potential impairments in memory, coordination, or drivingskills

• unsteadiness if they are awakened after having taken asleep aid

• tapering of medication before discontinuation; this will berequired if medication is used for more than a few nights


• avoiding the use of alcohol when taking a hypnotic, as theeffects are additive

• the propensity for medication abuse by patients with drugor alcohol problems

Universally accepted guidelines for dosing and duration of usefor hypnotics are not established. Both dose and duration mustbe individualized with the goal of finding the lowest dose andthe shortest duration. Short-term treatment (i.e., from one ortwo nights to one or two weeks) is reasonable for most patients.

Some patients with chronic insomnia may benefit fromlonger-term use, provided that there is careful monitoring bythe prescribing physician. No criteria are presently availableto identify this subpopulation. It seems reasonable to considerseveral short-term trials, with gradual tapering at the end ofeach period and a drug-free interval between each period, toestablish the patient’s need for and the appropriateness andvalue of continued therapy. Drug-free time intervals betweeninitial treatment periods should range from one to three weeks,depending on the half-life of the agent and its active metabo-lites and the rapidity of the taper schedule. Re-evaluation ofsuch a patient’s continued need for hypnotic medication at3- to 6-month intervals is also reasonable.

As the elderly are particularly susceptible to falls or confu-sion from hypnotic medication, use of the lowest availabledosage strength is advisable. The elderly should also avoid theuse of longer half-life agents or those with active metaboliteswith long half-lives because such medications tend to accumu-late due to pharmacodynamic differences in drug metabolismin the elderly.

For individuals who prefer behavioral and nonpharmaco-logical approaches to their insomnia, hypnotic use two orthree times per week may be beneficial.

ADVERSE EFFECTSBenzodiazepinesShorter half-life hypnotics, particularly triazolam, appear tocause anterograde amnesia more commonly compared to


longer half-life agents. Alarming reports in the lay pressregarding a greatly increased risk of hallucination, confusion,and anterograde amnesia with triazolam use are not supportedby all available data. All benzodiazepines can cause antero-grade amnesia, particularly at higher dosages. Anterogradeamnesia commonly occurs following benzodiazepine overdose.The prevalence of and the risk factors for anterograde amnesiawith appropriate dosing of sedative–hypnotic agents remainto be determined. Other adverse effects are discussed inChapter 4.

Chloral HydrateChloral hydrate has a narrow therapeutic index, causes gastricirritation, nausea, and vomiting, and, at high doses, maycause gastric necrosis. Overdose can be fatal due to respi-ratory depression. As with other sedative–hypnotics, depen-dence and tolerance may develop. Withdrawal symptoms upondiscontinuation, particularly following prolonged use, includeconfusion, hallucinations, stomach pain, and severe anxiety.Long-term use may lead to induction of hepatic microsomalenzymes as well. Any concomitant medication with significantsedating properties should be used with extreme caution, asthe effects are additive. When chloral hydrate is taken alongwith alcohol the effect can be supraadditive (as in a MickeyFinn or knockout drops).

ZolpidemThe elimination half-life of zolpidem is prolonged in theelderly and in patients with impaired hepatic or renal function.Zolpidem is a partial CYP 3A4 substrate. Zolpidem overdosecan cause respiratory depression or coma, especially whencombined with other CNS depressants. The benzodiazepineantagonist flumazenil can reverse the effects of an overdoseof zolpidem, reflecting its benzodiazepine-like mechanism ofaction.

Although zolpidem is classified as a schedule IV drug by theFDA, it appears to cause tolerance and withdrawal syndromessomewhat less frequently than benzodiazepine hypnotics do.Withdrawal symptoms occur more frequently if doses exceed


10 mg, the recommended maximum in the Physician’s DeskReference (PDR). Selectivity for �1 receptors appears to be lostat higher-than-standard doses, cross tolerance with alcohol andbenzodiazepines develops, and adverse effects are much morecommon. Anecdotal reports of hallucinations and confusionat standard hypnotic dosages should be evaluated across aspectrum of dosages and age groups.

ZaleplonLike zolpidem, this agent is classified as a schedule IV drugby the FDA. Higher than standard dosages have been asso-ciated with an abuse potential similar to that of triazolam.With dosages of 20 mg or less, abuse and dependence appearto be significantly less than that found with BZDs. Reboundinsomnia has been reported, but it occurs less frequently thanwith short acting BZDs such as triazolam.

EszopicloneGiven its relatively recent FDA-approval, eszopiclone’s sideeffect profile is yet to be assessed in clinical practice. Avail-able information is largely based on its similarity to the mixedisomer zopiclone. Most similar to zolpidem, eszopiclone alsoseems to be associated with minimal daytime residual somno-lence and significantly less rebound insomnia upon discontin-uation compared to BZDs. In fact, elderly insomniacs treatedwith zopiclone demonstrated improved cognitive functioningin one study. Dosed between 1 and 3 mg, there is no evidenceof tolerance or addiction after several weeks of continuoususe. Since it lacks the respiratory depressant effects of BZDs,eszopiclone is not contraindicated in COPD.

RamelteonA comprehensive and reliable assessment of the side-effectprofile of ramelteon is also limited by limited clinical expe-rience. Given its unique mechanism of action and seeminglyleast propensity to alter intrinsic sleep architecture, side effectsshould be minimal. The apparent absence of side effects atsignificantly higher than the recommended effective dose isparticularly promising.


DRUG INTERACTIONS AND SPECIAL PRECAUTIONSDrug interactions with benzodiazepines are common. Whenmixed with other sedating compounds the effects are additive.Narcotic medications, alcohol, and antihistamines are exampleswhere the additive sedation can lead to significant confusion.Alcohol ingestion slows hepatic metabolism and causestransiently higher concentrations of oxidatively metabolizedbenzodiazepines leading to further sedation. Antacids thatslow gastric emptying may decrease the rate of absorption of ahypnotic agent from its primary site, the small intestine. Thiscould alter both peak concentration and onset of effects.

Benzodiazepines all undergo some form of hepaticmetabolism, although some agents are much more extensivelymetabolized than others. Medications that inhibit hepaticmicrosomal oxidative metabolism may cause clinically mean-ingful drug interactions when combined with benzodiazepines.All currently available triazolobenzodiazepines (e.g., alpra-zolam, estazolam, midazolam, triazolam) and diazepam aremetabolized by hepatic microsomes and are complete or partialsubstrates of the CYP 3A4 isoform. Medications that arepotent inhibitors of this system will cause higher peakconcentrations of the triazolobenzodiazepine, particularly tria-zolam, which is highly hepatically metabolized. Nefazodone,ketaconazole, cimetidine, and macrolide antibiotics are exam-ples of clinically relevant CYP 3A4 inhibitors which may causea reduction in the clearance of these triazolobenzodiazepinesand an increase in their blood and brain concentrations. Thisincrease is greatest with higher hepatic clearance drugs suchas triazolam or midazolam and somewhat less for drugs suchas alprazolam or estazolam.

The CYP 1A2 isoform is also important in insomnia butfrom a different perspective. Methylxanthines (e.g., caffeine)and certain bronchodilators, such as theophylline, are signifi-cantly metabolized via CYP 1A2-mediated pathways. Inhibi-tion of CYP 1A2 activity by the SSRI fluvoxamine or by certainfluroquinolone antibiotics (e.g., ciprofloxacin) in patients alsoingesting caffeinated beverages or in those requiring theo-phylline may result in insomnia, agitation, or increased anxiety.

All benzodiazepine hypnotic agents are FDA PregnancyCategory X, meaning their use should be avoided during


pregnancy, especially during the first trimester due to anincreased risk of congenital malformations. A variety ofcongenital malformations, including cleft palate, delayed ossi-fication of a number of bony structures, and an increasedoccurrence of rudimentary ribs have been reported. Chronicusage of benzodiazepines during pregnancy may lead to phys-iologic dependence and withdrawal symptoms in the neonate.The use of benzodiazepine hypnotics during the last weeksof pregnancy may result in neonatal central nervous system(CNS) depression, and their use during labor may lead toneonatal flaccidity. Nursing also poses significant hazards, asall benzodiazepines are distributed into breast milk to varyingdegrees. Since infants metabolize benzodiazepines more slowlythan adults, accumulation may occur, leading to sedation andnursing difficulties in the infant. Pregnant women who haveinsomnia should be reminded of sleep hygiene techniquesand given any hypnotic agent only with extreme caution.Diphenhydramine can be used with relative safety, but onlyfor short periods of time since tachyphylaxis develops withprolonged use.

Benzodiazepines must be used in patients with obstruc-tive sleep apnea only with extreme caution. Benzodiazepinesmay cause respiratory depression and can render patients lesslikely to mount an appropriate respiratory response to hypoxia.Zolpidem, zaleplon, or trazodone are less likely to cause prob-lems for sleep apnea patients and may be preferable alter-natives. Aggressive evaluation and treatment of sleep apnea[e.g., continuous positive airway pressure (CPAP)] is extremelyimportant to ensure adequate restorative sleep.

Pharmacodynamic interactions between the Z-drugs andother medications are relatively rare, but cumulative seda-tive effects may occur in the presence of CNS depressants(ethanol, TCAs etc.). Compared to some BZDs that are almostentirely metabolized via CYP3A4, the Z-drugs are biotrans-formed by several CYP isoenzymes. Therefore, pharmacoki-netic interactions with CYP3A4 inhibitors and inducers are lessrelevant. For instance, phamacokinetic interactions, involvingthe CYP-450 enzyme system, seem to be limited to increasedplasma concentration of zopiclone when co-administered withcarbamazepine or erythromycin and decreased plasma levels


in the presence of rifampicin. Zaleplon has similarly uncom-plicated drug interaction profile, as it is metabolized by alde-hide oxidase. Since cimetidine inhibits aldehyde oxidase andCYP 3A4, and may increase zaleplon plasma concentrationsby 85%, the initial starting dose of zaleplon should be halvedin patients also taking cimetidine. Other potent inhibitors ofthe CYP 3A4 system such as ketoconazole and erythromycinalso may increase zaleplon levels. In contrast, rifampin seemsto increase zaleplon clearance.

SUMMARYInsomnia is common and disabling. Medical and psychiatricconditions are frequently associated with insomnia resultingin added impairment. Treatment of insomnia should beguided by careful assessment and needs to be individu-alized based on the specific needs of the patient. Effec-tive non-pharmacological treatments are available and shouldalways be considered first. Pharamcotherapeutic options haveexpanded dramatically over the past 30 years. Older agentshave been gradually replaced by safer, more effective andbetter tolerated hypnotics. Benzodiazepines and a series ofnon-benzodiazepine hypnotics acting on specific subunits ofthe benzodiazepine GABA receptor are currently the mostfrequently prescribed hypnotics in the United States. Anumber of other medications with prominent sedating prop-erties but FDA-approved indications other than insomnia arealso available and safely complement the benzodiazepine-typecompounds. Improved quality of sleep, fewer side effects,fewer drug-drug interactions, and the preservation of theintrinsic sleep cycle remain the primary goals of hypnotic-sedative drug development for the management of insomnia.At present, over 80% of those with insomnia problems arelikely to respond to treatments.

ADDITIONAL READING

1. Silber MH: Chronic insomnia. N Engl J Med 2005; 353:803–102. Roth T, Krystal A, Walsh J et al: Twelve months of nightly eszopi-

clone treatment in patients with chronic insomnia: assessment oflong-term efficacy and safety. Sleep 2004; 27:A260


3. Buysee DJ, Reynolds CF, Kupfer DJ et al: Clinical diagnoses in 216insomnia patients using the International Classification of SleepDisorders (ICSD), DSM-IV and ICD-10 categories: a report fromthe APA/NIMH DSM-IV Field Trial. Sleep 1994; 17:630–37

4. Morin CM, Colecchi C, Stone J et al: Behavioral and pharmacolog-ical therapies for late-life insomnia: a randomized controlled trial.JAMA 1999; 281:991–99

5. National Institutes of Health State of the Science Conference State-ment: Manifestations and Management of Chronic Insomnia inAdults. June 13–15, 2005. Sleep 2005; 28(9)

6PSYCHOSTIMULANTS

INTRODUCTIONThe psychostimulants methylphenidate, amphetamines, andatomoxetine reduce the core symptoms of childhood attention-deficit/hyperactivity disorder (ADHD) in approximately 70%of patients compared with 13% taking placebo. Short-termefficacy is more pronounced for behavioral rather than cogni-tive and learning abnormalities associated with ADHD. Thereis strong evidence for the efficacy of treatment in schoolage children but there is new evidence for efficacy in thetreatment of ADHD in preschoolers, adolescents, and adultsbecause of new large-scale randomized clinical trials for thoseage groups. Although psychostimulants are clearly effective inthe short term up through 14 months, concern remains thatlong-term benefits over years have not yet been adequatelyassessed.

PHARMACOLOGYChemistryDextroamphetamine (DEX) and methylphenidate (MPH) aresomewhat similar in chemical structure. Atomoxetine, thenewest agent in this class, is structurally unrelated to theamphetamine derivatives.



• MPH (methyl-a-phenyl-2-piperidineacetate hydrochloride)has two asymmetric carbon atoms, resulting in four opticalisomers: both d- and l-forms of the threo- and erythro- race-mates. The threo- isomer is biologically more potent than theerythro- structure. The commercial manufacturing processproduces the d� l-threo-MPH racemate exclusively.

• Amphetamine, (a-methylphenethylamine) is available inboth d-amphetamine and l-amphetamine forms, as wellas racemic mixtures. The DEX isomer appears to havemore potent CNS effects and is preferentially used in mostindications.

• Atomoxetine (N-Methyl-3-phenyl-3-(o-tolyloxyy)-propyla-minehydrochloride) is also structurally unrelated to otheragents in this class; it is formulated as the R�-� isomer.

Mechanism of ActionThe term psychostimulant refers to the ability of thesecompounds to increase CNS activity in some but not all brainregions. For example, while increasing the activity of striatumand connections between orbitofrontal and limbic regions,stimulants seem to have an inhibitory effect on the neocortex.Prominent central effects include activation of the medullaryrespiratory center and a lessening of central depression frombarbiturates.

• The blockade of the dopamine transporter has been cited asthe putative mechanism by which psychostimulants amelio-rate ADHD symptoms.

• DAT blockade is now regarded as the putative mecha-nism for psychostimulants: radioligand binding studies havedemonstrated the direct action of psychostimulants, partic-ularly MPH, on striatal DAT.

• Preliminary data suggest a role for 5-HT mechanisms insome components of the ADHD syndrome, even thoughSSRIs have not been helpful in treatment.

• Postsynaptically, stimulants are direct agonists at the adren-ergic receptor.

• They also block the action of a degradative enzyme,catechol-o-methyltransferase (COMT).

PSYCHOSTIMULANTS 177

• DEX and MPH are thought to act differently inside thepresynaptic neuron.

• Atomoxetine is a selective inhibitor of the presynaptic nore-pinephrine transporter, with no direct effect on serotoninor dopamine and minimal affinity for other norepinephrinereceptors.

• Peripherally, sympathomimetics have potent agonist effectsat alpha- and beta-adrenergic receptors: DEX stimulatescardiac muscle, raising systolic and diastolic blood pressure,with a reflex slowing of heart rate.

• Psychostimulants can also cause urinary bladder smoothmuscle to contract at the sphincter and will increase uterinemuscle tone and produce bronchodilatation.

PharmacokineticsThe pharmacokinetic characteristics of the stimulants aresummarized in Table 6-1. Because MPH’s short half-lifeprevents it from reaching steady state in the plasma, the stan-dard tablet must be given several times a day to maintainbehavioral improvement throughout the school day; a long-acting formulation is therefore preferred.

Atomoxetine has an elimination half-life of four hours inthe efficient metabolizers (90% of the population). The phar-macodynamic reaction at the receptor, however, can last for24 hours. The half-life of atomoxetine is greatly increased in5–10% of patients who have a polymorphism at the cytochromeP450 2D6 isoenzyme that makes them poor metabolizers.

Pharmacokinetic studies show that the sustained-releasepreparations have properties different from those of theimmediate-release (IR) tablets. The time course of peakplasma concentrations from the long-duration preparationdepends greatly on the design of the capsule. OROS MPH hasa slowly rising concentration over eight hours. The “beaded”long duration preparation, whether they involve MPH oramphetamine, provides an immediate peak and a second stim-ulant peak in plasma four hours later. A single daily dose of asustained release formulation may therefore be equivalent tomultiple doses of an IR preparation.

�TA

BLE

6-1.

Phar

mac

okin

etic

san

dPh

arm

acod

ynam

ics

PARA

MET

ERDE

XTRO

AM

PHET

AM

INE

MET

HYL

PHEN

IDA

TEA

DDER

ALL

STRA

TTER

A

Met

abol

ism

Hep

atic

toin

acti

vem

etab

olit

esE

xtra

hepa

tic

toin

acti

vem

etab

olit

esH

epat

icto

inac

tive

orw

eakl

yac

tive

met

abol

ites

Hep

atic

tolo

wle

vels

ofac

tive

met

abol

ite

Exc

reti

onR

enal

,of

met

abol

ites

;ra

teac

cele

rate

dby

acid

ific

atio

n

Ren

al,o

fm

etab

olit

esR

enal

;ra

teac

cele

rate

dby

acid

ific

atio

n

Ren

al(8

0%);

larg

ely

asm

etab

olit

es

Tm

ax2–

4h

1–3

h2–

4h

1–2

hH

alf-

life

6–8

h2–

4h

5–7

h5

h(2

1h

inpo

orm

etab

oliz

ers)

Eff

ect

onse

t30

–60

min

30–6

0m

in30

min

–E

ffec

tpe

ak1–

3h

1–3

h1–

3h

–E

ffec

tdu

rati

on3–

5h

2–4

h3–

5h

–

�TA

BLE

6-2.

Drug

sfo

rA

DHD:

Dose

san

dPh

arm

acod

ynam

ics

MED

ICA

TIO

NTA

BLET

S/DO

SAG

ESDO

SERA

NG

EA

DMIN

ISTR

ATI

ON

PEA

KEF

FECT

DURA

TIO

NO

FA

CTIO

N

Am

phet

amin

esD

exed

rine

®5

mg

10–4

0m

g/d

b.i.d

.or

t.i.d

.1–

3h

5h

(gen

eric

)A

dder

all

5,7.

5,10

,12.

5,15

,20,

30m

g10

–40

mg/

db.

i.d.o

rt.

i.d.

1–3

h5

h(g

ener

ic)

Dex

tros

tat®

5,10

mg

10–4

0m

g/d

b.i.d

.or

t.i.d

.1–

3h

5h

Lo

ng-

du

rati

on

typ

eD

exed

rine

Span

sule

®5,

10,1

5m

gsp

ansu

le10

–45

mg/

don

ce-d

aily

1–4

h6–

9h

Add

eral

lX

R®

10,2

0,30

mg

caps

ules

10–4

0m

g/d

once

-dai

ly1–

4h

9h

Met

hylp

heni

date

sR

ital

in®

5,10

,20

mg

10–6

0m

g/d

t.i.d

.1–

3h

2–4

hM

ethy

lphe

nida

te5,

10,2

0m

g10

–60

mg/

dt.

i.d.

1–3

h2–

4h

Met

hylin

®5,

10,2

0m

g10

–60

mg/

dt.

i.d.

1–3

h2–

4h

Foc

alin

®2.

5,5,

10m

g5–

30m

g/d

b.i.d

.1–

4h

2–5

hL

on

g-d

ura

tio

nty

pe

Rit

alin

-SR

®20

mg

20–6

0m

g/d

q.d.

inam

orb.

i.d.

