The Difficult-to-Treat Psychiatric Patient

The Difficult-to-Treat Psychiatric Patient

This page intentionally left blank

The Difficult-to-Treat Psychiatric Patient

Edited by

Mantosh J. Dewan, M.D. Ronald W. Pies, M.D.

Washington, DC London, England

Note: The authors have worked to ensure that all information in this book concerning drug dosages, schedules, and routes of administration is accurate as of the time of publication and consistent with standards set by the U.S. Food and Drug Administration and the general medical community. As medical research and practice advance, however, therapeutic standards may change. For this reason and because human and mechanical errors sometimes occur, we recommend that readers follow the advice of a physician who is directly involved in their care or the care of a member of their family. A product’s current package insert should be consulted for full prescribing and safety information. Books published by American Psychiatric Publishing, Inc., represent the views and opinions of the individual authors and do not necessarily represent the policies and opinions of APPI or the American Psychiatric Association. Copyright © 2001 American Psychiatric Publishing, Inc. 07 06

05

04 03

02

6 5

4

3

2 1

ALL RIGHTS RESERVED Manufactured in the United States of America on acid-free paper American Psychiatric Publishing, Inc. 1400 K Street, NW Washington, DC 20005 www.appi.org Library of Congress Cataloging-in-Publication Data The difficult-to-treeat psychiatric patient / edited by Mantosh J. Dewan, Ronald W. Pies. p. ; cm. Includes bibliographical references and index. ISBN 0-88048-949-9 (alk. paper) 1. Mental illness—Treatment. 2. Diagnostic errors. 3. Patient compliance. 4. Psychology, Pathological. I. Dewan, Mantosh J. II. Pies, Ronald W., 1952– [DNLM: 1. Mental Disorders—therapy. 2. Diagnostic Errors—prevention & control. 3. Medication Errors—prevention & control. 4. Patient Compliance. 5. Treatment Failure. WM 400 D5697 2001] RC454.4.D54 2001 616.89’1—dc21 00-069519 British Library Cataloguing in Publication Data A CIP record is available from the British Library.

To my parents, Jaimani and Sheel Dewan, with love and gratitude for giving me so much of what is important. —M.J.D To the memory of my father, Jacob Pies, and to my mother, Frances Pies Oliver. —R.W.P.

This page intentionally left blank

Contents

1

Contributors

xi

Introduction

1

Ronald W. Pies, M.D. Mantosh J. Dewan, M.D.

2

The Difficult-to-Treat Patient With Bipolar Disorder

7

Frederick K. Goodwin, M.D. S. Nassir Ghaemi, M.D.

3

The Difficult-to-Treat Patient With Schizophrenia

41

Ronald W. Pies, M.D. Mantosh J. Dewan, M.D.

4

The Difficult-to-Treat Patient With Depression Mantosh J. Dewan, M.D. Ronald W. Pies, M.D.

81

5

The Difficult-to-Treat Patient With Anxiety Disorder

115

Suzanne M. Sutherland, M.D. K. Ranga Krishnan, M.B., Ch.B.

6

The Difficult-to-Treat Patient With Posttraumatic Stress Disorder

149

Elizabeth A. Hembree, Ph.D. Randall D. Marshall, M.D. Lee A. Fitzgibbons, Ph.D. Edna B. Foa, Ph.D.

7

The Difficult-to-Treat Patient With Borderline Personality Disorder

179

Mary C. Zanarini, Ed.D. Kenneth R. Silk, M.D.

8

The Difficult-to-Treat Patient With Dissociative Disorder

209

Richard P. Kluft, M.D.

9

The Difficult-to-Treat Patient With Eating Disorder Wendy A. Harris, M.D. Claire V. Wiseman, Ph.D. Susan Wagner, Ph.D. Katherine A. Halmi, M.D.

243

10

The Difficult-to-Treat Patient With Substance Abuse

273

Mark Albanese, M.D. Edward Khantzian, M.D.

11

The Difficult-to-Treat Patient With Dementia or Traumatic Brain Injury

299

Jeffrey L. Cummings, M.D. Ronald E. Saul, M.D. Tomoko Nakawatase, M.D.

12

The Difficult-to-Treat Psychiatric Patient With Comorbid Medical Illness

325

David S. Harnett, M.D.

13

Special Somatic Treatments

359

D.P. Devanand, M.D. Sarah H. Lisanby, M.D. Harold A. Sackeim, Ph.D.

14

Summary: Clinical Wisdom in Psychiatry

393

Mantosh J. Dewan, M.D. Ronald W. Pies, M.D.

Index

403

This page intentionally left blank

Contributors

Mark Albanese, M.D. Associate Chief, Department of Psychiatry, Tewksbury Hospital, Tewksbury, Massachusetts; Assistant Clinical Professor, Department of Psychiatry, Harvard Medical School, Boston, Massachusetts Jeffrey L. Cummings, M.D. Professor, Department of Neurology and Psychiatry and Biobehavioral Sciences, UCLA, Los Angeles, California D.P. Devanand, M.D. Professor of Clinical Psychiatry, Department of Psychiatry, New York State Psychiatric Institute, New York, New York Mantosh J. Dewan, M.D. Professor and Chair, Department of Psychiatry, SUNY Upstate Medical University, Syracuse, New York Lee A. Fitzgibbons, Ph.D. Director of Children’s Program, Anxiety and Agoraphobia Treatment Center, Philadelphia, Pennsylvania Edna B. Foa, Ph.D. Professor, Department of Psychiatry, University of Pennsylvania, Philadelphia, Pennsylvania S. Nassir Ghaemi, M.D. Assistant Professor, Department of Psychiatry, Cambridge Hospital, Harvard Medical School, Boston, Massachusetts

❉

xi

❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

Frederick K. Goodwin, M.D. Professor, Department of Psychiatry, George Washington University, Washington, D.C. Katherine A. Halmi, M.D. Professor, Department of Psychiatry, Cornell Medical Center, New York Hospital, White Plains, New York David S. Harnett, M.D. Chief of Psychiatry and Director, Medical-Psychiatric Program, Lawrence Memorial Hospital of Medford/Hallmark Health, Medford, Massachusetts; Associate Clinical Professor, Department of Psychiatry, Tufts University School of Medicine, Boston, Massachusetts Wendy A. Harris, M.D. Instructor, Department of Psychiatry, Cornell Medical Center, New York Hospital, White Plains, New York Elizabeth A. Hembree, Ph.D. Assistant Professor, Department of Psychiatry, University of Pennsylvania, Philadelphia, Pennsylvania Edward Khantzian, M.D. Associate Chief, Department of Psychiatry, Tewksbury Hospital, Tewksbury, Massachusetts; Clinical Professor, Department of Psychiatry, Harvard Medical School, Boston, Massachusetts Richard P. Kluft, M.D. Clinical Professor, Department of Psychiatry, Temple University School of Medicine, Philadelphia, Pennsylvania K. Ranga Krishnan, M.B., Ch.B. Professor and Chairman, Department of Psychiatry and Behavioral Sciences, Duke University, Durham, North Carolina Sarah H. Lisanby, M.D. Assistant Professor of Clinical Psychiatry, Department of Biological Psychiatry, New York State Psychiatric Institute, New York, New York

❉

xii ❉

❉ Contributors

❉

Randall D. Marshall, M.D. Associate Professor, Department of Psychiatry, Columbia University College of Physicians and Surgeons, New York Tomoko Nakawatase, M.D. Staff Neurologist, Department of Neurology, UCLA, Los Angeles, California Ronald W. Pies, M.D. Clinical Professor, Department of Psychiatry, Tufts University School of Medicine; Lecturer on Psychiatry, Harvard University School of Medicine, Boston, Massachusetts Harold A. Sackeim, Ph.D. Professor of Clinical Psychology in Psychiatry, Department of Biological Psychiatry, New York State Psychiatric Institute, New York, New York Ronald E. Saul, M.D. Consultant Neurologist, Department of Neurology, UCLA, Los Angeles, California Kenneth R. Silk, M.D. Professor of Psychiatry, Associate Chair, Clinical and Administrative Affairs, Department of Psychiatry, University of Michigan Medical Center, Ann Arbor, Michigan Suzanne M. Sutherland, M.D. Assistant Consulting Professor, Department of Psychiatry and Behavioral Sciences, Duke University, Durham, North Carolina Susan Wagner, Ph.D. Clinical Assistant Professor, Department of Psychiatry, Cornell Medical Center, New York Hospital, White Plains, New York Claire V. Wiseman, Ph.D. Assistant Professor, Department of Psychiatry, Cornell Medical Center, New York Hospital, White Plains, New York

❉

xiii ❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

Mary C. Zanarini, Ed.D. Associate Professor of Psychology, Harvard Medical School; Director, Laboratory for the Study of Adult Development, McLean Hospital, Belmont, Massachusetts

❉

xiv

❉

1 Introduction Ronald W. Pies, M.D. Mantosh J. Dewan, M.D.

Why do some psychiatric patients fail to get better, even when in the care of competent clinicians? This is the central question addressed in this book, from a variety of perspectives. Whatever the answers, one thing is clear: treatmentrefractory patients are all too common in everyday clinical practice. In heterogeneous groups of depressed patients, for example, as many as 30%–40% fail to respond to an adequate course of treatment (Montgomery 1991). In the case of bipolar mood disorders, it appears that resistance to treatment has become increasingly common, as Drs. Goodwin and Ghaemi suggest in Chapter 2 of this book. Among schizophrenic patients, about 40% continue to show significant psychotic symptoms after treatment with conventional antipsychotics (American Psychiatric Association 1997). Whereas newer, atypical antipsychotics have led to improved outcome in schizophrenia, these agents are by no means “cure-alls”—even when patients treated with clozapine and other atypical antipsychotics show improvement on various research-oriented rating scales, they do not necessarily leave the hospital and ❉

1 ❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

become productive members of the community. The problem of treatment resistance is repeated across the spectrum of psychiatric disorders, as detailed by the authors of the chapters that follow. And whether we are considering the treatment-resistant patient with depression or anxiety, eating or substance use disorder, dementia or psychosis, we are facing a source of enormous emotional, economic, and social costs. In the United States, the direct and indirect cost of treating major depression alone amounts to more than $40 billion annually (Greenberg et al. 1993). But as with all major psychiatric disorders, the emotional cost to patients and their families is incalculable. Given all this, what are the main factors that lead to treatment resistance across the spectrum of mental illness?

Classification of Response Before abandoning or altering a treatment regimen, the clinician must determine whether the patient in question is a true nonresponder or partial responder. Many a promising treatment, in our experience, has been abandoned prematurely or inappropriately, when this distinction is not borne in mind. For example, a patient who has been taking an antidepressant for 3 weeks may report better sleep and improved energy yet still harbor intense suicidal thoughts. This is probably not a treatment failure; rather, it is typical of the time course for many somatic treatments of depression, including electroconvulsive therapy (ECT) (Pies 1997c; Van Praag 1992). The same principle applies to the use of antipsychotics. As a rule, it takes longer for negative than for positive symptoms of schizophrenia to respond to antipsychotics, sometimes 6 weeks or more, depending on the use of typical or atypical agents (Breier et al. 1987; Meltzer 1992). Again, a partial responder at week 2 or 3 should not be considered “treatment resistant.” And yet, too often in our experience, a promising antipsychotic regimen is abandoned after 2 or 3 weeks, perhaps as a result of pressures to get the patient out of the hospital or to satisfy the demands of anxious clinical staff. The issue of nonresponder versus partial responder is also critical with respect to the use of adjunctive strategies. For true nonresponders to a particular agent, it is usually inefficient and ineffective to use an augmentation strategy. (When this approach does work, it is usually because the added medication is itself acting as an effective primary therapeutic agent). On the other hand, partial responders can often be converted to full responders with the use of augmentation strategies (e.g., adding bupropion or thyroid medication to a selective serotonin reuptake inhibitor [SSRI]). ❉

2 ❉

❉

Introduction ❉

Diagnostic Factors The patient who is incorrectly diagnosed is unlikely to respond well to treatment. This seems too obvious to belabor; and yet, in our experience, many “treatment-resistant” patients have never been thoroughly reassessed, either to rule out an undetected comorbid condition or to revise the principal diagnosis. One of us (RP) frequently sees patients who have been declared “nonresponders” to antidepressant medication only to find, in some cases, that the patient has an undiagnosed bipolar spectrum disorder (see Akiskal 1996). The antidepressant, in such cases, is often worsening the patient’s cycling rate or producing a dysphoric manic state. Failure to detect a psychotic component in a patient with major depression can result in similar treatment errors. The issue of undetected medical illness in psychiatric patients has been exhaustively reviewed over the past 20 years (e.g., Koranyi 1980) and yet remains a source of apparent treatment resistance. For example, the depressed patient with undetected “subclinical” hypothyroidism may not respond to conventional treatment until the thyroid disorder is corrected (Pies 1997a). The presence of comorbid substance abuse, as detailed in Chapter 10 of this book, is also a source of apparent treatment resistance. In actuality, when the substance abuse disorder is successfully managed, the comorbid condition often responds quite well to conventional treatment approaches.

Medication-Related Factors With respect to mood disorders, many, if not most, cases of apparent treatment resistance stem from inadequate medication for inadequate periods of time. Although many practitioners seem to understand the need for an antidepressant trial of at least 3 weeks, a variety of pressures may conspire to shorten the treatment period to a week or less. It is virtually impossible to gauge the efficacy of an antidepressant in so short a time. This same pattern often occurs with the use of antipsychotics, as suggested earlier. A more subtle error occurs when the bioavailability of a medication is not considered. Patients differ markedly from one another in their ability to absorb and metabolize psychotropic agents. For example, the same oral dose of a phenothiazine may produce a wide range of plasma neuroleptic levels in a given cohort of patients. Some patients may be genetically slow or fast metabolizers of various psychotropics. They may achieve unusually high or low blood levels of a given agent, with toxicity or incomplete response, respectively, as frequent outcomes. In some cases, the bioavailability of a drug may be mark❉

3 ❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

edly altered by the coprescription of an agent that alters metabolism of the first (e.g., carbamazepine may drastically reduce the plasma level of haloperidol and other antipsychotics) (Arana et al. 1986). Finally, patients may vary on a molecular level with respect to drug sensitivity. Although we have no easy means of determining this at present, it seems likely to be a factor in patients who have been correctly diagnosed, who have been treated with adequate doses for adequate periods of time, and who have achieved normal to high blood levels of the particular agent. Whether these patients have an unusually “thick” blood-brain barrier or whether, on the neuronal level, they are not sensitive to the effects of psychotropic medications is almost never known. In such suspected cases, raising the medication dose or blood level is unlikely to be of help (Osser and Patterson 1996). Finally, medication side effects may be an important cause of poor or incomplete response in some patients. For example, the bipolar patient taking high doses of anticholinergic agents for extrapyramidal symptoms may respond poorly to antimanic agents as a consequence of the mood-elevating properties of anticholinergic drugs. The patient taking high doses of an SSRI may experience either apathy or extrapyramidal effects, leading to poor treatment outcome (HoehnSaric et al. 1991; Pies 1997b). The psychotic patient whose negative features are worsened by conventional neuroleptics is unlikely to do well in the community. The eating-disordered patient who gains weight on an SSRI is unlikely to continue taking it. Unless such side effects are recognized and managed, a patient may remain treatment resistant for months or even years.

Compliance-Related Factors A variety of factors peculiar to the patient may contribute to apparent treatment resistance. The issue of compliance is paramount, though this word is not terribly respectful of the patient’s autonomy. In any case, the patient who “cheeks” medication, or who simply does not take it once out of the hospital, is bound to do poorly. Why do patients undermine their own treatment in these ways? Side effects of psychotropic medications may be an important factor in medication noncompliance, but it is not the only one. Some patients have conscious or unconscious feelings about a medication that they do not share with the prescribing clinician. For example, the bipolar patient who believes that lithium “takes away my creative periods” or the paranoid patient who believes that antipsychotic medication is the psychiatrist’s means of “controlling my sexuality” is unlikely to comply with treatment. (Of course, all such complaints are not mere paranoia—many antipsychotic medications ❉

4 ❉

❉

Introduction ❉

have very real adverse effects on sexual function). Goodman (1992) has nicely summarized the many unconscious feelings and fantasies that affect compliance with medication. Unless the clinician is sensitive to the psychodynamic issues underlying pharmacotherapy, he or she is likely to miss important factors in poor medication compliance (Gutheil 1982). Finally, high medication costs and logistical barriers to obtaining them may be important reasons for poor compliance in disadvantaged patients (Perkins 1999).

Inappropriate Use of Psychosocial Interventions Sometimes psychosocial therapies are seen as innocuous “frills,” added onto somatic treatments because of medicolegal concerns or out of allegiance to some abstract biopsychosocial model. In reality, psychosocial therapies are powerful and effective interventions, often rivaling somatic treatments in terms of outcome measures. Like any effective treatment, psychotherapy may sometimes be associated with adverse effects. Even in carefully selected patients, psychotherapy may evoke powerful transference reactions, “acting out,” and periods of regression. When inappropriately selected patients enter some forms of psychotherapy, adverse outcomes are often seen at higher rates (e.g., patients with schizophrenia do not usually tolerate intense forms of “exploratory” or psychoanalytic psychotherapy). Whereas some seasoned clinicians find that modifications of psychoanalytic techniques can be used successfully with such patients (e.g., Karon and VandenBos 1972), inexperienced clinicians may make matters worse when treating fragile or regressed patients. Even the wrong type of day treatment setting may lead to regression in some schizophrenic patients, in our experience.

Conclusion The term treatment resistant should be used only when the clinician has systematically explored and ruled out the usual causes of poor treatment response in psychiatric patients. Whereas putative resistance at the neuronal level cannot be evaluated, inadequate dosing, low psychotropic blood levels, and poor compliance may be determined fairly easily in most cases. More subtle factors affecting response, such as the patient’s unspoken or unconscious fantasies about medication, require attention to the therapeutic alliance and the patient’s psychodynamic issues. Once genuine treatment resistance has been established, a variety of interventions is available to the clinician. That subject, indeed, is the focus of much that follows in this book. ❉

5 ❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

References Akiskal HA: The prevalent clinical spectrum of bipolar disorders: beyond DSM-IV. J Clin Psychopharmacol 116(suppl 1):4–14, 1996 American Psychiatric Association: Practice Guidelines for the Treatment of Patients With Schizophrenia. Am J Psychiatry 154(suppl):1–63, 1997 Arana GW, Goff DC, Friedman H, et al: Does carbamazepine-induced reduction of plasma haloperidol levels worsen psychotic symptoms? Am J Psychiatry 143: 650–651, 1986 Breier A, Wolkowitz OM, Doran AR, et al: Neuroleptic responsivity of negative and positive symptoms in schizophrenia. Am J Psychiatry 144:1549–1555, 1987 Goodman A: Medication noncompliance and the psychodynamics of pharmacotherapy. Integrative Psychiatry 8:181–190, 1992 Greenberg PE, Stiglin LE, Finkelstein SN, et al: The economic burden of depression in 1990. J Clin Psychiatry 54:405–418, 1993 Gutheil TG: The psychology of psychopharmacology. Bull Menninger Clin 46:321– 330, 1982 Hoehn-Saric R, Harris G, Pearlson G, et al: A fluoxetine-induced frontal lobe syndrome in an obsessive compulsive patient. J Clin Psychiatry 52:131–133, 1991 Karon BP, VandenBos GR: The consequences of psychotherapy for schizophrenic patients. Psychotherapy: Theory, Research, and Practice 9:11–119, 1972 Koranyi EK: Somatic illness in psychiatric patients. Psychosomatics 21:887–891, 1980 Meltzer HY: Treatment of the neuroleptic-nonresponsive schizophrenic patient. Schizophr Bull 18:515–542, 1992 Montgomery SA: Selectivity of antidepressants and resistant depression, in Advances in Neuropsychiatry and Psychopharmacology, Vol 2: Refractory Depression. Edited by Amsterdam JD. New York, Raven, 1991, pp 93–104 Osser D, Patterson R: Pharmacotherapy of schizophrenia, in Handbook for the Treatment of the Seriously Mentally Ill. Edited by Soreff S. Seattle, WA, Hogrefe & Huber, 1996, pp 91–155 Perkins DO: Factors affecting likelihood of adhering to treatment [slide]. J Clin Psychiatry (JCP Visuals) January 60:25–30, 1999 Pies R: The diagnosis and treatment of sublclinical hypothyroid states in depressed patients. General Hospital Psychiatry 19:344–354, 1997a Pies R: Must we now consider SRIs neuroleptics? J Clin Psychopharmacol 17:443– 45, 1997b Pies R: Time and the art of psychopharmacology. Harv Rev Psychiatry 5:36–39, 1997c Van Praag H: “Make-Believes” in Psychiatry, or the Perils of Progress. New York, Brunner-Mazel, 1992

❉

6 ❉

2 The Difficult-to-Treat Patient With Bipolar Disorder Frederick K. Goodwin, M.D. S. Nassir Ghaemi, M.D.

B

ipolar illness is one of the most challenging yet rewarding psychiatric conditions to treat. Patients who suffer from this condition are often quite high functioning at times when the illness is quiescent, and thus they have even more to lose when treatment fails to work. Bipolar disorder involves a spectrum from mild forms of illness that are sometimes difficult to distinguish from either unipolar depression or normal mood swings to more classic forms of type I bipolar disorder (or bipolar I) to mixed states clouded with substance abuse. As a result, the bipolar illness faced by clinicians today is often more complex to diagnose and difficult to treat. A major factor in treatment resistance in bipolar disorder involves its clinical complexity. Whereas patients with unipolar depression (except in the highly recurrent form) are either ill ❉

7 ❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

or well, patients with bipolar illness can be ill in four ways (mania, depression, mixed episodes, and continuous or rapid-cycling) but well in only one way (euthymia). Thus, while the patient with unipolar depression either responds to antidepressants (partially or completely) or does not, in bipolar illness, antidepressants may 1) fail to budge a depressive episode, 2) lead to partial response with residual depression, 3) lead to temporary euthymia followed by later relapse, 4) cause a transient hypomania followed by euthymia, 5) produce a transient hypomania followed by a full-blown acute mania, 6) lead directly to a full acute manic episode, or 7) lead to persistent euthymia. Euthymia is the goal of treatment, and although there are many ways to miss the mark of stable euthymia, there are few effective combinations of treatments that lead to stable euthymia. One tactic is to use stronger and stronger weapons when faced with resistance. This tactic is similar to treatment methods in unipolar depression, with treatment-resistant patients receiving one, two, or more antidepressant agents at increasing doses. An alternative tactic emphasizes precision. This strategy resembles the approach to treatment-resistant bipolar patients: one can conceptualize the matter by believing that any particular individual’s central nervous system responds to a single or limited number of combinations of mood stabilizers and mood-stabilizing adjuncts (other mood stabilizers, neuroleptic agents, antidepressant agents, and/or benzodiazepines). Thus, with the treatmentresistant bipolar patient, there are many ways in which the treatment can fail but few in which it can succeed. Only the right combination will hit the target. Given this complex background, in our opinion, the gravest error in treating patients with bipolar disorder is to focus on the immediate clinical situation at the expense of the longitudinal course of illness. Thus, when the patient is depressed, the clinician focuses on treating the depression, frequently with antidepressant agents; and when the patient is manic, the focus shifts to treating mania, frequently with neuroleptic agents. Instead, the focus should be on the fact that bipolar disorder represents an ongoing process, and it requires a long-term perspective for optimal treatment. Simply put, the only treatments that have been shown to work in both the short and long term for bipolar disorder are mood-stabilizing agents (lithium and some anticonvulsants). Thus, in treating a patient with bipolar disorder, the first priority is always mood stabilization. Although this may seem unnecessary to point out, a review of the evidence makes it clear that all too often mood-stabilizing agents are not used aggressively enough in the treatment ❉

8 ❉

❉

The Difficult-to-Treat Patient With Bipolar Disorder

❉

of bipolar disorder, whereas antidepressants and typical neuroleptics are used all too aggressively. This unbalanced approach does not work with the difficult-to-treat bipolar patient and indeed may be a major contributor to treatment resistance.

Prevalence The prevalence of bipolar disorder depends on the breadth of diagnostic criteria used. If limited to classic type I syndromes, prevalence rates around 1% have been reported in large epidemiologic studies (Goodwin and Jamison 1990). These estimates are based substantially on the Epidemiological Catchment Area (ECA) study using the lay-administered Diagnostic Interview Schedule (DIS) based on DSM-III (Regier and Kaelber 1995). However, clinician-administered research interviews correlated poorly with DIS-based diagnoses in one of the ECA sites (Anthony et al. 1985). More recently, ECAlike diagnostic methods were used in the National Comorbidity Survey; even with similar methods, the prevalence of mania was twice as high as in the ECA study (1.6% compared with 0.8%) (Kessler et al. 1994). The best clinical interview may still be insufficient to definitively rule out bipolar disorder. This is supported by an observation in the Iowa 500 study (Tsuang et al. 1980), where it was noted in the course of a family study that the diagnosis of mania was made more often in relatives of bipolar probands if hospital charts were used in addition to personal clinical interviews. In fact, simply relying on personal interview seemed to underestimate the incidence of mania in relatives by a factor of almost three (morbidity risk 1.9±1.07 excluding hospital charts vs. 5.3±1.73 including hospital charts). This finding suggests that many patients may deny such symptoms in the course of their clinical interview, and if investigators do not use outside sources of information, then the diagnosis of bipolar disorder would be underestimated compared with the true prevalence. One issue of relevance is the phenomenon of lack of insight, especially in manic phases of mood disorder (Ghaemi 1997), which obviously can contribute to misdiagnosis and may require sophisticated interview techniques that are difficult to standardize for research purposes. Estimating the prevalence of the full bipolar spectrum presents additional difficulties. For example, estimating the prevalence of bipolar II disorder (spontaneous hypomania and major depression) depends on the ascertainment of hypomania, which is more easy to miss than full-blown mania (Akiskal 1996). Similar problems exist with ascertainment of bipolar ❉

9 ❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

disorder, not otherwise specified (NOS) (antidepressant-induced mania or hypomania only and spontaneous major depression). Although it is not included in DSM-IV, we sometimes clinically apply the NOS diagnosis to patients with recurrent unipolar major depressive disorder with a family history of bipolar disorder in a first-degree relative and/or unipolar major depressive disorder with underlying hyperthymic personality. Given the clinical and genetic studies that suggest that these other forms of bipolar illness may be more common than type I illness, it is not unreasonable to estimate that the broad spectrum of bipolar disorder may have a prevalence in the 2%–5% range. When defined thus broadly, bipolar illness may be about as common as unipolar major depressive disorder.

Treatment Resistance in Bipolar Disorder: A Definition Routine treatment for bipolar disorder over the last three decades has consisted of mood stabilizing medications (initially lithium, with anticonvulsants broadening the armamentarium more recently). This has been augmented by antidepressant medications for breakthrough depressive episodes (initially tricyclic agents and more recently bupropion and serotonin reuptake inhibitors and the so-called broad-spectrum drugs such as venlafaxine and mirtazapine) and neuroleptic medications for breakthrough manias (initially typical and later atypical agents). The early trials with lithium reported overall response rates in the 60%–80% range for both antimanic and 1-year prophylaxis (reviewed in Goodwin and Jamison 1990). However, more recent clinical trials of lithium monotherapy for acute mania report lower response rates of about 50% (Bowden et al. 1994), and naturalistic follow-up studies in tertiary-care academic centers found response rates with lithium (sometimes supplemented with antidepressants and/or neuroleptic agents) of 4%–30% (Goldberg et al. 1995; Sachs et al. 1994b). On the other hand, some naturalistic studies using private practice samples continue to report response rates in the range of the earlier studies (Kukopulos et al. 1980; Page et al. 1987). Has the illness changed to the degree that lithium no longer works? First, the discrepancy between the academic and private practice results may reflect the fact that academic referral centers simply do not see very many lithium-responsive patients because practicing physicians have no reason to refer them. On the other hand, there is also evidence that the illness has changed in the last 30 years. Not only has the prevalence of the disorder among the ❉

10

❉

❉

The Difficult-to-Treat Patient With Bipolar Disorder

❉

young increased sharply, but also the median age at onset has decreased from 32 to 19 years (Goodwin and Jamison 1990). In the United States, these two trends parallel the cocaine epidemic. Other environmental factors have also been cited to explain this change (Wehr et al. 1988) as have dietary changes (Hibbeln and Salem 1995). Thus, there probably are more treatment-resistant patients overall, and they are more likely to be concentrated in referral centers. The currently understood factors associated with treatment resistance in bipolar disorder are described below, roughly in order of their importance, in our experience.

Medication Noncompliance Poor medication compliance (which covaries with other factors) is probably the most important factor in treatment nonresponse. The most common reason for noncompliance is side effects, with weight gain being the most frequent complaint. Next is memory impairment, particularly the noncued recall problem—the “tip of the tongue” phenomenon. Cognitive slowing and decreased energy are related side effects also noted both with lithium and valproate. Tremor is often an issue for lithium, whereas hair loss can sometimes lead to noncompliance with valproate. With both agents, patients not infrequently complain of a loss of creativity, especially among those who are used to functioning at a mildly hypomanic level much of the time. Working with the patient to relearn what is “normal” represents a psychotherapeutic challenge in management. It should be emphasized that many of these side effects, in fact the most common ones, can be treated. Thus, lithium tremor generally responds to low doses of propanolol or other β-blocking agents. Cognitive/memory problems respond to decreasing the dose, shifting it all to bedtime, supplementing with B vitamins (especially 400 mcg of folic acid), and correcting low normal thyroid function (Hatterer et al. 1989). Both lithium and valproate can be associated with weight gain. The effects of lithium on weight gain are complex and may be mediated by lithium’s probable mild anti-insulin effects (Peselow et al. 1980). The resultant state of mild reactive hypoglycemia following ingestion of simple carbohydrates (especially sugar) can produce subsequent carbohydrate craving. Thus, carbohydrate restriction, especially for sugar-containing foods, is important. Frequent small feedings may reduce carbohydrate cravings, and patients need to be warned to avoid high-calorie drinks with large concentrations of sugars when they try to ❉

11

❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

quench lithium-induced thirst. For patients who experience discrete periods of carbohydrate craving, the amino acid L-glutamine in doses of 500–1500 mg/day may prove helpful (Goodwin and Jamison 1990). Thyroid supplementation may also be helpful for weight control in bipolar patients. Psychological issues in poor compliance include cultural beliefs regarding medications “controlling” one’s moods, a belief that hypomania is necessary for success or attractiveness, and power struggles with the clinician. Psychoeducation and ongoing discussions between patients and physicians regarding patients’ feelings about taking medications should be part of any illness management strategy. Furthermore, it is important that the physician treat the patient as a collaborator, one from whom the clinician can learn. Compliance also involves practical matters such as dosing schedules. Oncea-day dosing at bedtime (or at most twice a day) is the easiest schedule with which to comply. There is no pharmacokinetic reason why these agents need to be dosed more frequently than once daily. The only reason to divide daily doses is for excessive sedation or nausea resulting from the acute bolus effect of a large single daily dose. With once-a-day lithium dosing, there is no evidence of increased recurrence of illness, and there is evidence of possible decrease in renal tubular problems (Hetmar et al. 1986). Thus, single daily dosing at bedtime is both safe and effective and indeed should be more effective to the extent that it enhances compliance. This simplification is particularly recommended for long-term use of mood stabilizers given the difficulty in remembering to take a medication in the absence of symptoms.

Comorbid Substance Abuse Comorbid substance abuse is an important risk factor for treatment resistance and is commonly associated with poor or uneven compliance. Data from the ECA study indicate that about 60% of patients with bipolar disorder will meet criteria for substance abuse at some point in their lives, and this is associated with poor outcome (Regier et al. 1990; Tohen et al. 1990). Substance abuse is relevant to treatment resistance in any number of ways. First, it may precipitate episodes of illness. For instance, alcohol abuse may contribute to depression, whereas cocaine, frequently used to “treat” a depressive episode, may very well precipitate mania through a “kindlinglike” process. Second, the impulsivity frequently associated with mania may predispose patients to alcohol or cocaine abuse, further impairing patients’ judgment, leading to medication noncompliance and a vicious regress of worsening symptoms. Substance abuse may mask symptoms of depression ❉

12

❉

❉

The Difficult-to-Treat Patient With Bipolar Disorder

❉

or mania, leading to misdiagnosis, inappropriate treatments, and a worsened course of illness. In general, when the differential diagnosis of bipolar disorder is raised in a patient with substance abuse, it is our rule of thumb to aggressively treat the bipolar disorder pharmacologically while vigorously pursuing substance abuse treatment. When the possibility of substance abuse arises in a patient already diagnosed with bipolar disorder, our practice is to institute a treatment plan for the substance abuse in order to prevent its long-term deleterious effects on the bipolar disorder.

Mood-Destabilizing Effects of Antidepressants Patients with bipolar disorder often are more aware of depression and its concomitant symptoms than they are aware of mood elevation. Hence, even in the midst of an episode of mood elevation, they may present requesting antidepressant treatment often while refusing mood stabilizers. Considerable evidence suggests a relationship between chronic antidepressant treatment, especially without concurrent mood-stabilizing treatment, and development of treatment resistance. Wehr and Goodwin (1987) have reviewed this matter and marshaled much of the relevant data. Briefly, in two double-blind outcome studies, the rate of manic episodes in patients with bipolar disorder treated with antidepressants and lithium is roughly twice the rate of those treated with lithium alone (Prien et al. 1974, 1984). In two studies (Altshuler et al. 1995; Kukopulos et al. 1980) of patients with rapid cycling, antidepressants were thought to be the likely causes of rapid cycling in 26%–35% of cases (n =85). Using mood charting, Wehr and Goodwin (1987) also documented increased frequency of affective cycles in patients treated with desipramine (and lithium) instead of lithium alone. Finally, when Quitkin and colleagues (1981) compared lithium alone with lithium plus imipramine in a double-blind, randomized, 2-year prospective study of bipolar patients, they found a 2.5-fold increase in breakthrough manic episodes in the combined treatment group with no difference in the frequency of breakthrough depressions. The absence of systematic or objective measures for cycling may account for the general underrecognition of this phenomena. Simple reliance on the patient’s subjective self-report often is insufficient. The limitations of self-report can be decreased by systematically collecting information from other sources, such as mood charting and family report. The above data largely involve tricyclic antidepressants. Two of three ❉

13

❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

double-blind studies suggest that some of the newer antidepressants may be less likely to cause mania and possibly less problematic in the long-term treatment of bipolar disorder as well. The first study (Sachs et al. 1994a), a double-blind comparison of bupropion and desipramine in 19 patients with bipolar disorder taking lithium, found a statistically significantly lower mania switch rate with bupropion (1/9, 11%) than desipramine (5/10, 50%) over 1 year. The second study (Cohn et al. 1989) was a double-blind comparison of fluoxetine versus imipramine versus placebo in bipolar patients, of whom about one-third were also taking lithium, whereas the others were not prescribed any mood stabilizer. The investigators reported better treatment response with fluoxetine than imipramine, but the mania switch rate was the same with both agents over 8 weeks. The third study was the largest and compared paroxetine versus imipramine versus placebo in 117 patients with bipolar I disorder unresponsive to lithium. The antidepressants were equally effective, but there was no mania switch with paroxetine, compared with 8% with imipramine over 10 weeks of treatment (Young et al. 1997). Although these studies suggest that at least bupropion and paroxetine may be less problematic in causing mania than tricyclic antidepressants, they are generally short-term studies (8–12 weeks), and the longer term risks of rapid cycling have not yet been specifically studied with the newer antidepressant agents. Some studies suggest that antidepressants may be somewhat safer in bipolar II disorder. Amsterdam and colleagues (Amsterdam 1998; Amsterdam et al. 1998) have reported efficacy with low mania switch rates with the use of venlafaxine and fluoxetine in bipolar II disorder. These findings may have underestimated mania switch rates, however, because they were not designed to assess manic symptoms. In a meta-analysis of selective serotonin reuptake inhibitor (SSRI) clinical trials of unipolar depression that included bipolar II patients, Peet (1994) found a lower mania switch rate with SSRIs (about 4%) than with tricyclic antidepressants (about 12%). Because these studies were not designed to assess switch rates and included only type II patients, their relevance to bipolar I disorder is uncertain. On the other side, Altshuler and colleagues (1995) found that the risk for antidepressant-induced rapid cycling is somewhat higher in patients with bipolar II disorder than bipolar I disorder. The risk here is not so much causing mania, but causing more depression in the long run by inducing more mood episodes with chronic antidepressant treatment. In the absence of prospective, controlled trials in bipolar II disorder, it would be unwise to assume that antidepressants do not possess the risks in that condition that they clearly ❉

14

❉

❉

The Difficult-to-Treat Patient With Bipolar Disorder

❉

possess in bipolar I disorder. A final note: it is our experience that many patients take or accept the label of “type II,” believing that unless they have severe dysfunction or have been hospitalized, their mood elevation symptoms do not meet criteria for mania. DSM-IV argues that “significant” social or occupational dysfunction results in a diagnosis of mania, not hypomania. The meaning of significant is not readily agreed upon by clinicians. Thus, hypomania is among the least reliable diagnoses in DSM-IV. Such uncertainty about who has hypomania versus mania magnifies the risks of using antidepressant monotherapy in a patient with presumed bipolar II disorder. We often find possible mood stabilizers with limited side effects, such as gabapentin, to be good initial low-risk, low-yield alternatives in these cases. Nonetheless, in some patients who are compliant and willing to follow instructions, if mood stabilizers are not acceptable to them, one can consider using an antidepressant alone after documenting the recommendation of a mood stabilizer. The patient should carefully chart his or her moods and agree beforehand to switching to or adding a mood stabilizer if the patient, his or her family, or the doctor note any symptoms of hypomania or rapid cycling. Although some clinicians describe generally good results when prescribing antidepressants to patients with bipolar disorder, it is difficult to completely ignore the apparent risks. Until better data are available, it seems prudent to limit exposure of bipolar patients to antidepressants to the periods of their acute depressive episodes and to provide concurrent moodstabilizing treatment.

Misdiagnosis Over the last few decades, there has been a well-documented decline in the age of onset of bipolar disorder, as well as a doubling of the prevalence of bipolar disorder among those under age 30, which in the United States has occurred at the same time as the prevalence of substance (especially cocaine) abuse was increasing. The comorbidity of bipolar disorder and substance abuse, as noted earlier, may thus be contributing again to misdiagnosis. Cocaine may also increase the rate or risk of switching into manic episodes. Experts have also noted that patients with bipolar disorder often underreport their symptoms, especially manic symptoms; thus their families may serve as a better source for such symptoms (Goodwin and Jamison 1990). An empirical study (Keitner et al. 1996) confirmed this observation, finding that families are twice as sensitive to behavioral symptoms of mania than ❉

15

❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

patients’ self-reports. This finding held only for behavioral effects in mania, not in depression. This study suggests that clinicians should obtain family report of behavioral symptoms in the assessment of mania or hypomania. Because insight is more absent in mania than in depression, it stands to reason that lack of insight is one factor leading to an underdiagnosis of mania and a relative overdiagnosis of depression. Recently, our research group found that about 40% of inpatients with bipolar I disorder had never before received that diagnosis when examined by psychiatrists before entering our research study. Of this group, an average of 8 years had elapsed from first mental health professional contact before the diagnosis of bipolar disorder was made (Ghaemi et al. 1999). All of those patients had previously been diagnosed instead with unipolar major depressive disorder and frequently were treated with antidepressant agents. Given the evidence presented earlier that treatment with antidepressant medications may be problematic, the misdiagnosis of bipolar disorder as unipolar depression is a potentially major source of treatment nonresponse. Even if complicating factors such as lack of insight are ignored, there is evidence that clinicians apply major depressive criteria more carefully and consistently in the evaluation of patients with mood disorders than they apply manic criteria (Sprock 1988). Thus, the careful application of criteria for the diagnosis of mania is an important aspect of the evaluation of every patient with depression. Manic criteria applied to the current mood episode will distinguish mixed episodes from pure depression, and if applied to past episodes, these criteria separate unipolar from bipolar depression. In Figure 2–1, we provide a mnemonic aid based on DSM-IV criteria, originally devised by Dr. William Falk at the Massachusetts General Hospital, that may promote more careful assessment of manic criteria. Returning to Kraepelin’s original broad notion of manic-depressive illness, we think one can conceive of all mood disorders as lying along a continuum. The variations of mood disorders can be conceived along one broad spectrum of manic-depressive illness (Figure 2–2), with bipolar I disorder and unipolar major depressive disorder at the extremes. Bipolar II disorder and cyclothymia display less severe manic symptoms. The area between cyclothymia and unipolar depression is controversial, corresponding to the DSM-IV diagnosis of bipolar disorder, NOS. We suggest that the diagnosis should include individuals who only experience hypomanic or manic episodes with antidepressant medications but not spontaneously, those with recurrent unipolar major depressive episodes and a first-degree relative with bipolar I disorder, and those with hyperthymic ❉

16

❉

❉

The Difficult-to-Treat Patient With Bipolar Disorder

❉

Distractibility—the inability to maintain one’s concentration, as opposed to the decreased concentration of depression, in which one is unable to initiate concentration. In mania, this leads to multiple tasks, none of which are finished, as opposed to depression, in which no task can be started easily. Insomnia—decreased need for sleep, as opposed to the decreased sleep of depressive insomnia. The patients sleeps less but has intact or increased energy the next day. Grandiosity—can be inflated self-esteem as well, need not be delusional. Flight of ideas—the subjective experience of racing thoughts. Activities—increased goal-directed activities (social, sexual, school, work, home activities); these are goal directed and thus not dysfunctional. Increased libido is either not expressed in activity or associated with increased activity with one’s usual sexual partner. Speech—pressured; this is an objective sign observed on the mental status examination. A subjective alternative is increased talkativeness, which is determined by asking the patient or others whether the patient has been more talkative than usual (when euthymic). Thoughtlessness—this refers to increased pleasurable activities with potential for painful consequences. Four stereotypic behaviors that may be asked about are sexual indiscretions, spending sprees, impulsive traveling, and reckless driving. Figure 2–1. The DIGFAST mnemonic for mania. Note. The diagnosis of mania is made when euphoric mood is present for 1 week with three of the DIGFAST symptoms, or irritable mood is present for 1 week with four of the DIGFAST symptoms, and there is significant social or occupational dysfunction. If there is no significant dysfunction, and the symptoms last 4 days at least, the diagnosis of hypomania is made. If the symptoms last less than 4 days, or if they are only present with antidepressant medications, a diagnosis of bipolar disorder, NOS, may be made. Source. Reprinted from Ghaemi SN, Sachs GS, Chiou AM, et al.: “Is Bipolar Disorder Still Underdiagnosed? Are Antidepressants Overutilized?” Journal of Affective Disorders 52:135–144, 1999.

personality at baseline and recurrent unipolar major depressive episodes. Recurrent, psychotic, and atypical unipolar depression may also be closer to the bipolar end of the spectrum, with similarities in underlying pathophysiology and treatment response. At the extreme of the bipolar end of the spectrum, schizoaffective disorder, bipolar type, might be viewed as a more severe psychotic form of bipolar illness. The spectrum concept helps guard against a common misperception: to ❉

17

❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

Figure 2–2. The Manic-depressive spectrum. Note. BP, NOS=bipolar disorder, not otherwise specified; MDD=major depressive disorder; MDE=major depressive episode; SA=schizoaffective disorder. BP, NOS could include mania or hypomania only on antidepressants, recurrent MDD with underlying hyperthymia, or recurrent MDD with a first-degree relative with bipolar disorder. Source. Reprinted from Goodwin FK, Ghaemi SN: “An Introduction to and History of Affective Disorders,” in New Oxford Textbook of Psychiatry, Vol 1. Edited by Gelder M, Lopez-Ibor J, Andreasen N. New York, Oxford University Press, 2000, pp. 677– 682.

think of “depression” as being equivalent to unipolar depression, which is then treated with antidepressants. There are a number of reasons for this phenomenon: the first is that patients often lack insight into their manic symptoms; not knowing that they are ill, they deny their manic symptoms to clinicians. Second, depressive symptoms tend to last longer than manic symptoms, sometimes are more frequent, and often are more psychically painful; thus, patients tend to seek assistance when depressed rather than when manic. Third, the many new antidepressants that have become available over the past 10 years have been extensively marketed to physicians at the same time that depression awareness programs have educated the public about the availability of safe and effective treatments. During this same time, while some new treatments for bipolar disorder have become available, there has been scant professional and public education about it. Lithium, the mainstay of bipolar treatment, is a generic drug with very little money available for its promotion or for educational efforts. As with the differential diagnostic process in any medical disease, the diagnosis of mood disorders should start with those disorders that must be ruled out first to those that remain afterwards (Figure 2–3). We believe this process should begin by ruling out depression that is clearly due to another medical or psychiatric disorder. Such “secondary depressions” usually ❉

18

❉

❉

The Difficult-to-Treat Patient With Bipolar Disorder

❉

Figure 2–3. The differential diagnosis of mood disorders: getting from “depression” to diagnosis. Note. The order in which diagnoses need to be ruled out is the following: 1) secondary depression, 2) bipolar depression, 3) unipolar depression. Thus, unipolar depression is a diagnosis of exclusion. Source. Reprinted from Goodwin FK, Ghaemi SN: “An Introduction to and History of Affective Disorders,” in New Oxford Textbook of Psychiatry, Vol 1. Edited by Gelder M, Lopez-Ibor J, Andreasen N. New York, Oxford University Press, 2000, pp. 677– 682.

involve a single major episode occurring in the absence of prior depressive symptoms or family history and at a later age at onset than is typical for primary depression. The second rule out diagnosis is bipolar disorder; first, bipolar I, then bipolar II, and then bipolar NOS (not otherwise specified) are sequentially ruled out before unipolar depression can be diagnosed. Unfortunately, many clinicians and patients jump from the recognition of a major depressive syndrome directly to a diagnosis of unipolar depression without the critical intermediate process of ruling out bipolar conditions.

Specificity of Diagnostic Subtype: Mixed Episodes and Rapid Cycling Mixed episodes may arise spontaneously (simple) or may result from the cooccurrence of a primary mood disorder and a secondary neuropsychiatric condition. Rapid-cycling bipolar disorder, defined as at least four affective episodes in a given year, has been described as less responsive to treatment with lithium and more likely to respond to anticonvulsants (Calabrese and ❉

19

❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

Delucchi 1990). One report (Levy et al. 1988) suggested that mood swings in some rapid-cycling patients may be the manifestation of a subclinical seizure disorder, thus accounting for anticonvulsant responsiveness. Rapid cycling has also been associated with antidepressant use, as noted earlier.

Comorbid Psychosis In the first episode mania prospective outcome studies from McLean Hospital, one of the other poor prognostic factors identified was the presence of psychosis during mania (Tohen et al. 1990). Concurrent psychosis may lead to misdiagnosis as schizophrenia or schizoaffective disorder, followed by inadequate mood-stabilizer treatment and thus a worsened course of illness. While the differential diagnosis of schizophrenia with bipolar disorder is probably less problematic now than it was 2 or 3 decades ago, the apparent underutilization of mood-stabilizing agents continues to occur. The new generation of atypical neuroleptic agents may be safer and possibly more effective than traditional neuroleptics and thus have an important role in treating psychotic bipolar disorder (Ghaemi and Goodwin 1999).

Comorbid Medical Conditions Another significant correlate of treatment resistance in bipolar disorder is the presence of comorbid medical or psychiatric conditions. Anxiety disorders and attention-deficit/hyperactivity disorder (Biederman et al. 1996) are associated with patients whose bipolar disorder began early in preadolescent childhood; this early-onset cohort tends to be more treatment resistant than patients whose bipolar disorder begins later in adulthood. Thyroid hormone abnormalities (Cowdry et al. 1983) and temporal lobe epileptiform abnormalities (Levy et al. 1988) also have been associated with rapid-cycling bipolar disorder, which other studies suggest can be refractory to lithium, as described earlier.

Clinical Steps for Treatment-Resistant Bipolar Disorder Acute Mania The treatment of acute mania begins with mood-stabilizing agents. Although in most double-blind circumstances lithium and the anticonvulsants tend to show similar overall response rates (Bowden et al. 1994), in moder❉

20

❉

❉

The Difficult-to-Treat Patient With Bipolar Disorder

❉

ate to severe mania the anticonvulsants appear to have a more rapid onset of action, especially when a loading dose is used (McElroy et al. 1996). Frequently, neuroleptic medications are added to the mood stabilizers. In treatment-resistant patients, if they are nonresponsive to lithium, the clinician should search for the reasonably well-established indicators for the use of anticonvulsants, such as rapid cycling, mixed states, comorbid substance abuse, and prior lithium failure. For typical manic symptoms, most of the research evidence to date would support the use of lithium as the initial treatment of choice. In treatment-resistant patients with typical mania, anticonvulsant agents should be added, when possible, to lithium, and lithium should be continued along with the anticonvulsants for combination mood-stabilizer treatment. In choosing the anticonvulsants, valproate and carbamazepine are both viable options. Whereas most recent research has focused on valproate, most of the previous research on carbamazepine supported moderate efficacy in combination with lithium for treatment-resistant patients (Post 1990). There are certain pharmacological issues that differentiate the two medications. Carbamazepine is a strong inducer of the hepatic cytochrome P450 enzyme system. Whereas this does not entail any relevant interactions with lithium, which is excreted unchanged by the kidney, carbamazepine would tend to reduce drug levels of other medications, such as neuroleptics, antidepressants, or warfarin sodium, by as much as one-half. Valproate mildly inhibits some hepatic cytochrome P450 enzymes and thus would have less effect on other medications, and would, if anything, mildly increase their levels. Carbamazepine also has a small risk of leukopenia and, rarely, aplastic anemia (1 in 500,000 risk). Valproate is associated with mild inhibition of certain clotting factors, thus leading to easy bruising and possibly a mild bleeding tendency; therefore, aspirin should be avoided with it and signs of hemorrhage and bleeding should be observed carefully. Valproate also reduces the platelet count and can increase liver enzyme levels, although risks of fatal hepatotoxicity are low with valproate monotherapy in adults. In children and adolescents, however, the hepatic effects of valproate should be followed carefully, and the medication should not be prescribed for children below age 2 years.

The Acute Mixed Episode Several investigators report poorer outcome with lithium therapy in patients with mixed episodes (dysphoric mania, depressive mania) compared with patients with pure affective episodes. Earlier open studies suggested that ❉

21

❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

such patients may respond better to anticonvulsants than to lithium (Post 1990; Post et al. 1989). This finding was confirmed in a double-blind comparison study (Swann et al. 1997) of valproate versus lithium, in which the treatment response rate for valproate was substantially better than lithium for mixed mania.

Breakthrough Manic/Mixed Episodes and Rapid Cycling The two best approaches to handling breakthrough mood elevation after adequate mood stabilizer monotherapy would be to combine two mood stabilizers or to add an atypical neuroleptic medication to a mood stabilizer. We should emphasize that this is our view, and it diverges from common clinical practice (IMS America 1998). In the standard view, antidepressant agents are often used for depressive symptoms occurring in the context of mixed episodes or rapid cycling. We provide our reasons for why we would instead avoid antidepressants and use combination mood stabilizing treatments and/or adjunctive neuroleptic agents instead. Combining Mood Stabilizers Most of the literature on combination treatments of mood stabilizers is uncontrolled (Post 1990). However, these studies seem to provide increasing evidence for the utility of combining lithium with either valproate or carbamazepine or both. In Post’s review of this literature, there were 15 uncontrolled studies of valproate augmentation of lithium in bipolar disorder that when combined yielded a 62% (246/400) success rate in acute treatment and a 48% (183/380) success rate in prophylaxis (Post 1990). A few more recent studies shed further light on mood stabilizer combination therapy. In one study, 75% (47/63) of patients with refractory affective disorders (mostly bipolar disorder) responded when valproate was added to either lithium or carbamazepine (Schaff et al. 1993). A few reports have also noted the efficacy and relative safety of combining valproate and carbamazepine in the treatment of refractory bipolar disorder (Ketter et al. 1992; Tohen et al. 1994). There is only one randomized prospective study of the combination of lithium and valproate in the treatment of bipolar I disorder. In that open study, 12 patients followed for 1 year were randomly assigned either placebo or valproate added to lithium. The combination treatment was associated with less relapse (P=0.01) but more side effects (P=0.04) (Solomon et al. 1997). ❉

22

❉

❉

The Difficult-to-Treat Patient With Bipolar Disorder

❉

In another report, 6 of 18 (33%) patients responded to valproate plus lithium to a moderate to marked degree in an open 1-year prospective treatment, and 4 of those 6 had been resistant to other mood-stabilizing treatments in the past. In 3 of 7 (43%) patients, triple therapy with lithium, carbamazepine, and valproate was the only effective treatment, although marked benefit was seen in only 1 patient (Denicoff et al. 1997a). In another study from the same group assessing lithium plus carbamazepine combination therapy, 52 patients were followed in a double-blind randomized manner on either lithium alone, carbamazepine alone, or the combination for up to 3 years. Whereas treatment response (based on moderate to marked improvement on the Clinical Global Impresssion scale) was higher on the combination treatment (55%) than lithium (33%) or carbamazepine (31%) monotherapy, this difference was not statistically significant, possibly due to the small sample sizes. However, statistically significant benefit to combination therapy was clearly present in patients with rapid-cycling bipolar disorder (56% combination therapy response versus 28% lithium response and 19% carbamazepine response, P<0.05) (Denicoff et al. 1997b). In most of these studies, each mood stabilizer was used in full dose. There are minimal, if any, research data on using reduced doses of lithium and an anticonvulsant in combination. After all, this is what would test the hypothesis of a synergy between them: does a low dose of lithium work in the presence of an anticonvulsant and vice versa? Taking into account the additive side effects in combination treatment, we sometimes use the agents in this manner, one in full dose and the other at a low (subtherapeutic) dose. Combining mood stabilizers requires attention to the pharmacokinetic effects of the medications. Carbamazepine, especially, should be cautiously introduced when using other medications that are metabolized by hepatic cytochrome P450 enzymes, because carbamazepine strongly induces those enzymes and thus lowers levels of medications metabolized by them (such as neuroleptics and antidepressants). Valproate, on the other hand, tends to inhibit those hepatic enzymes, thus potentially raising the levels of medications metabolized by this pathway. Enzyme induction can have important clinical consequences. In rare circumstances severe consequences may follow, as when a toxic metabolite is produced in significant quantities due to the induction of the P450 system. An interaction of this type can occur when valproate and carbamazepine are given concurrently. Under most circumstances in adults, valproate is metabolized mostly in the mitochondria by β-oxidation, and a relatively small amount is metabolized by the P450 metabolic pathway in the smooth endoplasmic reticulum. The latter path❉

23

❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

way produces potentially toxic metabolites—such as carbamazepine 10, 11 epoxide, and 4-en-valproate—but because the amount metabolized by this pathway typically is small, usually this effect is not clinically significant (Kesterson et al. 1984; Patsalos et al. 1985). If a patient is treated with carbamazepine as well, or another drug that induces the P450 pathway, the resulting increase in the toxic metabolite may reach potentially lethal levels. Whereas some suggest that patients often tolerate concurrent use of valproate and carbamazepine (Tohen et al. 1994), the use of this combination warrants caution and careful monitoring. New anticonvulsants may also be useful in the treatment of breakthrough episodes. In a large open naturalistic study, lamotrigine response, mainly as add-on therapy, was 50% for depressive symptoms and 48% for manic symptoms (based on 50% or greater decline in Young Mania Rating Scale or Hamilton Depression Rating Scale scores) in 40 rapid-cycling bipolar I patients (Calabrese et al. 1999a). A double-blind placebo controlled study (Calabrese et al. 1999b) of lamotrigine for acute bipolar depression also found it effective without increased risk of mania, whereas well-tolerated lamotrigine has a reported 1 in 1,000 (0.01%) risk of Stevens-Johnson syndrome. This generalized toxic epidermal rash can be fatal in some individuals. Lamotrigine also has a 10%–15% reported risk of a nonserious rash in all individuals. However, any rash is cause for concern, and the safest course is to discontinue the drug if any rash occurs. The risk of rash increases in combination treatment with valproate, however, and is lower with cotherapy with lithium. Thus, the former combination needs to be cautiously used; valproate reduces lamotrigine’s hepatic clearance, raising lamotrigine levels and thus leading to a lower necessary dose. Valproate also increases the half-life of lamotrigine from 25 to 70 hours. In the large open study, the mean dose of lamotrigine used was 287 mg/day in monotherapy and 105 mg/ day with valproate. The risk of rash is also lower if lamotrigine is titrated slowly; the method used in that study appears reasonable: 25 mg/day for the first 2 weeks, then 50 mg/day for weeks 3 and 4, then increasing 25 mg/week until clinical effect. This titration should be halved in dose with valproate. Some clinicians recommend an even slower titration, with an increase in daily dose of only 12.5 mg/week. This more conservative approach may be particularly useful in situations when the urgency of the clinical need for mood stabilization might allow a slower titration. A safer alternative is gabapentin, which has gained in popularity. Gabapentin’s advantages in safety are counteracted by its lack of evidence for efficacy, however. In two double-blind studies of treatment-refractory mania ❉

24

❉

❉

The Difficult-to-Treat Patient With Bipolar Disorder

❉

(Pande 1999) and bipolar depression (Frye et al. 2000), gabapentin was not more effective than placebo. Although it may have anti-anxiety effects, its mood-stabilizing effects, at least in refractory bipolar illness, have not been established. Naturalistic data suggest it may possess efficacy in nonrefractory bipolar II illness, but at this point gabapentin cannot be recommended as a mood-stabilizing agent for the refractory bipolar patient. Topiramate may also possess mood-stabilizing benefits as well as the useful side effects of promoting weight loss (Ghaemi and Gaughan 2000). Neuroleptic Agents The research literature on the use of typical neuroleptic agents for acute mania is limited. Early studies supported the use of haloperidol added to lithium for acute mania; this appeared to enhance rapidity of the antimanic response. Nonetheless, in the patient that is unresponsive to valproate or lithium, the addition of a neuroleptic is the first line of treatment. This is the case whether or not the patient is psychotic, because the neuroleptic agent is being used for its antimanic effect, not its antipsychotic effect. There is no evidence that the typical neuroleptic agents differ among themselves in antimanic effects, except in terms of side effects (Janicak et al. 1988a) and possibly increased speed of effect with haloperidol compared with chlorpromazine (Shopsin et al. 1975). Another study found chlorpromazine to be probably superior to lithium for initial control of psychomotor agitation (Prien et al. 1972), but the bulk of the literature supports lithium’s greater efficacy in the long-term treatment of mania (reviewed in Goodwin and Jamison 1990). However, there is evidence that continued typical neuroleptic treatment may be associated with a greater frequency or severity of postmania depressions (Kukopulos et al. 1980; Morgan 1972). Also, typical neuroleptic agents have elevated risks of extrapyramidal side effects and long-term tardive dyskinesia. For these reasons, recent researchers have begun to examine the use of atypical neuroleptic agents in acute mania. The atypical antipsychotic agents—clozapine, risperidone, olanzapine—may offer better results, but the literature available on the use of these agents with bipolar patients, until recently, was limited to case reports and open series. The reports were encouraging, however, as bipolar patients refractory to neuroleptics, anticonvulsants, and lithium were reported to derive substantial improvement on atypical antipsychotic agents that persist in the maintenance phase (Ghaemi and Goodwin 1999). There are now double-blind data available on olanzapine and risperidone in acute mania. In two studies, olanzapine was more effective than placebo in the inpatient ❉

25

❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

treatment of acute mania over 3 weeks using standard rating scales (Tohen 1999; Tohen et al. 1999). The Food and Drug Administration has approved olanzapine for the indication of acute mania based on these studies. In a study of risperidone (N=45), that agent was equally effective to lithium and haloperidol in the inpatient treatment of acute mania, with equivalent statistically significant declines on each drug in mania rating scale scores (Segal et al. 1998). These acute mania studies have been followed up with two open uncontrolled prospective studies of olanzapine and risperidone, mainly as add-on treatments to standard mood stabilizers, which provide suggestive evidence for antidepressant and antimanic (i.e., adjunctive mood-stabilizing) effects at 6 month follow-up (Sanger et al. 1999; Vieta et al. 1999). With risperidone and olanzapine, there have been case reports that these atypical antipsychotic agents may, in some cases, worsen manic symptoms (London 1998; Sajatovic et al. 1996). We believe that this is a controversy with little substance. The majority of researchers in this area appear near consensus that almost all such reports are due to monotherapy with atypical antipsychotic agents. Because these agents are not likely to be true mood stabilizers, of equal standing to lithium for instance, then it is not surprising that the use of ineffective treatments should lead to an unmasking of symptoms of an illness. The risk of tardive dyskinesia appears to be much lower with atypical than with typical neuroleptics and is apparently the same as the spontaneous yearly rate in patients with schizophrenia (Lemmens et al. 1999; Tollefson et al. 1997). Other Treatments Other treatments for breakthrough rapid cycling have been proposed, the most studied of which are calcium channel blockers. About one-third of patients have been reported to improve in a number of open naturalistic studies of treatment-resistant patients with bipolar disorder (Post et al. 1998). The specific agents used have included verapamil and nimodipine. Doubleblind data on verapamil have been mixed, however, with recent studies failing to find efficacy compared with placebo for acute mania (Janicak et al. 1998b). These agents may sometimes be useful, but this does not appear to be the case consistently. Benzodiazepines may be useful in the treatment of acute mania or in the control of agitation while using standard mood stabilizers for acute manic symptoms. This approach is often preferable to the use of traditional neuro❉

26

❉

❉

The Difficult-to-Treat Patient With Bipolar Disorder

❉

leptic agents in the control of acute mania because of the increased risks of neuroleptic agents in bipolar disorder. Clonazepam is an agent in particular that has been thought of as possibly possessing mood-stabilizing properties (Sachs 1990). However, a recent double-blind maintenance study found no added benefit to clonazepam when given with lithium, compared with placebo added to lithium, in the prophylaxis of bipolar mood episodes (Cosgrove et al. 1999). Thus, benzodiazepines probably have acute nonspecific antimanic/anti-agitation effects rather than true long-term mood-stabilizing properties. Omega-3 fatty acids have also generated interest. A recent double-blind study found benefit with these agents versus placebo, when added to standard mood stabilizers, in the prophylaxis of bipolar disorder (Stoll et al. 1999). However, that study was limited by using fatty acids in higher doses (9.6 g/day) than patients usually take (1–3 g/day) and by enrolling patients in variable mood states, thus complicating the assessment of improvement. Nonetheless, the limited risks of this natural substance make it appealing, and as an add-on treatment little could be said to argue against it other than cost and inconvenience at high doses.