3h

5h

Met

adat

e-E

R®

10,2

0m

g20

–60

mg/

dq.

d.in

amor

b.i.d

.3

h5

hM

edad

ate-

CD

®20

mg

20–6

0m

g/d

q.d.

inam

5h

8h

Con

cert

a®18

,36,

54m

g18

–54

mg/

dq.

d.in

am8

h12

hR

ital

in-L

A®

20,3

0,40

mg

20–6

0m

g/d

q.d.

inam

5h

8h

Ato

mox

etin

e10

,18,

25,4

0,60

,80,

100

mg

0.5

mg/

kg–1

00m

gq.

d.in

amor

b.i.d

––

Mod

afin

il85

mg

1.25

–4.5

mg/

kgb.

i.d.

2h

8h


Indications and ContraindicationsThe decision to use stimulant medication in the treatmentof ADHD employs different criteria for each developmentalstage. The criteria for use of stimulants in school-age childrenand adolescents are well described in DSM-IV. More diffi-culty arises when deciding whether to use these medications inpreschool children, adult patients, patients with mental retar-dation, or children with ADHD and comorbid disorders.

ADHD IN CHILDRENThe stimulants approved for the treatment of ADHD in chil-dren are summarized in Table 6-2.

New standards have been promulgated for the treatment ofADHD recommending multimodal therapy, and medication isone component of a therapeutic treatment plan. Other combi-nations, such as behavior modification plus medication, havebeen shown to be somewhat more effective than medicationalone but medication plus cognitive–behavioral therapy is notmore effective than medication alone.

Many stimulant treatment studies for school-age childrenwith ADHD exclude children with full-scale IQs below 70,who are considered to have mental retardation (MR). This isunfortunate, for the signs of ADHD exist in children with IQsin the MR range. Treatment with a stimulant is effective inchildren with ADHD who have mild or moderate MR but theymay be more vulnerable to adverse effects.

Stimulant treatment has also proved effective in childrenwith fragile X syndrome and in children with pervasive devel-opmental disorder, though the latter may be sensitive to stim-ulant side effects and may show increased irritability, motoractivity, and stereotypies.

DRUG SELECTION• Different stimulant preparations share the same indications.

Although MPH is regarded as the drug of choice for thetreatment of ADHD, few controlled comparative studieshave been conducted to determine which of these stimulantsworks best for which child.

• Pemoline, another medication for treatment of ADHD, hasrecently (December 2005) been withdrawn by the FDA.


Sustained-release preparations make it possible to give thepsychostimulant once in the morning and avoid administra-tion during the school day. There are three reasons for theimplementation of longer-acting stimulant preparations:

• Children who take medications in school are subject to peerridicule when they leave activities to go to the nurse.

• Some school officials refuse to allow school personnelinvolvement in the administration of medication.

• The time–action course of standard stimulant medicationsallows only a brief 1- to 3-hour window of effect, so somemedicated children may experience trough periods of littleor no drug action during important parts of the school day.

There are times when it is best to consider using nonstimulantmedications in the treatment of ADHD. Possible criteria are:

• an unsatisfactory response to two different stimulants• severe side effects• psychosis is a definite contraindication for the use of stimu-

lants.

Stimulant treatment has been associated with the appearanceof reversible facial and shoulder tics for a minority of children.Even with these minor motor movements, most children cancontinue to be treated cautiously with low-to-moderate dosesof stimulants, particularly MPH.

Other reasons for such a decision are listed in Table 6-3.Some alternative medications, including tricyclic antidepres-sants, bupropion, and clonidine, are listed in Table 6-4.

� TABLE 6-3. Indications for Using Nonstimulant Drugs

Unsatisfactory response to adequate trials of both MPH and to DEXInability to tolerate stimulantsComorbid condition(s) contraindicating stimulant treatment—e.g.,

psychosisMedical contraindication—e.g., tachyarrhythmiasRisk of stimulant abuse by parent or other person in the homeWhen treatment of comorbid condition with a stimulant alone is not

effective—e.g., tricyclic antidepressant for panic disorder andADHD

Combination strategies for ADHD—e.g., tricyclic antidepressant plusMPH

�TA

BLE

6-4.

Non

-stim

ulan

tDr

ugs

inA

tten

tion-

Defic

it/H

yper

activ

ityDi

sord

er

DRUG

USUA

LDO

SERA

NG

ESC

HED

ULE

COM

MEN

TS

Tri

cycl

ican

tide

pres

sant

s2–

5m

g/kg

;be

twee

n-su

bjec

tva

riab

ility

inle

vels

ata

cons

tant

mg/

kgdo

seis

com

mon

b.i.d

.or

bedt

ime

Ant

icho

liner

gic

but

gene

rally

wel

lto

lera

ted

orm

ildan

del

ectr

ocar

diog

ram

prec

auti

ons

are

man

dato

ryC

onsi

der

ther

apeu

tic

drug

mon

itor

ing

for

side

effe

cts

orla

ckof

bene

fit.

May

beco

mbi

ned

wit

hst

imul

ants

Clo

nidi

ne0.

1–0.

4m

g2–

6ti

mes

/day

Esp

ecia

llyus

eful

for

com

orbi

dti

cdi

sord

ers,

slee

ppr

oble

ms,

and

impu

lsiv

eag

gres

sivi

tyM

onit

orbl

ood

pres

sure

Tol

eran

cean

dse

dati

onar

eof

ten

prob

lem

atic

alM

aybe

com

bine

dw

ith

stim

ulan

tsB

upro

pion

150–

300

mg

b.i.d

.–t.

i.d.

Seiz

ure

hist

ory

isa

cont

rain

dica

tion

May

exac

erba

teti

csA

tom

oxet

ine

100–

300

mg

b.i.d

.So

me

wei

ght

loss

may

beex

peri

ence

dM

odaf

inil

85–4

20m

gb.

i.d.

Som

ein

som

nia

may

resu

lt


Atomoxetine fits into this treatment hierarchy as a secondline medication, as shown by comparative trials against OROS-MPH. Trials with atomoxetine have lasted for up to one yearin children and in adults. For those who respond, its efficacyduring long-term use remains stable. Atomoxetine is indicatedfor the treatment of ADHD only as part of a comprehensivetreatment program that includes psychological, educational,and social measures. In practice, atomoxetine can be used inchildren not on selective serotonin reuptake inhibitors aftertwo stimulants have been tried; it may also be an option forfamilies and physicians hesitant to use stimulant drugs becausethey are classified by the Drug Enforcement Administrationas drugs of potential abuse.

TREATMENT INITIATION AND DOSE TITRATION• An effective treatment strategy for ADHD requires a plan

for follow-up and monitoring.• Techniques involve regular follow-up visits, the use of rating

forms from parent and teacher, and the monitoring ofacademic progress in the school.

• Seeing the patient and a family member regularly is essen-tial, often on a once-monthly basis when the medicationprescription must be renewed.

• Once a child is diagnosed, the physician should have asystematic method for titration and adjustment of the dose.

Before the first pill is given, baseline data on height, weight,blood pressure, and heart rate should be collected as well as acomplete blood count. The Conners Teacher Rating Scale andConners Parent Rating Scale can also be collected. Standard-ized scoring methods for the teacher rating scale can be usedto generate a hyperactivity factor score.

There is no universally agreed method for dosing and thedose may be selected according to the child’s weight or bytitrating the dose through the approved dose range until clin-ical response occurs or side effects limit further dose increases.Clinically, each child requires an individually constructeddose–response curve, taking into consideration the time–actioneffects of drug at each dose before making dosage adjustments.


Titration acclimatizes the child to the drug and determineshis or her best dose:

• Children should be started with low doses to minimizeadverse effects.

• Psychostimulant medication should be taken at or just aftermealtime to lessen the anorectic effects; studies have shownthat food may enhance drug absorption.

• MPH treatment can be initiated with a single 10 mg dose oflong duration beaded methylphenidate at 8:00 am for 3 days;then a 15 mg dose at 8:00 am for the next 3 days; and finally20 mg at 8:00 am for 3 days are given and maintained for atleast 2 weeks.

• Preschoolers may start as low as 2.5 mg of MPH at 8:00 ambut build to the same total 20 mg/day dose.

• DEX is usually started at 2.5 to 5 mg/day and graduallyincreased in 2.5- to 5-mg increments.

• The dosing instructions should be written down for theparent, with dates and times specified in detail.

• A photocopy of the instructions should be kept in thepatient’s chart.

The Conners Teacher Rating Scale is then repeated. Furtherdose adjustments up or down depend on the rating scale’sscores, teachers’ verbal reports, parents’ comments, andadverse effects experienced by the child.

Stimulants with a shorter duration of action have beenadministered three times a day—before school, at lunch, andat home before homework. However, current practice is tostart with a long-duration preparation, such as Concerta orMetadate-CD. This once-daily regimen avoids the noontimedosing in school. Eventually, a standard formulation may haveto be combined with a sustained-release formulation at 8:00am to ensure early- and late-morning coverage.

Plasma level measurements are not helpful for adjustingthe dose of MPH because their inter- and intraindividual vari-ability in plasma level concentrations are large and dose–response relationships vary from individual to individual.Adverse reactions to medications show the same variabilityand may appear unpredictably during different phases of thedrug’s absorption or metabolic phases.


Specific recommendations are made for atomoxetine:

• Atomoxetine is effective when given once daily but the totaldaily dose may be equally divided into two to reduce sideeffects, given in the morning and late afternoon or earlyevening.

• Treatment should be initiated at a dose of 0.5 mg/kg for chil-dren weighing under 70 kg; this should be increased overa period of three days to target doses of approximately1.2 mg/kg.

• No therapeutic benefit has been shown at higher doses andthe total daily dose should not exceed 1.4 mg/kg or 100 mg,whichever is the greater.

• For children weighing over 70 kg and adults, treatmentshould be initiated at a total daily dose of 40 mg andincreased to 80 mg/day over three days; this may be increasedafter a further 2–4 weeks to a maximum of 100 mg if neces-sary.

• Atomoxetine should not be administered to patients withmoderate or severe hepatic insufficiency.

Treatment EvaluationThere is no agreement about:

• how many target symptoms must be reduced and by howmuch before a child crosses the threshold into “full clinicalresponse”;

• whether a full responder must improve in all settings (home,school, and the physician’s office);

• whether response is best defined by clinician scored ratingsor by parent and teacher ratings;

• whether impairment ratings, global ratings, or symptom scalescores produce the most valid and reliable indicator ofresponse.

Each child’s response to psychostimulants is different. Like-wise, each family’s needs are different.

• Maintenance plans should include schedules for the regularcollection of information that constitutes the child’s thera-peutic drug monitoring.


• Dosing plans for the child’s vacations, weekends, and after-school periods must be individualized.

• Many parents, concerned about long-term side effects, feelmost comfortable with their child not taking medication eachweekend and through the summer, despite the costs to familyharmony. The benefits depend on the balance of behavioralproblems and side effects in individual cases.

• Medication compliance should be monitored at each visit.• Height and weight should be taken every six months, and

the child’s pediatrician can be requested yearly to performa complete physical examination and blood work (completeblood count, liver function studies).

• The frequency of visits depends on the other therapiesrecommended. These may include once-monthly parentalcounseling, twice-monthly individual therapy, or weeklymeetings for individual psychotherapy or behavioral modi-fication management. A minimal frequency should beonce-monthly, particularly in the nine states requiringmultiple-copy prescription forms, which limit the amount ofpsychostimulant ordered to a 30-day supply.

A structured rating scale should be chosen that is easy tointerpret, convenient to use, and available in both parent andteacher formats. These scales, however, should not be used assubstitutes for an open discussion with the parent and teacher.They can be collected every four months or whenever thephysician needs to make decisions about dosage adjustment,time of dosing, or even continuation of medication. Examplesof these scales include:

• The 39-item Conners Teacher Rating Scale• The ADHD Rating Scale filled out by the clinician

Current thinking indicates that a 25% drop from baseline inthe mean of clinician-scored ADHD symptom ratings (with alower total score indicating improvement) is defined as apositive clinical response. Once the maintenance dose of thestimulant is set, the rating scales for teachers and parents canbe repeated at 4-month intervals.


Maintenance TreatmentThe predicted duration of treatment is an important step inthe generation of a treatment plan for a child with ADHD.Although there is no clear-cut recommendation for the lengthof psychostimulant treatment, many parents have a reasonableexpectation that it will not be open-ended.

• Treatment duration should be planned on an individual basisand planned in single school year units, starting with thepresent and projecting treatment to last through the presentschool year plus at least one additional month into thefollowing academic year.

• Once the child has responded to an initial level of medica-tion, maintenance doses can be set slightly lower. Treatmentcan be continued, if need be, and the next decision point canbe set for the following fall.

A trial off-medication can be used to determine if treatmentshould be continued for another year. Medication can bediscontinued during the school year, during some stable period.This should not be the beginning of the school year or duringcrucial placement examinations. Placebo tablets are no longeravailable from the companies producing MPH, so the psychos-timulants are simply discontinued. Most children do not needto taper their dose of medication at discontinuation, unlessthey show signs of marked afternoon rebound.

Treatment ResistanceMany apparent nonresponders to stimulants may simply havebeen treated with too low a dose or not treated with a secondstimulant when the first one fails. Both MPH and DEX shouldbe tried before moving to nonstimulant treatments for pedi-atric ADHD but specialist advice should be sought beforeventuring beyond a total dose of 60 mg/day of MPH and40 mg/day of DEX.

• Approximately 25% of children with ADHD are not helpedby the first psychostimulant given or experience side effectsso bothersome that meaningful dose adjustments cannotbe made.


• It is estimated that between 10 and 30% of children withADHD do not improve with stimulants or improve but expe-rience unmanageable side effects.

Effects of Treatment on SymptomsStimulants ameliorate disruptive ADHD behaviors cross-situationally (classroom, lunchroom, playground, and home)when repeatedly administered throughout the day:

• In the classroom, stimulants decrease interrupting, fidgeti-ness, finger tapping, and increase on-task behavior.

• In the playground, stimulants reduce overt aggression, covertaggression, signs of conduct disorder, and increase attentionduring baseball.

• At home, stimulants improve parent–child interactions, on-task behaviors and compliance; in social settings, stimulantsameliorate peer nomination rankings.

Although psychostimulants produce moderate to marked short-term improvement in motor restlessness, on-task behavior,compliance, and classroom academic performance, these effectshave been demonstrated convincingly in studies where themedications are given under protocol conditions. For example,academic improvement on tests of academic skills were main-tained over 14 months of the MTA protocol. When exam-ined over greater intervals, however, stimulant treatment plansin the community have generally failed to maintain academicimprovement or to improve the social problem-solving deficitsthat accompany ADHD.

Unfortunately, there are only a handful of published treat-ment studies on preschoolers.

• MPH produces improvements in structured situations, butnot in free play. Current labeling warns against using MPHin children below the age of six years. Even so, there was a180% increase between 1991 and 1195 in MPH prescriptionswritten for preschoolers.

• MPH appears to have a linear dose–response effect onimprovements in the mother–child interaction, perhapsrelated to increasing child compliance and decreased symp-tomatic intensity in the child.


ADHD in AdultsThe lifetime prevalence of ADHD in adults has been esti-mated to be at 4%, based on data from the National Comor-bidity Study. Although it had been assumed that childrenwith ADHD outgrow their problems, prospective follow-upstudies have shown that ADHD signs and symptoms continueinto adult life for as many as 60% of children with ADHD.However:

• Only a small percentage of adults impaired by residualADHD symptoms actually meet the full DSM-IV childhoodcriteria for ADHD.

• Adults with concentration problems, impulsivity, poor angercontrol, job instability, and marital difficulties sometimesseek help for problems they believe to be the manifestationof ADHD in adult life.

• Parents may decide that they themselves are impaired bythe same attentional and impulse control problems foundduring an evaluation of their ADHD children.

Alternative Preparations to the Standard StimulantMedicationsThe marketing of long-duration stimulant preparations waspreceded by a series of multisite, randomized, placebo-controlled, parallel-design clinicial trials involving hundreds ofschoolage children. These trials were used for Phase 3 registra-tion trials required for FDA approval. As a result, the stimulantmedication evidence base was expanded, and proved proof forthe safety and efficacy for mono-isomer MPH, racemic MPH,and mixed salts of amphetamine.

Second line medications have also been studied. The mostextensive controlled data exist for atomoxetine, involving overa thousand patients in controlled trials, and 3,000 in safetystudies. Atomoxetine was the first agent in this class to receiveapproval for ADHD in both children and adults; it improvesratings of core ADHD symptoms in adults.

The evidence for other medications is weaker. Fewer adultsstudies have involved stimulants, as shown in Table 6-5.Agents used for adult ADHD have included: MPH, 5 to 20 mg


� TABLE 6-5. Experimental Medication Therapies for Adultswith Attention-Deficit/Hyperactivity Disorder

POSSIBLE MEDICATION TREATMENTS SUGGESTED DOSE RANGE

Methylphenidate 5 mg b.i.d.–20 mg t.i.d.Amphetamine 5 mg b.i.d.–20 mg b.i.d.Mixed amphetamine salts 5 mg b.i.d.–20 mg b.i.d.Bupropion 100 mg b.i.d.–100 mg t.i.d.Selegiline 5 mg b.i.d. only

t.i.d.; DEX, 5 to 20 mg b.i.d.; methamphetamine, 5 to 25 mgonce in the morning; bupropion, 100 mg b.i.d. to 100 mg t.i.d.;and selegiline, 5 mg b.i.d. only.

The evidence for the efficacy of these alternative stimulantsfor ADHD in adult patients is limited, so practitioners shouldbe cautious in their use until more convincing proof is available.Of particular concern is the danger of using psychostimulantsin adults with comorbid substance abuse disorder. It would bewise to use OROS MPH, because it cannot be broken downinto a powder for internasal use, and because it has been shownin controlled studies to significantly reduce the symptoms ofADHD in adults. (See Tables 6-5 and 6-6.) Atomoxetine alsohas a low potential for abuse.

NarcolepsyNarcolepsy is a chronic neurological disorder that presentswith excessive daytime sleepiness and various problems ofrapid eye movement physiology, such as cataplexy (unexpecteddecreases in muscle tone), sleep paralysis, and hypnagogichallucinations, which are intense dream-like imagery beforefalling asleep. The prevalence is estimated to be 90 in 100,000.Treatment can include a regular schedule of naps; counselingof family, school, and patient; and use of medications, includingstimulants and rapid eye movement-suppressant drugs such asprotriptyline. Treatment may begin with standard, short-actingstimulants, such as MPH, 5 mg b.i.d., and increasing the doseup to 30 mg b.i.d. if necessary. Modafinal is also indicated forthe treatment of narcolepsy.

�TA

BLE

6-6.

Cost

sof

Psyc

host

imul

ants

BRA

ND

NA

ME

GEN

ERIC

NA

ME

BRA

ND

NA

ME

MA

NUF

ACT

URER

TABL

ETSI

ZEDO

SAG

ERA

NG

E

COST

PER

100

TABL

ETS∗ :

BRA

ND

(GEN

ERIC

)($

)

Met

hylp

heni

date

Nov

arti

sP

harm

aceu

tica

l,Su

mm

it,N

.J.

5m

g2.

5–60

mg

64(2

5)10

mg

91(3

5)20

mg

131

(69)

SR,2

0m

g19

8(9

6)L

A,1

0m

g30

2(N

A)

LA

,20

mg

302

(NA

)L

A,3

0m

g30

8(N

A)

LA

,40

mg

317

(NA

)D

exed

rine

(d-a

mph

etam

ine)

Smit

hKIi

neB

eech

amP

harm

aceu

tica

ls,

Phi

lade

lphi

a,P

A5

mg

(tab

let)

5m

g(s

pans

ule)

2.5–

40m

g54

(28)

130

(79)

10m

g(s

pans

ule)

162

(98)

15m

g(s

pans

ule)

207

(125

)A

dder

all

Shir

eR

ichw

ood

Inc.

,F

lore

nce,

KY

5m

g5–

40m

g22

7(6

1)7.

5m

g22

7(1

43)

�TA

BLE

6-6.