Acute and Breakthrough Depression Few clinicians doubt that antidepressants that effectively treat unipolar depressed patients are effective treatment for bipolar patients. Unfortunately, because bipolar patients are routinely excluded from most antidepressant treatment trials, empirical data are limited. Also, as noted previously, antidepressants have significant risks of acute mania and long-term rapid cycling. As reviewed by Goodwin and Jamison (1990), rates of treatment-emergent mania in bipolar I patients treated with tricyclic antidepressants alone range from 31% to 70%. Thus, it is important to recognize that the treatment of resistant bipolar depression, as with resistant mania or cycling, involves, first and foremost, adequate mood-stabilizing treatment. If lithium or one of the anticonvulsants fail alone in treating bipolar depressive symptoms, we recommend combining lithium with an anticonvulsant before proceeding to the addition of an antidepressant agent. In the case of lithium nonresponse, especially, one of the factors could be subclinical hypothyroidism, often related to lithium (Bauer and Whybrow 1988; Hatterer et al. 1989). This level of thyroid dysfunction can be subtle and is often not reflected in traditional laboratory screening of thy❉

27

❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

roid-stimulating hormone (TSH) values. Often, patients will demonstrate normal TSH values and low free thyroid (T4 or T3) values. Sometimes the TSH value is normal, and the free T4 or T3 values are in the normal range as well, but they are in the lower half of the normal range. It is our practice to correct low or low-normal free T4 or T3 levels, usually using L-thyroxine, in the bipolar depressed patient until the free T4 or T3 levels are in the upper quartile of the normal range. Sometimes this correction of subtle thryoid dysfunction can lead to response to lithium or other mood stabilizers that previously had failed to elicit treatment response. Bauer and Whybrow also have reported a possible role for supranormal thyroid levels in treating rapid-cycling patients. They found that bringing free T4 levels into the 12–15 range converted 10 of 11 previous lithium nonresponders to responders (Bauer and Whybrow 1990). Although this is a single study, it deserves replication. Frye and colleagues (1999) also found that free T4 levels in the upper ranges of normal were associated with better long-term response to mood stabilizers compared with low normal free T4 levels. This evidence would suggest that the hypothalamic-pituitary axis (HPA) may go subtly awry and influence mood symptoms. Whether thyroid supplementation is correcting such subtle HPA abnormalities or directly augmenting mood-stabilizing effects of other agents through its own cellular effects is unclear. Also, it should be noted that valproate and carbamazepine can affect laboratory readings of thyroid hormone and TSH in complicated ways, which might produce artificial evidence of hypothyroidism (Pies 1997). In terms of novel agents, breakthrough depression may respond to the adjunctive use of atypical neuroleptic agents, especially if psychotic features are present. Risperidone was effective in monotherapy in 7 of 10 patients with unipolar psychotic depression or schizoaffective disorder, depressed type (Hillert et al. 1995). In another report, olanzapine was also more effective than typical neuroleptics as add-on treatment to antidepressants (Rothschild et al. 1999). These data are uncontrolled. Duration of use of atypical agents in these cases is unclear. Our practice is to continues the atypical antipsychotic agents in the absence of significant extrapyramidal symptoms (EPS) or any evidence of tardive dyskinesia and to switch to other treatments if serious EPS or early tardive dyskinesia begin. If lithium and anticonvulsants fail, and atypical neuroleptic agents with mood stabilizers proves fruitless, then antidepressant medications can be used for bipolar depression. Again, these recommendations to use atypical neuroleptic agents before antidepressant agents in treating breakthrough de❉

28

❉

❉

The Difficult-to-Treat Patient With Bipolar Disorder

❉

pression in bipolar disorder are not in line with conventional practice. However, in keeping with the known risks of mania and rapid cycling associated with antidepressants, we recommend a more conservative course than is commonly practiced—that is, avoiding antidepressants until combined mood stabilizer and atypical neuroleptic treatments have been tried. As discussed previously, bupropion and paroxetine appear to be less likely to cause acute mania. Thus, we prefer to use those agents. How long to continue antidepressant treatment of breakthrough depression is much debated. There are limited data, and researchers and clinicians hold differing opinions. Our view is that patients should receive at least a trial of withdrawal of the antidepressant agent after euthymia is achieved, as discussed above in the treatment of acute depression. If depressive relapse occurs, then the antidepressant may need to be continued. Recent data from our center suggest that in many patients long-term antidepressant use decreases the severity and duration of depression but at the price of rapid cycling and a chronic instability of mood. In extremely refractory patients, sometimes this clinical outcome is all that can be achieved. Another option would be to consider maintenance treatment with electroconvulsive therapy (ECT). A course of ECT for acute depression or mania typically consists of 4 to 12 treatments. Numerous studies have reported that ECT is effective in treating depression, usually more so than tricyclic antidepressants (reviewed in Goodwin and Jamison 1990). However, most studies have not carefully separated patients with bipolar depression from unipolar depression. Those studies that have tried to differentiate bipolar from unipolar depression do not suggest a clearly different response pattern for acute depression (Abrams and Taylor 1974; Black et al. 1987; Perris 1966). However, a study by Kukopulos and colleagues (1980) suggests that ECT may cause less long-term rapid cycling than tricyclic antidepressants in the maintenance treatment of bipolar illness and thus it may be preferable for treating bipolar depression on a continuous basis. Bilateral ECT may be more effective than unilateral ECT, but unilateral ECT may have fewer cognitive side effects. Earlier work by Angst (1985) cautions against this conclusion, however, reporting a sharp increase in hospitalizations for mania at the Burgholzli Hospital during the decade after the introduction of ECT. On the other hand, ECT is an example of a malady that is itself the remedy, because ECTinduced mania or hypomania can be treated with continued ECT. In sum, whereas ECT appears to be a useful option for severe acute depression, the evidence for its utility in long-term therapy of treatment-resistant bipolar depression remains unclear. ❉

29

❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

The Role of Psychotherapy Most of this chapter has focused on pharmacological treatments for the resistant patient. One factor that should not be neglected, however, is the potential role of psychotherapies in the management of the treatment-resistant patient. Whereas medication is the main treatment for bipolar disorder, psychotherapy can help greatly with the interpersonal consequences of the illness, and recent research suggests that psychotherapy may even enhance the pharmacological effect of biological treatments. The first role for psychotherapy is to establish a therapeutic alliance. Research on depression suggests that the therapeutic alliance is an important factor in maximizing the beneficial effects of medication treatment (Krupnick et al. 1996). In the patient with bipolar disorder, with whom a good deal of psychoeducational work usually is required in distinguishing unipolar depression from bipolar depressive syndromes, having a therapeutic alliance is a necessary first step toward building a compassionate collaborative treatment. The alliance is established through attention to psychotherapeutic issues such as transference and countertransference, as well as defense mechanisms, in the course of psychopharmacology visits. It may be further enhanced through separate or concomitant psychotherapy sessions. An important goal in the treatment of patients with bipolar disorder is to address some of the common issues that arise. As discussed in greater detail elsewhere (Goodwin and Jamison 1990), these include the following: patient anger at the frustrations of the illness, denial about the nature of the illness, and ambivalence about accepting biological treatments for psychological experiences. Patients are often disappointed at partial treatment success, fear recurrence of the illness, and are concerned by the genetic implications of the diagnosis for their offspring. They also need to learn to discriminate their moods, distinguishing normal mood from hypomania, for instance, to better understand the effects of medications and the status of their condition. Some patients appear to enjoy their hypomanic periods. It is our experience, however, that many patients are quite willing to give up hypomanic symptoms if they are made to understand that it will also relieve their depressive symptoms. Patients should be taught that, by and large, they pay for their hypomanias with their depressions and that the only way to treat their depressions is to treat their overall illness, hypomania included. Miklowitz (1996) has suggested that the two main goals of psychotherapy for patients with bipolar disorder should be to directly reduce the ❉

30

❉

❉

The Difficult-to-Treat Patient With Bipolar Disorder

❉

impact of psychosocial stressors on triggering episodes of illness and to assist patients in adapting to the influence of such stressors on their lives. Miklowitz has reviewed two specific forms of manualized standardized psychotherapies that are being assessed for bipolar disorder: behavioral family management for bipolar disorder (BFM-BP) and interpersonal and social rhythm therapy (IP/SRT), to which we would add a third, cognitivebehavior therapy (CBT). BFM-BP (Falloon et al. 1985) grew out of work in schizophrenia and is a family therapy that focuses on social skills training and psychoeducation. Preliminary evidence suggests that it may reduce relapse when given with pharmacotherapy for patients with bipolar disorder (1/9 relapse in 9 months) compared with standard care (14/23 relapse in 9 months). IP/SRT (Frank 1995) grew out of research in unipolar depression and applies a focus on current interpersonal issues along with an emphasis on regularizing social rhythms (the daily course of social contacts and stimulation). Early results suggest that IP/SRT does indeed regularize social rhythms in patients with bipolar disorder; the effects of this intervention on relapse are currently being studied. CBT is being applied in individual and group formats to patients with bipolar disorder and focuses on cognitive restructuring of depressive and manic modes of thinking and behavioral retraining. An early study (Cochran 1984) found that 14 patients who received CBT with lithium showed a decline in rehospitalization and increased medication compliance compared with 14 patients who received lithium under standard care conditions. Current studies on the effects of individual and group CBT on relapse are under way. Another important technique for enhancing the psychotherapeutic treatment of bipolar disorder is mood charting. Mood charting is an essential aspect of engaging the patient with bipolar disorder in the treatment proactively while providing the clinician with probably more accurate diagnostic and course assessments than with traditional interview techniques alone. Mood charts consist essentially of daily ratings of mood (depressed, euthymic, elevated) along with recordings of sleep, anxiety, irritability, and medication use, all provided by the patient. This method often leads to a recognition by the patients of certain mood patterns that might otherwise have been ignored. It also allows the clinician to obtain information that is not biased by the retrospective recall over weeks (and sometimes months) on which one is forced to rely in typical office visit circumstances.

❉

31

❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

Clinical Vignette Mr. R is a 38 year-old white male diagnosed with bipolar disorder 3 years previously. He had first visited a psychiatrist during a major depressive episode 10 years ago and was diagnosed with unipolar major depressive disorder. He was treated with imipramine, fluoxetine, and sertraline in succession with no improvement in his depression and unstable mood swings. While taking sertraline, he developed a psychotic manic episode, and his diagnosis was changed to bipolar disorder. He had no history of psychosis except during his most severe manic episodes. Lithium was added to sertraline with little benefit. Valproate was added to lithium and sertraline with mild benefit. Both lithium (0.7 ng/dL) and valproate (71 ng/ dL) levels were therapeutic. He continued to have on average two manic, two hypomanic, and two major depressive episodes per year. Sertraline was tapered off, and he experienced a moderate reduction in frequency of his episodes, but he continued to be symptomatic. Because of the patient’s particular difficulty with insomnia, doxepin was added. He slept better, but the frequency of manic and hypomanic episodes increased. Doxepin was discontinued. Risperidone was added at 1 mg/day but then discontinued due to akathisia. At this point, clonazepam was added for insomnia, and gabapentin 600 mg/day was added to lithium and valproate . A low dose of synthroid 0.05 mg/day was also added because of a low normal free T4 level of 0.9 (range 0.7–1.5), although TSH was normal at 2.2. He developed nausea at 900 mg/day of gabapentin and did not benefit from 600 mg/day. Gabapentin was discontinued. He continued to have frequent episodes of hypomania and was fearful of the beginning of a severe manic episode. Olanzapine was added, with a slow titration to 10 mg/day. He experienced marked benefit, but then swung into a major depressive episode. He also experienced akathisia, but this was controlled with propanolol 20 mg twice daily. Olanzapine was reduced to 5 mg/day, and the depressive symptoms resolved with longer lasting euthymic periods than he had experienced in the past.

Recommended Treatment Algorithms The review of the literature presented in this chapter suggests a treatment algorithm (Figure 2–4). As described earlier, the first premise of this algorithm is that mood stabilizers need to be aggressively utilized in the treatment of bipolar disorder and that antidepressant agents should be avoided where possible and used briefly and intermittently where necessary. An increase in lithium dose should be considered in the absence of significant side effects if there is lack of response to low or low-therapeutic lithium levels. Also, thyroid function should be followed continuously and carefully, and free thyroid hormone levels should be corrected or optimized to the upper-half ❉

32

❉

❉

The Difficult-to-Treat Patient With Bipolar Disorder

❉

Figure 2–4. An algorithm for the treatment of bipolar disorder. Note. Stages II and III can be reversed in order depending on the clinical setting. Source. Adapted from Ghaemi et al. 1999.

of the normal range, even in the presence of normal TSH values. Thyroid function should be corrected before proceeding to combination treatments based on the algorithm. The algorithm emphasizes combining mood-stabilizing agents before adding antidepressant agents. Adding atypical neuroleptic agents is a useful alternative as well. The effectiveness of these strategies is enhanced by specific psychotherapeutic approaches. In summary, a good treatment paradigm for bipolar disorder is to aggressively maximize mood-stabilizing agents, combining them where needed, and adding atypical neuroleptic agents as preferred adjunctive medications. Antidepressant agents are reserved for those who do not show an antidepressant response to aggressive combination treatments after 1–3 months of pure depression, and then the antidepressant agents would be tapered off after 1– 3 months of euthymia. ❉

33

❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

Summary Bipolar disorder often is a refractory illness. Treatment resistance is particularly prominent in particular diagnostic subtypes of bipolar disorder, such as mixed episodes and rapid cycling, and it also appears to be associated with chronic antidepressant treatment. Clinical correlates of treatment resistance include onset of illness with a major depressive episode, substance abuse, mood-incongruent psychotic features, and psychiatric and medical comorbidities. Management involves replacing or combining lithium treatment with anticonvulsants or atypical antipsychotic agents. Other adjuncts include benzodiazepines, thyroid hormone, and ECT for the most refractory cases. Antidepressants should be used cautiously, mainly in the acute depressive episodes, and always with concomitant mood stabilizers. Above all, the treatment of bipolar disorder, whether refractory or not, is complex and requires careful attention to the therapeutic alliance.

References Abrams R, Taylor M: Unipolar and bipolar depressive illness: phenomenology and response to electroconvulsive therapy. Arch Gen Psychiatry 30:320–321, 1974 Akiskal HS: The prevalent clinical spectrum of bipolar disorders: beyond DSM-IV. J Clin Psychopharmacol 16(suppl 1):4S–14S, 1996 Altshuler LL, Post RM, Leverich GS, et al: Antidepressant-induced mania and cycle acceleration: a controversy revisited. Am J Psychiatry 152:1130–1138, 1995 Amsterdam J: Efficacy and safety of venlafaxine in the treatment of bipolar II major depressive episode. J Clin Psychopharmacol 18:414–417, 1998 Amsterdam JD, Garcia-Espana F, Fawcett J, et al: Efficacy and safety of fluoxetine in treating bipolar II major depressive episode. J Clin Psychopharmacol 18:435– 440, 1998 Angst J: Switch from depression to mania: a record study over decades between 1920 and 1982. Psychopathology 18:140–154, 1985 Anthony JC, Folstein M, Romanoski AJ: Comparison of lay DIS and a standardized psychiatric diagnosis. Arch Gen Psychiatry 42:667–675, 1985 Bauer M, Whybrow P: Thyroid hormones and the central nervous system in affective illness: interactions that may have clinical significance. Integrative Psychiatry 6:75–100, 1988 Bauer M, Whybrow P: Rapid cycling bipolar affective disorder, II: treatment of refractory rapid cycling with high-dose levothyroxine: a preliminary study. Arch Gen Psychiatry 47:435–440, 1990

❉

34

❉

❉

The Difficult-to-Treat Patient With Bipolar Disorder

❉

Biederman J, Faraone S, Mick E, et al: Attention deficit disorder and juvenile mania: an overlooked comorbidity. J Am Acad Child Adolesc Psychiatry 35:997–1008, 1996 Black D, Winokur G, Nasrallah H: The treatment of depression: electroconvulsive therapy vs. antidepressants: a naturalistic evaluation of 1495 patients. Compr Psychiatry 28:169–182, 1987 Bowden C, Brugger A, Swann A, et al: Efficacy of divalproex vs. lithium and placebo in the treatment of mania. JAMA 271:918–924, 1994 Calabrese J, Delucchi GA: Spectrum of efficacy of valproate in 55 patients with rapidcycling bipolar disorder. Am J Psychiatry 147:431–434, 1990 Calabrese J, Bowden C, McElroy S, et al: Spectrum of activity of lamotrigine in treatment-refractory bipolar disorder. Am J Psychiatry 156:1019–1023, 1999a Calabrese J, Bowden C, Sachs G, et al: A double-blind placebo-controlled study of lamotrigine monotherapy in outpatients with bipolar I depression. J Clin Psychiatry 60:79–88, 1999b Cochran S: Preventing noncompliance in the outpatient treatment of bipolar affective disorders. J Consult Clin Psychol 52:873–878, 1984 Cohn JB, Collins G, Ashbrook E, et al: A comparison of fluoxetine, imipramine and placebo in patients with bipolar depressive disorder. Int Clin Psychopharmacol 4:313–314, 1989 Cosgrove V, Ghaemi S, Baldassano C, et al: A double-blind placebo-controlled study of clonazepam as an adjunct to lithium maintenance treatment of bipolar disorder. Abstract presented at the 152nd Annual Meeting of the American Psychiatric Association, Washington, DC, May 15–20, 1999 Cowdry RW, Wehr TA, Zis AP, et al: Thyroid abnormalities associated with rapid cycling bipolar illness. Arch Gen Psychiatry 40:414–420, 1983 Denicoff K, Smith-Jackson E, Bryan A, et al: Valproate prophylaxis in a prospective clinical trial of refractory bipolar disorder. Am J Psychiatry 154:1456–1458, 1997a Denicoff K, Smith-Jackson E, Disney E, et al: Comparative prophylactic efficacy of lithium, carbamazepine, and the combination in bipolar disorder. J Clin Psychiatry 58:470–478, 1997b Falloon I, Boyd J, McGill C, et al: Family management in the prevention of morbidity in schizophrenia. Arch Gen Psychiatry 42:887–896, 1985 Frank E, Swartz HA, Kupfer DJ: Interpersonal and social rhythm therapy: managing the chaos of bipolar disorder. Biol Psychiatry 48:593–604, 2000 Frye MA, Denicoff KD, Bryan AL, et al: Association between lower serum free T4 and greater mood instability and depression in lithium-maintained bipolar patients. Am J Psychiatry 156:1909–1914, 1999 Frye MA, Ketter TA, Kimbrell TA, et al: A placebo-controlled study of lamotrigine and gabapentin monotherapy in refractory mood disorders. J Clin Psychopharmacol 20:607–614, 2000

❉

35

❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

Ghaemi SN: Insight and psychiatric disorders: a review of the literature, with a focus on its clinical relevance for bipolar disorder. Psychiatric Annals 27:782–790, 1997 Ghaemi SN, Gaughan S: Novel anticonvulsants: a new generation of mood stabilizers? Harv Rev Psychiatry 8:1–7, 2000 Ghaemi SN, Goodwin F: Use of atypical antipsychotic agents in bipolar and schizoaffective disorders: review of the empirical literature. J Clin Psychopharmacol 19:354–361, 1999 Ghaemi SN, Katzow JJ, Desai SP, et al: Gabapentin treatment of mood disorders: a preliminary study. J Clin Psychiatry 59:426–429, 1998a Ghaemi SN, Sachs GS, Chiou AM, et al: Is bipolar disorder still underdiagnosed? Are antidepressants overutilized? J Affect Disord 52:135–144, 1999 Goldberg JF, Harrow M, Grossman LS: Course and outcome in bipolar affective disorder: a longitudinal follow-up study. Am J Psychiatry 152:379–384, 1995 Goodwin FK, Ghaemi SN: An introduction to and history of affective disorders, in New Oxford Textbook of Psychiatry, Vol 1. Edited by Gelder M, Lopez-Ibor J, Andreasen N. New York, Oxford University Press, 2000, pp 677–682 Goodwin FK, Jamison KR: Manic Depressive Illness. New York, Oxford University Press, 1990 Hatterer J, Koscis J, Stokes P: Thyroid function in patients maintained on lithium. Psychiatry Research 26:249–258, 1989 Hetmar O, Bolwig T, Brun C, et al: Lithium: long-term effects on the kidney, I: renal function in retrospect. Acta Psychaitrica Scandinavica 73:574–581, 1986 Hibbeln JR, Salem N: Dietary polyunsaturated fatty acids and depression: when cholesterol does not satisfy. Am J Clin Nutr 62:1–9, 1995 Hillert A, Maier W, Wetzel H, et al: Risperidone in the treatment of disorders with a combined psychotic and depressive syndrome: a functional approach. Pharmacopsychiatry 25:213–217, 1995 IMS America: National Disease and Therapeutic Index. New Jersey, IMS America, 1998 Janicak P, Bresnahan D, Sharma R, et al: A comparison of thiothixene with chlorpromazine in the treatment of mania. J Clin Psychopharmacol 8:33–37, 1988a Janicak P, Sharma R, Pandey G, et al: Verapamil for the treatment of acute mania: a double-blind placebo controlled trial. Am J Psychiatry 155:972–973, 1998b Keitner GI, Solomon DA, Ryan CE, et al: Prodromal and residual symptoms in bipolar I disorder. Compr Psychiatry 37:362–367, 1996 Kessler RC, McGonagle KA, Zhao S: Lifetime and 12-month prevalence of DSM-IIIR psychiatric disorders in the United States. Arch Gen Psychiatry 51:8–19, 1994 Kesterson J, Granneman G, Machinist J: The hepatotoxicity of valproic acid and its metabolites in rats. Hepatology 4:1143–1152, 1984 Ketter TA, Pazzaglia PJ, Post RM: Synergy of carbamazepine and valproic acid in affective illness: case report and review of the literature. J Clin Psychopharmacol 12:276–281, 1992

❉

36

❉

❉

The Difficult-to-Treat Patient With Bipolar Disorder

❉

Krupnick J, Sotsky S, Simmens S, et al: The role of the therapeutic alliance in psychotherapy and psychopharmacology outcome: findings in the National Institute of Mental Health Treatment of Depression Collaborative Research Program. J Consult Clin Psychol 64:532–539, 1996 Kukopulos A, Reginaldi P, Laddomada G, et al: Course of the manic-depressive cycle and changes caused by treatments. Pharmakopsychiatrie 13:156–167, 1980 Lemmens P, Brecher M, Van Baelen B: A combined analysis of double-blind studies with risperidone vs. placebo and other antipsychotic agents: factors associated with extrapyramidal symptoms. Acta Psychiatr Scand 99:160-170, 1999 Levy A, Drake M, Shy K: EEG evidence of epileptiform paroxysms in rapid cycling bipolar patients. J Clin Psychiatry 49:232–234, 1988 London J: Mania associated wtih olanzapine (letter). J Am Acad Child Adolesc Psychiatry 37:997–1008, 1998 McElroy S, Keck P, Stanton S, et al: A randomized comparison of divalproex oral loading versus haloperidol in the initial treatment of acute psychotic mania. J Clin Psychiatry 57:142–146, 1996 Miklowitz DJ: Psychotherapy in combination with drug treatment for bipolar disorder. J Clin Psychopharmacol 16(suppl 1):56S–66S, 1996 Morgan H: The incidence of depressive symptoms during recovery from hypomania. Br J Psychiatry 120:537–539, 1972 Page C, Benaim S, Lappin F: A long-term retrospective follow-up study of patients treated with prophylactic lithium carbonate. Br J Psychiatry 150:175–179, 1987 Pande A: Combination treatment in bipolar disorder (abstract). Bipolar Disorders 1(suppl 1):17, 1999 Patsalos R, Stephenson T, Krishna S, et al: Side effects induced by carbamazepine 10,11 epoxide (letter). Lancet 2:496, 1985 Peet M: Induction of mania with selective serotonin re-uptake inhibitors and tricyclic antidepressants. Br J Psychiatry 164:549–550, 1994 Perris C: A study of bipolar (manic-depressive) and unipolar recurrent depressive psychoses. Acta Psychiatr Scand 42(suppl 194):15–152, 1966 Peselow E, Dunner D, Fieve R: Lithium carbonate and weight gain. J Affect Disord 2:303–310, 1980 Pies RW: The diagnosis and treatment of subclinical hypothyroid states in depressed patients. Gen Hosp Psychiatry 19:344–354, 1997 Post RM: Non-lithium treatment for bipolar disorder. J Clin Psychiatry 51(suppl 8):9– 16, 1990 Post RM, Rubinow DR, Uhde TW, et al: Dysphoric mania: clinical and biological correlates. Arch Gen Psychiatry 46:353–358, 1989 Post RM, Frye MA, Leverich GS, et al: The role of complex combination therapy in the treatment of refractory bipolar illness. CNS Spectrums 3:66–86, 1998

❉

37

❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

Prien R, Caffey EJ, Klett C: Comparison of lithium carbonate and chlorpromazine in the treatment of mania: report of the Veterans Administration and National Institute of Mental Health Collaborative Study Group. Arch Gen Psychiatry 26:146– 153, 1972 Prien RF, Klett CJ, Caffey EM: Lithium prophylaxis in recurrent affective illness. Am J Psychiatry 131:198–203, 1974 Prien R, Kupfer D, Mansky P, et al: Drug therapy in the prevention of recurrences in unipolar and bipolar affective disorders. Arch Gen Psychiatry 41:1096–1104, 1984 Quitkin FM, Kane J, Rifkin A, et al: Prophylactic lithium carbonate with and without imipramine for bipolar 1 patients. Arch Gen Psychiatry 38:902–907, 1981 Regier DA, Kaelber CT: The epidemiologic catchment area (ECA) program: studying the prevalence and incidence of psychopathology, in Textbook in Psychiatric Epidemiology. Edited by Tsuang MT, Tohen M, Zahner GEP. New York, Wiley, 1995, pp 133–157 Regier DA, Farmer ME, Rae DS, et al: Comorbidity of mental disorders with alcohol and other drug abuse: results from the Epidemilogic Catchment Area (ECA) Study. JAMA 264:2511–2518, 1990 Rothschild AJ, Bates KS, Boehringer KL, et al: Olanzapine response in psychotic depression. J Clin Psychiatry 60:116–118, 1999 Sachs G: Use of clonazepam for bipolar affective disorder. J Clin Psychiatry 51(5 suppl):31–34, 1990 Sachs GS, Lafer B, Stoll AL, et al: A double-blind trial of bupropion versus desiprimine for bipolar depression. J Clin Psychiatry 55:391–393, 1994a Sachs GS, Lafer B, Truman CJ, et al: Lithium monotherapy: miracle, myth and misunderstanding. Psychiatric Annals 24:299–306, 1994b Sajatovic M, DiGiovanni SK, Bastani B, et al: Risperidone therapy in treatment refractory acute bipolar and schizoaffective mania. Psychopharmacol Bull 32:55–61, 1996 Sanger T, Tohen M, Jacobs T, et al: Long-term olanzapine treatment of mania: improvement in mood symptoms and cognitive function. Abstract presented at the 152nd annual meeting of the American Psychiatric Association, Washington, DC, May 15–20, 1999 Schaff MR, Fawcett J, Zajecka JM: Divalproex sodium in the treatment of refractory affective disorders. J Clin Psychiatry 54:380–384, 1993 Segal J, Berk M, Brook S: Risperidone compared with both lithium and haloperidol in mania: a double-blind randomized controlled trial. Clin Neuropharmacol 21:176–180, 1998 Shopsin B, Gershon S, Thompson H, et al: Psychoactive drugs in mania: a controlled comparison of lithium carbonate, chlorpromazine, and haloperidol. Arch Gen Psychiatry 32:34–42, 1975

❉

38

❉

❉

The Difficult-to-Treat Patient With Bipolar Disorder

❉

Solomon DA, Ryan CE, Keitner GI, et al: A pilot study of lithium carbonate plus divalproex sodium for the continuation and maintenance treatment of patients with bipolar I disorder. J Clin Psychiatry 58:95–99, 1997 Sprock J: Classification of schizoaffective disorder. Compr Psychiatry 29:55–71, 1988 Stoll AL, Severus WE, Freeman MP, et al: Omega 3 fatty acids in bipolar disorder: a preliminary double-blind, placebo-controlled trial (see comments). Arch Gen Psychiatry 56:407–412, 1999 Swann A, Bowden C, Morris D, et al: Depression during mania: treatment response to lithium or divalproex. Arch Gen Psychiatry 54:37–42, 1997 Tohen M: Use of antipsychotics in bipolar disorder (abstract). Bipolar Disorders 1:18, 1999 Tohen M, Waternaux CM, Tsuang M: Outcome in mania: a 4-year prospective followup of 75 patients utilizing survival analysis. Arch Gen Psychiatry 47:1106–1111, 1990 Tohen M, Castillo J, Pope H, et al: Concomitant use of valproate and carbamazepine in bipolar and schizoaffective disorders. J Clin Psychopharmacol 14:67–70, 1994 Tohen M, Sanger T, McElroy S, et al: Olanzapine vs. placebo in the treatment of acute mania. Am J Psychiatry 156:702–709, 1999 Tollefson GD, Beasley CM Jr, Tamura RN, et al: Blind, controlled, long-term study of the comparative incidence of treatment-emergent tardive dyskinesia with olanzapine or haloperidol. Am J Psychiatry 154:1248–1254, 1997 Tsuang MT, Winokur G, Crowe RR: Morbidity risks of schizophrenia and affective disorders among first degree relatives of patients with schizophrenia, mania, depression, and surgical conditions. Br J Psychiatry 137:497–504, 1980 Vieta E, Herraiz M, Fernandez A, et al: Efficacy and safety of risperidone in bipolar disorders. Abstract presented at the 152nd annual meeting of the American Psychiatric Association, Washington, DC, May 15–20, 1999 Wehr TA, Goodwin FK: Can antidepressants cause mania and worsen the course of affective illness? Am J Psychiatry 144:1403–1411, 1987 Wehr TA, Sack DA, Rosenthal NE, et al: Rapid cycling affective disorder: contributing factors and treatment responses in 51 patients. Am J Psychiatry 145:179–184, 1988 Young ML, Pitts CD, Oakes R, et al: A double-blind placebo-controlled trial comparing the effect of paroxetine and imipramine in the treatment of bipolar depression. Article presented at the 2nd International Conference on Bipolar Disorder. Pittsburgh, PA, June 19–21, 1997

❉

39

❉

This page intentionally left blank

3 The Difficult-to-Treat Patient With Schizophrenia Ronald W. Pies, M.D. Mantosh J. Dewan, M.D.

Introduction It is arguable that the patient with schizophrenia is inherently “difficult to treat” because schizophrenia is a chronic, debilitating, and often incapacitating mental illness. However, experienced clinicians know that schizophrenia falls along a continuum of severity and that numerous intrinsic and extrinsic factors make some patients much more difficult to manage than others. In this chapter, we review the expected outcome of somatic and psychosocial therapies for schizophrenia, factors that contribute to treatment resistance, and specific somatic and psychosocial strategies that we have found useful in refractory cases. We also present two treatment algorithms that integrate the use of antipsychotic medication with psychosocial interventions. ❉

41

❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

Incidence/Prevalence/Course of Schizophrenia The lifetime prevalence of schizophrenia is approximately 1%, with rates across four U.S. sites ranging from 0.6%–1.9% of the population (American Psychiatric Association 1997). Schizophrenia has a variable course, with some individuals free of further episodes, a small proportion remaining chronically psychotic, and the majority displaying exacerbations and remissions (American Psychiatric Association 1997). Complete remission is uncommon. There is some evidence of a triphasic symptom progression in schizophrenia, with early negative features appearing as a prodrome, followed by prominent positive symptoms (hallucinations, delusions), and, finally, a predominance of negative features later in the course of illness (American Psychiatric Association 1997). Positive symptoms tend to be more responsive to classical antipsychotics (neuroleptics) than negative symptoms, perhaps accounting for the predominance of negative symptoms between acute psychotic episodes. Better outcome in schizophrenia appears to be associated with female sex, family history of affective disorder, lack of family history of schizophrenia, good premorbid functioning, higher IQ, marriage, acute onset with precipitating stress, and paranoid subtype (American Psychiatric Association 1997). Unfortunately, many patients continue to have marked disability despite the use of standard treatments for schizophrenia.

Expected Response Rates With Conventional Somatic Treatments Both positive and negative symptoms may initially respond to classical neuroleptics (Breier et al. 1987), although most studies show a strongly preferential long-term effect on positive symptoms. Around 60% of patients will show substantial improvement over 6 weeks of treatment with conventional antipsychotics versus around 20% treated with placebo. During the first year of treatment, only approximately 30% of neuroleptic-treated patients relapse versus approximately 65% on placebo (American Psychiatric Association 1997). Despite the evident benefits of neuroleptics, about 40% of neuroleptic-treated patients continue to show moderate to severe psychotic symptoms, and 8% show no improvement or become worse (American Psychiatric Association 1997). Further, Weiden et al. (1996) have stated that under “real world conditions,” schizophrenic patients maintained on depot neuroleptics will show a 1-year rehospitalization rate of about 50%, with loss of medication response accounting for about 60% of this relapse rate and medication noncompliance accounting for the remaining 40%. ❉

42

❉

❉

The Difficult-to-Treat Patient With Schizophrenia

❉

Response Rates to Newer and Atypical Agents Arguably the gold standard for treatment, clozapine is effective in about 40%–78% of neuroleptic-refractory patients (Buckman and Malan 1999). Rates of response in well-defined refractory populations have not yet been determined for the other atypical antipsychotics (AAPs), although clinical data look promising for risperidone and olanzapine (Chakos et al. 2001). In general, AAPs are as effective as conventional agents for the treatment of positive symptoms and are superior to conventional agents for negative, affective, and cognitive features of schizophrenia, with improved cognitive function in a number of domains (Berman et al. 1999). This has important implications for social and vocational function. Moreover, adequacy of response in schizophrenic patients is increasingly being seen in terms of general well-being and quality of life, rather than simply in terms of symptom reduction (Noordsy et al. 2000). With respect to relapse rates on AAPs, Weiden et al. (1996) have noted that rehospitalization rates are about 8% and 14% lower with olanzapine and clozapine, respectively, than with conventional neuroleptics. On the other hand, olanzapine was not effective in a recent open trial of patients with treatment-refractory schizophrenic or schizoaffective psychoses (Sanders and Mossman 1999); indeed, several patients became more agitated within the first several weeks. Similar reports of olanzapine-related agitation suggest that high doses (>15–20 mg/day) may be countertherapeutic in some patients. There are few controlled studies comparing one AAP with another (Chakos et al. 2001). However, one study comparing clozapine with risperidone in partial responders to conventional agents found that clozapine was superior to risperidone for positive symptoms and extrapyramidal side effects (EPS) but did not differ from risperidone with respect to negative symptoms or depression (Breier et al. 1999). Studies comparing the efficacy of risperidone with olanzapine have yielded conflicting results (Ho et al. 1999; Tran et al. 1997) and have numerous methodological weaknesses. It is clear, however, that EPS (including akathisia) is more common with risperidone than with olanzapine (Ho et al. 1999).

Response to Conventional Psychosocial Therapies Despite early promise (Karon and VandenBos 1972), controlled outcome studies failed to find significant benefit from psychotherapy alone when ❉

43

❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

compared with medication (American Psychiatric Association 1997). However, both McGlashen (1986) and the present authors affirm the critical importance of the relationship between doctor and patient in the treatment of schizophrenia. Furthermore, adding supportive psychotherapy to medications may be helpful. Stanton et al. (1984) found that reality-oriented supportive strategies appeared to be as good as or better than insight-oriented approaches in subacute schizophrenic outpatients. Hogarty et al. (1997a) compared a “disorder relevant” personal therapy with supportive therapy and family therapy in 151 patients followed for 3 years after discharge. All patients were also on medications. For patients receiving personal therapy, the researchers reported an impressively low attrition rate of 8% (versus 23% in the comparison group) and decreased relapse rates, but only in patients living with their families. Further, personal therapy produced improved social and personal adjustments, which were independent of relapse prevention. Compared with supportive therapy that led to peak improvement at 12 months with a plateau thereafter, personal therapy continued to improve functioning in the second and third years (Hogarty et al. 1997b). Following earlier promising reports that cognitive-behavioral therapy may decrease symptoms, lessen exacerbations, and reduce days spent in a hospital, Sensky et al. (2000) showed in a randomized controlled trial that cognitive-behavioral therapy improved positive, negative, and depressive symptoms in patients with psychosis that was resistant to medications, an improvement that was sustained at 9 months. With respect to group therapy of schizophrenia, the American Psychiatric Association (APA) Practice Guidelines concluded that “the evidence for the efficacy of group therapy is not strong, possibly because of the serious methodological flaws in many of the studies” (1997, p. 28). In contrast, having reviewed the results of 10 controlled studies of various family management interventions in schizophrenia, the APA Practice Guidelines conclude that it was helpful and “relapse rates typically have been halved” (p. 27), down to about 15%, with one study even reporting a relapse rate of 5%. MacFarlane et al. (1995) report that multiple-family groups are even more time efficient and effective in preventing relapse. Of all the strategies that can be added to medications, family therapy is distinctly the best at preventing relapse (Thornicroft and Susser 2001) and in decreasing family burden and cost. Gabbard et al. (1997) conclude that “schizophrenia appears to show the most evidence that a psychotherapeutic intervention can have a positive effect on costs. This may be surprising, since common lore holds that psychotherapy is least effective for patients with the most serious psychiatric ❉

44

❉

❉

The Difficult-to-Treat Patient With Schizophrenia

❉

disorders” (p. 153). Social skills training may also reduce relapse rates and symptom severity, especially when combined with medication supplementation at the time of prodromal worsening of psychotic symptoms (American Psychiatric Association 1997; Marder et al. 1996). However, social skills training seems to lose its prophylactic effect at 2-year follow-up.

What Causes Treatment-Refractory Schizophrenia? Causes of Poor Responsiveness to Somatic Treatments Chronic patients identified as neuroleptic nonresponders fall into two main categories: total nonresponders, and partial or suboptimal responders. Often, these subgroups require different management approaches. Before discussing such issues, however, we need to examine the causes for both partial and total lack of drug responsiveness. Inherent Biological Factors (Patient Related) Individuals differ markedly with respect to how completely they absorb and/ or metabolize neuroleptics and other psychotropics. Plasma neuroleptic levels may vary from 10-fold to 100-fold from individual to individual, and these variations are not eliminated by controlling for the dose, timing, or prior exposure to the drug. And whereas plasma neuroleptic levels may not necessarily reflect brain neuroleptic levels (Baldessarini 1985), it is reasonable to assume that patients differ in the latter as well. Osser (1989) has cited several idiosyncratic factors that may interfere with neuroleptic effects in certain individuals (e.g., abnormalities in intestinal permeability have been noted in refractory psychotic patients who lack neuroleptic-related parkinsonian side effects and schizophrenic individuals who show inhibited neuroleptic activity at the level of the dopamine-2 [D2] receptor). Neuroleptic resistance may be associated with early onset of schizophrenia, particularly in women, and Tandon (1997) has suggested that once “neurobiologic deterioration” occurs, “no treatment can reverse it” (p. 17). Whereas this echoes our clinical experience, it remains to be seen if this prognosis applies to the use of newer AAPs as well. Medication-related issues. Baldessarini (1986) has noted two apparent phases of response to neuroleptics: rapid improvement over the first week (acute cases) or month (chronic cases) followed by slower improvement over the course of the next 3–25 weeks. Thus, discontinuing an antipsychot❉

45

❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

ic because a patient “hasn’t gotten any better” after 3 or 4 days is usually counterproductive unless the patient is clearly intolerant of the agent (Pies 1997). Indeed, for some agents, such as clozapine, a full response may not be seen for up to 6 months (Meltzer 1992). In practice, we recommend a 6– 8 week trial but would consider a change in treatment if a schizophrenic patient showed absolutely no improvement after 3 weeks. When clinicians do change after only a few days or weeks, they should understand that they may or may not be dealing with true neuroleptic resistance. Whereas historically 450 mg or more of chlorpromazine or 15 mg and more of haloperidol per day have been regarded as appropriate and/or necessary for the treatment of acute schizophrenia, recent evidence suggests that doses of neuroleptics greater than 40 mg/day of haloperidol equivalents are no more efficacious than doses of 15–20 mg/day (Kane and Marder 1993). Notwithstanding some positron emission tomographic evidence that haloperidol doses of 2–4 mg/day may suffice in the initial treatment of psychotic episodes (Kapur et al. 1996), the authors regard 4–8 mg/day of haloperidol equivalents as a reasonable initial dose (average over first 3 days of treatment) for the majority of new-onset, “early” cases of schizophrenia. We recognize that upward titration to a dose of 10–15 mg/day may be necessary in some acute, and many chronic, cases. A few patients who absorb antipsychotics poorly, or who metabolize them rapidly, may require oral doses in excess of 20 mg/day of haloperidol equivalents (see section “Dosage Increase or Decrease” later in this chapter). Dewan and Koss (1995) have summarized the conflicting data regarding equivalent (or equipotent) doses of various antipsychotic agents. For the treatment-resistant schizophrenic patient, we believe that obtaining a plasma antipsychotic level can often be informative and useful, particularly for agents with putative therapeutic plasma levels (see Table 3–1). A very low or negligible plasma level may point to either noncompliance or pharmacokinetic factors that limit drug bioavailability. The latter may require either intramuscular administration (e.g., to avoid extensive “firstpass” hepatic extraction of the agent) or increased oral dosage in order to achieve adequate blood levels (Table 3–1). Patient compliance factors. Perkins (1999) has pointed out that among patients treated with conventional neuroleptics, approximately 40% stop taking their medication within 1 year, and about 75% stop within 2 years. Patient compliance factors include

❉

46

❉

Table 3–1.

Dosages and therapeutic levels of currently available antipsychotics

47 ❉

Haldol

2–30

4–8

2–12

Moban

20–225

50–75

—

Serlect

12–24

4–8

—

Seroquel

300–500

50–100

—

Loxitane

30–150

50–60

—

2–10

4–6

—

Clozaril

75–800

25

350 (?threshold) [?250–420 for most patients]

Zyprexa

7.5–20

5–7.5

>23ng/mL

Risperdal

1–10

2–4

28–42ng/mLb

Geodon

80–160

80

—

Orap

❉

Thienobenzodiazepine Olanzapine Benzisoxazole Risperidone Benzisothiazolyl piperazine Ziprasidone

Putative therapeutic plasma level (ng/mL)

Brand name

The Difficult-to-Treat Patient With Schizophrenia

❉

Butyrophenones Haloperidol Dihydroindolones Molindone Phenolindole Sertindole Dibenzothiazepine Quetiapine Dibenzoxazepines Loxapine Diphenylbutylpiperidines Pimozide Dibenzodiazepines Clozapine

Standard initiala dose range for young, healthy patient (mg/day)

❉

Class/subclass, generic name

Daily dose range usually reported in literaturea (mg/day)

Table 3–1.

Dosages and therapeutic levels of currently available antipsychotics (continued)

❉

Thorazine Vesprin Sparine

150–1000 20–150 25–1000

300–600 50–100 100–200

30–100

Prolixin, Permitil Trilafon Stelazine Tindal

2–50 8–60 5–60 40–80

8–12 16–20 10–30 50–60

0.2–2 0.8–3 1–2.5

Serentil Mellaril

75–300 100–800

100–200 300–600

Taractan

30–600

200–300

Navane

6–50

12–30

Brand name

Putative therapeutic plasma level (ng/mL)

The Difficult-to-Treat Psychiatric Patient

48

❉

Perphenazine Trifluoperazine Acetophenazine Piperidine Mesoridazine Thioridazine Thioxanthenes Aliphatic Chlorprothixene Piperazine Thiothixene

Standard initiala dose range for young, healthy patient (mg/day)

❉

❉

Class/subclass, generic name Phenothiazines Aliphatic Chlorpromazine Triflupromazine Promazine Piperazine Fluphenazine

Daily dose range usually reported in literaturea (mg/day)

2–15

Table 3–1.

Dosages and therapeutic levels of currently available antipsychotics (continued)

Putative therapeutic plasma level (ng/mL)

❉ 49 ❉

Initial dose range refers to average daily dose during the first 3 days of treatment. The very first dose of an antipsychotic is often 50% of the low end of this range in order to minimize the risk of acute dystonic reactions, hypotension, etc. Initial dose range may vary depending on severity of symptoms, degree of agitation, and risk factors for acute EPS (which are relatively frequent in young males). Initial doses of antipsychotics that induce hypotension (e.g., mesoridazine, chlorpromazine, risperidone, clozapine) may need to be significantly lower than ultimately effective dose. Doses during an acute psychotic episode are generally higher than maintenance doses for many patients and may need to be substantially higher than the initial dose range for some patients. Dosage recommendations vary considerably based on age of patient, diagnosis, and plasma levels attained. Examples: 1) Initial and final doses of risperidone may need to be lower in elderly patients due to risk of hypotension or in those with bipolar disorder (because high risperidone doses may exacerbate manic symptoms). 2) Maintenance clozapine dose may be substantially lower in elderly Parkinson’s patients with psychosis than in young, healthy patients with schizophrenia. 3) For a patient who achieves a plasma haloperidol level of only 1 ng/mL at 10 mg per day, a dosage increase may be necessary. bRisperidone + 9-hydroxy-risperidone. Source. Adapted from Bork et al. 1999; Buckley and Meltzer 1995; Janicak et al. 1997; Jenkins and Hansen 1995; Kaplan et al. 1994; Lindenmayer and Apergi 1996; Perry et al. 2001; Physicians Desk Reference 2001; Pies 1998a; Shader 1994a; Van Putten et al. 1990. See also Dewan and Koss 1995 regarding variance in reported clinical equivalence (potency) of antipsychotic agents.

The Difficult-to-Treat Patient With Schizophrenia

a

Brand name

Standard initiala dose range for young, healthy patient (mg/day)

❉

Class/subclass, generic name

Daily dose range usually reported in literaturea (mg/day)

❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

• Medication side effects • The patient’s perception of the direct and indirect benefits of antipsychotic medication • Worsening of negative features due to standard neuroleptics • Unconscious or idiosyncratic feelings and beliefs about taking medication • Postpsychotic reactions, including depression and so-called awakenings • Various practical barriers to medication use, such as cost or complexity of the regimen (Dewan 1992; Perkins 1999) Medication side effects appear to be a major contributor to poor medication compliance in some (Van Putten 1974). In our experience, excessive sedation, anticholinergic side effects (dry mouth, constipation), akathisia, and weight gain are common causes of noncompliance in psychotic populations. Akathisia is particularly distressing to patients and may elicit both noncompliance and aggressive or suicidal ideation. Despite their overall benefits and favorable side-effect profile, AAPs are also associated with noncompliance, in some cases, because of unpleasant side effects such as weight gain (Perkins 1999). However, Adams and Howe (1993) found that among acutely psychotic inpatients, side effects were less important and “the greater the number of secondary benefits, e.g., keep you out of the hospital, save your marriage, that a neuroleptic medication provides, the more compliant a patient will be” (p. 560). This finding may have important therapeutic implications. Many unconscious factors are involved in medication noncompliance, particularly issues of dependence and control. Thus, the pill may be perceived as something that incorporates the “good,” nurturing doctor and that fills the patient’s profound sense of inner emptiness. Alternatively, as “bad” object, the pill may represent bodily invasion, violation of boundaries, or even dissolution of the self (Goodman 1992). A psychotic patient may have very idiosyncratic associations to the name of a particular antipsychotic, and this may affect compliance. “For instance, a patient may be vehemently opposed to taking Stelazine (“I hate Stella,” said one) but happily agree to taking Mellaril (“It makes me mellow”) (Dewan 1992, p. 104). Finally, Weiden et al. (1996) have characterized “awakening” experiences: an initial dramatic improvement with olanzapine, followed by a severe, postpsychotic depression following the sudden realization—after many years of psychotic denial—that he or she suffers from an extremely debilitating chronic illness. The consequences of such awakenings include suicidal ideation or behavior, ❉

50

❉

❉

The Difficult-to-Treat Patient With Schizophrenia

❉

denial of the need for further antipsychotic medication, or refusal to take olanzapine or other AAPs. Patient comorbidity. Schizophrenia may be “mimicked” by a plethora of neurological and medical conditions (e.g., Wilson’s disease, temporal lobe tumors) that, if undetected, may contribute to poor outcome (Pies 1994). The Expert Consensus Guideline Series for schizophrenia (McEvoy et al. 1999) recommends a drug screen, general chemistry screen, complete blood count, and urinalysis as the “first-line” medical workup of first-episode schizophrenic patients, with other tests (magnetic resonance imaging, electroencephalogram, etc.) indicated in selected cases. Comorbid substance abuse and dependence account for a large proportion of refractory cases and generally predict poor prognosis in schizophrenia (see Chapter 10 of this book). Some 30%–50% of schizophrenic patients have alcohol abuse or dependence, and an additional 20%–30% have cannabis and cocaine abuse or dependency. Prominent obsessive-compulsive symptoms in schizophrenia may worsen overall psychosocial function and prognosis and require specialized pharmacological approaches (Sasson et al. 1997), which are discussed later in this chapter. Similarly, the presence of depressive features often presents diagnostic and treatment dilemmas (see Figure 3–1).

Poor Response to Psychosocial Treatments Whereas some authors believe that insight-oriented psychotherapy may be useful in selected schizophrenic patients (American Psychiatric Association 1997), most of the literature suggests that social-skills training, occupational therapy, and other supportive approaches yield better results with lower risk of relapse. This is not surprising, given the “problems in thinking, poor attention, poor memory, difficulty in concentrating, and distorted or inaccurate perceptions” (Herz 1996, p. 532) often seen in schizophrenic individuals. In our experience, such cognitive deficits and distortions—coupled with the risk of psychotic regression—make an exploratory approach inappropriate for the vast majority of schizophrenic patients, regardless of the phase of their illness. Indeed, overly intrusive exploratory psychotherapy, in our experience, can be a cause of treatment resistance in schizophrenia. The same may be said for day treatment programs, groups, family therapy, and social skills training that overestimate the patient’s ego strength or require excessive affective expression, confrontation, and “probing” (American Psychiatric Association 1997), all of which can lead to adverse reactions. ❉

51

❉

❉

52

❉

The Difficult-to-Treat Psychiatric Patient

❉

❉

❉ The Difficult-to-Treat Patient With Schizophrenia

❉ 53 ❉

❉ Figure 3–1. Algorithm for treatment-refractory patients taking an atypical antipsychotic. aTrying a fourth or fifth atypical agent (e.g., ziprasidone) is reasonable before moving to unconventional treatments. ECT=electroconvulsive therapy; NLP=neuroleptic.

❉

The Difficult-to-Treat Psychiatric Patient

❉

Options for Treatment-Refractory Schizophrenia Pharmacological Options Dosage Increase or Decrease As noted previously, there is little evidence that—for the average patient with schizophrenia—trial on high doses of conventional neuroleptics is a useful strategy (Kane and Marder 1993; Kapur et al. 1996). Indeed, as Osser and Patterson (1996) observe, “greater D2 receptor occupancy will either have no effect, or increase disabling side effects more than it will increase clinical benefits.” As Lieberman (1996) has shown, keeping the patient on a standard neuroleptic dose but extending the duration of treatment (e.g., from 4 to 8 weeks) may be at least as useful as increasing the neuroleptic dose or changing agents. Nevertheless, a few patients with unusual pharmacokinetic or pharmacodynamic responses to neuroleptics may benefit from high-dose strategies (American Psychiatric Association 1997). On the other hand, in our experience, some patients may actually benefit from a reduction of their neuroleptic/antipsychotic dose provided that they are already within the putative therapeutic range (dose/plasma level) of that agent. Lower doses of conventional agents may be associated with improved compliance, better subjective state, and possibly better adjustment in the community (American Psychiatric Association 1997). Of course, these possible benefits must be weighed carefully against the risk of relapse or exacerbation of symptoms. Change to Another Class of Typical Agent The American Psychiatric Association Practice Guidelines (1997) suggest that if a dosage increase is not useful in treatment-refractory schizophrenia (TRS; as outlined previously), “an antipsychotic from a different class should be considered.” However, the guidelines go on to note that “in most cases, patients who have a poor therapeutic response to one conventional antipsychotic medication also respond poorly to others.” This is our general experience, although, of course, occasional patients do seem to benefit from such a switch. Osser and Patterson (1996) suggest the utility of what they call “somewhat atypical neuroleptics” (SANs), which include loxitane, mesoridazine, and pimozide; however, with the availability of the newer AAPs, the use of SANs prior to a clozapine trial becomes harder to justify. Anecdotal data suggest that pimozide may be especially useful in treating the somatic ❉

54

❉

❉

The Difficult-to-Treat Patient With Schizophrenia

❉

type of delusional disorder, which has affinities with both body dysmorphic disorder and so-called monosymptomatic hypochondriasis Depot Neuroleptics Glazer and Kane (1992) have argued that depot neuroleptic therapy represents “a valuable treatment option for many patients and [merits] wider use” (p. 426). These authors advocate trial on depot neuroleptics not only for noncompliant patients, but also for “patients who are refractory to oral agents or who have been hospitalized repeatedly while taking oral neuroleptics” (p. 426). They cite six double-blind, random-assignment, prospective studies, finding that, on average, relapse rates are 15% lower with depot neuroleptics than with oral agents. They also suggest that depot neuroleptic therapy may reduce variability in plasma neuroleptic levels. Although discounting the putative risks of depot neuroleptic therapy (e.g., increased rates of tardive dyskinesia and increased mortality due to neuroleptic malignant syndrome [NMS]), Glazer and Kane (1992) do acknowledge that the slow elimination half-life of depot agents “could in theory increase the risk” of NMS (p. 428). The Expert Consensus Guidelines for the treatment of schizophrenia (McEvoy et al. 1999) support the use of a depot neuroleptic when medication noncompliance interferes with treatment, especially “if the patient has lack of insight into the need for medication” (p. 13). At least with respect to haloperidol decanoate versus oral haloperidol the rates of EPS may actually be lower with the depot preparation than with orally administered neuroleptics. This pattern may be reversed with respect to fluphenazine, although unnecessarily high doses of fluphenazine may have been used in some of the depot neuroleptic studies (Glazer and Kane 1992). We believe depot preparations have an important place for refractory patients who lack any insight into their illness, and/or refuse all antipsychotic medications or comply poorly with oral medication. Of course, appropriate judicial authorization must be obtained for involuntary intramuscular administration. Fortunately, depot preparations of several AAPs are expected soon. Atypical Agents We regard AAPs as the mainstay of treatment for schizophrenic patients refractory to standard neuroleptics as well as those with tardive dyskinesia and related syndromes. The use of AAPs has been thoroughly reviewed (e.g., Marder 1996; Weiden et al. 1996), and we restrict our comments here to a few central points. First, the work of VanderZwaag et al. (1996) strongly suggests the utility of plasma clozapine levels and the existence of a thresh❉

55

❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

old plasma level at around 200–300 ng/mL. The mean dose of clozapine needed to achieve plasma drug levels within this range was just over 370 mg/ day. A few patients may need clozapine doses up to 900 mg/day for optimal response. We regard an adequate trial of clozapine as one achieving blood levels of at least 250 ng/mL for a period of at least 3–4 months. Some would argue that 6 months or more should be permitted to assess clozapine adequately, but we would consider a change in strategy (i.e., either augmentation or change of agents) if a patient has shown minimal response to clozapine after 4 months. Second, if one AAP fails, there is no reason to assume that another AAP will not be effective. With the addition of quetiapine and ziprasidone, our AAP armamentarium is quite extensive. However, it is not clear what advantages these newer AAPs will have over clozapine beyond their improved side-effect profile (e.g., little weight gain with ziprasidone). The possibilities of augmentation within this group of agents (i.e., adding one atypical to another) is discussed in the section “Combining Antipsychotics” later in this chapter. In our experience, the sudden conversion from a standard to an AAP can lead to a “stormy” course or frank decompensation. This can also occur when switching suddenly from one AAP to another, particularly from clozapine to another AAP. We thus advise a gradual conversion to the AAP, usually over a period of 3–6 weeks and often via gradual addition to, followed by tapering of, the first (typical or atypical) agent (Pies 1998a). This is particularly important in the severely ill patients and may not always apply to stable, moderately ill patients. This latter group may often be switched rapidly from a typical or atypical agent to an atypical (Kinon et al. 2000) without serious adverse effects. Augmentation Strategies The Patient Outcomes Research Team Treatment Recommendations (Lehman and Steinwachs 1998) contain a useful summary of adjunctive pharmacotherapy in schizophrenia: “Persons who experience persistent and clinically significant associated symptoms of anxiety, depression, or hostility, despite an adequate reduction in positive symptoms with antipsychotic therapy, should receive a trial of adjunctive pharmacotherapy. A trial of a benzodiazepine or propranolol is merited for persistent anxiety. An antidepressant trial should be considered for persistent depression...[and] lithium, a benzodiazepine, or carbamazepine should be considered for persistent hostility or maniclike symptoms” (p. 6). We now consider these and other adjunctive pharmacotherapies in refractory schizophrenia. ❉

56

❉

❉

The Difficult-to-Treat Patient With Schizophrenia

❉

Antidepressants. The combined use of antidepressants and conventional neuroleptics is comprehensively reviewed by Sussman (1997). He notes that antidepressants have been used to ameliorate persistent negative symptoms, persistent anxiety, and postpsychotic depression. With tricyclic antidepressants, Sussman concludes that “overall, studies...show a trend toward modest improvement in negative symptoms, with little or no exacerbation in psychosis” (p. 24). With respect to selective serotonin reuptake inhibitor (SSRI) augmentation of standard neuroleptics, the results of several controlled trials involving fluvoxamine, fluoxetine, and citalopram have been generally encouraging, particularly with respect to reduction of negative symptoms. However, as Sussman points out, the tendency of some SSRIs to elevate plasma neuroleptic levels cannot be ruled out as a factor in clinical improvement. Moreover, worsening of EPS and (rarely) exacerbation of psychosis have been reported with SSRI-neuroleptic combinations (Sussman 1997). One encouraging finding was the reduction in aggression seen in a study (Vartiainen et al. 1995) in which citalopram was added to neuroleptics in the treatment of chronically violent schizophrenic patients. Only a few studies have looked at the addition of SSRIs to clozapine in clozapine-resistant schizophrenia (Buchanan et al. 1996; Silver et al. 1992). Fluvoxamine (25–100 mg/day) improved negative but not positive symptoms (Silver et al. 1992), whereas fluoxetine augmentation of clozapine produced no improvement in either positive or negative symptoms (Buchanan et al. 1996). We regard the use of antidepressants as a potentially useful augmentation strategy in chronic schizophrenic patients with treatment-resistant negative features or prominent depression, with the caveat that the patient must be closely monitored for adverse pharmacokinetic and pharmacodynamic effects. Finally, ziprasidone may have antidepressant benefits in depressed schizophrenic patients (Labbate and Ayd 2001). Mood stabilizers. Marder (1996) has summarized the studies of lithium and carbamazepine as adjunctive agents for treatment-resistant schizophrenic patients. Both agents seem useful primarily in patients with an affective or “excited” component to their clinical picture, although lithium may produce improvement in core symptoms of schizophrenia even in patients lacking such features. Lerner et al. (1988) compared the addition of lithium or placebo with haloperidol in a population of schizophrenic patients and found that lithium produced significantly greater reductions in Brief Psychiatric Rating Scale (BPRS) depression-anxiety scores in depressed schizophrenic patients. However, in the nondepressed patients, there was no ❉

57

❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

difference in BPRS depression-anxiety ratings between lithium and placebo. Small et al. (1975) found rather broad symptomatic improvement when lithium was compared with placebo as a neuroleptic augmenting agent. Compared with placebo, the addition of lithium was associated with significantly greater improvements in overt psychotic symptoms, excitement, mannerisms/posturing, irritability, cooperation, personal neatness, and social competence (American Psychiatric Association 1997). Toxic/neurological side effects may be seen in anywhere from 5% to 50% of patients treated with lithium-neuroleptic combinations (American Psychiatric Association 1997), an effect that may be mitigated by using both agents in the lower therapeutic dosage ranges and by carefully monitoring the patient’s response. Carbamazepine may be useful in psychotic patients with abnormal electrocardiograms or “impulsive or excited behaviors” (Marder 1996, p. 28); however, one double-blind study of adjunctive carbamazepine (added to an antipsychotic) failed to show statistically significant advantages over placebo on “excited states” in schizophrenic and schizoaffective patients (Okuma et al. 1989). Carbamazepine must be used cautiously in combination with neuroleptics because it may markedly reduce neuroleptic plasma levels. With respect to valproate, there appear to be few double-blind controlled studies of its use as an adjunctive agent in well-defined schizophrenic populations, and the evidence that exists does not show impressive effects (American Psychiatric Association 1997). Nevertheless, the authors have occasionally found adjunctive valproate useful in some schizophrenic or schizoaffective patients with excited, aggressive, or manic features and in some cases of exacerbations of chronic schizophrenia (Wassef et al. 2000). The role of newer anticonvulsant mood-stabilizers such as lamotrigine, gabapentin, and topiramate has not yet been defined in schizophrenia. However, one open-label study found that gabapentin (mean dose=1,400 mg/ day) was effective for hypomania and mania in patients with bipolar and schizoaffective disorder (bipolar type) when added to ongoing neuroleptics. Lamotrigine was found useful in two cases of “epileptic psychosis” and also improved BPRS scores when added to clozapine in chronic schizophrenic patients partially responsive to clozapine (Dursan et al. 1999). Lamotrigine’s ability to inhibit glutamate release might confer antipsychotic properties on it (Anand et al. 2000), but further research is needed to define lamotrigine’s role in refractory schizophrenia. Combining antipsychotics. There are almost no controlled studies of combining two conventional neuroleptics, two AAPs, or a conventional neu❉

58

❉

❉

The Difficult-to-Treat Patient With Schizophrenia

❉

roleptic and an AAP (Pies 2001). Nevertheless, in the latest Expert Consensus Guidelines (McEvoy et al. 1999), combining a conventional neuroleptic or a newer AAP with clozapine are considered second-line options for selected refractory patients. There are numerous uncontrolled reports of combination treatment with antipsychotics with varying degrees of success (e.g., clozapine or risperidone plus a conventional neuroleptic), and clozapine or olanzapine plus the D2 receptor blocker, pimozide (see Pies 2001 for a review). It is too soon to recommend any of these regimens with confidence, although some refractory patients may benefit. It has been our experience that the addition of a low dose of neuroleptic can occasionally enhance the response to one of the AAPs. Our preference is to use 1–2 mg of haloperidol or thiothixene in combination with one of the AAPs (e.g., haloperidol 1 mg qd plus olanzapine 10 mg qd; we favor doses in the lower ranges for each agent on the premise that synergistic effects often occur). Monitoring the patient for new onset of EPS and drug-drug interactions is important (e.g., akathisia, prolactin elevation, or prolonged QT interval; the latter may be especially critical if ziprasidone is used [Labbate and Ayd 2001]). Henderson and Goff (1996) reported that risperidone added to clozapine led to improvement in both positive and negative symptoms, as well as depressive symptoms, in 10 of 12 patients with chronic schizophrenia. This regimen was generally well tolerated, with no increase (as might be expected) in clozapine blood levels or complaints of dizziness. However, 4 patients experienced mild akathisia, and hypersalivation recurred in 5 patients after risperidone was added to clozapine. Of some concern is a case report (Godleski and Sernyak 1996) of agranulocytosis following (in 6 weeks) the addition of risperidone to a stable regimen of clozapine (22-month duration), although a causal role for risperidone was not evident. John et al. (1998) reported two cases in which high doses of olanzapine appeared to be associated with maniclike reactions. However, one of the patients was successfully stabilized with a combination of olanzapine 10 mg qhs and risperidone 3 mg bid. Although one of us (R.P.) has seen some cases in which the olanzapine-risperidone combination has been successful, there are too few studies at this time to recommend this combination with confidence. However, in extremely refractory cases, this combination could be considered, with use of low doses for both agents and careful monitoring for EPS, akathisia, and signs of either sedation or hyperarousal. Attempts at combining newer agents, such as quetiapine, with the “older” AAPs must take into account the potential for pharmacokinetic interactions via the cytochrome ❉

59

❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

P450 system. For example, because clozapine and quetiapine are both metabolized (in part) via the CYP 3A4 system, it is possible that elevated blood levels of one or both agents could be seen when used in combination. Nevertheless, one of us (R.P.) has seen good results in extremely agitated psychotic patients when low doses of quetiapine (25–50 mg bid prn) are combined with ongoing atypical antipsychotic (e.g., risperidone) treatment. A recent case report (Koziupa 1999) found that when quetiapine (500 mg/ day) was added to ongoing olanzapine (15 mg/day) and lithium carbonate (300 mg tid), a patient previously unresponsive to risperidone and clozapine showed modest improvement. Pharmacodynamic interactions—such as increased sedation, cardiac arrhythmia, or hypotension—must also be considered with such combinations (see Pies 2001 for a review). Dopamine agonists/psychostimulants. Diminished prefrontal activity in schizophrenia may be associated with regional deficits in dopaminergic function, with resultant anhedonia, apathy, and restricted affect (Goldberg et al. 1991). Thus, the use of dopamine agonists has some theoretical appeal. In a double-blind, placebo-controlled, crossover study of 21 patients with chronic schizophrenia, Goldberg et al. (1991) found that the addition of a single oral dose of dextroamphetamine (0.25 mg/kg) resulted in clinical improvement in affect, cooperation, and engagement with the environment. No patient clearly worsened, contrary to the hyperdopaminergic model of schizophrenia. However, owing to the unusual design of this study, the authors did not advocate amphetamine as a routine treatment of schizophrenia. The APA Practice Guidelines (1997) do not advocate the use of dopaminergic agents for highly refractory schizophrenic patients with prominent negative features. However, the Expert Consensus Guidelines (McEvoy et al. 1999) list “add a dopaminergic agent to clozapine” as a second-line option for patients with persistent prominent negative symptoms who have not responded to “sequential trials of conventional antipsychotics, one or more of the newer atypical antipsychotics, and clozapine” (p. 39). Finally, the selective MAO-B inhibitor selegiline—which has dopamine-augmenting effects—was modestly effective in one open study of negative symptoms in schizophrenia but not in a recent controlled study (Jungerman et al. 1999). Alprazolam or other benzodiazepines plus neuroleptic. As Janicak et al. (1997) summarize the use of benzodiazepines (BZDs) as either sole or adjunctive agents in psychotic patients, effects range “from deterioration, to no change in most patients, to striking improvement in a rare patient . . .” ❉

60

❉

❉

The Difficult-to-Treat Patient With Schizophrenia

❉

(p. 176). BZDs may ameliorate superimposed anxiety, auditory hallucinations, and perhaps negative symptoms in a few schizophrenic patients; however, sedation, ataxia, cognitive impairment, and behavioral disinhibition may occur (Janicak et al. 1997; Pies 1997). A number of BZDs (lorazepam, clonazepam, diazepam) have proved useful in catatonic patients, including some with catatonic schizophrenia (Martenyi et al. 1989). Generally, these results have been obtained using intramuscular or intravenous BZDs, although some studies point to continued benefit when the patient is maintained on oral BZDs (Martenyi et al. 1989). (Keep in mind that “catatonia” is a symptom, not a diagnosis, and that treatment must be directed at the underlying pathology.) BZDs also may be helpful in managing acute psychotic states and may reduce the need for higher doses of neuroleptics in agitated psychotic patients. One study of alprazolam augmentation of neuroleptics (Douyon et al. 1989) found a 20%–30% mean reduction in positive and negative symptoms in nine schizophrenic patients whose changes were correlated with higher alprazolam levels. Recently, Carpenter et al. (1999) found that diazepam was superior to placebo and comparable to fluphenazine in preventing symptom exacerbation/progression in schizophrenia. The combination of BZDs with clozapine has been associated with respiratory suppression or arrest and is discouraged in the current PORT Treatment Recommendations (Lehman and Steinwachs 1998; Physicians Desk Reference 2001). However, we believe this is a safe combination for most younger, medically stable treatment refractory patients, provided the BZD dose is kept low and the patient’s vital signs are carefully monitored.

Electroconvulsive Therapy The rationale for use of electroconvulsive therapy (ECT) in schizophrenia and delusional states has recently been summarized by Fink (1997). Fink and Sackheim (1996), after reviewing the past 60 years of ECT use in schizophrenia, concluded that “ECT is particularly applicable in patients with first-break episodes, especially those marked by excitement, overactivity, delusions, or delirium; in young patients, to avoid the debilitating effects of chronic illness; and in patients with syndromes characterized by catatonia, positive symptoms of psychosis, or schizoaffective features” (p. 27). Fink also found evidence that the concurrent use of neuroleptic drugs and ECT elicits better clinical results than either treatment alone. There have been noncontrolled reports that augmentation of clozapine with ECT is useful in schizophrenic patients refractory to either treatment alone (Fink 1997). ❉

61

❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

Frankenburg et al. (1993) reported on the safety and efficacy of combined ECT-clozapine treatment in 12 TRS patients; the combined treatment was very effective in 4 of the patients, and none of the patients experienced adverse effects such as spontaneous seizures, elevated blood pressure, tachycardia, or prolongation of seizure time. However, some clinicians have expressed concerns about prolonged seizures with the ECT-clozapine combination and have hypothesized that the combination of ECT and risperidone might be an effective and safer alternative. However, we know of no studies examining this regimen. We have found that the more acute, “excited,” and catatonic the presentation of schizophrenia, the better the response to ECT. Similarly, the APA Practice Guidelines note that “catatonic and treatment-resistant schizophrenic patients may be candidates for ECT, as are patients with severe depression that cannot be managed with medication.... [T]hose who do not respond or are intolerant to clozapine may [also] benefit from a trial of ECT (8–20 sessions), especially when affective symptoms are prominent” (p. 39). Schatzberg et al. (1997) note, “ECT can reverse psychotic excitements and catatonic stupor but has no real value in preventing future episodes of psychosis” (p. 172). Nevertheless, in our view, the possibility of utilizing maintenance ECT in selected cases should not be ruled out. The indications for ECT are discussed further in Chapter 13 of this book.

Miscellaneous Agents Goff et al. (1995b, 1996) have pointed to the possible role of abnormal Nmethyl-D-aspartate (NMDA)–receptor function in schizophrenia, and have studied D-cycloserine—a partial agonist at this receptor—as an augmenting agent. Their initial findings suggested that D-cycloserine (50 mg/day) added to conventional neuroleptics significantly improved negative symptoms (Goff et al. 1995b). Subsequent work (Goff et al. 1996) did not replicate this finding when D-cycloserine was added to clozapine. Recently, Goff et al. (1999a) showed that D-cycloserine (50 mg/day) added to conventional neuroleptics significantly reduced negative symptoms when compared with placebo in schizophrenic patients meeting criteria for “deficit syndrome.” Similarly, Heresco-Levy et al. (1999) found that glycine—an obligatory coagonist at NMDA receptors—was useful in reducing negative symptoms in a group of treatment-resistant schizophrenic patients. There is considerable interest in the investigational agent CX516, which acts selectively at the AMPA type of NMDA receptor (Goff et al. 1999a). ❉

62

❉

❉

The Difficult-to-Treat Patient With Schizophrenia

❉

The 5-hydroxytryptamine (serotonin) type 3 receptor (5-HT3) antagonist, ondansetron, is actively being investigated as an adjunctive treatment in schizophrenia. Briskin and Curtis (1997) found that ondansetron (4 mg bid) augmented the effects of clozapine (750 mg/day). Reserpine has weak antipsychotic effects, but Schatzberg et al. (1997) note that reserpine’s onset of action may be delayed for 2 months and that “schizophrenic patients may pass through a stage of behavioral turbulence before improving” (p. 172). Other nonstandard adjunctive treatments for schizophrenia have been reviewed by Rifkin (1997) (see Table 3–2). Whereas Rifkin viewed the supporting evidence for propranolol as moderately strong, Janicak and colleagues (1997) concluded that “propranolol’s efficacy remains doubtful, but there is sufficient positive evidence to consider it potentially helpful for some patients” (p. 178). Although we do not advocate the use of cycloserine, methadone, famotidine, or ondansetron for most patients with refractory schizophrenia, we do include these as agents worth considering if better validated treatments fail. Finally—although not included in our table—estrogen is the focus of great interest in refractory psychosis. Some data suggest that high levels of estrogen protect against symptom exacerbation in women with schizophrenia, and preliminary data in psychotic women suggest at least a transient augmenting effect when estradiol is added to ongoing neuroleptics (Kulkarni et al. 1996).

Optimizing Psychosocial Interventions The schizophrenia PORT survey (Lehman and Steinwachs 1998) found that the majority of patients did not receive appropriate individual, family, or rehabilitation therapies, which have been shown to improve outcome. Individual Therapy We believe that many treatment-resistant, chronically relapsing schizophrenic patients have been exposed to inappropriate forms of individual therapy or to therapists who have misapplied general psychoanalytic principles to the unique needs of the schizophrenic patient. Weiden and Havens (1994) and Weiden (1996a, 1996b) have cogently described the psychotherapeutic management of schizophrenia in both acutely psychotic individuals and more stable outpatients. Based on their and our own experience, we have summarized some of these pitfalls and corresponding management strategies in Table 3–3. As Weiden and Havens (1994) have noted, “specific and sometimes counterintuitive techniques are needed to work successfully ❉

63

❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

Table 3–2. Nonstandard adjunctive treatments for schizophrenia

Proposed agent

Usual dose range (added to antipsychotic)

Strength of evidence

High-dose propranolol

Up to 1200 mg/day

Moderate

Electroconvulsive therapy

Up to 12 treatments

Moderate

Lithium

Dose to obtain level of 0.6–1.0 mEq/L

Moderate

Benzodiazepines

Up to 300 mg/day chlordiazepoxide

Modest

Methadone

Up to 100 mg/day

Modest

Carbamazepine

Dose to obtain level of 6–12 mcg/mL

Weak

Famotidine

40 mg/day

Weak

Ondansetron

4 mg bid

Weak

Cycloserine (added to typical antipsychotic) 50 mg/day Source.

Modest

Adapted from Rifkin 1997.

with patients with schizophrenia” (p. 549). Hogarty et al. (1997a, 1997b) have shown success with disorder and person-specific “personal therapy,” which is being developed into a more sophiscticated version called “cognitive enhancement therapy” (Thornicroft and Susser 2001). Cognitivebehavioral therapy geared toward the treatment of patients with medicationresistant psychotic symptoms was eight times more likely to reduce symptoms, prevent exacerbations, and decrease days in hospital (Sensky et al. 2000). Goodman (1992) has cogently explained how psychodynamic factors may be an important source of medication noncompliance. Huszonek (1987) has suggested that working with schizophrenic patients entails first recognizing the patient’s “need-fear dilemma” and second providing the patient with a “real” relationship that permits a more complete sense of “self and other” to develop. Group Therapy We generally agree with the seven basic goals of group therapy for schizophrenia, outlined by Mosher (1982) (i.e., focusing on management of reallife problems, sharing relevant experiences, listening to the needs and feelings of others, offering new perspectives, asking questions, expressing opin❉

64

❉

Table 3–3.

Countertherapeutic factors in individual therapy with schizophrenic patients

Overly “close” physical and emotional interaction leads to overstimulation; “probing” for unconscious material in an actively psychotic patient.

Pay careful attention to “body language” (e.g., maintain appropriate physical boundaries, avoid excessive eye contact). Avoid exploration of putative unconscious sexual issues or inappropriate self-disclosure of highly personal material.

❉ 65

Overly challenging or “head-on” approach to patient’s paranoid or Use a quasiparadoxical approach, such as “sharing mistrust” (e.g., other delusional beliefs; attempt to “persuade” the patient to be saying to a paranoid patient, “Well, the world can feel pretty threatmore trusting or “reality centered.” Premature “psychoeducation,” ening at times.”). [NB: The therapist should not veer into the other use of the “medical model,” etc., while patient is floridly extreme of colluding with the patient’s delusional system]. psychotic.

❉

Overzealous attack on patient’s use of denial (e.g., when denial occurs during “well” period, and patient is compliant with all treatment modalities).

If patient is otherwise doing well (e.g., is free of florid positive symptoms of psychosis, complying with treatment), it may be prudent to ignore a statement such as “I’m not gonna need this medication forever, you know!”

Excessive “pathologizing” of symptom that may not necessarily be It may be more useful to “normalize” the patient’s experience (e.g., psychotic (e.g., “I feel like people look at me funny when I go saying “You know, we all get a little self-conscious in public now and to the store.”). then.”). Some extremely aggressive, agitated patients will require involuntary administration of antipsychotics. However, long-term medication compliance in less emergent cases requires rapport and trust between patient and therapist.

❉

With medication noncompliant patients, insistently “pushing” medical treatment before establishing even a rudimentary degree of rapport or trust with the patient.

The Difficult-to-Treat Patient With Schizophrenia

Management

❉

Countertherapeutic factor

Table 3–3.

Countertherapeutic factors in individual therapy with schizophrenic patients (continued)

Insisting that the patient see you only at a specific time for the usual 20- to 50-minute session can be impractical and countertherapeutic. Consider brief, informal meetings in an unconventional but “safe” setting, such as a unit lounge.

Directly confronting the patient’s delusional ideation without first attempting to empathize with the underlying affect.

Identify and resonate with the affect underlying the delusion (e.g., for a paranoid patient who is convinced “the CIA is out to get me,” one might say, “It sounds like your life has been hell for you...always having to be on your guard.”). Never collude with delusional ideation.

66

Source.

Adapted from Weiden and Havens 1994; Weiden 1996a, 1996b; Huszonek 1987.

❉

The Difficult-to-Treat Psychiatric Patient

Insisting on the standard “rules” of psychotherapy.

❉

Management

❉

Countertherapeutic factor

❉

❉

The Difficult-to-Treat Patient With Schizophrenia

❉

ions without fear of criticism, and keeping a topic in focus until a conclusion is reached). We believe clinicians should have a low threshold for removing a disturbed schizophrenic patient from a stressful group session. We also believe that group (and individual) therapy aimed at strengthening cognitive skills and problem-solving strategies may help reduce treatment resistance and relapse rates (Mosher 1982; Pallanti et al. 1997). In one such approach—integrated psychological therapy—specific curricula are designed to correct corresponding cognitive and social deficits in the patient (Brenner et al. 1992). Other Adjunctive Programs Programs for Assertive Community Treatment (PACTs) are important but infrequently available resources that are effective in helping the severely ill and even the homeless. PACTs are comprehensive, integrated programs that function 24 hours a day year-round and aggressively support the patient in the community. They are particularly helpful for the treatment-resistant and noncompliant patient and have been shown to decrease the length of readmissions and to improve living conditions (American Psychiatric Association 1997; Lehman and Steinwachs 1998). Vocational rehabilitation is another scarce resource that is often integral to recovery. Aquila et al. (1995) suggest that a “rehabilitation alliance” is essential. After basic trust has been painstakingly established, patients are encouraged to see themselves as capable of holding a real job and to think of how their symptoms may interfere with this project, thereby also increasing compliance. Transitional employment and job banks, as put into practice by agencies such as Fountain House in New York City, report success rates of 5%–40%, which is laudable given the severity of illness in their population. Perhaps 10% of difficult-to-treat schizophrenic patients need long-term hospitalization. Studies show that they are most effectively treated in programs that “emphasize highly structured behavioral techniques, including a token economy, point systems, and skills training. Paradoxically, despite its demonstrated efficacy, the token economy is not often used in clinical settings” (American Psychiatric Association 1997). In summary, after optimizing medications, noncompliance can be addressed by adding supportive therapy and treating residual symptoms with cognitive-behavioral therapy. Patients living with their family benefit from family psychoeducational programs; those living alone who are difficult to treat require the addition of PACT. Cognitive deficits may respond to cognitive rehabilitation strategies, and social skills training can address deficits in ❉

67

❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

everyday social functioning. Vocational rehabilitation helps provide hope and jobs. Bustillo and colleagues (2001) provide an excellent review of psychosocial treatment of schizophrenia.

Special Cases Several “special cases” are worth brief mention (see Pies 1998a and Stowe et al. 1998 for reviews). The pregnant or lactating patient with schizophrenia poses many medicolegal challenges, including the risk of teratogenesis, neonatal toxicity, and neonatal effects caused by antipsychotics in breast milk. And yet, as Stowe and colleagues (1998) conclude, “the paucity of data linking these [antipsychotic] agents to either congenital or neurobehavioral deficits suggests that the risk of these medications is minimal” (p. 987). Nevertheless, use during the first trimester is best avoided, if possible, with ECT providing a good alternative for some patients (see Chapter 13). If an antipsychotic must be used in the first trimester, piperazine phenothiazines may have less teratogenic potential than alternatives (Stowe et al. 1998), although not all data support this claim. Because antipsychotics are variably excreted in breast milk, breast-feeding is best avoided if the mother remains on the medication. The teratogenic risk associated with AAPs is not wellresearched in humans. One report (Stoner et al. 1997) on two psychotic women treated with clozapine during pregnancy noted essentially normal births; however, one child had a seizure 8 days after delivery. These authors made the important point that “clinical judgment determined that failure to treat the patients’ symptoms of schizophrenia presented a greater risk of harm to ‘self and fetus’ than maintaining treatment with clozapine . . .” (p. 364)—a position we find defensible in many similar cases. The noncompliant outpatient poses a particular challenge, because even a court-ordered “Rogers” guardianship (permitting involuntary administration of depot neuroleptics) may be difficult to carry out (e.g., when the patient refuses to come to appointments). Assertive community treatment programs may be useful in such cases, often by providing case management and on-site medication to homeless patients and those in community residences. Some states also permit so-called outpatient civil commitment, which may allow authorities to bring in recalcitrant patients for administration of depot neuroleptics and other urgent care under certain life-threatening circumstances. We have sometimes been surprised to find that simply switching a noncompliant patient to clozapine—with its mandatory weekly or biweekly blood drawings—seems to improve compliance beyond the ❉

68

❉

❉

The Difficult-to-Treat Patient With Schizophrenia

❉

drug’s antipsychotic effects. Although this “switch” must sometimes be initiated as part of an involuntary inpatient admission, it is our impression that some previously noncompliant patients subsequently become acculturated to the outpatient mental health system—perhaps because they appreciate the structure and nurturing of regular biweekly appointments.

Algorithms for Treatment-Refractory Schizophrenia In recent years, many treatment algorithms for TRS have been promulgated; each of them has advantages and disadvantages, and—to our knowledge— none has been proved superior to competing algorithms in prospective controlled trials (Janicak and Sadek 1996; Miller et al. 1999; Osser and Patterson 1996). We offer the following algorithms with the caveat that they are, at best, provisional guidelines. A nonresponder is defined as a patient who shows virtually no improvement after adequate dose-time (DT) treatment; a partial responder is one who shows at least a 25% improvement, as determined either by standardized rating scales, the clinician’s judgment, or both. Our algorithms make early use of plasma antipsychotic levels, even while recognizing that the putative therapeutic ranges for these levels are far from certain (see Table 3–1); however, we believe that plasma levels are useful in 1) pointing out extreme “outliers” from the expected values, and 2) serving as a useful guide to therapeutic levels of haloperidol, clozapine, and the other agents shown in Table 3–1. Unlike most algorithms for use of antipsychotic agents, our algorithm includes assessment of psychosocial therapies and comorbid diseases or major symptom clusters. We operationally define major symptom clusters as including prominent catatonic, affective (manic or depressed), and obsessive-compulsive features. We use the short-hand acronym CAOCD for such prominent symptoms that lie outside the criteria for schizophrenia and its subtypes. Adjunctive treatments for CAOCD syndromes include 1) BZDs for catatonic features, 2) lithium or an anticonvulsants for “excited” and manic features, and 3) an SSRI or clomipramine for obsessive-compulsive features. The hierarchy of decisions in these algorithms reflect primarily our clinical confidence in the strategies used (e.g., in the neuroleptic-refractory patient, we have greater confidence in going to an AAP early in treatment than in trying various augmenting strategies, absent a clear CAOCD syndrome [see Figure 3–2]). In our algorithm for patients refractory to an AAP (see Figure 3–1), we emphasize augmentation strategies for partial responders, and switching to another AAP for complete nonresponders. Like the Texas ❉

69

❉

❉

70

❉

The Difficult-to-Treat Psychiatric Patient

❉

❉

❉

71 ❉

The Difficult-to-Treat Patient With Schizophrenia

❉

❉

Figure 3–2. Algorithm for treatment-refractory patients taking a standard neuroleptic. aArguably, any nonresponder to a standard NLP should simply be started on an atypical agent; however, doing so hastily may undermine efforts to detect factors leading to poor response. We do regard atypicals as the drugs of choice for the vast majority of schizophrenic patients. bIt is very rare that a change to depot NLP would result in inadequate blood levels. NLP=neuroleptic.

❉

The Difficult-to-Treat Psychiatric Patient

❉

algorithm (Miller et al. 1999), we generally do not recommend combining antipsychotics (typical plus atypical, or two AAPs) until clozapine has been given an adequate trial. Given the availability of five AAPs, this algorithm clearly requires a long period of drug trials, extending over a year or more in some cases. The addition of ECT to an AAP is also worth considering, although, to our knowledge, published studies are limited to clozapine. Finally, we would consider the use of “unconventional” approaches with a very limited data base to support them, such as reserpine, ondansetron, and cycloserine. Clinical Vignette Mr. H is a 37-year-old unmarried male diagnosed with “chronic undifferentiated schizophrenia.” Mr. H has had more than 40 hospitalizations over the past 20 years and has been treated with most of the commonly used neuroleptics. Mr. H had a normal pre- and perinatal history and, for the most part, reached developmental milestones normally. He had few friends as an adolescent and had difficulty maintaining passing grades in junior high and high school. At age 17, Mr. H had his first psychotic episode, leading to a prolonged hospitalization. At that time he was treated with chlorpromazine and released after 1 month. He was eventually tapered off the chlorpromazine and followed as an outpatient. However, he relapsed within 6 months, required rehospitalization, and was maintained on various standard neuroleptics over the ensuing 15 years. These neuroleptics included thiothixene, haloperidol, fluphenazine (oral and depot preparations), and molindone. Mr. H was only partially responsive to these medications and displayed prominent negative features of schizophrenia, including social withdrawal, alogia, and flat, apathetic affect. He was also quite sensitive to the parkinsonian side effects of these agents and frequently complained of “ants in my pants,” which was diagnosed as akathisia. In 1987, a haloperidol level was obtained at 8 ng/mL, which was interpreted as “adequate.” Mr. H had been an outpatient at the same mental health clinic for most of the past 20 years; however, he had been treated by at least seven different psychiatrists. In the 1970s and early 1980s, Mr. H had attended a group described as “supportive” but that frequently involved very intense “uncovering work.” Mr. H found this very stressful and usually walked out of the meetings. The first family intervention—in the late 1970s—informed Mr. H’s family that schizophrenia was often caused by “double binds” and “mixed messages” transmitted to the patient by parents. One of the therapists suggested psychotherapy for Mr. H’s mother, since she seemed “schizophrenogenic.” This approach only alienated Mr. H’s family, who refused to have anything more to do with the mental health center. In 1990, during an inpatient admission, Mr. H was started on cloza-

❉

72

❉

❉

The Difficult-to-Treat Patient With Schizophrenia

❉

pine. At doses of 250 mg/day, Mr. H showed almost no improvement after 4 weeks. He continued to experience both positive symptoms (auditory hallucinations, delusions that “the FBI is bugging my room,” and thought process disorder) and negative symptoms (apathy, social withdrawal, alogia). One of the nurses on the unit also noted that Mr. H had “peculiar rituals,” in which he had to count each item of food on his plate 13 times, as well as obsessive preoccupations of a nonpsychotic nature (e.g., he needed to check his hair 10–20 times per day “to see if I got it parted the right way”). At this point, a serum clozapine level was obtained and came back at 185 ng/mL. Mr. H’s clozapine dose was increased to a total of 500 mg/day, and a blood level came back at 350 ng/mL (within the therapeutic range). Mr. H began to show marked improvement in his negative and positive symptoms. He tolerated the clozapine reasonably well, although he did have mild drooling and some complaints of feeling overly sedated. The clozapine dose was reduced to 400 mg/day, which yielded a plasma level of 280 ng/mL and reduced side effects. Mr. H’s “obsessive” symptoms persisted, however, and his psychiatrist added sertraline 50 mg/day. His obsessional preoccupations and rituals diminished after 5 weeks of sertraline treatment. Unfortunately, Mr. H began to complain to this psychiatrist—a male resident—that “You’ve taken away my fun.” After very gentle exploration of this, his psychiatrist learned that Mr. H was no longer able to masturbate, owing to decreased libido and erectile dysfunction. This was attributed to the combined effects of sertraline and clozapine. The sertraline dose was reduced to 25 mg/day, with improvement in Mr. H’s sexual function and only modest loss of anti-obsessional effect. Mr. H’s family was invited to meet with his psychiatrist and psychiatric social worker and agreed to attend three “family educational” sessions prior to his discharge. These sessions emphasized that schizophrenia is primarily a biological illness but that families can help their afflicted relatives by recognizing early signs of relapse, by helping the patient maintain a stable medication regimen, and by reducing undue emotional stress in the patient’s life. The family was given some reading material on schizophrenia, and commented, “This is the first time anyone really explained this thing to us.” Meanwhile, Mr. H entered a new day treatment program that emphasized social skills and vocational rehabilitation. Mr. H began to develop an interest in computer programming and was able to learn some rudimentary skills in this area. He continued on his regimen of clozapine and sertraline for the subsequent 6 years. During that time, Mr. H required only one hospitalization.

Clinical Wisdom and Summary We believe that the successful treatment of schizophrenia rests on two key elements: the critical role of the psychiatrist in coordinating somatic and psychosocial treatments and the avoidance of either well-intentioned under❉

73

❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

treatment or haphazard overtreatment. Because of “cost-cutting” strategies associated with third-party payers and the diffusion of authority endemic in many mental health systems, patients with schizophrenia are often deprived of the most effective treatment modalities (e.g., the early use of an AAP or the provision of family therapy). Too few patients, in our experience, are given the benefits of combined therapeutic modalities, such as the provision of family or vocational therapy in concert with appropriate medication. Psychiatrists must be able to utilize antipsychotics and adjunctive agents in creative but well-reasoned strategies, avoiding irrational polypharmacy while always considering promising combination treatments. The use of depot antipsychotics and outpatient commitment may be necessary for some severely ill patients who simply refuse, or are unable, to comply with treatment. In extremely refractory cases, investigational agents should not be ruled out. Finally, we should always keep in mind—however clichéd the sentiment may seem in this age of “managed care”—that patients with schizophrenia are fundamentally suffering persons in need of a human and humanizing therapeutic relationship.

References Adams SG, Howe JT: Predicting medication compliance in a psychotic population. J Nerv Ment Dis 181:558–560, 1993 American Psychiatric Association: Practice Guidelines for the Treatment of Patients with Schizophrenia. Am J Psychiaty 154 (April suppl):1–63, 1997 Anand A, Charney DS, Oren DA: Attenuation of the neuropsychiatric effects of ketamine with lamotrigine. Arch Gen Psychiatry 57:270–276, 2000 Aquila R, Weiden PJ, Emanuel M: Compliance and the rehabilitation alliance. J Clin Psychiatry 60 (suppl 19):23–27, 1999 Baldessarini RJ: Chemotherapy in Psychiatry. Cambridge, MA, Harvard University Press, 1985 Berman I, Klegon D, Fiedosewicz H, et al: The effect of novel antipsychotics on cognitive function. Psychiatric Annals 29:643–646, 1999 Bork JA, Rogers T, Wedlund PJ, et al: A pilot study on risperidone metabolism: the role of cytochromes P450 2D6 and 3A. J Clin Psychiatry 60:469–476, 1999 Breier A, Wolkowitz OM, Doran AR, et al: Neuroleptic responsivity of negative and positive symptoms in schizophrenia. Am J Psychiatry 144:1549–1555, 1987 Breier AF, Malhotra AK, Su TP, et al: Clozapine and risperidone in chronic schizophrenia. Am J Psychiatry 156:294–298, 1999 Brenner HD, Hodel B, Roder V, et al: Treatment of cognitive dysfunctions and behavioral deficits in schizophrenia. Schizophr Bull 18:21–26, 1992

❉

74

❉

❉

The Difficult-to-Treat Patient With Schizophrenia

❉

Briskin JK, Curtis JL: Augmentation of clozapine therapy with ondansetron (letter). Am J Psychiatry 154:1171, 1997 Buchanan RW, Kirkpatrick B, Bryant N, et al: Fluoxetine augmentation of clozapine treatment in patients with schizophrenia. Am J Psychiatry 153:1625–1627, 1996 Buckman RW, Malan RD: Clozapine for refractory schizophrenia: the Illinois experience. J Clin Psychiatry 60 (suppl 1):18–22, 1999 Bustillo J, Lauriello J, Horan W, et al: The psychosocial treatment of schizophrenia: an update. Am J Psychiatry 158:163–175, 2001 Carpenter WR Jr, Buchanan RW, Kirkpatrick B, et al: Diazepam treatment of early signs of exacerbation in schizophrenia. Am J Psychiatry 156:299–303, 1999 Chakos M, Lieberman J, Hoffman E, et al: Effectiveness of second-generation antipsychotics in patients with treatment-resistant schizophrenia: a review and metaanalysis of randomized trials. Am J Psychiatry 158:518–526, 2001 Dewan MJ: Adding medications to ongoing psychotherapy: indications and pitfalls. Am J Psychotherapy 46:102–110, 1992 Dewan MJ, Koss M: The clinical impact of reported variance in potency of antipsychotic agents. Acta Psychiatr Scand 91:229–232, 1995 Douyon R, Angrist B, Peselow E, et al: Neuroleptic augmentation with alprazolam: clinical effects and pharmacokinetic correlates. Am J Psychiatry 146:231–234, 1989 Dursan SM, McIntosh D, Milliken H: Clozapine plus lamotrigine in treatment-resistant schizophrenia (letter). Arch Gen Psychiatry 56:950, 1999 Fink M: ECT update—1997. Psychiatric Times, April 1997, pp 39–41 Fink M, Sackheim HA: Convulsive therapy in schizophrenia? Schizophr Bull 22:27– 39, 1996 Frankenburg FR, Suppes T, McLean PE: Combined clozapine and electroconvulsive therapy. Convulsive Therapy 9:176–180, 1993 Gabbard GO, Lazar SG, Hornberger J, et al: The economic impact of psychotherapy: a review. Am J Psychiatry 154:147–155, 1997 Glazer WM, Kane JM: Depot neuroleptic therapy: an underutilized treatment option. J Clin Psychiatry 53:426–433, 1992 Godleski LS, Sernyak MJ: Agranulocytosis after addition of risperidone to clozapine treatment. Am J Psychiatry 153:735–736, 1996 Goff DC, Tsai G, Manoach DS, et al: A dose-finding trial of D-cycloserine added to neuroleptic for negative symtoms in schizophrenia. Am J Psychiatry 152:1213– 1215, 1995a Goff DC, Midha KK, Sarid-Segal O, et al: A placebo-controlled trial of fluoxetine added to neuroleptic in patients with schizophrenia. Psychopharmacol (Berl) 117:417– 423, 1995b Goff DC, Tsai G, Manoach DS, et al: D-cycloserine added to clozapine for patients with schizophrenia. Am J Psychiatry 153:1628–1630, 1996

❉

75

❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

Goff DC, Tsai G, Levitt J, et al: A placebo-controlled trial of D-cycloserine added to conventional neuroleptics in patients with schizophrenia. Arch Gen Psychiatry 56:21–27, 1999a Goff DC, Bagnall AL, Perlis RH: Glutamergic augmentation strategies for cognitive impairment in schizophrenia. Psychiatric Annals 29:649–654, 1999b Goldberg TE, Bigelow LB, Weinberger DR, et al: Cognitive and behavioral effects of the co-adminstration of dextroamphetamine and haloperidol in schizophrenia. Am J Psychiatry 148:78–84, 1991 Goodman A: Medication noncompliance and the psychodynamics of pharmacotherapy. Integrative Psychiatry 8:181–190, 1992 Henderson DC, Goff DC: Risperidone as an adjunct to clozapine therapy in chronic schizophrenics. J Clin Psychiatry 57:395–397, 1996 Heresco-Levy U, Javitt DC, Ermilov M, et al: Efficacy of high-dose glycine in the treatment of enduring negative symptoms of schizophrenia. Arch Gen Psychiatry 56:29–36, 1999 Herz MI: Psychosocial treatment. Psychiatric Annals 26:531–535, 1996 Ho B-C, Miller D, Nopoulos P, et al: A comparative effectiveness study of risperidone and olanzapine in the treatment of schizophrenia. J Clin Psychiatry 60:658–663, 1999 Hogarty GE, Kornblith SJ, Ulrich RF, et al: Three-year trials of personal therapy among schizophrenic patients living with or independent of family, I: description of study and effects on relapse rates. Am J Psychiatry 154:1504–1513, 1997a Hogarty GE, Kornblith SJ, Ulrich RF, et al: Three-year trials of personal therapy among schizophrenic patients living with or independent of family, II: effects on adjustment of patients. Am J Psychiatry 154:1514–1524, 1997b Huszonek JJ: Establishing therapeutic contact with schizophrenics: a supervisory approach. Am J Psychotherapy 41:185–193, 1987 Janicak PG, Sadek HS: Psychopharmacotherapy for acute and recurrent psychotic disorders. Psychiatric Annals 26:68–77, 1996 Janicak PG, Davis JM, Preskorn SH, et al: Principles and Practice of Psychopharmacotherapy, 2nd Edition. Baltimore, MD, Williams & Wilkins, 1997, pp 97–188 John V, Rapp M, Pies R: Aggression, agitation, and mania with olanzapine (letter). Can J Psychiatry 43:1054, 1998 Jungerman T, Rabinowitz D, Klein E: Deprenyl augmentation for treating negative symptoms of schizophrenia: a double-blind, controlled study. J Clin Psychopharmacol 19:522–525, 1999 Kane J, Marder SR: Psychopharmacologic treatment of schizophrenia. Schizophr Bull 19:287–302, 1993 Kapur S, Remington G, Jones C, et al: High doses of dopamine D2 receptor occupancy with low-dose heloperidol treatment: a PET study. Am J Psychiatry 153:948– 950, 1996

❉

76

❉

❉

The Difficult-to-Treat Patient With Schizophrenia

❉

Karon BP, VandenBos GR: The consequences of psychotherapy for schizophrenic patients. Psychotherapy: Theory, Research, and Practice 9:111–119, 1972 Kinon BJ, Basson BR, Gilmore JA, et al: Strategies for switching from conventional antipsychotic drugs or risperidone to olanzapine. J Clin Psychiatry 61:833–840, 2000 Koziupa DM: The efficacy of quetiapine in relieving persistent symptoms in a patient with poor response to both conventional and atypical antipsychotic therapy (abstract). J Clin Psychiatry 60 (suppl 23):13, 1999 Kulkarni J, de Castella A, Smith D, et al: A clinical trial of the effects of estrogen in acutely psychotic women. Schizophr Res 20:247–252, 1996 Labbate LA, Ayd FJ Jr: Ziprasidone: the latest FDA-approved atypical antipsychotic. International Drug Therapy Newsletter, April 2001, pp 25–30 Lehman AF, Steinwachs: At issue: translating research into practice: the schizophrenia patient outcomes research team (PORT) treatment recommendations. Schizophr Bull 24:1–10, 1998 Lerner Y, Mintzer Y, Schestatzky M: Lithium combined with haloperidol in schizophrenia patients. Br J Psychiatry 153:359–362, 1988 Lieberman JA: Pharmacotherapy for patients with first-episode, acute, and refractory schizophrenia. Psychiatric Annals 26:515–518, 1996 Marder SR, Wirshing WC, Mintz J, et al: Two year outcome of social skills training and group psychoterhapy for outpatients with schizophrenia. Am J Psychiatry 153:1585–1592, 1996 Martenyi F, Harangozo J, Laszlo M: Clonazepam for the treatment of catatonic schizophrenia. (letter) Am J Psychiatry 146:1230, 1989 McEvoy JP, Scheifler PL, Frances A (eds): The Expert Consensus Guideline Series: Treatment of Schizophrenia 1999. J Clin Psychiatry 60 (suppl 11):8–80, 1999 McFarlane WR, Lukens E, Link B, et al: Multiple-family groups and psychoeducation in the treatment of schizophrenia. Arch Gen Psychiatry 52:679–687, 1995 McGlashen TH: Schizophrenia: psychosocial treatments, in Psychiatry Update: The American Psychiatric Association Annual Review, Vol 5. Edited by Frances AJ, Hales RE. Washington, DC, American Psychiatric Press, 1986, pp 96–111 Meltzer HY: Treatment of the neuroleptic-nonresponsive schizophrenic patient. Schizophr Bull 18:515–542, 1992 Miller AL, Chiles JA, Chiles JK, et al: The Texas medication algorithm project (TMAP). J Clin Psychiatry 60:649–657, 1999 Mosher LR: A psychosocial approach to returning schizophrenia. The Schizophrenic Outpatient 1:1–11, 1982 Noordsy DL, Torrey WC, Mead S, et al: Recovery-oriented psychopharmacology: redefining the goals of antipsychotic treatment. J Clin Psychiatry 61 (suppl 3):22– 29, 2000

❉

77

❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

Okuma T, Yamashita I, Takahashi R, et al: A double-blind study of adjunctive carbamazepine versus placebo on excited states of schizophrenic and schizoaffective disorders. Acta Psychiatr Scand 80:250–259, 1989 Osser DN: A systematic approach to pharmacotherapy in patients with neurolepticresistant psychoses. Hospital and Community Psychiatry 40:921–926, 1989 Osser DN, Patterson RD: Pharmacotherapy of schizophrenia, in Handbook for the Treatment of the Seriously Mentally Ill. Edited by Soreff SM. Seattle, WA, Hogrefe & Huber, 1996, pp 91–155 Pallanti S, Quercioli L, Pazzagli A: Relapse in young paranoid schizophrenic patients: a prospective study of stressful life events, P300 measures, and coping. Am J Psychiatry 154:792–798, 1997 Perkins DO: Adherence to antipsychotic medications. J Clin Psychiatry 60 (suppl 21):25–30, 1999 Perry PJ, Lund BC, Sanger T, et al: Olanzapine plasma concentrations and clinical response: acute phase results of the North American Olanzapine Trial. J Clin Psychopharmacol 21:14–20, 2001 Physician’s Desk Reference, 55th Edition. Montvale, NJ, Medical Economics, 2001 Pies R: Clinical Manual of Psychiatric Diagnosis and Treatment. Washington, DC, American Psychiatric Press, 1994 Pies R: Time and the art of psychopharmacology. Harvard Rev Psychiatry 5:1–4, 1997 Pies R: Handbook of Essential Psychopharmacology. Washington, DC, American Psychiatric Press, 1998a Pies R: Converting from one antipsychotic to another. Psychiatric Times November, 1998b, pp 36–38 Pies R: Combining antipsychotics: risks and benefits. International Drug Therapy Newsletter, April 2001, pp 9–13 Rifkin A: Experimental therapeutics: nonstandard drug treatments, in Psychiatry. Edited by Tasman A, Kay J, Leiberman JA. New York, Guilford, 1997, pp 1702–1716 Sanders RD, Mossman D: An open trial of olanzapine in patients with treatmentrefractory psychoses. J Clin Psychopharmacol 19:62–66, 1999 Sasson Y, Bermanzohn PC, Zohar J: Treatment of obsessive-compulsive syndromes in schizophrenia. CNS Spectrums 2:34–45, 1997 Schatzberg AF, Cole JO, DeBattista C: Manual of Clinical Psychopharmacology. Washington, DC, American Psychiatric Press, 1997, p 172 Sensky T, Turkington D, Kingdon D, et al: A randomized controlled trial of cognitive behavioral therapy for persistent symptoms in schizophrenia resistant to medication. Arch Gen Psychiatry 57:165–172, 2000 Silver H, Nasser A: Fluvoxamine improves negative symptoms in treated chronic schizophrenia: an add-on double-blind, placebo-controlled study. Biol Psychiatry 31:698–704, 1992

❉

78

❉

❉

The Difficult-to-Treat Patient With Schizophrenia

❉

Small JG, Kellams JJ, Milstein V, et al: A placebo-controlled study of lithium combined with neuroleptics in chronic schizophrenic patients. Am J Psychiatry 132:1315– 1317, 1975 Stanton AH, Gunderson JG, Knapp PH, et al: Effects of psychotherapy in schizophrenia, I: design and implementation of a controlled study. Schizophr Bull 10:520– 563, 1984 Stoner SC, Sommi RW Jr, Marken PA, et al: Clozapine use in two full-term pregnancies. J Clin Psychiatry 58:364–365, 1997 Stowe ZN, Strader JR, Nemeroff CB: Psychopharmacology during pregnancy and lactation, in The American Psychiatric Press Textbook of Psychopharmacology, 2nd Edition. Edited by Schatzberg AF, Nemeroff CB. Washington, DC, 1998, pp 979– 996 Sussman N: Augmentation of antipsychotic drugs with selective serotonin reuptake inhibitors. Primary Psychiatry 4:24–31, 1997 Tandon R: Rebuttal to the article by Dr. Keshavan. The Journal of Psychotic Disorders: Reviews and Commentaries 1:17–18, 1997 Thornicroft G, Susser E: Evidence-based psychotherapeutic interventions in the community care of schizophrenia. Br J Psychiatry 178:2–4, 2001 Tran PV, Hamilton SH, Kuntz AJ, et al: Double-blind comparison of olanzapine versus risperidone in the treatment of schizophrenia and other psychotic disorders. J Clin Psychopharmacol 17:407–418, 1997 Van Putten T: Why do schizophrenic patients refuse to take their drugs? Arch Gen Psychiatry 31:67–72, 1974 VanderZwaag C, McGee M, McEvoy JP, et al: Response of patients with treatmentrefractory schizophrenia to clozapine within three serum level ranges. Am J Psychiatry 153:1579–1584, 1996 Vartiainen H, Tiihonen J, Putkonen A, et al: Citalopram, a selective serotonin reuptake inhibitor, in the treatment of aggression in schizophrenia. Acta Psychiatr Scand 91:348–351, 1995 Wassef A, Dott S, Harris A, et al: Randomized, placebo-controlled pilot study of divalproex sodium in the treatment of acute exacerbations of chronic schizophrenia. J Clin Psychopharmacol 20:357–361, 2000 Weiden PJ: Communicating with acutely psychotic patients: the initial evaluation. Journal of Practical Psychiatry and Behavioral Health 2:47–50, 1996a Weiden PJ: Communicating with psychotic patients during the course of acute treatment. Journal of Practical Psychiatry and Behavioral Health 2:122–124, 1996b Weiden P, Havens L: Psychotherapeutic management techniques in the treatment of outpatients with schizophrenia. Hospital and Community Psychiatry 45:549– 555, 1994 Weiden P, Aquila R, Standard J: Atypical antipsychotic drugs and long-term outcome in schizophrenia. J Clin Psychiatry 57 (suppl 11):53–60, 1996

❉

79

❉

This page intentionally left blank

4 The Difficult-to-Treat Patient With Depression Mantosh J. Dewan, M.D. Ronald W. Pies, M.D.

P

atients with difficult-to-treat depression are a common challenge in clinical practice and have high morbidity and mortality associated with them. With an initial trial of an antidepressant (AD), 50% of depressed patients achieve full remission, 10%–15% show significant improvement but not remission, and the remaining 35%–40% have an inadequate response. Similarly, just over half of all depressed patients treated with interpersonal therapy (IPT), cognitive therapy (CT), or behavior therapy (BT) show a robust response (Thase 1995). Half of all depressed patients, therefore, will show some degree of treatment resistance. Parts of this chapter are adapted from Dewan MJ, Masand P: “Pharmacological Treatment of Resistant Depression,” in Textbook of Psychopharmacology. Edited by Ananth J. Delhi, India, Jaypee Brothers, 1999.

❉

81

❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

Factors Associated With Treatment Resistance Poor response to drug treatment is associated with using an AD drug at too low a dose or for too short a time. Therefore, the goals of initial AD treatment focus on using full doses for at least 6–8 weeks (Quitkin et al. 1996) and perhaps 10 weeks for chronic depressions (Kocsis et al. 1996). Poor response can also be due to poor absorption, noncompliance due to side effects, or secondary to psychological issues such as patients not wanting to accept that they need medication (Dewan 1992; Dewan and Koss 1989). Over 50% of all patients who prematurely decrease or discontinue medication within 6– 9 months relapse (Frank et al. 1992). Chronic psychosocial stressors, an Axis II diagnosis, or comorbid substance abuse may worsen prognosis (American Psychiatric Association 2000; Frank et al. 1992). A poor response to psychotherapy has been associated with several factors, including previous episodes not responsive to treatment. Psychotherapy (of any kind) alone with severely ill depressed patients with psychotic or melancholic features is not recommended (Agency for Health Care Policy and Research 1993; American Psychiatric Association 2000), but this has been challenged on the basis of a meta-analysis (DeRubeis et al. 1999). Correlates of poor response to CT include being single or unmarried, chronicity, initial severity, and premorbid dysfunctional attitudes (Thase 1995) as well as treatment by inexperienced therapists with poor technical skills (Burns and Nolen-Hoeksma 1992; DeRubeis et al. 1990). Abnormal results on two biological markers, baseline sleep electroencephalogram (EEG) and the dexamethasone suppression test, predict a poor response to either CT or IPT alone (Thase 1999). Patients with abnormally fewer delta slow waves during the first rapid eye movement (REM) period were more likely to do poorly with IPT alone. Endogenously depressed patients with reduced REM latency, increased REM density, and decreased sleep efficiency had poorer responses to CT than those patients who had normal baseline sleep EEGs (Kupfer et al. 1990; Thase et al. 1993). We recommend that, as with medications, depressed patients being treated with psychotherapy alone should be reevaluated after 6 weeks, because the largest proportion of change in therapy occurs in the first 6–8 sessions (Koss and Shiang 1994). Some improvement warrants extending treatment to 12 weeks; if there is no improvement, psychotherapy should be replaced by medication as the primary treatment because this is likely to be effective (Agency for Health Care Policy and Research 1993; Stewart et al. 1993). However, nonmedical psychotherapists usually take 6–14 months ❉

82

❉

❉

The Difficult-to-Treat Patient With Depression

❉

before considering this step (Kendall et al. 1992). Predictors of a poor response to combined treatment have been reported in a study of 95 elderly depressed patients. Predictors were 1) higher pretreatment levels of acute and chronic stressors, 2) poorer social supports, 3) younger age at first depressive episode, 4) endogenous depression, 5) higher current anxiety, 6) older current age, and 7) poorer subjective and objective (EEG) sleep (Dew et al. 1997). When a patient is not responding adequately to competent treatment, a complete review is imperative. Is the diagnosis correct? Bipolar disorder or a psychotic component should be especially ruled out. Is there an additional comorbid diagnosis or ongoing psychosocial stressor that has not been appreciated and that needs additional attention? Is there a contributing medical factor such as hypothyroidism? Are there medications (e.g., steroids) that may be exacerbating the depression or significant drug-drug interactions interfering with AD efficacy?

Subtypes of Depression Accommodations must be made for subtypes of depression. Psychotic depression, which occurs in up to 15% of severely ill patients, is often misdiagnosed. Patients do not volunteer their psychotic symptoms, and clinicians often do not associate psychosis with depression, thereby laying the foundation for the underdiagnosis of psychotic depressions. Direct inquiry is recommended. Patients with psychotic depression are more likely to be bipolar, with recurrent episodes and a positive family history of depression. Mood incongruence suggests a poor prognosis. In unipolar psychotic depression, the prognosis is stark. Two years after the first episode, although 94% were nonsymptomatic, only 29% had recovered baseline function judged by occupational and residential status (Tohen et al. 2000). Psychotherapy alone is contraindicated and tricyclic antidepressants (TCAs) alone may exacerbate symptoms; however, ADs, such as amoxapine or the selective serotonin reuptake inhibitors (SSRIs) fluvoxamine, sertraline, and paroxetine, may be effective by themselves. A combination of an AD plus an antipsychotic is proven to be effective and is recommended. If symptoms persist, lithium augmentation is helpful. Electroconvulsive therapy (ECT) is very effective and sometimes the preferred treatment (American Psychiatric Association 2000; Crismon et al. 1999). Bipolar depression is usually more difficult to treat than unipolar depression. Further, bipolar depression is often misdiagnosed as unipolar, leading ❉

83

❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

to mistreatment. Bipolar depression must be treated with mood stabilizers; add an AD only if necessary. Treatment with an AD alone carries a significant risk of precipitating mania and rapid cycling. Therefore, the differential diagnosis between bipolar and unipolar depression is important (some differences are described in Table 4–1). Depression during mania (mixed mania) responds poorly to lithium and well to divalproex sodium (Swann et al. 1997). (Bipolar depression is also addressed in Chapter 2 of this book.) Atypical depression is characterized by mood reactivity, increases in both sleep and appetite, sensitivity to rejection, and leaden paralysis (Pies 1988). Atypical depression shares many features with personality disorders, especially borderline personality disorder, and careful attention to diagnosis is important. It is responsive to monoamine oxidase inhibitors (MAOIs), SSRIs, and CT (Crismon et al. 1999; Jarrett et al.1999; Pies 1988). Table 4–1. Bipolar versus unipolar major depression

Bipolar depression

Unipolar major depression

Family history of mania

Prominent

Minimal or none

Psychomotor retardation

Likely

Less likely

Sleep

Hypersomnia likely

Insomnia likely

Weight gain

Likely

Less likely

Episodes

Short, frequent episodes

Longer episodes

Frequent

Infrequent

Postpartum depression Source.

Adapted from Goodwin and Jamison 1990.

Dysthymia is often undiagnosed and/or undertreated because the longstanding, milder symptoms of depression are explained away as “everyone has their ups and downs.” The results of long-term, naturalistic follow-ups conflict with the more optimistic outcomes seen in short-term research studies. Two large naturalistic studies indicate that “although dysthymic disorder often presents with only mild or moderate symptoms, from a longitudinal perspective, it is more severe than episodic major depressive disorder” (Klein et al. 2000, p. 937). In a 2-year study of 626 subjects, dysthymic patients had the poorest functional status (but not the most severe depression) at the beginning and the worst prognosis compared with patients with major depressive disorder (MDD) and subthreshold depression (Wells et al.1992), with the suggestion that dysthymic patients with their milder symptoms ❉

84

❉

❉

The Difficult-to-Treat Patient With Depression

❉

were often not treated. In the longest (5 years) prospective, naturalistic follow-up of 86 early-onset dysthymic patients and 39 patients with episodic major depression, Klein and colleagues (2000) confirmed the more chronic and severe course of dysthymic disorder despite “usual” treatment, which appears to be suboptimal. Only 50% of these patients were receiving treatment at the end of the first year; this percentage declined to 41% at 5-year follow-up. Of note, only a minority received adequate levels of ADs; 26% and 30% at the end of 1 and 5 years, respectively. The number of patients getting psychotherapy decreased from 38% at the end of year 1 to 23% at the 5-year mark. Klein and colleagues (2000) state that “it is interesting that the patients with dysthymic disorder received significantly more treatment than did patients with episodic major depressive disorder across the follow-up period. Thus, treatment probably cannot account for the poorer course and outcome of the dysthymic disorder group” (p. 937). They report that dysthymic patients were significantly more likely to have a lifetime history of substance abuse or dependence and a comorbid personality disorder, both being usually associated with poor outcome. The 5-year recovery rate was 53%, with a median time to recovery of 58 months. The rate of recovery was steepest in the first 7 months (22%), steady through month 35 (47%), and slow after that. Of those who recovered, 45% relapsed over a mean of 23 months’ follow-up. Compared with patients with MDD, patients who had been dysthymic for over 5 years were more likely to be depressed (70% of the time versus 25% for MDD), attempt suicide, be hospitalized, and have a lower level of functioning, all attesting to their greater burden. In contrast, several short-term, controlled studies have reported a 60%– 65% response rate (comparable with MDD) in dysthymic patients treated with adequate doses of TCAs, SSRIs, and MAOIs. Stable remission was maintained at 5-month follow-up with improvement in psychosocial functioning and mood (Kocsis et al.1997; Thase et al. 1996). It should be noted that chronic dysthymic patients may take up to 10 weeks on an AD to show a response. The specific therapies (IPT, CT, BT and CBT) all effectively treat dysthymia “although responses have been somewhat smaller than...with major depression” (American Psychiatric Association 2000, p. 16). The best outcomes are seen when psychotherapy and medications are combined (American Psychiatric Association 2000; Keller et al. 2000). In a 12-week study comparing the effect of nefazodone, the cognitive behavioral analysis system (CBAS), and their combination on 681 patients with chronic depression (42% of whom had double depression [i.e., a current major depression superimposed on dysthymia]) Keller and colleagues (2000) report ❉

85

❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

a response rate of 55% in the medication group, 52% in the therapy group, and 85% in the combined treatment group. Long-term outcome data for this combined treatment are awaited. It is apparent that although efficacy in research protocols has been established, there is a need for improved effectiveness in the usual clinical settings. Clinicians need to actively evaluate patients for this serious and disabling disorder and then treat it vigorously with adequate, and perhaps maximum, doses of ADs plus competently delivered specific psychotherapy. Seasonal depression is characterized by recurrent episodes, with its onset in fall and remission or hypomania in spring. It is common mainly in the higher, nonequatorial latitudes and responds well to phototherapy and the SSRIs. Comorbidity with bipolar II disorder may delay or complicate response to appropriate treatment.

Comorbidity The frequently present substance abuse, medical, or psychiatric comorbidity enhances the complexity of the clinical presentation, diminishes treatment response, and worsens prognosis.

Comorbid Substance Abuse A large study of depressed patients in primary care and psychiatric practices reported a 18%–29% lifetime prevalence and a 14%–19% current prevalence for comorbid alcohol abuse (Sherbourne et al.1993). In depressed patients, even occasional alcohol use (even when not abuse) interfered with AD treatment and was associated with poorer functioning and well-being (Worthington et al. 1996). Alcoholism led to more primary care visits, an increased risk of hospitalization, and suicide attempts but to a decrease in mental health treatment (American Psychiatric Association 2000). The relationship between mood disorders, substance abuse, and suicide is particularly strong in the teenage years and in young adults (Akiskal 1995).

Comorbid Medical Disorders In a widely cited prospective study, Hall and colleagues (1981) found that about half of all psychiatric admissions were prompted by unrecognized medical conditions causing or exacerbating psychiatric symptomatology. In another study, 58% of depressed patients in treatment reported a comorbid ❉

86

❉

❉

The Difficult-to-Treat Patient With Depression

❉

medical disorder (American Psychiatric Association 2000). Many medical diseases and medications (see Tables 4–2 and 4–3) may be responsible and should be specifically considered in all treatment-resistant patients. The role of folate deficiency in depression is underrecognized. Folatedeficient depressed patients are more likely to have melancholic depression, suffer longer episodes, and are less likely to respond to SSRIs (Alpert and Fava 1997). An estimated 10%–30% of depressed patients have low folate levels independent of poor nutritional status. Folate levels should therefore be measured particularly in patients who are treatment refractory. If low, the addition of 15 mg/day of methylfolate (which is preferred over folic acid) to their AD has been shown to improve outcome (Godfrey et al. 1990). Initial data suggest that the addition of folate to patients with normal folate levels may augment AD response. Coppen and Bailey (2000) report significant improvement in 127 depressed patients when 500 mcg of folic acid was added to fluoxetine irrespective of baseline blood folic levels. Women improved more than men, who may need higher doses (up to 1 mg) of folic acid. Folate is required for synthesis of S-adenosylmethionine (SAMe), which plays a key role in the production of serotonin, norepinephrine, and dopamine. The few, small studies of SAMe suggest that it may be a physiologically based AD superior to placebos and comparable to TCAs. It may have a quicker onset of action and fewer side effects (primarily headache, gastrointestinal, and hypomanic/manic symptoms). Use with caution because there are no long-term studies of SAMe in bipolar or psychotically depressed patients (Pies 2000). Available over the counter and already widely used, Brown and colleagues (1999) suggest that SAMe can be used alone (800 mg bid) or combined with most ADs—cautiously with the SSRIs because of the potential of a serotonin syndrome—but not with the MAOIs.

Table 4–2. Medical conditions associated with depression: DEMONS mnemonic Deficiency states (i.e., B12, folate, iron, minerals) Endocrinopathies (i.e., thyroid, pituitary, adrenal) Malignancies (i.e., pancreas, stomach) Others (i.e., infections, toxins) Neurological illnesses (i.e., Parkinson’s disease, stroke, AIDS) Seizures (i.e., complex partial seizures)

❉

87

❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

Table 4–3. Medications associated with depression Anticonvulsants

Phenobarbitol, clonazepam

Antihypertensives

Reserpine, clonidine, α-methyldopa

β-blockers

Propranolol

Steroids

ACTH, anabolic steroids, glucocorticoids

Drugs

Cocaine, amphetamine, alcohol

Histamine blockers

Cimetidine, ranitidine

Others

Ethambutol, levodopa, cycloserine

Note.

ACTH=adrenocorticotropic hormones.

Similar to folate, vitamin B12 deficiency is associated with a twofold increased risk of severe depression. In community-dwelling older women, B12 deficiency was found in 15% of 478 nondepressed women, 17% of the 100 mildly depressed subjects, and 27% of 122 severely depressed women (Penninx et al. 2000). Vitamin B12 deficiency may also result in an earlier onset of depression (Levitt and Joffe 1989) but was not related to depressive subtype or treatment response (M. Fava et al. 1997). Since B12 and folate deficiency often coexist, and because treatment with folate alone (without B12) can further neurological deterioration if B12 is low, both levels should be checked. A minority of depressed patients have increased cortisol levels and abnormal dexamethasone suppression. These patients respond poorly to placebo or psychotherapy but adequately to pharmacotherapy (Ribeiro et al. 1993; Thase 1999). Could baseline markers (e.g., cortisol or EEG abnormalities) help guide our initial choice between psychotherapy or pharmacotherapy?

Comorbid Psychiatric Disorders Anxiety disorders are frequently comorbid with depression. Sherbourne and colleagues (1996) reported that up to 54% of 2,494 depressed patients seen in primary care also had generalized anxiety disorder, and an additional 9.4% had panic disorder, both of which increase the risk of suicide and decrease the response to ADs. The 30% of depressed patients with a comorbid personality disorder are less likely to respond to an AD (Frank and Kupfer 1990). Often, after depression remits, unrecognized personality traits not responsive to medications ❉

88

❉

❉

The Difficult-to-Treat Patient With Depression

❉

become apparent. Psychotherapy is helpful for these patients. On the other hand, chronic depressive symptoms and dysfunction may be mistakenly labeled a personality disorder and thus prescribed psychotherapy but not medication, thereby undertreating a treatable disorder.

Treatment Options for Resistant Depressions Currently, our options include pharmacotherapy, psychotherapy, a combination of these two, or somatic treatments such as ECT or phototherapy.

Pharmacotherapeutic Options Pharmacological options for the treatment of resistant depressions focus on optimization of AD monotherapy, switching to another AD, augmentation, or a combination of two (or more) ADs. Optimization of Initial Antidepressant Treatment In suboptimal responders to TCAs, measuring blood levels is a useful first step. In a seminal article, Glassman et al. (1977) showed that levels of imipramine plus desipramine above 225 ng/mL yielded a 93% response rate, between 150 ng/mL and 225 ng/mL a rate of more than 60%, and below 150 ng/mL a mere 30% response rate. High levels may be needed for a good response in severely ill, resistant patients as well (Kupfer and Frank 1996). The utility of blood levels for the SSRIs and other non-TCA antidepressants is less clear. Noncompliance is a major factor in poor outcome. In a 3-year maintenance trial, only 1.5% of patients doing well for all 3 years were “clearly noncompliant,” compared with 58% of those who had a recurrence (Frank et al. 1992). Direct questioning, repeated education of patients about their illness, and involvement of significant caretakers may be needed to improve compliance. Psychotherapy can be added with the primary objective of “increased adherence to medication prescription” (Agency for Health Care Policy and Research 1993). Optimization also includes managing side effects (e.g., by splitting the dose, moving a sedating drug from the A.M. to P.M., or even decreasing the total dose). In contrast, an increase above the standard dose of an AD may be helpful. M. Fava and colleagues (1994) showed that increasing the dose of fluoxetine from 20 mg to 40 mg per day was more effective than adding lithium or desipramine. This was also effective for patients who relapsed on ❉

89

❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

fluoxetine (the so-called poop-out effect) after a good initial response (M. Fava et al. 1994). It may be necessary to cautiously increase to above the recommended upper dose limits if there are no intervening side effects. For instance, Amsterdam and Berwish (1989) reported a 70% response rate in treatment-resistant patients treated with high doses of tranylcypromine. The dose range of 90–170 mg/day to 170 mg/day, average 113 mg/day, was well above the usual upper limit of 60 mg/day. Careful monitoring for hypotension and other side effects is important. Often the best results can be obtained by merely adding the “tincture of time” (Pies 1997), a lesson that is especially pertinent in today’s managed care climate. The Next Step: Augment or Switch? If there is an inadequate response even after optimization of monotherapy, does the clinician switch to another AD or augment the existing AD? There are no controlled studies directly comparing the relative efficacy of these strategies, and there is no consensus on this important clinical question (Crismon et al. 1999; Pies 1998; Posternak and Zimmerman 2001). Based on our clinical experience and some data, we recommend that this next step should depend on whether there is no response or a partial response to the first AD, with nonresponders being switched to another class of ADs and partial responders being augmented. Pies (1998) summarizes data based on studies by Keller (1995) and others that suggest that the worse the response to the first AD, the more likely a response to the second class of AD. Therefore nonresponders are most likely to benefit from switching to another AD. Other factors that favor switching from one AD to another include a good (50%–70%) response rate, the potential for fewer side effects, absence of drug-drug interactions, lower medication costs, and increased patient adherence with monotherapy. Arguments in support of augmentation of partial responders include a good (50%–60%) response rate, shorter response time of 1 day to 3 weeks, avoidance of abandoning partial response with monotherapy and the patient discouragement that may result from a failed trial, fear of the depression worsening when the partially effective AD is discontinued, and evidence that some augmenting agents convert partial responders to full remitters (Crismon et al. 1999). Lastly, a prospective, naturalistic comparison found a trend favoring augmentation (56% improvement) over switching (45%). In nonresponders, a second switch or augmentation was equally effective, with a 50% response rate (Posternak and Zimmerman 2001). ❉

90

❉

❉

The Difficult-to-Treat Patient With Depression

❉

Switch to another antidepressant. If optimization is ineffective, it is beneficial to switch nonresponders to monotherapy with another AD agent. If the first AD is an SSRI—as is increasingly the case—the second AD could be another SSRI (preferred because of their lower side-effect potential) or a TCA. Pies (1998) reports on several studies that found a switch within the SSRI class may be effective in 40%–70% of patients: for instance, 70% of patients who were intolerant to fluoxetine responded to sertraline, and two separate studies showed that 42% and 63% responded when sertraline failures were switched to fluoxetine. If two trials of SSRIs are not successful, a novel AD rather than a third SSRI should be tried (discussion follows). A switch from an SSRI to a TCA is also very effective (60%–70% response rate) but liable to produce more side effects. For example, a study by Peselow and colleagues (1989) found that 73% of patients resistant to paroxetine responded to imipramine, which was also effective in about 60% of sertraline nonresponders (Keller 1995). If the first AD is a TCA, a switch to an SSRI is recommended, with an expected response rate of 45%–55% (Keller 1995; Preskorn 1996). One multicenter study that crossed over imipramine nonresponders to sertraline found that patients switched to sertraline had a 55% response rate. Also, patients who showed no response to imipramine did better on sertraline than those who had some benefit from imipramine treatment (Keller 1995). A switch within the tricyclic class, however, is not recommended because this has a mere 22% response rate (Preskorn 1996). A switch from a TCA to an MAOI is especially effective in depression with atypical features. For instance, there was a 70% response to phenelzine in patients who were nonresponders to imipramine (Mueller et al. 1996) or who had failed a combination of interpersonal psychotherapy and imipramine (Thase et al. 1996). The introduction of novel ADs provides another ray of optimism. A switch to venlafaxine, with its serotonin and norepinephrine reuptake inhibiting activity, may be particularly helpful in refractory patients. Phillips and Nierenberg (1994) reported a 40% response rate to venlafaxine in patients who had failed adequate trials of other treatments, including MAOIs and ECT. Further, Thase and colleagues (2001) analyzed eight studies involving 2,045 patients and found remission rates for venlafaxine to be significantly higher (45%) than those for SSRIs (35%) and placebo (25%). Remission was defined as a Hamilton Rating Scale for Depression score of ≤ 7 as opposed to response, which is commonly defined as a 50% reduction from baseline and often does not reach the remission score of ≤ 7. Mirtazapine, which also ❉

91

❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

facilitates both serotonin and norepinephrine, is another good option for severe depression (Kasper l997; Megna and Dewan 1999). Reboxetine, the first selective noradrenergic antidepressant, may represent another option for switching or for combination (Kent 2000). Exactly when should clinicians switch ADs? Quitkin and colleagues (1996) suggest two chronological milestones. First, if there is no improvement after 4 weeks, as occurs in about 10% of patients, switch. Second, if there is minimal improvement initially but no significant improvement after 5 weeks, switch to another AD agent. On the other hand, if there is some improvement, patients may need up to 10 weeks for a full effect, especially with depressions of long duration (Kocsis et al. 1996). Augmentation. A partial response to the initial AD warrants augmentation. Augmentation is the addition of an agent that by itself does not have AD efficacy. If augmentation is effective, both agents should be continued for the entire duration of treatment unless side effects become troublesome. This recommendation is supported only by a small study of lithium augmentation in elderly patients (Hardy et al. 1997) and has not been adequately tested. Augmentation of ADs, primarily the TCAs, has been tried with a number of agents. The most widely used strategy is to add lithium at 600– 900 mg/day (serum level of lithium ≥ 0.4 mEq/L) for a 3-week trial. Some 50%–60% of patients previously resistant to TCA or MAOIs respond (Post 1995), and improvement can occur in a day or two but may take 2 weeks or longer (Thase et al. 1989). A double-blind study found that the addition of lithium or desipramine was equally effective in fluoxetine nonresponders but somewhat less efficacious than increasing the dose of fluoxetine (M. Fava et al. 1994). Thyroid supplements are generally safe and easy to use as augmenting agents. In unipolar depressions, triiodothyronine (T3) is preferred over thyroxine (T4) because of a better response rate (50% vs. 20%). Also, in one study, no patients got worse on T3, whereas 30% did on T4. A review of studies reporting the potentiation of TCAs with T3 reveals a wide range of response rates, from 50% to 91% in open trials, 25% to 70% in partially controlled trials, to 0% to 67% in controlled trials (Nemeroff et al. 1996). A meta-analysis by Aronson and colleagues (1996) of pooled data from 292 patients showed a doubling of the response rate from 24% in control subjects to 57% in T3-augmented patients. Therefore, 50 mcg/day of T3 is recommended for a 3-week trial. Improvement often occurs within days and usually within the first week or two (Post 1995). If there is no improvement ❉

92

❉

❉

The Difficult-to-Treat Patient With Depression

❉

after 3 weeks, T3 must be discontinued. Long-term use may cause rebound hypothyroidism when T3 is stopped. Case reports suggest the utility of T3 augmentation with other classes of ADs as well, including the SSRIs and MAOIs (Aronson et al. 1996). A history of atrial fibrillation is a relative contraindication to the use of T3. Although lithium is better studied and usually preferred over T3 as an initial augmenter, a controlled study (Joffe et al. 1993) found lithium and thyroid augmentation to be equally effective and superior to placebo. Contrary to most reports, the study also found T4 to be as effective as T3. Numerous other agents are being experimented with as augmenting agents. In a controlled trial, Shelton and colleagues (2001b) added an atypical antipsychotic, olanzapine, to fluoxetine in treatment resistant but nonpsychotic depressed patients. At 8 weeks, fluoxetine augmented by olanzapine demonstrated superior efficacy over either agent alone. Alternative augmentation strategies that have merely anecdotal or conflicting support for their efficacy are presented in Table 4–4. These agents are often used in difficult-to-treat patients as a last resort and should be considered therapeutic trials. Table 4–4. Alternative augmentation strategies Medication

Dosage

Pergolide

0.05–1 mg/day for 1 week

Bromocriptine

2.5–20 mg/day for 1 week

Amantidine

100–200 mg/day for 1 week

Estrogen

0.125–0.375 mg/day for 2 weeks

Sodium valproate

750–1500 mg/day for 3 weeks

Carbamazepine

600–1200 mg/day for 3 weeks

Pindolol

7.5–15 mg/day for 1 week

Buspirone

30 mg/day for 3 weeks

Modafinil

200–400 mg/day for 1 week

Combining Antidepressant Medications Combination therapy refers to the simultaneous use of two standard ADs. Despite a lack of data, there is increasing use of theoretically sound “rational polypharmacy,” and Stahl’s (1996) schema is helpful. Stahl suggests that the first step is monotherapy, followed by augmentation in the “classic combo” ❉

93

❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

(AD plus lithium) and the “hormone combo” (AD plus thyroid). This is followed by combining two ADs. Controlled studies have shown that combining SSRIs with TCAs (in either order) is effective (Phillips and Nierenberg 1994). Because SSRIs (especially fluoxetine and paroxetine) tend to increase TCA levels via P450 system activity, TCA doses must be decreased and blood levels must be checked to prevent toxicity. The “cautious combo,” a combination of TCAs and MAOIs, has been reported to be effective (although this has not been established in controlled studies) in treatment-resistant depressed patients (Feighner et al. 1985). However, there have been reports of toxicity (e.g., serotonin syndrome) with this combination. If it is used, an MAOI should be cautiously added to the TCA or the two should be started together; adding a TCA to an MAOI has a high risk of toxicity. The combination of MAOIs and SSRIs should be avoided because of the risk of a serotonin syndrome. Anecdotal reports indicate that stimulants are effective in combination with other ADs. Methylphenidate or dextroamphetamine, usually 5–10 mg twice a day, are useful ADs by themselves, particularly in depressed medically ill patients. Case reports suggest they can be added to TCAs (“adrenergic combo”), SSRIs, or MAOIs with good results. Their quick onset of action, usually within a day or two, is an advantage. The Physicians Desk Reference (2001) does warn of the risk of hypertension when MAOIs are combined with stimulants, and Post (1995) recommends obtaining informed consent before gingerly combining them. There are sparse data supporting the “serotonin combo” of an SSRI plus nefazodone, buspirone, or trazodone, but this may be an effective step, particularly if trazodone is also being used as a hypnotic. It is increasingly common to affect both the serotonin and adrenergic systems in difficult-to-treat depressions, leading to full doses of a TCA (with blood levels monitored) plus SSRI, or SSRI plus bupropion or stimulant or reboxetine, or venlafaxine plus mirtazapine, because each drug has a different mechanism of action and may be complementary to the other. The most aggressive treatment, which may involve combining three or even four ADs, is aptly dubbed the “heroic combo” (Stahl 1996). However, ECT is arguably the preferred option at this point (see Chapter 13 of this book).