(Con

tinue

d)

BRA

ND

NA

ME

GEN

ERIC

NA

ME

BRA

ND

NA

ME

MA

NUF

ACT

URER

TABL

ETSI

ZEDO

SAG

ERA

NG

E

COST

PER

100

TABL

ETS∗ :

BRA

ND

(GEN

ERIC

)($

)

10m

g22

7(6

1)12

.5m

g22

7(1

43)

15m

g22

7(1

43)

20m

g22

7(6

1)30

mg

227

(61)

XR

,5,1

0,15

,20,

25,3

0m

g37

3(N

A)

Stra

tter

a(a

tom

oxet

ine)

Eli

Lill

y&

Com

pany

,Ind

iana

polis

,IN

10m

g[0

.5m

g/kg

]–10

0m

g37

5(N

A)

18m

g37

5(N

A)

25m

g37

5(N

A)

40m

g38

5(N

A)

60m

g38

5(N

A)

SR

,Sus

tain

edre

leas

e;L

A,L

ong

Act

ing;

XR

,ext

ende

dre

leas

e;N

A,n

otav

aila

ble

inth

isfo

rm.

∗P

rice

sar

ees

tim

ates

base

don

2006

Car

dina

lHea

lth

who

lesa

lepr

ice

guid

epr

ovid

edby

Pha

rmac

yD

epar

tmen

tofN

ewY

ork

Stat

eP

sych

iatr

icIn

stit

ute,

New

Yor

k,N

Y,r

ound

edto

the

near

est

who

ledo

llar.


Other IndicationsPsychostimulants have been used to treat depression, as aprovocative test in schizophrenia, and to treat the cognitive–affective dysfunction found in patients with acquired immun-odeficiency syndrome-related complex. Anecdotal reports,open studies, and small controlled studies suggest that stim-ulants may help patients with depression and mania; in geri-atric patients who are withdrawn and apathetic; medically illpatients who are depressed; and in patients with patholog-ical fatigue or neurasthenia. Open studies using 20 mg/day ofMPH as an adjuvant to tricyclic antidepressants in refrac-tory depression showed improvement, probably because of anelevation in plasma level of tricyclic antidepressants. However,controlled studies have not been done to show that combina-tion tricyclic antidepressant–stimulant regimens do better thantricyclic antidepressants alone.

CostsThe costs of the principle treatments for ADHD are summa-rized in Table 6-6.

ADVERSE EFFECTSStimulant side effects are dose-dependent and range frommild to moderate in most children (Table 6-7). Management

� TABLE 6-7. Stimulant Side Effects and Their Management

SIDE EFFECT MANAGEMENT

For all sideeffects

Unless severe, allow 7–10 days for tolerance to develop.Evaluate dose–response relationships.Evaluate time–action effects and then adjust dosing

intervals or switch to sustained-release preparation.Evaluate for concurrent conditions, including

comorbidities and environmental stressors.Consider switching stimulant drug.

Anorexia ordyspepsia

Administer before, during, or after meals.With atomoxetine, consider drug-induced jaundice.



SIDE EFFECT MANAGEMENT

Weight loss Give drug after breakfast and after lunch.Implement calorie enhancement strategies.Give brief drug holidays.

Slowed growth Apply weight loss remedies, such as late-nightsnacks.

Give weekend and vacation (longer) drugholidays.

Consider another stimulant or nonstimulantdrug.

Dizziness Monitor blood pressure and pulse.Encourage adequate hydration.If associated with only Tmax, change to

sustained-release preparation.Insomnia or

nightmaresAdminister earlier in day.Omit or reduce last dose.If giving sustained preparation, switch to IR

preparation.Consider adjunctive antihistamine or clonidine.

Dysphoricmood oremotionalconstriction

Reduce dose or switch to long-actingpreparation.

Switch stimulants.Consider treatment of comorbid condition

requiring alternative or adjunctive therapies.Rebound Switch to sustained-release preparation.

Combine long- and short-acting preparations.Tics Firmly establish correlation between tics

and pharmacotherapy by examiningdose–response relationship, includingno-medication condition.

If tics are mild and abate after 7–10 days withmedications, reconsider risks and benefits ofcontinued stimulant treatment and renewinformed consent.

Switch stimulants.Consider nonstimulant treatment (e.g., clonidine

or tricyclic antidepressant).If tic disorder and ADHD are severe, consider

combining stimulant with a high-potencyneuroleptic.

Psychosis Discontinue stimulant treatment.Assess for comorbid thought disorder.Consider alternative treatments.


generally involves a temporary reduction of dose or a changein time of dosing. MPH has an excellent safety record, prob-ably because the duration of action of standard formulationsis so brief.

Appetite suppression can appear when children withADHD begin stimulant treatment. For this reason, dosingshould optimally occur after breakfast and lunch. Even thoughthe daytime appetite is reduced, hunger rebounds in theevening.

• The effects on appetite often weaken within the firstsix weeks of treatment and are reversed when treatment isdiscontinued.

• DEX, whose half-life is two to three times that of MPH,produces more sustained effects than MPH on weight velocity.

• MPH-treated children with ADHD followed up for two tofour years show dose-related decreases in weight velocity,with some tolerance to the suppressive effect developing inthe second year.

• About 20% of children treated with 1.2 mg/kg/day of atom-oxetine lose 3.5% of their body weight; long-term follow-up over 18 months reported slight decreases in height andweight percentiles.

Height and weight should be measured at 6-month inter-vals during stimulant treatment and recorded on age-adjustedgrowth forms to determine the presence of a drug-relatedreduction in height or weight velocity. If such a decrement isdiscovered during maintenance therapy with psychostimulants,a reduction in dosage or change to another class of medicationcan be carried out.

The safety of stimulants in preschoolers is now under study.Although no multiple-sample pharmacokinetic studies havebeen done in preschool children to determine if younger chil-dren, with their larger liver-to-body size ratio, might requiredifferent doses for maximum efficacy and safety, one multisitecontrolled trial revealed that the optimal total daily dose ofMPH ranged around 15 mg of the immediate release prepa-ration. Therefore, it is wise to start preschool children withADHD on low doses of the IR preparation. The study also


� TABLE 6-8. Side Effects Associated with Atomoxetine

abdominal paindecreased appetitevomitingnauseasomnolencefatiguesexual dysfunction (erectile dysfunction, impotence, abnormal

orgasm)loss of body weightslight decreases in height and weight percentiles over 18 monthsincreased risk of mydriasisallergic reactions, including angioneurotic edemavery rare cases of severe liver injury.

showed higher rates of adverse events in the younger group,even on lower doses of MPH.

The incidence of atomoxetine side effects in school age chil-dren appears to be low but comparative trials with other treat-ments for ADHD are lacking (Table 6-8). Atomoxetine hasrecently been found to cause rare cases of jaundice. Treatmentwith atomoxetine should be discontinued if hepatotoxicity issuspected.

DRUG INTERACTIONSInteractions between psychostimulants and other medicationsmay be pharmacokinetic or pharmacodynamic and are poten-tially serious (Table 6-9). In particular, the addition of a

� TABLE 6-9. Drug Interactions

MEDICATION STIMULANT EFFECT

Guanethidine MPH Decreased hypotensive effectof guanethidine

Coumarin MPH Decreased metabolism ofcoumarin

Antiepileptic medications(phenobarbital,diphenylhydantoin,primidone)

MPH Decreased metabolism ofantiepileptic agents


Phenylbutazone MPH Decreased metabolism ofphenylbutazone

Tricyclic antidepressants MPH Decreased metabolism ofantidepressant

Ammonium chloride DEX Increased elimination of DEXSodium phosphate DEX Increased elimination of DEXAcetazolamide DEX Delayed elimination of DEXThiazide diuretics DEX Delayed elimination of DEXSympathomimetics MPH, DEX Increased sympathomimetic

effectsSympatholytic

antihypertensives(beta blockers,others)

MPH, DEX Decreased antihypertensiveeffect of antihypertensives

Albuterol atomoxetine Potentiates increased heartrate and blood pressure

From Waslick B and Greenhill LL (2000) Psychostimulants. In PsychiatricDrugs, Lieberman JA and Tasman A (eds.) WB Saunders, pp. 128–155. © 2000,with permission from Elsevier.

psychostimulant to a monoamine oxidase inhibitor antide-pressant regimen is a potentially lethal combination that canelevate blood pressure to dangerous levels. Other medica-tions have been combined successfully with MPH, such asclonazepam to reduce tics and clonidine to reduce sleepdisturbances.

SUMMARYPsychostimulant medications are a mainstay in the treatmentof ADHD. This popularity has resulted from their proven effi-cacy during short-term controlled studies, as shown by lowerADHD symptom scores given by teachers and parents. In fact,the majority of children with ADHD respond to either MPHor DEX, so nonresponders are infrequent. Yet, the responseof ADHD children to MPH and other psychostimulants formore than two years is not yet known. Optimal treatmentinvolves the planning for a multimodal treatment programthat combines educational and psychosocial interventions withmedication therapy.

However, treatment plans that center on psychostimu-lant medication have flourished for a number of reasons.


The effects of psychostimulants are rapid, dramatic, andnormalizing. The risk of long-term side effects remains low,and no substantial impairments have emerged to lessen theremarkable therapeutic benefit–risk ratio of these medications.More expensive and demanding treatments, including behaviormodification and cognitive–behavioral therapies, have, at best,only equaled the treatment with psychostimulants. In the MTAstudy, the combination of behavioral and medication therapieswas only slightly more effective than the medication alone.

ADDITIONAL READING

1. Schatzberg AF and Nemeroff CB (Eds) (2004) The AmericanPsychiatric Press Textbook of Psychopharmacology, 3rd edition,American Psychiatric Press, Washington, DC.

2. Greenhill LL, Shockey E, Halperin J, March J. (2003) Stimulants.In Psychiatry, 2nd edition (Eds Tasman A, Kay J, Lieberman JA),John Wiley & Sons, Ltd, London, 2062–2095.

3. MTA Cooperative Group (1999) 14 month randomized clin-ical trial of treatment strategies for children with attentiondeficit hyperactivity disorder. Archives of General Psychiatry 561073–1086.

4. MTA Cooperative Group (2004) National Institute of MentalHealth multimodal treatment study of ADHD follow-up: 24-monthoutcomes of treatment strategies for attention-deficit/hyperactivitydisorder. Pediatrics 113(4) 754–761.

5. Greenhill LL, Pliszka S, Dulcan MK, Bernet W, Arnold E,Beitchman J. (2002) Practice parameter for the use of stimulantmedications in the treatment of children, adolescents, and adults.Journal of the American Academy of Child and Adolescent Psychi-atry 41 26S–49S.

6. Greenhill L. (2002) Childhood attention deficit hyperactivitydisorder: pharmacological treatments. In Treatments that Work,2nd edition (Eds Nathan PE, Gorman J), Saunders, Philadelphia,25–55.

7COGNITIVE ENHANCERSAND TREATMENTS FORALZHEIMER’S DISEASE

INTRODUCTIONBy the late 1980s the advent and general acceptanceof research-based diagnostic criteria for the dementia ofAlzheimer’s disease (AD), and an understanding of its under-lying pathology along with mechanism-based pharmacolog-ical therapeutics, provided the framework for clinical trialsto exploit a variety of new treatment strategies that mightpositively impact the illness. During the 1990s research- andconsensus-based criteria were proposed to include vasculardementia, dementia with Lewy bodies, and frontotemporaldementia. These later criteria are evolving and have not beenwell-accepted and thus clinical trials including these popula-tions have been limited.

PHARMACOLOGYMechanism of ActionAcetylcholine is inactivated when it is hydrolyzed tocholine and acetate by acetylcholinesterase (AChE) andbutyrylcholinesterase (BChE). By inhibiting the actions of



AChE, cholinesterase inhibitors (ChIs) effectively increase theamount of ACh available for intrasynaptic cholinergic receptorstimulation.

The ChIs differ among themselves in selectivity for AChEand BChE, mechanism of inhibition, competition with AChfor binding, and in pharmacokinetics. An unresolved questionis whether or not these differences result in differential clin-ical efficacy, and different clinically observable adverse events.A summary of pharmacokinetics and pharmacodynamics is inTable 7-1.

• An acetylcholinesterase inhibitor can work at either of twosites on AChE, an ionic subsite or a catalytic esteraticsubsite, to prevent the interaction between ACh and AChE.Tacrine and donepezil act at the ionic subsite; physostigmineand rivastigmine act at the catalytic esteratic subsite.

• Tacrine is an example of a non-selective inhibitor of AChEand BChE.

• Specific inhibition of AChE can occur with relatively littleinhibition of when the side chains of the ChI interacts withthe peripheral anionic site of AChE. Donepezil has thisproperty and is therefore selective for AChE.

• Binding to the AChE sites may be either reversible or irre-versible, and may be competitive or noncompetitive withacetylcholine. Galantamine is an example of a competitiveChI, competing with acetylcholine for AChE; tacrine is anon-competitive inhibitor.

In addition, AChE is present in a few molecular forms: one, atetramer, G4, is located on the presynaptic membranes withinthe cholinergic synaptic cleft; another a monomer, G1, is foundon postsynaptic membranes. Although G4 is decreased alongwith the neuronal loss of presynaptic cholinergic neurons, post-synaptic cholinergic receptor neurons and G1 ACh are notdecreased significantly with AD or aging. Rivastigmine is aChI that is highly selective for the postsynaptic G1 monomerform of AChE, while galantamine is less so, and donepezil isnot selective.

L-glutamate is the main excitatory neurotransmitter in thecentral nervous system. Enhancement of its activity at theN-methyl-D-aspartate (NMDA) receptor may contribute to

�TA

BLE

7-1.

Phar

mac

odyn

amic

and

Phar

mac

okin

etic

sof

Mar

kete

dCh

olin

este

rase

Inhi

bito

rs

DRUG

PHA

RMA

CODY

NA

MIC

SA

BSO

RPTI

ON

BIO

AVA

IL-

ABI

LITY

PEA

KPL

ASM

A(H

)EL

IMIN

ATI

ON

HA

LF-L

IFE

(H)

PRO

TEIN

BIN

DIN

GM

ETA

BOLI

SM/C

OM

MEN

TS

Tac

rine

(Cog

nex)

Non

com

peti

tive

,re

vers

ible

ChI

,bot

hbu

tyry

lan

dac

etyl

ChI

,al

som

ulti

ple

othe

rac

tion

s

Del

ayed

byfo

od17

%1–

22–

455

%V

ia1A

2,no

nlin

ear

phar

mac

okin

etic

s;he

pato

xici

tyre

quir

esre

gula

rm

onit

orin

gof

seru

mal

amin

eam

inot

rans

fera

ses.

Don

epez

il(A

rice

pt)

Non

com

peti

tive

,re

vers

ible

acet

ylC

hI.

Not

affe

cted

byfo

od10

0%3–

470

96%

Via

2D6,

3A4.

Non

linea

rph

arm

acok

inet

ics

at10

mg/

dR

ivas

tigm

ine

(Exe

lon)

Non

com

peti

tive

ChI

,bo

thbu

tyry

lan

dac

etyl

ChI

may

diff

eren

tial

lyaf

fect

diff

eren

tac

etyl

ChI

s

Del

ayed

byfo

od40

%1.

4–2.

6<

540

%H

ydro

lysi

sby

este

rase

san

dex

cret

edin

urin

e(n

onhe

pati

c).D

urat

ion

ofch

olin

este

rase

inhi

biti

onlo

nger

than

plas

ma

half

-lif

e.N

onlin

ear

phar

mac

okin

etic

sG

alan

tam

ine

(Rem

inyl

)C

ompe

titi

ve,r

ever

sibl

eC

hI,

mod

ulat

esni

coti

nere

cept

ors

Del

ayed

byfo

od90

%1

718

%V

ia2D

6,3A

4

No

te:

Pha

rmac

odyn

amic

effe

cts

ofso

me

ChI

sar

elo

nger

than

thei

rel

imin

atio

nha

lf-l

ives

.D

rugs

that

inhi

bit

orin

duce

the

cyto

chro

me

enzy

mes

abov

em

ight

beex

pect

edto

incr

ease

orde

crea

sebl

ood

leve

ls.

For

the

mos

tpa

rt,

how

ever

,dr

ugin

tera

ctio

nsw

ith

done

pezi

l,ri

vast

igm

ine,

and

gala

ntam

ine

have

not

been

clin

ical

prob

lem

s.


the pathogenesis of Alzheimer’s disease, a phenomenon knownas excitotoxicity. Memantine is a low affinity antagonist thatis believed to reduce NMDA receptor overstimulation andrestore receptor signaling function to more normal levels. Itis also possible that reducing excitotoxicity may be neuropro-tective by preventing neuronal calcium overload though thereis no evidence that memantine modifies neurodegeneration inpatients with Alzheimer’s disease.

Several other strategies have been evaluated or may havepotential as treatments for Alzheimer’s disease (Table 7-2)though none are currently supported by adequate evidence ofefficacy.

� TABLE 7-2. Other Treatment Strategies

DRUG MECHANISM OF ACTION EVIDENCE

Dopamine precur-sors/agonists(clonidine,guanefacine,amantadine,bromocriptine)

Enhance dopaminefunction

Studies suggest largelyineffective

Selegiline MAO-B inhibitor No adequate evidenceof efficacy;see below forcombination withvitamin E

Hydergine,nicergoline

Metabolic enhancement No adequate evidenceof efficacy

Nootropics (e.g.,piracetam,oxiracetam)

Neuroprotectionand cognitiveenhancement

Non-specific effect onmemory but noadequate evidenceof efficacy indementia

Neurotrophicfactors(e.g., nervegrowth factor)

May counteract atrophyof cholinergicneurones

Tested in small numberof patients withlimited success; noclinical trials

Estrogens May enhance cognitivefunction and exertneurotrophic andneuroprotectiveeffects

Epidemiologicalevidence suggestspossible benefit butintervention studiesshow no efficacy

TREATMENTS FOR ALZHEIMER’S DISEASE 203

NSAIDs May reduceinflammationassociated withneurodegeneration

Epidemiologicalevidence suggestspossible benefit butintervention studiesshow no efficacy

Vitamin E (withselegiline),Ginkgo biloba

Antioxidants Vitamin E plus selegilinemaintains activities ofdaily living andprolongs survival inthe community butwithout measurableimprovement incognitive testperformance; Ginkgobiloba used in Europefor dementiasyndromes butevidence for itscognitive enhancingproperties is weak andinconsistent

Calcium channelblockers

Block increasesintracellular calciumthat may mediatecell death

May delay progression ofmemory impairmentand onset of dementiabut there is a lack ofcontrolled data onefficacy

PharmacokineticsThe cholinesterase inhibitors differ in their pharmacokineticcharacteristics.

• Tacrine is poorly absorbed and has a short half-life. It ismetabolized by CYP 1A2 hepatic enzymes.

• Donepezil has an oral bioavailability approaching 100%,with peak concentrations in four hours and linear pharma-cokinetics. The drug accumulates at a constant rate, reachingsteady state in about two weeks. Its relatively slow clear-ance, with a long half-life of 70 hours, enables once-dailyadministration, with little plasma level fluctuation. The drugis extensively bound to plasma proteins, including �1-acidglycoproteins. It is both excreted unchanged in the urine


and extensively metabolized by CYP 2D6 and 3A4 hepaticenzymes to active and inactive metabolites.

• The oral bioavailability of rivastigmine is about 40% up to adose of 3 mg, after which AUC increases non-linearly. Thepeak plasma concentration occurs after one hour; maximumcholinesterase inhibition is about 60%, occurring after fivehours and lasting for about ten hours. The eliminationhalf-life is 1.5 hours. Rivastigmine undergoes hydrolysisby cholinesterase, with minimal hepatic involvement. It isexcreted almost entirely in the urine as the sulfate conjugateof the decarbamylated metabolite.