Psychotherapeutic Options Optimization In acute mild to moderate depression, a depression-specific psychotherapy alone is the preferred treatment because there is strong support for the effi❉

94

❉

❉

The Difficult-to-Treat Patient With Depression

❉

cacy of IPT, CT, and BT in this condition. Brief dynamic therapy, although widely used in practice, may be less effective (Agency for Health Care Policy and Research 1993). In severe depression, medications are generally preferred as a primary treatment (American Psychiatric Association 2000). However, there are data that suggest that CT may be equally effective in severe depression (DeRubeis et al. 1999). Despite widespread use and eloquent theoretical arguments for its utility, data suggest that brief dynamic therapy is less effective than the specific therapies. For instance, in one meta-analysis the response rate was 35% for brief psychodynamic therapy compared with 47% with CT, 52% with IPT, and 55% with BT (American Psychiatric Association 2000). In addition, the efficacy of long-term psychodynamic psychotherapy or psychoanalysis in either the acute or maintenance phase of major depression, either in conjunction with pharmacotherapy or alone, has not been adequately studied (American Psychiatric Association 2000). Eclectic therapy or even therapies that purportedly integrate IPT, CT, and BT have not been found to be better than the individual, specific models (Jacobsen et al. 1991; Rehm et al. 1987; Rude 1986). We conclude that replacing the most commonly used therapy, generic psychodynamic psychotherapy—which is more effective than no treatment for control subjects on a waiting list (American Psychiatric Association 2000)—with depression-specific therapies (IPT, CT, BT) would lead to improved outcomes and would optimize treatment. In the maintenance phase, treatment with CT has the advantage of lower relapse rates when compared with medications after both treatments have been terminated (Paykel et al. 1999), although relapse rates after successful CT in the acute phase are still troublesome (50% at 1 year, 74% over 2 years). Jarrett and colleagues (2001) therefore developed a 10-session relapse prevention CT program for use in the continuation phase. Unmedicated patients with recurrent depression in this program had a markedly reduced relapse rate (16% vs. 67% in control subjects) at the end of 2 years. In women patients who suffered marital discord and depression, O’Leary and Beach (1990) found that behavioral marital therapy alone treated depression as effectively as, and increased marital satisfaction more than, CT alone. These gains were maintained at 1-year follow-up. In the only controlled study, Covi and colleagues (1974) found that group therapy alone was not effective in the treatment of depression. A favorable outcome with psychotherapy has been highly related to the competence of the therapist. Above average IPT was as effective as medications, whereas below average IPT was as poor as placebo (Frank et al. 1990; ❉

95

❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

Thase 1995). Therefore, optimal psychotherapy would consist of a specific therapy (e.g., IPT or CT) provided by a competent therapist. The relationship of the “dose” of psychotherapy to efficacy has not been well studied. Clinical practice and research protocols suggest once weekly therapy in the acute phase and once a month or even less frequent “booster” sessions during the maintenance phase (American Psychiatric Association 2000; Elkin et al. 1989; Frank et al. 1990). How long should we treat a patient with psychotherapy alone? We know that most of the positive change occurs in the first 6–8 sessions, especially in mild to moderately ill patients (Koss and Shiang 1994). The Agency for Health Care Policy and Research (1993) guidelines state that “if the patient being treated with psychotherapy fails to show improvement in depressive symptoms by 6 weeks or only partial response by 12 weeks, a reevaluation and potential switch to, or addition of, medication are indicated. Medication is almost always recommended for those who do not respond to therapy at all” (p. 4). However, Kendall and colleagues (1992) found that nonmedical therapists wait an average 6–14 months, rather than the 2–3 months recommended by the Agency for Health Care Policy and Research (1993) before considering treatment a failure and pursuing additional options such as medication. Thase and colleagues (1992) suggest that brief therapies be sensitive to time as well as to endpoint clinical improvement. They report that in patients with an endpoint Hamilton Depression Rating Scale score of 6 or below, there was a 10% relapse rate after 1 year. If a partial or less-sustained endpoint response was obtained, the relapse rate was more than 50%. Switch to Another Psychotherapy Technique Theoretically, a switch from a more generic “psychodynamic” therapy to a specific therapy (CT, IPT, or BT) would enhance efficacy (American Psychiatric Association 2000), particularly for the long-term, maintenance phase. However this strategy has yet to be adequately studied. Augmentation There are no studies that have found effective augmenters of the specific psychotherapies. Group therapy by itself is not an effective treatment in depression (Covi et al. 1974), but clinical experience suggests that it may be a useful adjunct in the treatment of bereavement or chronic illness or as a support group for medication maintenance (American Psychiatric Association 2000). ❉

96

❉

❉

The Difficult-to-Treat Patient With Depression

❉

Combining Psychotherapies There are no studies in which two specific and effective individual therapies have been combined. There is theoretical justification for doing so because these therapies potentially target different aspects of a patient’s psychological difficulties. However, in a study (Keller et al. 2000) of the cognitive behavioral analysis system of psychotherapy (CBASP), an approach that “draws on many behavioral, cognitive, and interpersonal techniques used in other forms of psychotherapy” (p. 1463), CBASP was effective in about half the patients, which is equivalent to results with the specific therapies in other studies. Besides the specific individual therapies (IPT, CT, and BT), there is some evidence that marital therapy is effective (Agency for Health Care Policy and Research 1993). It has been suggested that marital therapy be combined with individual therapy when marital conflict precedes either the onset of depression or relapse (Paykel et al. 1969). Results from the three studies that did combine these modalities suggest that marital and family therapy may reduce the depressive symptoms and the risk of relapse in patients with marital and family problems. These modalities may be less useful in depressed patients without specific family or marital discord (American Psychiatric Association 2000).

Combined Treatment With Pharmacotherapy and Psychotherapy Pharmacotherapy and psychotherapy can be combined sequentially (in either order) or prescribed simultaneously. Sequential Therapy Two preliminary studies lend support for the sequential use of pharmacotherapy and psychotherapy. G. Fava and colleagues (1997) studied “drugresistant” patients who had failed at least two adequate trials of medications. When treated with a minimum of 10 (mean of 15) sessions of CBT, 63% of these 19 patients showed a good response. Paykel and colleagues (1999) studied 158 partial responders to ADs randomized to CT or clinical management. The addition of CT to AD therapy increased rates of full remission and decreased relapse rates (29% vs. 47% for control subjects) at 68 weeks. Therefore, after a fair trial of AD medication alone, if there are difficulties with adherence, persistent psychosocial stressors, or residual symptoms, the addition of a specific psychotherapy is advisable. The reverse sequence was studied by Stewart and colleagues (1993). In their double-blind study, nonresponders to CT were randomly assigned to either imipramine (up to a ❉

97

❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

maximum of 300 mg/day) or placebo for 6 weeks. Of the 12 patients completing the trial, all 5 on imipramine responded, whereas none of the 7 on placebo improved. If depression responds to medication added to adequate psychotherapy, the patient is often able to competently deal with associated psychosocial problems with minimal or no further psychotherapy (Kocsis et al. 1997; Mintz et al. 1992). However, patients with comorbid personality disorders may be unable to tolerate a reduction. We recommend that, after the full effects of the added medications are felt, the need for continuing psychotherapy should be reevaluated in all patients. Simultaneous Therapy Acute phase. Despite a lack of supporting evidence, expert opinion has long held that severely depressed patients in the acute phase should be treated simultaneously with a combination of psychotherapy and medication (Agency for Health Care Policy and Research 1993; American Psychiatric Association 2000), because combined treatment may increase the magnitude, probability, and breadth of response as well as the acceptability of treatment (i.e., psychotherapy enhances compliance to medications, and medications speed symptomatic response and facilitate psychotherapeutic work) (Hollon and Fawcett 1995). Further, the Agency for Health Care Policy and Research guidelines (1993) suggested an advantage for patients who have responded partially to either treatment alone or who have a history of chronic episodes, chronic psychosocial problems, or treatment adherence difficulties. Hollon and Fawcett’s (1995) review found trends favoring combined treatment and concluded that “[d]rugs appear to work more rapidly than most types of psychotherapy (CT excluded) and may depend less on the skill of the practitioner for their effect. Conversely, psychotherapy appears to do more to enhance social functioning or reduce subsequent risk (particularly CT)” (p. 1231). For instance, one study (DiMascio et al. 1979) found that medication and IPT combined reduced patient dropout rates, improved symptom relief, and increased social adjustment more than either treatment alone. Only recently have there been data clearly supporting combined treatment over monotherapy. Thase and colleagues (1997) conducted a metaanalysis of the treatment outcome of nearly 600 depressed outpatients and compared psychotherapy alone to combined therapy. In severely depressed patients (i.e., those with a Hamilton Depression Rating Scale score of over 20), they found that combined therapy was more effective (43% compared ❉

98

❉

❉

The Difficult-to-Treat Patient With Depression

❉

with 25%) and led to faster recovery (the 75th percentile of the time to recovery was 9 weeks compared with 13 weeks). In less severely depressed patients, combined therapy was not significantly better than psychotherapy alone (although the recovery rate was 48% compared with 37%). Therefore, these patients may be well served by CT or IPT alone. The results from perhaps the definitive study of the efficacy of combined treatment clearly favor its use in the acute treatment of chronic depression. Keller and colleagues (2000) randomly assigned 681 chronically depressed patients to 12 weeks of either nefazodone monotherapy, CBASP alone, or combined nefazadone plus CBASP. Combined treatment resulted in “higher rates of response and remission than any previously reported treatment for chronic depression” (85%) compared with monotherapy with nefazodone (55%) or CBASP (52%). Maintenance phase. In the maintenance phase, “assuming a full response to combined treatment in acute or continuation phases, the evidence shows that medication alone may be all that is necessary to prevent a recurrence. A substantial number of clinicians believe that some patients will benefit from targeted psychotherapies in continuation or maintenance phases if only partial symptomatic response or psychosocial restoration was obtained with medication alone” (Agency for Health Care Policy and Research 1993, p. 122). Further, “maintenance psychotherapy as the sole treatment to prevent recurrence is generally not recommended unless the patient, for some reason (e.g., pregnancy) needs to avoid medication” (Agency for Health Care Policy and Research 1993, p. 118). A nonrandomized study found a reduction in self-reported depressive symptoms at 20 weeks with the addition of CT to medication, an advantage that was maintained at 12 months (Miller et al. 1989). Two other naturalistic studies suggest that the addition of CT may reduce the rate of relapse over a 1-year follow-up period (Blackburn et al. 1986; Jarrett et al. 1999). Two early controlled studies of combined treatment for maintenance in depressed patients were not supportive of it being superior to either treatment alone although there was a nonsignificantly better breadth of response in one study (Weissman et al. 1974) and a lower dropout rate of 8% on combined treatment versus 21% on medication alone in the second study (Frank and Kupfer 1990). More recently, three randomized controlled studies evaluated the rates of relapse in chronically depressed patients and found support for the efficacy of combined treatment. Frank and colleagues (1990) followed 128 patients with highly recurrent depression for 3 years. They ❉

99

❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

were treated with imipramine and IPT. G. Fava and colleagues (1998) studied 40 patients who were treated with medication and CT over a 6-year period, and Paykel and colleagues (1999) reported on the 1-year outcome of 158 patients also treated with medications and CT. These studies found that patients in combined treatment continued to suffer from depressive symptoms despite adequate treatment, but risk of relapse decreased by about 50%, benefits were maintained even after treatment was discontinued, and this group had significantly fewer unscheduled visits to their psychiatrist (Paykel et al. 1999) When combined treatment with medication and psychotherapy is indicated, both treatments provided by one person (i.e., a psychiatrist) is preferred over split treatment (M.D. psychopharmacologist plus non-M.D. therapist) on both theoretical and economic grounds (Dewan 1999). Medicine and managed care organizations espouse the “primary care” model, in which biopsychosocial care provided by a single physician replaces fragmented care from several specialists. Ironically, psychiatry, which has long practiced the biopsychosocial paradigm, is being fragmented, and the “split model” of M.D. psychopharmacologist plus a non-M.D. psychotherapist is being promoted supposedly because it is less expensive. However, Dewan (1999) used managed care rates to show that integrated treatment (10 medication visits plus 15 psychotherapy sessions) with a psychiatrist was less expensive than split treatment with either a psychologist ($1,331 vs. $1,547) or a social worker ($1,331 vs. $1,392). If treatment consisted of 3 medication visits and 5 psychotherapy sessions, the cost would be $456 for a psychiatrist doing both compared with split treatment costs of $491 with a social worker and $544 with a psychologist. Group and family therapy have also been combined with medication. Neither the addition of once weekly group therapy or family therapy improved outcome. However, family therapy was able to enhance the breadth and magnitude of response, primarily in women patients (Covi et al. 1974). Combined treatment in the elderly. Combined treatment may be particularly effective for the elderly (who tend to have a lower response rate) in the acute phase but may need to be applied for 12 weeks or more to optimize results. In one study of combined treatment with IPT and nortriptyline in elderly patients with a history of recurrent depressive episodes, remission was achieved by 66%, a better-than-usual response for either modality alone (Miller et al. 1997). Dew and colleagues (1997) alert us to the much wider variance in the temporal profiles of recovery in the elderly. They note that a ❉

100 ❉

❉

The Difficult-to-Treat Patient With Depression

❉

rapid response group had 100% recovery at week 10, the gradual or delayed response group achieved this by week 12, a third group showed a partial response, and the fourth group was nonresponsive. Further, Reynolds and colleagues (1999) showed that combined treatment was better for maintenance as well. They report that nortriptyline plus IPT prevented relapse better than either alone, a finding that was most marked among patients over 70 years of age. Can we recommend combined treatment? In sum, “specific effective psychotherapy...is preferred as solo treatment” for mild to moderate depressions (American Psychiatric Association 2000). For severe and chronic depressions—and if there is a history of poor compliance, personality disorder, or interpersonal/psychosocial issues—combined treatment provides the best short- and long-term results. Scott (2000) suggests that, in these patients, the effects of combined treatment are additive. However, he points out that the delivery of competent combined treatment poses a formidable challenge. There is a shortage of therapists who can provide “depression-specific” short-term therapies competently, and therapist competence accounts for 30% of variance in outcomes and correlates with lower dropout rates. Whenever available, then, aggressive combined treatment provided by a psychiatrist or as “split treatment” is recommended for severe, chronic depression.

Other Somatic Approaches Electroconvulsive Therapy When depression is refractory to pharmacotherapeutic and psychotherapeutic strategies, it is important to note that ECT is proven to be both safe and effective. It is a logical next step to consider in the management of a treatment-refractory depressed patient (Crismon et al. 1999). About 60% of pharmacotherapy-resistant depressed patients respond to ECT, but the relapse rate is very high (50%–70% within a year) despite maintenance AD medications. However, recent studies indicate that maintenance treatment with either ECT plus antidepressants (Gagne et al. 2000) or a combination of nortriptyline and lithium (Sackeim et al. 2001) markedly decrease relapse rates and are distinctly more effective than an antidepressant alone (see Chapter 13 of this book). Transcranial Magnetic Stimulation Transcranial magnetic stimulation (TMS) is an investigational tool that holds promise for the effective treatment of severe depression, including ❉

101 ❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

drug treatment and ECT-resistant depression. Preliminary data suggest it may be as effective as ECT with the important advantage of a lower relapse rate (Grunhaus et al. 2000). This is discussed in greater detail in Chapter 13. Phototherapy Exposure to 2,000–10,000 lux of full-spectrum light for 1–2 hours per day has been proven to be effective in seasonal affective disordered patients and as an augmenting agent in nonseasonal depressed patients (American Psychiatric Association 2000; Kripke et al. 1992). Psychosurgery As neurosurgical techniques have become more refined and accurate, psychosurgery has once again become a viable treatment option of last resort for a number of intractable disorders. For instance, Spangler and colleagues (1996) report that, with magnetic resonance image–guided stereotaxic cingulotomy, one-third of 34 patients demonstrated significant improvement, and another one-third showed some benefit (see also Chapter 13). Vagal Nerve Stimulation Vagal nerve stimulation (VNS) by a cardiac pacemaker-like device is a procedure that is used for patients with intractable seizure disorder. VNS has been shown in open trials to be effective in 40%–50% of patients with severe depression who had been resistant to treatment with several ADs and to ETC. Significantly, responders maintained their symptom and functional improvements over 4- to 9-month follow-up (Rush et al. 2000). Alternative Medicine Therapies Although increasingly popular with patients, complementary and alternative therapies such as exercise, herbal therapy, acupuncture, and relaxation therapy are poorly studied and lack evidence for efficacy in depression (Ernst et al. 1999). For instance, SAMe and St. John’s Wort are widely used and may be useful in some cases (Pies 2000), but the first large controlled U.S. study of St. John’s Wort in major depression failed to show benefit over placebo (Shelton et al. 2001a). Clinical Vignette Mr. J is a 40-year-old white married male with a history of “lifelong depression” that worsened 10 years ago. A college-educated successful business-

❉

102 ❉

❉

The Difficult-to-Treat Patient With Depression

❉

man, he failed to buy out his brother’s share in the business after his father reneged on the deal. He left the business and was unemployed. Mr. J became depressed, with sleep disturbance, the need to lie on the couch all day because of fatigue, anergia, decreased concentration, poor memory, increased appetite, and a plan for suicide (revolver or car “accident”). He felt hopeless and was hospitalized in April, June, and August of 1992. He was treated sequentially with nortriptyline, trazodone, and bupropion without effect. Adjunctive alprazolam was also ineffective. Fluoxetine produced a good response but was discontinued because of a rash. Intensive psychodynamic therapy was of no benefit. Thirteen bilateral ECTs produced an initial positive response (increased energy and function) and a brief period of hypomania. He became depressed again despite being maintained on phenelzine in increasing doses up to 90 mg/day and flurazepam 30 mg hs for severe insomnia. Because he was “just not getting better,” he was hospitalized in another city for consultation. Phenelzine was tapered and discontinued, and flurazepam was replaced by temazepam. He was evaluated, educated about the nature of his disorder, and discharged with the recommendation that he have a trial of fluoxetine (or another SSRI), continuing individual psychotherapy to deal with his unexpressed anger and family therapy to directly address issues pertaining to his father. Given the history of a rash with fluoxetine and poor response to monotherapy, Mr. J was put on paroxetine 20 mg and nortriptyline 25 mg. There was distinct improvement but with frequent breakthrough symptoms. This led to steadily increasing doses of paroxetine to 60 mg and nortriptyline to 50 mg qhs. Because there was a partial positive effect with no intervening side effects, paroxetine was pushed beyond the usual recommended dose of 50 mg. Blood levels for nortriptyline (which had been significantly elevated by the effect of paroxetine on the hepatic enzymes) were monitored as the dose was raised. These were within the therapeutic window on 50 mg/day. Depression improved and breakthrough symptoms were infrequent and appeared only under clear psychosocial stress. Weekly psychotherapy focused on unresolved grief and anger regarding the loss of his family, business, income, and status. Symptoms of depression appeared to worsen each time there was a confrontation regarding these issues, leaving him feeling angry, impotent, and depressed. Psychotherapy was tapered down to once a month. A year later, his depression became manageable and he got a good job with the state government but continued to complain about a lack of drive, decreased concentration, and memory problems. Mr. J sued his father and brother for his share of the business and was retraumatized. This evoked his history of severe, daily physical abuse by his brother (who also abused the father) and guilt/distress regarding his role in “standing guard” while his brother sexually abused their younger sister. While a combination of medications and psychotherapy helped maintain him free from severe depression and functioning at work, he was

❉

103 ❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

unable to feel the anger or deal with it constructively, being afraid that he would become violently destructive. This left him periodically frustrated and depressed. He was referred to another therapist for eye movement desensitization and reprocessing (EMDR), which, after some initial hesitation, he found helpful. The long struggle with anergic depression and the 4-year court battle strained the marriage. His wife and he seriously considered divorce. They were referred for, and benefited from, marital counseling while he continued on the full dose of medications (paroxetine 60 mg/ day and nortriptyline 50 mg/day) and sporadic individual psychotherapy. After being in the hospital four times in one year, the combination of medications (at a dose higher than recommended) along with individual therapy, marital therapy, and EMDR has enabled him to stay out, change careers, and achieve a fairly high level of function although he continues to complain of concentration and memory problems. These could be secondary to his ADs; however, given his hard-earned relief from depression, he has opted to tolerate these symptoms and continue with his current regime.

Summary Treatment-refractory depressions are common, and sometimes stubborn, but usually respond to vigorous treatment. Moreover, 10% of these patients remit and get better spontaneously every year. In a naturalistic study of 431 depressed patients, 93% had recovered by the 10th year; of those who were unremittingly ill for the first 5 years, 38% recovered in the following 5 years. A shorter duration of illness prior to treatment and being married were associated with recovery. Despite the 5-year history of severe, chronic illness, treatment in this naturalistic study was at a low level, 100 mg imipramine equivalent (Mueller et al. 1996), which is consistent with reports of the undertreatment of depression. Our recommendation, based on more recent studies reviewed in this chapter, is to begin with careful attention to diagnosis. Next, medical concerns should be addressed, particularly low levels of thyroid and folate. This is followed by systematic aggressive treatment—undertreatment is a common cause of poor response—of both the depressive disorder and comorbid conditions (see Figure 4–1). Severe depression is preferentially treated with both medication and psychotherapy from the start. Weekly interpersonal, behavioral, or cognitive psychotherapy for up to 12 weeks is useful, and long-term continuation at a lower frequency is usually necessary. It is imperative that therapy be provided in a competent manner. Medications, too, must be prescribed according to guidelines. Currently, an SSRI is preferred as the initial medication choice. For the 50% of patients who show a subop❉

104 ❉

❉

The Difficult-to-Treat Patient With Depression

❉

timal response, increasing the dose (e.g., from 20 mg/day to 40 mg/day of fluoxetine) is helpful. Doses above recommended maximums are useful if side effects do not interfere. Partial responders can often be converted to complete responders by augmentation with lithium or another AD. Nonresponders would benefit from a switch to another SSRI and then to another class of AD. ECT is a very effective option. It is essential to maintain a supportive therapeutic relationship and a consistent hopeful attitude. Educating the patient and the family about the richness of our armamentarium is important. Utilizing all of the many psychotherapeutic, pharmacological, and somatic treatment options available holds promise for the effective treatment of resistant depression, and very few patients should make it into the dreaded refractory category.

References Agency for Health Care Policy and Research: Depression in Primary Care, Vol 2: Treatment of Major Depression (Clinical Practice Guideline No 5). Rockville, MD, U.S. Department of Health and Human Services, 1993 Akiskal H: Mood disorders: Clinical features, in Comprehensive Textbook of Psychiatry/VI, 6th Edition. Edited by Kaplan HI, Sadock BJ. Baltimore, MD, Williams & Wilkins, 1995, pp 1123–1152 Alpert J, Fava M: Nutrition and depression: the role of folate. Nutr Rev 55:145–149, 1997 American Psychiatric Association: Practice guideline for the treatment of patients with major depressive disorder (revision). Am J Psychiatry 157 (suppl):1–45, 2000 Amsterdam JD, Berwish NJ: High dose tranylcypromine therapy for resistant depression. Pharmacopsychiatry 22:21–25, 1989 Aronson R, Offman HJ, Joffe RT, Naylor CD: Triiodothyronine augmentation in the treatment of refractory depression. Arch Gen Psychiatry 53:842–848, 1996 Blackburn I, Eunson K, Bishop S, et al: A two year naturalistic follow-up of depressed patients treated with CT, pharmacotherapy and a combination of both. J Affect Disord 10:67–75, 1986 Brown R, Bottiglieri T, Colman C: Stop Depression Now. New York, GP Putnam, 1999 Burns D, Nolen-Hoeksema S: Therapeutic empathy and recovery from depression in cognitive-behavioral therapy: a structural equation model. J Consult Clin Psychol 60:441–449, 1992 Coppen A, Bailey J: Enhancement of the antidepressant effect of fluoxetine by folic acid: a randomized placebo controlled trial. J Affect Disord 60:121–130, 2000 Covi L, Lipman R, Derogatis L, et al: Drugs and group psychotherapy in neurotic depression. Am J Psychiatry 131:191–198, 1974

❉

105 ❉

❉

106

❉

The Difficult-to-Treat Psychiatric Patient

❉

❉

❉ The Difficult-to-Treat Patient With Depression

❉ 107 ❉

❉

Figure 4–1. Algorithm for severe depression. a Plus supportive therapy to maintain hope, deal with intercurrent stressors, and increase compliance.

❉

108 ❉

The Difficult-to-Treat Psychiatric Patient

❉ Figure 4–1. Algorithm for severe depression. (continued) a Plus supportive therapy to maintain hope, deal with intercurrent stressors, and increase compliance.

❉

❉

The Difficult-to-Treat Patient With Depression

❉

Crismon M, Trivedi M, Pigott T, et al: The Texas Medication Algorithm Project: major depressive disorder. J Clin Psychiatry 60:142–156, 1999 DeRubeis R, Evans M, Hollon S, et al: How does cognitive therapy work? Cognitive change and symptom change in cognitive therapy and pharmacotherapy for depression. J Consult Clin Psychol 58:862–869, 1990 DeRubeis R, Gelfand L, Tang T, et al: Medication versus cognitive behavioral therapy for severely depressed outpatients. Am J Psychiatry 156:1007–1013, 1999 Dew M, Reynolds C, Houck P, et al: Temporal profiles of the course of depression during treatment. Arch Gen Psychiatry 54:1016–1024, 1997 Dewan M: Adding medication to ongoing psychotherapy. Am J Psychother 46:102– 110, 1992 Dewan M: Are psychiatrists cost-effective? An analysis of integrated versus split treatment. Am J Psychiatry 156:324–326, 1999 Dewan M, Koss M: The clinical impact of the side effects of psychotropic drugs, in The Limits of Biological Treatments for Psychological Distress. Edited by Fisher S, Greenberg R. Hillsdale, NJ, Lawrence Erlbaum, 1989, pp 189–234 DiMascio A, Weissman M, Prusoff B, et al: Differential symptom reduction by drugs and psychotherapy in acute depression. Arch Gen Psychiatry 36:1450–1456, 1979 Elkin I, Shea M, Watkins J, et al: NIMH treatment of depression collaborative research program 1: general effectiveness of treatments. Arch Gen Psychiatry 46:971–982, 1989 Ernst E, Rand J, Stevinson C: Complementary therapies for depression. Arch Gen Psychiatry 55:1026–1032, 1999 Fava G, Savron G, Grandi S, et al: Cognitive-behavioral management of drug resistant major depressive disorder. J Clin Psychiatry 58:278–282, 1997 Fava G, Rafanelli C, Grandi S, et al: Six-year outcome for cognitive behavioral treatment of residual symptoms in major depression. Am J Psychiatry 155:1443–1445, 1998 Fava M, Rosenbaum JF, McGrath PJ, et al: Lithium and tricyclic augmentation of fluoxetine treatment for resistant major depression: a double-blind, controlled study. Am J Psychiatry 151:1372-1374, 1994 Fava M, Borus JS, Alpert JE, et al: Folate, vitamin B12, and homocysteine in major depressive disorder. Am J Psychiatry 154:426–428, 1997 Feighner JP, Herbstein J, Damlouji N: Combined MAOI, TCA and direct stimulant therapy of treatment-resistant depression. J Clin Psychiatry 46:206–209, 1985 Frank E, Kupfer DJ: Axis II personality disorders and personality features in treatmentresistant and refractory depression, in Treatment Strategies for Refractory Depression. Edited by Roose SP, Glassman AH. Washington, DC, American Psychiatric Press, 1990, pp 205–221 Frank E, Kupfer D, Perel J, et al: Three year outcomes for maintenance therapies in recurrent depression. Arch Gen Psychiatry 47:1093–1099, 1990

❉

109 ❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

Frank E, Perel J, Malinger A, et al: Relationship of pharmacologic compliance to longterm prophylaxis in recurrent depression. Psychopharmacology Bulletin 28:231– 235, 1992 Gagne G, Furman M, Carpenter L, et al: Efficacy of continuation ECT and antidepressant drugs compared to long-tern antidepressants alone in depressed patients. Am J Psychiatry 157:1960–1965, 2000 Glassman A, Perel J, Shostak M, et al: Clinical implications of imipramine plasma levels for depressive illness. Arch Gen Psychiatry 34:197–204, 1997 Godfrey P, Toone B, Carney M, et al: Enhancement of recovery from psychiatric illness by methylfolate. Lancet 336:392–395, 1990 Goodwin F, Jamison K: Manic Depressive Illness. New York, NY, Oxford University Press, 1990 Grunhaus L, Dannon P, Schreiber S, et al: Repetitive transcranial magnetic stimulation is as effective as electroconvulsive therapy in the treatment of nondelusional major depressive disorder: an open study. Biol Psychiatry 47:314–324, 2000 Hall RC, Gardner ER, Popkin MK, et al: Unrecognized physical illness prompting psychiatric admission: a prospective study. Am J Psychiatry 138:629–635, 1981 Hardy B, Shulman K, Zucchero C: Gradual discontinuation of lithium augmentation in elderly patients with unipolar depression. J Clin Psychopharmacol 17:22–26, 1997 Hollon S, Fawcett J: Combined medication and psychotherapy, in Treatments of Psychiatric Disorders, 2nd Edition, Vol 1. Gabbard GO, Editor-in-Chief. Washington, DC, American Psychiatric Association, 1995, pp 1221–1236 Jacobson N, Dobson K, Fruzzetti A, et al: Marital therapy as a treatment for depression. J Consult Clin Psychol 59:547–557, 1991 Jarrett R, Schaffer M, McIntyre D, et al: Treatment of atypical depression with cognitive therapy or phenelzine. Arch Gen Psychiatry 56:431–437, 1999 Jarrett R, Kraft D, Doyle J, et al: Preventing recurrent depression using cognitive therapy with and without a continuation phase. Arch Gen Psychiatry 58:381–388, 2001 Joffe RT, Singer W, Levitt AJ, et al: A placebo-controlled comparison of lithium and triiodothyronine augmentation of tricyclic antidepressants in unipolar refractory depression. Arch Gen Psychiatry 50:387–393, 1993 Kasper S: Efficacy of Antidepressants in the treatment of severe depression: the place of mirtazapine. J Clin Psychopharmacology 17 (suppl 1):195–285, 1997 Keller M: Depression in adults. Data presented at the Eighth Annual U.S. Psychiatric Congress, New York city, November 1995 Keller M, McCullough J, Klein D, et al: A comparison of nefazodone, the cognitive behavioral analysis system of psychotherapy, and their combination for the treatment of chronic depression. N Engl J Med 342:1462–1470, 2000 Kendall P, Kipris D, Otto-Salaj L, et al: When clients don’t progress. Cognitive Therapy and Research 16:269–281, 1992

❉

110 ❉

❉

The Difficult-to-Treat Patient With Depression

❉

Kent J: SNaRIs, NaSSAs, and NaRIs: new agents for the treatment of depression. Lancet 355:911–918, 2000 Klein D, Schwartz J, Rose S, et al: Five-year course and outcome of dysthymic disorder: a prospective, naturalistic follow-up study. Am J Psychiatry 157:931–939, 2000 Kocsis, JH, Friedman RA, Markowitz, JC, et al: Maintenance therapy for chronic depression: a controlled clinical trial of desipramine. Arch Gen Psychiatry 53:769– 774, 1996 Kocsis JH, Zisook S, Davidson J: Double-blind comparison of sertraline, imipramine, and placebo in the treatment of dysthymia: psychosocial outcomes. Am J Psychiatry 154:390–395, 1997 Koss E, Shiang K: Research on brief therapy, in Handbook of Psychotherapy and Behavioral Change, 4th Edition. Edited by Bergin AE, Garfield SL. New York, Wiley, 1994, pp 664–700 Kripke DF, Mullaney DJ, Klauber MR, et al: Controlled trial of bright light for nonseasonal major depressive disorders. Biol Psychiatry 31:1119–1134, 1992 Kupfer DJ, Frank E: Maintenance therapy for chronic depression: a controlled clinical trial of desipramine (commentary). Arch Gen Psychiatry 53:775–776, 1996 Kupfer D, Frank E, McEachran A, et al: Delta sleep ratio: biological correlate of early recurrence in unipolar affective disorder. Arch Gen Psychiatry 47:1100–1105, 1990 Levitt A, Joffe R: Folate, B12, and life course of depressive illness. Biol Psychiatry 25:867–872, 1989 Megna J, Dewan M: The role of mirtazapine in the treatment of depression Current Practice of Medicine 2:181–183, 1999 Miller I, Norman W, Keitner G, et al: Cognitive behavioral treatment of depressed inpatients. Behavior Therapy 20:25–47, 1989 Miller M, Wolfson L, Frank E, et al: Using IPT in a combined psychotherapy/medication research protocol with depressed elders. J Psychother Pract Res 7:47–55, 1997 Mintz, J, Mintz L, Arruda M, et al: Treatments of depression and the functional capacity to work. Arch Gen Psychiatry 49:761–768, 1992 Mueller TI, Keller MB, Leon AC, et al: Recovery after 5 years of unremitting major depressive disorder. Arch Gen Psychiatry 53:794–799, 1996 Nemeroff CB, DeVane CL, Pollack BG: Newer antidepressants and the cytochrome P450 system. Am J Psychiatry 153:311–320, 1996 O’Leary K, Beach S: Marital therapy: a viable treatment for depression and marital discord. Am J Psychiatry 47:183–186, 1990 Paykel E, Myers J, Deinelt M, et al: Life events and depression: a controlled study. Arch Gen Psychiatry 21:753–760, 1969 Paykel E, Scott J, Teasdale J, et al: Prevention of relapse in residual depression by cognitive therapy. Arch Gen Psychiatry 56:829–835, 1999

❉

111 ❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

Penninx B, Guralnik J, Ferrucci L, et al: Vitamin B12 deficiency and depression in physically disabled older women. Am J Psychiatry 157:715–721, 2000 Peselow E, Filippi A, Goodnick P, et al: The short and long term efficacy of paroxetine Hcl. Psychopharmacology Bulletin 25:272–276, 1989 Phillips KA, Nierenberg AA: The assessment and treatment of refractory depression. J Clin Psychiatry 55:20–26, 1994 Physicians’ Desk Reference, 55th Edition. Montvale, NJ, Medical Economics, 2001 Pies R: Atypical depression, in Handbook of Clinical Psychopharmacology, 2nd Edition. Edited by Tupin JP, Shader R, Harnett D. Northvale, NJ, Jason Aronson, 1988, pp 329–356 Pies R: Time and the art of psychopharmacology. Harv Rev Psychiatry 5:36–39, 1997 Pies R: Handbook of Essential Psychopharmacology. Washington, DC, American Psychiatric Press, 1998 Pies R: Neuropsychiatric side effects of herbal and over-the-counter antidepressants. J Clin Psychiatry 61:815–820, 2000 Post R: Mood disorders: somatic treatment, in Comprehensive Textbook of Psychiatry/ VI. 6th Edition. Edited by Kaplan HI, Sadock BJ. Baltimore, MD, Williams & Wilkins, 1995, pp 1152–1177 Posternak M, Zimmerman M: Switching versus augmentation: a prospective, naturalistic comparison in depressed treatment-resistant patients. J Clin Psychiatry 62:135–142, 2001 Preskorn SH: Why did Terry fall off the dose-response curve? Journal of Practical Psychiatry and Behavioral Health 2:34–93, 1996 Quitkin FM, McGrath PJ, Stewart JW, et al: Chronological milestones to guide drug change. Arch Gen Psychiatry 53:785–792, 1996 Rehm L, Kaslow N, Rabin N: Cognitive and behavioral targets in a self control therapy program for depression. J Consult Clin Psychol 55:60–67, 1987 Reynolds C, Frank E, Perel J, et al: Nortriptyline and interpersonal psychotherapy as maintenance therapies for recurrent major depression JAMA 281:39–45, 1999 Ribeiro S, Tandon R, Grunhaus L, et al: The DST as a predictor of outcome in depression: a meta-analysis. Am J Psychiatry 150:1618–1629, 1993 Rude S: Relative benefits of assertion or cognitive self-control treatment for depression as a function of proficiency in each domain. J Consult Clin Psychol 54:390–394, 1986 Rush A, George M, Sackeim H, et al: Vagus nerve stimulation for treatment resistant depression: a multicenter study. Biol Psychiatry 47:276–286, 2000 Sackeim H, Haskett R, Mulsant B, et al: Continuation pharmacotherapy in the prevention of relapse following electroconvulsive therapy. JAMA 285:1299–1307, 2001 Scott J: Treatment of chronic depression. N Engl J Med 342:1518–1520, 2000 Shelton RC, Keller M, Gelenberg A, et al: Effectiveness of St. John’s Wort in major depression: a randomized controlled trial. JAMA 285:1978–1986, 2001a

❉

112 ❉

❉

The Difficult-to-Treat Patient With Depression

❉

Shelton RC, Tollefson GD, Tohen M, et al: A novel augmentation strategy for treating resistant major depression. Am J Psychiatry 158:131–134, 2001b Sherbourne CD, Hays RD, Wells KB, et al: Prevalence of comorbid alcohol disorder and consumption in medically ill and depressed patients. Arch Fam Med 2:1142– 50, 1993 Sherbourne CD, Jackson CA, Meredith LS, et al: Prevalence of comorbid anxiety disorders in primary care outpatients. Arch Fam Med 5:27–34, 1996 Spangler WJ, Cosgrove GR, Ballantine HT, et al: Magnetic resonance image–guided stereotactic cingulotomy for intractable psychiatric disease. Neurosurgery 38: 1071–1076, 1996 Stahl S: Essential Psychopharmacology. Cambridge, UK, Cambridge University Press, 1996 Stewart J, Mercier M, Agosti V, et al: Imipramine is effective after unsuccessful cognitive therapy. J Clin Psychopharmacol 13:114–119, 1993 Swann AC, Bowden CL, Morris D, et al: Depression during mania. Arch Gen Psychiatry 54:37–42, 1997 Thase ME: Reeducative therapies, in Treatments of Psychiatric Disorders, 2nd Edition, Vol 1. Gabbard GO, Editor-in-Chief. Washington, DC, American Psychiatric Association, 1995, pp 1169–1204 Thase ME: Psychotherapy of depression. CNS Spectrums 4:62–66, 1999 Thase ME, Kupfer EJ, Frank E, et al: Treatment of imipramine-resistant recurrent depression, II: an open clinical trial of lithium augmentation. J Clin Psychiatry 50:413–417, 1989 Thase ME, Simons A, McGeary J, et al: Relapse after cognitive behavior therapy of depression: potential implications for longer courses of treatment. Am J Psychiatry 149:1046–1052, 1992 Thase ME, Simons A, Reynolds C, et al: Physiological correlates to poor response to cognitive behavior therapy: potential indications for antidepressant pharmacotherapy. Psychopharmacology Bulletin 29:293–301, 1993 Thase ME, Fava M, Halbreich U : A placebo-controlled, randomized clinical trial comparing sertraline and imipramine for the treatment of dysthymia. Arch Gen Psychiatry 53:777–784, 1996 Thase ME, Greenhouse J, Frank E, et al: Treatment of major depression with psychotherapy or psychotherapy-pharmacotherapy combinations. Arch Gen Psychiatry 54:1009–1015, 1997 Thase M, Entsuah R, Rudolph R: Remission rates during treatment with venlafaxine or selective serotonin reuptake inhibitors. Br J Psychiatry 178:234–241, 2001 Tohen M, Hennen J, Zarate C, et al: Two-year syndromal and functional recovery in 219 cases of first-episode major affective disorder with psychotic features. Am J Psychiatry 157:220–228, 2000 Wells KB, Burnam MA, Rogers W, et al. The course of depression in adult outpatients: results from the medical outcomes study. Arch Gen Psychiatry 49:788–794, 1992

❉

113 ❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

Weissman M, Klerman GL, Paykel ES, et al: Treatment effects on the social adjustment of depressed patients. Arch Gen Psychiatry 30:771–778, 1974 Worthington J, Fava M, Agustin C, et al: Consumption of alcohol, nicotine, and caffeine among depressed outpatients. Psychosomatics 37:518–522, 1996

❉

114 ❉

5 The Difficult-to-Treat Patient With Anxiety Disorder Suzanne M. Sutherland, M.D. K. Ranga Krishnan, M.B., Ch.B.

In this chapter, we review the current knowledge concerning treatment of the four anxiety disorders: social phobia, obsessive compulsive disorder, panic disorder, and generalized anxiety disorder. We also discuss problems in treating anxiety disorders and difficulties specific to each of these disorders as well as clinical approaches to understanding and treating those patients who do not respond to standard approaches.

Social Phobia Social phobia, with a 13.3% lifetime prevalence, is the third most common psychiatric disorder in the United States. The onset is typically between 13 and 20 years of age; onset is rare after 25 years (Davidson et al. 1993). The ❉

115 ❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

discrete subtype involves excessive fear of observation or scrutiny in performance situations, such as presenting in a classroom or work setting, giving a toast at a wedding, or playing in a recital. The more common generalized subtype includes experiencing excessive anxiety in a wide variety of social situations (e.g., making phone calls; introducing oneself to a stranger; and participating in business, family, or social parties).

Overview of Treatment Pharmacotherapy Selective serotonin reuptake inhibitors. Controlled trials of selective serotonin reuptake inhibitors (SSRIs) in social phobia have established efficacy for fluvoxamine, sertraline, and paroxetine (Allgulander 1999; Baldwin et al. 1999; Katzelnick et al. 1995; van Vliet et al. 1994), and some have labeled them the “pharmacological treatment of choice.” Paroxetine, the most extensively studied SSRI for social phobia, is also the only one with response rates at or near the 63%–75% response seen in trials with phenelzine (Davidson 1998). However, in the trials reporting response rates to paroxetine of 66%–69%, the placebo rate was higher at 29%–32%, compared with 20% or less in the phenelzine trials. Response rates for the other SSRIs were 40%– 50% for fluvoxamine and 50% for sertraline, with placebo responses of 7%– 23%, and 9%, respectively. Monoamine oxidase inhibitors. Phenelzine, with its well-established efficacy for social phobia, is not recommended as a first-line drug because of its troublesome side effects. Benzodiazepines. There are three main studies using benzodiazepines in the treatment of social phobia, all with high-potency drugs. One doubleblind, placebo-controlled trial with alprazolam, phenelzine, cognitive-behavioral group therapy (CBT group), and placebo showed only a 38% response rate for alprazolam, with a 20% placebo response, and the alprazolam group had a high relapse rate within 2 months of discontinuing treatment (Gelernter et al. 1991). In a larger controlled trial of 75 social phobia subjects, there was a 78% response rate for clonazepam versus 20% for placebo (Davidson et al. 1993). A more recent study comparing clonazepam and CBT group in social phobia also showed a robust response in the clonazepam group, with that group showing an advantage over the CBT group at week 12 (Otto et al. 2000). ❉

116 ❉

❉ The Difficult-to-Treat Patient With Anxiety Disorder

❉

Other medications. Gabapentin was shown to be effective for social phobia in a double-blind study using dose ranges from 600 to 3600 mg/day (Pande et al. 1999). The benefits of gabapentin were most apparent in those subjects with more severe symptoms. Some of the newer antidepressants may also prove to be of value in treating social phobia. Nefazodone has been effective in a clinical case series and a 12-week open trial (Van Ameringen et al. 1999). Reports of venlafaxine treatment in social phobia patients who did not respond to SSRIs show promise at low to moderate doses (Altamura et al. 1999). Buspirone, reported to be effective in two open-label studies, failed to show efficacy in a controlled study at a mean dose of 32 mg/day (Clark and Agras 1991). A more recent small open trial successfully used buspirone as augmentation in patients who had a partial response to SSRIs (Van Ameringen et al. 1996).

β-Blockers. There are no controlled studies supporting the widely held view, based on case reports and open trials, that β-blockers should be used to treat performance anxiety in students and musicians. An early negative report of propranolol (Falloon et al. 1981) was followed by a controlled study with phenelzine and atenolol that showed no benefit for atenolol over placebo in the small subset of patients with discrete (performance) social phobia (Liebowtiz et al. 1992). In spite of this lack of evidence, many patients report that β-blockers are helpful in maintaining a sense of control in performance situations. The reduction in autonomic arousal symptoms (e.g., rapid heart rate or tremor) provides a greater sense of security, thus preventing the positive feedback loop of increasing anxiety in response to the physiological arousal symptoms. Psychotherapy CBT has been shown to be effective for social phobia. Heimberg et al. (1998) reported a 58% response rate for patients treated with CBT group in a controlled study comparing CBT group to phenelzine. Recent literature reviews and meta-analyses have demonstrated efficacy for exposure intervention strategies over cognitive restructuring alone, with suggestive evidence favoring the combination of behavior techniques with cognitive therapy and/or social skills training (Gould et al. 1997; Taylor 1996). More recently, Shear and Beidel (1998) reviewed the literature and concluded that the key factor influencing outcome in CBT therapies was the exposure to feared situations. The studies used different forms of exposure, including gradual, intensive, ❉

117 ❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

in vivo, and imaginal. Other procedures used in conjunction with exposure, such as relaxation training, cognitive restructuring, and social skills training, were not shown to enhance efficacy. Although CBT without exposure has not been shown to be helpful in most trials, cognitive interventions may reduce the negative critical thoughts that feed anxiety in social situations, thus increasing the likelihood of exposure to feared situations. Relaxation techniques, on the other hand, may be counterproductive if applied successfully in feared situations; if attention is directed away from a feared object or situation, habituation will not occur. Although medication provides a quicker onset of action, data suggest better maintenance of treatment gains with CBT. Most practitioners, however, unlike in these studies, would tend to keep a patient on medication for an extended period. Treatment periods of at least 1 year may maintain improvement and reduce the rate of subsequent relapse (Davidson 1998).

Factors Associated With Lack of Response Therapeutic Alliance The patient with social phobia has strong expectations of not being liked or respected or even of being considered worthy of acknowledgment. For many, there is also a fear or distrust of any suggestion of intimacy. It is helpful for the clinician to maintain an open and respectful stance. Fear of authority figures and of criticism are common and can affect the relationship with the clinician. Any perception on the part of the patient of lack of concern, of criticism, or of dismissal may cause a break in the therapeutic alliance. Conversely, excessive warmth and an overly complimentary approach may also hurt this often fragile alliance. Anger may also be a hidden or an overt factor. The person with social phobia has exaggerated expectations of rejection or criticism. The perceived slights are internalized and the person may feel hurt or may respond with anger at the injustice. This righteous anger may surface in therapy or may cause the patient to discontinue treatment. An attempt to process the anger with the patient may be felt as further criticism. Acknowledgment of the anger and then a return to the standard cognitive approach in the therapy session is the judicious approach. Comorbidity Patients with social phobia often present for treatment because of comorbid symptoms of depression. Targeting the dose to the response of the depressive symptoms, which should resolve earlier, and then following the patient and ❉

118 ❉

❉ The Difficult-to-Treat Patient With Anxiety Disorder

❉

assessing changes in social anxiety symptoms will prevent excessive dosing. Again, it is important to encourage the patient to engage in exposure to feared situations. Doing imaginal exposure in the office can be helpful, and role play increases the chance of success outside of the office. Many patients will also be more amenable to entering into formal CBT group or social effectiveness therapy (SET) once some of the depressive and anxiety symptoms have resolved. If comorbid depression does not respond to SSRIs, CBT should be added. The next step should include standard augmentation of antidepressants with lithium, thyroxine, stimulants, or low-dose tricyclic antidepressants (TCAs). If the depression remains, a trial of a monoamine oxidase inhibitor is indicated (see “Alternate medications” later in this section). Substance abuse has been associated with poor outcome in drug trials. A long period of alcohol abuse or other substance abuse may have intensified the anxiety symptoms, thus making the disorder more difficult to treat. If the abuse is addressed directly in treatment, especially if comorbid depressive symptoms are also dealt with, there is a better chance of success. Using gabapentin along with an SSRI is recommended, especially for alcohol abusers, as it can reduce craving while also treating social phobia. Judicious use of a benzodiazepine may sometimes be necessary to achieve progress. Group psychotherapy should also be a part of the treatment. Patients with comorbid anxiety disorders, particularly panic disorder, may be particularly challenging. For those with prominent physiological symptoms who are willing to take medication, clonazepam should be used initially and other medications added as indicated for the symptom complex. Thorough psychoeducation and readily available support are necessary for these patients who are generally phobic of medications. The use of a medicine may not be possible initially, and much individual support may be needed to engage the patient in treatment. Comorbid personality disorders, specifically borderline or passivedependent, have been cited as negative predictors in drug trials. Addressing traits that specifically interfere with the treatment of social phobia can increase the chances for success. Avoidant personality disorder may not be a confounding factor, as it responds to treatments for social phobia in many medication trials.

Treatment Recommendations Treatment of the patient with social phobia starts with a good assessment. The interview process may be difficult for some patients because of the ❉

119 ❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

nature of the symptoms inherent in the disorder. Even those patients who are more comfortable may find it difficult to convey specific symptoms. Use of a rating scale such as the Liebowitz Social Anxiety Scale or the Duke Brief Social Phobia Scale may help to elicit symptoms and to objectify and quantify the disorder. This may be particularly helpful in a patient who has trouble seeing the social anxiety symptoms as a treatable illness and believes that the various behaviors are “just who I am.” An important consideration is that the fear of scrutiny, as well as typical cognitions such as “That doctor probably thought I was an idiot” or “I knew he noticed how much I was blushing—I’m too embarrassed to go back,” may prevent the person from returning for needed help if the initial approach does not alleviate some of the anxiety. The clinician must take care to be reassuring and be careful not to be abrupt or dismissive. Psychoeducation is an important part of the initial treatment of all anxiety disorders and is particularly helpful for patients whose sense of embarrassment or humiliation may be connected to a belief that their problems reflect a unique defect. Understanding that many people suffer from similar fears can cause an immediate reduction in general anxiety and embarrassment. Understanding social phobia as a treatable illness also provides hope and motivation for further treatment in many individuals who have been imprisoned in their wall of anxiety. Once the patient has been educated about the different choices of treatment, the formal treatment process is chosen. The initial treatment is generally determined by the patient, indirectly or directly. If the patient presents to a psychiatrist initially, there is greater likelihood of a medication being tried, although those who are trained in specific CBT techniques for social phobia will likely offer a choice. Presentation to a psychologist is more likely to result in a CBT trial. Psychotherapy The patient can be entered into individual or group therapy. CBT group is especially helpful in social phobia as it provides an opportunity for exposure practice in the presence of the therapist. Feedback from the group as well as from the therapist can help in identifying distorted negative cognitions and developing new, more adaptive ones. Simulated exposure situations should be individualized and their difficulty graded by modifying the behavior of the group (audience) or simulated potential romantic partner or authority figure. Obviously, persuading a socially phobic individual to engage in a series ❉

120 ❉

❉ The Difficult-to-Treat Patient With Anxiety Disorder

❉

of group sessions can be challenging. We have found that many of these patients have enough insight to realize that this is what they need and can be persuaded on this basis. Reassuring them about the shared fears and the safety and confidentially of the setting is also helpful. For those who are unwilling to engage in CBT group, individual CBT is appropriate. This is especially useful in the discrete subtype of social phobia. Distorted cognitions can be analyzed and reframed, and appropriate exposure situations can be planned. At times the therapist may need to be present in these situations, at least initially, as the patients’ ingrained avoidance behavior may make it impossible for them to engage in exposure without the safety of the trusted therapist. SET is useful for patients who have identifiable social skills deficits. This is a formalized CBT program that involves exposure to feared situations but also uses social skills training (Turner et al. 1994). Rather than focusing on cognitive errors, as is done in CBT group, it emphasizes development of social skills. The skills training component is done in weekly group sessions, and weekly individual sessions focus on each individual’s specific fears. Typically treatment courses for CBT group and SET are 12–16 weeks with one to two sessions each week. Once patients have improved social skills, exposure therapy will be less anxiety provoking and will result in more positive reinforcement. Pharmacotherapy Medication may be the treatment of choice for many patients and may be the only choice available, as there is a lack of clinicians skilled in psychotherapy strategies effective for social phobia. Our first choice of medication is an SSRI. The consistently good response in controlled trials, reports of improved functioning in social and work arenas of life, and our own clinical experience lead to a choice of paroxetine as a first-line treatment for social phobia. Unlike patients with panic disorder, there is no initial exacerbation of symptoms, so starting at 20 mg from the beginning is possible and generally adequate. Social phobia does not respond quickly to SSRIs, and the clinician should not increase the dose unless there has been an adequate trial of 10–12 weeks. Clonazepam is appropriate for those who are intolerant of SSRIs, are without comorbid depression, and have no history of substance dependence. Substance abuse is a relative contraindication; sometimes a patient truly has been self-medicating in order to function and, with careful monitoring, can benefit from this drug. There may be a need for a medication that can act quickly, in which case clonazepam alone or in conjunction with an ❉

121 ❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

antidepressant is most likely to be successful. Patients often notice differences after the first week or two of taking clonazepam and generally tolerate the medication without extreme sedation or incoordination. An appropriate initial dose is 0.5 mg at bedtime, which can be increased to twice a day as soon as it is apparent that sedation is not a problem. The target dose is quite variable, from 0.75 to 6 mg/day in bid dosing. Over time, patients tend to need smaller doses, and some eventually can do well without medication. A slow taper over many weeks should reduce rebound anxiety. Gabapentin is an alternate choice. The response is slower than with clonazepam, but the benefits of a lack of addiction potential, no potentially dangerous drug interactions, minimal side effects, and possible reduction in alcohol craving, make it a contender for an initial drug choice. Inadequate response to medication. As noted previously, social phobia tends to respond slowly to treatment with SSRIs. If there is some response, we recommend continuing at standard antidepressant doses for 12–16 weeks. A dose increase at this point may help some patients. However, if there is significant improvement at this stage, an additional period of weeks to several months may result in further gains without the need for higher dosages or different medication choices. Social phobia tends to be chronic, and most patients will have experienced symptoms for years and tend to be accepting of slow response. In the initial treatment phase, the clinician should educate the patient to expect a gradual response while encouraging exposure to feared situations as soon a reduced level of anxiety is apparent. Paroxetine should be tried as an alternative to any of the other SSRIs. Adding a benzodiazepine or other anxiolytic early in treatment can minimize the exacerbation of anxiety and speed anxiolysis during the initial therapeutic lag period. This can also provide additional benefit throughout the treatment period for the many patients who continue to have some anxiety symptoms. Clonazepam is the clear first choice given its proven efficacy in social phobia. Generally, doses range from 0.5 to 2.5 mg/day. A second choice is gabapentin, 900–3600 mg/day. A third alternative is buspirone at doses of 30–60 mg/day. Higher baseline levels of sympathetic arousal may be a negative predictor of outcome (Davidson 1998), and addition of a β-blocker may be helpful. β-Blockers can also be used for specific physiological symptoms such as tremor and blushing. Alternate medications. Venlafaxine, shown to be effective for some social phobia patients who do not respond to SSRIs, is our first choice for a substi❉

122 ❉

❉ The Difficult-to-Treat Patient With Anxiety Disorder

❉

tute antidepressant. Nefazodone or bupropion at standard doses are alternatives to an SSRI and can also be combined with clonazepam, gabapentin, or even buspirone. We have found the combination of bupropion and clonazepam to be effective for a number of patients. If the above drug choices still provide a suboptimal response and CBT has failed or is unavailable, a trial of an MAOI, preferably phenelzine, is recommended. Alternative Approaches Computerized assessment has been used in some clinical trials with positive response from socially anxious patients. Patients may feel safe interacting with a machine that cannot watch and make judgments about them. This assessment should prove to be particularly helpful in the future as a therapy tool for these patients. Clinical Vignette Ms. M, a 58-year-old married woman with a lifelong history of severe social anxiety, presented with complaints of difficulty interacting with residents of her new community. She had a history of severe performance anxiety and avoided any situation in which she would be the center of attention. She believed that people only pretended to like her and voiced their true dislike and criticism behind her back. On presentation, she described frequent periods of anhedonia, low energy, loss of appetite, and restless sleep with feelings of hopelessness and despair. She had experienced similar episodes (lasting 2–3 days) many times over the years, but the episodes were beginning to last longer and come more frequently. Throughout the initial evaluation, she appeared uncomfortable and was apologetic but became somewhat agitated and defensive at times. Initial sessions were marked by the patient’s reluctance to take “addictive medicines” or to commit to long-term medication. She was unable to participate in cognitive or behavioral therapy, and asserted that her mind just went blank in any anxiety situation. Also, she often wished to end sessions early because of her extreme discomfort with attention being focused on her. Psychotherapy was not profitable because of the extreme level of anxiety that prevented her from engaging. She did reluctantly agreed to try paroxetine, but, even at the low initial dose, was unable to tolerate it because of stomach distress. Ms. M then agreed to clonazepam and, at a dose of 0.5 mg/day, quickly noticed improvement in her periods of extreme depressive symptoms. She also improved in her ability to recognize her distorted thoughts and quickly became skilled at identifying irrational thoughts and substituting more

❉

123 ❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

accurate ones. At 1 mg/day of clonazepam, the patient was able to do a very successful presentation to the community board of directors and basked in the positive attention received from neighbors following the meeting. After 4 months of treatment, the patient pronounced herself cured, tapered off the clonazepam, and stopped therapy. Two months later she reported continued success with using the cognitive therapy skills she had acquired. She no longer experienced anxiety or depressive episodes. Again, 1 year later she reported continued good health with a very active social life. This patient had difficulty tolerating standard drug treatments and was initially unable to engage in CBT. However, the ability to tolerate clonazepam at a low dose for a few weeks caused enough improvement in her anxiety that she was able to expose herself to the feared process of revealing her irrational cognitions and to learn how to change her automatic thoughts. Treatment also allowed her to expose herself to feared situations and to have a healing experience in which she was able to accept the positive attention of others.

Obsessive-Compulsive Disorder Obsessive-compulsive disorder (OCD), with an estimated lifetime prevalence of 2.5%, generally begins in adolescence or early adulthood. It is a chronic illness that has a waxing and waning course, has a high comorbidity with other anxiety and mood disorders, and can be disabling. Some OCD patients also have comorbid psychotic symptoms, which include delusions, hallucinations, and/or thought disorder.

Overview of Treatment The two standard accepted treatments for OCD are pharmacological treatment with serotonin reuptake inhibitors (SRIs) and behavioral treatment. There are few data comparing the two forms of treatment, but both have shown efficacy and there is some evidence that combining the treatments may be more effective, at least in severely ill patients with depression or with predominant obsessions (Hohagen 1999). In clinical trials a decrease of 25%–35% of the Yale-Brown Obsessive-Compulsive Scale (Y-BOCS) is typically considered a categorical treatment response, with very few patients becoming symptom free. Pharmacotherapy The SRIs, which include the SSRIs and clomipramine, are the first-line choice for OCD treatment. Clomipramine has been used in many clinical tri❉

124 ❉

❉ The Difficult-to-Treat Patient With Anxiety Disorder

❉

als as the standard by which newer treatments are measured and remains an alternative primary agent. The other TCAs and MAOIs have shown little efficacy in OCD. All of the SSRI antidepressants are effective in the treatment of OCD. The first SSRI trial reported was with fluoxetine and demonstrated efficacy at doses of 20 mg/day, 40 mg/day, and 60 mg/day, with a tendency for greater efficacy with increased doses (Tollefson et al. 1994). A fixed-dose trial of paroxetine showed no effect at 20 mg/day but did show efficacy at 40 mg/day and 60 mg/day (Wheadon et al. 1993), and a flexible dose study found 40 mg/day of paroxetine to be effective and comparable with clomipramine (Zohar and Judge 1996). Sertraline was studied in a large, multicenter, fixed-dose trial and found to be effective at the lowest dose of 50 mg/day and the highest of 200 mg/day (Greist et al. 1995). Another multicenter trial demonstrated efficacy for sertraline at a mean dose of 129 mg/day and reported an impressive 51% reduction in Y-BOCS with sertraline compared with 43% with 90 mg/day of clomipramine (Bisserbe et al. 1997). Studies have reported efficacy for fluvoxamine at doses of 200–250 mg/day (Goodman et al. 1998) and citalopram at 20 mg/day, 40 mg/day, and 60 mg/day (Montgomery 1998). Behavioral Therapy Exposure and response prevention (ERP), a form of behavioral therapy widely used in OCD, significantly reduces OCD symptoms in many controlled studies and is equivalent in efficacy to medication treatment (Christensen et al. 1987; Cox et al. 1993). The greater efficacy of behavioral therapy in the long term was supported in a more recent review (Foa and Kozak 1996). Controlled studies of ERP have reported improvement of as much as 60%–85%, with follow-up studies reporting maintenance of gains for as long as 6 years (O’Sullivan et al. 1991) and relapse rates of about 25% across a spectrum of 3 months to 6 years following treatment (Hiss et al. 1994). It should be noted, however, that the majority of the controlled studies that have validated ERP as an effective treatment have included patients with cleaning and checking rituals as their primary problem. Behavior therapy is most effective for the compulsive behaviors and rituals with a secondary effect on the associated obsessions.

Factors Associated With Lack of Response In spite of advances in the treatment of OCD, between 40% and 60% of patients are nonresponders in trials of SRIs, and very few “responders” become ❉

125 ❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

asymptomatic (Goodman et al. 1998). Many patients refuse behavioral therapy or are nonresponsive to it, and skilled behavior therapists are not always readily available to those who would participate. Inaccurate or incomplete diagnosis, comorbidity with other psychiatric or medical conditions, poor compliance, and inappropriate application of behavioral therapy all contribute to poor response. Perhaps 20%–25% of patients do not initially acknowledge the unreasonableness of their symptoms, a factor identified as limiting response to treatment. In contrast, many are reluctant to talk about their symptoms because of embarrassment or fear that others will think they are “crazy.” As the treatment progresses and the therapeutic alliance strengthens, the OCD patient may be more willing to reveal additional symptoms. Doing a thorough assessment when progress appears to have stalled or plateaued can be helped by using a measurement tool such as Y-BOCS, which also helps to depersonalize the symptoms. If a reassessment of symptoms reveals that compulsive symptoms predominate, focusing on behavioral approaches is indicated. If the core symptomatology is the obsessions, the focus should be on enhancing medication treatment. Comorbidity Depression is the most common coexisting disorder. Although response to OCD with medication is generally independent of coexistent depressive symptoms, patients who have opted for a course of behavioral therapy and have refused medication may progress slowly if there is persistent depression. An SSRI should be the drug of choice, but for those who do not respond or are intolerant of SSRIs, other antidepressants can be used. Cognitive therapy or interpersonal therapy can also be utilized to address the depressive symptoms. A coexisting personality disorder may predict a poorer treatment response. Baer and colleagues (1992) found in a controlled trial of clomipramine, that subjects with schizotypal, borderline, and avoidant personality disorders and those with more than one personality disorder had poorer outcomes. When schizotypal, schizoid, or paranoid personality disorder is present, an antipsychotic should be added. Avoidant personality disorder should be treated with clonazepam or gabapentin. Patients with borderline personality disorder can benefit from mood stabilizers and/or antipsychotics. These drugs should be combined with psychotherapy appropriate to the comorbid disorder. Other disorders that are likely to be a factor in the difficult-to-treat OCD ❉

126 ❉

❉ The Difficult-to-Treat Patient With Anxiety Disorder

❉

patient are the schizophrenic spectrum disorders, the impulsive cluster disorders, and other anxiety disorders. Treatment with a neuroleptic and/or mood stabilizer is indicated for patients with impulse cluster disorders and schizophrenia spectrum disorders. An atypical agent should be used first, beginning at low doses and gradually increasing over weeks to months. If unsuccessful, a typical agent can be tried, always in conjunction with a serotonergic drug.

Treatment Recommendations Initial assessment should identify the particular obsessions and compulsions that are most concerning. Education about the disorder and use of the YBOCS will help to objectify the disease and increase the chance of disclosure. If obsessions are predominant, beginning with medication is indicated. For compulsions, ERP is preferable. Pharmacotherapy Patients can respond differently to the various SSRIs. At least three should be tried before declaring a patient refractory to these agents. Trials at the target dose should last at least 10–12 weeks. It is also important to keep in mind that continual improvement is often noted over many months. Therefore, if a given treatment has shown a partial response, staying with that same treatment longer is indicated. The TCA clomipramine was the first medication proven effective in the treatment of OCD, but because of its high side-effect profile, the SSRIs are now considered the first-line medication treatment. Clomipramine remains an alternative as a primary medication for those patients who are resistant to or cannot tolerate the side effects of SSRIs. It can also be used to augment an SSRI, taking care to monitor blood levels as the SSRI may increase its levels. Many clinicians have been reluctant to use antipsychotic medications in OCD in spite of evidence of their usefulness. There have been reports of worsening of OCD symptoms during monotherapy with risperidone and clozapine (Levkovitch et al. 1995; Patel and Tanden 1993) along with reports of the usefulness of risperidone as adjunctive treatment for OCD in SSRI-refractory patients (McDougle et al. 1995). Baker and colleagues (1996) reported no exacerbation of OCD symptoms in a controlled trial of patients with schizophrenia who were given olanzapine at doses of 1 mg/day or 10 mg/day. Pato (1998) added olanzapine to an SSRI in treating six patients with delusional/psychotic OCD and reported some improve❉

127 ❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

ment in the Y-BOCS score, with obsessions improving more than compulsions. A trial of olanzapine or risperidone is warranted for patients refractory to SRIs, especially those with clear psychotic symptoms, and with coexisting tics and Tourette syndrome. Low doses, beginning with 0.25 mg/day or 0.5 mg/day for risperidone and 2.5 mg/day for olanzapine should be used and can be increased slowly over many weeks as tolerated. Clonazepam is another augmenting agent that can be used alone or with an antipsychotic. Although other benzodiazepines have received mixed reports, clonazepam was found to be effective in augmenting SRIs in OCD patients (Hewlett 1993). There is considerable evidence that trazodone has anti-obsessive properties and that trazodone may be particularly helpful in reducing SSRI side effects such as nausea, gastrointestinal distress, weight gain, and sexual dysfunction, in addition to its well known efficacy in reducing sleep disturbance (Marazziti et al. 1999). Two other augmenting agents that have not proven to be beneficial in controlled trials are lithium and buspirone. Anecdotal evidence combined with our clinical experience nevertheless leads us to recommend these as possible second-line agents in difficult-to-treat patients. Psychotherapy ERP consists of exposure to a feared activity, situation, or thought and then prevention of the normal compulsive activity that is a typical response. The therapist assists the patient with the exposure and also instructs the patient in continuation of the behavior therapy practice outside the office. The most important factor in predicting positive outcome is compliance with ERP. A total of 15–30 hours is necessary for an adequate trial (Baer et al. 1992). The usefulness of cognitive therapy alone for OCD has been questioned. A meta-analysis of 15 published studies of CT found no additional beneficial effect (James and Blackburn 1995). Existing research data and clinical experience support the use of both SRI treatment and behavior therapy as first-line treatments in OCD, and most clinicians offer both when initiating treatment. Although approximately 25% of OCD patients refuse to comply with ERP initially, 12% dropout of treatment, and 20%–30% do not respond to ERP (Hohagen 1999), success generally depends on ERP being included prior to completion of treatment. Response to the standard medication treatments for OCD is typically in the range of 40%–60%. Existing data suggest that 60%–70% of patients willing to engage in ERP will respond and that relapse in the long term may ❉

128 ❉

❉ The Difficult-to-Treat Patient With Anxiety Disorder

❉

be lower with ERP than with medication alone. The clinician must not lose sight of the need for behavioral therapy even in those patients who have rejected it initially. Each patient should be encouraged to engage in this additional therapy as an adjunct to medications at every opportunity. An initial partial response to medication may allow the patient to be more accepting of this important part of the treatment. Additional Psychotherapy Strategies Salkovskis and colleagues have developed a cognitive approach to obsessional thoughts that has been shown in one study to equal and, in some areas, exceed ERP in efficacy (Van Oppen et al. 1995). This method helps distinguish between anxiety-producing thoughts and neutralizing thoughts. Patients are exposed to their anxiety-producing thoughts either by mental initiation or by a “loop tape,” with thoughts recorded in the patient’s own voice. They are then not allowed to avoid or respond with neutralizing thoughts. There is increasing interest in the use of multimodal therapy, which may include cognitive and interpersonal therapy techniques along with the standard ERP and which has been found effective in the treatment of outpatients and severely ill inpatients with OCD (Hohagen 1999). OCD patients with depression may benefit especially from this multimodal approach. Novel and Experimental Treatments For patients who have difficulty tolerating side effects of oral medications, intravenous clomipramine is an option. Although considered investigational, there are several cases and open trials reported. Recently, Koran and colleagues (1997) reported benefit early in treatment with this method. Patients are generally able to convert to oral medication after initial response. Novel medication interventions include immunomodulatory therapies in children with autoimmune triggers and agents affecting second messenger systems. Psychosurgery is available to treatment-refractory OCD patients in specialized centers. The gamma knife is a noninvasive and promising tool. Transcranial magnetic stimulation, now being tested in depression and OCD, may prove helpful for refractory patients. Another noninvasive technique of promise is kundalini yoga, which compared favorably to a control treatment of relaxation response plus mindfulness meditation (ShannahoffKhalsa et al. 1999). The following is an example encountered by one of the authors (S.M.S.) of a difficult-to-treat OCD patient. ❉

129 ❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

Clinical Vignette Ms. V, a 26-year old married woman, presented to our clinic with her husband of 3 years and her mother, both of whom had been very involved in her treatment. She had been diagnosed with OCD 2 years earlier but remained severely ill in spite of treatment with medication and behavioral therapy. The patient was currently on a daily medication regimen of fluvoxamine 150 mg, bupropion 150 mg, methylphenidate 40 mg, and clonazepam 2 mg. Competent CBT was being provided at a local anxiety disorder clinic. A referral letter stressed that the marriage would end without some early improvement in the patient’s symptoms, including her sexual dysfunction. The patient had great difficulty describing her symptoms and spoke very slowly and haltingly, with frequent pauses. There was no thought blocking, poor reality testing, or psychomotor retardation. With the help of family, I was able to elicit symptoms of excessive cleaning and attention to hygiene and estimated that these symptoms took up at least 10–12 hours daily. Before the end of the initial evaluation, I asked to meet separately with the couple and explored their concerns about their sexual relationship. The husband expressed his frustration and his concerns that the medicines may be contributing to this problem. We decided to make this issue a focus of treatment concurrently with the OCD symptoms. I also requested that in future sessions the couple return without the patient’s mother. With the hope for good response to the continuing ERP therapy, we elected to direct medication toward treating the mild depressive symptoms and longstanding lack of libido. Bupropion was increased to 300 mg/day and fluvoxamine decreased to 100 mg/day. Within weeks the patient had brightened considerably and was more fluent in her speech but had no improvement in libido. We reduced the fluvoxamine further to 50 mg and increased bupropion to 450 mg/day. Methylphenidate was also tapered and discontinued. The patient soon reported some increase in libido but worsening mood. She continued to be more fluent in her speech, and information concerning her OCD symptoms was more readily attainable. There was little noticeable improvement in compulsions despite months of ERP therapy. Fluvoxamine was discontinued, and clomipramine was initiated and slowly increased over the next several months to a level of 75 mg/day. At this point, some of the OCD behavior had improved, with the patient reporting as little as 6–8 hours each day of OCD symptoms. The sexual problems remained as well as periods of depressed mood. Nefazodone was added, increasing to a level of 400 mg/day. There was no change in OCD symptoms over the next 3 months, and the patient began to be less fluent in her speech. A change back to fluvoxamine to a dose of 100 mg/day resulted in resolution of some depressive symptoms. About 8 months into treatment, the couple left for the summer. Upon

❉

130 ❉

❉ The Difficult-to-Treat Patient With Anxiety Disorder

❉

their return they reported that the patient’s OCD symptoms had been less problematic away from home. They had engaged in sex therapy upon their return, and the patient had resumed ERP. Within weeks, the patient reported an increase in depressed mood and a return of OCD symptoms that were worse than at the time of the couple’s departure for the summer. Olanzapine was initiated at a dose of 2.5 mg/day. Two weeks later the patient reported improved mood and less obsessional checking. With an increase of olanzapine to 2.5 mg twice a day, definite improvement was noted in task efficiency, and the patient’s speech was more fluent and her affect brighter. The patient continues to slowly improve in terms of OCD symptoms and continues ERP therapy. Her medications are fluvoxamine 100 mg/day, bupropion 450 mg/day, clomipramine 75 mg/day, clonazepam 2 mg/day, and olanzapine 5 mg/day. The couple also continues in sex therapy and reports some progress. The husband expresses less frustration and remains committed to the marriage. Obsessional slowness is difficult to treat, and this case also includes the complicating factor of the family’s participation in the patient’s OCD behaviors. Even though there were no overt psychotic symptoms present in this case, the addition of the atypical antipsychotic olanzapine appeared to herald a turning point in this patient’s treatment.