• The oral bioavailability of galantamine is about 90%;peak serum concentrations occur after one hour after theimmediate-release formulation, corresponding to the time ofmaximum inhibition of cholinesterase of 40%. The modified-release formulation is bioequivalent with the immediate-release tablet, though peak levels are both delayed(to 4.5–5 hours) and reduced (by 25%). The pharmacoki-netics of galantamine are linear over the recommended doserange. It has low protein binding (18%) and 53% of theconcentration in whole blood is distributed into red cells. Itundergoes metabolism by CYP2D6 and CYP3A4 enzymes;about one-third of the dose is excreted unchanged in theurine and 12% as the glucuronide. Although total drug expo-sure (AUC) is increased by 35% in poor metabolizers, andby a similar amount in elderly patients, this is not clinicallysignificant because the dose is titrated according to patienttolerability.

• Peak blood levels of memantine occur 3–7 hours after oraladministration. It has low protein binding (45%) and a longhalf-life of 60–80 hours. Total exposure to the drug is 45%greater among women than men due to their lower bodymass but the recommended dose is the same. Approxi-mately half the dose of memantine is excreted unchangedin the urine; the remainder undergoes hepatic conversionto inactive metabolites. The metabolism of memantine isnot affected by other drugs that induce or inhibit hepaticenzymes but drugs that alkalinize the urine (e.g., carbonicanhydrase inhibitors) reduce its clearance.


INDICATIONSThe ChIs are approved for the treatment of mild to moderateAlzheimer’s disease; memantine is currently the only agentapproved for the treatment of moderate to severe Alzheimer’sdisease (defined as an MMSE score of less than 15).

Because of the actions of ChIs, these drugs requirecaution when used in patients with significant asthma, signif-icant chronic obstructive pulmonary disease, cardiac conduc-tion defects, or clinically significant bradycardia. Appropriateconsiderations are involved in general anesthesia as well sincethey may prolong the effects of succinylcholine-type drugsand memantine may increase the risk of adverse effects withketamine.

DRUG SELECTIONApproaches to the treatment of AD can be grouped intoseveral conceptual categories (Table 7-3).

• One approach attempts to treat the behavioral symptomssuch as agitation, aggression, psychosis, depression, anxiety,apathy, and sleep or appetite disturbances.

• A second approach attempts to treat the cognitive orneuropsychological signs of the illness such as memory,language, praxis, attention, orientation, and knowledge.

• A third approach attempts to slow the rate of progression ofthe illness, preserving patients’ quality of life or autonomy.(Slowing the rate of decline might also be related to treatingsymptoms.)

• A fourth conceptual treatment approach is primary preven-tion, to delay the time to onset of illness. Success at thisapproach could have considerable impact: for example,delaying the onset of AD by five years would halve its inci-dence.

Thus far, only ChIs and memantine have shown generallyconsistent symptomatic efficacy in standardized, well-controlled multicenter trials lasting from six months to occa-sionally 12 months. Tacrine is associated with a relatively highrisk of hepatotoxicity and is now little used.


� TABLE 7-3. Conceptualized Treatment Strategies for Patientswith Cognitive Impairment

Symptomatic and/or Restorative

Targets: impaired cognition, depression, psychosis, agitation, aggression,anxiety, insomnia

Examples: cholinesterase inhibitors, various cholinergic agonists,antidepressants, antipsychotics, mood stabilizers, antianxiety agents,hypnotics, NMDA and AMPA receptor modulators, angiotensinconverting enzyme inhibitors

Neurotrophic factors: nerve growth factor, brain-derived neurotrophicfactors, estrogens

Notes: some substances may have symptomatic or restorative effects but areunproven. These may include hydergine and neotropics such as piracetam.

Pathophysiologically-directed

Targets: underlying pathophysiology of neurodegeneration, includinginflammation, production of oxidizing free radicals, excitatory amino acids.

Examples: anti-inflammatory agents, calcium channel blockers, NMDAand AMPA receptor modulators. Transplantation of hormonally activetissues, or NGF (nerve growth factor) gene therapy using viral vectorshave been undertaken experimentally.

Etiologically-directed

Targets: �-amyloid formation or hyperphosphorylated tau protein.

Examples: modulators of APP expression, �- and �-secretase inhibitors,inhibitors of beta-amyloid protein aggregation or deposition,immunization with antibodies to beta-amyloid.

Note: The interventions listed above include some that are available andmarketed, as well as some that have not been demonstrated effective orsafe, and some conceptual treatments not yet developed.

TREATMENT INITIATIONThe typical candidates for ChIs are as follows:

• Outpatients with AD of mild to moderate cognitive severity• Those usually living at home or in an assisted living facility• Those suffering from dementia as their main clinical problem• Those in which behavioral syndromes such as psychosis,

agitation, or significant insomnia, apathy, or depression donot dominate. ChIs can be given in the presence of most ofthese comorbid symptoms.


Treatment with cholinesterase inhibitors or memantine shouldbe initiated at a low dose and, depending on tolerability,titrated to a target dose.

• Donepezil is initiated at 5 mg/day and then increased to10 mg/day after four to six weeks. Raising the dose earlierincreases the risk for cholinergic adverse events. The effec-tive dose is 5–10 mg/day; the higher dose tends to be some-what more effective when the various trials are evaluated asa group.

• The recommended starting dose of rivastigmine is 1.5 mgb.i.d., taken with meals. If this dose is well-tolerated aftera minimum of two weeks of treatment, it may be increasedto 3 mg b.i.d. Subsequent increases to 4.5 mg and then 6 mgb.i.d. should be based on good tolerability of the currentdose and may be considered after a minimum of two weeksof treatment. Higher daily doses, averaging about 9 to 10 mgare associated with better efficacy than lower doses.

• The initial dose of galantamine is 4 mg b.i.d., and should beraised to 8 mg b.i.d. after two to four weeks. For patientswho are tolerating medication but not responding, the dosecan be raised to 12 mg b.i.d. after another four weeks.

• Treatment with memantine should be initiated at a dose of5 mg/day and increased in increments of 5 mg at intervals ofno less than one week to a target dose of 20 mg/day; dosesgreater than 5 mg/day should be divided into two and givenin the morning and evening. Memantine is formulated astablets and as a liquid that is administered by the carer viaa syringe-like dosing device.

Maintenance TreatmentOptimal duration of treatment with continuing efficacy isunknown but overall efficacy extends at least 9 to 12 monthsbased on the clinical trials and open-label extension phases.

• It is not possible to predict individual patient responses toChIs because of the great interpatient variability of response.

• Maintenance treatment can be continued as long as a ther-apeutic benefit for the patient seems apparent.


• The potential clinical benefit of ChIs should be reassessedon a regular basis.

• Discontinuation should be considered when evidence of atherapeutic effect is no longer present. However, there isconsistent evidence from several studies that discontinuationusually leads to minimal to mild worsening on cognitive testscores, and the patient returns to the same level of cognitiveimpairment as the patient who took placebo all along inthese clinical trials.

Treatment EvaluationClinical experience suggests that ChIs may be effective at leastfor mildly disturbed behavior, and in delaying the onset oftroublesome behaviors, perhaps by maintaining cognitive func-tion or perhaps through enhancing attentional processes andactivation. The evidence for this effect is based on incidentalfindings in cognitive enhancing studies and the symptoms thatshow the greatest improvement are depression and apathy.

It is difficult to assess individual patient response because ofthe variability of the deteriorating course of AD, and becausemost of the effect of medication is due to a stabilization or lackof worsening of symptoms or cognitive function while placebo-treated patients continue to decline. Therefore, the clinicalobservations of minimal or no clinical worsening may be suffi-cient reasons to continue medication treatment if patients aretolerating therapy.

Virtually all clinical trials of antidementia drugs undertakenin the US for regulatory purposes have used the Alzheimer’sDisease Assessment Scale-cognitive subscale (ADAS-cog) asthe index of cognitive change and the clinical global impressionof change (CGI-C) as the “global” clinical measure. Secondarymeasures in AD clinical trials often include the Mini-MentalState Examination (MMSE), a brief, physician-administered,structured examination of cognitive function, and a varietyof functional activity scales, such as the Alzheimer’s DiseaseCooperative Study-ADL (ADCS-ADL) Scale to assess aspectsof daily functioning.

In clinical trials lasting up to six months, memantine hasbeen shown to achieve small improvements in measures of


� TABLE 7-4. FDA-Approved Drugs to Improve Cognitive Functionin Alzheimer’s Disease

DRUG HOW SUPPLIED INITIAL DOSAGEMAINTENANCE

DOSAGE

Tacrine 10, 20, 30,and 40 mgcapsules

10 mg q.i.d. 30 or 40 mgq.i.d.

Donepezil 5 and 10 mgtablets

5 mg q.i.d. 5–10 mgq.i.d.

Rivastigmine 1.5, 3, 4.5,and 6 mgcapsules

1.5 mg b.i.d. 3, 4.5, or6 mg b.i.d.

Galantamine 4, 8, and 12 mgtablets,solution4 mg/mL

4 mg b.i.d. 8 or 12 mgb.i.d.

Memantine 5 and 10 mgtablets, oralsolution2 mg/mL

5 mg q.i.d. 10 mg b.i.d.

Initial dosages should be maintained for at least 2 and preferably 4–6 weeksbefore increasing. Adverse events may occur with dosage titration.

cognition and activities of daily living and behavior. Patientshad a clinically noticeable reduction in deterioration over thisperiod with less functional and cognitive decline comparedwith placebo; they were also less likely to become agitated.Memantine also improves cognitive function, activities of dailyliving and behavior to a very small extent when added toestablished treatment with donepezil in patients with moderateto severe Alzheimer’s disease.

ADVERSE EFFECTSCholinesterase InhibitorsMost adverse events from ChIs are cholinergically mediated,and are characteristically mild in severity and short-lived,


lasting only a few days (Table 7-5). Significant cholinergic sideeffects can occur in up to about 25% of patients receivinghigher doses; often they are related to the rate of initial titra-tion of medication.

Patients tend to become tolerant to the adverse eventsrapidly. However, anorexia and weight loss may be clini-cally significant problems over the longer term, especially inolder, more medically ill, and nursing home patients, so these

� TABLE 7-5. Adverse Effects of Cholinesterase Inhibitors

Summary of adverse event data in placebo-controlled, randomized clinicaltrials. The method of obtaining adverse events and their reporting varyamong trials.

Drug Adverse Events

Tacrine Nausea, vomiting, diarrhea, dyspepsia, myalgia, anorexia,dizziness, confusion, insomnia, rare agranulocytosis

Approximately 50% of patients will develop direct,reversible hepatotoxicity manifested by elevatedtransaminases.

Drug interactions may include increased cholinergiceffects with bethanecol; increased plasma tacrine levelswith cimetidine or fluvoxamine. This may occur byinhibition of P450 1A2. The association of tacrine withhaloperidol may increase parkinsonism and tacrineincreases theophylline concentration.

Donepezil Nausea, diarrhea, insomnia, vomiting, muscle cramps,fatigue, anorexia, dizziness, abdominal pain, myasthenia,rhinitis, weight loss, anxiety, syncope (2 vs. 1%)

Rivastigmine Nausea, vomiting, anorexia, dizziness, abdominal pain,diarrhea, malaise, fatigue, asthenia, headache, sweating,weight loss, somnolence, syncope (3 vs. 2%). Rarely,severe vomiting with esophageal rupture

Galantamine Nausea, vomiting, diarrhea, anorexia, weight loss,abdominal pain, dizziness, tremor, syncope (2 vs. 1%)

Adverse event estimates vary widely among the cholinesteraseinhibitors from study to study and thus relative adverse event ratesamong drugs are difficult to estimate. Cholinergic side effects generallyoccur early and are related to initiating or increasing medication. Theytend to be mild and self-limited. Medications should be restarted atlowest doses after temporarily stopping.


General precautions with cholinesterase inhibitors (as indicated in theprescribing information)

By increasing central and peripheral cholinergic stimulationcholinesterase inhibitors may:

1. increase gastric acid secretion, increasing the risk for GI bleedingespecially in patients with ulcer disease or those takinganti-inflammatories

2. produce bradycardia, especially in patients with sick sinus or othersupraventricular conduction delay, leading to syncope, falls, andpossible injury

3. exacerbate obstructive pulmonary disease4. cause urinary outflow obstruction5. increase risk for seizures6. prolong the effects of succinylcholine-type muscle relaxants

parameters should be monitored and medication reduced ordiscontinued to assess if appetite returns, if anorexia or weightloss become clinically significant.

Tacrine is poorly tolerated by 10–20% of patients and isassociated with reversible hepatotoxicity. Transaminases maybe elevated above three times the upper limit of normal inapproximately 30% of patients within 6 to 12 weeks andreversed within 6 weeks of discontinuing medication. Althoughpatients may be able to tolerate tacrine if it is reintroduced, itis now seldom prescribed.

Memantine appears to be relatively well-tolerated. Overall,the profile of reported adverse events is similar to that occur-ring in patients assigned to placebo and include dizziness,confusion, headache, hallucinations, tiredness; less commonly,vomiting, anxiety, hypertonia, cystitis, increased libido, andseizures have been reported. Some studies suggest a beneficialeffect in reducing agitation but other studies do not show thiseffect.

DRUG INTERACTIONS• Inhibitors of CYP450 3A4 and 2D6 (such as ketoconazole

and quinidine) and inducers of CYP 2D6 and CYP 3A4(such as phenytoin, carbamazepine, dexamethasone, andphenobarbital) could either inhibit donepezil metabolism orincrease the rate of elimination, but the clinical significanceof this is not known.


• Rivastigmine’s metabolism does not depend on liver P450enzymes, and therefore no drug interactions related to theP450 system have been observed.

• Inhibitors of CYP3A4 or CYP2D6 may increase total expo-sure to galantamine; examples include paroxetine, fluoxe-tine, fluvoxamine, amitriptyline, cimetidine, and quinidine.The clinical significance of these interactions is uncertainbut changes to concomitant drug therapy during treatmentmay result in a loss of efficacy or an increased risk ofadverse effects. Galantamine appears to have no effect onthe metabolism of other hepatic substrates.

• Memantine does not inhibit or induce hepatic microsomalenzymes; because it is excreted in the urine predominantlyas unchanged drug, it is unlikely to be affected by drugsthat affect hepatic enzyme function. Drugs that alkalinizethe urine may reduce renal excretion of memantine andincrease the risk of adverse effects. Memantine may increasethe risk of central nervous system toxicity if administeredwith amantadine or dextromethorphan.

SUMMARYChIs and memantine are the best proven efficacious symp-tomatic treatments for AD. They provide consistent, but small,effects in many patients with mild to moderate dementia, andhave become the current pharmacological standard of treat-ment. Other therapeutic approaches are not as well tested oras clearly efficacious. Therefore, ChIs are likely to be activelyused clinically for at least the next several years. Memantineis the only treatment approved for moderate to severe AD.

However, therapeutic results of these treatments are usuallymodest, affecting a minority of patients. In trials with ChIs,patients assessed were usually outpatients with mild to moder-ately severe dementia and few concomitant medical illnesses.Duration of effect beyond one year and long-term safety arenot known, except for the uncontrolled observations of patientswho continue on these drugs after the controlled trial. It isessential to understand the broad magnitudes of effects andthe range of clinical utility. It often takes time, experience,and further studies for clinicians to appreciate the overall


effectiveness and utility of new drugs. Long-term trials of thecholinesterase inhibitors in patients with mild cognitive impair-ment will help define the extent and limits of their efficacy, aswill trials in vascular dementia.

ADDITIONAL READING

1. Kaduszkiewicz H et al: Cholinesterase inhibitors for patients withAlzheimer’s disease: systematic review of randomised clinical trials.BMJ 2005; 331:321–327

2. Tasman A, Kay J, Lieberman JA (Eds): Psychiatry, 2nd edition,John Wiley & Sons, Ltd, London, 2003

8DRUGS FOR TREATING

SUBSTANCE ABUSEDISORDERS

INTRODUCTIONDisorders of substance abuse are one of the commonest ofpsychiatric illnesses. They often complicate other psychiatricdisorders and, together with complications of their own, theymay also present in general medical practice. Clinicians musttherefore be prepared to consider such a diagnosis in all clinicalsettings. Substance abuse disorders fall within the competencyof psychiatry because they involve distressing behavioral andpsychological syndromes but their medical sequelae mean thatclose co-operation between psychiatrists and other clinicians isessential.

The following summary focuses on the pharmacologicalmanagement of intoxication and overdose syndromes; with-drawal syndromes; relapse prevention; and the treatment ofcomorbidity in substance abusers. Non-pharmacological strate-gies are very important components of management and theireffects are additive to those of drug treatment. However, thefocus of this book is drug treatment and these strategies arenot discussed in depth.



SYNDROMES ASSOCIATED WITH INTOXICATIONAn accurate diagnosis is of fundamental importance in theassessment of patients who may be intoxicated. A full historymust be obtained in the awareness that progression to poten-tially fatal overdose or withdrawal syndromes may occur. Thefocus of physical examination should be the acute effects ofintoxication (such as vital and neurologic signs) and chronicsigns and symptoms associated with drug dependence (size ofliver, evidence of venipuncture). Drug use within the previous4–12 hours can be determined by analysis of blood and breathsamples whereas urinalysis is useful for assessing substance usewithin the preceding 24–72 hours. Urinalysis is the preferredoption (with the exception of determining alcohol use) thoughsaliva testing is becoming more widely used. Immediate resultscan be obtained with urine testing kits suitable for use in anoffice or clinic.

Alcohol IntoxicationIntoxication with alcohol is associated with:

• maladaptive mental state (increased aggression)• neurologic signs such as incoordination, unsteady gait,

slurred speech• impaired attention and memory

Stupor, coma, and cardiovascular collapse occur at high bloodlevels of alcohol (400–800 mg/dL) but the threshold for sucheffects depends on individual tolerance.

The rate at which blood alcohol levels decline averages15 mg/dL/hour. Overall, nonpharmacological management ispreferred because it avoids the risk of interactions betweendrugs and alcohol. Lorazepam 1–2 mg orally may be effectivein belligerent patients who cannot be managed by supportivelimit setting. If, despite these measures, the patient’s conditionworsens over the next 1–2 hours, an intramuscular injection ofhaloperidol 5 mg can be safely given.

In patients with significant mental status changes or alter-ations in sensorium, clinicians should be alert to the possibility

DRUGS FOR TREATING SUBSTANCE ABUSE DISORDERS 217

of other causes such as head trauma or metabolic disturbance(e.g., thiamine deficiency). Furthermore, intoxicated patientswith a recent history of regular heavy use and likely phys-iologic dependence on alcohol may be at risk from seriousalcohol withdrawal (seizures or delirium) as the blood levelclears, particularly if there is a past history of such complica-tions. In that case, clinicians should consider starting taperingdoses of a long-acting benzodiazepine (e.g., chlordiazepoxide)as acute intoxication begins to clear (see the section on alcoholwithdrawal on page 223).

Sedative–Hypnotic IntoxicationBenzodiazepine intoxication can be associated with behav-ioral disinhibition, potentially resulting in hostile or aggres-sive behavior; this effect is most commonly seen in patientscombining benzodiazepines with alcohol. Benzodiazepine useis also frequently observed in combination with cocaine oropioids. Although benzodiazepines alone do not stronglysuppress respiration (and pure benzodiazepine overdoses donot usually produce respiratory arrest and death), benzodi-azepines may augment the respiratory depressant effects ofother drugs such as alcohol, barbiturates, or opioids.

Suspected benzodiazepine overdosage can be reversed withthe benzodiazepine antagonist flumazenil, which should beadministered intravenously, beginning with 0.2 mg slow pushover 30 seconds, followed by increments of 0.3 mg or 0.5 mg ifno response, with total dose not to exceed 3 mg to 5 mg. If noresponse is obtained at those total doses, then another causeof stupor or coma should be considered.

When barbiturates are taken in relatively low doses,intoxication can be indistinguishable from alcohol intoxica-tion. Symptoms include incoordination, sluggishness, poormemory, slow speech, poor comprehension, distorted mood,poor attention span, faulty judgment, and emotional lability.Other potential symptoms include hostility, moroseness,argumentativeness, and occasionally paraniod and suicidalideation. As with alcohol, clinicians should be alert to othercauses of mental status changes, and for patients with recentregular use and physiologic dependence on sedative–hypnotics,


particularly short-acting agents (e.g., alprazolam, lorazepam),serious withdrawal can ensue, and prophylactic treatment withtaper of a long-acting agent should be considered (see thesection on sedative–hypnotic withdrawal on page 230).

Opiate IntoxicationAlthough mild opiate intoxication can be stimulating, severeintoxication causes a maladaptive mental state with symptomssuch as apathy; this is associated with pupillary constriction(characteristically midpoint with meperidine) or dilation ifthe patient has anoxia from severe overdose, central nervoussystem depression, and respiratory depression. Meperidine,propoxyphene, or pentazocine have active metabolites thatmay cause seizures.

Severe opiate intoxication is a life-threatening conditionbecause of the high likelihood of respiratory depression orarrest. Opiate overdose is a common cause of death, especiallyamong teenagers and young adults, and is particularly likelyamong individuals with low levels of tolerance, including inex-perienced users, or patients who have recently detoxified orbeen abstinent for a period of time. Death from opiate over-dose is an underappreciated risk, and, just as one would assessrisk of suicide in depressed patients, clinicians evaluating apatient presenting with opiate intoxication should evaluate therisk of overdose, including level of tolerance and past historyof overdose episodes.

The presence of miosis and respiratory depression atpresentation is an indication for immediate treatment. Anintravenous dose of the pure opiate antagonist naloxone HCl0.4–0.8 mg usually reverses opiate-induced respiratory andCNS depression in two minutes. The dose may be repeatedevery 2–3 min if previous dose was not effective. Responseoccurs in the majority of patients after up to four doses butlarger doses may be needed in cases of intoxication by highlypotent opiates such as fentanyl or long-acting agents such asmethadone. Naloxone has a duration of action of 1–2 h, whichis shorter than that of opiates. When a response has beenachieved, a naloxine infusion should therefore be initiated(initial dose 0.4 mg/h) and maintained for a minimum of 12 h.


When using naloxone it is important to note that:

1. CNS depression in patients with opiate overdose may be dueto other causes, particularly if standard doses of naloxoneare not effective.

2. Naloxone may suddenly precipitate an opiate withdrawalsyndrome and should therefore be administered cautiously.Patients with precipitated withdrawal may become suddenlyanxious, irritable, or combative, and try to leave the emer-gency ward against advice; such patients should be treatedsupportively, and naloxone discontinued temporarily untilsymptoms clear, after which naloxone may need to beresumed in response to resumed intoxication and respira-tory depression, especially in the presence of longer-actingagonists. Such patients should be prevented, if at all possible,from leaving the emergency ward or clinic, since respira-tory depression may return quickly once the naloxone wearsoff, or the patients may seek out and take more opiates.If opiate withdrawal persists, treatment should be initiated(see below).

3. Naloxone may not reverse the effects of buprenorphine.This partial opiate agonist, long available in parenteral formfor analgesia, is now marketed in sublingual form [brandnames Suboxone (buprenorphine-naloxone) and Subutex(buprenorphine)] as an alternative to methadone for agonistmaintenance treatment of opioid dependence. Buprenor-phine by itself produces less respiratory depression thanother full opiate agonists, even at high doses, but deathfrom overdose has been associated with combinations ofbuprenorphine and benzodiazepines.

Cocaine and Amphetamine IntoxicationThe effects of cocaine, amphetamines, and similar stimulantsare characterized by the following:

• maladaptive psychological or behavioral changes such asanxiety, paranoid delusions, and hallucinations

• physical signs such as hypertension, tachycardia, mydriasis,perspiration, psychomotor agitation, and dyskinesia


• mild intoxication may be associated with behavior similar tohypomania, such as increased activity, gregariousness, andtalkativeness; or irritability, anxiety, or agitation

• severe effects include respiratory depression, cardiacarrhythmias, delirium, and seizures

The behavioral, psychologic, and physical effects ofamphetamine are longer-lasting than those of cocaine, typicallyresolving within 24–48 h.

Intoxication by cocaine and other stimulants may thereforecause serious physical effects; neurologic and cardiac shouldalways be assessed and a medical plan to manage emergencies(hypertension, arrhythmias, seizures) should be developed.

The behavioral and psychologic effects of stimulant intox-ication should, if possible, be managed in a reassuring andstraightforward manner in a quiet environment. Lorazepam2 mg is often successful if drug treatment is necessary. In fact,benzodiazepines are preferred even for more severe behavioraldisturbances such as psychosis because antipsychotic agentsmay complicate management by their cardiovascular effects(such as tachycardia) or neurological complications (hyper-thermia, seizures). If behavior escalates despite treatment witha benzodiazepine, cautious use of a high-potency antipsychoticsuch as haloperidol 5 mg may be appropriate. The clearance ofamphetamine is increased by acidification of the urine but thistechnique is not recommended in cases of cocaine intoxication.

Intoxication by LSD, Mescaline, MDMA (‘Ecstasy’),and PsilocybinIntoxication by an hallucinogen is associated with:

• maladaptive psychologic effects (anxiety, paranoia, belief ofgoing insane)

• changes of perception (perception becomes more intense,depersonalization, hallucinations, synesthesia)

• physical signs (mydriasis, tachycardia)

The usual treatment option is the oral administration of20 mg of diazepam; this should attenuate the LSD experienceand alleviate any associated panic to a halt within 20 min and is


considered a superior treatment option than the once-common‘talking down’ method.

‘Flashback’ (a transient perceptual abnormality reminis-cent of a previous episode of intoxication by an hallucinogen)can be triggered by anxiety, fatigue or use of drugs suchas cannabis. A flashback may be distressing, particularly ifthe original episode was distressing. These events are short-lived and usually remit spontaneously. Supportive reassuranceis therefore usually sufficient but a benzodiazepine may beuseful in carefully selected patients who experience anticipa-tory anxiety.

MDMA (‘Ecstasy’) is reported to produce increased feel-ings of well-being, interpersonal warmth, and connectedness,and the drug is often taken in social situations includinglarge gatherings (‘raves’). Hyperthermia and occasional deathshave been reported, perhaps abetted by overactivity anddehydration.

Phencyclidine IntoxicationIntoxication by phencyclidine (PCP) is associated with:

• maladaptive changes in behavior, such as belligerence,assaultiveness, and impulsivity

• physical signs such as nystagmus, hypertension, ataxia,muscle rigidity, diminished responsiveness to pain, seizures,and coma

• lower doses typically cause excitation whereas higher dosesare dangerously sedative

The duration of symptoms of intoxication varies: PCPundergoes enterohepatic recirculation, resulting in symptomrecurrence after a period of quiescence. Patients should there-fore be observed for 12 h before they are discharged.

Interventions other than drug treatment include a quietenvironment and close observation for the behavioral andphysical effects of intoxication. ‘Talking down’ is reportedlyless useful than in the management of hallucination intoxica-tion. Patients may have a diminished response to pain (PCP hasanesthetic activity) and if they become belligerent they may


struggle violently against restraint. In such cases there is a riskof rhabdomyolysis and patients should be closely monitored.

When pharmacologic agents are used to treat PCP intoxica-tion, the aim is usually to produce sedation or relieve psychosis;no useful PCP-receptor antagonist has been developed. Forgeneral sedation, a benzodiazepine such as lorazepam may beused orally or intramuscularly. Haloperidol is a reasonableneuroleptic agent when psychotic agitation is unresponsive tobenzodiazepines. More anticholinergic neuroleptics may actto enhance the anticholinergic properties of PCP itself. Ifneuroleptics are used, one must remember that PCP itself canproduce muscle rigidity and acute dystonias. Urinary acidi-fication has been recommended to increase the excretion ofPCP in the urine. However, this may also exacerbate a devel-oping metabolic acidosis and increase the risk of renal failureresulting from rhabdomyolysis.

The aim of drug treatment is to relieve symptoms throughsedation and reduce psychosis. There is no specific antag-onist for PCP. A benzodiazepine such as lorazepam (oralor intramuscular) is preferred for general sedation; in caseswhere psychotic agitation does not respond to a benzodi-azepine, haloperidol is a reasonable choice. There is a risk thatsome antipsychotic agents may worsen some of the effects ofPCP, which has anticholinergic activity and may cause musclerigidity and acute dystonias.

DRUG TREATMENT OF WITHDRAWAL SYNDROMESSome general principles are important when considering phar-macologic treatments for particular withdrawal syndromes:

When monitoring treatment:

• set clear targets• make serial assessments and modify on the basis of these

assessments

Psychosocial factors:

• prepare the patient• place emphasis on detoxification as a beginning to treatment


From a pharmacologic standpoint, an ideal agent for the treat-ment of withdrawal should have the following characteristics:

• efficacy in relieving the complete range of abstinence signsand symptoms for a given type of withdrawal

• a relatively long duration of action and gradual offset ofeffects

• a high degree of safety in the dosage needed to suppresswithdrawal (i.e., high therapeutic index)

• it should be available by a variety of routes of administrationand have little abuse potential in itself

Some equally general important aspects that may be over-looked include:

• keep clear treatment targets in mind• consider structured rating scales for measuring symptom

severity• treatment must be guided by serial assessment of the clinical

response: protocols offer useful guidance but orders must bereviewed and rewritten, frequently daily. It is dangerous todetoxify a patient on autopilot

From the patient’s perspective:

• they should be told what to expect from the experience ofdetoxication

• physicians should make the effort to engage them in a jointeffort to alleviate symptoms safely

• they should be awarae that they are unlikely to be free ofdistress

• it may be possible to accomplish detoxication on an outpa-tient basis for those who have relatively good health andsufficient social stability

• most importantly, they must understand that detoxication isthe beginning of treatment of their chronic problems withsubstance dependence – it is not, by itself, a treatment foraddiction

Alcohol WithdrawalAlcohol withdrawal symptoms typically occur 6–12 h after theend of, or a reduction in, heavy and prolonged drinking;


they may also occur despite the presence of significant bloodalcohol levels if those levels are falling.

Alcohol withdrawal symptoms include:

• autonomic hyperactivity• increased muscle tremor (may affect the hand, eyelids, or

tongue)• insomnia• nausea or vomiting• transient hallucinations or illusions• agitation• anxiety

The alcohol withdrawal syndrome also includes generalizedseizures. These may occur 12–48 h after the end of alcoholconsumption; they are typically nonfocal and confined to oneor two episodes.

Alcohol withdrawal delirium (delirium tremens) usuallyoccurs 3–5 days after the end of consumption. This is associ-ated with:

• disturbance of consciousness• change in cognition• perceptual disturbance• agitation, belligerence, and elevation in vital signs may

also occur

This delirium is more likely to occur in the presenceof a comorbid physical disorder or in someone with ahistory of seizures or delirium during previous withdrawalsyndromes. The occurrence of delirium is therefore an indi-cation for reassessing the patient’s medical status to identifypossibly undiagnosed illness. Alcohol withdrawal is a medicalemergency and can be life threatening without appropriatesupportive medical treatment.

Appropriate management greatly reduces the risk thatuncomplicated withdrawal will progress to seizures or delirium.Patients should undergo systematic assessment when theybegin detoxication and during the process. The Clinical Insti-tute Withdrawal Assessment for alcohol scale is a simple toolthat is widely used (Table 8-1).


� TABLE 8-1. Clinical Institute Withdrawal Assessment for AlcoholScale∗

Patient: _________________Time: ____: ____(24 hour clock, midnight = 00:00)Date: ___ /___ /___

y m d

NAUSEA AND VOMITING: Ask, “Do you feel sick to your stomach?Have you vomited?” Observation.0—no nausea and no vomiting1—mild nausea with no vomiting234—intermittent nausea with dry heaves567—constant nausea, frequent dry heaves and vomiting

TREMOR: Arms extended and fingers spread apart. Observation.0—no tremor1—not visible, but can be felt fingertip to fingertip234—moderate, with patient’s arms extended567—severe, even with arms not extended

PAROXYSMAL SWEATS: Observation.0—no sweat visible1—barely perceptible sweating, palms moist234—beads of sweat obvious on forehead567—drenching sweats

Pulse or heart rate, taken for1 minute: ___________________Blood pressure: ____ /____

TACTILE DISTURBANCES: Ask, “Have you any itching,pins-and-needles sensations, any burning, any numbness, or do you feelbugs crawling on or under your skin?” Observation.0—none1—very mild itching, pins and needles, burning, or numbness



2—mild itching, pins and needles, burning, or numbness3—moderate itching, pins and needles, burning, or numbness4—moderately severe hallucinations5—severe hallucinations6—extremely severe hallucinations7—continuous hallucinations

AUDITORY DISTURBANCES: Ask, “Are you more aware of soundsaround you? Are they harsh? Do they frighten you? Are you hearinganything that is disturbing to you? Are you hearing things that you knowaren’t there?” Observation.0—not present1—very mild harshness or ability to frighten2—mild harshness or ability to frighten3—moderate harshness or ability to frighten4—moderately severe hallucinations5—severe hallucinations6—extremely severe hallucinations7—continuous hallucinations

VISUAL DISTURBANCES: Ask, “Does the light appear to be toobright? Is its color different? Does it hurt your eyes? Are you seeinganything that is disturbing you? Are you seeing things that you knowaren’t there?” Observation.0—not present1—very mild sensitivity2—mild sensitivity3—moderate sensitivity4—moderately severe hallucinations5—severe hallucinations6—extremely severe hallucinations7—continuous hallucinations

ANXIETY: Ask, “Do you feel nervous?” Observation.0—no anxiety, at ease1—mildly anxious234—moderately anxious, or guarded, so anxiety is inferred567—equivalent to acute panic states as seen in severe delirium or acuteschizophrenic reactions

AGITATION: Observation.0—normal activity1—somewhat more than normal activity


234—moderately fidgety and restless567—paces back and forth during most of the interview, or constantly thrashesabout

HEADACHE, FULLNESS IN HEAD: Ask, “Does your head feeldifferent? Does it feel like there is a band around your head?” Do notrate dizziness or lightheadedness. Otherwise, rate severity.0—not present1—very mild2—mild3—moderate4—moderately severe5—severe6—very severe7—extremely severe

ORIENTATION AND CLOUDING OF SENSORIUM: Ask, “Whatday is this? Where are you? Who am I?”0—oriented and can do serial additions1—cannot do serial additions or is uncertain about date2—disoriented for date by no more than 2 calendar days3—disoriented for date by more than 2 calendar days4—disoriented for place and/or person

Total CIWA-A score ____Rater’s initials ____

Maximum possible score—67

∗ This scale is not copyrighted and may be used freely.

See Sullivan JT, Sykora K, Schneiderman J, Narango CA, Sellers EM (1989).Assessment of alcohol withdrawal: the revised Clinical Institute WithdrawalInstrument for Alcohol Scale (CIWA-AR). British Journal of Addiction84:1353–1357.

The scale should be administered at regular intervals(initially every 1–2 h) 6–24 h after the patient’s last drink untilthere are at least two consecutive assessments with scores lessthan 8–10; structured assessment can then safely end. Closermonitoring is indicated if the score exceeds 15; the interpreta-tion of intermediate scores depends on clinical judgment, takinginto account issues such as the history of withdrawal episodesand the degree of discomfort experienced by the patient.


Provided serial clinical assessments are carried out, it isrelatively straightforward to strike a balance between with-drawal and intoxication. The most common error associatedwith drug treatment of withdrawal is prescribing doses that aretoo small or dose intervals that are too long.

Benzodiazepines are the treatment of choice for alcoholwithdrawal. They are relatively safe compared with othersedative–hypnotic drugs; and they reduce the frrequency ofseizures and delirium. Alcoholics are cross-tolerant to theireffects and may therefore require doses that would be consid-ered high for non-tolerant individuals.

The benzodiazepines can be dividied into two classesaccording to their duration of action:

• Longer-acting agents include chlordiazepoxide (which haslower abuse potential and may therefore be preferred)and diazepam. These agents undergo both oxidation andglucuronidation. It is an advantage that blood levels ofthese agents decline gradually during the tapering processbecause this is associated with greater between-dose patientcomfort and possibly better seizure prophylaxis. The disad-vantage is that if elimination is delayed (for example, inelderly patients or patients with impaired liver or pulmonaryfunction), there is a risk that accumulation will causetoxicity.

• Shorter-acting agents include oxazepam and lorazepam,which undergo only glucuronidation. Their advantage is thatthey are well metabolized and eliminated by elderly people,with less chance of accumulation and toxicity in patients withliver disease. The disadvantage is that between-dose bloodlevels may decline abruptly, increasing the risk of break-through symptoms and seizures.

All may be administered orally or intravenously; onlylorazepam is available as an intramuscular injection.

The patient’s response should be determined by serialassessments and drug treatment should be adjusted accordingto clinical need. For example, increasing signs of withdrawalindicate the need for an increase in dose or a decrease in the


dose interval. A degree of sedation is desirable but if over-sedation occurs drug treatment should be withheld until it isagain clinically indicated. Recommendations for medicationare summarized in Figure 8-1.

Hallucinations that develop during delirium are relativelyrefractory to benzodiazepines alone; if they occur despitesubstitution therapy with a sedative–hypnotic, adjunctivehaloperidol 2–5 mg may be administered.

If persistent tachycardia or hypertension occur the possi-bility should be excluded that they may be due to inadequatelytreated withdrawal. Treatment with a beta-blocker or cloni-dine has been successful; these agents may also decrease vitalsigns and tremor but they do not prevent seizures.

As an alternative management strategy, a patient attendingfor detoxication may be ‘frontloaded’ with 1–2 hourly dosesof diazepam or chlordiazepoxide to achieve sedation. Because

Patient in AlcoholWithdrawal

<65 years old and no evidenceof liver or pulmonary dysfunction

>65 years old or evidence ofliver or pulmonary dysfunction

Day 1:

Day 2:

Day 3:

Day 4:

Chlordiazepoxide, 50–100 mg q6h



Chlordiazepoxide, 50–100 mg hs

Day 1:

Day 2:

Day 3:

Day 4:

Lorazepam, 2 mg q4h

Lorazepam, 1.5 mg q4h

Lorazepam, 1 mg q4h

Lorazepam, 0.5 mg q4h

FIGURE 8-1. Medication recommendations for treating alcoholwithdrawal. From Selzer J (2000). Drugs for Treating SubstanceAbuse. In Psychiatric Drugs, Lieberman JA and Tasman A (eds.) WBSaunders, pp. 214–241. © 2000, with permission from Elsevier.


these drugs have a long elimination half-life, their effectsdiminish slowly over the period of withdrawal. However, theneed remains for careful serial assessment while there is a riskof withdrawal complications.

Reports of experience with carbamazepine and valproatein the treatment of alcohol withdrawal are promising and theymay have a role in special circumstances. However, the safetyand efficacy of benzodiazepines is currently unsurpassed.

Thiamine is indicated for every patient with alcoholism toprevent Wernicke’s encephalopathy and Korsakoff’s amnesticsyndrome. The typical dose is 100 mg/day; replacement withthiamine should begin immediately and must precede admin-istration of intravenous glucose. Nutritional managementis completed with the addition of folate and multivitaminsupplements, though the need is less acute. Although somebelieve that magnesium supplementation prevents withdrawalseizures, there is no consensus on its role in the absence ofproven magnesium deficiency.

Withdrawal from Sedative–HypnoticsSymptoms of sedative–hypnotic withdrawal include:

• tremor• anxiety• insomnia• anorexia• nausea and vomiting• postural hypotension• seizures

These symptoms are qualitatively similar to those of alcoholwithdrawal and the Clinical Institute withdrawal assessmentscale is therefore useful to document symptom progression andthe response to treatment. Withdrawal from sedative–hypnoticdrugs is a medical emergency: if untreated there is a risk ofserious complications (such as hyperpyrexia) and death. Thenature of withdrawal symptoms is similar for all sedative–hypnotics but the rate of onset and decline depends on the


half-life of individual drugs. For short-acting agents (e.g., seco-barbital), withdrawal symptoms begin 12–24 h after the lastdose. By contrast, symptoms develop more slowly for long-acting agents (e.g., diazepam) and may become maximal oneweek after the last dose.