Generalized Anxiety Disorder The lifetime prevalence for generalized anxiety disorder is 5.1%, with 90% of these subjects having at least one other lifetime psychiatric diagnosis, most commonly depression. It tends to be a chronic illness, with most patients remaining symptomatic for much of their lives (Rickels and Schweizer 1997). Generalized anxiety disorder is defined as excessive anxiety and worry about a number of events or activities, accompanied by physiological symptoms. This disorder usually occurs before the onset of major depression and is associated with roughly the same amount of role impairment by itself as is major depression without generalized anxiety disorder (Kessler et al. 1999).

Overview of Treatment People with pure generalized anxiety disorder rarely seek treatment, and there is as yet no generally agreed upon treatment approach. Benzodiazepines, antidepressants, and buspirone have all been purported to be effective drug treatments for generalized anxiety disorder, and the benzodiazepines remain the treatment choice for many physicians (Rickels and Schweizer 1997). Low-dose antipsychotic agents, anticonvulsants, and the ❉

131 ❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

antihistamine hydroxyzine have also been used for their anxiolytic properties. Various psychotherapy techniques have been shown to be useful, including nondirective psychotherapy, CBT, and specifically targeted anxiety management training. Pharmacotherapy Benzodiazepines, the most widely prescribed and best-studied anxiolytic agents, have been the mainstay of treatment for both acute and generalized anxiety over the last several decades. There is a lack of adequate long-term and follow-up studies, but benzodiazepines are generally accepted as invaluable short-term treatment, assuaging anxiety and tension, lessening physical symptoms, and inducing a state of well-being (Lader 1994). For chronic anxiety, however, many physicians are reluctant to prescribe benzodiazepines because of their perceived “addictive” potential. Psychiatric and psychopharmacological experts are more positive than general practitioners and the general public regarding the risk/benefit ratio of benzodiazepines and see physical dependence as neither a common problem nor a serious one (Balter et al. 1993). Benzodiazepines may be less effective in treating the psychic symptoms of anxiety, including irritability and the core symptom of apprehensive worry, than the somatic/autonomic symptoms (Connor and Davidson 1998). Some studies (e.g., Hoehn-Saric and McLeod 1988) have shown worsening of irritability with high-potency benzodiazepines such as alprazolam and better response of psychic symptoms to drugs such as imipramine. With increased emphasis on the psychic component of generalized anxiety disorder, the use of drug classes other than benzodiazepines is warranted for longterm therapy, with delegation of the benzodiazepines more to the treatment of acute anxiety symptoms or to intermittent therapy for chronic anxiety. Buspirone, an azapirone, has become widely prescribed, especially because of its lack of withdrawal symptoms and “addictive” potential. Clinical studies, mostly lasting 4–6 weeks, have shown it to be slower in onset of action and less effective than benzodiazepines on somatic and autonomic symptoms (Lader 1998). It is recommended when there are conspicuous symptoms of worry, apprehension, tension, and irritability (Connor and Davidson 1998). Because of its slow onset of action and weaker effect on physiological symptoms, patients often reject it early in treatment, and generally it is not effective for more severe patients. However, with its minimal sedation and minimal psychomotor or cognitive impairment, it may be safer than a benzodiazepine for long-term use (Lader 1998). An ade❉

132 ❉

❉ The Difficult-to-Treat Patient With Anxiety Disorder

❉

quate trial requires 3–4 weeks at up to 60 mg/day. One of the earliest medications found useful in the treatment of anxiety symptoms was the antihistamine hydroxyzine. A recent article (Ferreri and Hantouche 1998) summarized early trials, comparison trials with benzodiazepines and antidepressants, and large multicenter trials. The authors suggest that hydroxyzine is reasserting itself as an anxiolytic, with recent data showing good cognitive tolerance; rapid relief of psychic and anxiety symptoms; and early improvement in apprehension, irritability, and concentration difficulties. In practice, antidepressants have not become well established as firstline treatments for generalized anxiety disorder in spite of controlled studies demonstrating efficacy for several classes and recent regulatory approval for venlafaxine and paroxetine. Venlafaxine, a serotonin and norepinephrine reuptake inhibitor, received FDA approval for the treatment of generalized anxiety disorder after proving effective in several large multicenter trials of DSM-IV generalized anxiety disorder. One trial found venlafaxine-XR to be effective at doses of 75 mg/day and 150 mg/day for psychic anxiety symptoms, whereas no significant differences were noted for buspirone (Gelenberg et al. 2000). Imipramine has been shown to be more effective than diazepam for psychic symptoms of anxiety, and a large generalized anxiety disorder trial comparing imipramine, trazodone, and diazepam showed efficacy for both antidepressants (Rickels et al. 1993). Nefazodone has proven beneficial for generalized anxiety disorder in an open-label study (Hedges et al. 1996). In a meta-analysis of six controlled trials, nefazodone was shown to be equal to imipramine in relieving depression and anxiety, and more effective for agitation and for psychic and somatic anxiety (Fawcett et al. 1995). β-Blockers, long used in the treatment of physical symptoms of anxiety in spite of little formal evaluation, continue to be used for such symptoms as tremor, rapid heart rate, and flushing (Morgan and Tyrer 1994). Nonpharmacological Interventions Unlike drug treatments, the nonpharmacological treatment of generalized anxiety disorder has some long-term data, with many of these therapies evaluated over several months. A review of outcome studies concluded that cognitive therapy was most effective, with about one-half of patients attaining normal functioning (Durham and Allan 1993). Generalized anxiety disorder has also been helped by supportive, nondirective therapies and anxietymanagement training (Blowers et al. 1987). ❉

133 ❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

Complicating Factors in Treatment Patients with a recent history of drug or alcohol abuse/dependence present a special problem in anxiety treatment. Anxiety and insomnia are common problems in this population, both during active use and in the withdrawal phase. Accurate diagnosis can be difficult because those without a prior history of generalized anxiety disorder may present with symptoms that meet the criteria but that are substance induced. Psychoeducation and pychotherapeutic interventions directed at the anxiety symptoms should be incorporated into the substance abuse treatment program. Early medication intervention, sometimes including benzodiazepines, may be necessary to ensure abstinence and compliance. A second group of patients with generalized anxiety disorder that is challenging to treat with medication is the geriatric population, with their frequently altered drug metabolism and multiple medical problems and medications. Psychotherapy, often underutilized in this population, can be quite effective. One of the most controversial issues in treating chronic anxiety is that of whether to use benzodiazepines and, if so, when, for how long, and at what dose. Both clinicians and patients are often reluctant to use them at all and may underdose when using them. A common problem is using a medication with a relatively short half-life, such as lorazepam or alprazolam, and dosing it only once or twice a day or “as needed.” The patient then suffers daily episodes of rebound anxiety between doses, often resulting in approved or nonapproved escalations in the amount of each dose and increased side effects. Alternatively, it may be decided that the drug is not helpful when it may actually be an excellent choice if dosed at more frequent intervals.

Treatment Recommendations Anxiety management techniques and psychological therapies should, whenever possible, be used in treating generalized anxiety disorder. Relaxation training, using such methods as progressive muscle-relaxation, fantasy methods, hypnosis, meditation, and biofeedback, can produce dramatic improvement in generalized anxiety symptoms. Generally, patients present during the periods of acute exacerbation and will benefit by at least a short course of medication. Although there is no research support for combined therapy, our recommendation is to start with this approach whenever possi❉

134 ❉

❉ The Difficult-to-Treat Patient With Anxiety Disorder

❉

ble. During subsequent exacerbation, brief courses of drug intervention may be necessary and appropriate. For the generalized anxiety disorder patient with prominent psychic symptoms of worry, apprehension, and tension and no comorbid substance abuse or depressive symptoms, hydroxyzine is recommended. If there is no response within 4 weeks, another agent should be tried. For those with generalized anxiety disorder and a depressive illness, an antidepressant is the initial treatment of choice. Venlafaxine-XR and paroxetine are first-line agents, and the other SSRI drugs also may be used. A TCA such as imipramine or amitryptyline may be preferred for those who have prominent insomnia, elevated blood pressure, or a prior history of good response to these agents. Trazodone and nefazodone are alternative choices and often can be tolerated at high doses of 400–600 mg daily. A benzodiazepine can also be used adjunctively early in treatment. If there is a partial response, hydroxyzine or a benzodiazepine can be added. Patients with severe somatic symptoms can benefit from a benzodiazepine. Those who have been successfully treated with benzodiazepines in the past and have no recent substance abuse are appropriate for a course of benzodiazepines. If depressive symptoms emerge or the psychic symptoms of poor concentration and cognitive difficulties remain problematic, an antidepressant should be added and the benzodiazepine reduced or withdrawn. As indicated earlier, intermittent use of benzodiazepines for brief periods is recommended over the course of the illness. For geriatric patients and others with multiple problems and potential drug interactions, medication should be used at low doses and titrated slowly. TCAs are not recommended because of their cardiac side effects and the anticholinergic activity that can negatively affect cognition. If a benzodiazepine is used, those that are metabolized by conjugation, such as lorazepam and oxazepam, are recommended, especially for patients with liver dysfunction. Hydroxyzine and trazodone are useful agents in this population. We also suggest low doses of nefazodone and mirtazepine, carefully avoiding potential drug interactions with nefazodone. Additional Medication Interventions Neuroleptics, historically referred to as “major tranquilizers,” continue to have a place in the treatment of anxiety. The newer atypical antipsychotics can be especially helpful in anxiety accompanied by chronic pain or other somatic symptoms. Our first choice would be olanzapine at doses of 2.5–7.5 mg/day. Quietiapine at doses of 25–150 mg/day can relieve agitation and ❉

135 ❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

insomnia and is well tolerated by the geriatric population. Risperidone at doses of 0.25–1.0 mg/day may also be useful. These should be used as adjunctive medicines for patients with chronic depression, especially those with somatic complaints and/or history of psychosis who continue to experience impairment due to anxiety. For pure generalized anxiety disorder, atypical antipsychotics can be used alone and are relatively safe for longterm treatment. Combination medication can be used for patients who fail trials of single drugs and certainly should be considered initially for severe symptoms with high risk of suicide or when there is a comorbid condition present. The predisposition to depression is an important and common complication and, even when an antidepressant is not the first choice for the anxiety symptoms per se, it is frequently prudent to initiate a course of treatment along with the anxiolytic of choice. β-Blockers, at low doses, can be used for motor restlessness, tremor, flushing, or other symptoms of sympathetic arousal. For patients with comorbid panic disorder, a benzodiazepine along with an antidepressant is the recommended choice. If there is significant mood lability and/or extreme anger and irritability, an anticonvulsant such as gabapentin or carbamazepine may be a better choice than a benzodiazepine for the somatic anxiety symptoms and can be combined with an antidepressant. Novel and Alternative Treatments Newer drugs that are partial agonists of the benzodiazepine receptor are being studied (e.g., abecarnil, which has shown promise in several large clinical trials [Lader 1998]). Some herbal remedies have long been used to relieve symptoms of anxiety. Piper methysticum (Kava Kava) is a plant that is native to the Pacific Islands where it has been used to relieve stress and anxiety for centuries. Its use spread to Europe in the 1880s, and it is often prescribed in several European countries, including Germany, where clinical studies have found it comparable to many prescription sedatives in relieving stress and anxiety. This herb is sold in capsules generally containing 60–70 mg of kavalactones and may be helpful at doses of one to three capsules daily. It should not be combined with benzodiazepines as it may potentiate their sedative effects. Hypericum perforatum (St. John’s Wort) has a long history as a natural remedy for nervousness, anxiety, and sleep problems and more recently for mild to moderate depressive illness. One to three tablets of 300 mg daily is an effective dose for most people. There are many traditional and some newer stress-relieving activities. ❉

136 ❉

❉ The Difficult-to-Treat Patient With Anxiety Disorder

❉

Aerobic exercise, yoga, various meditation techniques, body therapies such as polarity and Alexander technique, therapeutic massage, and aromatherapy are among the many possibilities. Clinical Vignette Ms. A, a 24-year-old married woman with generalized anxiety disorder, was referred after not responding to treatment with various antidepressants, including nefazodone, paroxetine, and imipramine, combined with the benzodiazepine alprazolam, as well as cognitive therapy. She was unable to tolerate nefazodone and paroxetine due to severe stomach distress and imipramine due to sedation and dizziness. She was reluctant to use alprazolam because of concerns about addiction and a subjective sense of loss of control when using it, even at low doses. During the evaluation it became clear that there was a great deal of conflict between the patient and her physician husband, whom she was constantly trying to please. He was critical of her difficulties with her anxiety and disparaging of her need for medication. During this initial session, psychoeducation about her illness and also about medication was a priority. It became immediately clear that gastrointestinal distress was integral to her illness. She was terrified of being in public and having an attack of diarrhea and also very sensitive to gastrointestinal side effects of medication. She had never had a panic attack but reacted with intense anxiety to physiological side effects of medications and admitted to being noncompliant with her prescribed drugs. We first targeted her diarrhea and stomach distress. She agreed to take lorazepam at 0.5 mg before going out in public and also to daily doses of desipramine at 10 mg, which would cause few side effects but might provide some benefit in terms of reducing intestinal motility. With much support and encouragement, she gradually was able to increase lorazepam to a dose of 0.5 mg twice a day over the next several weeks. She was uncomfortable with the increased heart rate caused by increasing desipramine to an ultimate dose of 35 mg/day but tolerated it because of the reduction in her diarrhea. The stomach distress, tightness “like a knot,” and pain was alleviated with lorazepam. Once the somatic symptoms were resolved, the patient was able to engage in activities outside the home more easily and was better able to engage in cognitive therapy involving other issues. Her constant worry about multiple issues gradually lessened, and her medications were eventually able to be reduced to an occasional dose of lorazepam and a daily dose of desipramine 25 mg. With the generalized anxiety disorder symptoms resolved, the patient began to work on therapeutic issues concerning her family, using an interpersonal approach. As of this writing, she continues on desipramine, has infrequent appointments, and is functioning fully in all aspects of her life.

❉

137 ❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

This case involved significant somatic symptoms that prevented progress in treating the whole spectrum of generalized anxiety disorder symptoms. Both a benzodiazepine and a TCA were necessary to treat the specific complaints. Once the patient was able to trust her body, she proceeded with treatment to address her multiple other issues and made good progress over the next several months.

Panic Disorder Panic disorder, with a 3.5% lifetime prevalence, is less common than social phobia or generalized anxiety disorder, yet it carries with it a high risk for dysfunction. Individuals with panic disorder tend to be high users of health services and are at high risk for substance abuse, suicide attempts, marital problems, and financial dependency (Markowitz et al. 1989). Comorbid major depression, personality disorders, and medical problems as well as other anxiety disorders are common. The essential feature in panic disorder is the unexpected panic attack, which is a discrete period of intense fear and discomfort with symptoms such as dizziness, increased heart rate, sweating, shortness of breath, and paresthesias, as well as fear of dying, losing control, or going crazy. Some patients have recurrent episodes, whereas many remain symptomatic chronically, usually with waxing and waning intensity of symptoms. Both CBT and pharmacotherapy have proven effective in the treatment of panic disorder, with some data suggesting greater efficacy for the combination of drug and CBT than for CBT alone. Psychoeducation is also a valuable part of the treatment.

Overview of Treatment Pharmacotherapy Alprazolam was the first medication to gain regulatory approval for treatment of panic disorder after a large controlled study (Ballenger et al. 1988) conclusively demonstrated efficacy. Two other high-potency benzodiazepines, clonazepam and lorazepam, have also proven to be effective. The lower-potency benzodiazepines are not as effective but can be helpful, especially in combination with antidepressants. The SSRIs have become established over the last decade as first-line treatments for panic disorder. Clinical trials have reported efficacy for fluvoxamine (Black et al. 1993), paroxetine (Oehrberg et al. 1995), and sertra❉

138 ❉

❉ The Difficult-to-Treat Patient With Anxiety Disorder

❉

line (Gorman and Wolkow 1994). The final target dose of antidepressants in panic disorder may be higher than that needed for depression. A dose-ranging study of paroxetine for panic disorder found that 40 mg/day was the minimum effective dose (Ballenger et al. 1998), and studies using fluvoxamine report effective doses from 150 to 300 mg/day. A meta-analysis of 27 controlled studies involving 2,348 patients with panic disorder found that the effects sizes for the SSRIs were larger than those for imipramine or alprazolam (Boyer 1995). The TCAs imipramine and clomipramine now have well-established efficacy in panic disorder (Lydiard et al. 1996), and other TCAs have also been widely used. Although effective, the TCAs have disadvantages; in panic disorder patients, the dizziness and increased heart rate can be especially upsetting as they mimic some of the same symptoms experienced in panic attacks. In contrast to the SSRIs, effective TCA doses are often lower than those used for depression. The MAOIs have not been as widely used in the United States as the TCAs but may be more effective than other treatments for coexisting major depression and panic disorder (Lydiard et al. 1996). The potential for serious side effects has limited their usefulness. Psychotherapy CBT is reported to be superior to relaxation techniques and nondirective psychotherapies and equal in efficacy to antidepressant medications (Black et al. 1993; McNally 1990). Much of the research has been done in major centers with clinicians who specialize in this modality. However, a recent report of CBT group in a general clinical setting summarized the results of groups done by several clinicians not formally trained in CBT. The overall response rate for those completing treatment was 89% (Martinsen et al. 1998). Our experience also suggests that with a brief introduction to CBT techniques, most experienced clinicians can utilize this method.

Factors Associated With Lack of Response Comorbid depression generally responds to the treatments described above, but the other common comorbid disorders, such as severe personality disorders and ongoing substance abuse or dependence, are predictive of failure. These illnesses should be addressed directly by a clinician skilled in their treatment. Severe borderline personality disorder patients are often referred for dialectical behavior therapy programs, a specialized form of CBT. Sub❉

139 ❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

stance abuse programs are also available in most communities and should be utilized for the patients with ongoing symptoms. Social phobia also has high comorbidity with panic disorder. If a thorough assessment has been done, the clinician will have targeted both disorders from the beginning. However, for many panic disorder patients, the panic attacks are such an overwhelming focus that the coexistence of social phobia symptoms are not initially addressed. For this comorbidity, the medications that specifically treat social phobia should be prescribed, as they also have been found beneficial in panic disorder. Dosing may be different, however. For example, paroxetine 20 mg/day is generally adequate for social phobia, but 40 mg/day is considered the minimum dose in panic disorder.

Treatment Recommendations Psychoeducation in the initial sessions is critical in engaging panic disorder patients in treatment. The connection between perceived threat and the physiological symptoms of arousal, psychic anxiety, and automatic thoughts should be explained. Whether in individual therapy or group therapy, each patient should be helped to recognize his or her own patterns. For those patients who elect medication as part of the treatment, an antidepressant or benzodiazepine or both are possible first choices. Given the better side-effect profile and possible greater efficacy for SSRIs, we recommend starting with one of this class of drugs, using low doses such as 5 mg of fluoxetine, 10 mg of paroxetine or citalopram, and 25 mg of sertraline or fluvoxamine daily. Most panic disorder patients are fearful of medication and need a great deal of encouragement and reassurance. The patient must be carefully educated concerning expected side effects and should be offered the opportunity to call with any concerns. Education of a family member can also be extremely useful in providing the patient with support in this process. Maintaining a sense of control is an important issue for these patients and allowing them control over dosing of the medication may add to their sense of security. Most panic disorder patients are sensitive to caffeine and over-thecounter cold medications and also have increased anxiety symptoms when withdrawing from even moderate alcohol intake. Education about the potential for such products to worsen their illness will help panic disorder patients to monitor more carefully these environmental stressors and can cause an early improvement in symptoms. For those patients who insist on continuing some caffeine intake, avoidance of fluvoxamine is judicious because ❉

140 ❉

❉ The Difficult-to-Treat Patient With Anxiety Disorder

❉

it can increase the half-life of caffeine several fold. Many patients find it impossible to tolerate even the minimum starting doses of antidepressants. Using a benzodiazepine early in treatment will increase the likelihood of success. For many patients and their families who are fearful of “addiction” potential, explaining the concept of physiological dependence and the need for careful withdrawal after chronic use as a separate issue from addiction, which rarely occurs outside of the substance abuse population, can ease their anxiety. If a benzodiazepine is chosen as a single medication, the side effects of sedation, ataxia, and sometimes irritability should be explained. Even if formal CBT is not engaged in, having the patient keep a diary of anxiety ratings on a daily basis along with a record of panic attacks and dysfunctional thoughts can help to objectify the disease. This can give the patient a measure of control and also serve as a way to monitor progress. Agoraphobic symptoms can be addressed by helping the patient to create a hierarchy of feared situations. The symptoms can then be successfully dealt with by exposure to the least feared ones first, gradually progressing to the most feared. For patients on medication, there is a choice of chronic administration or episodic use for those patients who experience long panic-free periods. The initial course of medication is for a minimum of 6 months. Also, for patients who have opted for CBT, follow-up “booster” sessions should continue periodically for 6 months following the initial 8- to 16-week treatment course. Other Approaches for Treatment-Resistant Patients Nefazodone or venlafaxine are alternative antidepressant choices. For those with nocturnal panic attacks, nefazodone can be especially effective and can be given at bedtime. For those with severe comorbid depressive symptoms, venlafaxine would be our first choice. The anticonvulsants generally are not a first-line choice in drug treatment of panic disorder, but emerging evidence supports the use of valproic acid for those who cannot tolerate antidepressants. For those patients with comorbid social phobia, gabapentin should be tried. For those with a comorbid mood-cycling illness or posttraumatic stress disorder, gabapentin again would be the first choice and lamotrigine the second choice. Carbamazepine has not been shown to be helpful in panic disorder. ❉

141 ❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

Clinical Vignette Ms. K, a 49-year-old married woman with panic disorder and generalized anxiety disorder presented with exacerbation of her symptoms over the previous several months, including constant worrying about multiple issues, increased partial symptom panic attacks occurring most days of the week, severe anticipatory anxiety, and increasing agoraphobia. Past treatments included several courses of psychotherapy, both psychodynamic and cognitive, first at the age of 5, and multiple periods of medication intervention. Recent medications were 1) alprazolam 1 mg to be used as needed at bedtime and up to 1 mg during the day as needed and 2) paroxetine 20 mg/day. She had discontinued use of the alprazolam for the last 8 weeks on the recommendation of her physician and in response to her own concern about addiction. Other recent interventions included discontinuing all use of caffeine and alcohol about 3 months previously and increasing exercise. The assessment included discussion of issues of benzodiazepine use, covering risks and benefits. The initial plan was to continue to abstain from alcohol and to resume alprazolam on a regular dosing schedule of 0.25 mg three times a day. Therapeutic support was provided initially. The longer term plan was to add an SSRI after the acute stabilization of fears and anxiety symptoms. An increase in alprazolam to 0.25 mg five times a day resulted in remission of virtually all acute anxiety symptoms and improvement in functioning within 4 weeks. Sertraline was then added, started at 25 mg/day, and increased to 50 mg/day after 2 weeks. Eight weeks after presentation, Ms. K was functioning well. More focused, cognitive techniques were added during the first 4 months. Sertraline was increased to 150 mg/day over 2 months, and the patient was virtually symptom free after 5 months of treatment. At that point, the patient’s husband began to pressure her to taper off the alprazolam because of the “addiction potential.” More education about chronic anxiety disorders and the risks and benefits of benzodiazepine use was done for the couple, and the patient was maintained at her current medication dosages. After a period of 6 months of normal functioning with no acute symptoms, the patient gradually decreased alprazolam to 0.25 mg three times a day without adverse effects. Four years later, she is stable on a dose of sertraline 150 mg/day and alprazolam once or twice a day as needed, up to 1 mg at a time for such situations as overseas airplane flights.

The case demonstrates the value of using benzodiazepines on a regular basis in doses given frequently enough to stabilize the anxiety symptoms. When prescribed only “as needed” for anxiety disorders, there is generally a need for higher total daily dosages. Also, stabilizing the acute symptoms prior to introducing an SSRI, which is likely to exacerbate them, increases the ❉

142 ❉

❉ The Difficult-to-Treat Patient With Anxiety Disorder

❉

likelihood of the patient tolerating the SSRI. In this case with comorbid panic disorder and generalized anxiety disorder, intermittent use of a benzodiazepine can easily control the mild acute exacerbations while the antidepressant prevents recurrence of panic attacks.

Summary The anxiety disorders respond well to medications and to psychotherapy, but high comorbidity complicates treatment decisions. Careful attention to the presenting symptoms as well as the history is necessary for an accurate diagnosis. The anxiety symptoms per se often interfere with the ability of the clinician to establish a relationship and also with the patient’s ability to process information. Psychoeducation can help to reduce anxiety, but often the patient is unable to benefit from this until the acute anxiety is addressed. Patients are often reluctant to take any medication, and it may take several sessions of support and reassurance to initiate treatment. The various psychotherapies that are effective for anxiety disorders include exposure to anxiety-producing situations, and patients are understandably reluctant to participate in this treatment that predictably increases the symptoms for which they are seeking treatment. The more difficult cases of anxiety disorders generally require combination treatments for full remission of symptoms and frequently continue to have some residual symptoms and/or recurrent periods of increased symptoms. It is not clear from the literature that a combination of psychotherapy and medication is more beneficial than either one alone for these disorders, but clinical experience suggests that this is generally true. Medication combinations also are often needed, with the medication choices carefully tailored to the individual patient. The other issue with medications that is somewhat unique to this population is that they need to be adjusted over time as symptoms wax and wane. These illnesses are very reactive to both internal and external environmental stressors, and the distress engendered by these can develop quickly and can be quite debilitating. Early and appropriate intervention with each can enable these patients to live productive and satisfying lives in spite of their chronic illnesses.

References Allgulander C: Paroxetine in social anxiety disorder: a randomized placebo-controlled study. Acta Psychiatr Scand 100:193–198, 1999

❉

143 ❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

Altamura AC, Pioli R, Vitto M, et al: Venlafaxine in social phobia: a study in selective serotonin reuptake inhibitor nonresponders. Int Clin Psychopharmacol 14:239– 245, 1999 American Psychiatric Association: The Diagnostic and Statistical Manual for Mental Disorders, 4th Edition. Washington, DC, American Psychiatric Association, 1994 Baer L, Jenike MA, Black DW, et al: Effect of axis II diagnosis on treatment outcome with clomipramine in 55 patients with obsessive-compulsive disorder. Arch Gen Psychiatry 49:862–866, 1992 Baker RW, Ames D, Umbricht DSG, et al: Obsessive compulsive symptoms in schizophrenia: a comparison of olanzapine and placebo. Psychopharmacology Bulletin 32:89–93, 1996 Baldwin D, Bobes J, Stein DJ, et al: Paroxetine in social phobia/social anxiety disorder. Br J Psychiatry 175:120–126, 1999 Ballenger JC, Burrows GD, Dupont RL, et al: Alprazolam in panic disorder and agoraphobia: results from a multicenter trial. Arch Gen Psychiatry 45:413–422, 1988 Ballenger JC, Wheadon DE, Steiner M, et al: Double-blind, fixed-dose, placebo-controlled study of paroxetine in the treatment of panic disorder. Am J Psychiatry 155:36–42, 1998 Balter MB, Ban TA, Unlenhuth EH: International study of expert judgement on therapeutic use of benzodiazepines and other psychotherapeutic medications, I: current concerns. Human Psychopharamcology 8:253–261, 1993 Bisserbe J, Lane R, Flament M, et al: A double-blind comparison of sertraline and clomipramine in outpatients with obsessive-compulsive disorder. Eur Psychiatry 12:82–93, 1997 Black DW, Wesner P, Bowers W, et al: A comparison of fluvozamine, cognitive therapy and placebo in treatment of panic disorder. Arch Gen Psychiatry 50:44–50, 1993 Blowers C, Cobb J, Mathews A: Generalized anxiety: a controlled treatment study. Behav Res Ther 25:493–502, 1987 Boyer W: Serotonin uptake inhibitors are superior to imipramine and alprazolam in alleviating panic attacks: a meta-analysis. Int Clin Psychopharmacol 10:45–49, 1995 Chouinard G: Sertraline in the treatment of obsessive-compulsive disorder: two double-blind placebo controlled studies. Int Clin Psychopharmacol 7 (suppl 2):37– 41, 1992 Christensen H, Hadzi-Pavlovic D, Andrews G, et al: Behavior therapy and tricyclic medication in the treatment of obsessive-compulsive disorder: a quantitative review. J Consult Clin Psychol 55:701–711, 1987 Clark DB, Agras WS: The assessment and treatment of performance anxiety in musicians. Am J Psychiatry 148:598–605, 1991 Cobb JP: The nonpharmacological management of anxiety. Br J Clin Pract Suppl 38:42–47, 1985

❉

144 ❉

❉ The Difficult-to-Treat Patient With Anxiety Disorder

❉

Connor CM, Davidson JRT: Generalized anxiety disorders: neurogiological and pharmacotherapies perspectives. Biol Psychiatry 44:1286–1294, 1998 Cox BJ, Swinson RP, Morrison B, et al: Clomipramine, fluoxetine, and behavior therapy in the treatment of obsessive-compulsive disorder: a meta-analysis. J Behav Ther Exp Psychiatry 24:149–153, 1993 Davidson JRT: Pharmacotherapy of social anxiety disorder. J Clin Psychiatry 59 (suppl 17):47–51, 1998 Davidson JRT, Hughes DL, George LK, et al: The epidemiology of social phobia: findings from the Duke Epidemiological Catchment Area Study. Psychol Med 23:709– 718, 1993 Durham RC, Allan T: Psychological treatment of generalized anxiety disorders. Br J Psychiatry 163:19–23, 1993 Falloon IR, Lloyd GG, Harpin R: The treatment of social phobia: real life rehearsal with nonprofessional therapists. J Nerv Ment Dis 169:180–184, 1981 Fawcett J, Marcus RN, Anton SF, et al: Response of anxiety and agitation symptoms during nefazodone treatment of major depression. J Clin Psychiatry 56 (suppl) 6:37–42, 1995 Ferreri M, Hantouche EG: Recent clinical trials of hydroxyzine in generalized anxiety disorder. Acta Psychiatr Scand 98 (suppl 393):102–108, 1998 Foa EB, Kozak MJ: Psychological treatment for obsessive-compulsive disorder, in Long-Term Treatments of Anxiety Disorders. Edited by Mavissakalian MR, Prien RF. Washington, DC, American Psychiatric Press, 1996, pp 285–309 Gelenberg AJ, Lydiard RB, Haskins JT, et al: Efficacy of venlafaxine extended release capsules in nondepressed outpatients with generalized anxiety disorder: a 6month study. JAMA 283:3082–3088, 2000 Gelernter CS, Uhde TW, Cimbolic P, et al: Cognitive-behavioral and pharmacological treatments of social phobia. Arch Gen Psychiatry 48:938–945, 1991 Goodman WK, Ward HE, Murphy TK: Biologic approaches to treatment refractory obsessive-compulsive disorder. Psychiatric Annals 28:641–649, 1998 Gorman J, Wolkow R: Sertraline as a treatment for panic disorder. Presented at the 19th Annual Meeting of the College of the International Neuropsychopharmacology. Washington, DC, June–July 1994 Gould R, Buckminster S, Pollack M, et al: Cognitive behavioral and pharmacological treatment for social phobia: a meta-analysis. Clinical Psychological Science and Practice 4:291–306, 1997 Greist JH, Chouinard G, DuBoff E, et al: A double-blind comparison of three doses of sertraline in outpatients with OCD. Arch Gen Psychiatry 52:289–295, 1995 Hedges DW, Reimherr FW, Strong RE, et al: An open trial of nefazodone in adult patients with generalized anxiety disorder. Psychopharmacol Bull 32:671–676, 1996 Heimberg RG, Liebowitz MR, Hope DA: Cognitive behavioral group therapy vs. phenelzine therapy for social phobia. Arch Gen Psychiatry 55:1133–1141, 1998

❉

145 ❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

Hewlett WA: The use of benzodiazepines in obsessive compulsive disorder and Tourette’s syndrome. Psychiatric Annals 23:309–316, 1993 Hiss H, Foa EB, Kozak MJ: A relapse prevention program for the treatment of obsessivecompulsive disorder. J Clin Psychol 62:801–803, 1994 Hoehn-Saric R, McLeod DR, Zimmerli WD: Differential effects of alprazolam and imipramine in generalized anxiety disorder: somatic versus psychic symptoms. J Clin Psychiatry 49:293–301, 1988 Hohagen F: Cognitive-behavioral therapy and integrated approaches in the treatment of obsessive-compulsive disorder. CNS Spectrums 5 (suppl 3):35–40, 1999 James IA, Blackburn I-M: Cognitive therapy with obsessive-compulsive disorder. Br J Psychiatry 166:444–450, 1995 Katzelnick DJ, Kobak KA, Greist JH, et al: Sertraline for social phobia: a double-blind, placebo-controlled crossover study. Am J Psychiatry 152:1368–1371, 1995 Kessler RC, DuPont RL, Berglund P, et al: Impairments of twelve-month independent and comorbid generalized anxiety disorder and major depression in two national surveys. Am J Psychiatry 155:1915–1923, 1999 Koran LM, Sallee FR, Pallanti S: Rapid benefit of intravenous pulse loading of clomipramine in obsessive-compulsive disorder. Am J Psychiatry 154:396–401, 1997 Lader MH: Benzodiazepines: a risk benefit profile. CNS Drugs 1:377–387, 1994 Lader MH: The nature and duration of treatment of GAD. Acta Psychiatr Scand 98 (suppl 393):109–117, 1998 Levkovitch Y, Kronnenberg Y, Gaoni B: Can clozapine trigger OCD? J Am Acad Child Adolesc Psychiatry 34:263, 1995 Liebowitz MR, Schneier F, Campeas R, et al: Phenelzine vs. atenolol in social phobia. Arch Gen Psychiatry 40:290–300, 1992 Lydiard RB, Brawnan-Mintzer O, Ballenger JC: Recent developments in the psychopharmacology of anxiety disorders. J Consult Clin Psychol 64:660–668, 1996 Marazziti D, Gemignani A, Dell Osso L: Trazondone augmentation in OCD: a case series report. CNS Spectrums 4:48–49, 1999 Markowitz JS, Weissman MM, Ouellette R, et al: Quality of life in panic disorder. Arch Gen Psychiatry 46:984–992, 1989 Martinsen EW, Olsen T, Tonset E, et al: Cognitive behavioral group therapy for panic disorder in the general clinical setting. J Clin Psychiatry 59:437–442, 1998 McDougle C, Fleischmann R, Epperson C, et al: Risperidone addition in fluvoxaminerefractory obsessive-compulsive disorder: three cases. J Clin Psychiatry 56:526– 528, 1995 McNally RJ: Psychological approaches to panic disorder. Psychol Bull 108:403–419, 1990 Montgomery SA: Citalopram treatment of obsessive-compulsive disorder. Abstract presented at the 37th Annual Congress of the American College of Neuropsychopharmacology, Honolulu, HI, February 1–3, 1998

❉

146 ❉

❉ The Difficult-to-Treat Patient With Anxiety Disorder

❉

Morgan J, Tyrer P: Treating the somatic symptoms of anxiety. CNS Drugs 1:427–434, 1994 Oehrberg S, Christiansen PE, Behnke K, et al: Paroxetine in the treatment of panic disorder: a randomized, double-blind, placebo-controlled study. Br J Psychiatry 167:374–379, 1995 O’Sullivan G, Nashirvani H, Marks I: Six-year follow-up after exposure and clomipramine therapy for obsessive-compulsive disorder. J Clin Psychiatry 52:150– 155, 1991 Otto MW, Pollack MH, Gould RA: A comparison of the efficacy of clonazepam and cognitive-behavioral group therapy for the treatment of social phobia. J Anxiety Disord 14:345–358, 2000 Pande AC, Davidson JRT, Jefferson JW, et al: Treatment of social phobia with gabapentin: a placebo-controlled study. J Clin Psychopharmacol 19:341–348, 1999 Patel B, Tanden R: Development of obsessive-compulsive symptoms during clozapine treatment (letter). Am J Psychiatry 150:836, 1993 Pato M: Delusional OCD treated with olanzapine augmentation. Paper presented at 3rd International Obsessive-Compulsive Disorder Conference, Madeira, Portugal, September 11–12, 1998 Rickels K, Schweizer E: The clinical presentation of generalized anxiety in primarycare settings. J Clin Psychiatry 58 (suppl 11):4–10, 1997 Rickels K, Case WG, Downing RW, et al: Indications and contraindications for chronic anxiolytic treatment: is there tolerance to the anxiolytic effect? in Chronic Treatments in Neuropsychiatry. Edited by Kemal D, Racagni G. New York, Raven, 1985, pp 193–204 Rickels K, Downing R, Schweizer E, et al: Antidepressants for the treatment of generalized anxiety disorder: a placebo-controlled comparison of imipramine, trazodone, and diazepam. Arch Gen Psychiatry 50:884–895, 1993 Shannahoff-Khalsa D, Ray L, Levine S, et al: Randomized controlled trial of yogic meditation techniques for patients with obsessive-compulsive disorder. CNS Spect 4:34–47, 1999 Shear MK, Beidel DC: Psychotherapy in the overall management strategy for social anxiety disorder. J Clin Psychiatry 59 (suppl 17):39–44, 1998 Taylor S: Meta-analysis of cognitive-behavioral treatments for social phobia. J Behav Ther Exp Psychiatry 26:1–9, 1996 Tollefson GC, Rampey AH Jr, Potvin JH, et al: A multi-center investigation of fixeddose fluoxetine in the treatment of obsessive-compulsive disorder. Arch Gen Psychiatry 51:559–567, 1994 Turner SM, Beidel DC, Cooley MR: A multi-component behavioral treatment for social phobia: social effectiveness therapy. Behav Res Ther 32:381–390, 1994 Van Ameringan M, Mancini C, Wilson C: Buspirone augmentation of selective serotonin reuptake inhibitors (SSRIs) in social phobia. J Affect Disord 39:115–121, 1996

❉

147 ❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

Van Ameringan M, Mancini C, Oakman JM: Nefazodone in social phobia. J Clin Psychiatry 60:96–100, 1999 Van Oppen P, De Haan E, Van Ballcom A, et al: Cognitive therapy and exposure in vivo in the treatment of obsessive-compulsive disorder. Behav Res Ther 33:379– 390, 1995 van Vliet IM, den Boer JA, Westenberg HG: Psychopharmacological treatment of social phobia; a double blind placebo controlled study with fluvoxamine. Psychopharmacology (Berl) 115:128–134, 1994 Wheadon D, Bushnell W, Steiner M: A fixed-dose comparison of 20, 40, or 60 mg paroxetine to placebo in the treatment of OCD. Abstract presented at the 32nd Annual Meeting of the American College of Neuropharmacology, December 1993, Honolulu, Hawaii Zohar J, Judge R: Paroxetine versus clomipramine in the treatment of obsessive-compulsive disorder. Br J Psychiatry 169:468–474, 1996

❉

148 ❉

6 The Difficult-to-Treat Patient With Posttraumatic Stress Disorder Elizabeth A. Hembree, Ph.D. Randall D. Marshall, M.D. Lee A. Fitzgibbons, Ph.D. Edna B. Foa, Ph.D.

Knowledge regarding effective treatment of posttraumatic stress disorder (PTSD) has advanced considerably in recent years. However, the existing body of literature informs us little about the characteristics of PTSD sufferers who are difficult to treat or who respond poorly to interventions of proven effectiveness. In this chapter we combine knowledge from empirical studies with clinical wisdom and experience to make recommendations for working with difficult-to-treat PTSD patients. ❉

149 ❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

Diagnostic Criteria and Prevalence According to the DSM-IV (American Psychiatric Association 1994), PTSD may develop in those who experience or witness with “horror, terror, or helplessness” a traumatic event that involves real or perceived threat to life or physical integrity. In PTSD, posttrauma sequelae manifest in three clusters of symptoms: reexperiencing the trauma, avoidance behaviors, and hyperarousal. DSM-IV stipulates that the diagnosis be made when symptoms have been present for more than 1 month and cause clinically significant distress or impairment in functioning. PTSD is specified as chronic when the symptoms persist for 3 months or more. The lifetime prevalence of PTSD has been estimated at 24% among trauma survivors and at 9% in the general population (Breslau et al. 1991).

Comorbid Disorders The majority of PTSD sufferers meet criteria for other disorders. Breslau and colleagues (1991) found that 83% of subjects with PTSD in an urban population also met criteria for one or more additional DSM disorders. Kessler and colleagues (1995) reported that 52% of men and 28% of women with lifetime PTSD also met criteria for alcohol abuse/dependence. Other common comorbid disorders are major depression (48%) and other anxiety disorders (e.g., social phobia, 28%; simple phobia, 30.2%; men and women combined). As will be illustrated in a clinical vignette later in this chapter, comorbidity often introduces difficulty in treating patients with chronic PTSD.

Treatment of Chronic PTSD Pharmacological Treatment At least fifteen randomized, placebo-controlled trials for PTSD have been conducted and presented to date. In this chapter, we focus only on those pharmacological treatments that have shown the relatively greatest efficacy in controlled trials: the serotonin reuptake inhibitors, tricyclic antidepressants, and monoamine oxidase inhibitors. (For a complete review of the literature see Yehuda et al., in press.) Selective Serotonin Reuptake Inhibitors Four multicenter trials with selective serotonin reuptake inhibitors (SSRIs) (sertraline, paroxetine, and fluoxetine) have demonstrated superiority over ❉

150 ❉

❉

The Difficult-to-Treat Patient With Posttraumatic Stress Disorder ❉

placebo (Brady et al. 2000; Davidson et al., in press; Judge et al. 2000; Marshall et al. 2000a, 2000b), confirming earlier single-site trials of fluoxetine (Connor et al. 1999a; Van der Kolk et al. 1994). There are also negative trials with SSRIs (sertraline and fluoxetine); 3 of 4 involved primarily U.S. war veterans (Hertzberg et al. 2000; Nagy et al. 1996; sertraline studies cited in Phycisians’ Desk Reference 2000). The discrepancy in outcome may be because of the treatment-refractory status of U.S. veterans at tertiary treatment centers rather than drug efficacy per se, because almost all studies with noncombat-related trauma have shown superiority over placebo for the SSRIs. In summary, SSRIs are the first-line medication treatment for PTSD (Foa et al. 1999b). However, the considerable range of treatment responses with SSRIs suggests there is a wide range of treatment responsiveness in PTSD populations. Possible explanations for these findings are discussed later in the chapter. Tricyclic Medications An early trial with desipramine found no difference between drug and placebo in male veterans with chronic PTSD (Reist et al. 1989). Dosages were probably subclinical (100–200 mg/day), and the trial may have been too brief to detect treatment effects. In contrast, Davidson and colleagues (1990) found amitriptyline produced modest but significant improvement in 50% of subjects treated in an 8-week trial, compared with 17% of those receiving placebo. Symptoms were reduced by about 20%, but the majority of individuals receiving amitriptyline still met criteria for PTSD posttreatment. Notably, response was greater in less severely ill veterans. Imipramine was also shown to be more effective than placebo among combat veterans (Kosten et al. 1991). The only study to date comparing two active pharmacotherapies to placebo was an 8-week trial of imipramine and phenelzine (a monoamine oxidase inhibitor) in veterans with chronic PTSD (Kosten et al. 1991). Both drugs were significantly more globally effective for PTSD symptoms than placebo for about two-thirds of subjects. Phenelzine appeared significantly better than imipramine for overall symptoms of PTSD and specifically for intrusive symptoms (56% reduction vs. 27% reduction, respectively). Earlier trials of phenelzine found minimal to no efficacy (Lehrer et al. 1987; Shestatzky et al. 1988). As with SSRI trials, the disparate findings across treatment trials suggest the existence of a subgroup of PTSD patients who are not pharmacotherapy responsive. Other categories of medications whose efficacy has been studied in con❉

151 ❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

trolled or open trials include benzodiazepines (alprazolam, temazepam), inositol, noradrenergic suppressors (propranolol, clonidine, guanfacine), and mood stabilizers (lithium, carbamazepine, valproic acid). (For a detailed review of this literature, see Marshall and Pierce 2000).

Psychosocial Treatment A variety of psychotherapeutic approaches have been employed with chronic PTSD patients. We briefly review here the cognitive-behavioral treatments and their reported response rates, because this approach has been most widely and rigorously studied. (For an extensive review of psychosocial treatments for PTSD, see Foa and Meadows 1997.) Cognitive-Behavioral Treatment: Theory and Clinical Practice The cognitive-behavioral interventions for PTSD generally employ, either alone or in combination, exposure procedures (imaginal and in vivo), anxiety management procedures (such as stress inoculation training, or SIT), and cognitive restructuring procedures. A brief description of these interventions is followed by a partial review of the literature on efficacy and response rates. Exposure therapy. Foa and colleagues (e.g., Foa and Jaycox 1999; Foa and Riggs 1993) have suggested that the presence of PTSD reflects impairment in the emotional processing of a traumatic event and the formation of a pathological cognitive structure of the trauma memory. Specifically, Foa and colleagues (1989) proposed that the presence of pathological stimuli associations as well as erroneous evaluations of danger and oneself distinguish a pathological trauma memory from normal trauma memory. Foa and Kozak (1986) proposed that successful treatment modifies the pathological elements of a cognitive structure by activating the structure and providing new information that is incompatible with its pathological elements. To successfully reduce symptoms, psychotherapy for PTSD must produce changes in the patient’s erroneous associations and evaluations. Exposure procedures activate the trauma memory structure through confrontation with trauma-related memories and thoughts (i.e., imaginal exposure) or through situations that trigger such memories (i.e., in vivo exposure). For example, if a rape survivor deliberately and vividly recalls (relives) the memory of her assault and describes what happened during the assault, she is likely to initially feel distress and anxiety. Repeated reliving of ❉

152 ❉

❉

The Difficult-to-Treat Patient With Posttraumatic Stress Disorder ❉

the assault results in diminishing anxiety and in a realization that emotional engagement (rather than avoidance) with the memory of the rape is not dangerous. Repeated reliving also helps the patient make sense of the traumatic event and distinguish it from similar but safe events. (For a more elaborated exposition of the mechanisms involved in exposure therapy see Foa and Rothbaum 1998.) Anxiety management/stress inoculation training. Anxiety management approaches provide tools for the patient to use for managing his or her anxiety as it occurs. These approaches include coping skills or strategies that help the patient manage his or her reactions to trauma-related cues or situations. A commonly used and relatively well-studied anxiety management treatment for PTSD is Meichenbaum’s (1975) SIT, adapted by Veronen and Kilpatrick (1983) for use with rape victims. Veronen and Kilpatrick (1983) viewed trauma (specifically, rape) as a situation in which emotional, cognitive, and behavioral fear and anxiety responses are evoked by the experience or threat of physical pain or death. Through classical conditioning, neutral stimuli (e.g., places, people, certain activities, time of day) acquire the capacity to elicit anxiety via their association with trauma-related stimuli. Treatment is aimed at teaching the patient specific skills for managing this rape-related fear and anxiety. Veronen and Kilpatrick’s SIT program includes education, deep muscle relaxation, breathing exercises, role playing, covert modeling, thought stopping, and guided self-dialogue. Cognitive therapy. Cognitive theories (e.g., Foa et al. 1989; Horowitz 1976) hold that traumatic experiences have a profound effect on the way the survivor perceives him- or herself, other people, and the world. Specifically, a traumatic event may shift one’s perceptions and beliefs about the world from safe to dangerous, about the self from competent to incompetent, and about other people from good or trustworthy to bad and untrustworthy. Foa and Riggs (1993) suggested that a traumatic event may change cognitions not only by shifting previously positive beliefs into negative ones but also by strengthening and intensifying already existing negative perceptions and thoughts. Cognitive therapy (Beck 1976) is based on the idea that it is our interpretation of events, rather than events themselves, that leads to specific emotional responses. For example, if a person interprets a harmless event as threatening, emotions like fear or anger are likely to be experienced. The ❉

153 ❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

goal of cognitive therapy for individuals with chronic PTSD is to teach the patient to identify trauma-induced irrational or unhelpful beliefs that may influence his or her response to a situation and lead to intense negative emotion. The patient learns to challenge these thoughts or beliefs in a rational manner. Relevant facts that support or do not support the irrational beliefs are examined, and alternative ways to view the eliciting situation are considered. The therapist helps the patient to weigh all of this information and conclude whether the thought or belief is helpful and accurate, and, if not, replace or modify it. Such a modification is thought to lead to reduction of PTSD symptoms, including fear and anxiety. Eye movement desensitization and reprocessing. Eye movement desensitization and reprocessing (EMDR; Shapiro 1995) is considered by many to be a variant of imaginal exposure because it requires the patient to focus on a traumatic memory and the associated thoughts and feelings. As the patient engages the disturbing memory, the therapist elicits rapid saccadic eye movements by instructing the patient to visually track a finger rapidly waved back and forth in front of the patient’s face. Many studies (see Lohr et al. 1998) failed to support the claim that rapid eye movements are an essential component of the treatment. Overall, several studies seem to demonstrate that EMDR is an effective treatment for PTSD, but more well-controlled studies are needed for a more definitive conclusion. Outcome Studies of Cognitive-Behavioral Treatments Cognitive-behavioral treatments have been employed with a variety of trauma populations, including combat veterans, rape and nonsexual assault victims, survivors of childhood sexual abuse, and accident and disaster survivors. Although many studies have shown that cognitive-behavioral interventions are effective at reducing symptoms of chronic PTSD in trauma survivors, some have also found that a significant minority of patients respond minimally to the treatments. Early, small studies of veterans found that patients treated with imaginal exposure improved more on measures of PTSD symptoms, depression, and anxiety compared with waitlist control subjects (Keane et al. 1989) and standard treatment (Cooper and Clum 1989). However, in the study by Keane and colleagues, 36% (4/11) of the exposure patients did not respond significantly to treatment. Boudewyns and colleagues (1990) found at 3 months posttreatment that whereas more veterans treated with exposure were considered “successes” (defined as being in the top 25% of improvement) com❉

154 ❉

❉

The Difficult-to-Treat Patient With Posttraumatic Stress Disorder ❉

pared with those treated with conventional treatment, more than half the exposure group did not reach this criterion. Cognitive-behavioral treatment of rape-related PTSD has also been found effective but again with the caveat that, in many studies, a minority of patients do not show a significant response to treatment. Foa and colleagues have found in two controlled studies of female assault victims with chronic PTSD (Foa et al. 1991, 1999a) that cognitive-behavioral treatments including prolonged exposure, SIT, exposure and SIT combined, and supportive counseling produced significant improvement in PTSD symptoms, depression, and anxiety, relative to waitlist control subjects. However, depending on which treatment they received, at posttreatment 36%–70% of patients who completed therapy in these two studies achieved a clinically significant response, with exposure therapy showing somewhat superior outcome. It should be noted that these studies used stringent criteria for improvement. For example, Foa and colleagues (1999a) classified patients as achieving good end-state at posttreatment if they met three criteria: a score of 20 or less on the Posttraumatic Stress Diagnostic Scale (self-report version; PDS-SR; Foa et al. 1995) and normal scores on depression and anxiety. Recent investigations of treatment for PTSD that results from other types of trauma (e.g., motor vehicle accidents, disasters) have produced similar findings. Marks and colleagues (1998) randomized patients with chronic PTSD resulting from mixed trauma to either exposure, cognitive restructuring, exposure and cognitive restructuring combined, or relaxation (an attention control condition). They found that exposure and cognitive restructuring were equally effective alone and in combination and that all three treatments were superior to relaxation. At posttreatment, 32%–53% of patients in the cognitive restructuring, exposure, and combined groups showed clinically significant improvement as measured by a composite endstate criterion similar to that used in Foa and colleagues (1999a): 50% or greater reduction in PTSD severity and normal scores on depression and anxiety. These studies inform us that even the best available psychosocial treatments for chronic PTSD do not effect a clinically significant response in many patients. Information about predictors of response to treatment is still quite scarce. Given the prevalence of PTSD in the general population, it is important to determine the characteristics of the difficult-to-treat PTSD patient and to develop specific strategies for intervening with them. We focus this examination on the biological and exposure-based treatments because most of our knowledge of treatment response comes from these interventions. ❉

155 ❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

Factors That Interfere With Treatment Response Biological Factors Factors associated with nonresponse to pharmacological treatment have not been well studied. Perhaps the most widely discussed factors are the severity of the stressor and chronicity of illness. Davidson and colleagues (1993) studied predictors of treatment response in veterans with PTSD who participated in a placebo-controlled trial of amitryptiline and found that greater symptom severity and the presence of comorbid disorders were negative predictors of treatment response. Marshall and colleagues (1998a) found that cumulative childhood trauma was negatively correlated with treatment response to paroxetine in adults with chronic PTSD. The potential neurophysiological consequences of years of symptomatology and impairment may include loss of previous functional capacity, relatively indelible alterations in synaptic connections, and damage to cells and the functional integrity of neural circuitry. The mechanisms of kindling and sensitization have been invoked to explain the process whereby repeated presentations of a stimulus can provoke a progressively more robust response in PTSD. Taken together, several factors may theoretically contribute to treatment-refractory status in patients treated with medication: severity of trauma, severity and chronicity of illness, comorbidity, and, possibly, developmental phase at which the trauma occurred. Implications for treatment response. It is possible that there are biologically distinct subtypes of PTSD, and this may explain the dramatic range of responsiveness to treatment in chronic PTSD. For example, Griffin and colleagues (1997) studied rape victims with high peritraumatic dissociation scores versus those with low scores. The authors found significant differences in heart rate, skin conductance, and nonspecific movement between the two groups. High-dissociation subjects were also more likely to meet criteria for PTSD and had higher PTSD symptom scores on all measures. Thus, prominent dissociation may be associated with a physiological subtype of severe PTSD, which may have treatment implications. Determination of treatment modality on the basis of biological status may be possible but not necessary. For example, psychosocial interventions can also lead to significant normalization of biological alterations, as was found in successful behavioral treatment of obsessive-compulsive disorder (Schwartz et al. 1996). ❉

156 ❉

❉

The Difficult-to-Treat Patient With Posttraumatic Stress Disorder ❉

Psychological and Behavioral Factors A few reports have contributed to our understanding of the psychological and behavioral characteristics of PTSD patients who respond poorly to psychosocial treatment interventions. Jaycox and Foa (1996) discussed extreme anger, emotional numbing, and overwhelming anxiety as obstacles that sometimes arise in the implementation of exposure therapy for individuals suffering from chronic PTSD. Foa and colleagues (1995) also found extreme anger to be related to treatment outcome: patients reporting high pretreatment levels of anger showed fewer facial expressions of fear during imaginal exposure to traumatic memories and had poorer outcome compared with patients who reported less anger. The notion that emotional numbing interferes with treatment response is supported by studies showing that emotional engagement with the traumatic memory promotes recovery. Foa and colleagues (1995) found that patients who displayed more facial fear expressions during the first imaginal exposure had superior outcome compared with those who displayed less facial fear expressions. Jaycox and colleagues (1998) reported that patients who reported high distress levels in their first imaginal exposure session followed by a gradual decrease in distress over subsequent sessions showed greater improvement after treatment compared with those who reported either high or moderate initial distress and no habituation in distress. Patients who exhibit extremely high levels of distress and fail to habituate within and between exposure sessions often appear overwhelmed with anxiety, as described by Jaycox and Foa (1996). That this might interfere with emotional processing of the traumatic experience is supported by preliminary analyses of predictors of treatment response from the Foa and colleagues (1999a) study (see Foa and Cahill, in press). In examining unique predictors of outcome for specific treatment conditions, we have found that for patients treated with prolonged exposure, higher pretreatment levels of general anxiety were associated with more severe PTSD symptoms at posttreatment. In contrast, higher levels of pretreatment PTSD and depression predicted greater PTSD severity following treatment with stress inoculation training. These findings suggest that prolonged exposure may be more suitable for clients who initially present with severe PTSD or severe depression. In contrast, SIT may be more suitable for patients who present with high levels of general anxiety.