Similar to the treatment for alcohol withdrawal, benzodi-azepine taper (as described above) is a good choice, particu-larly if the patient is dependent on benzodiazepines.

An excellent alternative for substitution therapy insedative–hypnotic withdrawal is phenobarbital. This long-acting agent is associated with low between-dose fluctuation inblood levels; it has low potential for abuse and a wide marginbetween therapeutic and lethal blood levels. The symptoms ofphenobarbital intoxication (ataxia, slurred speech, nystagmus)are readily apparent and easy to respond to in a detoxicationprotocol.

In detoxication of sedative–hypnotic dependency, the firststep is to obtain a history of drug use. This enables the physi-cian to estimate the dose of phenobarbital equivalent to thetotal dose of sedative–hypnotic, as specified in Table 8-2.

The total phenobarbital equivalent dose is then summedand divided into a three-times daily dose regimen. The totaldaily dose of phenobarbital rarely exceeds 500 mg even forpatients with extreme dependence.

In the event that acute withdrawal symptoms emerge beforesubstitution therapy has begun, the first dose of phenobarbitalmay be administered by intramuscular injection. Withdrawalor intoxication symptoms should then be reassessed 1–2 h laterto determine the next dose of phenobarbital.

The degree of dependence can be assessed by serial admin-istration of pentobarbital 200 mg but it is not clear whether thisapproach is superior to direct substitution with phenobarbital,which is speedy, simple, and safe.

There should be no signs of sedative–hypnotic withdrawalor phenobarbital toxicity 24–48 h after beginning substitutiontherapy with phenobarbital and the dose reduction phase maythen begin. This involves a reduction in phenobarbital doseof 30 mg/day, maintaining the three times daily dose regimen.If phenobarbital toxicity occurs (slurred speech, nystagmus,ataxia), the next dose should be withheld and the total daily


� TABLE 8-2. Phenobarbital Withdrawal Equivalents of Sedative–Hypnotics

CLASS/GENERIC NAME DOSE (mg)∗

BENZODIAZEPINESAlprazolam 1Chlordiazepoxide 25Clonazepam 2Clorazepate 7�5Diazepam 10Estazolam 1Flurazepam 15Halazepam 40Lorazepam 2Oxazepam 10Prazepam 10Quazepam 15Temazepam 15Triazolam 0�25

BARBITURATESAmobarbital 100Butabarbital 100Butalbital 100Pentobarbital 100Secobarbital 100

OTHERChloral hydrate 500Ethchlorvynol 500Glutethimide 250Meprobamate 1200Methyprylon 200Zolpidem 5

∗ Dose equivalent to 30 mg of phenobarbital for withdrawal.

Adapted from Wesson DR, Smith DE, Ling W, Seymour RB: Chapter 17:Sedative-Hypnotics. In: Lowinson JH, Ruiz P, Millman RB, Langrod JG,editors: Substance Abuse: A Comprehensive Textbook, 4th ed. Philadelphia:Lippincott Williams and Wilkins; 2005.

dose should be reduced. In the event of objective signs ofwithdrawal, the daily dose of phenobarbital is increased andthe dose reduction phase is delayed until the patient is stabi-lized again.


Withdrawal from OpiatesOpiate withdrawal, by contrast with withdrawal from sedative–hypnotics, may be intensely uncomfortable but for most adults(with the important exceptions of adults with little reserve,such as those with advanced AIDS, and newborn infants) it isnot usually life-threatening. Nevertheless, easing symptoms canbe an important part of the process of engaging an individualin treatment for their opiate addiction and to facilitate othermedical treatment.

The time to onset of withdrawal symptoms depends onthe duration of action of the opiate. Symptoms typically begin6–24 h after the last dose of a short-acting agent such as heroinbut 48–72 h after longer-acting opiates like methadone. Theadministration of an opiate antagonist (e.g., naloxone) canprecipitate severe withdrawal symptoms, including: dysphoria,nausea and vomiting, muscle aches, lacrimation or rhinorrhea,dilated pupils, piloerection, diarrhea, diaphoresis, yawning,fever, and insomnia.

Methadone is approved by the FDA for the treatmentof opiate withdrawal. Regulations for its use differ betweenstates but they typically allow its use for inpatient detoxicationand methadone maintenance. Its use in an outpatient settingfor the management of opiate withdrawal is not permittedexcept as part of a licensed methadone maintenance treatmentprogram.

Oral methadone has a long duration of action. The initialdose of 15–20 mg is given when signs of opiate withdrawal(not simply reports of drug craving) are evident. An addi-tional 5–10 mg may be given 1–2 h later if symptoms persistor worsen. A dose of 40 mg/day usually controls signs of with-drawal well (note that the dose for the different indicationof long-term methadone maintenance may often be higher).When oral administration is impossible due to withdrawalsymptoms, methadone 5 mg may be administered as an intra-muscular injection. Having reached a dose that relieves with-drawal symptoms, the daily dose can be tapered by a gradual10–20% for full detoxication.

A newly available option for treatment of opioid with-drawal is the schedule III opioid partial agonist buprenor-phine, which is now available for use in office-based practice


by any physician who has taken a brief training and certi-fication. Although the use of schedule II drugs such asmethadone for the treatment of opiate dependence is restrictedto hospitals or specially licensed clinics (leading to a shortageof facilities where opiate-dependent individuals can receiveappropriate treatment), the Drug Addiction Treatment Actof 2000 introduced less stringent regulations, allowing theuse of narcotic drugs for the treatment of addiction in theoffice or any other health-care setting by any licensed physi-cian who has taken a brief training course and obtainedregistration, thereby increasing access to treatment. Twonew formulations of buprenorphine (Subutex and Suboxonesublingual tablets) were the first products to be approvedby the FDA under this Act for the treatment of opioiddependence. Subutex contains only buprenorphine in dosesof 2–8 mg; Suboxone also contains the opioid antagonistnaloxone (0.5 and 2 mg respectively); the purpose of thenaloxone is to discourage diversion of the medication forabuse intravenously (crushing the pills and injecting them),since naloxone is poorly absorbed after oral or sublingualadministration, but if injected it would produce precipitatedwithdrawal.

Buprenorphine is an excellent detoxification agent becauseof its long duration of action owing to very high affinity forand very slow dissociation from opioid receptors. Startingbuprenorphine must be done carefully, because of the partialagonism. If administered too close in time to the last doseof a full agonist such as heroin, buprenorphine will precipi-tate withdrawal, and the withdrawal produced can be atypicaland in rare cases has been observed to include delirium.Precipitated withdrawal is more likely among patients depen-dent on long-acting agonists (e.g., methadone) or high dailydoses of a shorter-acting agonist such as heroin. Thus, whenstarting buprenorphine, the clinician should wait for symp-toms of opioid withdrawal to begin to appear before givingthe first dose of buprenorphine, which should be a test dose of2 mg [Subutex 2 mg, or Suboxone (2 mg buprenorphine/0.5 mgnaloxone)]. If this dose is well tolerated, administer another2 mg 1 h later, and up to 8 mg total on the first day, and up


to 16 mg on the second day. After this, buprenorphine can betapered slowly to zero over 10 days to 2 weeks. Considerableflexibility in the taper schedule is possible, including a muchfaster taper, since the slow dissociation from receptors in itselfeffectively produces a taper. However, clinicians should alsobe alert for the emergence of low-grade withdrawal symptoms(fatigue, anxiety, mild flu-like physical symptoms) in the weeksafter discontinuing buprenorphine. This subacute or protractedwithdrawal can be observed from any opioid drug but canseem surprising with buprenorphine because the taper phaseof a buprenorphine detoxification is usually comfortable anduneventful.

If one of the first few doses of buprenorphine administeredis followed by a rapid worsening of withdrawal symptoms,this is precipitated withdrawal, and no further buprenorphineshould be given; at this point it is probably best to treat witha full agonist (methadone), although one could also wait forprecipitated withdrawal to clear and full-blown opiate with-drawal (from whatever the patient was addicted to) to emerge,after which one can try again beginning with a test dose of2 mg buprenorphine.

A number of nonnarcotic medications are useful intreating the symptoms of opiate withdrawal. These include the�-adrenergic receptor agonist and antihypertensive clonidine,which is particularly helpful with the autonomic symptomsof withdrawal as well as the anxiety, benzodiazepines (clon-azepam is typically used), which are particularly helpful foranxiety and insomnia, antiemetics, and NSAIDs for muscleaches (oral agents such as ibuprofen, or toradol which canbe given parenterally). Although with clonidine alone theautonomic symptoms may be well controlled, patients oftencomplain of greater subjective distress with clonidine than withan agonist such as methadone or buprenorphine.

Clonidine is used to assist detoxication in individualsusing illegal opiates in circmstances where methadone is notpermitted (e.g., outpatient settings) and to relieve abstinencesymptoms in a patient stopping methadone maintenance.In clonidine-assisted detoxication, the main side effects are


hypotension, which may cause doses to be withheld or treat-ment discontinued, and sedation. Hypotension may be wors-ened by diarrhea or vomiting, which are common in opiatewithdrawal. Patients should be encouraged to take plenty offluids, and sports drinks such as Gatorade are particularlyhelpful because they also supply electrolytes. Because cloni-dine by itself does not treat all aspects of withdrawal, cloni-dine is most effective in combination with the other agentsmentioned above. A protocol for clonidine-assisted detoxica-tion is given in Table 8-3.

A combination of clonidine with naltrexone has been usedto achieve a more rapid withdrawal followed by maintenancewith naltrexone. This strategy requires close monitoring ofthe patient and an experienced clinician to titrate the doseof clonidine against naltrexone-induced withdrawal symptoms.The acceptability of maintenance therapy with naltrexone toopiate addicts is disappointing.

With opiate detoxification, it is particularly important toconsider the indications for it, and to establish an adequatetreatment plan after detoxification is completed. Chronicopioid use induces tolerance, and detoxification reducesor eliminates tolerance. Because of the loss of tolerance,

� TABLE 8-3. Clonidine Detoxification

DAY SHORT-ACTING OPIATE∗ LONG-ACTING OPIATE†

Day 1 0.1 mg q4h 0.1 mg q4hDays 2–4 0.1–0.2 mg q4h (depending

on symptoms)0.1–0.2 mg q4h (depending

on symptoms)Days 5–10 Reduce daily dose by

0.2–0.4 mg/dMaintain on 0.4–1.2 mg/d

Day 11 Reduce daily dose by0.2–0.4 mg/d

Adjunctive medication: oxazepam (in limited supplies) for sleep or agitation;ibuprofen or toradol for muscle or bone pain; bismuth subsalicylate for diarrhea;prochlorperazine or odansetron for nausea.∗ For example, heroin.† For example, methadone.From Selzer J (2000). Drugs for Treating Substance Abuse. In PsychiatricDrugs, Lieberman JA and Tasman A (eds.) WB Saunders, pp. 214–241. © 2000,with permission from Elsevier.


detoxified opiate addicts are at increased risk of death fromopiate overdose; doses that they routinely self-administeredpreviously when tolerant could now be lethal, a problem exac-erbated by the variable and sometimes high potency of illicitheroin. In general, the risk of relapse following opioid detoxifi-cation is high. Therefore, patients should be assessed for over-dose risk, and those with a history of past overdoses, or withmultiple past relapses, should be encouraged to take agonistmaintenance treatment, with methadone or buprenorphine,rather than undergoing detoxification. For those not enteringagonist maintenance, a strong plan for psychosocial treatmentis important, for example long-term residential treatment ortherapeutic community, a good outpatient treatment program,supplemented by a self-help group (Alcoholics Anonymous,or Narcotics Anonymous).

Management of Withdrawal in Patients withMultiple DependenciesPatients may have dependencies on drugs from more than onepharmacological class. It is safest first to suppress each of theabstinence syndromes then attempt detoxication from the mostdangerous class of drugs first. In patients who are dependenton both opiates and sedative–hypnotics or alcohol, withdrawalfrom a sedative–hypnotic or alcohol may cause convulsions ordeath and is clearly more important than opiate withdrawal,which is associated with discomfort. In such cases, a steady doseof methadone or buprenorphine should be maintained whiledetoxication from sedative–hypnotics or alcohol is achieved;methadone detoxication may then proceed.

AGENTS TO AID RELAPSE PREVENTIONThe emergence of treatments to prevent post-detoxicationrelapse in patients with dependency is an exciting developmentin psychopharmacology. A wide variety of strategies has beendeveloped:

• aversive therapy depends on creating a biochemical environ-ment in which the effect of a substance is unpleasant ratherthan gratifying


• antagonist therapy blocks the gratifying effect of a substance(without replacing it with an aversive one)

• agonist therapy involves substituting a drug of abuse fora medicine which acts at the same receptor but has saferpharmacological properties

• less widely used, medications may be used to correct theputative neurochemical changes associated with drug abuse

• new agents are being developed that act at the neurophys-iological steps intermediate between the use and effect ofabused substances, such as synaptic transporters

There is evidence that psychosocial interventions increasethe effectiveness of drug treatment and psychopharmacologicalstrategies for relapse prevention are not intended to replacethem. In patients who receive intensive psychosocial inter-ventions that are relevant to drug dependency problems, theincremental benefit of adding medication to prevent relapseprevention is reduced. On the other hand, a psychosocial treat-ment tailored to take advantage of the effect of the medicationmay synergize to produce a stronger effect. The physician maysometimes be expected to justify—to both the patient andother clinicians—substituting one drug for another as part ofthe treatment for drug dependence. The argument rests onthe clear difference between taking medication prescribed bya physician and using drugs of abuse to which the patient isaddicted.

Medications for Alcohol DependenceThree medications are now FDA approved for treatment ofalcohol dependence: disulfiram, naltrexone, and acamprosate.

Disulfiram binds irreversibly to aldehyde dehydrogenase,the enzyme that catalyzes the oxidation of acetaldehyde (ametabolite of alcohol) to acetic acid. It therefore changes thebody’s response to alcohol. Ingestion of alcohol results in theaccumulation of acetaldehyde, causing an extremely dysphoricexperience of palpitations, hypotension, nausea and vomiting,and diaphoresis. This is self-limiting but it can produce substan-tial stress to the cardiovascular system, particularly with highdoses both of alcohol and of disulfiram. This is generally notdangerous in young, healthy individuals, but could lead to


cardiovascular collapse and death in patients with significantcardiovascular or renal disease. Disulfiram has also been asso-ciated with liver toxicity, and education of the patient aboutsigns of liver failure and periodic monitoring of liver functionsare advisable.

When the FDA approved disulfiram as a treatment foralcoholism, standards for efficacy studies were lower than theyare today. The best evidence suggests that, although disulfiramis associated with a reduction in the number of drinking daysin alcoholics who drink while taking it, compliance with medi-cation, not the effects of disulfiram, is probably important fora better outcome.

In the view of many clinicians, disulfiram may nonethelessprovide an important disincentive to drink in well-motivatedpatients. There is evidence that using disulfiram as part ofa contract involving the spouse offers the prospect of goodresults: the alcoholic agrees to allow the spouse to witness disul-firam ingestion and the spouse agrees to refrain from commenton the alcoholic’s drinking. If disulfiram is taken regularly, itwill prevent drinking, since even a small amount of intake willmake the patient sick.

The usual dose of disulfiram is 250 mg/day; this maintainsinhibition of acetaldehyde dehydrogenase while minimizingside effects such as lethargy. Some patients will report beingable to drink without consequence on 250 mg per day. This maysimply mean the patient is not actually taking the medicationregularly, and should prompt gentle inquiry into compliance.However, some patients do need higher doses, up to 500 mgper day. In rare cases, drug-induced psychosis may occur dueto inhibition of dopamine �-hydroxylase by disulfiram. Thereis also a risk of interactions between disulfiram and medica-tions (e.g., tricyclic antidepressant and phenytoin levels areincreased). Patients should be made aware that alcohol is acomponent of products other than beverages (e.g., shavinglotions). Significant inhibition of acetaldehyde dehydrogenase,and therefore the risk of reactions if alcohol is ingested, lastsfor up to two weeks after stopping disulfiram.

It is believed that the reinforcing effects of alcohol may bemediated by endogenous opiate systems and this hypothesisunderlies the use of naltrexone in alcohol relapse prevention.


Naltrexone combined with structured psychosocial treatmenthas been associated with improvements in complete absti-nence, less craving for alcohol when the patient was abstinent,and less drinking once drinking began. Psychosocial treat-ments that have proven particularly effective in combinationwith naltrexone are those, such as cognitive behavioral relapseprevention, that emphasize coping skills for handling varioussources of relapse risk. The recommended dose is 50 mg perday. This agent differs from disulfiram in that there is noadded ill effect once drinking begins. It may limit the severityof binges and diminish preoccupation with alcohol in absti-nent alcoholics. Nausea has been a common side effect earlyin naltrexone treatment for alcoholism, but is generally mildand clears with continued use. Naltrexone can produce livertoxicity, which is dose dependent, has mainly been observed atmuch higher doses (200 or 300 mg per day), and resolves withdose reduction. Elevated liver enzymes, which do not resolvewith dose reduction or discontinuation, represent anotherlikely cause of hepatitis (e.g., viral, alcoholic).

Acamprosate has been approved for a number of years inEurope, and it recently received FDA approval. Its mecha-nism is not well understood but is thought to involve inter-ference with excitatory amino acid mechanisms that may beinvolved in relapse. Clinical trials have shown a modest effectin reducing relapse risk, and studies are underway to determineif it may have complementary effects to those of naltrexoneand whether the combination of the two may increase effec-tiveness. Acamprosate is supplied in 333 mg tablets, and therecommended dosage is two tablets, three times daily. Blisterpacks, organized according to the daily schedule of dosing,can be helpful to patients in maintaining compliance. Acam-prosate is generally safe and well tolerated. Diarrhea is themost common side effect.

In the same way that any given antidepressant medica-tion will be helpful to some depressed patients but not others,naltrexone and acamprosate may similarly have little effectin some patients but work well in others. As of yet, thereare no reliable predictors of which alcohol-dependent patientwill benefit from these medications, but clinicians should beencouraged to attempt adequate trials (effective dose, for at


least a month) of these medications for patients with problemdrinking, and not be discouraged by the fact that some patientswill not benefit.

Medications for Cocaine DependenceGiven the high relapse rates with psychosocial treatment alone,there has been considerable interest in the development ofpharmacotherapies for cocaine dependence. Case reports andopen pilot studies have suggested that a variety of agentsmay be effective but double-blind trials have provided disap-pointingly inconsistent results. Surveys have demonstrated thatmany addiction medicine specialists are using medicines totreat cocaine dependence despite the lack of evidence of astandard sufficient for FDA approval.

The most widely prescribed medicines for cocaine depen-dence are antidepressants. Depression occurs frequently incocaine addicts and it is possible that antidepressants maycorrect neurotransmitter deficiencies associated with cocaineuse. The best evidence relates to desipramine but early reportsof its effectiveness in preventing relapse were subsequentlycontradicted and its efficacy for more than 6–12 weeks wasquestioned. Typical desipramine doses are the same as for thetreatment of depression. Concurrent use of an antidepressantand cocaine increases the risk of additive cardiotoxicity andpatients who recommence cocaine use during treatment shouldbe assessed for this possibility.

Although antipsychotic drugs have been proposed on thegrounds that blockade of dopamine receptors would attenuatethe euphoric response to cocaine, experience treating patientswith schizophrenia and cocaine dependence do not supportthis hypothesis. Furthermore, cocaine use is associated withdopamine depletion and may therefore increase the risk ofextrapyramidal effects.

Psychomotor stimulants have been suggested as an agonistmaintenance treatment strategy. Although early clinicalreports suggested stimulants might exacerbate cocaine depen-dence, more recent placebo-controlled trials suggest oraldexedrine may have promise. Patients with attention deficitdisorder, which commonly co-occurs with cocaine dependence,


may benefit from stimulants. Similarly, the effectiveness oflithium in reducing cocaine use is confined to patients withcomorbid bipolar disorder. The anticonvulsant carbamazepinehad disappointing effects in controlled trials after initiallypromising results in open studies. However, more recently,some of the newer anticonvulsants with GABA-enhancingor excitatory amino acid inhibiting effects (e.g., topiramate,gabapentin) have shown promise in small placebo-controlledtrials.

Interestingly, perhaps the most consistent data so farrelate to disulfiram, which has shown promise in a series ofsmall placebo-controlled trials for cocaine dependence, andthe effect does not seem to depend on concurrent alcoholuse. Since disulfiram inhibits dopamine a-hydroxylase, whichcatalyzes the conversion of dopamine to norepinephrine in thebrain, it may promote increased levels of brain dopamine andcombat the dopamine depletion engendered by cocaine.

Another pharmacotherapeutic strategy for cocaine depen-dence is to treat comorbid psychopathology, most commonlyunipolar depression, bipolar disorder, attention deficit hyper-activity disorder, and anxiety disorders. Such treatment is mostlikely to have a direct impact on the psychopathology, followedby indirect effects of improved functioning and reduced druguse. The benefits of drug treatment in the absence of comorbidpsychopathology is unclear. An effective pharmacotherapy forcocaine dependence is urgently needed.

Medications for Opiate DependenceAgonist maintenance treatment for opiate dependence is apowerful treatment with a large effect size. When properlyprescribed, methadone will rapidly induce a dramatic remis-sion in 50% or more of patients. It prevents or reduces illicitopiate use, craving for illicit opiates, criminal behavior associ-ated with acquisition of illicit opiates, and diseases associatedwith illicit opiate use (such as illness related to infection withhuman immunodeficiency virus), and improves employmentand other aspects of social functioning. Methadone has alsobeen shown to reduce mortality rates among opioid dependentpatients, in part by protecting against overdose. Methadone


is also sometimes misunderstood as ‘substituting one drug foranother.’ In fact, methadone works by inducing marked toler-ance such that effects of other opiates are blocked, and noeuphoric effects of the methadone itself are experienced. Whenprescribed with careful titration, methadone is neither intoxi-cating nor sedating, and it does not interfere with performanceof functions that are important for responsible adult roles (e.g.,studies have shown that methadone does not impair drivingability).

Methadone is well-suited for use as maintenance treatment.It is effective after oral administration and has a long half-life�>24 h�; it suppresses opiate withdrawal syndrome for up to36 h and it blocks euphoria induced by other opiates. Chronicadministration is associated with minimal side effects, the mostfrequent of which are constipation, excess sweating, reducedsexual interest, but these rarely result in discontinuation oftreatment.

The initial dose of methadone maintenance lies in therange 20–40 mg and this is mainly to relieve symptoms dueto abstinence. This is followed by an ’induction period’, inwhich the dose may be increased in increments of 5–10 mguntil a dose is achieved that prevents opiate craving and blocksthe euphoric effects of illicit opiates. Measures for patientmonitoring include subjective reports, interval history, andregular frequent usine toxicology analysis. Although a doseof 40 mg/day may be sufficient for some patients, there isnow evidence from several trials that most require 80 mg/day.According to Federal regulations, doses of methadone greaterthan 120 mg/day require permission, although individual statescan specify lower ceiling doses. As with many psychophar-macologic treatments, the most common reason for treat-ment failure in methadone maintenance is inadequte dose,and continued opiate use should prompt consideration of adosage increase. Some evidence suggests that optimal treat-ment response requires trough methadone blood levels inthe range 200–400 ng/mL, and blood level monitoring may beuseful in nonresponders. Some patients are rapid metabolizers,which can be assessed by comparing peak and trough bloodlevels. Rapid metabolizers may benefit from a divided, twicedaily dose schedule.


A pragmatic drawback of methadone maintenance is thatregulations stipulate that it can only be administered atspecially licensed clinics that require frequent attendance(daily at the outset) and that are not even available in manygeographic regions. This can be a practical constraint or adisincentive for many patients. On the other hand, it hasbeen shown that the effectiveness of methadone maintenancedepends upon regular counseling in conjunction with the medi-cation, which is a requirement of methadone clinics, and moreseverely dysfunctional patients probably benefit from the struc-ture imposed by clinic rules. Furthermore, many of the bettermethadone clinics offer primary medical and psychiatric care,which is important since chronic opiate-dependent patientsoften have multiple medical problems (e.g., hepatitis B and C,HIV) and psychiatric problems (e.g., depression, PTSD).

A recent development is the approval and marketing of thelong-acting opiate partial agonist buprenorphine (see also thesection above on opiate withdrawal on page 233), which hasbeen shown in clinical trials to have effectiveness equivalent tothat of methadone for maintenance treatment. A major differ-ence is that buprenorphine can be prescribed by any physicianwho has taken a brief training course and received certification,making it more widely available than methadone maintenance.The main regulatory restriction is that individual physicianspracticing in an office-based setting are restricted to treatingno more than 30 patients with maintenance buprenorphine atany one time.

Buprenorphine has interesting pharmacologic properties.It is a partial agonist, meaning that it binds opiate recep-tors but only partially activates them. This may translate intolower abuse potential compared with full agonists, althoughbuprenorphine has been abused by intravenous injections inother countries where it was widely available. The sublingualformulations marketed for treatment of opioid dependence doappear to have limited abuse potential by themselves, and theSuboxone formulation, which includes naloxone, discouragesattempts to extract and inject the contents, since intravenousnaloxone will precipitate withdrawal; the naloxone is poorlyabsorbed by the sublingual or oral routes. Buprenorphine bindsalmost irreversibly to opiate receptors, and dissociates very


slowly, accounting in part for its long duration of action. Whenproperly dosed, similarly to methadone, it induces tolerance,blocks the effects of other opiates, and produces little or nosedating or intoxicating effects.

The buprenorphine/naloxone combination (Suboxone) ispreferred for both detoxification and maintenance treat-ment, although some patients may be more sensitive to thepresence of the antagonist (naloxone) and tolerate straightbuprenorphine (Subutex) better. Because it is a partial agonist,buprenorphine will precipitate withdrawal in individuals whohave recently used any opioid drug; treatment should there-fore begin when there are clear signs of withdrawal (or at least4 h after last use of a short-acting opioid) (see also the sectionon buprenorphine for detoxification on page 234). There isless experience of induction with buprenorphine in individ-uals using long-acting agents such as methadone, but the riskof precipitated withdrawal is greater. The daily methadonedose should be below 40 mg per day before buprenorphineinduction is attempted, and a delay of around 48 h or moreis advisable to allow withdrawal symptoms from methadoneto manifest clearly. Induction is completed over 2–4 days,depending on the target dose. The recommended dose onday 1 is 16 mg, increasing to 16 mg on day 2 and there-after, and more gradual induction may be associated with ahigher risk of drop-out. The dose should be adjusted in incre-ments of 2–4 mg to that which keeps the individual in thetreatment program and suppresses withdrawal symptoms; thetarget maintenance dose is 16 mg/day but may range from4 to 32 mg/day. Buprenorphine should be administered aspart of a psychosocial treatment program. The relative easeof withdrawing from buprenorphine may result in a greatertendency to leave the treatment program compared withmethadone.

Owing to its long duration of action, buprenorphine can beadministered every other day (e.g., 32 mg every other day), oreven twice per week; this property can be useful for patientswhere there are concerns about compliance, since the medi-cation can be held at a clinic or by a significant other andadministered under observation on a less than daily basis.


Buprenorphine can produce sedation. However, emergenceof sedation should also raise suspicion of use of other drugsor alcohol. Unlike full opiate agonists (heroin, methadone,other narcotic analgesics), where respiratory depression is aserious risk, buprenorphine by itself produces less respiratorydepression. The rate of deaths from drug overdose droppedsubstantially in France after buprenorphine was introduced fortreatment of drug dependence. The one exception was over-doses of buprenorphine in combination with benzodiazepines,where deaths were observed. This has led to an exaggeratedconcern that buprenorphine is contraindicated in patients whouse benzodiazepines. For patients using benzodiazepines atregular, modest doses, which is the most common pattern evenamong opiate addicts, buprenorphine is safe. Patients who takelarge doses or binges of benzodiazepines are at risk of over-dose in combination with a variety of other drugs, includingbuprenorphine, and alcohol. It is likely that the risk of over-dose in such patients would be the same on either methadonemaintenance or buprenorphine maintenance.

Naltrexone is a long-acting (24–48 h duration) opioid antag-onist available in 50 mg tablets. It is effective in blocking theeffects of opioids and can be used as a maintenance treat-ment, but its effectiveness has been limited by poor compli-ance. Compliance can be improved with behavioral therapy,but rates of retention in treatment still remain well below whatcan be expected from agonist maintenance with methadoneor buprenorphine. Furthermore, naltrexone does not protectagainst opiate overdose; patients who stop naltrexone are nottolerant and are therefore vulnerable to overdose. Naltrexoneis also complicated to manage. It cannot be started until apatient has been fully detoxified, in order not to precipitatewithdrawal. Rapid induction methods using buprenorphine,clonidine, and clonazepam have been described, but gener-ally require 5–7 days to carry out. Anesthesia-assisted rapiddetoxification and induction onto naltrexone has been shownto involve the same level of discomfort, with increased riskof serious adverse events, and is not recommended. Once apatient is inducted onto naltrexone, if they stop taking thenaltrexone and relapse, naltrexone cannot be resumed withoutprecipitating withdrawal, and repeat detoxification is needed.


In summary, although some patients benefit from naltrexone,it is considered a second-line agent for patients who have failedor refuse agonist treatment. Patients maintained on naltrexoneshould be warned about the risk of fatal drug overdose ifnaltrexone is discontinued.

Special ConsiderationsTwo patient populations warrant particular comment: patientson medical-surgical units who are methadone-maintained andpatients who are pregnant. When patients are admitted tohospital their maintenance program should be contacted toverify their methadone dose. This dose should be maintained,not reduced, during the stress of an illness and its treat-ment. Maintenance methadone will prevent opiate withdrawalbut it will not provide analgesia. Patients who are in severepain should therefore receive analgesia with a non-opiate or,bearing in mind that higher doses and shorter dose inter-vals may be needed, a short-acting opioid analgesic. Opioidswith mixed agonist-antagonist properties, such as pentazocineand buprenorphine may precipitate withdrawal and should beavoided.

Longitudinal studies have shown that in utero exposureto methadone is not associated with abnormal development.Methadone maintenance should therefore be continued bywomen who become pregnant, though the dose may need to bereduced during the third trimester. The neonate may developabstinence symptoms and this should be expected and treatedafter birth.

Drug Treatments for Nicotine DependenceNicotine resembles other addictive drugs in that it inducessimilar addictive behavior patterns (such as inability to controlconsumption) and similar neuroadaptive changes (e.g., toler-ance). Nicotine intoxication and withdrawal are not associatedwith the acutely harmful behavior that other drugs of abusecause but the health consequences of nicotine make it themost important drug of abuse. Withdrawal symptoms associ-ated with nicotine addiction include anxiety, restlessness, diffi-culty concentrating, and irritability. Drug treatment to prevent


relapse in individuals with nicotine addiction must reduce thesesymptoms and in addition decrease craving for nicotine. Thismust be achieved in an environment with multiple conditionalstimuli for nicotine use, such as the morning cup of coffee.

Nicotine can be delivered safely and its use graduallyreduced using nicotine replacement therapy. By contrast withmethadone maintenance, nicotine replacement therapy is notan indefinite maintenance treatment but it is worthwhile spec-ulating whether nicotine replacement reduces harm comparedwith cigarette smoking. Nicotine replacement therapy iscurrently available as a transdermal patch and polacrilex ’gum’and both are effective in promoting abstinence.

The FDA recently approved a sustained-release formu-lation of bupropion for the treatment of nicotine depen-dence. It is as effective as nicotine replacement therapy. Thesuggested dose is 150 mg twice daily, beginnning two weeksbefore smoking cessation is attempted. In patients who deriveno benefit from nicotine replacement or bupropion alone, acombination of the two therapies has been shown to be safeand more effective than monotherapy.

Many patients who succeed in initiating abstinence fromcigarettes will relapse within 3–6 months. Clinicians andpatients should not be discouraged by this. The data suggestthat most patients who make repeated quit attempts eventuallysucceed in achieving sustained abstinence.

PHARMACOTHERAPIES FOR SUBSTANCE ABUSERSWITH ADDITIONAL PSYCHIATRIC ILLNESSThere is consistent evidence that patients with psychiatricillness have higher than expected rates of substance abuse, andthat patients undergoing treatment for substance dependencehave higher than expected rates of psychiatric illness.

In this challenging patient population, clinical experiencehas confirmed that:

• treatment should be administered by health profes-sionals with expertise in general psychiatry and chemicaldependence

• it is unrealistic to insist on abstinence as a condition fortreatment


• abstinence is not necessary for the safe and effective use ofmost psychotropic drugs

• symptom stability may be needed before a reduction insubstance use is apparent

• substance use disorder tends to persist unless it is treated asan illness in itself, regardless of the effective management ofother psychopathology

This section summarizes the management of particularpsychiatric illnesses complicated by substance use and endswith a summary of important drug interactions in substanceuse disorders.

Pharmacotherapy for Specific Psychiatric DisordersIt is difficult to be definitive about when a substance-inducedmood disorder becomes a major depressive episode. Accordingto most experts, there should be a period of sobriety lastingtwo weeks before a second diagnosis can be made in apatient presenting with a depressive syndrome. However, clin-ical trials have shown the benefit of antidepressant medica-tion in patients who are not abstinent to begin with. Thus,there is room for clinical judgment, and, if significant depres-sion persists and the patient cannot achieve abstinence, trialsof antidepressant medications can be attempted. In a patientvulnerable to substance abuse relapse, the most worryingeffects of antidepressant medication are cardiotoxicity (e.g.,arrhythmias), neurotoxicity (e.g., seizures) or death from inten-tional overdose. These concerns are greatest with tricyclicantidepressants and the selective serotonin reuptake inhibitorshave a safer pharmacologic profile. Although it is important todiscourage individuals from using drugs of abuse, a significantnumber of people taking an antidepressant use alcohol andother potential drugs of abuse in moderation without ill effects.There seems to be little abuse potential with antidepressantsthough abuse of amitriptyline for its sedative effects and, morerecently, fluoxetine for its stimulant effects, has been reported.

Substance abuse or dependence are frequent complica-tions of schizophrenia; conversely, the psychotic symptoms thatoccur during drug intoxication and withdrawal complicate thediagnosis of schizophrenia. These patients need a particularly


high degree of patience and sophisticated psychiatry services.Long-term follow-up (more than one year) of patients withschizophrenia and substance abuse suggests the need for ener-getic engagement in treatment for psychosis with an emphasison compliance with antipsychotic medication and the maladap-tive effects of substance misuse. The problem of substanceabuse declines for most patients after a year of medicationcompliance and treatment for addiction. Long-acting depotantipsychotics are a rational component of such a strategy. Acombination of medication and contingency contracting hasalso been used for patients with schizophrenia, in which unsu-pervised use of disability benefits is contingent on negativeurine toxicology.

Substance-induced disorders (particularly stimulant intox-ication and alcohol or sedative–hypnotic withdrawal) canresemble generalized anxiety disorder or panic attacks, andthus, as with depression, at least a two week period ofabstinence is preferable prior to initiating pharmacotherapy,although again there is room for judgment. Other anxietydisorders, such as social phobia, agoraphobia, PTSD, or OCD,have distinctive symptoms that do not overlap with symptomsof toxicity or withdrawal. Behavioral approaches are effec-tive for many anxiety disorders and should be considered, firstalone and then as a supplement to pharmacotherapy. Antide-pressants are effective for panic disorder or generalized anxietydisorder and have less abuse potential than the benzodi-azepines. Buspirone may be beneficial for generalized anxietydisorder at a dose of at least 45 mg/day. If a benzodiazepineis still preferred, expert opinion (supported by some experi-mental evidence) suggests that the safest agents are oxazepamand chlordiazepoxide because their onset of action is moregradual and they have a lesser tendency to produce euphoria.In general, however, benzodiazepines should be avoided owingto their abuse potential.

Substance abuse and bipolar disorder (particularly in themanic phase) are frequently comorbid disorders but there islittle information about the effects of treatments for bipolardisorder on substance abuse. Research in the treatment ofalcohol and cocaine dependence suggests that concurrent useof lithium is relatively safe with regard to drug interactions but


little is known about the risk of interactions between carba-mazepine and valproate and drugs of abuse in patients withbipolar disorder.

Follow-up of children with attention deficit–hyperactivitydisorder (ADHD) and conduct disorder has revealed highrates of substance abuse, and ADHD may be over-representedin substance abuse treatment settings. It is possible thatpatients with ADHD may seek stimulants for self-medication.The diagnosis of ADHD requires evidence of illnessduring childhood, preferably with verification from anotherperson. These patients have benefited from methylphenidate,according to some reports. Desipramine and bupropion havebeen used successfully in treating children with ADHD andboth have been advocated in patients with comorbid substanceabuse.

Drug Interactions in Chemical DependencyDrug interactions in patients with chemical dependency maybe pharmacodynamic or pharmacokinetic, and may occurbetween drugs of abuse or between drugs of abuse andprescribed medications.

ALCOHOLThe combination of alcohol and cocaine has been associatedwith fatal cardiac events and increased hepatoxicity. Alcoholin combination with an opiate or cannabis cause greatersedation and greater neurologic impairment than when usedalone. Acute alcohol intoxication is associated with inhibi-tion of hepatic enzymes and may increase blood levels ofconcurrent medication. Conversely, chronic alcohol use causesenzyme induction, leading to lower blood levels of antipsy-chotic drugs, antidepressants, valproic acid, carbamazepine,and some benzodiazepines. Griseofulvin, metronidazole, chlo-ramphenical and oral hypoglycemic agents may cause milddisulfiram-like reactions with alcohol. Alcohol and chloralhydrate compete for alcohol dehydrogenase; concurrent inges-tion leads to higher blood levels of both and increased intoxi-cation (this combination is known as a ‘Mickey Finn’).


COCAINE AND OTHER STIMULANTSThe risk of tachycardia is greater with a combination of cocaineand cannabis than with either drug alone. A ‘speedball’ (heroinand cocaine) more easily produces respiratory depression thanthe single drugs. Taking cocaine or amphetamine during orshortly after treatment with a monoamine oxidase inhibitorcan lead to hypertensive reactions.

OPIATESMethadone increases blood levels of concurrently takenantidepressants and zidovudine. Conversely, blood levels ofmethadone may be reduced to the point where abstinencesymptoms occur by concurrent use of rifampin, phenytoin,barbiturates, and carbamazepine. The combination of meperi-dine and a monoamine oxidase inhibitor may cause extremereactions ranging from collapse (hypotension and coma) toexcitation (hypertension and convulsions).

NICOTINECigarette smoking reduces blood levels of caffeine; patientswho stop smoking may therefore experience caffeine toxicityand this could be mistaken for nicotine withdrawal. Smokingalso reduces blood levels of antipsychotic drugs, tricyclicantidepressants, theophyline, and propranolol.

HALLUCINOGENSA serotonin reuptake inhibitor reportedly exacerbatedhallucinogen-persisting perception disorder (flashbacks) inadolescents with a history of LSD use. This is not surprisinggiven the role of serotonergic mechanisms in hallucinogenicdrug effects.