❉

157 ❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

Treatment Strategies for Difficult-to-Treat PTSD Patients Comorbid Disorders It is imperative to conduct a careful assessment of all current symptoms to determine primary and secondary diagnoses. This assessment should include information about the onset and course of each problem and which disorder(s) currently cause the most distress and interference in the patient’s life. In our experience, the depression, anxiety, panic, phobic avoidance, and alcohol or substance abuse commonly seen in patients with PTSD is often secondary to the PTSD, especially in long-term cases. In these patients, the secondary symptoms may lessen or remit along with the target symptoms as PTSD-focused treatment progresses. If another disorder is primary or causes relatively greater interference, it is often best to focus treatment on that problem(s) first and then reevaluate the PTSD symptoms. It is usually recommended that severe alcohol and substance dependence be treated before proceeding with treatment for chronic PTSD. One reason for this order of intervention is that alcohol and drugs are often used to reduce distress and anxiety and thus are a means of avoidance. Avoidance must be decreased in order to confront and emotionally process the traumatic experience(s). For PTSD patients with less severe or episodic alcohol or drug abuse, or in cases where the PTSD is primary but the abuse behavior has serious psychosocial consequences, concurrent treatment for both disorders may be best. The PTSD treatment can be augmented with alcohol treatments such as coping skills training (Monti et al. 1989) or motivational interviewing (Miller and Rollnick 1991). Pharmacological Treatment The pharmacotherapy literature in PTSD does not yet allow development of an empirically based treatment algorithm. Nevertheless, clinical necessity dictates that clinicians make treatment decisions based on available evidence to date. The SSRIs are considered a first-line treatment for chronic PTSD, based on recent controlled trials, and the FDA has given sertraline an indication for adult women with PTSD (Brady et al. 2000; Connor et al. 1999b; Davidson et al., in press; Judge et al. 2000; Marshall et al. 2000a, 2000b; Van der Kolk et al. 1994). Treatment response in acute trials appears to plateau at 8–10 weeks on effective dosages, but a new continuation study suggests that gains may continue to accrue over 24 weeks (Londborg et al., in press) (as seen after exposure therapy). If one SSRI is not well tolerated, we recommend a different SSRI before abandoning this category of medication. If an ❉

158 ❉

❉

The Difficult-to-Treat Patient With Posttraumatic Stress Disorder ❉

SSRI is not effective, it seems advisable to recommend medications with some empirical support for efficacy in PTSD. Although tricyclic antidepressants showed only modest efficacy in two controlled trials, this type of medication has not been adequately studied in noncombat-related PTSD. MAOIs can also be effective. A number of open trials and case reports have suggested efficacy for some of the newer antidepressants, including venlafaxine (Hamner and Frueh 1998) and mirtazepine (Connor et al. 1999b), bupropion (Canive et al. 1998), moclobemide (Neal et al. 1997), and nefazodone (summarized in Hidalgo et al. 1999). A common clinical dilemma arises when an individual shows a partial response to pharmacotherapy. The decision to discontinue the medication or add a second medication depends on a number of factors such as history of prior treatment response, degree of clinical response, severity of illness, and comorbid symptoms or problems. If there is a significant partial response, most experts advise identifying symptom domains that may be responsive to a particular class of medication. The open trials literature provides some guidance for treatment-refractory PTSD. In particular, trials with mood stabilizers (i.e., lithium, valproic acid, carbamazepine, lamotrigine, topiramate, and gabapentin) were typically conducted and reported to be sometimes helpful with the subgroup of PTSD patients manifesting impulsive, aggressive, affectively labile, and/or self-destructive behavior. Drugs that suppress noradrenergic activity such as guanfacine or clonidine can be helpful for persistent increased arousal, insomnia, and nightmares. Early reports found propranolol useful for such symptoms also. Finally, recent case reports and open trials have also suggested a role for some of the atypical antipsychotics, including risperidone (Hamner and Ulmer 1998) and olanzapine (Burton and Marshall 1999). A small controlled trial of olanzapine as a stand-alone treatment, however, did not show drug superiority over placebo (Butterfield et al., in press). Combination with a psychosocial treatment should always be considered. For example, medication may improve insomnia, hyperarousal, and intrusive memories, but persistent avoidance of situations that evoke intrusive symptoms may be best treated with a behavioral approach to promote habituation to these stimuli (rather than a series of pharmacotherapy trials). If, on the other hand, there is only modest response in all symptom domains, adding a second medication with a complementary mechanism of action (e.g., a tricyclic antidepressant) to an SSRI is often attempted. Care must be taken to consider drug-drug interactions, including a potential increase in blood level when drugs with similar catabolic pathways are combined. ❉

159 ❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

For the subgroup of highly treatment-refractory individuals with PTSD, the goals of pharmacotherapy are usually palliative and supportive. It is important that the objectives of treatment be discussed with patients to avoid unrealistic expectations and rejection of a partially helpful treatment. When to discontinue an effective medication is an important question that has not been studied. The clinical decision to reduce and/or discontinue a medication needs to be based on considerations such as the length of time the patient has been symptom free, available social support, consequences of a potential relapse, and prior history of treatment response and need for ongoing pharmacotherapy. Combining Pharmacotherapy and Psychotherapy There are no controlled studies examining the relative efficacy of medication and psychosocial treatment, nor are there systematic studies of either sequential or simultaneous use of these two modalities. Many clinicians recommend that medications be prescribed in combination with psychotherapy. Anecdotally, severely symptomatic individuals may benefit from stabilization of affective and anxiety symptoms before initiating traumafocused psychotherapy. Medication should also be considered if symptoms have not responded to psychotherapy, and it should be kept in mind that studies to date demonstrate that treatments such as prolonged exposure produce benefits relatively quickly (within 2–3 months). Psychosocial Treatment We begin with an overview of a prolonged exposure treatment program for patients who suffer chronic, assault-related PTSD. We then discuss problems that may arise in using this treatment with difficult patients. Finally, we will present the case of a very difficult-to-treat patient who received this treatment. (For a detailed description of the treatment program see Foa and Rothbaum 1998.) Prolonged exposure treatment program. The treatment consists of 9 or 12 individual sessions 90–120 minutes in length. Treatment is completed in 9 sessions if self-reported PTSD symptom severity has decreased by 70% or more by session 8; if not, treatment is extended to 12 sessions. The goal of the treatment program is to help the patient acquire and master specific skills that are used to reduce PTSD symptoms. The central components of therapy are as follows: education about PTSD symptoms and common reactions to trauma, breathing retraining, in vivo exposure, and imaginal expo❉

160 ❉

❉

The Difficult-to-Treat Patient With Posttraumatic Stress Disorder ❉

sure. Homework, consisting of these skills, is an integral part of the treatment. Education and treatment rationale. Treatment begins with a discussion of the impact of traumatic experiences and the development of PTSD. The patient is presented with the idea that symptoms have persisted because avoidance of thinking about the trauma and avoidance of reminders, although common, have prevented the event from being emotionally processed and integrated. The therapist explains that exposure to the memory (imaginal exposure) and to reminders (in vivo exposure) blocks avoidance and thus promotes emotional processing of the trauma. This processing results in increased coherence and organization of the memory, habituation of distress when thinking about the trauma, realization that the memory itself is not dangerous and that anxiety does not last forever, and increased confidence in competence and coping skills. The therapist provides information about common reactions to trauma, and the patient is encouraged to describe his or her own experience of these reactions. This discussion is intended to help the patient to comprehend that symptoms of PTSD and related problems (e.g., shame, guilt, sadness, disruptions in relationships, reduced sexual interest) are common reactions to trauma. The goal is to reduce negative appraisal of PTSD symptoms and to set the focus of treatment on PTSD. Breathing retraining. Breathing retraining is taught as an anxiety management tool. The therapist presents a rationale for its usefulness and explains that the goal of the breathing retraining is to slow respiration rate and reduce oxygen intake. The therapist models then instructs the patient to practice this slow breathing pattern while observing his or her breathing and providing appropriate feedback. Finally, an audio loop tape is made of the therapist guiding the patient through 10–15 such respiratory cycles. The patient is encouraged to practice the skill three times daily to develop its use for managing anxiety. In vivo exposure. The therapist explains that fears associated with the trauma are often unrealistic or excessive and lead to avoidance, which prevents important learning from occurring. Prolonged and repeated confrontation (i.e., in vivo exposure) with situations that are anxiety arousing but not objectively dangerous will result in habituation or reduction of anxiety. The beliefs that particular situations are dangerous and that avoidance or ❉

161 ❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

escape is necessary for anxiety reduction are disconfirmed. Following the rationale, situations or activities that trigger traumarelated anxiety are identified. A hierarchy of avoided situations is constructed with rankings based on the patient’s anticipated distress level if confronting the situation. Distress is rated on a 0–100 scale of subjective units of discomfort (SUDS). Typical situations may be very specific, such as visiting an area related to the incident or reading about a similar assault in the newspaper. Depending on the particular trauma, more general fear-evoking triggers may include making eye contact with unfamiliar men, being alone, sitting in the dark, or standing in line in a crowded store. A careful assessment of objective danger is made in developing the hierarchy so that realistically dangerous situations are not targeted for in vivo exposure practices. Beginning in session 2 and continuing throughout treatment, the patient chooses situations to confront for homework each week, starting with the relatively low items on the hierarchy. The patient is instructed to stay in each situation for 30–45 minutes or until his or her anxiety drops considerably (by at least 50%). By the end of treatment, patients are generally able to comfortably confront most of the situations originally avoided. The therapist recommends that any remaining situations that are still avoided be confronted in continued in vivo exposure work to continue the process of recovery. Imaginal exposure. The detailed rationale for exposure is repeated again in the beginning of the first imaginal exposure session. The therapist explains that the imaginal reliving of the trauma will help to emotionally process and organize the memory. During imaginal exposure, the patient is instructed to describe aloud what happened during the trauma while visualizing it as vividly as possible. The patient is asked to keep his or her eyes closed; to use the present tense; and also to include the thoughts, emotions, and sensory experiences that occurred during the traumatic event. Throughout the reliving, the therapist monitors the patient’s SUDS level as well as his or her rating of how vivid the image seems. Imaginal exposure is continued for 45–60 minutes and includes multiple repetitions of the memory if necessary. The therapist prompts the patient for additional details as needed, especially when some aspect (thoughts, feelings, sensory) is omitted. The goal is to help the patient access the memory and maximize emotional engagement. Immediately following the reliving, the patient and therapist discuss the experience, any new understanding that may have emerged, and patterns of habituation that have occurred both within and between sessions. Imaginal ❉

162 ❉

❉

The Difficult-to-Treat Patient With Posttraumatic Stress Disorder ❉

reliving begins in session 3 and is conducted in each treatment session thereafter. Homework consisting of daily listening to audio tapes of the imaginal exposure continues the work of emotionally processing of the trauma. As therapy proceeds, the imaginal exposure progressively focuses on the most distressing parts of the trauma memory, or “hot spots,” in a repetitive fashion. Repeatedly reliving those parts associated with the highest levels of distress facilitates processing of these memories. Throughout treatment, the therapist must be aware of the patient’s distress level and may need to direct the patient’s reliving to titrate distress—prompting for more detail and feelings if the patient is not emotionally engaged with the memory or helping him or her to decrease detail and engagement if the distress feels overwhelming. Implementing Exposure Therapy With Difficult-to-Treat PTSD Patients Importance of the therapeutic alliance. The alliance between patient and therapist is an important component in any psychotherapy. We posit that this relationship is particularly important in the treatment of PTSD, especially in cases of rape, abuse, or interpersonal violence. In traumatic experiences of this nature, trust, the very fabric of human relationships, has often been deeply torn. Yet to help the patient emotionally process the traumatic event and recover from PTSD, the therapist must gain the patient’s trust and instill at least a seed of faith that the treatment will be effective. A strong and collaborative therapeutic relationship is first forged by the therapist conveying that the therapy is an empirically validated, effective treatment for chronic PTSD and that he or she is an expert in its use. It is further strengthened throughout treatment by the therapist and patient making decisions together regarding in vivo assignments and which aspects of the trauma require attention in the imaginal exposure. In decision making, the therapist makes recommendations but also trusts the patient’s judgment and does not push too far too fast. It is essential throughout treatment that the therapist maintain a nonjudgmental attitude, display compassion and strength when listening to the patient’s traumatic experience, consistently express confidence in the treatment program, and actively praise the patient for his or her courage and the unique strengths and coping resources that the patient brings to treatment. With some PTSD patients, forging a strong and effective alliance can prove difficult because of mistrust, anger, fear, and avoidance. It is important for the therapist to remember that these characteristics are often part of the ❉

163 ❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

PTSD. Secondly, despite these characteristics, the therapist must develop a strong and connected relationship with the patient. The relationship often provides the patient with the support he or she needs to comply with treatment procedures. In fact, in some cases the experience of sharing the pain and horror of the trauma memory with a compassionate, understanding, nonjudgmental person is a powerfully healing experience in itself and reduces the patient’s fear and shame. Importance of rationale. It is frightening and difficult for most PTSD sufferers to confront the memory of their trauma and/or reminders of it, and urges to avoid are common. Thus, it is imperative that patients embrace the rationale, because this acceptance will guide them to confront the memory and reminders when the therapist is not present. With many PTSD patients, the rationale immediately makes sense, the wisdom of the procedures is intuitively grasped, and they comply with little resistance beyond normal trepidation. However, with patients who are excessively fearful or avoidant or extremely intolerant of anxiety, selling the rationale becomes even more critical. General guidelines in these cases include asking the patient to explain the rationale and to discuss any concerns or questions they have regarding the treatment. Frequently, feeling anger, experiencing anticipatory anxiety, feeling blamed for his or her symptoms, or believing that the rationale does not take into account the uniqueness of his or her own situation can delay a patient’s acceptance of the rationale. Additional time, attention, and tailoring of the discussion of the rationale to the patient’s own circumstances can be helpful. Finally, the patient’s intellectual understanding of the rationale is often followed by an emotional understanding during the process of therapy. Some patients are unable to endorse the rationale before they experience the benefit of the procedures. Obstacles to Emotional Engagement: Anger, Numbing, and Anxiety Intolerance As described earlier in this chapter, the goal of exposure treatment is the modification of pathological cognition and fear. Emotion processing theory postulates that modification of a fear structure requires activation of the trauma memory so that corrective information can alter its pathological elements. For this reason, excessive anger, numbing, and affect intolerance can all impede exposure therapy because in various ways they interfere with activation and modification of the trauma memory. When anger is dominant, the fear network is not being accessed. Similarly, numbing can be construed ❉

164 ❉

❉

The Difficult-to-Treat Patient With Posttraumatic Stress Disorder ❉

as a response (often but not always involuntary) that shuts down the “felt experience” and blocks engagement with important aspects of the trauma memory, thus preventing its modification. Intolerance of anxiety leads to avoidance and escape and interferes with treatment primarily by preventing processing of the trauma. Anger, emotional numbing, and intolerance of anxiety are present to some degree in most patients with chronic PTSD. When one of these characteristics is present to an extreme degree, treatment procedures should be modified so the patient can emotionally engage in and process the trauma memory. Anger. When patients experience and display anger with an intensity and frequency that interferes with the emotional processing of their trauma, our general strategy is to first validate the anger they feel about having been assaulted and traumatized. In addition to being a valid and reasonable emotional response to trauma, we remind them that anger is a symptom of PTSD. We empathize with their frustration and pain. This is followed by a discussion of how anger, despite its validity, can be an obstacle to engagement in the fear and distress that is also part of their response to the assault. The rationale is reviewed and the suggestion made to the patient that he or she try to fully engage emotionally in the exposure procedures and maximize the benefit of the treatment. Numbing and intolerance of anxiety. These characteristics, when present to an extreme degree, prevent effective emotional engagement in treatment. Our strategy in responding to them is similar: as always, the rationale for processing traumatic events via exposure is reviewed, and the importance of effective and prolonged emotional engagement is stressed. We then modify the exposure procedures so as to titrate the patient’s level of distress accordingly. If the patient is numbing during exposures and avoiding engagement, the therapist prompts for more details. This may include probe questions about details of what happened and what the patient was feeling, thinking, or sensing physically. In cases of intolerance of anxiety, the patient often feels overwhelmed by fear and fears losing control or experiencing prolonged flashbacks if he or she engages in the trauma memory. While it is important that the patient engage in the memory in order to process it, it is equally essential that the patient maintain a sense of his or her own safety in the therapist’s office. To accomplish this the patient must be firmly “grounded” in present reality. In ❉

165 ❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

cases of severe PTSD accompanied by severe flashbacks, grounding the patient can be difficult. Modifications of the exposure procedure to titrate the patient’s distress to manageable levels are often necessary. Such modifications include instructing the patient to keep his or her eyes open, using the past tense, making eye contact with and conversing with the patient about the story he or she is relating, and providing frequent, grounding-type comments (e.g., “you are safe here...the memory can not hurt you...I’m here to help you”). Maintaining focus on PTSD. As previously discussed, comorbidity is high in patients with chronic PTSD. It is also not unusual for patients with severe PTSD to have numerous crises during the treatment course, especially if their traumatic experiences began at an early age and PTSD has interfered with development of healthy coping skills. In such cases, excessive affective lability and intolerance, poor impulse control, self-destructive behaviors, suicidality, high conflict with significant others, and other problems can emerge and potentially derail the PTSD treatment. If careful assessment has determined that chronic PTSD is the primary problem, as opposed to borderline pathology, the general guideline is to attempt at all times to keep the PTSD in focus. When patient behaviors threaten personal safety or the safety of others, obviously the therapist must respond by attending to the current crisis. However, in general, the therapist will serve the patient best by adhering to the treatment procedures that will resolve PTSD symptoms rather than following each crisis. The most effective means of maintaining such focus is for the therapist to align with the patient’s desire to get well, to communicate to the patient at all times a firm and strong belief that the patient wants to reduce PTSD, and to lavishly praise every movement (however small) in the direction of healthy coping and treatment compliance. Aligning with the patient and building on his or her desire to get well can often be supported by labeling and externalizing the current crises as related to the PTSD rather than by interpreting crisis as “resistance” or “avoidance.” The latter is liable to result in a power struggle with the patient and to direct the focus of treatment away from PTSD. The choice to err in the direction of being overinclusive regarding the PTSD symptoms provides emotional support through the crisis, maintains the therapeutic alliance, and keeps PTSD in focus. The following is a detailed case description of a woman with chronic PTSD who was treated in the prolonged exposure program described previously. Her case illustrates many of the factors that make treatment difficult. ❉

166 ❉

❉

The Difficult-to-Treat Patient With Posttraumatic Stress Disorder ❉

Clinical Vignette History Ms. E was a 37-year-old, white, unemployed store clerk who was living with her boyfriend at the time of her treatment. She had been married and divorced twice. Her three children lived with their father, her first husband. Ms. E sought treatment for symptoms related to a prolonged assault perpetrated by a male acquaintance when she was 25 years old. He abducted her off the street, imprisoned her in his house for 4 days, and tortured her with repeated beatings and sexual assaults. He kept her tied up, injected her with drugs against her will, and threw knives at her. She was raped orally, vaginally, and anally. Ms. E, who was convinced that he intended to eventually kill her, escaped when her assailant left the house to buy food. After her escape, Ms. E was hospitalized for a period of time during which she received several surgeries for internal and facial injuries. Her assailant was apprehended, convicted, and sentenced to 20 years imprisonment. Ms. E was also raped when she was 12. Although she attempted suicide after this first assault, she said that she had not been disturbed by this earlier rape for many years. Additionally, Ms. E reported a history of severe physical and verbal abuse in her family while growing up and drug abuse throughout her teenage years. During the year following the abduction, Ms. E abused alcohol and a variety of drugs to the point of dependence. At some point after this year she quit using these substances and was able to push the horrors out of her mind because “my babies needed me.” She said that by an effort of will she made herself not think of it; she “wanted to get on with life.” Her strategy was successful for many years. She reported that for about 10 years she only thought of the trauma during the anniversary week. During those times, she said that the memories were too intense, and she would check herself into a motel and drink alcohol heavily. This changed when an attempted sexual assault on her 14-year-old son by a stranger triggered in Ms. E an onslaught of PTSD symptoms stemming from her own terrible ordeal. During the year prior to her treatment in our program, Ms. E sought mental health treatment and was initially diagnosed with bipolar disorder because she was unable to sleep, was emotionally labile and angry, and had severe flashbacks that were thought to be hallucinations. She was treated with lithium, neuroleptics, and several different SSRIs. She reported no benefit from the medications and believed that she had been misdiagnosed and ineffectively treated. At the time of her intake for the PTSD program, she reported extreme distrust of mental health professionals and had discontinued all psychotropic medications without consultation. She refused to consider further treatment with medications and stated at intake that this study (i.e., the PTSD treatment study she had seen advertised in the newspaper) was the last treatment she would risk.

❉

167 ❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

Initial Assessment Ms. E was diagnosed with PTSD (lifetime and current), recurrent major depression (current), alcohol dependence in early partial remission, and polydrug substance abuse with dependence in full remission. Her primary diagnosis was determined to be PTSD, although her depression was also quite severe (Beck Depression Inventory=45 [Beck et al. 1961]). Although she reported frequent thoughts of her own death, she denied any suicidal ideation or intent. Her global functioning was estimated at 35 on Axis V. She was unable to work and had sent her children to live with their father as she felt unable to care for them. Her social life was limited to interactions with her boyfriend and one neighbor. Her relationship with her boyfriend was characterized by great conflict especially surrounding sexual contact. She rarely left her house alone. The Difficult-to-Treat Patient Ms. E qualified as a difficult-to-treat patient for a number of reasons. First, she professed strong distrust of mental health professionals, which could easily have interfered with her accepting the rationale and complying with treatment procedures. Her hostility toward mental health professionals threatened the formation of a therapeutic alliance and increased the probability that anger would prevent emotional engagement. Second, the severity of her PTSD resulted in extreme distress and severe flashbacks, which were triggered by the treatment procedures themselves. She frequently became overwhelmed with fear and anxiety during exposure. She reported extreme fear of her own fear. Third, she became intensely angry at times, both in and out of treatment sessions. Fourth, despite an initial commitment to refrain from alcohol use, several times during her treatment she binged for a night. Alcohol Anonymous (AA) meetings were recommended at the beginning of treatment to help her abstain from drinking. Ms. E did not comply with this suggestion; she reported that she often went months without drinking alcohol and that AA had not been helpful to her in the past. Finally, she frequently came to therapy sessions in the midst of an emotional crisis. This made it difficult to implement tasks designated for specific sessions. In addition to her episodic drinking, she struggled with suicidality, felt hostile toward her partner, and felt anger at her therapist such that she considered stopping treatment. Course of Treatment Ms. E was treated with the standard prolonged exposure protocol described earlier in this chapter. Session length was on average 1½–2 hours long. Whereas providing treatment within a study required the therapist to adhere closely to the therapy protocol, working with Ms. E required great flexibility, and several sessions were extended out of clinical necessity. Additionally, her therapy was extended beyond the usual 12 sessions by 3 sessions to address all of the events contributing to her symptoms. The

❉

168 ❉

❉

The Difficult-to-Treat Patient With Posttraumatic Stress Disorder ❉

continual challenge for Ms. E’s therapist was to provide sufficient support and exposure to facilitate emotional processing of her trauma memories while simultaneously titrating the exposures to prevent her from feeling overwhelmed by distress or flashbacks. Establishing Therapeutic Alliance and Promoting the Rationale The first obstacle confronting the therapist was Ms. E’s distrust of mental health professionals. In her first session she clearly stated that she believed mental health professionals did not know what they were doing. She felt betrayed and misunderstood by having been diagnosed with bipolar disorder. She resented the medications and the “runaround” she had received. In the year prior to entering treatment she was seen by numerous therapists and several different psychiatrists, who prescribed a variety of medications. Throughout this process she insisted that she knew she “was not crazy.” However, at the same time she reported that she “felt crazy” because she could not control her own thoughts. The therapist responded to Ms. E’s barrage with sympathy and a confident and calm resolve that Ms. E had found a place that had diagnosed her difficulties and could provide her with expert treatment. Her problems were externalized as PTSD that had not surprisingly followed from her extreme trauma. The labeling of these problems as PTSD greatly reduced Ms. E’s distrust and anxiety. She became tearful as she admitted that she may have finally come to the “right place.” The therapist then presented the rationale for the treatment and procedures. Ms. E grasped the treatment rationale easily and immediately recognized that for years she had attempted to keep the memory from her mind. However, she reported that while she could “understand” the rationale, she could not “believe” it. She felt unable to accept that deliberately “processing” the trauma memory could free her from it. The therapist did not push for her to accept the rationale emotionally but requested only that she attempt to grasp it intellectually and suspend judgment. Ms. E expressed a willingness to try the therapy “because no one else has been able to help me.” The therapist’s tack of externalizing Ms. E’s difficulties and normalizing them within the context of PTSD without arguing about her previous treatment greatly reduced Ms. E’s distrust and aligned her with the part of Ms. E that wanted desperately to be well. Additionally, the therapist’s acceptance of Ms. E’s emotional reaction to the treatment rationale supported Ms. E while also clearly delineating the challenge and promise of the treatment. Determining a Manageable Starting Point The therapist and Ms. E developed a hierarchy of trauma memories that consisted of particular events that occurred during her terrible ordeal and that caused a high level of distress when recalled or spontaneously reexperienced. The top three memories of Ms. E’s hierarchy were the anal rape,

❉

169 ❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

having knives hurled at her, and being forcibly injected with large amounts of drugs. In her first imaginal exposure session, Ms. E predicted that she could manage the knife-throwing memory prior to beginning the reliving. She was provided with the standard instructions to close her eyes, to describe the event in the present tense, including events, thoughts, feelings and sensations while remembering that she was in the room with the therapist and was safe. Ms. E complied with the instructions and was almost immediately overwhelmed with flashbacks. She rubbed her wrists frantically trying to free them from “ropes” and repeatedly jerked around to see the “knives” as they flew past her. The therapist curtailed the procedure quickly and talked Ms. E through and out of the flashback. Exposure in the second session was adjusted to a less distressing point in the story, when Ms. E was actually picked up by her assailant in his car. At this time she again became overwhelmed with flashbacks from later points in the assault. At the third exposure session, still another point in the story was selected, the morning of her escape after her assailant had left. While she still experienced flashbacks during the retelling of this part of the story, she was able to maintain her sense of safety and repeatedly informed the therapist, “I’m okay. It’s just a flashback. I know I am here.” In subsequent sessions, the therapist worked on this part of the memory until Ms. E’s distress while retelling it was negligible and her flashbacks did not occur. Then the therapist progressed by moving backward from that point, usually finishing with the end part. It is notable that this memory reinforced the fact of her eventual escape to safety. In final sessions, Ms. E had developed enough control over her memories that she could work solely with the most distressing parts. Severe Flashbacks The most difficult obstacle confronting Ms. E’s therapist was the extremity of her reexperiencing symptoms and the overwhelming fear and anxiety they generated. Ms. E suffered from extended flashbacks that included strong “body memories.” That is, Ms. E’s flashbacks were not only visual and auditory but she also sometimes reexperienced the painful physical sensations that she had suffered during the assaults. The first challenge of working with such extreme symptoms was to educate Ms. E about the flashbacks and physical sensations: they did not signify “real danger” and were not evidence that she was loosing her mind, as she feared. Ms. E articulated her struggle with this fear clearly when she said “I know it’s not 12 years ago, but it feels like it is. How can I be having this confusion? I must be going crazy.” The symptoms Ms. E was experiencing were explained as common responses to extreme trauma. In every imaginal exposure session Ms. E was urged to engage the trauma memory while maintaining an awareness that she was with the therapist and was safe. Ms. E’s success at this task was enhanced by her developing ability to dismiss the reality of the physical sensations as “just a flashback” and by the therapist’s success in titrating the exposure experience and keeping flashbacks to a minimum.

❉

170 ❉

❉

The Difficult-to-Treat Patient With Posttraumatic Stress Disorder ❉

Titrating Exposure by Modifying Procedures and Promoting Grounding Because of the extreme flashback Ms. E experienced during her initial session of imaginal exposure, her therapist immediately implemented a standard modification by instructing her to keep her eyes open and describe her traumatic experiences in a conversational manner, including usage of the past tense. This procedure was continued in subsequent imaginal exposures throughout the therapy. Second, when Ms. E’s rising distress level or the beginnings of a flashback threatened her sense of control, the therapist repeatedly instructed Ms. E to “look at me” in a firm voice to maintain her awareness of being in the room with the therapist. A third standard tactic utilized by Ms. E’s therapist was to instruct Ms. E to use the breathing retraining skill, taught to her in session 1, during the imaginal exposure. This helped Ms. E to manage the experience without spiraling into panic. A fourth modification adopted by Ms. E’s therapist was to maximize Ms. E’s sense of safety by frequently talking to her while she told her story, saying “I’m here with you...You are safe...This is not happening now...You are doing a really good job.” Ms. E reported that she was able to use the therapist’s voice to stave off the flashbacks. Ms. E’s therapist also used other “grounding” techniques to help her to manage her flashbacks. Ms. E was able to maintain her sense of the present reality better when she was able to touch something tangible. Her therapist provided her with a variety of articles during the imaginal exposure such as clay for her to mold, a stress ball, or a towel that she could wring. Ms. E also reported that it helped her if she could move while telling her story, so during particularly difficult work Ms. E would tell her story to her therapist while walking outside so that Ms. E could talk while walking and looking at the trees around her. A final modification of the imaginal exposure was to harness Ms. E’s considerable skill at imagery and direct it in a helpful vein. The therapist identified the beneficial power of Ms. E’s imagery skills during her first exposure session when Ms. E was lost in a flashback. In order to ground Ms. E, the therapist attempted to engage her in a conversation about her children. Ms. E mentioned that her son had won a soccer game the day before. The therapist directed her to image and describe the game. Ms. E soon began to smile and then reported that the flashback had dissipated. From then on, the therapist urged Ms. E to end spontaneous, distressing flashbacks by deliberately imaging a more pleasing scene. Later the therapist suggested to Ms. E that she “put the flashback on a television screen.” When Ms. E successfully accomplished this, Ms. E was instructed to “turn the television off” and “watch the image fade as the screen goes black.” Again, Ms. E was able to accomplish this. Ms. E’s success with imagery to control the flashbacks prompted her to elaborate the strategy on her own. When listening to her exposure tapes between sessions Ms. E reported that she imaged the events “as if I were standing behind a large picture window, watching it outside.” If she was

❉

171 ❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

dealing with a particularly distressing part, Ms. E would “put window panes up in front of the window.” This action created more distance from the memory and provided Ms. E with a greater sense of control. Finally, when Ms. E felt she was in danger of being “sucked in” she would place her hands on the imaginary glass so that she knew she was not in the picture. These modifications helped Ms. E to approach even the most difficult portions of the memory without losing her sense of safety. Communicating That Flashbacks Are Not Dangerous As mentioned earlier, Ms. E’s flashbacks were severe. Typically on telling a portion of her story for the first or second time, Ms. E would reexperience many of the physical sensations that had accompanied that particular incident. The most extreme example of this was when she was describing being injected with heroin and cocaine. The first time she told this portion of the memory, Ms. E was immediately overcome with nausea so strong that she vomited as she had when she was injected. When Ms. E vomited the therapist provided her with a trashcan and some tissues and laid a hand gently on her back. She sympathetically urged her to deal with the flashback in a calm and soothing voice. Ms. E’s therapist never responded to her flashbacks with fear. She always matter-of-factly labeled the experience as a flashback or a body memory and urged Ms. E to breathe, prompting her to manage the experience with “You are here with me, you are safe,” “What can you do to make yourself feel better?” and “You are OK, just try to go on.” Similarly, when the flashback was over, the therapist would praise Ms. E for her ability to soothe herself and regain her sense of reality. Importantly, Ms. E’s therapist never responded to the flashbacks as a reason to discontinue the therapy but rather communicated a desire to help Ms. E manage them so that the processing work might continue. Notably, Ms. E began to regard her flashbacks as not dangerous and did not let their presence frighten her from approaching the memory. Frequently, after the third or fourth retelling, the physical sensations abated. Keeping PTSD in Focus As mentioned earlier, Ms. E frequently opened the session with a crisis. She relapsed to episodic alcohol abuse, had ongoing domestic conflicts, had suicidal crises, and felt hostile to her therapist. With each crisis, the therapist responded by empathizing with the patient’s distress, normalizing her response by suggesting that the crisis was that much more difficult for her because of her PTSD symptoms, praising her for coming to the session despite her understandable distress, and challenging Ms. E to not let the crisis distract her from her treatment goals. The therapist would then initiate brief problem solving with Ms. E so that the crisis would not continue to “get in her way.” The therapist utilized written contracts to promote abstinence from alcohol and in response to sporadic suicidal ideation. For about 1 week mid-

❉

172 ❉

❉

The Difficult-to-Treat Patient With Posttraumatic Stress Disorder ❉

way through treatment Ms. E’s suicidal urges became strong, and the option of hospitalization was discussed and framed as the best treatment course if she truly believed she could not manage her current distress without harming herself, but that this would derail the current treatment plan. Ms. E elected to contract for safe behavior and keep working on the PTSD. Several times the therapist met with Ms. E and her partner to educate him about PTSD symptoms, explain treatment, and coach him to engage in more supportive behaviors throughout the course of treatment. When Ms. E became excessively angry, the therapist validated her anger and also suggested to Ms. E that it was an indication that the emotional processing of her trauma was occurring and this signified good progress. She additionally explained to Ms. E that while her anger about the assaults was entirely justified, the anger could interfere with her processing of other emotions and prevent her from reaching the goals she wanted. In each case, the therapist treated Ms. E’s report of distress with empathy and respect and did not second guess her motivation. Instead the therapist built on the assumption that Ms. E wanted to get better and that these crises were real problems interfering with the work that Ms. E wanted to do. This consistent communication tended to align the therapist with Ms. E and to redirect Ms. E toward problem solving that allowed her to refocus on her PTSD treatment. Treatment Outcome Ms. E was evaluated at the conclusion of the research protocol (session 12). This posttreatment assessment was followed by three additional “booster” sessions working with parts of the trauma memory that remained to be addressed. At the time of her postassessment, which unfortunately coincided with the anniversary week of her abduction, Ms. E’s symptoms had not improved from her pretreatment levels. Her scores on the PSS-I (Post-Traumatic Stress—Interview version) remained exactly the same, with a score of 41. While her self-report measures indicated a slight drop in PTSD symptoms (about 16% reduction in severity), they showed a more substantial decrease in depression (Beck Depression Inventory) and general anxiety (Beck Anxiety Inventory). However, Ms. E saw herself as much improved and more in control of herself and her symptoms. She reported that for the first time she felt equipped with “tools” to use to manage her PTSD. Approximately 5 weeks after the posttreatment evaluation, Ms. E’s level of depression became significantly worse again. She was referred for psychiatric evaluation and ongoing outpatient therapy with a new therapist. She was prescribed sertraline and attended weekly counseling sessions with a broader focus than that provided in the PTSD study. At the 3-month follow-up evaluation, her symptoms reflected a significant improvement. At that point her PTSD symptom severity was 70% below pretreatment level and she no longer met criteria for PTSD. While we can not discern the relative contribution of each intervention, it seems reasonable to conclude that some combination of exposure treatment, medi-

❉

173 ❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

cation, and ongoing psychotherapy proved effective. It is interesting to note that prior treatment with SSRIs had been ineffective for Ms. E. At her 1-year follow-up evaluation, Ms. E’s PTSD symptom severity had increased again but remained about 40% below her pretreatment level. This was despite again being evaluated in the week of the anniversary of her abduction. At this time she met criteria for current PTSD, but her self-reported depression and general anxiety remained quite low.

Conclusion PTSD is a complex and often difficult-to-treat psychiatric disorder. Empirically validated psychosocial and pharmacological treatment interventions yield a significant number of patients with minimally improved or poor outcome. Patients such as Ms. E suggest that combined therapeutic intervention often may be most effective. Studies are clearly needed that assess the efficacy of medication, cognitive-behavioral treatment, and their combination. It is also clear that we need to expand our knowledge of predictors of treatment response and outcome. We need to pay greater attention to the individual characteristics of those who suffer from chronic PTSD and to the interaction of these characteristics with treatment interventions. As discussed by Bowman (1999) and others, the prevalence of exposure to seriously traumatic events is probably far greater than the prevalence of PTSD. Individual differences in biological vulnerability, emotionality and expressiveness, cognitive schemas and beliefs, prior traumatic experience, resilience, and coping strategies mediate the impact of traumatic events. Gaining a better understanding of these differences, and how they affect the development of PTSD as well as treatment response, will enable more effective interventions.

References American Psychiatric Association: Diagnostic and Statistical Manual of Mental Disorders, 4th Edition. Washington, DC, American Psychiatric Association, 1994 Beck AT: Cognitive Therapy and the Emotional Disorders. New York, International Universities Press, 1976 Beck AT, Ward CH, Mendelsohn M, et al: An inventory for measuring depression. Arch Gen Psychiatry 4:561–571, 1961 Boudewyns PA, Hyer L, Woods MG, et al: PTSD among Vietnam veterans: an early look at treatment outcome using direct therapeutic exposure. J Trauma Stress 3:359–368, 1990

❉

174 ❉

❉

The Difficult-to-Treat Patient With Posttraumatic Stress Disorder ❉

Bowman ML: Individual differences in posttraumatic distress: problems with the DSMIV model. Can J Psychiatry 44:21–33, 1999 Brady K, Pearlstein T, Asnis GM, et al: Effeciacy and safety of sertraline treatment of posttraumatic stress disorder: a randomized controlled trial. JAMA 283:1837– 1844, 2000 Breslau N, Davis GC, Andreski P, et al: Traumatic events and posttraumatic stress disorder in an urban population of young adults. Arch Gen Psychiatry 48:216– 222, 1991 Burton JK, Marshall RD: Categorizing fear: the role of trauma in a clinical formulation. Am J Psychiatry 156:761–766, 1999 Butterfield MI, Becker ME, Connor KM, et al: Olanzapine in the treatment of PTSD: a pilot study. (in press) Canive JM, Clark RD, Calais LA, et al: Buproprion treatment in veterans with posttraumatic stress disoder: an open study. J Clin Psychopharamcol 18:379–383, 1998 Connor KM, Sutherland SM, Tupler LA, et al: Fluoxetine in post-traumatic stress disorder: randomised, double-blind study. Br J Psychiatry 175:17–22, 1999a Connor KM, Davidson JRT, Weisler RH, et al: A pilot study of mirtazepine in posttraumatic stress disorder. Int Clin Psychopharmacol 14:29–31, 1999b Cooper NA, Clum GA: Imaginal flooding as a supplementary treatment for PTSD in combat veterans: a controlled study. Behavior Therapy 20:381–391, 1989 Davidson JRT, Kudler HS, Smith R, et al: Treatment of posttraumatic stress disorder with amitriptyline and placebo. Arch Gen Psychiatry 47:259–266, 1990 Davidson JRT, Kudler HS, Saunders WB, et al: Predicting response to amitriptyline in posttraumatic stress disorder. Am J Psychiatry 150:1024–1029, 1993 Davidson JRT, Rothbaum BO, van der Kolk B, et al: Multi-center, double-blind comparison of sertraline and placebo in the treatment of posttraumatic stress disorder. Arch Gen Psychiatry (in press) Eidelman I, Seedat S, Stein DJ: Risperidone in the treatment of acute stress disorder in physically traumatized inpatients. Depress Anxiety 11:187–188, 2000 Foa EB, Cahill S: Specialized treatment for PTSD: matching survivors to the appropriate modality, in Treating Trauma Survivors With PTSD: Bridging the Gap Between Intervention Research and Practice. Edited by Yehuda R. Washington, DC, American Psychiatric Press (in press) Foa EB, Jaycox LH: Cognitive-behavioral theory and treatment of posttraumatic stress disorder, in Efficacy and Cost-Effectiveness of Psychotherapy. Edited by Spiegel D. Washington, DC, American Psychiatric Press, 1999, pp 23–61 Foa EB, Kozak MJ: Emotional processing of fear: exposure to corrective information. Psychol Bull 99:20–35, 1986 Foa EB, Meadows EA: Psychosocial treatments for post-traumatic stress disorder: a critical review, in Annual Review of Psychology. Edited by Spence J, Darley JM, Foss DJ. Palo Alto, CA, Annual Reviews, 1997 pp 449–480

❉

175 ❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

Foa EB, Riggs DS: Posttraumatic stress disorder and rape, in American Psychiatric Press Review of Psychiatry, Vol 12. Edited by Oldham JM, Riba MB, Tasman A. Washington, DC, American Psychiatric Press, 1993, pp 273–303 Foa EB, Rothbaum BO: Treating the Trauma of Rape. New York, Guilford, 1998 Foa EB, Steketee G, Rothbaum B: Behavioral/cognitive conceptualizations of posttraumatic stress disorder. Behavior Therapy 20:155–176, 1989 Foa EB, Rothbaum BO, Riggs D: Treatment of posttraumatic stress disorder in rape victims: a comparison between cognitive-behavioral procedures and counseling. J Consult Clin Psychol 59:715–723, 1991 Foa EB, Riggs DS, Massie ED, et al: The impact of fear activation and anger on the efficacy of exposure treatment for PTSD. Behavior Therapy, 26:487–499, 1995 Foa EB, Dancu CV, Hembree EA, et al: The efficacy of exposure therapy, stress inoculation training, and their combination in ameliorating PTSD for female victims of assault. J Consult Clin Psychol 67:194–200, 1999a Foa EB, Davidson JRT, Frances A: The Expert Consensus Guidelines Series: treatment of posttraumatic stress disorder. J Clin Psychiatry 60:4–76, 1999b Griffin MG, Resick PA, Mechanic MB, et al: Objective assessment of peritraumatic dissociation: psychophysiological indicators. Am J Psychiatry 154:1081–1088, 1997 Hamner M, Ulmer H: Risperidone for positive symptoms of psychosis in PTSD: a preliminary open trial (abstract no 160). Poster presented at the NCDEU 38th Annual Meeting, Boca Raton, FL, 1998 Hertzberg MA, Feldman ME, Beckham JC, et al: Lack of efficacy for fluoxetine in PTSD: a placebo controlled trial in combat veterans. Annual Clinical Psychiatry 12:101–105, 2000 Horowitz MJ: Stress-Response Syndromes. Northvale, NJ, Aronson, 1976 Jaycox LH, Foa EB: Obstacles in implementing exposure therapy for PTSD: case discussions and practical solutions. Clinical Psychology and Psychotherapy 3: 176– 184, 1996 Jaycox LH, Foa EB, Morral A: The influence of emotional engagement and habituation on exposure therapy for PTSD. J Consult Clin Psychol 66:185–192, 1998 Judge R, Martenyi F, Brown E, et al: Fluoxetine versus placebo in posttraumatic stress disorder. New Research Poster presented at the ACNP Annual Meeting, San Juan, PR, December 11, 2000 Keane TM, Fairbank JA, Caddell JM, et al: Implosive (flooding) therapy reduces symptoms of PTSD in Vietnam combat veterans. Behavior Therapy 20:245–260, 1989 Kessler RC, Sonnega A, Bromet E, et al: Posttraumatic stress disorder in the National Comorbidity Survey. Arch Gen Psychiatry 52:1048–1060, 1995 Kosten TR, Frank JB, Dan E, et al: Pharmacotherapy for posttraumatic stress disorder using phenelzine or imipramine. J Nerv Ment Dis 179:366–370, 1991 Lehrer B, Bleich A, Kotter M, et al: Posttraumatic stress disorder in Israeli combat veterans. Arch Gen Psychiatry 44:976–981, 1987

❉

176 ❉

❉

The Difficult-to-Treat Patient With Posttraumatic Stress Disorder ❉

Lohr JM, Tolin DF, Lilienfeld SO: Efficacy of eye movement desensitization and reprocessing. Behavior Therapy 29:123–156, 1998 Londborg PD, Hegel MT, Goldstein S, et al: Sertraline treatment of posttraumatic stress disorder: results of 24 weeks of open label continuation treatment. J Clin Psychiatry (in press) Marks I, Lovell K, Noshirvani H, et al: Treatment of posttraumatic stress disorder by exposure and/or cognitive restructuring. Arch Gen Psychiatry 55:317–332, 1998 Marshall RD, Pierce D: Posttraumatic stress disorder: recent findings and implications for pharmacotherapy. Harvard Review of Psychiatry Jan/Feb:247–256, 2000 Marshall RD, Schneier FR, Fallon BA, et al: An open trial of paroxetine in patients with noncombat-related chronic PTSD. J Clin Psychopharmacol 18:10–18, 1998a Marshall RD, Beebe KL, Oldham M, et al: Efficacy and safety of paroxetine treatment of chronic PTSD: a fixed-dosage, multi-center, placebo-controlled study. Scientific presentation, International Society of Traumatic Stress Studies, San Antonio, TX, November 18, 2000a Marshall RD, Schneier F, Simpson B, et al: Interim Report of a controlled trial of paroxetine in PTSD. Symposium presentation, International Society for Traumatic Stress Studies, San Antonio, TX, November 18, 2000b Meichenbaum D: Self-instructional methods, in Helping People Change. Edited by Kanfer FH, Goldstein AP. New York, Pergamon, 1975, pp 357–391 Miller RW, Rollnick S: Motivational Interviewing: Preparing People To Change Addictive Behavior. New York, Guilford, 1991 Monti PM, Abrams DB, Kadden RM, et al: Treating Alcohol Dependence. New York, Guilford, 1989 Nagy LM, Southwick SM, Charney DS: Placebo-controlled trial of fluoxetine in PTSD. Article presented at the International Society for Traumatic Stress Studies Annual Meeting, San Francisco, CA, November 9–13, 1996 Neal LA, Shapland W, Fox C: An open trial of moclobemide in the treatment of posttraumatic stress disorder. Int Clin Psychopharmacol 12:231–237, 1997 Physicians’ Desk Reference, 54th Edition. Montvale, NJ, Medical Economics, 2000 Reist C, Kauffman CD, Haier RJ, et al: A controlled trial of desipramine in 18 men with posttraumatic stress disorder. Am J Psychiatry 146:513–516, 1989 Schwartz JM, Stoessel PW, Baxter LR, et al: Systematic changes in cerebral glucose metabolic rate after successful behavior modification treatment of obsessivecompulsive disorder. Arch Gen Psychiatry 53:109–113, 1996 Shapiro F: Eye Movement Desensitization and Reprocessing: Basic Principles, Protocols, and Procedures. New York, Guilford, 1995 Shestatsky M, Greenberg D, Lerer B: A controlled trial of phenelzine in posttraumatic stress disorder. J Psychiatr Res 24:149–155, 1988 Van der Kolk BA, Dreyfuss D, Michaels M, et al: Fluoxetine in posttraumatic stress disorder. J Clin Psychiatry 55:517–522, 1994

❉

177 ❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

Veronen LJ, Kilpatrick DG: Stress management for rape victims, in Stress Reduction and Prevention. Edited by Meichenbaum D, Jaremko ME. Boulder, CO, Perseus Publishing, 1983, pp 341–374 Yehuda R, Marshall RD, Giller EL: Psychopharmacological treatment of posttraumatic stress sisorder, in A Guide to Treatments That Work. Edited by Nathan PE, Gorman JM. New York, Oxford University Press (in press)

❉

178 ❉

7 The Difficult-to-Treat Patient With Borderline Personality Disorder Mary C. Zanarini, Ed.D. Kenneth R. Silk, M.D.

M

any clinicians believe that the terms difficult and borderline are synonymous. In addition, the life-and-death struggles and the intense transferencecountertransference paradigms that often arise in the treatment of borderline patients make many clinicians reluctant to engage such patients in psychotherapy and/or pharmacotherapy. There is an assumption underlying this view that all borderline patients are equally disturbed. However, clinical experience suggests that there is a continuum of borderline psychopathology, and for heuristic purposes, we propose three distinct subtypes of borderline patients. Type I patients have mild cases of borderline personality disorder (BPD) or perhaps more aptly, borderline trait disorder. These patients manifest the same dysphoria, the ❉

179 ❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

same cognitive disturbances, and the same interpersonal difficulties as more severely ill borderline patients. However, what distinguishes them is their lack of impulsivity, particularly in the areas of self-mutilation and suicidal efforts, and their greater ability to use the treatment relationship to enhance their functioning in the wider world. Type II borderline patients are intermittently self-destructive, particularly when they are fearful of being abandoned by someone on whom they depend (Gunderson 1984). However, they function well for months or even years at a time, as long as they feel stably “held” in at least one important relationship. These borderline patients are also able to use a therapeutic relationship well, although they are typically more fragile and rely on the therapeutic relationship to fulfill more of their emotional needs than the Type I borderline patients described above. Put another way, Type I borderline patients want to understand and overcome their problems so that they can function better and fulfill the goals that they have for themselves. Type II borderline patients, on the other hand, have often come to a more limited adjustment after a very difficult life struggle and see their treatment as their lifeline to stability. They too wish to understand their problems but may feel it is impossible for them to overcome their problems. Type III borderline patients lead very chaotic lives, with areas of strength intermingled with a wide-ranging and chronic pattern of selfdefeating behaviors. These patients typically use a tremendous amount of psychiatric treatment and over the course of their disorder may well give up both their determination and ability to function in the real world. Many of these patients abandon the structure of work or school and end up supporting themselves on disability. They may also relinquish important relationships, such as those with a spouse or children, and end up living lives of almost complete social isolation. In their case, months of functioning well are often interspersed with years of varying degrees of serious dysfunction. The only relevant epidemiological study has estimated that at least 2% of American adults between the ages of 19–55 meet research criteria for BPD (Swartz et al. 1990). However, there are no good figures on what percentage of borderline patients, 75% of whom are female (Widiger and Weissman 1991), fall into each of these three groups. It is highly probable that almost all of the research on BPD has dealt with Type II and Type III borderline patients. This is so because most of this research has been conducted at tertiary care facilities, which until recently have focused on the inpatient treatment of BPD. In this chapter, we deal with the last of these patient groups. We focus ❉

180 ❉

❉

The Difficult-to-Treat Patient With Borderline Personality Disorder

❉

on the most severely ill borderline patients because it is Type III patients who use the most mental health care resources and who lead to the most distress among clinicians. We will try to answer the following two related questions: First, what makes these patients so difficult to treat? In this regard, we review affective and cognitive symptoms as well as self-destructive forms of impulsivity and interpersonal survival strategies that have outlived their usefulness. Second, what can be done to deal with these vexing areas of psychopathology? In this regard, we focus on the role of psychotherapy, pharmacotherapy, and psychiatric hospitalizations in the treatment of Type III borderline patients.

Particularly Problematic Sectors of Psychopathology Chronic and Intense Dysphoria Borderline patients suffer from a range of intense dysphoric affects (Gunderson and Kolb 1978; Zanarini et al. 1990b). These affects include depression and sorrow; anger and rage; anxiety and panic; and feelings of helplessness, hopelessness, worthlessness, emptiness, and loneliness. What distinguishes borderline patients from other patient groups is the number of dysphoric affects they feel at the same time and the overall amplitude of this pain (Zanarini et al. 1998a). It is difficult to know what causes these varied and shifting affects. It may be that they are the sequelae of childhood experiences of loss, neglect, and/or abuse (For a review of this literature, see Zanarini and Frankenburg 1997). Alternatively, these affects may represent the underlying hyperbolic temperament that one of us (M.C.Z.) has suggested is core to the borderline diagnosis (Zanarini and Frankenburg 1994). In this view, borderline patients have a temperament that leads them to convert unbearable feelings of rage, sorrow, shame, and/or terror into unremitting attempts to get others to pay attention to the enormity of the emotional pain that they feel. These attempts are usually indirect and involve a covert reproach of the listener’s “insensitivity,” “stupidity,” or “malevolence.” It may also be that these dysphoric affects are due to some type of biological dysfunction. For example, the results of biochemical studies have typically found reduced serotonergic activity in criteria-defined borderline patients (Coccaro et al. 1989; Hollander et al. 1994). In reality, these factors are not totally separate, and thus the tendency toward intense dysphoria may be due to a combination of all three factors. ❉

181 ❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

One of the most difficult aspects of the dysphoria of borderline patients is that clinicians often feel compelled to try to lessen its intensity or to eliminate it. Clearly, the problem with this is that no medication is going to “cure” someone of BPD, and medication may not even affect the target symptom at which it is aimed. (We discuss the pharmacotherapy of borderline patients later in this chapter.) Another reaction is to mistake the dysphoria (and the lability) of borderline patients for any one of a number of comorbid conditions. Research clearly indicates that borderline patients are prone to a number of Axis I disorders, particularly unipolar mood disorders, substance use disorders, posttraumatic stress disorder, and eating disorders (Zanarini et al. 1998b). This too may lead some clinicians to mistake BPD for some type of comorbid condition. This is problematic as it leaves many aspects of BPD untreated, particularly those of an interpersonal and a temperamental nature. Additionally, it often comes as a surprise to these clinicians how treatment resistant these symptoms are. This may lead to heroic efforts to medicate away what is really the patient’s “dis-ease.” Clearly, patients who do not feel that people care about them or hear them are not going to be happy that someone, however well intentioned, is trying to take away their pain. In response to these efforts, borderline patients may develop other symptoms or seem to have their Axis I pathology transformed into yet another disorder. The best approach to dealing with this type of intense and shifting dysphoria is to empathize with how hard it is to deal with these feelings and to acknowledge the likelihood that they will persist, to one degree or another, for the foreseeable future. This lets the patient know that others are aware of his or her suffering. It also helps to reassure the severely ill borderline patient who typically is fearful that his or her pain will both never end and end too soon. This does not mean that medications should not be used if these feelings escalate into a treatable Axis I disorder. It does mean that grief will probably not yield to pharmacotherapy, and, as in treating other bereaved patients, patience is more likely to win the day than aggressive polypharmacy.

Cognitive Problems Borderline patients suffer from three levels of cognitive symptomatology: 1) troubling but nonpsychotic problems, such as overvalued ideas of worthlessness and guilt, experiences of depersonalization and derealization, and nondelusional suspiciousness and ideas of reference; 2) quasipsychotic or psychoticlike symptoms (i.e., transitory, circumscribed, and somewhat real❉

182 ❉

❉

The Difficult-to-Treat Patient With Borderline Personality Disorder

❉

ity-based delusions and hallucinations); and 3) genuine delusions and hallucinations. The last category is rare and almost always occurs in the context of a psychotic depression (Pope et al. 1985; Zanarini et al. 1990a). The other two categories or levels are ongoing problems for many severely disturbed borderline patients. The literature is contradictory with regard to the issue of whether these symptoms are trauma related (Brodsky et al. 1995; Shearer 1994; Zanarini et al. 2000; Zweig-Frank et al. 1994a, 1994b). The best evidence seems to suggest that most severely disturbed borderline patients suffer from these symptoms but that these symptoms are more intense in borderline patients reporting childhood histories of sexual abuse. Thus, the symptoms seem both intrinsic to the disorder and stress related. As for managing these symptoms, the cognitive distortions implicit in overvalued ideas can be explored. Clearly, it is not uncommon for those who have been the object of some type of neglect and/or abuse, however subtle or commonplace, to blame themselves for what has gone wrong in their lives and the lives of those they love and on whom they depend. Their suspiciousness may also have a basis in reality. When the high rates of childhood abuse and neglect reported by borderline patients in 10 studies were unknown (Herman et al. 1989; Links et al. 1988; Ogata et al. 1990; Paris et al. 1994a, 1994b; Salzman et al. 1993; Shearer et al. 1990; Westen et al. 1990; Zanarini et al. 1989, 1997) and psychodynamic theories posited that the etiology of BPD could be found in subtle problems in parenting, their distrust of others was thought of as a near psychotic symptom (Adler and Buie 1979; Kernberg 1975; Masterson 1972). Today, their mistrust and suspiciousness seems more justified or at least understandable, and clinicians are hopeful that patients with BPD can be helped to distinguish between the dangers of their childhood and the usually safer circumstances of adulthood. Quasipsychotic symptoms in borderline patients may be related to the felt belief that they have been abandoned; in this view, the symptoms can be seen as a form of cognitive restitution, providing contact with a loved one whom the patient fears she has lost (Gunderson 1984). It may also be that the severely disturbed borderline patient is suffering from an unrecognized major depression or is reliving some type of body memory as a consequence of physical and/or sexual abuse or assault that may have occurred in childhood and/or adulthood. It may also be that the patient has learned to signal others that he or she is worried about his or her pain being forgotten by complaining of psychotic or psychoticlike symptoms. This pattern and the view that these symptoms are acts of restitution can be handled by pointing out the pattern and its meaning for the patient. They can also be helped by adding more support and ❉

183 ❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

structure to the patient’s life. The severely ill borderline patient will not abandon these symptoms easily or quickly, but at least they will be able to be discussed. Body memories can be normalized by letting the patient know that they are common among adults who have been abused or neglected in severe ways. Quasipsychotic experiences related to depression may need to be treated with medication. However, it is clear that all of these factors may be at play in the sort of psychotic symptoms exhibited by severely disturbed borderline patients. An approach that combines useful new information, more support, and, possibly, medications is necessary.

Self-Mutilation Self-mutilation in borderline patients is both an unusual form of self-soothing and an indirect, although very effective, manner of expressing rage. It goes against everything therapists believe in and is often mistakenly taken as a personal affront. This is, in part, the way the patient intended it. But in a larger sense, it is meant to protect the therapist (and others who the patient loves and needs) from the ravages of the patient’s rage and self-hatred; a rage and self-hatred that is truly excoriating in nature. Borderline patients may cut and burn themselves because they are dissociated and need to “feel real,” or because they need to relieve a tremendous amount of anxiety. They also hurt themselves as a way of managing a murderous degree of frustration and rage. Rather than wasting time feeling upset or even horrified, therapists wishing to work with severely disturbed borderline patients need to remember that, in many cases, this type of behavior began in childhood. At that time, it had nothing to do with therapy and was secretive in nature. Rather than seeing self-harm as an iatrogenic behavior with all its meaning tied to treatment, it is probably more accurate to see it as 1) a long-standing form of self-soothing, 2) a protective reaction in which the patient’s volcanic rage is turned back upon him- or herself, and 3) as an addiction, with all the power that implies. As one 29-year-old borderline patient said recently: “I plan to give up my eating disorder when I turn 30, smoking when I turn 40, and cutting and burning when I turn 50. At least I’m trying, aren’t I?” The therapist’s first task is to empathize with the pain that led to selfinjurious behavior, whereas the second task is to make it plain to the patient that he or she will not struggle with the patient over the patient’s self-mutilative efforts. If the self-harm is medically insignificant, the patient should see a general or emergency room physician. However, if the self-harm is medically serious and the patient can not control it, a brief hospitalization is ❉

184 ❉

❉

The Difficult-to-Treat Patient With Borderline Personality Disorder

❉

indicated, founded on a previously discussed and mutually agreed upon treatment plan. Once hospitalized, the patient can have time to think and “cool down,” a medication consultation can take place, and the patient and therapist can have a brief but needed respite from one another. The third thing to do is to act in a respectful manner toward this powerfully addictive form of behavior. Much as alcoholics do not give up drinking until they are ready, borderline patients will not give up self-mutilation until they have found something or someone who can give them more of the comfort they need and crave. Much as adolescents can not be cajoled into growing up faster because it is more convenient for their parents, borderline patients can not be nagged into giving up a form of self-soothing than can be both physically agonizing and the source of feelings of euphoria. This should be one of the goals of treatment, not a precondition for undertaking such a venture.

Suicidality Many borderline patients threaten suicide on a regular basis when they are worried about being abandoned (Gunderson 1984). Others make numerous suicide gestures. Serious attempts are not infrequent, and anywhere from 3% to 10% of borderline patients with a history of being hospitalized at least once go on to commit suicide (McGlashan 1986; Paris et al. 1987; Stone 1990). The treater of a seriously disturbed borderline patient needs to take a careful history of the patient’s suicidality, both threats and efforts. If a patient is prone to suicide threats but has no history of gestures or attempts, a useful approach is to ask what the patient thought he or she was conveying by saying: “If my mother isn’t more understanding in the future, I am going to take the family gun and shoot myself at Thanksgiving dinner.” This borderline patient may believe he or she was talking about frustrations with his or her mother and may not realize that he or she was subtly shifting responsibility for his or her anger to the listener, typically the therapist. Often a psychoeducational approach that lets the patient “hear” how he or she sounds and that provides the patient with alternative ways of expressing his or her feelings will be very helpful, if not openly appreciated. Much the same approach can be taken with the borderline patient who takes several aspirin and then goes to an emergency room or cuts him- or herself with a plastic knife and hopes to die. However, a very different approach needs to be taken with the borderline patient with a history of serious suicide attempts. Often such a patient will have a striking history of mood ❉

185 ❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

disorder (typically unipolar in nature), an ongoing substance abuse problem, and a family history of suicidality. All three of these factors need to be taken into account when contingency plans are being made. When dealing with this triad of risk factors, the patient should be encouraged to get aggressive treatment for both the mood disorder and substance abuse. All too often clinicians overlook a borderline patient’s emerging major depression because it arises insidiously out of his or her chronic dysphoria, and time and again therapists make far too little of their borderline patients’ drinking problems or abuse of prescription anxiolytics. Clearly, these patients need to get clean and sober and stay that way until their lives are under better control. Preliminary data from a longitudinal study of BPD that one of us (M.C.Z.) is conducting suggest that borderline patients who episodically abused alcohol during their early 20s may be able to drink socially in the future. However, chronic abusers need to seek out specialized treatment for their substance abuse problems (e.g., Linehan’s treatment aimed at decreasing drug abuse among borderline women [Linehan et al. 1999]). Only after sobriety is achieved and maintained for a substantial amount of time should a more traditional psychotherapy be undertaken. In addition, lifetime sobriety must be a goal for these patients. For the borderline patient with a history of deep and prolonged depressions associated with serious suicide attempts, it is important to empathize with the difficulty of managing such a serious mood disorder in addition to all of the patient’s other problems. It is both frightening and discouraging to patient and clinician that depressive episodes keep recurring and that they may be deeper and more prolonged as time goes by. Often simply having a clinician address the implications of having a personal and family history of depression and suicidality will be a relief to a borderline patient. Relentlessly looking for interpersonally based precipitating factors and encouraging the patient to believe that one day he or she will no longer have to worry about being depressed may only be making the patient feel more discouraged and alone. It may well be that the patient will not get depressed again if all goes well for him or her, but it may be liberating to share the possibility that this is a life long struggle and one in which the therapist is prepared to participate. Clearly, there is no known way of predicting which seriously ill borderline patients will commit suicide. Plainly, it is a bad prognostic sign if a family member has tried or succeeded in committing suicide. To the patient, this represents the breaking of a taboo, such that what was once unthinkable now enters the realm of the acceptable. Often times a seriously ill borderline patient will begin to play a game ❉

186 ❉

❉

The Difficult-to-Treat Patient With Borderline Personality Disorder

❉

of “cat and mouse” with the therapist (and with his or her own life) around the issue of suicidality. What can and should be done in such a situation? Termination always comes to mind and may realistically be an option if carried out in an appropriate and ethical manner. Bringing other people onto the treatment team, getting a consultation, and convincing the patient to live in a residential center are probably more helpful. It is also important to talk with such a patient about the real possibility that the patient will die as a result of all of his or her suicidal efforts and for the therapist to let the patient know that the therapist would miss him or her, although the therapist will certainly go on. Although some clinicians might feel uncomfortable with this degree of self-disclosure, our experience suggests that most severely disturbed borderline patients have no idea of the affection and respect in which they are held. On the other hand, it is important to address any misconceptions about who would actually be dying by reminding the patient that any vengeful feelings he or she has toward the clinician or clinicians will only be very partially satisfied. If the patient has children, you may also want to discuss with the patient how his or her death might affect them, particularly given his or her own knowledge of the pain of being abandoned. In the most refractory of circumstances, anticipatory mourning (or a deeply felt acknowledgment by patient and clinician alike of the grief that is sure to accompany suicide) may help to alleviate some of the desperation driving such a patient. It also allows the therapist to come to grips with this very real possibility and may even help the patient’s family accept that death may be on the horizon for their child.

Devaluation Devaluation has been viewed as one of the primary defenses of borderline patients (Kernberg 1975) or, alternatively, as a particularly annoying form of “misbehavior” (Gunderson 1984). In reality, if properly understood, devaluation is useful for patient and clinician alike—at least in the initial stages of treatment. The value of devaluation is that the borderline patient uses it to negotiate intimacy, albeit in a very awkward manner. After all, no one bothers to put someone down unless they are emotionally important to them in some way. Borderline patients are affiliative by nature and wish to be close to others but are hindered in their quest for closeness by their fears of being unwanted or used. Devaluation is also potentially useful for the clinician in that it is a window into the patient’s own extraordinarily low self-esteem. Devaluation is an ❉

187 ❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

emotionally taxing but still useful crash course in what the borderline patient had to face on a day-to-day basis while growing up. In our experience, devaluation is learned at home while young, much like other overlearned forms of behavior. And much like other forms of overlearned behavior, the borderline patient is oblivious to the actual content of his or her behavior and its effects on others. This is so because the families of many borderline patients expressed their love and devotion in a “biting,” indirect manner, and thus the borderline patient is not fully and consistently aware that implying or stating that someone is stupid, mean, or completely unhelpful engenders hurt feelings in friends and clinicians alike. It is as though borderline patients speak a special language and clinicians must learn to translate this language rapidly for the therapeutic relationship to proceed with a certain degree of smoothness and emotional tone. Thus, a borderline patient may say: “No one loves me or listens to me. No one cares.” The therapist, who is meeting with the patient at 7 P.M. and feeling tired and hungry, can take this comment personally and angrily confront the patient with his or her lack of gratitude, or the therapist can quickly translate this verbal missive into what the patient actually meant: “I care so much for you. I wish that you needed me as much as I need you.” This process of translation allows the clinician to respond to what the patient thought he or she said with a careful clarification that protects the patient’s fragile self-esteem while providing the patient with useful new information in a manner that the patient can actually “hear.” Such a clarification might be: “Sometimes people find it hard to let people know that they care for them.” This type of comment gives the borderline patient time and room to think about his or her behavior without being criticized for it. The comment is put somewhat in the third person and directed at the middle distance. It or a variant on it will have to be repeated many times on many occasions before the borderline patient seems to “get it.” However, it is our clinical experience that even the most severely disturbed borderline patient has typically noted this comment from the beginning, thought about it often, but will not acknowledge it or its usefulness until the desperation that underlies devaluation has abated.

Manipulation Most borderline patients are highly skilled at manipulation, which we define here as trying to get what one wants from others in an indirect manner. Needs are not directly expressed and wishes are not completely owned. ❉

188 ❉

❉

The Difficult-to-Treat Patient With Borderline Personality Disorder

❉

Rather, borderline patients try to maneuver people in much the same manner as small children put their mother’s hand in the cookie jar when they are hungry for a snack. Much as children “use” their mother because they are small and relatively weak and powerless, borderline patients try to take from people that which might ordinarily be freely and openly given or at least refused in a polite and respectful manner. But clinicians often take manipulation personally and respond by trying to control or “set limits” with borderline patients. This is so because manipulation is both a felt assault on one’s generosity and a seeming indictment of one’s naïveté. Most therapists find the borderline patient’s propensity to try to manipulate them demeaning and hateful (as well as unnecessary), particularly as they have worked so hard to give up these “uncivil and childish” forms of behavior themselves. Here again, as with devaluation, borderline patients are unaware of the maladaptive aspects of their behavior and take a confrontation on their manipulativeness as an assault on their right to live. It is far better to use a clarification aimed at the middle ground between patient and clinician: “It is hard for some people to believe that they can ever get anything they want.” Plainly, most people do not steal food at a buffet dinner. Nor do most hosts get angry at a guest who mistakes a finger bowl for a glass of water. However, clinicians often get so angry and outraged at a borderline patient’s manipulative behavior that they respond with a chilly superego stance that has more in common with Cotton Mather than Sigmund Freud. Borderline people, like other people stuck in the awkward behavior of adolescence, will respond better to a psychoeducational approach than a clerical edict. After all, most of us learn more effectively while sitting in a comfortable chair than while trapped in a walk-in refrigerator.

Demandingness Borderline patients often are very insistent about getting what they want when they want it. This behavior is often taken as an affront by their overly civilized therapists who have worked hard to get what they want by asking politely and waiting patiently. In this view, demandingness is both boorish and embarrassingly obvious. Less-experienced therapists often cringe in horror at these naked demands, while quietly wondering if there is something to this brazen, if annoying, approach. More experienced therapists often intuitively realize that only the weak and powerless demand, whereas the powerful only need ask. Here too a psychoeducational approach is useful as is an admiring view ❉

189 ❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

of the patient’s “prowess.” One might comment: “Good for you. You are finally taking care of yourself,” or, alternatively, one might say: “It’s hard to believe that polite people ever get anything.” Certainly, borderline patients have been told repeatedly by their family and their previous clinicians that demandingness is off-putting. But rarely is this behavior accepted as a way station on the road to true assertiveness. Whereas their false power should not be allowed to rule the day, their efforts at being assertive should be encouraged. This is an important goal in the work with severely disturbed borderline patients as they are often timid and self-defeating while at the same time they rarely go unnoticed due to their insistent bravado.

Entitlement To most therapists, entitlement is the evil sister to demandingness. In this view, entitlement is the inner state that underlies demanding behavior. Borderline patients are seen as believing that they have a “divine” right to everything they want and that the “rules of life” do not apply to them. In many ways this perception is true. However, in our experience, it is important to remember that the more “entitled” the borderline patient, the more likely the patient is to be starving him- or herself because he or she believes that he or she does not deserve to eat, or cutting him- or herself because the patient thinks that he or she is bad or evil. In many ways, what borderline patients feel entitled to is life the way it used to be before everything went wrong, or life the way it should have been if everything had not always been wrong. Like most grieving people, borderline patients are obsessed with the past and are unable to let go. The more depleted they feel, the more they believe that only the impossible will make them whole. Thus, borderline patients who have been abused as small children want to be treated in the endlessly loving way that they believe attentive mothers treat their toddlers, or borderline patients who were neglected when young want an always available parental figure who is never tired, sad, or frustrated. Much as latency age children long to be the child of a famous person, borderline patients feel that they deserve the childhood they never had. And so they do. But reality dictates that dead people can not live again and childhood is reserved for the very young. The job of a therapist of a severely disturbed borderline patient is not to attack the patient for his or her heartfelt wish to be cared for in a loving and tender manner but to mourn with the patient the simple fact that the emotional carousel only goes around once for ❉

190 ❉

❉

The Difficult-to-Treat Patient With Borderline Personality Disorder

❉

each of us. However, it is also the job of the therapist of the severely disturbed borderline patient to point out that life offers many kinds of reparations but that these disguised adult substitutes can only be found in the real world and are not hidden in the therapist’s office in some emotional toy chest that only the lucky few get to open. Neither persistence nor insistence will bring back the childhood one never had.

Treatment Regressions Just when a therapist feels that he or she has done a particularly good piece of therapeutic work, the severely borderline patient may undergo a serious regression in outpatient treatment. Typically there are two reasons for this. The first is that the therapist has, or the patient believes that the therapist has, inadvertently “promised” the patient more than he or she can or will deliver. In this situation, the patient’s hopes for a perfect relationship have been raised only to be dashed, and the patient feels profoundly embarrassed, betrayed, and enraged. And like anyone about to lose his or her spot in first class, borderline patients hang on for all they are worth. Unfortunately for the therapist, they hang on with all the maladaptive survival skills they have at their command, and the therapeutic relationship can soon take on the appearance of a World War I battlefield strewn with the innocent being punished by the terrified. The second reason that a serious behavioral regression can occur in outpatient treatment is that the borderline patient believes that he or she is making too much progress in the real world and that his or her therapist now believes that the patient is “all better.” Given the centrality of their emotional pain to their sense of themselves (Zanarini and Frankenburg 1994), this can seem like an affront to all that the severely disturbed borderline patient holds most dear. And like a small child with a new brother or sister, lessons recently learned can be abandoned in the desperate emotional struggle that ensues. This emotional recrudescence of their abandonment fears can unleash a panic and rage that can not be controlled in an outpatient setting. This regression, of course, may lead to an inpatient stay, which may eventually lead to an even more profound behavioral regression. When cornered and made to feel powerless, the severely ill borderline patient may fight back and end up hurting him- or herself to get back a sense of control (Gunderson 1984). Or alternatively, the patient may become despairing and abandon all pretence of control and end up in seclusion and restraints. Clearly, few of us do our best learning when tied up and tied down. And for ❉

191 ❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

severely disturbed borderline patients, this type of experience may reenact childhood experiences of physical and/or sexual abuse. Much as competent therapists do not offer needy borderline patients an unlimited spending spree at an emotional toy store, a well-run inpatient unit does not put a borderline patient in restraints for countertransferentially driven reasons, such as “she said she was unsafe” or “she got really angry and I thought she might lose control.” Much of the chaos on an inpatient unit trying to work with a borderline patient arises from the inner feelings of the patient that he or she is being controlled or that he or she has lost all control and is, again, small and powerless. However, a large measure of the chaos arises because so many young people—patients and staff alike—are trapped in a small space with all of their conflicted longings to be taken care of and to take care of others casually intermingled (Burnham 1966; Main 1957). In such a situation, borderline patients are going to notice that some staff members are warm and nurturing, and others are more distant and controlling. Soon, through the magic of projective identification, these staff members are quarreling with one another over their basic worth. In the end, the severely ill borderline patient will be punished in some way for being inconsiderate enough to notice the real differences in attitudes that always exist on an inpatient unit. The patient will be seen as deliberately “splitting” staff, and nurses and mental health workers alike will have forgotten that much as it takes two to tango, it takes two (or more) to engender a regression. Put another way, it is hard to imagine a borderline patient bothering to regress on a desert island. After all, regression is both an awkward and indirect form of communication as much as an act of desperation. Whereas both patient and therapist or inpatient staff are usually implicated in some way when a treatment regression occurs, it is also possible and even probable that the patient has experienced an upsurge in Axis I psychopathology that is autonomous in nature. Much as sometimes a cigar is really just a cigar, often a depression is just a constitutionally driven vulnerability gone awry, and it is difficult to overestimate the degree to which a serious major depressive episode can exacerbate already existing borderline psychopathology.

Special Relationships As we have said in many places in this chapter, borderline patients long for a warm and loving relationship with a generous and kind-hearted mother figure. They may also long to recapture the “special” but inappropriate and/ ❉

192 ❉

❉

The Difficult-to-Treat Patient With Borderline Personality Disorder

❉

or unhelpful relationship that they once had with a parent or parental figure. Both of these factors make them vulnerable to both unwitting and predatory therapists. The unwitting therapist may believe that all the severely ill borderline patient needs is to be reparented. Often this belief goes hand-in-hand with the belief that the patient is not really borderline but rather suffering from a chronic form of PTSD and that all of the patient’s problems are because of a childhood history of abuse. However, what this evangelical therapist forgets or overlooks is that the borderline patient is far more in need of the lessons most people learn in latency and adolescence rather than the emotional swaddling clothes of early childhood. As Maltsberger and Buie (1974) have pointed out, such emotionally tender psychotherapeutic trysts are often followed by enraged reenactments. This is so because borderline patients, no matter what their wishes, are not infants and know in some deep and pervasive way that they are not nearly as appealing when balled up in a corner of a quiet room at the age of 25 as they were when they were taking a nap in the back of the family car at the age of 2. Whereas experiences of too much giving and then an angry retrenchment are common when treating borderline patients, actual friendships and love affairs between borderline patients and their clinicians can be catastrophic for all the reasons that incest is so destructive. Love affairs are by definition mutual and reciprocal. Borderline patients, although they certainly need supportive life partners, need their therapist as a guide to the rules of life, not the disappointments of love.

Dependency and Counterdependency Borderline patients are typically very dependent and prone to helping others at the same time. Their counterdependency can be seen as shame over normal wishes to be taken care of and a deeply held belief that their dependency needs will not be met. It may also be seen as more adaptive and a personality trend that may some day evolve into altruism. In this regard, Vaillant (1977) has found that creativity and generosity are the only two highly adaptive orientations that people with a very troubled childhood are likely to achieve, and thus this trend should be encouraged in borderline patients. However, they will encounter special difficulties in entering the helping professions and their potential for envy-driven regressions will need to be anticipated and addressed. Clearly, it is very difficult to help others if one is hungry for help oneself. ❉

193 ❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

The dependency of borderline patients often joins with their desperation, leading to requests or demands for more contact with their therapist than is easy or appropriate to give. Sometimes therapists succumb to these demands and end up exhausted and resentful. Other times they set very firm “limits” on their availability. Unfortunately, these limits are often set in such a way that they imply that the patient is misbehaving and needs to be controlled. Our experience is that it is more helpful to teach borderline patients about boundaries. These boundaries are areas that they draw around their own space and time. As so many borderline patients did not learn about the most appropriate interpersonal distances as children and vacillate between clinging and fleeing behaviors, this can be a useful set of lessons. It also puts the issue under the patient’s control and protects his or her dignity. In addition, it makes it clear that the attainment and maintenance of appropriate boundaries is as much for the patient as for the clinician. This is not to say that there should be no limitations on the therapy or the demands that a therapist can and should fulfill. Clearly, patients should attend all of their appointments, come on time; leave when scheduled; make some effort, no matter how awkward, to discuss what is actually going on in their lives; and pay their bills in full. Usually problems arise as much because of a therapist’s guilt and his or her countertransferential absorption of the patient’s abandonment fears as because of the unrealistic demands that a severely disturbed borderline patient makes. Put another way, therapists treating these patients need to give themselves permission to sleep through the night and to enlist the help of others when a particular patient is putting more strain on them than can reasonably be borne. However, this type of freedom requires that a therapist know that he or she may be operating on the patient’s “borrowed” fears and that neither of them will die as a result of acknowledging their separate identities. Modern technology also provides clinicians with help in dealing with the borderline patient’s wish for frequent extrasession contact. One of us (K.R.S.) has found that e-mail is a useful tool in this regard (Silk 1997). Voice mail is also useful in providing borderline patients with the contact they crave and perhaps need because of their problems with object constancy (Adler and Buie 1979). Voice mail allows them to call at all hours, listen to their therapist’s voice, and in this way restabilize themselves. It also allows them the opportunity to record, replay, and then erase angry messages that they do not want their therapist to listen to. This process, while initially secret, typically finds its way into therapy and then can be discussed in a useful way. ❉

194 ❉

❉

The Difficult-to-Treat Patient With Borderline Personality Disorder

❉

Distortions of the Truth Borderline patients often lie or at least, tell different people different versions of their truth. While some borderline patients have strong sociopathic tendencies, the majority do not. Rather, they lie for three reasons. The first reason is to bolster their shaky sense of self by coming up with a personal truth that is more bearable to them and perhaps more appealing to others. In this regard, they may be boastful and claim accomplishments that are not really theirs. The second reason is that they are particularly afraid of disappointing others and/or being punished for being less than perfect. Thus, they bend the truth, often through critical omissions, to avoid losing the support of those they care about and trust. The third reason for their tendency to make misrepresentations is that they do not know what the shared or objective truth is. This can be because of their varied and shifting sense of identity. It can also be because of the fact that clinicians have been so insistent that abuse is at the root of all of their problems, even if they disagree, that they no longer have a personal narrative in which they believe and from which they derive a sense of continuity and identity. A useful approach when dealing with this tendency is for the therapist to consistently point out the inconsistencies in what the patient is telling him or her. This will put the patient on notice that others pay attention to what the patient is saying and hold him or her accountable for at least trying to make sense in a shared manner. The second useful approach is not crowding the patient with an externally imposed version of the truth of his or her life but instead listening carefully as the patient develops or, more probably, simply relates a personal narrative he or she has been quietly, even secretly, working on for years.

Sadomasochistic Tendencies Borderline patients often alternate between behaving like a victim and victimizing others. They are both submissive and cruel by nature and training. Therapists are quick to notice how cruelly they can behave and therapists rightly resent trying to be controlled or treated in a cruel manner. Here too therapists are likely to react by reinforcing or imposing overly strict “limits.” When using this approach, therapists often end up with a more compliant patient or no patient at all as borderline patients often prematurely terminate treatment when their feelings are hurt too deeply. In our experience, therapists are not as good at noticing how masochistic their severely disturbed ❉

195 ❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

borderline patients are. When angry and frustrated enough, therapists can overlook the self-destructive aspects of many behaviors, such as bingeing and purging, having unprotected sex, and routinely driving at 90 MPH. Therapists can also come to view patients’ self-mutilation and even very serious suicide attempts as forms of misbehavior meant only or mainly to ruin their evening, weekend, or vacation. Clearly, in this situation, the therapist is out of touch with his or her own cruelty and has come to view the patient as all bad. Such simplistic thinking is dangerous when treating severely ill borderline patients as it reinforces their sense of their own inner badness. A more useful approach is to refuse to struggle with them or to try to avoid attempts to control them. Rather, an educational approach that points out how it is in their interest to take care of themselves and not alienate others by treating them in the unfortunate way that they remember being treated. Often appeals to self-interest and the hopeful power of the golden rule are effective tools when working with a severely disturbed borderline patient.

Common Treatment Modalities Psychotherapy Most severely disturbed borderline patients have a long history of psychotherapy. Oftentimes during the course of this psychotherapy or therapies, the patient has gone from being a person in deep pain but still functioning in the world of relationships and work to being a chronic patient with little structure to his or her life and few social contacts outside of treatment. Other times, the severely disturbed borderline patient is still functioning in the real world but is contemplating giving up because the struggle is so difficult. The first question that usually arises is how often the patient should be seen. The answer needs to be individualized to the patient, but a rule of thumb would be often enough to stabilize the patient’s functioning but not so often that regression is an ongoing threat. This type of patient tends to believe that the more treatment the better. Often clinicians will agree with this assessment. This can be because the patient is articulate and appealing; because the therapist holds the discredited view that to get better a patient must first get worse; or because the therapist has some opening in his or her schedule, and a loyal borderline patient who attends all of his or her appointments will provide a financial incentive to overly intense psychotherapy. The next question that usually arises is what the focus of the treatment should be. This question is relatively easy to answer. The focus on the treat❉

196 ❉

❉

The Difficult-to-Treat Patient With Borderline Personality Disorder

❉

ment should be to help the patient to function better in the real world. This means supporting the functioning of the patient who is considering giving up and going on disability. And for the patient who has given his or her life over to “patienthood,” it means encouraging the patient to recapture as much of his or her previous life as possible. It is critical that one take a careful history of psychosocial functioning over the course of the patient’s life, and it is important to know how competent the patient was as a child and an adolescent. Clearly, the patient who had no friends, never participated in extracurricular activities, and did poorly in school may have a worse prognosis than a patient who had several friends who really mattered to him or her, enjoyed playing sports after school, and did well academically. In the same vein, a borderline patient who had a good school or work record as a young adult, who dated regularly or married, and who had children may have a better prognosis than a borderline patient who has never worked or had an intimate relationship. Unfortunately, many severely ill borderline patients are so dilapidated after years of treatment that clinicians actually “forget” that the patient had a life before becoming a chronic patient. In this situation, the clinician is oblivious to all that the patient has given up—either as a result of the extreme severity of his or her illness or the collaboration with prior clinicians who mistakenly thought that they were being helpful. If the patient still has a life outside of being a patient, the focus should be on bringing all the resources to bear so that he or she can continue to function in the real world. The patient may need to take time off from work because of a comorbid condition, most typically major depression. However, the leave should be circumscribed and the expectation clearly laid down that the patient will be going back to work or school after a period of weeks to months. Putting a patient still able to function on federal disability is one of the most unkind things that a therapist can do. It is our experience that once on such disability payments, most patients loose all motivation to return to a productive life. Few borderline patients who once had a life will thank their therapist for making them a chronic patient. However, if the patient has had severe problems from childhood or early adolescence, disability may be the only stable source of income that these patients can obtain. Therapists should still expect that such a patient will work in a sheltered environment, do volunteer work, or attend school on a part-time basis. The third question that typically arises in the outpatient treatment of the severely ill borderline patient is the role of exploring sexual abuse issues. Our experience suggests that those borderline patients who have such a history (about ❉

197 ❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

50%–60% [Zanarini 1997]) will describe their experiences in some detail in the early sessions of a therapy. The clinician should avoid any efforts to go over all the events and reactions in great detail. This only leads, in most cases, to an increasingly overwhelmed and dysfunctional patient. If the patient could have incorporated such a set of experiences, he or she would have already. This does not mean that the affective and behavioral sequelae of such experiences should not be discussed. In fact, fuller appreciation of these sequelae are key to a patient’s feeling understood and emotionally held. They are also key to a patient’s recovery from maladaptive forms of impulsivity and interpersonal survival strategies that have long since outlived their usefulness. Clearly, no one can stop selfdefeating forms of behavior unless they realize the self-destructiveness of their behavior patterns and the reasons why they so energetically cling to these outmoded survival strategies. The approach outlined above might be called supportive, eclectic, psychoeducational, or practical. It is informed both by a psychodynamic understanding of the dynamics of BPD and an appreciation of the success of cognitive-behavioral therapy in effecting changes in symptom expression. Today, within the confines of psychodynamically oriented therapy, controversy still exists about the role of early interpretation and the management of transference. Masterson (1976) and Gunderson (1984) emphasize the need to identify, confront, and interpret the aggressive motives that exist in the BPD patient, whereas Gabbard and colleagues (1994) have described transference interpretation as a “high-risk, high-gain” intervention with BPD patients (e.g., might lead to a serious upsurge in symptomatology or a serious behavioral regression). In terms of cognitive-behavioral therapy, Linehan’s (1993) dialectical behavior therapy has become enormously popular in a short period of time. We believe that this is so for two reasons. First, Linehan provides clinicians with clear-cut guidelines for how to treat borderline patients. Second, Linehan is actually a gifted, intuitive psychodynamic thinker who truly understands borderline patients and their conflicts. Her manualized treatment approach is also the only psychotherapeutic strategy for borderline patients to have empirical validation, having been shown in randomly controlled trials to be more effective than “treatment as usual” in reducing parasuicidal and drug-abusing behaviors (Linehan et al. 1991, 1999).

Psychotropic Medication Most severely ill borderline patients have been tried on a wide variety of medications. In our experience, they are often on three or more standing ❉

198 ❉

❉

The Difficult-to-Treat Patient With Borderline Personality Disorder

❉

medications at any one time. Not only is this expensive, but it also is not clear how effective this approach is. In addition, there are serious side effects to most psychiatric medications, particularly weight gain. This is a serious problem for many severely ill borderline patients as their obesity or near obesity may interfere with their social life and may, over the long run, be deleterious to their physical health as well. Another consideration is that having ready access to a number of different medications can result in more serious physical damage should the decision be made, either in a planned or impulsive manner, to ingest a large and indiscriminant quantity of those medications. A good rule of thumb is the rule of two—two concurrent medications are probably the maximum in most instances and may be one too many in many others. Often the number of medications added slowly over the years comes as a surprise to the treating physician. It is not uncommon for severely ill borderline patients to be on anywhere from three to eight medications at any one time. Clearly, it is impossible to tell which of these medications, if any, is effective. The patient, although devoted to his or her psychiatrist and fearful of losing his or her care, can be so sedated and confused as a result of this drug medley that he or she can not function in the real world. Our experience suggests that medication trials are useful for the treatment of borderline patients with clear-cut comorbid conditions. Whereas research suggests that serious Axis II pathology lessens the effectiveness of these trials (Reich and Vasile 1993), any borderline patient with a major depression, bipolar disorder, panic disorder, obsessive-compulsive disorder, social phobia, bulimia, or binge-eating disorder should be tried on appropriate medications that have proven effectiveness for these conditions. The use of medications targeted at particular symptoms is more controversial, and yet all of the numerous medication trials that have been conducted have been directed at symptoms of BPD rather than at comorbid conditions. These studies will be briefly reviewed by class of medication: antipsychotics, antidepressants, anticonvulsants or mood stabilizers, anxiolytics, and, lastly, naltrexone. Early work by Serban and Siegel (1984), Goldberg and colleagues (1986), Cowdry and Gardner (1988), and Soloff and colleagues (1989) found that BPD patients show a “broad spectrum” response to standard antipsychotic medication (i.e., both affective and cognitive symptoms were affected). However, a long-term (16-week) placebo-controlled study found little lasting benefit from haloperidol in doses up to 6 mg/day, except on measures of irritability (Cornelius et al. 1993). ❉

199 ❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

Recently, there have been reports of positive results with the use of atypical antipsychotics. Frankenburg and Zanarini (1993) studied the effects of clozapine in 15 borderline patients with atypical psychotic symptoms. They found a decrease in overall symptomatology and an increase in psychosocial functioning. These findings have been replicated by Benedetti and colleagues (1998) and Chengappa and colleagues (1999). Schulz and colleagues (1999) have found that olanzapine is both safe and efficacious in an open-label study of borderline patients with comorbid dysthymia. Schulz and colleagues (1999) have found that olanzapine is both safe and efficacious in an open-label study of borderline patients with comorbid dysthymia, whereas Zanarini and Frankenburg (in press), who conducted a double-blind, placebo-controlled trial, have found that olanzapine is more effective than placebo in reducing symptoms in all four core areas of borderline psychopathology. Both tricyclic antidepressants and monoamine oxidase inhibitors were studied in the 1980s and 1990s. Soloff and colleagues (1986) found that amitriptyline was surprisingly unhelpful and in some cases actually worsened suicidality. Cowdry and Gardner (1988) found that tranylcypromine was highly rated by patients. However, Soloff and colleagues (1993) studied another monoamine oxidase inhibitor, phenelzine, and found that this agent was only moderately useful. Furthermore, these clinically modest effects wore off over time (Cornelius et al. 1993). Recently, interest has grown in the use of serotonergic agents in BPD, partly because there is some evidence of abnormal serotonergic function in BPD and related impulsive/aggressive conditions (Coccaro et al. 1989; Hollander et al. 1994). Norden (1989) studied 12 patients with BPD but without concurrent major depression in an open-label trial of fluoxetine and rated 75% as much or very much improved. Salzman and colleagues (1995) carried out a 13-week, double-blind study of subjects with mild to moderately severe BPD. A clinically and statistically significant decrease in anger was seen among the fluoxetine subjects. However, the number of subjects in this study was small, the placebo responsiveness was high, and the clinical features were in the mild to moderate range of severity. Gardner and Cowdry (1986) also studied the efficacy of carbamazepine and found that it markedly reduced severity of behavioral dyscontrol in their sample of severely disturbed borderline patients. Similarly, Stein et al. (1995) found that valproate led to overall improvement in about 50% of a small sample of BPD patients (N=11) in an open-label, 8-week trial. Valproate was modestly helpful for mood, irritability, anxiety, anger, rejection ❉

200 ❉

❉

The Difficult-to-Treat Patient With Borderline Personality Disorder

❉

sensitivity, and impulsivity; however, the results were not robust. More recently, Pinto and Akiskal (1998) found that lamotrigine was an effective treatment in three of eight borderline patients participating in an open-label study. Gardner and Cowdry (1985) also studied the effects of alprazolam and found that it was associated with a marked disinhibitory response. They found that behavioral dyscontrol actually worsened, a finding occasionally reported with benzodiazepines in other patient populations. There are also a growing number of open-label or clinical trial studies that suggest a possible role for naltrexone to diminish self-cutting behavior (Pies and Popli 1995; Roth et al. 1996; Sonne et al. 1996). The purported reason for the success of naltrexone may have to do with blocking the pleasure or relief response that some self-mutilating patients claim is the reason for their self-mutilation (Leibenluft et al. 1987). However, it is important to note that, at least in our experience, naltrexone is not widely used by clinicians treating self-harming borderline patients.

Psychiatric Hospitalizations Sometimes borderline patients are so depressed, out of control, and/or psychotic in their thinking that they need to be hospitalized. Very few clinicians still believe that a hospitalization should be prolonged or that it constitutes a “growth” experience. Rather, because of economic pressures, clinicians have come to see hospitalizations as brief experiences whose main purpose is to stabilize the functioning of a borderline patient. When borderline patients say they feel much better the day after being admitted for cutting themselves, they are likely to be believed. In most cases, this works out well, but in some cases, such a rapid move toward discharge can have tragic consequences. Brief hospitalizations promise less and typically lead to less regression (Silk et al. 1994; Springer et al. 1996). However, there are times when a borderline patient may need to stay in the hospital for weeks or even months at a time. This is so when they are deeply and perhaps psychotically depressed. It may also be so when they have managed to alienate everyone in their life to the extent that they have no meaningful relationships and suicide is a real option for them. Regardless of the length of the hospitalization, borderline patients often resist being discharged. Sometimes they are able to discuss this with staff, and sometimes they act out to stay in. Often these end-stage regressions can ❉

201 ❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

be avoided by anticipating them. If their fear and anger at being discharged is predicted and discussed throughout their inpatient stay, patients may be able to have a cognitive map of what is going on and state their objections in a straightforward manner or simply go ahead with the discharge without resorting to regressive delaying tactics.

Long-Term Course and Outcome To date, the results of 17 small-scale, prospective studies of the short-term course and outcome of BPD in adult patients have been published. (See Zanarini et al. 1991 for a review of both short and long-term studies of the course of BPD.) In general, these studies have found that most borderline patients continue to have substantial difficulty functioning both socially and vocationally 2–7 years after their initial evaluation. In addition, they tend to remain highly symptomatic and to need continuing psychiatric care. There have also been four large-scale, follow-back studies of the longterm course and outcome of BPD in adult patients. The results of these studies, which were conducted by McGlashan (1986), Paris and colleagues (1987), Plakun (1985), and Stone (1990), have usually been interpreted to mean that the average borderline patient is functioning reasonably well a mean of 14–16 years after his or her index admission. However, a closer look at the results of these studies reveals that only 53%–72% of these patients were functioning in the good to recovered range. In addition, the average borderline patient in these studies socialized only about once a month and had a relatively unstable work record. These patients also continued to use high levels of mental health services. In addition, two of the three studies that have assessed mortality rates have found that 9%–10% of their traced borderlines eventually committed suicide, usually within the first 5 years after their index hospitalization. Thus, BPD seems to be a disorder with a wide range of outcomes: some patients leading productive lives and being basically asymptomatic, many continuing to have difficulty functioning in a number of areas of their lives, and a substantial minority committing suicide. Whereas it might be easy to assume that the severely disturbed borderline patient will have a poor long-term course, it is impossible to predict this at this time. This is so because the follow-back studies described above only studied patients at one point in time, and thus it is impossible to know if those patients who were judged to be doing well had always done well or if this was a more recent development. Our clinical experience suggests that some severely disturbed borderline patients will function better over time and be less symptomatic as well. ❉

202 ❉

❉

The Difficult-to-Treat Patient With Borderline Personality Disorder Clinical Vignette Ms. D is a highly intelligent, 28-year-old, single white woman. She comes from a well-to-do intact family. After a childhood and adolescence marked by achievement at school and struggles with her mother at home, she graduated from college and got a good job. She entered outpatient treatment at the age of 23 because she felt depressed and because she recognized that her loneliness had led her to be less than discreet in choosing her sexual partners. Soon after entering treatment with a male therapist, she began to feel suicidal and started to make desperate suicidal gestures that involved calling her therapist in the middle of the night and expecting him to find her and save her. This process went on for several months, and the patient was finally hospitalized. Once hospitalized, she recovered memories of childhood sexual abuse by her older brother. She felt that this explained many of her troubled feelings and self-destructive patterns. She wanted to “work through her issues” and convinced her second therapist, a woman with a strong interest in PTSD, to start to see her three times a week on an outpatient basis. She also started to take medications for the first time and was placed on an antidepressant, an anxiolytic, and a mood stabilizer. Within a few weeks of her discharge from the hospital, she felt that she could no longer handle her job and asked her clinicians to sign forms for her to go on disability. They did so and she left her job and started day treatment two days a week. Soon she began to make suicidal gestures again. Her time at day treatment was increased to 6 hours per day, Monday through Friday. She continued to make suicidal gestures of increasing severity and was repeatedly hospitalized to try to contain what she described as her feelings of dysphoria and desperation. Her medications were changed often and with little symptom relief. After her sixth hospitalization, Ms. D agreed to live in a residential program. She stayed at this program for 2 years. During this time, she increased the frequency of her therapy appointments to five times per week. She was also prescribed an antipsychotic medication, which she took regularly and without objection. After her discharge from this program, Ms. D again decompensated. A consultation was arranged, and it was recommended that she be sent to a different mental health system closer to her family home. Ms. D was extremely upset that her care was being transferred to another mental health system and made a very serious suicide attempt. She was admitted to the general hospital psychiatric unit that served this system and while there met her new therapist, an older female social worker who agreed to see her once a week. She also met her new psychopharmacologist who agreed to meet with her once a month. Ms. D moved home with her parents upon her discharge and spent much of her time watching television. Her new therapist focused on her daily functioning and refused to review her abuse history in detail until she

❉

203 ❉

❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

was much more stable. Her psychopharmacologist carefully reviewed her medications and asked Ms. D to let him know which ones she wanted to continue to take. After much thought, she decided that the antipsychotic medication and the mood stabilizer were of little help and stopped taking both. At her therapist’s suggestion, she bought a dog and started to go for long walks with her pet on a daily basis. She also enrolled in an adult education course in creative writing at a nearby community college and received praise for her diary entries that she had been writing throughout her psychiatric career. After 11 months of this routine, Ms. D got a part-time job in a local bookstore and felt better that she had more structure in her life. She still is troubled on a daily basis with self-destructive thoughts but has not acted on them. Ms. D plans to continue to work in the bookstore on a part-time basis for another 6 months and then is planning to work full-time. She has few friends and does not date but feels she is on the road to recovery.

Conclusions Severely ill borderline patients are among the most challenging in the field of mental health. Often they have suffered deeply as children and developed a series of symptoms and behaviors that helped them survive while young but that limit them as adults. Clinicians wishing to work with these patients need to have an optimistic nature, a good sense of humor, and an iron constitution. Common sense is probably a more important tool for their clinicians to possess than a wish to conduct a sophisticated psychodynamic psychotherapy. Whereas many clinicians view such patients with anxious dread, we suggest that one can admire the integrity with which they have dealt with their pain. After all, not many people remain so loyal to and so respectful of such disheartening experiences.

References Adler G, Buie D: Aloneness and borderline psychopathology: the possible relevance of child developmental issues. Int J Psychoanal 60:83–96, 1979 Benedetti F, Sforzini L, Colombo C, et al: Low-dose clozapine in acute and continuation treatment of severe borderline personality disorder. J Clin Psychiatry 59:103– 107, 1998 Brodsky BS, Cloitre M, Dulit RA: Relationship of dissociation to self-mutilation and childhood abuse in borderline personality disorder. Am J Psychiatry 152:1788– 1792, 1995

❉

204 ❉

❉

The Difficult-to-Treat Patient With Borderline Personality Disorder

❉

Burnham D: The special problem patient: victim or agent of splitting? Psychiatry 29:105–122, 1966 Chengappa KN, Ebeling T, Kang JS, et al: Clozapine reduces severe self-mutilation in psychotic patients with borderline personality disorder. J Clin Psychiatry 60:477– 484, 1999 Coccaro EF, Siever LJ, Klar HM, et al: Serotonergic studies in affective and personality disorders: correlates with suicidal and impulsive aggressive behavior. Arch Gen Psychiatry 46:587–599, 1989 Cornelius JR, Soloff PH, Perel JM, et al: Continuation pharmacotherapy of borderline personality disorder with haloperidol and phenelzine. Am J Psychiatry 150:1843– 1848, 1993 Cowdry RW, Gardner DL: Pharmacotherapy of borderline personality disorder; alprazolam, carbamazepine, trifluoperazine, and tranylcypromine. Arch Gen Psychiatry 45:111–119, 1988 Frankenburg FR, Zanarini MC: Clozapine treatment of borderline patients: a preliminary study. Compr Psychiatry 34:402–405, 1993 Gabbard GO, Horwitz L, Allen JG, et al: Transference interpretation in the psychotherapy of borderline patients: a high-risk, high-gain phenomenon. Harv Rev Psychiatry 2:59–69, 1994 Gardner DL, Cowdry RW: Alprazolam-induced dyscontrol in borderline personality disorder. Am J Psychiatry 142:98–100, 1985 Gardner DL, Cowdry RW: Positive effects of carbamazepine on behavioral dyscontrol in borderline personality disorder. Am J Psychiatry 143:519–522, 1986 Goldberg SC, Schulz SC, Schulz PM, et al: Borderline and schizotypal personality disorders treated with low-dose thiothixene vs. placebo. Arch Gen Psychiatry 43:680–686, 1986 Gunderson JG: Borderline Personality Disorder. Washington, DC, American Psychiatric Press, 1984 Gunderson JG, Kolb JE: Discriminating features of borderline patients. Am J Psychiatry 135:792–796, 1978 Herman JL, Perry JC, van der Kolk BA: Childhood trauma in borderline personality disorder. Am J Psychiatry 146:490–495, 1989 Hollander E, Stein DJ, Decaria CM, et al: Serotonergic sensitivity in borderline personality disorder: preliminary findings. Am J Psychiatry 151:277–280, 1994 Kernberg O: Borderline Conditions and Pathological Narcissism. New York, Jason Aronson, 1975 Leibenluft E, Gardner DL, Cowdry RW: The inner experience of the borderline selfmutilator. J Personality Disord 1:317–324, 1987 Linehan MM: Cognitive-Behavioral Treatment of Borderline Personality Disorder. New York, Guilford, 1993 Linehan MM, Armstrong HE, Suarez A, et al: Cognitive-behavorial treatment of chronically parasuicidal borderline patients. Arch Gen Psychiatry 48:1060–1064, 1991

❉

205 ❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

Linehan MM, Schmidt H, Dimeff LA, et al: Dialectical behavioral therapy for patients with borderline personality disorder and drug-dependence. Am J Addict 8:279– 292, 1999 Links PS, Steiner M, Offord DR, et al: Characteristics of borderline personality disorder: a Canadian study. Can J Psychiatry 33:336–340, 1988 Main T: The ailment. Br J Med Psychol 30:129–145, 1957 Maltsberger J, Buie D: Countertransference hatred in the treatment of suicidal patients. Arch Gen Psychiatry 30:625–633, 1974 Masterson J: Treatment of the Borderline Adolescent: A Developmental Approach. New York, Wiley, 1972 Masterson J: Psychotherapy of the Borderline Adult. New York, Brunner/Mazel, 1976 McGlashan TH: The Chestnut Lodge follow-up study, III: long-term outcome of borderline personalities. Arch Gen Psychiatry 43:20–30, 1986 Norden MJ: Fluoxetine in borderline personality disorder. Biol Psychiatry 13:885– 893, 1989 Ogata SN, Silk KR, Goodrich S, et al: Childhood sexual and physical abuse in adult patients with borderline personality disorder. Am J Psychiatry 147:1008–1013, 1990 Paris J, Brown R, Nowlis D: Long-term follow-up of borderline patients in a general hospital. Compr Psychiatry 28:530–536, 1987 Paris J, Zweig-Frank H, Guzder J: Psychological risk factors for borderline personality disorder in female patients. Compr Psychiatry 35:301–305, 1994a Paris J, Zweig-Frank H, Guzder J: Risk factors for borderline personality disorder in male outpatients. J Nerv Ment Dis 182:375–380, 1994b Pies R, Popli A: Self-injurious behavior: pathophysiology and implications for treatment. J Clin Psychiatry 56:1–9, 1995 Pinto OC, Akiskal HS: Lamotrigine as a promising approach to borderline personality: an open case series without concurrent DSM-IV major mood disorder. J Affect Disord 51:333–343, 1998 Plakun EM, Burkhardt PE, Muller JP: 14 year follow-up of borderline and schizotypal personality disorders. Comp Psychiatry 26:448–455, 1985 Pope HG, Jonas JM, Hudson JI, et al: An empirical study of psychosis in borderline personality disorder. Am J Psychiatry 142:1285–1290, 1985 Reich JH. Vasile RG: Effect of personality disorders on the treatment outcome of axis I conditions: an update. J Nerv Ment Dis 181:475–484, 1993 Roth AS, Ostroff R B, Hoffman R E: Naltrexone as a treatment for repetitive selfinjurious behavior: an open-label trial. J Clin Psychiatry 57:233–237,1996 Salzman JP, Salzman C, Wolfson AN, et al: Association between borderline personality structure and history of childhood abuse in adult volunteers. Compr Psychiatry 34:254–257, 1993

❉

206 ❉

❉

The Difficult-to-Treat Patient With Borderline Personality Disorder

❉

Salzman C, Wolfson A, Schatzberg A, et al: Effect of fluoxetine on anger in symptomatic volunteers with borderline personality disorder. J Clin Psychopharmacol 15:23– 29, 1995 Schulz SC, Camlin KL, Berry SA, et al: Olanzapine safety and efficacy in patients with borderline personality disorder and comorbid dysthymia. Biol Psychiatry 46:1429–1435, 1999 Serban G, Siegel S: Response of borderline and schizotypal patients to small doses of thiothixene and haloperidol. Am J Psychiatry 141:1455–1458, 1984 Shearer SL: Dissociative phenomena in women with borderline personality disorder. Am J Psychiatry 151:1324–1328, 1994 Shearer SL, Peters CP, Quaytman MS, et al: Frequency and correlates of childhood sexual and physical abuse histories in adult female borderline inpatients. Am J Psychiatry 147:214–216, 1990 Silk KR: E-mail, transitional relatedness, and borderline patients. Article presented at the 5th International Congress on the Disorders of the Personality, Vancouver, British Columbia, Canada, June 25–27, 1997 Silk KR, Eisner W, Allport C, et al: Focused time-limited inpatient treatment of borderline personality disorder. J Personal Disord 8:268–278, 1994 Soloff PH, George A, Nathan R, et al: Paradoxical effects of amitriptyline in borderline patients. Am J Psychiatry 143:1603–1605, 1986 Soloff PH, George A, Nathan RS, et al: Amitriptyline versus haloperidol in borderlines: final outcomes and predictors of response. J Clin Psychopharmacol 9:238–246, 1989 Soloff PH, Cornelius J, George A, et al: Efficacy of phenelzine and haloperidol in borderline personality disorder. Arch Gen Psychiatry 50:377–385, 1993 Sonne S, Rubey R, Brady K, et al: Naltrexone treatment of self-injurious thoughts and behaviors. J Nerv Ment Dis 184:192–195, 1996 Springer T, Lohr NE, Buchtel HA, et al: A preliminary report of short-term cognitivebehavioral group therapy for inpatients with personality disorders. J Psychotherapy Pract Res 5:57–71, 1996 Stein DJ, Simeon D, Frenkel M: An open trial of valproate in borderline personality disorder. J Clin Psychiatry 56:506–510, 1995 Stone MH: The Fate of Borderline Patients. New York, Guilford, 1990 Swartz M, Blazer D, George L, et al: Estimating the prevalence of borderline personality disorder in the community. J Personal Disord 4:257–272, 1990 Vaillant GE: Adaptation to Life. Boston, MA, Little, Brown, 1977 Westen D, Ludolph P, Misle B, et al: Physical and sexual abuse in adolescent girls with borderline personality disorder. Am J Orthopsychiatry 60:55–66, 1990 Widiger TA, Weissman MM: Epidemiology of borderline personality disorder. Hospital and Community Psychiatry 42:1015–1021, 1991

❉

207 ❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

Zanarini MC: Evolving perspectives on the etiology of BPD, in The Role of Sexual Abuse in the Etiology of Borderline Personality Disorder. Edited by Zanarini MC. Washington, DC, American Psychiatric Press, 1997, pp 1–14 Zanarini MC, Frankenburg FR: Emotional hypochondriasis, hyperbole, and the borderline patient. J Psychother Pract Res 3:25–36, 1994 Zanarini MC, Frankenburg FR: Pathways to the development of borderline personality disorder. J Personal Disord 11:93–104, 1997 Zanarini MC, Frankenburg FR: Olanzapine in the teatment of female borderline patients: a double-blind, placebo-controlled pilot study. J Clin Psychiatry (in press) Zanarini MC, Gunderson JG, Marino MF, et al: Childhood experiences of borderline patients. Compr Psychiatry 30:18–25, 1989 Zanarini MC, Gunderson JG, Frankenburg FR: Cognitive features of borderline personality disorder. Am J Psychiatry 147:57–63, 1990a Zanarini MC, Gunderson JG, Frankenburg FR, et al: Discriminating borderline personality disorder from other Axis II disorders. Am J Psychiatry 147:161–167, 1990b Zanarini MC, Chauncey DL, Grady TA, et al: Outcome studies of borderline personality disorder, in Psychiatric Treatment: Advances in Outcome Research. Edited by Mirin SM, Gossett JT, Grob MC. Washington, DC, American Psychiatric Press, 1991, pp 181–194 Zanarini MC, Williams AA, Lewis RE, et al: Reported pathological childhood experiences associated with the development of borderline personality disorder. Am J Psychiatry 154:1101–1106, 1997 Zanarini MC, Frankenburg FR, DeLuca CJ, et al: The pain of being borderline: dysphoric states specific to borderline personality disorder. Harv Rev Psychiatry 6:201–207, 1998a Zanarini MC, Frankenburg FR, Dubo ED, et al: Axis I comorbidity of borderline personality disorder. Am J Psychiatry 155:1733–1739, 1998b Zanarini MC, Ruser TF, Frankenburg FR, et al: Risk factors associated with the dissociative experiences of borderline patients. J Nerv Ment Dis 188:26–30, 2000 Zweig-Frank H, Paris J, Guzder J: Dissociation in female patients with borderline and nonborderline personality disorders. J Personal Disord 8:203–209, 1994a Zweig-Frank H, Paris J, Guzder J: Dissociation in male patients with borderline and nonborderline personality disorders. J Personal Disord 8:210–218, 1994b

❉

208 ❉

8 The Difficult-to-Treat Patient With Dissociative Disorder Richard P. Kluft, M.D.

T

he DSM-IV (American Psychiatric Association 1994) dissociative disorders include dissociative amnesia, dissociative fugue, dissociative identity disorder, depersonalization disorder, dissociative disorder not otherwise specified, and the proposed classification of dissociative trance disorder. Patients with dissociative disorders suffer transient or chronic “disruption in the usually integrative functions of consciousness, memory, identity, or perception of the environment” (p. 477). Relatively little research on the dissociative disorders has been funded over the last two decades. Therefore, recommendations for their treatment have been drawn from clinical experience or from theoretical arguments or generalizations from other fields of study. For example, although there are several studies (Coons 1986; Ellason and Ross 1997; Kluft 1984b, 1986, 1993b; Ross 1989, 1997) on the treatment of dissociative identity disorder (DID), there are no controlled clinical trials in the treatment of this or any other dissociative condition. ❉

209 ❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

Incidence, Prevalence, and Course of Dissociative Disorders Epidemiological studies have been done in general populations and in clinical populations. Ross studied a sample of 1,055 respondents in Winnipeg, Canada, screening them with the Dissociative Experiences Scale (DES; Bernstein and Putnam 1986) and then with the Dissociative Disorders Interview Schedule (DDIS; Ross 1989). A reevaluation of the DES data (Waller cited in Ross 1997) demonstrated that 3.3% had pathological dissociative experiences consistent with a serious dissociative disorder. The DDIS data found that 12.2% had a dissociative disorder of some kind. More specifically, dissociative amnesia was found in 6%, dissociative identity disorder in 3%, depersonalization disorder in 2.8%, and dissociative disorder not otherwise specified in 0.2%; dissociative fugue was not found. Studies on general inpatient groups in the United States (Saxe et al. 1993), Canada (Horen et al. 1995; Ross 1991), Norway (Knudsen et al. 1995), The Netherlands (reported in Boon and Draijer 1993), and Turkey (Tutkun et al. 1998) have found prevalences of 8.2%–20.7% for dissociative disorders in general and 4%–6% for DID in particular. These and other studies in several settings and nations (summarized in Kluft 1999b) demonstrate that dissociative disorders are more common than had been appreciated. The courses of the various dissociative disorders vary widely. Dissociative amnesia involves the “inability to recall important personal information, usually of a traumatic or stressful variety, that is too extensive to be explained by normal forgetfulness” (American Psychiatric Association 1994, p. 478). Usually minutes to several hours are lost, although much longer losses are known. One or more episodes may occur. Spontaneous recovery of the missing material may occur, from shortly after the amnestic episode to many years later. Therapeutic resolution of the amnesia is often possible with support, psychotherapy, and often the use of hypnosis or drug-facilitated interviews. Dissociative fugue occurs most commonly as a single episode in relation to overwhelming events or profound intrapsychic conflict. It usually lasts from hours to months, but longer episodes are encountered. Spontaneous recovery of one’s baseline identity is commonplace, but amnesia for time spent in an altered identity often persists and resists efforts to access and address it. Dissociative identity disorder runs a chronic and recurrent polysymptomatic pleiomorphic course; its diverse manifestations often obscure its core ❉

210 ❉

❉ The Difficult-to-Treat Patient With Dissociative Disorder ❉ criteria. An average of 6–7 years go by between a DID patient entering the mental health system and the patient receiving the DID diagnosis (Putnam et al. 1986). Its onset is in childhood, but it is usually diagnosed in the fourth decade of life. DID usually remains covert with regard to the full manifestations of the alters, although some patients remain overt in their manifestations throughout. I (Kluft 1985) studied 210 DID patients and discovered that although the disorder waxes and wanes, total spontaneous remission does not occur. Treatment is often successful in eliminating or palliating DID, but there appear to be several subgroups of DID that vary widely in their therapeutic response (Horevitz and Loewenstein 1994; Kluft 1984b, 1994d). Some DID patients are not treatment responsive. Depersonalization disorder usually has its onset in adolescence or adult life and may not be recognized because of other comorbid psychopathologies involving anxiety, panic, or depression. When this disorder develops as a response to trauma it usually occurs rather suddenly after the exposure. The symptoms may be episodic, recurrent, or chronic. In some cases the duration of the symptoms is quite brief; in some they return, often in the context of some stressor; and in some patients the manifestations may persist for years. The response to treatment is quite varied, with some patients responding rapidly, some requiring work on their vulnerability to particular stressors, and some failing to respond. Dissociative disorder not otherwise specified (NOS) encompasses a wide variety of conditions, many of which have only been outlined recently (Coons 1992; Ross et al. 1992). Their course and treatment responsiveness parallel those of the conditions they most closely resemble. Dissociative trance disorder encompasses two groups of conditions. The first, trance, involves trance phenomena in the sense of temporary marked alterations in the state of consciousness or loss of the customary sense of personal identity without replacement by an alternate identity. The second, possession trance, involves either one or more episodes in which the state of consciousness is altered by the replacement of the customary sense of identity with a new identity, which is attributed to the influence of a supernatural entity or another person. There is insufficient information to allow the characterization of the course of either form of this condition. Contemporary summaries of treatments of the dissociative disorders are available (Kluft 2001; Loewenstein 2001; Steinberg 2001). Many of the dissociative disorders are acute, and their episodes are self-limited. The majority of dissociative disorder patients who are understood as such, and are difficult to treat, emerge from those who suffer DID or those forms of disso❉

211 ❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

ciative disorder NOS that most resemble DID. They are the focus of this chapter.

Characteristics of Conventional DID Treatment The treatment of DID follows the paradigm of the treatment of trauma outlined by Herman (1992). A phase of safety is followed by a phase of remembrance and mourning and next by a phase of reconnection. That is, after the patient is helped to feel safe and supported, and is strengthened, the patient is helped to face his or her traumatic experiences and to grieve their impact. Finally, the patient is helped to integrate his or her identity and understanding of the narrative of his or her life and to reconnect with others and with his or her social roles and relationships. In work with patients with DID, several efforts have been made to characterize the stages of therapy (e.g., Braun 1986; International Society for the Study of Dissociation 1997; Kluft 1991; Putnam 1989). Kluft’s stages are 1) establishing the psychotherapy (International Society for the study of Dissociation 1997), 2) preliminary interventions, 3) history gathering and mapping, 4) metabolism of the trauma, 5) moving toward integration/ resolution, 6) integration/resolution, 7) learning new coping skills, 8) solidification of gains/working through, and 9) follow-up. (For a more detailed description of these phases and the typical conduct of treatment see Kluft 1991, 1999.) In essence, Herman’s safety corresponds to Kluft’s stages 1–3. Herman’s remembrance and mourning equate with Kluft’s stage 4, metabolism of the trauma. Herman’s reconnection becomes Kluft’s stages 5–9. A wide variety of techniques has been used in the treatment of DID. A survey by Putnam and Loewenstein (1993) found that the most commonly applied modalities were psychodynamic psychotherapy, often facilitated with hypnosis. A wide variety of hypnotic techniques useful for work with DID has been described (e.g., Kluft 1994a), most of which are designed to help the patient cope with and contain dysphoria and potentially disruptive material. Most commonly, the therapy will involve an overall orientation (such as psychodynamic or cognitive behavioral) into which are imbricated techniques useful in addressing particular problems such as hypnosis and, more recently, eye movement desensitization and reprocessing (EMDR). The role of psychopharmacology in the management of DID is ancillary. The core symptoms of the condition do not respond to medication on a reliable basis. However, most DID patients do present target symptoms for psychopharmacological intervention (Barkin et al. 1986; Kluft 1984a; Loe❉

212 ❉

❉ The Difficult-to-Treat Patient With Dissociative Disorder ❉ wenstein 1991; Putnam 1989; Ross 1989, 1997). Most of the hallucinatory phenomena do not respond reliably to major tranquilizers because they are dissociative phenomena. Anecdotally, the novel antipsychotics are occasionally able to help overwhelmed DID patients to restabilize.

Expected Response Rates to Conventional Treatments Few outcome studies are available, and those available are difficult to compare. I (Kluft 1982, 1984b, 1986, 1993b, 1994b, 1994d) have reported on a series of cases in my private practice. In 1984, I (Kluft 1984b) reported that of 123 monitored treatments, 83 reached integration (67%) and 33 (27%) fulfilled the rigorous stable integration criteria for my study. By 1986, I (Kluft 1986) reported that 106 of these monitored patients (86%) had reached integration and that 52 (42%) fulfilled the more demanding criteria. As of this writing, 109 (89%) of the initial group have achieved integration; 3 (2%) remain in active treatment; and the remainder (11 [9%]) are either deceased, have left treatment, or have terminated without achieving integration, although approximately half of them are much improved. Coons (1986) followed 20 DID patients in treatment with 20 therapists for an average of 39 months. Five had integrated completely, and approximate two-thirds were considerably improved. Ellason and Ross (1997) reported on a 2-year follow-up on 54 of 103 DID patients admitted to a dissociative disorders unit. Those available for follow-up and those who could not be followed up were demographically identical groups. Twelve (23%) were reported by their therapists to have achieved integration by my (Kluft 1984a) criteria. As a group, they were very much improved. Based on treatment response, there are three groups of DID patients: one which responds rapidly to specific treatment; one which responds more slowly, more erratically, and at times less completely; and one which responds still more slowly and is unlikely to have a complete resolution of DID (Horevitz and Loewenstein 1994; Kluft 1984b, 1994d). The essential characteristics of the rapid responders, or high-functioning DID patients, are that they have at least some significant resources (psychological, interpersonal, social, vocational, and financial) and relatively little psychological comorbidity or personality disorder. If they have such conditions, they are related to traumatic phenomena that give the appearance of representing borderline personality disorder, but they respond far more rapidly than core character disorder phenomena. These patients promptly demonstrate the capacity to achieve mastery over strong affects and distress and can form a good thera❉

213 ❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

peutic alliance. Although they may have some hospitalizations, and their hospital stays may prove to be turning points, they are infrequently in the hospital and are treated as outpatients. They constitute from between 20% to 60% of DID patients, depending on the setting of the study. Many of them respond rapidly in any setting, and others seek out experts and are overrepresented in those therapists’ private practices. An intermediate or complex group with comorbidity is characterized by fewer resources and/or more problematic comorbidity and personality characteristics that are associated with a poorer prognosis. A patient should not be designated a member of this group without having had a considerable trial of adequate therapy, because many apparently very chaotic DID patients rapidly mobilize in appropriate therapy. Such patients may seem more convincingly borderline; may suffer comorbid substance abuse disorders, brain damage, eating disorders, medical illnesses, and more complicated affective disorders; and be troubled by severe marital and family difficulties. They may demonstrate “dependency, low autonomy, external locus of control, blaming, and self-preoccupation” (Horevitz and Loewenstein 1994, p. 292). They constitute the majority of DID patients in many settings. Their treatment will be much slower than the former group, their potential for overall growth may be less pronounced, and their capacity to attain integration may be compromised. This group fares much better with experienced and skilled therapists. A final group, called “enmeshed” by Horevitz and Loewenstein (1994), “poor prognosis” by Caul (1988), and referred to as “low-functioning” or “chronic” in workshop settings, is characterized by more severe manifestations of the features noted in the intermediate group and/or may have chronic or intermittent psychotic features that suggest bipolar, schizophrenia spectrum, or hysterical psychotic (psychosis NOS) diagnoses. Furthermore, they are often “enmeshed in abusive relationships, have a ‘dissociative’ lifestyle, and actively participate in self-destructive and/or antisocial behaviors and habits” (Horevitz and Loewenstein 1994, p. 292). They are not candidates for the traditional integration-oriented therapy unless they improve tremendously in a supportive therapy format and generally require help with basic coping with life and with their disorder on a sustained basis. They are estimated to constitute between 10% and 33% of DID patients, depending on the setting. It generally takes 6 months to a year of competent DID treatment to determine how a DID patient is likely to progress (Kluft 1994b, 1994d). An instrument for monitoring treatment, the Determinants of Therapeutic ❉

214 ❉

❉ The Difficult-to-Treat Patient With Dissociative Disorder ❉ Movement Instrument (DTMI), is available (Kluft 1994d) and offers clinicians a means of quantifying the DID patient’s progress in therapy. It is possible that a patient who initially appears to be an excellent candidate for a definitive treatment may prove to require supportive treatment, and a patient who at first appears too chaotic or damaged for anything but supportive treatment may demonstrate the capacity to proceed toward integration.

Specific/Pertinent Causes of Poor Response to Initial Treatment Numerous factors may play a role in the failure of the DID patient to respond to the initial approach to treatment. The following italicized factors may be extracted from the text and used as a checklist.

Process/Interaction Variables The most crucial concern is the state of the therapeutic alliance. Without a strong and reliable therapeutic alliance that extends across the majority of the major personalities, it is difficult for the therapy to be safe, stable, secure, and contained. A therapeutic alliance that is strong enough to sustain supportive work may lack the power to support more painful exploratory work. A therapeutic alliance with some alters that does not involve other “major players” may make the therapist a participant in a civil war among the alters rather than the therapist of the whole human being, and/or alters that are not identified with the goals of therapy may act to undermine the process. It is important to note and attend to any real or potential difficulty in the therapist-patient match. Some DID patients are critical or apprehensive about their therapists and test them and query them at length; others are so fearful of rejection or desperate for treatment that they try to deny or minimize any concerns. Also, it is important to ascertain whether the stance, pace, and modalities of the therapy are optimally matched to the patient. Work with a patient who has had such difficulties in a prior therapy, or responds poorly to initial approaches, should be studied to detect these types of difficulties. When a patient encounters difficulties managing traumatic material, it may be advisable to use the more gradual, fractionated approaches (Fine 1991; Kluft 1988b, 1990, in press). Two common factors that may impede therapy are apprehension about dealing with traumatic material and fears on the part of some alters of losing ❉

215 ❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

control to others. Dysphoria about either or both of these two issues may cause the patient to become reluctant to proceed with treatment. A most common cause of poor response is the failure of the therapy to complete the tasks of the stage of safety before proceeding into trauma work. The therapist may inadvertently have given short shrift to the early stages of the treatment, either through inexperience or because he or she believes the “real” work of the treatment consists of addressing the traumatic material. In fact, without achieving the goals of the early stages, trauma work is generally contraindicated. Techniques useful to reground and restabilize the patient are generally taught in these early stages, and moving forward without mastering them builds an instability into the therapeutic enterprise (Boon 1997; Kluft 1993a, 1997). In his or her eagerness to please or driven by a sense of inner urgency, the patient may have represented him- or herself more prepared to go forward than is truly the case, may be overly eager to get into the traumatic material and get it “over with,” or may have overridden the voices of alters that are not prepared to go forward. Cooperative alters may not wish to acknowledge those that oppose or harbor serious reservations about the therapy and/or the therapist. There is controversy over whether it is wise to try to explore the alter system prior to proceeding or whether to allow the alters to emerge gradually. Those that argue for gradual emergence are often afraid of exerting undue suggestion or creating iatrogenic alters. Those that favor exploring reason that because DID is a condition of hiddenness, and many alters rarely emerge spontaneously or exist within the “third reality” of an inner world (Kluft 1998), many important alters will remain unknown if they are not accessed and may undermine the psychotherapy from behind the scenes. In 22 years of offering consultation to therapists treating DID, my most common finding in cases that are not progressing well is that major alters have not been discovered and/or brought into the treatment (see Kluft 1988a, 2000). Often a patient will need a moratorium before moving into trauma work and ask for it by a failure to progress. Remaining focused on supportive goals for a period of time usually allows the patient a chance to regroup and begin to move forward. Another common problem occurs when the therapist and patient believe in good faith that they are ready to pursue a particular therapy goal, but a life goal or concern of the patient’s is so compelling that it makes progress with the therapy goal unlikely to succeed. If such concerns are not accorded a respectful hearing, and reassurances or interventions designed to address such concerns are not made, treatment may be stalemated or may become a battleground. ❉

216 ❉

❉ The Difficult-to-Treat Patient With Dissociative Disorder ❉ Finally, it is not uncommon for the DID patient to withhold feelings about the therapist that he or she feels he or she should not have, does not want to have, feels he or she needs to protect the therapist from, or fear will upset the therapist or cause the therapist to be upset with the patient. Because many of the transferences of DID patients are based on traumatic expectations, a failure to air and explore them may impede therapeutic progress.

Therapist Variables There are several therapist factors that may play a role in impeding therapeutic progress, but rigidity, fear, inexperience, skepticism or credulity, countertransference difficulties, and difficulty working with the alters are major concerns. Certain therapists labor strenuously to make their preferred techniques, paradigms, and theories work, whether or not they are proving effective with a DID patient. Flexibility and a willingness to work with many models and approaches characterize successful therapists, partially because many patients require an eclectic approach and partially because imposing one’s paradigm upon a patient often recapitulates an abusive interaction with a prior authority figure and reenacts a sadomasochistic scenario, regardless of the therapist’s good intentions. Many therapists are simply scared of the DID patient, fearing violence, experiencing therapeutic impotence, having to deal with upsetting material, anticipating having to master many new skills, and anticipating a drain upon their time and energy. The treatment of DID raises so many issues and complexities that only a minority of psychotherapists, regardless of their general competence and experience, are either “naturals” or adapt smoothly and rapidly to this work. For most, work with DID proves to be an unsettling and unpredictable amalgam of the familiar and the novel. Until a therapist worked with three or four DID patients for 3–4 years, he or she is likely to continue to encounter situations that prove challenging and raise important questions about how to proceed. If the inexperienced therapist is willing to ask questions of those with more experience and expertise, that therapist is likely to do well. Egregious therapist errors are not uncommon among neophytes, but most will be forgiven or tolerated if the therapist is perceived as good hearted and willing to learn. Conversely, the neophyte who retreats into a defensive authoritarian stance and is unable to acknowledge error is unlikely to succeed in correcting the direction of the treatment. If the therapist holds either a skeptical or credulous stance toward what ❉

217 ❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

the DID patient states, or toward the condition of DID, therapeutic misadventure is just over the horizon. Both stances superimpose a therapist attitude on the therapeutic encounter and disable the therapist’s capacity for accurate empathy. The patient who feels embroiled in an ongoing battle over the credibility of what he or she says in therapy will not feel well received. Conversely, if a patient comes to find there are no efforts to explore what he or she says, that patient will feel unprotected from his or her inner turmoil and its products, which may be the cause of great uncertainty. The therapist should be aware that in most DID patients there will be alters that take different stances about the reality of an account or an experience (reflecting either their different perceptions or their different motivations) and that the therapist’s being either skeptical or credulous on an ongoing basis is sure to complicate the treatment (Kluft 1994c). Countertransference errors are extremely common in work with DID patients because they are so complex, their crises can prove difficult to manage, their traumatic material is so affectively charged, their attachment and dependency issues are often stressful to address, the issues surrounding their treatment are so fraught with controversy, and the therapist so often finds him- or herself feeling exasperated or deskilled (Coons 1986; Kluft 1994c; Loewenstein 1993). Addressing childlike, seductive, and aggressive aspects of DID behavior can also prove problematic. Suffice it to say that the therapist’s issues will interact with abuse patterns in the patient’s past to create a difficult ambience within the therapy.

Patient Variables, Characteristics, and Attributes and Choices The DID patient brings much to the therapy that may prove challenging or problematic and may make deliberate choices that complicate the therapeutic process. In some instances, the nature of the patient and the choices that stem from this are hard to distinguish. Comorbidity is a crucial concern. The failure to diagnose and address a comorbid mental or medical disorder can render the DID inaccessible. Do the manifestations of another disorder constitute another freestanding disorder or represent a further manifestation of the DID, which may take myriad forms and mimic many disorders? Substance abuse may prove a particularly difficult issue here, with some DID patients appearing to be genuine substance abusers and others abandoning substances when the issues of particular alters who used the substances have been addressed. Lists of comorbid phenomena have been published by Ross (1997) and me (Kluft ❉

218 ❉

❉ The Difficult-to-Treat Patient With Dissociative Disorder ❉ 1991), but both are incomplete, especially with regard to somatoform manifestations. Suffice it to say the DID patient should receive a careful physical examination, with follow-up of all significant findings. In some cases neurological and/or neuropsychological examinations are useful. In some consultation cases, I have found that comorbid cognitive difficulties virtually precluded a DID patient’s profiting from verbal psychotherapy. A switch to communication in writing, however cumbersome, led to gratifying results. Many DID patients are reluctant to accept comprehensive medical assessments and interventions; such refusal, whatever its motivation, often is a poor prognostic sign. Although DID itself is not invariably associated with a general compromise of ego strength, ego strength may be compromised either by the fragmentation associated with extreme complexity, by rapid switching, or by the impact of comorbid conditions. The optimal treatment of DID requires considerable ego strength to form a reasonable therapeutic alliance, participate actively in the treatment, use self-control and containment techniques, maintain organizational coherence in the face of stress, and participate in the disassembling of the dissociative psychopathology. All conditions that might adversely affect ego strength must be treated, and general psychotherapeutic and specific cognitive efforts to build ego strength may be advisable when indicated. To participate in an exploratory psychotherapy, the patient must display a “healthy masochism,” an aspect of ego strength that is a willingness to endure the pain that is inherent in undergoing the treatment. Many DID patients are pain-phobic and characteristically seek comfort, nurture, and support. Whereas periods of such behavior are encountered in virtually all DID treatments, some patients hold this stance to an extreme extent and may even, in protector alters, threaten to harm the therapist or to withdraw from treatment if particular other alters are caused distress. Such patients, if they cannot be helped over a period of time to acknowledge the benefits of a definitive treatment, may require purely supportive efforts. Managing the DID patient who threatens harm to self, others, or the therapist requires a matter-of-fact assessment of the reality of the risk. DID patients’ threats to harm themselves may have many motivations, among which alters’ punishing one another for revelations or anticipated revelations, distracting from one type of pain by creating another, or associating self-harm with a relief of tension are quite common. Usually safety contracts against such behaviors can be obtained, but when this is not possible, the clinician must decide, based on the patient’s prior pattern of behavior and cur❉

219 ❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

rent situation, whether hospitalization is warranted. Hospitalization under such circumstances is rarely beneficial in the long run and may be traumatizing to the patient, who rarely will be treated in a manner suitable to address the DID diagnosis, but may be advisable to address an acute crisis. When threats are made against another person and appear to be serious, appropriate warnings should be made, and the patient should be hospitalized, involuntarily if necessary. Threats made against the therapist usually occur when protector parts or parts based on an identification with the aggressor fear that the therapist is getting too close and that other alters’ regard for the therapist is undermining the protective roles that they play or the power that they wield. In the vast majority of cases, conversation with the alters involved will lead to a safety contract. However, attacks on therapists by DID patients do occur; in those unusual cases in which a therapist is genuinely concerned, and/or cannot enter a dialogue with the parts that are problematic, it may be wise to obtain consultation, to have a colleague present in sessions, or even to consider a period of hospital care. It is often assumed that DID patients are fantasy prone (Wilson and Barber 1983) and that the capacity to visualize phenomena and accord veracity to them accounts for their alleged proclivity to confabulated pseudomemories. Unpublished work by Loree Little casts doubt upon those assumptions as general principles: DID patients were no more likely than normal subjects to misattribute the source or reality of perceptions and less prone to do this than a group of mixed psychiatric patients. However, when definite fantasy proneness is encountered it can complicate treatment because fantasy material may be represented as memory, including fantasies in the transference! The therapist should neither assume nor discount that the DID patient’s material reflects such an origin. When present, considerable amounts of therapeutic time may be consumed by work on material that is not historically accurate but which is experienced as such by the patient. It is occasionally possible to break through such confabulations with evidence to the contrary or reassurance. It is often suggested that the therapist should consult family members under these circumstances. However, such recommendations bypass considerations of confidentiality and the negative impacts of such efforts on the therapeutic alliance and the transference. They also implicitly assume the veracity of persons who may be the subject of allegations of abuse, a questionable premise. Such contacts with family members may be held (in legal situations) to indicate that the psychiatrist has a therapeutic relationship with third parties to the treatment, has a duty to them, and may be held liable for any harm that they believe the therapist and/or the therapy ❉

220 ❉

❉ The Difficult-to-Treat Patient With Dissociative Disorder ❉ may have done to them. Overall, there is no compelling reason to assume that contact with family members will result in information any more accurate than that which has been provided by the patient. Such contacts should be undertaken only with the patient’s informed consent. A major consideration may prove to be the importance and the intrusiveness of what has been called the “third reality” (Kluft 1998). This refers to the inner world of the DID patient, in which the alters are understood to interact with one another and to which some DID patients accord a reality equally or more compelling than external reality. Many times frustration in psychotherapy is related to the patient’s greater allegiance to or responsiveness to pressures from this inner world, in comparison to which the therapist and external reality are dismissed. It is always useful to ask the alters what factors they are taking into consideration when their decisions are contrary to those that would advance the treatment. It is not uncommon to discover that alters fear, resist, or sabotage therapy because they are apprehensive that the treatment will destroy them or interfere with what they understand to be their vested interests, which may not be the same as those of the patient as a whole or of other alters. For example, alters based on abusers may experience an attempt to relieve general suffering as interfering with their own defense of turning passive suffering into identification with the aggressor and perpetuating suffering in other alters within the “third reality” of the patient’s inner world. Fearing being rendered passive and/or unable to externalize their own pain, they may undermine the therapy and terrorize alters that attempt to cooperate with it. A classic arena for such difficulties regards alters’ fear that integration means their death. Although such concerns are usually dealt with straightforwardly late in therapy, early in treatment they can lead to major roadblocks and resistances. A number of issues surround the management of traumatic material other than the painfulness of the material and the vicissitudes of traumatic memory. Some patients develop a preoccupation with knowing “the truth.” Psychotherapy is far more effective in bringing about recovery than it is in determining the historical accuracy of autobiographical memory. Some patients refuse to do any work on material that they do not know to be true, lest they be “unfair” to or betray those in the apparent memories. For others, “the truth” means credulous acceptance of all that is alleged by all alters, and the pursuit of “the truth” becomes a destabilizing search for every last experience of alleged mistreatment. These are parallels to the poles of skepticism and credulousness described among therapists; the patient must be helped ❉

221 ❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

to achieve a more moderate perspective and appreciate that therapy is a process of healing, not a process of investigation. I find it useful to remind the patient that if the treatment frame is honored, all material and allegations remain within the treatment and can do no harm to the person or reputation of any third party to the treatment. Issues of loyalty to persons who are alleged abusers, fearful concerns of being punished (by external others or inner alters) for making revelations, and shame-driven pressures toward privacy can paralyze DID treatments. Some patients have had experiences that are so mortifying that they cannot bear to reveal them or fear their revelation would disgust, drive off, or sexually stimulate the therapist. At times patients fear if they were to learn that they were mistreated they would be unable to control their rage and therefore do not dare to discover the contents of their minds. Patients from cultures that demand extreme loyalty toward and subservience to family members may have terrible difficulty in dealing with intrafamilial abuse. I have treated several southeast Asian women whose therapies progressed with agonizing slowness for this reason, even in instances in which the abuse was confessed and apologies had been made. Attachment issues (Barach 1991; Liotti 1992) may also prove difficult, because often DID patients anticipate that making revelations will lead to the loss of major figures in their lives. For many patients attachment issues so dominate their concerns that they are unable to address themselves to any other considerations, and a supportive focus may be necessary for some time, if not for the duration of treatment. Previous life experiences may determine or contribute to problems in treatment. Typically, the DID patient has suffered severe traumatization and betrayal and develops a number of negative transferences toward the therapist. These transferences include the traumatic transference, in which active harm from the therapist is anticipated, and transferences based on failed protectors, which lead the patient to feel that the therapist will prove inadequate to protect and to provide help (Kluft 1994c; Loewnstein 1993). Consequently, as the DID patient becomes closer to the therapist, not only do positive connections become stronger, but the negative ones are enhanced as well. Operationally, this may mean that as some alters come to trust the therapist, others fear this will pave the way for their exploitation and destruction by the therapist. This is especially difficult when the DID patient has been exploited by a previous therapist and is somewhat problematic when prior therapists have missed or mistreated the diagnosis. Often issues of control and fears of dependency are problematic, and the ❉

222 ❉

❉ The Difficult-to-Treat Patient With Dissociative Disorder ❉ DID patient will attend therapy but will decline optimal medication or refuse to allow the use of a useful treatment modality, such as hypnosis. This may place major obstacles in the way of the therapeutic work. Finally, for some DID patients dissociation is not so much a symptom as a characterologic adaptation (e.g., Brenner 1994). Such patients are unlikely to progress until treatment has provided them with alternative strategies for managing their distress.

External Factors DID is difficult to treat even under optimal conditions, and optimal conditions rarely prevail. Treatment may be compromised, even stalemated, by factors extrinsic to the therapy itself. First, logistics may be problematic. DID does best with intense individual psychotherapy. Twice-weekly treatment is the minimum intensity with which most DID patients can both be stable and make progress, and three sessions per week or more may be optimal, preferable, or even necessary for some DID patients. Some patients require or do better with a single extended session per week or more. Hospital milieus capable of supporting intensive therapy or even dedicated dissociative disorder programs can be very helpful. In the 1970s and early 1980s, I succeeded in integrating a large cohort of DID patients very rapidly (e.g., Kluft 1984b, 1986). I was able to see my patients several times per week, use extended sessions when indicated, and hospitalize patients to sustain them during difficult moments in their therapy. Despite advances in the field, much greater experience, and, on the whole, less ill patients, I cannot duplicate my earlier results when restricted to once or twice weekly treatments and with minimal hospital support. Managed care organizations generally have not provided therapist expertise in DID to their DID subscribers and have not supported the standards for the treatment of DID published in the literature. Furthermore, by labeling change-oriented treatment as regressive, many of these organizations have rationalized withholding all but the most minimal supportive treatments. As a result, the average DID patient is undertreated. Also, the media’s coverage of issues closely related to DID, including the false memory controversy and legal charges against therapists who work with DID, frequently impinges on the therapy. It is not uncommon for therapeutic work to be damaged if not undone by the assaults on psychotherapy of the abused and the credibility of those who allege that they have been abused. I have had DID patients insist, on the basis of media accounts, that they must be mistaken in their allegations, even though I have been in pos❉

223 ❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

session of evidence that documents the fact of their abuse. Coverage of highprofile legal cases often makes it difficult for patients to trust their therapists. Furthermore, other third parties to the treatment may prove problematic. Significant others in the patient’s life often oppose the treatment or insist that the patient’s recollections of abuse are erroneous. Frequently they protest the changes in the patient, complain about the length of treatment, or despair over the patient’s symptoms or chances of recovery. At times third parties can play a constructive role, but this cannot be assumed a priori. It is best to proceed with caution. Closely related is the problem of revictimization. Many DID patients are not able to protect themselves from being reabused or are enmeshed in abusive relationships. These exploitations and retraumatizations wreak havoc with the capacity of the DID patient to participate in treatment. In addition, intercurrent stressors may make it necessary to prioritize stabilization over moving forward. Often management of the patient’s life issues will have a higher priority than addressing his or her DID, and work will have to focus on its containment.