ADDITIONAL READING

1. Substance Abuse: A Comprehensive Textbook (4th Edition).Edited by Lowinson JH, Ruiz P, Millman RB, Langrod JG.Lippincott Williams and Wilkins, Philadelphia, 2005

2. Principles of Addiction Medicine (3rd Edition). Edited by GrahamAW, Schultz TK, Mayo-Smith MF, Ries RD, Wilford BB. AmericanSociety of Addiction Medicine, Inc., Chevy Chase, Maryland, 2003


3. Textbook of Substance Abuse Treatment (3rd Edition). Edited byGalanter M, Kleber HD. American Psychiatric Publishing, Inc.,Washington, D.C., 2004

4. Clinical Textbook of Addictive Disorders (3rd Edition). Editedby Frances RJ, Miller SI, Mack AH. The Guilford Press, NewYork, 2005

5. International Handbook of Alcohol Dependence and Problems.Edited by Heather N, Peters T, Stockwell T. John Wiley and SonsLtd., Chichester, UK, 2001

INDEX

acamprosate 240–1acute depression 99–109acute mania 90–9ADHD see attention

deficit/hyperactivitydisorder

age factors 57agitation 21agranulocytosis 35, 91akathisia 25–6, 73alcohol 129–30, 216–17,

223–30, 238–41, 251allergic effects 75alprazolam 132, 134, 138,

145–6, 153, 218Alzheimer’s disease 199–213amantadine 26, 202amisulpride 11, 16, 24, 30amitriptyline 42, 45, 143,

249amoxapine 45amphetamines 22–3, 175–98,

219–20anesthesia 38anorexia nervosa 51, 193antacids interactions 38anticholinergic effects 70–1anticonvulsants

acute depression 99–103,105, 108–9

acute mania 91–2, 94–7

anxiety 135, 138–9, 147drug interactions 36–7,

120–3indications 83, 90pharmacology 86–9sedative–hypnotic agents

166side effects 116–18, 124–5

antidepressants 41–80acute depression 99–102,

105acute period 63anxiety 129, 131–4, 139–45,

149attention deficit/hyperactivity

disorder 181–2, 194childhood disorders 52combination therapies

19–20, 64–6continuation period 67discontinuation 69dosages 58–61drug interactions 36, 76–9drug monitoring 58, 62drug selection 55–76early period 62–3indications 49–55initiation of treatment 58maintenance period 67–9pharmacology 41–9


256 INDEX

antidepressants (Continued)predictors of poor

response 57pregnancy 57recommendations 80responsiveness 63–7sedative–hypnotic agents

165side effects 41, 43, 58, 62,

69–76, 149–51substance abuse 241, 249–51suicidality 74–5

antiepileptics 196antihistamines 158, 165antipsychotics 1–39

agitation 21anxiety 135, 139, 147–8bipolar disorder 19clinical profiles 15–19combination therapies 8–10,

12, 19–20delirium 20delusional disorder 20dementia 21developmental disorders

21–2dosages 5–6, 8–12drug interactions 36–9,

76–8, 123drug selection 7–23drug switching 12–13Huntington’s disease 22indications 4–7initiation of treatment 8–12maintenance treatment

13–14major depression 19–20mental retardation 21–2mood stabilizers 94, 105,

109, 118–19, 123pharmacology 1–4pregnancy 38–9psychosis 19–20, 21

schizoaffective disorder 15schizophrenia 1, 4, 7–19side effects 1–3, 8–10, 14,

18, 20–36, 118–19, 125substance abuse 241,

250–1substance-induced psychoses

22–3Tourette’s disorder 22treatment evaluation 12–13treatment resistance 14–15violent patients 15see also atypical

antipsychoticsanxiety 53, 58, 226anxiolytics 129–54

combination therapies 148contraindications 152–3dosages 148–50drug interactions 136,

151–2drug selection 148–50indications 139–48initiation of treatment

148–50mood stabilizers 93–4pharmacology 131–9pregnancy 152side effects 136, 142, 149special precautions 152–3

aripiprazoleanxiety 135bipolar disorder 19costs 6dementia 21drug interactions 36–8mood stabilizers 84, 87, 92,

98, 101pharmacology 1–4schizophrenia 7–11, 16side effects 24–5, 28, 30

atomoxetine 43, 175–7, 179,182–5, 189–90, 192, 196

INDEX 257

attention deficit/hyperactivitydisorder (ADHD)

adults 189antidepressants 52psychostimulants 175,

179–83, 186–90, 193–7substance abuse 251

atypical antidepressants 43,47–9, 52–4, 61

atypical antipsychoticsacute mania 92, 94–5,

97–9anxiety 135, 139, 147–8drug interactions 123indications 83, 90pharmacology 87, 90side effects 118–19, 125

autism 52autonomic effects 70azopyrones 130

barbiturates 129, 165–6,217–18, 231–2

BDD see body dysmorphicdisorder

benzodiazepinesanxiety 129–30, 132, 136,

140, 145–6combination therapies 8–10,

12, 15, 58drug interactions 136, 151–2extra-pyramidal symptoms

26mood stabilizers 90, 92–3sedative–hypnotic agents

158–9, 161–4, 166–8,170–1

side effects 150–2, 166–7substance abuse 217, 219,

222, 229–33, 236, 250substance-induced psychoses

22–3withdrawal syndrome 153

benztropine 10, 26

beta-blockersanxiety 136, 138, 146–7drug interactions 36extra-pyramidal symptoms

26lithium toxicity 113substance abuse 229

bethanechol 70binge eating 52bipolar disorders

antipsychotics 19mood stabilizers 101, 106,

126substance abuse 242, 250–1

body dysmorphic disorder(BDD) 51

breastfeeding 57bromocriptine 29, 202bulimia 41, 51buprenorphine 219, 234–5,

237, 244–6bupropion

anxiety 137, 144, 149attention deficit/hyperactivity

disorder 181–2combination therapies 66drug interactions 76indications 51, 54initiation of treatment 58maintenance therapy 68pharmacology 43, 47–8side effects 70–5substance abuse 248, 251

buspirone 130, 134, 136, 138,146, 148, 250

calcium channel blockers 203,206

carbamazepineanxiety 147drug interactions 76mood stabilizers 84–97,

103–5, 108–11, 115, 117,120, 124–5

258 INDEX

cardiovascular effects 32–3CBT see cognitive-behavioral

therapycentral serotonergic syndrome

75chloral hydrate 159–60, 163,

168, 232, 251chlordiazepoxide 229, 250chlorpromazine 2–3, 5, 10,

32–3, 34cholinesterase inhibitors

200–1, 205–13chronic obstructive pulmonary

disease (COPD) 152, 156,169

cimetidine 77citalopram 44, 46, 48, 133, 137,

141–2clomipramine 45, 143, 148clonazepam

anxiety 132, 134, 138, 145–6mood stabilizers 93substance abuse 236, 246

clonidineAlzheimer’s disease 202attention deficit/hyperactivity

disorder 181–2, 194substance abuse 229, 235–6,

246clozapine 1, 6, 10–13, 16–17, 19,

24–5, 28–38cluster headaches 53co-morbidities 57cocaine 22–3, 219–20, 241–2,

252cognitive enhancers 199–213

costs 209dosages 209drug interactions 211–12drug selection 205–6indications 205initiation of treatment 206–7maintenance treatment

207–8

pharmacology 199–204side effects 209–11treatment evaluation 208–9

cognitive-behavioral therapy(CBT) 166

Conners Teacher/Parent RatingScales 183–4, 186

COPD see chronic obstructivepulmonary disease

coumarin 196

dantrolene 29delirium 20delirium tremens 224–5delusional disorder 20dementia 21depakene 110dermatological effects 115desipramine 45, 58, 77, 251developmental disorders

21–2dextroamphetamine 175–9,

184, 187, 190, 195, 197diazepam

anxiety 132, 138, 146substance abuse 220–1, 229,

231diphenhydramine 26, 157–8diphenylbutylpiperidines 20disulfiram 238–9, 242diuretics 113divalproex see valproatedonezepil 200–1, 203–4, 207,

209–11dopamine agonists 29, 38doxepin 45, 72doxylamine 158duloxetine

antidepressants 43, 47, 49,53–5, 70, 72, 79

anxiety 137, 143dysphoric mania 83dystonia 25–6

INDEX 259

ecstasy 220–1electroconvulsive therapy

(ECT)acute depression 100, 105acute mania 94major depression 67

endocrine effects 30epilepsy 72EPS see extrapyramidal

symptomsescitalopram 44, 46, 48, 133,

137, 141estazolam 163estrogens 202eszopiclone 160–1, 163–4, 169,

171–2ethchlorvynol 232ethnicity 56–7euphoric mania 83extrapyramidal symptoms (EPS)

1–3, 18, 20, 24–6, 118

fentanyl 218flashbacks 221, 252flumazenil 168, 217fluoxetine

antidepressants 42, 44–6,49–50, 52–4, 66, 72, 79

anxiety 133, 137, 141–2combination therapies 84mood stabilizers 84substance abuse 249

fluphenazine decanoate 4, 6, 25flurazepam 163fluvoxamine 44, 46, 133, 137,

141–2

gabapentin 87, 135, 138–9, 147,150

GAD see generalized anxietydisorder

galantamine 200–1, 204, 207,209–10, 212

gastrointestinal effects 34, 73,115

gender 56generalized anxiety disorder

(GAD)antidepressants 50anxiolytics 130, 136, 139–40,

142–3, 145–6Gingko biloba 203glutethimide 232guanefacine 202guanethidine 77, 196

hallucinogens 220–1, 252haloperidol

antipsychotics 2–6, 10, 20,22, 25, 27–8, 36

mood stabilizers 92substance abuse 216, 222,

229hematological effects 34–5, 75,

116hepatic effects 34, 76, 116heterocyclic antidepressants 36Huntington’s disease 22hydergine 202, 206hydroxyzine

hydrochloride/pamoate165

hypnotics see sedative–hypnoticagents

iloperidone 16, 24imipramine 41–2, 45, 58, 143insomnia 194

see also sedative–hypnoticagents

intoxication syndromes 215–22iproniazid 41isocarboxazid 45

Korsakoff’s amnestic syndrome230

260 INDEX

lamotrigineanxiety 147mood stabilizers 84–7, 89,

99–103, 105, 109, 120–7leukopenia 34–5levodopa 23, 38lithium

acute depression 99–108acute mania 90–6antidepressant augmentation

63–4costs 110drug interactions 119–20,

122indications 83–4, 90pharmacology 85–8side effects 111–16, 124substance abuse 242

lorazepamanxiety 134, 138, 146mood stabilizers 93–4substance abuse 216, 218,

222, 229loxapine 2lysergic acid diethylamide (LSD)

220–1, 252

major depression 19–20, 53,55–76

MAOIs see monoamine oxidaseinhibitors

maprotiline 45MDMA 220–1memantine 202, 204–5, 207–9,

211–12mental retardation 21–2, 180meperidine 78, 218meprobamate 129, 232mescaline 220–1metabolic effects 31–2methadone 233, 237,

242–4methamphetamine 22–3

3, 4-methylenedioxy-N-methylamphetamine(MDMA) 220–1

methylphenidates (MPH) 77,175–81, 184, 187–91,195–6

methyprylon 232migraines 53milnacipran 43mirtazapine

antidepressants 43–4, 47,49–50, 53, 79

anxiety 133–4, 137, 144,149

sedative–hypnotic agents165

mixed episodes 83mixed serotonin

antagonist/reuptakeinhibitors 43, 47, 49,61, 78

mixed serotonin/noradrenalineantagonists 43–4, 47, 49,61, 78

moclobamide 144modafinil 182monoamine oxidase inhibitors

(MAOIs)acute depression 102anxiety 129, 132, 137, 140,

144, 149drug interactions 76drug selection 56, 58–9, 62,

66, 69, 75indications 50–1, 54pharmacology 41–2, 44–5side effects 149

mood stabilizers 83–127acute depression 99–109acute mania 90–9anxiolytics 93–4breakthrough episodes

103–4costs 110–11

INDEX 261

dosages 94–9, 102–3, 107drug interactions 119–23drug selection 90–111indications 83–5, 90initiation of treatment 94–9,

102–3maintenance therapy 104–9pharmacology 85–90pregnancy 117, 124–6rapid cycling 83, 110side effects 111–19steroid-induced psychosis 23

MPH see methylphenidate

naloxone 218–19, 244–5naltrexone 236, 239–41, 246–7narcolepsy 190narcotics 78nefazodone

antidepressants 43, 47,49–50, 53, 75

anxiety 133–4, 137, 144,149–50

neuroleptic malignant syndrome(NMS) 29–30

neuroleptics see antipsychoticsneurologic disease 83neurologic effects 72–3neurotrophic factors 202nicergoline 202nicotine 38, 247–8, 252NMS see neuroleptic malignant

syndromenon-steroidal anti-inflammatory

drugs (NSAIDs) 203,235

norepinephrine reuptakeinhibitors (NRIs) 43

nortriptyline 45, 56, 58, 72NRIs see norepinephrine

reuptake inhibitorsNSAIDs see non-steroidal

anti-inflammatory drugs

obesity, antidepressants 52obsessive–compulsive disorder

(OCD)antidepressants 41, 50anxiolytics 131, 140, 142–6,

148oestrogens 202olanzapine

anxiety 135, 138, 147bipolar disorders 19combination therapies 84dementia 21dosages 6, 8, 11drug interactions 36–8mood stabilizers 84, 87, 92,

98, 101, 105, 119schizophrenia 16–19side effects 27–32, 34–5, 119

opiates 218–19, 233–7, 242–7,252

overdose syndromes 215, 217,218

oxcarbazepine 85, 118oxiracetam 202

panic disorder (PD) 41, 50,130, 139–48

Parkinsonism 25–6Parkinson’s disease 23paroxetine

antidepressants 44–6, 48, 51,57, 66, 70, 78

anxiety 133, 137, 141–2, 150PCP see phencyclidinePD see panic disorderpemoline 180pentazocine 218perphenazine 5phencyclidine (PCP) 22–3,

221–2phenelzine 45phenobarbital 78, 231–3phenylbutazone 197

262 INDEX

physostigmine 200pilocarpine 70pimozide 22piracetam 202PMDD see premenstrual

dysphoriapolacrilex 248posttraumatic stress disorder

(PTSD) 41, 50–1, 131,141–8

pregabalin 135, 139pregnancy

antidepressants 57antipsychotics 38–9anxiolytics 152mood stabilizers 117, 124–6

premenstrual dysphoria(PMDD) 51–2

propoxyphene 218propranolol 113, 146protriptyline 45, 53psilocybin 220–1psychiatric disorders 248–51psychoeducation 131psychosis 19–20, 21, 194

see also antipsychoticspsychostimulants 175–98

alternative preparations189–90

antidepressant augmentation63–4

antipsychotics 22–3attention deficit/hyperactivity

disorder 175, 179–83,186–90, 194–7

contraindications 180costs 191–2dosages 183–5, 191–2drug interactions 196–7drug selection 180–3indications 180–1, 193initiation of treatment 183–5maintenance treatment 185,

187

narcolepsy 190pharmacology 175–80resistance to treatment

187–8side effects 193–6substance abuse 219–20treatment evaluation 185–6

psychotherapy 65–6, 105PTSD see posttraumatic stress

disorderpyrazolopyrimidine indiplon

159

quazepam 163quetiapine

antipsychotics 4–6, 11, 16,19–20, 24–5, 29–32,36–7

anxiety 135, 147mood stabilizers 84, 87, 92,

98–103, 111, 127

racial factors 56–7ramelteon 161, 163–4, 169rapid cycling 83, 110rapid neuroleptization 10reboxetine 43relapse 215, 237–48restless leg syndrome 53risperidone

antipsychotics 4–6, 8, 11,16–18, 20–5, 28–31, 37

anxiety 135, 138, 147–8mood stabilizers 84, 87, 92,

98, 101ritalin 179rivastigmine 200–1, 204, 207,

209–10, 212

schizoaffective disorder 15schizophrenia 1, 4, 7–19,

249–50sedation 35, 52–3

INDEX 263

sedative–hypnotic agents155–73

diagnoses 155–6drug interactions 170–2drug selection 161–6non-prescription 157–8pharmacology 159–61side effects 159, 161, 164–9special precautions 170–2substance abuse 217–18, 229,

230–3treatment implementation

166–7treatment options 156–7

seizures 35–6, 129selective serotonin reuptake

inhibitors (SSRIs)anxiety 130, 132–3, 139–42,

148–52drug interactions 36, 78–9drug selection 56–60, 62–3,

66, 68–75indications 50–5pharmacology 42–4, 46, 48–9side effects 149–51substance abuse 249

selegiline 202–3serotonin–norepinephrine

reuptake inhibitors (SNRIs)anxiety 130, 132, 142–3, 149drug interactions 76drug selection 61, 66indications 55pharmacology 43, 47, 53

sertindole 16, 24sertraline

antidepressants 44, 46, 51–2,54, 66, 78

anxiety 133, 137, 141–2, 148,150

sexual effects 30–1, 73–4SJS see Stevens-Johnson

syndromesleep apnea 53, 152, 171

sleep hygiene techniques 157SNRIs see

serotonin–norepinephrinereuptake inhibitors

social phobiasantidepressants 50–1anxiolytics 131, 139–40, 143

SSRIs see selective serotoninreuptake inhibitors

steroid-induced psychosis 23Stevens-Johnson syndrome

(SJS) 102–3stimulants see psychostimulantssubstance abuse 215–53

alcohol 216–17, 223–30,238–41, 251

amphetamines 219–20antipsychotics 22–3anxiolytics 130cocaine 219–20, 241–2, 252comorbidities 215, 242drug interactions 251–2hallucinogens 220–1, 252intoxication syndromes

215–22mood stabilizers 83multiple dependencies 237nicotine 247–8, 252opiates 218–19, 233–7,

242–7, 252overdose syndromes 215,

217, 218phencyclidine 221–2psychiatric disorders 248–51relapse 215, 237–48sedative–hypnotic agents

167sedative–hypnotics 217–18,

229, 230–3special considerations 247withdrawal syndromes 215,

218–19, 222–37, 247–8suicidality 18–19, 74–5

264 INDEX

tachycardia 33tacrine 200–1, 205, 209–11tardive dyskinesia (TD) 1, 2,

20, 24, 26–9TCAs see tricyclic

antidepressantsTD see tardive dyskinesiategretol XR 111temazepam 163thiamine 230thioridazine 32thyroid hormones 64–5tiagabine 135, 147topiramate 87, 135, 147–8torsade de pointes 33Tourette’s disorder 22, 52tranylcy-promine 45trazodone

antidepressants 43, 47–9,52–4, 58, 72, 74, 79

anxiety 145sedative–hypnotic agents

164–5triazolam 163, 167–8tricyclic antidepressants

(TCAs)acute depression 99–100,

102anxiety 129, 131–2, 137, 140,

143, 149attention deficit/hyperactivity

disorder 181–2, 194drug interactions 76–8drug selection 56–9, 62, 66,

69–75indications 50–4pharmacology 41–5sedative–hypnotic agents

165side effects 149substance abuse 249

trileptal 111trimipramine 45

valproateanxiety 135, 147costs 110–11drug interactions 120drug selection 91–4initiation of treatment 94–7maintenance treatment 105,

108–9pharmacology 84–7pregnancy 124–5side effects 114

venlafaxineantidepressants 43, 47,

49–50, 55, 72, 79anxiety 130, 132–4, 137,

142–3, 150vigabatrin 139violent patients 15vitamin E 203

weight gain 73, 116, 119weight loss 194Wernicke’s encephalopathy

230withdrawal syndromes 215,

218–19, 222–37, 247–8

zaleplon 160, 163–4, 169, 171–2ziprasidone

antipsychotics 4–6, 8–11, 16,24–5, 28, 30–3, 36–7

anxiety 135, 171–2mood stabilizers 84, 87,

97–8, 111zolpidem 160, 163–4, 168–9,

232zotepine 11, 35zyprexa 111

Index prepared by Neil Manley

Date post:	20-Jun-2020
Category:	Documents
Upload:	others
View:	0 times
Download:	0 times

the-eye.eu of Psychiatric Dru… · CONTENTS Preface xv Acknowledgements xvii Disclaimer xix 1...

Documents