The Next Clinical Steps (for Acute and Maintenance Phase) Determining the next clinical steps in work with a DID patient who is proving refractory or difficult to treat involve the systematic assessment of potential impasses and misalliances in the therapy, potential failures to address comorbidity, and potential inaccuracies in the therapist’s estimation of the patient’s readiness to move forward in treatment. Ironically, because most DID patients are considered difficult to treat, and the majority of DID treatments go through difficult phases, a literature that specifically addresses the refractory or difficult-to-treat DID patient has not developed. Virtually every anecdotal account of a therapy, whether in the lay or scientific literature, details the struggles of a therapist-patient dyad to surmount a series of obstacles that is impeding the progress of the treatment. The items italicized in the previous section offer a framework for reviewing a DID treatment with regard to potential obstacles and therapeutic impasses. This approach may be used by the therapist to study the treatment, by the therapist in a dialog with the patients as they review the treatment together, or by a consultant to whom a treatment is being presented, from the perspective of the treating therapist or the patient. Every area that is problematic should be addressed to maximize the likelihood of therapeutic suc❉

224 ❉

❉ The Difficult-to-Treat Patient With Dissociative Disorder ❉ cess. Most of these areas are self-explanatory or self-evident, but a small number address issues related to the alter personalities. They are unique to the treatment of DID and may not be familiar to the general psychiatrist. Here they will be addressed further. Pragmatically speaking, although there is much theoretical controversy over whether therapists should address and work with the alter personalities, those therapists who have been most successful in the treatment of DID do work with the alters. To summarize an argument presented in detail elsewhere (Kluft 2000), the successful therapist addresses the DID patient both as a single human being and an aggregation of alters, each with its own identity, self-representation, autobiographic memory, and sense of ownership of its own activities, conation, and mentation (Kluft 1991); that is, without challenging the subjective reality of each alter’s self and experience, the therapist keeps before the patient (when therapeutically meaningful) that all alters are aspects of a single human being. Because the alters are understood to be personified alternative approaches to coping with life circumstances, they are not regarded as extrinsic to a core personality but as aspects of it. The alters generally trace their origin to specific traumatic or overwhelming circumstances that may be segregated into their individual autobiographic memories. To integrate the identity and autobiographical narrative of the DID patient, the contents and attributes of the alters must be shared with one another. In Coons’ (1984) terms, the personality of a DID patient is to have multiple personalities. Therefore, the personalities are to be addressed and worked with. Because many alters emerge infrequently or exist in the inner world of the third reality (Kluft 1998), it may be necessary to make efforts to access them to address the issues with which they are associated. In clinical practice, although working with alters may initially make them appear to become more three-dimensional and active, appearing to support the idea that working with them reinforces them, it sets the stage for their identification, empathy, collaboration, and support for one another, which paves the way for their more smooth functioning as a system of alters, and, in complete cures, for their ultimate integration. Three activities are very useful in working with the alters: mapping the alters system, accessing the alters, and reconfiguring the alter system to address certain clinical contingencies. Without the therapist’s possessing these skills, the treatment of DID becomes very difficult, and it is hard to protect the patient optimally in the course of the therapeutic work. Mapping involves efforts to discover the “roster” of the alter system and the alters’ patterns of interaction. There are several ways of going about this, including ❉

225 ❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

asking each alter to tell what it knows of the others and simply requesting alters to come out one after another and introduce themselves. Unfortunately, the interaction patterns of the alters often make these straightforward methods problematic (e.g., an alter asked to describe others that do not want to be discovered may be threatened or attacked by the others). An approach to mapping first described by Fine (1991) is often helpful. Fine instructs the alter that is out to place its name on a sheet of paper and then invites the others to place their names on the sheet as well, next to those others to whom they feel closest. Alters without names or who do not want to reveal their names can be encouraged to make a mark to indicate their presence. Even if such an effort is only partially successful, most experts find it is preferable to working in the dark. I once was referred an adolescent girl with DID who had been hospitalized for the better part of 2 years because of self-destructive acting out by a second personality and childlike behavior by a third. When mapped, within 15 minutes she described a system of over 60 alters, all but 2 unknown to her prior psychiatrist. With her situation better appreciated, she became treatable, and did well. When therapy is not progressing well, it is important to consider discussing the situation with all alters and negotiating agreements to address any areas of difficulty and concern. It is not at all unusual to find that apprehensions, misgivings, and tensions in particular alters or the alter system as a whole may be playing a part. One fairly common finding is that whereas some alters are acting as if they were ready to go forward, other alters have not really mastered the goals of the first portion of therapy, and therefore have no sense of safety about moving forward. Furthermore, in clinical practice, one of the most common findings in treatments that have not prospered is the presence of unsuspected additional alters, as in the previously illustrated instance. A vigorous search should be made for such parts. Accessing alters is often essential. Often the alters with which therapy must be done most urgently are not in control of the body for long periods of time, and their emergence may be opposed by other alters. Simply asking to speak to a particular alter, or whichever alter was involved in a particular incident, is often effective. When the personality that is out is unable or unwilling to get out of the way, simple eye closure or the induction of hypnosis to facilitate the therapist’s request is often successful. When such efforts are undertaken but fail to access the alter that is sought, asking other alters to direct the therapist to the sought alter is often helpful. As a general rule, whichever alter enters the session should leave it. Exceptions occur when ❉

226 ❉

❉ The Difficult-to-Treat Patient With Dissociative Disorder ❉ the alter that arrives is or becomes dysfunctional with respect to a crucial activity or integrates in the course of the session. It also is useful to be able to reconfigure an alter system to facilitate its function. A common example of this is hypnotically placing those alters asleep between sessions whose overwhelmed state compromises the function of the patient as a whole. Another is to request personalities to help one another or to share time. For example, a law student failed his examinations because he could not recall his class work or what he studied. It emerged that a child personality was often coming out, uncomprehending of his schoolwork or his circumstances. Once accessed and worked with, this alter was able to remain asleep during grownup activities, and the patient achieved excellent grades. In addition to reviewing the treatment for potential causes of impasse and addressing matters of skill acquisition, it becomes essential to reassess whether the patient is in the right type of DID therapy, a supportive treatment or a treatment attempting to work toward the resolution of the DID and the integration of the alters. As noted earlier, if the DID patient cannot master the tasks of the first stages of therapy, or does so but repeatedly fails to handle work directed at the resolution of the impact of traumatic experiences, with few exceptions the patient should be retained in a supportive format. More structured criteria for retaining a patient in a supportive treatment have been put forward by Boon (1997; illustrated in van der Hart and Boon 1997) and this author (Kluft 1994b, 1994d, 1997). In general, the only reasons to go forward with trauma work with patients who do not fulfill criteria for doing so are 1) that the clinician has come to a reasoned clinical decision that the patient cannot be stabilized until a particular bit of trauma work was successfully concluded and 2) traumatic material is intrusive and containment measures are proving unsuccessful (Kluft 1997). (Detailed studies of early stage tasks are available in Boon 1997 and Kluft 1993a.) When treatment is not progressing adequately, it is appropriate to revisit the biological and medication aspects of treatment and to make sure that they have been thoroughly considered and optimized if indicated. Few psychopharmacological studies of the treatment of DID are available, and psychopharmacology does not address the core phenomena of the dissociative disorders. Ross’s 1989 observation that every medication treatment of DID is an uncontrolled clinical trial still is relevant. Loewenstein (1991) is an excellent resource on using medications to address target symptoms thoughtfully. In general, it is not typical for medications to lead to completely classical responses in this patient population. For example, where there is ❉

227 ❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

depression, there also is human misery that remains largely unaffected by medications. Where there is anxiety or panic, there is also the terror of passive influence experiences (when another alter or another alter’s issues or affects intrude into the experience of another alter) and the misery of flashback and reenactment phenomena. Despite the difficulty in being sure what is the true impact of any medication intervention, it is often appropriate to make clinical trials of medications that seem likely to affect particular target symptoms. A common error made by psychopharmacologists involves the conceptualization of alters’ voices or flashback phenomena as the auditory hallucinations of psychosis and prescribing major tranquilizers. Although this may appear to help initially, in the long run it is often more effective for the therapist to access the alters whose voices are being heard and negotiate with them for less intrusive behavior (Kluft 1983). If such interventions are not possible, the novel antipsychotics, in usual doses, may be useful for a period of days or weeks. Thus far, experience is anecdotal and has not established a body of experience that allows the recommendation of particular medications or dosage regimens. van der Kolk (1987) established the rationale for using selective serotonin reuptake inhibitor (SSRI) medications to diminish the disruptive impact of trauma on thinking patterns. Nathanson (1991) has speculated that these medications may interrupt pathological shame cycles. Anecdotally, I have found that the use of SSRIs for these purposes is often quite rewarding. Standard dosages are useful and should be continued for a period of months before tapering. It is useful to bear in mind that many DID patients have or have had problems with prescription medication abuse or substance abuse (Putnam et al. 1986; Ross 1989, 1997). Although many DID patients will occasionally misuse medications, on the whole this does not constitute a sufficient general rationale for withholding compassionate psychopharmacology. When treatment has not progressed adequately, it becomes even more imperative to address all comorbid conditions, medical and mental. The number of patients I have seen in consultation who refused to take medications for comorbid depressions or who refused physical assessments and spent years compromised by easily addressed medical conditions is high. Because our medical colleagues often give psychiatric patients a poor evaluation and do not take their subjective complaints seriously, it is imperative that the therapist become an advocate for the DID patient’s optimal medical care. ❉

228 ❉

❉ The Difficult-to-Treat Patient With Dissociative Disorder ❉ It becomes essential to optimize the therapy that is being delivered. Consultation often helps to sharpen the therapist’s focus and to pick up ways in which the treatment has wandered off course. If the therapist is using approaches with which he or she has minimal experience, this advice is all the more relevant. It is often useful for the therapist to switch stances to assess whether the perspective from which the therapy is being conducted should be altered. The treatment of DID can be conducted from a number of stances (Kluft 1988c, 1993b), most of which are useful from time to time in any therapy. However, treatment usually is dominated by a particular stance. The major stances in DID therapy are strategic integrationalism, tactical integrationalism, adaptationalism, personality-centered therapy, and minimization. Strategic integrationalism is an exploratory therapy that focuses on the undoing of dissociative defenses so that the dissociative disorder collapses from within. It may use technique-oriented interventions but prefers a processoriented situation. It addresses the alters but is more concerned with overall dynamics. Its roots are in psychoanalysis and psychoanalytic psychotherapy. It requires a relatively strong patient who can retain good control and stability. Although it is often promoted, it simply requires more ego strength and containment than many DID patients can bring to therapy. It is well described in the work of Wilbur (1986), Marmer (1980, 1991), and Brenner (1994). Tactical integrationalism involves the application of technical interventions in the context of a process-oriented therapy. It consists of a series of short-term therapies directed at the attainment of specific goals imbricated within the context of a single ongoing long-term psychotherapy. It works toward integration in a step-wise fashion, objective by objective. Its origins are within the hypnotherapeutic tradition, but it marries well with a cognitivebehavioral orientation. It requires less ego strength of the patient, and implies a far more active and interventive stance by the therapist. It is best described in the work of Fine (1991, 1993). Personality-oriented therapy has been used to refer to a problematic treatment approach in which each alter is regarded as a person, many of whom are thought to require a very tangible reparenting by the therapist. This boundary-violating form of therapy is contraindicated. However, personality-oriented therapy also refers to those types of treatment in which working with the alters and the alter system to facilitate a more stable and functional pattern of interactions among the alters is the major goal. Integration may or may not be pursued. It can be understood as a group or family therapy of the self. It can be used with patients at all levels of function and ❉

229 ❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

ego strength, and is best illustrated in the work of Caul (1984) and the Watkins (1997). Adaptationalism focuses on meeting day-to-day problems by encouraging the patient as a whole and the alter system to prioritize containment (minimizing the expression of dissociative phenomenology) and function (the capacity to fulfill the patient’s social and occupational [or educational] roles). It can be used with patients at all levels of function and ego strength. It does not necessarily address trauma or pursue integration. Although it has not developed a literature or demonstrated its efficacy, it is often endorsed implicitly by those who oppose the funding of intensive individual therapy. Minimization is an approach that supposes that if not reinforced, the DID will cease to manifest itself. Although it often achieves that goal within therapy sessions, it rarely carries over and is not associated with successful treatment outcomes. When therapy conducted from a particular stance has not prospered, it is useful to consider, or even make a trial, of another stance. Often the patient will appreciate the change, if only in retrospect. One of my patients who had not done well in a strategic integrationalist therapy, prospered in a tactical integrationalist model, feeling much more secure with my being more active and directive. Another, switched from tactical integrationalist to strategic integrationalist therapy, reflected that when I had been more interventive, she had felt intruded on and invaded. Often we find that the modalities of treatment that have been employed have ceased to be effective or have not achieved the anticipated results. For example, hypnotic interventions may be blocked by autohypnotic defenses, alters may decline to do recommended cognitive-behavioral prescriptions (such as refusing to report automatic thoughts because they include warning threats from persecutor alters), and coping strategies that were adequate to buffer the patient from materials encountered during the early stages of trauma work may prove inadequate to contain the dysphoria of the more deeply hidden and more severe traumata. Often a challenged modality can be reestablished by making a minor alteration in its application. For example, when a patient begins to associate a hypnotic induction with the painful reexperiencing of trauma, and balks at hypnosis, either proposing to use hypnosis to do ego strengthening or support or introducing a very different induction may be sufficient. In general, the therapist is well served by being proficient in many approaches to therapy and in many techniques for achieving a given objective. Bringing an additional modality to bear is often effective, even if it does no more than indicate that the therapist is actively trying to turn ❉

230 ❉

❉ The Difficult-to-Treat Patient With Dissociative Disorder ❉ things around. Often it will remain unclear whether such an intervention has worked because of its intrinsic power or because of the heightened expectations and anticipations mobilized by the patient who believes that he or she will be the recipient of something special, new, and powerful. In recent years, I have had tremendous success in breaking through impasses with EMDR (Shapiro 1995) and dialectic cognitive-behavioral therapy (Linehan 1993) techniques. Hypnosis can play an extremely helpful role in the treatment of DID. If it has not already been employed in a treatment that fails to progress, its addition should be considered. DID patients are a highly hypnotizable population. Even if a therapist declines to use heterohypnosis (hypnosis that is deliberately induced by another), autohypnosis and spontaneous trance phenomena play a major role in DID phenomenology and function. Although much controversy surrounds the use of hypnosis in memory retrieval, there is ample evidence that memory distortion is far from inevitable and that accurate memory recovery is often possible (Kluft 1998). In any case the use of hypnosis for memory work should be done under the aegis of informed consent, which should specifically address the vicissitudes of autobiographical memory. The controversy surrounding memory has obscured the fact that hypnosis is useful in accessing alters, reconfiguring the alter system, ego strengthening, soothing overwhelmed alters, isolating material that threatens to unsettle the patient, and improving the patient’s functioning. For example, one useful technique is to create a hypnotic “safe place” and then to suggest to alters who are overwhelmed and/or whose feelings would flood the patient and compromise function that they go to the safe place and enter a state of dreamless sleep between sessions. This often enables the DID patient to continue to function despite doing difficult trauma work in the therapy sessions. Another important use of hypnosis is to contain and control the abreaction of traumatic material (see Kluft 1988b, 1989, 1990, 1994a, in press). In general, when a therapy is not progressing, not only should there be a vigorous effort to find specific problem areas but there should be ongoing efforts to build strength and enhance coping. There also should be efforts to revisit issues thought to have been addressed to check whether these apparent successes were “flights into health” that bypassed rather than resolved the issues under consideration.

❉

231 ❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

Clinical Vignette Ms. H, a successful but desperately miserable professional woman in her mid-30s, had sought therapy in her 20s for recurrent depression, anxiety, and difficulty in relationships. She was suicidal. She was the child of a depressed mother and an alcoholic father. Her brother was a substance abuser, and her sister was chronically depressed. Ms. H always had recalled childhood sexual abuse and recovered additional memories of mistreatment in the course of her therapy. Both conventional psychotherapy and psychopharmacology had preserved her safety and function and somewhat eased her depression, but she remained uncomfortable. After several years, her therapist had begun to notice dissociative symptoms, the first of which was that Ms. H did not recall her sexual experiences with her husband and invariably cried after intercourse, and ultimately arrived at the diagnosis of DID. Ms. H and her therapist began to work with the DID, and after several years, although a dozen alters had been discovered and a considerable amount of traumatic material had been addressed, both Ms. H and her therapist concurred that she remained very uncomfortable and her dissociative symptoms had not improved. In fact, they were increasing. After a suicide attempt by ingestion for which Ms. H had no recall, her therapist requested consultation from a clinician experienced with dissociative disorders. In the initial consultation the therapist presented Ms. H’s case. The consultant observed that the therapist had begun trauma work without having completed the tasks of the stage of safety and had not made efforts to establish the extent of the patient’s alter system. Next, the therapist observed the consultant administer a Structured Clinical Interview for the Diagnosis of DSM-IV Dissociative Disorders, Revised (Steinberg 1994). Ms. H received a maximum score of 20, indicating she was much more disrupted by dissociative symptomatology than either Ms. H or her therapist had appreciated. Furthermore, the mapping exercise uncovered eight additional alters, all of which indicated that they were not prepared to participate in the treatment. One of them had made the disremembered suicide attempt, another participated in sex, and still a third (a traumatized child) was responsible for the weeping after intercourse. The consultant advised that therapy take a supportive stance focused on the goals of the stage of safety, that trauma work be avoided whenever possible, and that efforts be made to build an alliance with the newly discovered alters. The consultant suggested the use of antidepressant medication be continued but be augmented by strategies used to address refractory depression and recommended a trial of hypnosis to contain traumatic material and to work with particular alters. However, neither Ms. H nor her therapist found these recommendations congenial. A year later the consultant was called in again. The alter system had addressed the stage one, or safety and strengthening, concerns, and the therapeutic alliance was enhanced. However, the patient remained miserable and unwilling to proceed with treatment. She wanted to drop out or to

❉

232 ❉

❉ The Difficult-to-Treat Patient With Dissociative Disorder ❉ kill herself because she saw no end to her pain. It was agreed that the consultant would see Ms. H every other week to attempt to discover what was making her course so refractory. Using a checklist including the italicized elements of a previous section (specific/pertinent causes of poor response to initial treatment), the consultant made a systematic assessment. He found that Ms. H was lax about her medical care, continued to fear medications and hypnosis, and had still other alters. He found that the alters were terrified both of reliving traumatic material and of being “killed” by the therapy. Furthermore, he found that Ms. H was behind on her bill and secretly held the fear that she would be asked to pay off the balance of her bill with sexual favors. Over a period of 6 months (12 visits) the consultant systematically addressed all of Ms. H’s concerns, sharing his efforts with her primary therapist. First, the consultant helped Ms. H accept overdue medical evaluation. She was found to have a mild thyrotoxicosis that exacerbated her anxiety. She also was hypokalemic. As potassium supplementation was discussed, Ms. H revealed that she had been abusing diuretics and laxatives to control her weight. When the alters involved with this were accessed and worked with, they agreed to desist from this behavior and to go into hypnotic “sleep” between sessions until their issues had been worked through. This proved the pathway to help Ms. H accept hypnosis. The consultant taught Ms. H’s therapist to use relevant hypnotic techniques and how to approach traumatic material in a gentle fractionated manner (Fine 1991; Kluft 1988b, 1990). In conjunction with these steps, he persuaded Ms. H to accept psychopharmacology. Her depression responded to a combination regimen of fluoxetine 60 mg/day plus venlafaxine 75 mg extended release/day. The consultant then discussed the concept of integration with the alters, indicating that it could not be imposed and would occur spontaneously or with facilitation when the time was right and that if the time were never right, it would not occur. He assured the parts that integration involved a process of flowing together and “being there all the time,” as opposed to a process in which they were destroyed or eliminated; furthermore, he indicated that their cooperation toward a goal of maximal functioning and comfort was more important than integration per se. Ms. H continued in therapy with her primary therapist, and the consultant gradually reduced his involvement to as required. Over the next two years, Ms. H’s mood was euthymic unless she was working with particularly traumatic material, she achieved continuous contemporary memory, and as material was addressed with fractionated abreaction several of her alters spontaneously integrated. She ceased to show spontaneous dissociative phenomena. She continued her therapy and remained on medication. She is currently struggling with whether to proceed toward further integration or to attempt to stabilize at her current level of comfort and function.

❉

233 ❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

Summary and Recommended Treatment Algorithm The treatment of complex chronic dissociative conditions such as DID is an ongoing effort to anticipate and prevent or to address and resolve actual and potential therapeutic impasses. Because these conditions do not respond to medication and require extensive long-term psychotherapy, the treating clinician must be alert to the state of the therapy, monitoring its progress as a process, and be sensitive to the patient’s responses, as both an overall person and an aggregate of alters, to the material that emerges in the therapy and to the therapist and the therapist’s interventions. The therapist must bear in mind that circumstances that might translate into a stalemate for a DID patient able and willing to work toward integration (e.g., the inaccessibility of traumatic material) might indicate a desirable outcome for a patient for whom such ambitious therapeutic goals would be contraindicated and that what might constitute a breakthrough in a patient able and willing to work toward integration (the emergence of new material) could be the prelude to decompensation in a patient for whom a supportive focus would be preferable. Furthermore, the therapist must contemplate the possibility that many patients who initially appear poor candidates for a definitive cure may, after considerable strengthening, be able to be “promoted” into a more ambitious therapy and must be mindful that patients who are candidates for definitive treatment to the goal of integration nonetheless have moments of exquisite vulnerability and cannot simply “march through” their traumatic pasts. All treatments of dissociative patients must be paced with great gentleness and compassion. It is difficult to reduce the treatment of DID to an algorithm, but the general pattern is rather straightforward and is represented schematically in Figure 8–1. Each step of therapy should be assessed as it is attempted. If it is accomplished successfully and if the patient is motivated to proceed, the next step is attempted. If progress is problematic, the treatment should be reviewed using the previously italicized potential problems as an outline or protocol, and each area of difficulty should be addressed. If they can be addressed successfully, treatment moves to the next phase; if they cannot, treatment remains supportive while these areas are addressed. If it becomes clear that the problems cannot be resolved, treatment retains a supportive focus and occasionally revisits the problem areas to see if they now can be addressed. Some treatments will be unable to move beyond a supportive endeavor. Some issues, however, require further commentary. A DID patient who is distressed but unmotivated for specific DID treatment should receive ❉

234 ❉

❉ The Difficult-to-Treat Patient With Dissociative Disorder

❉ 235 ❉

❉

Figure 8–1. An algorithm for the treatment of dissociative identity disorder (DID). a The list of concerns are italicized on pp. 215–224.

❉

236

❉

The Difficult-to-Treat Psychiatric Patient

❉

❉

237 ❉ An algorithm for the treatment of dissociative identity disorder (DID). (continued)

❉ The Difficult-to-Treat Patient With Dissociative Disorder

❉ Figure 8–1.

❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

general supportive psychotherapy. However, a DID patient who has begun therapy, and whose alters have entered the therapy, who then decides against definitive treatment or who is unable to proceed in a definitive treatment is in a different situation. It would be necessary to address the DID while one remains in a supportive stance (supportive DID treatment), as described by Boon (1997; van der Hart and Boon 1997), or to use ego-state therapy (Watkins and Watkins 1997) supportively. Also, because DID patients who apparently cannot or will not go forward in therapy and represent themselves as feeling well usually are involved in a flight into health, it is recommended that their treatment be tapered and that they be periodically reassessed. Often, in these reassessments, they are able to talk about why they “shut down” and return to treatment. Again, the therapist must always be prepared to reassess the DID patient and to recalibrate the therapy to the patient’s situation. Time and time again it may be necessary to return to concerns of the phase of safety, and frequently patients who have required supportive work for years become able to move into definitive treatment. The treatment of complex chronic dissociative patients can be long, grueling, and arduous for patient and therapist alike. Nonetheless, many of these patients can, if they receive appropriate treatment, achieve complete cures or attain the substantial amelioration of their conditions. The therapist should not be daunted or deterred by these patients. Their treatment, although challenging, is one of the most optimistic areas of psychiatry.

References American Psychiatric Association: Diagnostic and Statistical Manual of Mental Disorders, 4th Edition. Washington, DC, American Psychiatric Association, 1994 Barach P: MPD as an attachment disorder. Dissociation 3:117–123, 1991 Barkin R, Braun B, Kluft R: The dilemma of drug therapy for MPD, in Treatment of Multiple Personality Disorder. Edited by Braun B. Washington, DC, American Psychiatric Press, 1986, pp 107–132 Bernstein E, Putnam F: Development, reliability, and validity of a dissociation scale. J Nerv Ment Dis 174:727–735, 1986 Boon S: The treatment of traumatic memories in DID: indications and contraindications. Dissociation 10:67–81, 1997 Boon S, Draijer N: Multiple Personality Disorder in the Netherlands: A Study on Reliability and Validity of the Diagnosis. Amsterdam, The Netherlands, Swets and Zeitlinger, 1993

❉

238 ❉

❉ The Difficult-to-Treat Patient With Dissociative Disorder ❉ Braun B: Issues in the psychotherapy of multiple personality disorder, in Treatment of Multiple Personality Disorder. Edited by Braun B. Washington, DC, American Psychiatric Press, 1986, pp 1–28 Brenner I: The dissociative character. J Am Psychoanal Assoc 42:819–846, 1994 Caul D: Group and videotape techniques for multiple personality disorder. Psychiatric Annals 14:46–50, 1984 Caul D: Determining the prognosis in the treatment of multiple personality disorder. Dissociation 1:24–26, 1988 Coons P: The differential diagnosis of multiple personality disorder. Psychiatr Clin North Am 7:51–67, 1984 Coons P: Treatment progress in 20 patients with multiple personality disorder. J Nerv Ment Dis 174:715–721, 1986 Coons P: Dissociative disorder not otherwise specified: a clinical investigation of 50 cases with suggestions for typology and treatment. Dissociation 5:187–195, 1992 Ellason J, Ross C: Two-year follow-up of inpatients with dissociative identity disorder. Am J Psychiatry 154:832–839, 1997 Fine C: Treatment stabilization and crisis prevention: pacing the therapy of the multiple personality disorder patient. Psychiatr Clin North Am 14:661–675, 1991 Fine C: A tactical integrationalist perspective on the treatment of MPD, in Clinical Perspectives on Multiple Personality Disorder. Edited by Kluft R, Fine C. Washington, DC, American Psychiatric Press, 1993, pp 135–153 Herman J: Trauma and Recovery. New York, Basic Books, 1992 Horen S, Leichner P, Lawson J: Prevalence of dissociative symptoms and disorders in an adult psychiatric inpatient population in Canada. Can J Psychiatry 40:185– 191, 1995 Horevitz R, Loewenstein R: The rational treatment of multiple personality disorder, in Dissociation: Clinical and Theoretical Perspectives. Edited by Lynn S, Rhue J. New York, Guilford, 1994, pp 289–316 International Society for the Study of Dissociation: Guidelines for Treating Dissociative Identity Disorder (Multiple Personality Disorder) in Adults. Glenview, IL, International Society for the Study of Dissociation, 1997 Kluft R: Varieties of hypnotic interventions in the treatment of multiple personality. Am J Clin Hypn 24:230–240, 1982 Kluft R: Hypnotherapeutic crisis intervention with multiple personality. Am J Clin Hypn 26:73–83, 1983 Kluft R: Aspects of the treatment of multiple personality disorder. Psychiatric Annals 14:51–55, 1984a Kluft R: Treatment of multiple personality disorder. Psychiatr Clin North Am 7:9–29, 1984b Kluft R: The natural history of multiple personality disorder, in Childhood Antecedents of Multiple Personality. Edited by Kluft R. Washington, DC, American Psychiatric Press, 1985, pp 197–238

❉

239 ❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

Kluft R: Personality unification in multiple personality disorder, in Treatment of Multiple Personality Disorder. Edited by Braun B. Washington, DC, American Psychiatric Press, 1986, pp 29–60 Kluft R: On giving consultations to therapist treating multiple personality disorder: fifteen years’ experience—Part I (diagnosis and treatment). Dissociation 1:23– 29, 1988a Kluft R: On treating the older patient with multiple personality disorder: race against time or make haste slowly? Am J Clin Hypn 30:257–266, 1988b Kluft R: Today’s therapeutic pluralism. Dissociation 1:1–2, 1988c Kluft R: Playing for time: temporizing techniques in the treatment of multiple personality disorder. Am J Clin Hypn 32:90–98, 1989 Kluft R: The fractionated abreaction technique, in Handbook of Hypnotic Suggestions and Metaphors. Edited by Hammond D. New York, Norton, 1990, pp 527–528 Kluft R: Multiple personality disorder, in Annual Review of Psychiatry, Vol 10. Edited by Tasman A, Goldfinger S. Washington, DC, American Psychiatric Press, 1991, pp 161–188 Kluft R: The initial stages of psychotherapy in the treatment of multiple personality disorder patients. Dissociation 6:145–161, 1993a Kluft R: Treatment of dissociative disorder patients: an overview of discoveries, successes, and failures. Dissociation 6:87–101, 1993b Kluft R: Applications of hypnotic interventions. Hypnos 21:205–283, 1994a Kluft R: Clinical observations on the use of the CSDS Dimensions of Therapeutic Movement Instrument (DTMI). Dissociation 7:272–283, 1994b Kluft R: Countertransference in the treatment of multiple personality disorder, in Countertranference in the Treatment of Post-Traumatic Stress Disorder. Edited by Wilson J, Lindy J. New York, Guilford, 1994c, pp 122–150 Kluft R: Treatment trajectories in multiple personality disorder. Dissociation 7:63– 76, 1994d Kluft R: On the treatment of the traumatic memories of DID patients: always? never? sometimes? now? later? Dissociation 10:82–92, 1997 Kluft R: Reflections on the traumatic memories of dissociative identity disorder patients, in Truth in Memory. Edited by Lynn S, McConkey K. New York, Guilford, 1998, pp 304–322 Kluft R: An overview of the psychotherapy of dissociative identity disorder. Am J Psychother 53:289–319, 1999a Kluft R: Current issues in dissociative identity disorder. Journal of Practical Psychiatry and Behavioral Health 15:3–19, 1999b Kluft R: The psychoanalytic psychotherapy of dissociative identity disorder in the context of trauma therapy. Psychoanalytic Inquiry 20:259–286, 2000 Kluft R: Dissociative identity disorder, in Treatments of Psychiatric Disorders, 3rd Edition, Vol 2. Edited by Gabbard G. Washington, DC, American Psychiatric Publishing, 2001, pp 1653–1693

❉

240 ❉

❉ The Difficult-to-Treat Patient With Dissociative Disorder ❉ Kluft R: The management of abreactions, in Multiple Personality Disorder: Continuum of Care. Edited by Turkus J, Cohen B. Northvale, NJ, Aronson (in press) Knudsen H, Draijer N, Haslerud J, et al: Dissociative disorders in Norwegian psychiatric inpatients. Article presented at the Spring Meeting of the International Society for the Study of Dissociation, Amsterdam, The Netherlands, 1995 Linehan M: Cognitive-Behavioral Treatment of Borderline Personality Disorder. New York, Guilford, 1993 Liotti L: Disorganized/disoriented attachment in the etiology of dissociative disorders Dissociation 5:196–204, 1992 Loewenstein R: Rational psychopharmacology in the treatment of multiple personality disorder. Psychiatr Clin North Am 14:721–740, 1991 Loewenstein R: Posttraumatic and dissociative aspects of transference and countertransference in the treatment of multiple personality disorder, in Clinical Perspectives on Multiple Personality Disorder. Edited by Kluft R, Fine C. Washington, DC, American Psychiatric Press, 1993, pp 51–85 Loewenstein R: Dissociative amnesia and dissociative fugue, in Treatments of Psychiatric Disorders, 3rd Edition, Vol 2. Edited by Gabbard G. Washington, DC, American Psychiatric Publishing, 2001, pp 1623–1652 Marmer S: Psychoanalysis of multiple personality. Int J Psychoanal 61:439–459, 1980 Marmer S: Multiple personality disorder: a psychoanalytic perspective. Psychiatr Clin North Am 14:677–693, 1991 Nathanson D: Shame and Pride. New York, Norton, 1991 Putnam F: The Diagnosis and Treatment of Multiple Personality Disorder. New York, Guilford, 1989 Putnam F, Loewenstein R: Treatment of multiple personality disorder: a survey of current practices. Am J Psychiatry 150:1048–1052, 1993 Putnam F, Guroff J, Silberman E, et al: The clinical phenomenology of multiple personality disorder: review of 100 recent cases. J Clin Psychiatry 47:285–293, 1986 Ross C: Multiple Personality Disorder: Diagnosis, Clinical Features, and Treatment. New York, Wiley, 1989 Ross C: Epidemiology of multiple personality disorder and dissociation. Psychiatr Clin North Am 14:503–517, 1991 Ross C: Dissociative Identity Disorder: Diagnosis, Clinical Features, and Treatment of Multiple Personality. New York, Wiley, 1997 Ross C, Anderson G, Fraser G, et al: Differentiating multiple personality disorder and dissociative disorder NOS. Dissociation 5:88–91, 1992 Saxe G, van der Kolk B, Berkowitz R, et al: Dissociative disorders in psychiatric inpatients. Am J Psychiatry 150:1037–1042, 1993 Shapiro F: Eye Movement Desensitization and Reprocessing: Basic Principles, Protocols, and Procedures. New York, Guilford, 1995 Steinberg M: Structured Clinical Interview for DSM-IV Dissociative Disorders (SCIDD), Revised. Washington, DC, American Psychiatric Press, 1994

❉

241 ❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

Steinberg M: Depersonalization, in Treatments of Psychiatric Disorders, 3rd Edition, Vol 2. Edited by Gabbard G. Washington, DC, American Psychiatric Publishing, 2001, pp 1695–1714 Tutkun H, Sar V, Yargiuc I, et al: Frequency of dissociative disorders among psychiatric inpatients in a Turkish university clinic. Am J Psychiatry 55:800–805, 1998 van der Hart O, Boon S: Treatment strategies for complex dissociative disorders: two Dutch case examples. Dissociation 10:157–165, 1997 van der Kolk B: Psychological Trauma. Washington, DC, American Psychiatric Press, 1987 Watkins J, Watkins H: Ego States: Theory and Therapy. New York, Norton, 1997 Wilbur C: Psychoanalysis and multiple personality disorder, in Treatment of Multiple Personality Disorder. Edited by Braun B. Washington, DC, American Psychiatric Press, 1986, pp 133–142 Wilson S, Barber T: The fantasy-prone personality: implications for understanding imagery, hypnosis, and parapsychological phenomena, in Imagery: Current Theory, Research, and Application. Edited by Sheikh A. New York, Wiley, 1983, pp 340–390

❉

242 ❉

9 The Difficult-to-Treat Patient With Eating Disorder Wendy A. Harris, M.D. Claire V. Wiseman, Ph.D. Susan Wagner, Ph.D. Katherine A. Halmi, M.D.

Eating disorders seem to have earned the reputation for difficult if not impossible to treat among the general public as well as the medical and psychiatric communities. The validity of this opinion is most apparent when applied to anorexia nervosa. The very essence of this eating disorder is a passionate refusal to change in conjunction with a profound denial of illness. However, an informed perspective requires an acknowledgment of the relatively wide range of symptom pictures as well as variations in severity represented by individuals with eating disorders. Treatment and its success seem to parallel such differences in the disorders themselves. In this chapter, we emphasize the treatment modalities that may be useful in more refractory cases. ❉

243 ❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

Diagnosis Anorexia Nervosa Anorexia nervosa is best described in the DSM-IV (American Psychiatric Association 1994). The treatment-resistant features of anorexia nervosa include absolute refusal to maintain weight within a normal range for height and age and an intense fear of weight gain. These two symptoms combined usually result in the third diagnostic criterion: absence of the menstrual cycle or amenorrhea. Body image disturbance complicates the disorder and contributes to the inevitable denial of illness that supports the patients’ refusal to change. DSM-IV describes two subtypes of anorexia nervosa: 1) the restricting type and 2) the binge-eating/purging type. The more traditional concept of anorexia nervosa refers to the restricting type, in which the subject severely limits intake of food and may engage in some form of excessive exercise. In contrast, binge-purge anorectic subjects will eat large amounts of food in a short time period and then aggressively vomit until they believe they are empty.

Bulimia Nervosa Bulimia nervosa was first described as a separate entity 100 years after anorexia nervosa. This eating disorder has some similarities to anorexia nervosa, but significantly distinct characteristics differentiate it. As with binge-purge anorectic subjects, patients with bulimia nervosa binge eat with a sense of loss of control. The binges are distinct overeating episodes (more food consumed than the normal amount for that setting). Following the binge, the bulimic engages in some type of compensatory behavior with the goal of preventing weight gain. Examples of compensatory behavior include the purging type: self-induced vomiting, laxative abuse, and diuretic abuse; or the nonpurging type: excessive exercise, fasting, or strict diets. The binges and compensatory behavior occur at least twice a week for a period of 3 months. As with the anorectic subjects the bulimic subjects have significant dissatisfaction with their body shape and weight.

Binge-Eating Disorder Binge-eating disorder is a provisional diagnosis at this time but is recognized as an eating disorder by most clinicians. Binge-eating disorder shares some symptomatology with bulimia nervosa and is characterized mainly by recur❉

244 ❉

❉

The Difficult-to-Treat Patient With Eating Disorder ❉

rent binge episodes followed by significant distress. Unlike bulimia, subjects with binge-eating disorder do not participate in any compensatory behavior following the binge and therefore tend to become obese. Because binge-eating disorder remains a provisional diagnosis and treatment techniques continue to be refined, the treatment of binge-eating disorder will not be discussed further here. Approximately 90% of people who have eating disorders are female (Andersen and Holman 1997). There are significant differences in the development of eating disorders between genders. However, there is not much difference in the treatment approaches. Some differences occur in the area of hormone replacement, with males requiring testosterone instead of estrogen (Andersen and Holman 1997). Overall, males respond to the same treatments, therefore treatments for eating disorders in general will be discussed here without specific mention of gender differences.

Incidence/Prevalence The study of incidence and prevalence of the eating disorders has presented researchers with formidable problems (e.g., methodological inconsistencies and inadequate definition of terms). Nevertheless, the prevalence rates for anorexia nervosa are estimated at 0.28% (Hoek 1993), with young women between the ages of 15 and 24 years found to be most at risk. Although there are far fewer epidemiological studies of bulimia nervosa, most studies conducted between 1981 and 1989 indicate a 1% prevalence rate (reviewed by Fairburn and Beglin 1990), with a typical age of onset between 18 and 19 years (Keel and Mitchell 1997). Finally, there is clear evidence of a dramatic rise in the number of eating disorder cases between 1931 and 1986 (Eagles et al. 1995; Jones et al. 1980).

Outcome There are recent well-designed studies in which patients were followed for periods of up to 10 (Eckert et al. 1995), 12 (Herzog et al. 1997), and 20 years (Ratnasuriya et al. 1991). However, because this type of study is usually conducted at large academic centers, sampling biases are likely. For example, patients studied in academic settings have often failed treatment elsewhere (Theander 1985). In the well-designed studies, the assessment of outcome included the central facets of the eating disorder: weight, menstrual status, disordered eating ❉

245 ❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

behavior, and attitudes about food and weight (Theander 1985). Several studies have also examined the psychosocial and occupational adjustment of the patients at follow-up as well as comorbid diagnoses (Eckert et al. 1995; Hsu 1980) Although similar studies of bulimia nervosa are fewer in number, reliable findings can be reported about this eating disorder as well.

Outcome in Anorexia Nervosa Given the complexity of anorexia nervosa, the “typical” course is actually quite variable, ranging from full recovery to death. One overriding finding is that anorexia nervosa is a chronic illness. The prevailing outcome of longterm follow-up is that most patients have ongoing problems with the illness (Eckert et al. 1995; Halmi et al. 1991; Theander 1985). According to Kaye and colleagues (1998), 50% of anorectic patients recover, 30% have lingering episodic symptoms, 10% continue to have unremitting symptoms, and 10% die. Chronicity is also implied in Russell’s (1992) finding that meaningful recovery data can only be collected at least 4 years after patients’ initial assessment. It may be the case that recovery (partial or full) is more likely to appear the greater the time span between initial presentation and follow-up. Course of illness also includes changeover from anorexia nervosa to bulimia nervosa for at least 50% of anorectic patients (Eckert et al. 1995). Finally, mortality rates are higher than the early studies suggested. Current reports give rates that range from 6.6% at the 10-year point to 18% at 30 years after presentation (Halmi et al. 1991; Theander 1985). Several outcome studies have found that although many of the patients were considerably improved medically, their disordered cognitions persisted (Halmi et al. 1991; Hsu et al. 1979; Morgan and Russell 1975; Theander 1985). Ratnasuriya and colleagues (1991) found that after 20 years, of those who had not died, one-third were socially isolated, with one-half having difficulty with attitudes toward sexuality, marriage, and childbearing. Patients generally have better adjustment in the occupational sphere on follow-up than in the psychosocial realm (Hsu 1980). In summary, these studies underscore the fact that with our current methods of treatment, anorexia nervosa is often a chronic illness with high morbidity and mortality, and when recovery is achieved, it is often only after several years.

Outcome in Bulimia Nervosa Bulimia nervosa has only relatively recently been defined as an entity separate from anorexia nervosa, and the diagnostic criteria have been changing, ❉

246 ❉

❉

The Difficult-to-Treat Patient With Eating Disorder ❉

complicating outcome research (Russell 1979). Keel and Mitchell (1997) reviewed 88 articles reporting outcome in bulimia nervosa. Of major importance is the finding that relapse is common in this illness. This finding has many implications, but one particularly important implication informs treatment of bulimia nervosa: relapse prevention work must be part of any treatment approach and should include training in which patients learn to recognize early signs of the disorder. Keel and Mitchell also found that recovery rates were approximately 50%, with about 20% of patients still meeting full criteria for diagnosis of bulimia nervosa at follow-up 5 and 10 years after initial evaluation. They also concluded that at 4 years about one-third of the bulimic patients who were considered recovered had relapsed. Mortality rates ranged from 0 to 3%. Hsu (1979) determined that approximately 50% of patients were recovered when reassessed between 2 and 10 years after initial presentation. When these patients were separated into two groups, symptom free and non–symptom free, about 30% in the latter group were appropriately considered “subsyndromal.” Consistent with the Keel and Mitchell review, Hsu concluded that about 20% of bulimic patients continued to maintain the full disorder.

Response to Treatment Pharmacotherapy of Anorexia Nervosa The efficacy of pharmacotherapy for anorexia nervosa has not been extensively studied in controlled trials. This lack of research may reflect clinically disappointing results of medication on the symptoms of anorexia nervosa. Neuroleptics were among the first medications used empirically to treat anorexia nervosa. In particular, low-dose chlorpromazine was chosen, but no double-blind studies have been done. However, two double-blind studies comparing pimozide and sulpiride in anorectic patients (Vandereycken and Pierloot 1982) showed that patients on pimozide evidenced modest improvement in weight gain and weight phobia and demonstrated better motivation for treatment. The fact that traditional antipsychotics have significant side effects and a narrow margin of success makes them an undesirable first choice in treating anorexia nervosa. However, the unrelenting and overwhelming panic of anorectic patients facing weight gain and the profoundly distorted thinking inherent in the illness make antipsychotic agents a compelling intervention ❉

247 ❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

to consider. The atypical neuroleptics have not been studied to date: however, their relatively milder side-effect profile may encourage researchers to test their usefulness. Anorectic patients also often have concomitant depression (25%–80% of eating disorder patients) (Strober and Katz 1988; Swift et al. 1986) or an anxiety disorder (20%–65%) (Fornari et al. 1992; Halmi et al. 1991; Herzog et al. 1993). The presence of these other disorders has guided most of the drug studies, in which positive responses are moderate at best. In addition, 22.9% of binge-purge anorectic patients had a comorbid alcohol dependence or abuse (Braun et al. 1994; Holderness et al. 1994), but there is a lower incidence among restricting type anorectic patients (Herzog et al. 1993). In general, comorbidity does not seem to affect treatment outcome. Fluoxetine and clomipramine have reportedly improved weight gain and reduced depressive symptoms (Crisp et al. 1987; Gwirtsman et al. 1990). Kaye and colleagues (1991) tested the impact of fluoxetine on maintenance of weight gained in an inpatient setting. They found that 29 of 31 patients had maintained their weight above 85% average body weight, with restricting type anorectic patients showing a better response than the bingepurge type. Kaye and colleagues (1998) concluded that fluoxetine significantly reduces the core symptomatology of anorexia nervosa, specifically anxiety, depression, and obsessions and compulsions. Fluoxetine also reduces the rate of relapse: 63% versus 16% without relapse for control subjects. In one study (Lacey and Crisp 1980) using low-dose clomipramine, there was no improvement over placebo. These selective serotonin reuptake inhibitors (SSRIs) do not appear to be as effective in severely underweight or malnourished anorectic patients (Attia et al. 1998) and are best avoided during the low weight phase. Amitriptyline was found to have minimal effect on symptoms of anorexia nervosa (Biederman et al. 1985; Halmi et al. 1986). Open trials of monoamine oxidase inhibitors (MAOIs) and trazodone have shown no benefit (Hudson et al. 1995). Lithium has been shown to have some modest benefit in increasing weight gain (Gross et al. 1981). It should be noted that the risk to anorectic patients, who are prone to dehydration, makes lithium a poor choice. Clonidine (Casper et al. 1986) and metoclopramide, which is an antiemetic agent (Moldofsky et al. 1977), are also ineffective. There is no outstanding traditional antidepressant for treating anorexia nervosa (Walsh 1992). However, cyproheptadine, an antihistamine and 5hydroxytryptamine antagonist, has been found safe and effective in treating the restricting subtype of anorexia nervosa. Patients treated with cyprohep❉

248 ❉

❉

The Difficult-to-Treat Patient With Eating Disorder ❉

tadine had faster rates of weight gain and at doses of 24–32 mg/day had significant improvement in their level of depression. However, the binge-purge anorectic patients showed a negative effect compared with placebo (Halmi et al. 1986).

Pharmacotherapy of Bulimia Nervosa Antidepressants have been widely used for the treatment of bulimia nervosa, which has a 25%–80% comorbidity of depression (Strober and Katz 1988; Swift et al. 1986). Also, antidepressants help reduce binge-eating behavior independent of their effect on mood (Walsh et al. 1991). Most classes of antidepressants have been tested in double-blind, placebo-controlled studies, and the results have been remarkably similar across studies (Walsh 1997). These medications include desipramine, imipramine, amitriptyline, nortriptyline, phenelzine, bupropion, trazodone, and fluoxetine. About 22% of patients become abstinent from bingeing. Goldbloom and Olmsted (1993) studied the effects of fluoxetine on bulimic symptoms and found that it was also effective in reducing core psychological symptoms. The SSRIs seem to be best tolerated by bulimic patients, which makes them a logical first choice. Tricyclic antidepressants (TCAs) have more side effects and are generally less well tolerated. In particular, weight gain associated with the TCAs reduces compliance in bulimic patients. MAOIs are contraindicated because of the required dietary restrictions, which bulimic patients are not likely to follow (especially during binge episodes). Bupropion is also inappropriate because seizures have been reported in a significant number of cases (Horne et al. 1988). Trazodone has been reported to cause delirium in bulimic patients (Damlouji and Fergusen 1984) and thus also carries more risk than the SSRIs. There is evidence to support sequential trials of antidepressants for nonresponders (Agras 1997; Mitchell et al. 1989; Pope and Hudson 1982; Pope et al. 1985). Other considerations in using antidepressants for bulimia nervosa include optimal dose and duration of treatment. Enas and colleagues (1989) found that 60 mg of fluoxetine yielded a therapeutic effect, whereas 20 mg showed no difference compared with placebo. There are no similar data on the other antidepressants. Therefore, dosing usually follows recommendations for treating depression. Studies evaluating the long-term outcome of treatment with antidepressants in patients with bulimia nervosa have shown that the majority of patients relapse (Pyle et al. 1990; Walsh et al. 1991). However, Agras and colleagues (1994) demonstrated that patients treated with desipramine for 24 ❉

249 ❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

weeks had significantly better maintenance at 1 year than those treated with desipramine for 16 weeks. We recommend a 6-month trial at a minimum. Although antidepressants are the most effective and widely used pharmacologic intervention for bulimia, mood-stabilizing agents have also been studied. Lithium was found to have little benefit (Hsu et al. 1991), and its use in this population is risky (dehydration due to purging is predictable among bulimic patients). Studies of carbamazepine (Kaplan et al. 1983) and valproate indicated that only a small number of patients may respond to these medications. Naltrexone, the opioid antagonist used to treat addictive behavior (Mitchell et al. 1989) in higher doses (200–300 mg/day), did have a significant effect on bingeing. Such doses are associated with hepatotoxicity, which calls for extremely judicious use of this medication.

Psychotherapy of Anorexia Nervosa There are few studies of the use of psychotherapy for anorexia nervosa. Cognitive-behavioral therapy (CBT), considered the standard in treatment of bulimia nervosa, has not been studied in anorectic patients. However, behavioral therapy, administered in a controlled inpatient setting, has been shown to effectively induce weight gain (Wulliemier et al. 1975). There are no published findings of behavior therapy on long-term effects, relapse prevention, or improvements in “anorectic thinking.” The effects of behavioral therapy in less restrictive settings have not been evaluated in large studies. Channon and colleagues (1989) used a small sample to compare CBT with a combined behavioral treatment and routine outpatient management. They found no significant differences between groups. Russell and colleagues (1987) did a controlled trial comparing individual supportive psychotherapy to family therapy with 80 patients, including both anorectic patients and low-weight bulimic patients. Patients were evaluated after 1 year. They found that family therapy was more effective than individual therapy for patients who had early onset (under the age of 19 years) with a short duration of the illness. Individual therapy was more effective in inducing weight gain than family therapy for those with a later onset of the illness (over the age of 19 years). Few patients reached a state of full recovery.

Psychotherapy of Bulimia Nervosa There are numerous studies of psychotherapy for bulimia nervosa. The therapy modalities have included cognitive therapy, behavioral therapy, or the ❉

250 ❉

❉

The Difficult-to-Treat Patient With Eating Disorder ❉

combination CBT; psychodynamic therapy; interpersonal therapy; and psychoeducational therapy. CBT is the most effective psychosocial treatment for bulimia nervosa (Fairburn et al. 1993b). In a review of outcome studies using CBT, Craighead and Agras (1991) found 79% of patients had a reduction in purging, and 57% were recovered. Some 30%–40% of the patients that did not improve immediately after the treatment did so 1 year after. Some had even recovered completely at 1-year follow-up. There is high maintenance of change at 6-year follow-up (Fairburn et al. 1995). As CBT has become the gold standard, over 35 studies have compared other psychological treatments with CBT (see Fairburn et al. 1993b; Mitchell and Raymond 1992; Wilfley and Cohen 1997). CBT is clearly superior to other treatments. Agras (1989) also evaluated the efficacy of adding a response prevention component to standard CBT. This involved exposure to feared food with prevention of vomiting during the session. CBT alone was found to be more effective than CBT with added response prevention after treatment and at 6-month follow-up (Agras et al. 1989). Cooper and Steere (1995) found that although immediately posttreatment the CBT and behavioral therapy groups had similar outcomes, at 1-year follow-up the CBT group had maintained their gains but there was significant relapse in the behavioral therapy group. Other studies have replicated this finding. Fairburn and colleagues (1995) found an 86% relapse rate with behavioral therapy compared with 37% with CBT at 6-year follow-up. CBT, when compared with nondirective psychotherapy, clearly has been proven to be superior both posttreatment and at 3- to 6-month follow-up. The finding was the same whether the CBT was given in individual or group format.(Agras et al. 1989; Kirkly et al. 1985; Walsh et al. 1997). When psychodynamically oriented psychotherapy has been compared with CBT (Garner et al. 1993), CBT was more effective than Luborsky’s expressive supportive psychotherapy on eating disorder behavior as well as cognitions, and also in improvement in psychopathology such as depression. Walsh and colleagues (1997) found similar results when CBT was compared with generic short-term supportive psychotherapy in which patients identified and explored the emotional roots of their eating disorder. A group therapy with an educative focus was compared with CBT by Olmstead and colleagues (1991). CBT proved superior except for the 25%– 45% of the sample that was least symptomatic. For these patients the treatments were equally effective, but no long-term follow-up was done. In their studies of a self-help manual, Cooper and colleagues (1994, 1996) found ❉

251 ❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

that 50% of the patients had recovered at 4- to 6-month follow-up, with the frequency of bingeing and purging in the second sample down by 80%. Those who had poor outcomes had greater than 2 times the rate of a previous history of anorexia nervosa than those who had a good outcome. Twothirds of the patients that were followed up at 1 year were completely recovered. Thus, the educative and self-help approaches may have some benefit in certain patients, and further study is needed. Interpersonal therapy is the only other therapy that has rivaled CBT in its efficacy. Fairburn and colleagues (1993a, 1995) have found that interpersonal therapy was equal to CBT in reducing core bulimic symptoms, and at 1-year follow-up the patients fared better with interpersonal therapy and maintained improvement at 6-year follow-up. Further studies are needed to confirm these findings. In addition, Wilfley and Cohen (1997) suggest a sequential approach. One example of this is to have patients participate in a course of CBT and, if they are unsuccessful, to move on to interpersonal therapy. Based on our experience such an approach would likely be beneficial, but further data are needed.

Relative Efficacy of Pharmacotherapy Versus Psychotherapy Because of the lack of effective pharmacological treatments for anorexia nervosa, studies comparing pharmacotherapy with psychotherapy in the treatment of anorexia nervosa have not been done. However, in bulimia, studies comparing the treatment of bulimia nervosa with psychotherapy versus medication suggest that medication offers little over the treatment with CBT. Mitchell and colleagues (1990) found group CBT superior to imipramine in reducing bingeing and purging, and the addition of imipramine to the CBT group proved of no further benefit. Agras and colleagues (1992) found that individual CBT alone was as effective as desipramine or the combination of medication and CBT at 16 weeks of treatment. However, at 32 weeks, the combined treatment had the best outcome. In a 1-year follow-up study, Agras and colleagues (1994) found that the group that fared the worst in follow-up was the group that had desipramine only for 16 weeks. The combined treatment group had the most gains. Leitenberg and colleagues (1994) replicated the finding that with a shorter treatment interval, in this case 20 weeks, desipramine combined with CBT did not improve the outcome over CBT alone. Walsh and colleagues (1997) used a more complicated medication regimen using desipramine followed by fluoxetine in cases where the de❉

252 ❉

❉

The Difficult-to-Treat Patient With Eating Disorder ❉

sipramine was ineffective or not tolerated. They compared this medication regimen alone to CBT alone and to the combination of the two. They found the medication-only group to be as successful as the CBT group. The combination was most effective. They also added psychodynamic psychotherapy to the medication regimen and found that the outcome of this group was similar to medication only. In a recent study of fluoxetine and CBT, Goldbloom and colleagues (1997) also found that the combination of fluoxetine and CBT was superior to medication alone, but in this study they found that psychotherapy alone was as effective as the combination. Thus study results to this point support that medication, although sometimes improving outcome, provides only modest benefit over CBT alone. Further investigation is needed to clarify the issue.

Summary of Treatment Recommendations: Anorexia Nervosa Anorexia nervosa typically follows a chronic course and can take many years from the time of onset to near recovery. Clearly effective treatment methods have not yet been established. Two broad treatment categories, inpatient and outpatient treatment, should be considered. There are some indications for inpatient treatment that are absolute. In most instances, however, clinical judgment will be required to make that determination. Outpatients should be weighed weekly to monitor weight, and periodic CBC and electrolytes should be taken to monitor medical status. Amylase should be followed in those who purge, as elevated levels can signal increases in vomiting. Medical necessity must always dictate treatment decisions. For example, a weight of 20% or less than ideal body weight, potassium less than 2.5, heart rate less than 50, fainting episodes, dehydration, or prolonged QT intervals measured by electroencephalogram require immediate medical attention. Failure to maintain a minimally safe weight should signal the need for inpatient stabilization. A contract should be set up at the outset whereby it is clear to both the patient and the clinician when hospitalization will be considered medically necessary. If a patient for whom hospitalization is medically indicated is unable to be convinced and he or she is in immediate medical danger, then the patient may meet criteria for commitment to the hospital against his or her will. When a patient’s medical status is reasonably stable, the following should be considered in making treatment recommendations regarding the appropriate level of care: severity of behavioral and cognitive symptoms, history of appropriate treatment and response to treatment, duration of the disorder, and sources of motivation to change. ❉

253 ❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

Anorectic patients typically present in a psychiatric setting at the insistence of others. This circumstance must be acknowledged immediately. At the start of treatment in either setting, the patient’s attitude toward therapy will always be ambivalent (if not entirely rejecting). Nevertheless, patients should be encouraged to join in a positive alliance with their treatment team. The most obvious and universal source of anorectic patients’ treatment resistance is their awareness that treatment will, if successful, interfere with their allegiance to thinness. In fact, the first and ongoing goal of treatment must be nutritional rehabilitation at a rate that is responsive to the patient’s medical, psychological, and environmental necessities. Efforts made to engage the patient in his or her own treatment may help the patient develop a somewhat positive alliance. One useful technique is to ask the patient to first list the benefits and then list costs to him or her in maintaining his or her anorectic disorder. In being asked to list these benefits, the patient is validated in his or her experience of the anorexia nervosa as important in his or her life. The patient’s list of the costs in maintaining the illness provides a foundation of personal reasons from which questions can be raised about the desirability of this disorder. This same list can also enable the patient to see the value of treatment. Although some patients may claim that there are no costs in having anorexia nervosa, most will admit to some negative experiences that they know are the result of the illness. Typical examples include poor concentration and decreased energy. Anorexia nervosa is best treated using a team approach. Although a psychiatrist can mange both the medical and psychological aspects of the treatment, it is often useful to have the treatment split between a medical professional—either an internist, pediatrician, or psychiatrist—and a psychotherapist—either a psychologist, psychiatrist, or social worker supervised by a psychiatrist. With this arrangement, the treatment alliance can sometimes be better maintained in the therapy, as a team member other than the primary therapist is responsible for making the ultimate decision on medical necessity for hospitalization, a decision the patient may resist. A nutritionist should also be involved, as nutritional recommendations for the gaining process are advisable. Although patients with anorexia nervosa may appear to have extensive knowledge regarding nutrition, often they are filled with misconceptions, such as how many calories it takes to gain a pound or how much fat is needed in the diet to stay healthy. In the inpatient setting, the team should also include social workers and nursing staff. In both settings family therapy or parent counseling may also be useful, particularly to those patients living with their parents and in patients with early onset ❉

254 ❉

❉

The Difficult-to-Treat Patient With Eating Disorder ❉

younger than age 19 years and recent onset in the past 3 years (Russell et al. 1987). To help patients gain weight, behavioral strategies are usually necessary. Specific minimal goals should be set for weight gain along with contingency planning to support success. Contingency planning is setting up concrete consequences if the patient does not make his or her weight requirement. For instance, a patient may lose his or her privilege to participate in a favorite sport if the patient is not gaining a pound per week. Contingencies are needed as the fear of weight gain is so strong in these patients that a consequence more upsetting than gaining weight is often the only way to achieve progress. (More specific information about behavioral strategies for weight gain can be found in Halmi 1984.) Weight gain is considered a necessary but insufficient condition for good outcome. The “anorectic mindset” must be treated as well. The timing of this endeavor is dependent on the patient’s cognitive capabilities. Generally, it seems that the lower the weight, the less access the patient will have to higher level thought processes. When cognitively capable, CBT techniques can be introduced. Although there are no definitive data that indicate CBT is preferred over other structured therapies in anorexia nervosa, there is reason to expect this given the success of CBT with bulimia nervosa. Anorexia nervosa and bulimia nervosa share many of the same cognitions and behaviors that CBT addresses in the treatment of bulimia nervosa. The patients are asked to identify assumptions regarding food, shape, and weight, and challenge these ideas rather than automatically accepting them. The validity of their beliefs about shape and weight are examined using objective data that support or refute the thoughts. The patients are also asked to begin to make changes in their anorectic behaviors. These would include increasing caloric intake and types of foods eaten and may also include other eating-disordered behaviors such as repetitive weighing, measuring of wrist size, or excessive exercise. The reader is referred to Kleifield and colleagues’ (1996) summary of CBT for anorexia nervosa for more detail on this treatment approach. If a patient is not responding to these interventions, more intensive treatment, such as inpatient hospitalization may be necessary. Hospitalization in a specialized eating-disorder program will allow for the potential roadblocks to treatment to be addressed. The patient is removed from his or her social environment, and these interpersonal dynamics can be evaluated and family therapy attempted while the patient is removed and interactions are less intense. The structure of the inpatient unit, where the patient is ❉

255 ❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

monitored constantly for eating-disordered behavior, effectively extinguishes the majority of these behaviors. The patient’s intake is closely monitored, and through a comprehensive behavioral program he or she is encouraged to eat increasing amounts and varieties of food. Specific food phobias are addressed much easier in this setting where the patient has 24-hour structure and support to help cope with the anxiety of gaining weight or giving up eating-disordered behaviors. Weight gain can be accomplished more rapidly than in the outpatient setting as the patient can be closely monitored for medical complications of more rapid refeeding. The patient can be assessed constantly to ascertain if pharmacotherapy is needed, and, if so, he or she can be monitored closely for side effects. Partial hospitalization, although offering more intense treatment and monitoring, loses the advantage of inpatient treatment in that it cannot offer 24-hour support and monitoring. However, it is a useful step to try with some patients who are resisting inpatient treatment or who are in transition to outpatient treatment after hospitalization. Cyproheptadine, the only medication that has been proven helpful in the treatment of anorexia nervosa, may be a useful adjunct to psychotherapy. The reader is referred to the earlier discussion of the details of using cyproheptadine in this population. When anorexia nervosa occurs with a comorbid psychiatric illness, that illness should be treated as well. Special care must be taken in diagnosing the comorbid diagnosis. Emaciation in itself reproduces many of the symptoms of depression, such as insomnia, irritability, indecision, and dysphoria. At a better weight, such symptoms will either recede or remain, and thereby demonstrate their source. For this reason, as well as the likelihood that low weight increases problems with side effects, it is recommended that psychotropic medications be introduced at healthier weights.

Summary of Treatment Recommendations: Bulimia Nervosa The bulimic patient is somewhat more amenable to treatment than the patient with anorexia nervosa. Bulimic patients overvalue thinness, as do anorectic patients, but are unable to maintain a very low weight. These patients are typically distressed by their binges, terrified that the binges will make them fat, that they are out of control, and if extremely symptomatic, experience deterioration in their functioning. For these reasons bulimic patients are more likely to seek treatment on their own and to look for help in changing their binge behavior. ❉

256 ❉

❉

The Difficult-to-Treat Patient With Eating Disorder ❉

Medical complications associated with bulimia nervosa are usually less drastic than with anorexia nervosa. Medical crises can arise from extreme cases of self-induced vomiting, ipecac abuse, and laxative abuse. Electrolyte imbalances can occur, including life-threatening hypokalemia. In normalweight bulimic patients, weekly weight monitoring is not a necessity but may provide useful data for the patient as he or she normalizes eating behavior. The patient can see that his or her weight remains essentially the same, even when the patient is eating well. Contingency planning can be useful for the bulimic patient as well. Weight cannot be used as the measure of progress, so the goal may be that the patient must maintain normal electrolytes and amylase. One mistake that some doctors make is to give the vomiting patient potassium supplements. This colludes with the patient’s illness and his or her denial of its severity, allowing the patient to continue to practice dangerous behaviors and depriving the doctor of an objective measure of the amount of vomiting occurring. CBT is the treatment of choice for bulimia nervosa. The manual by Fairburn and colleagues (1993a) is the basis of most CBT treatments. Its aim is to interrupt the cycle of restricting/bingeing/purging so that the patient can learn and use cognitive techniques to establish new and healthier responses to emotional triggers (and to eliminate triggers that arise from restricting). CBT is often successful when applied in an outpatient setting (Fairburn et al. 1993a). If a bulimic patient is medically compromised or if he or she has failed to respond to appropriate outpatient treatment, hospitalization should be recommended. In the hospital, the patient should be prevented from continuing the habit cycle of restricting/bingeing/purging while engaging in normal eating behavior. This practice can give the patient a better chance to respond to CBT outside a hospital setting. As described earlier, medications have not been shown to increase the effectiveness of CBT. Therefore medications should be reserved for those that do not respond to CBT alone or who have comorbid psychiatric disorders. For treating bulimia nervosa or concurrent depression, antidepressants have consistently been shown as the most effective medications. SSRIs are the least likely to cause weight gain or other side effects particularly intolerable to the eating-disorder patient, such as dry mouth or constipation, and thus are the first-line choice. For patients who do not respond to CBT, a trial of interpersonal therapy is advised (Fairburn et al. 1993b). Adjunctive therapies can include pharmacotherapy, nutritional counseling, and family therapy.

❉

257 ❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

Factors Associated With Outcome Prognostic Factors in Anorexia Nervosa Studies of prognostic factors in anorexia nervosa have reported somewhat contradictory results. Poor outcome was predicted by older age at onset (Steinhausen and Glanville 1983), distorted body image (Morgan and Russell 1975), bingeing and purging (Eckert et al. 1995), the length of time the patient was ill before referral for treatment (Hsu et al. 1979; Steinhausen et al. 1991; Theander 1970), and poor familial relationships (Hsu et al. 1979; Morgan and Russell 1975; Morgan et al. 1983; Ratnasuriya et al. 1991; Theander 1970). Wonderlich and Mitchell (1997) report that the presence of comorbidity, specifically depression, anxiety, or personality disorders, has little impact on treatment outcome.

Prognostic Factors in Bulimia Prognostic factors in bulimia nervosa have not been extensively researched. Comorbid Axis II diagnoses are reported to have significant association with outcome in bulimia nervosa (Keel and Mitchell 1997). Specific personality traits, such as impaired impulse control (typically seen in the borderline personality disorders), have been found to correlate with poor outcome (Brotman et al. 1988; Fahy et al. 1993; Fairburn et al. 1993a; Nash and Colborn 1994). There was no association between outcome and symptom severity, age at onset, duration of illness, comorbid depression or anxiety (Wonderlich and Mitchell 1997), prior history of anorexia nervosa (Abraham et al. 1983; Collings and King 1994; Fallon et al. 1991; Hsu 1980; Rossiter et al. 1993; Walsh et al. 1991), or history of comorbid substance abuse, although no studies have been completed looking at current substance abuse (Wonderlich and Mitchell 1997).

Biological Factors Researchers have been equally interested in the influence of biological factors that predispose individuals to the eating disorders and that may also affect outcome. Familial studies suggest that genetics contribute to the development of eating disorders. Twin studies have shown that between 55% and 65% of monozygotic twins are concordant for anorexia nervosa. Sisters of patients with anorexia nervosa have a 7% rate of development of anorexia ❉

258 ❉

❉

The Difficult-to-Treat Patient With Eating Disorder ❉

nervosa (Holland et al. 1988). In addition, the families of patients with anorexia nervosa have been shown to have high rates of anxiety and mood disorders (Halmi et al. 1991). Physiological factors are discussed in detail elsewhere (Heebink and Halmi 1994). Potential biological markers include disordered energy regulation and metabolism, abnormalities of insulin metabolism, elevations of corticotrophin-releasing hormone, abnormalities in the central dopamine system, and abnormalities of serotonin, norepinephrine, and opioids (Heebink and Halmi 1994). The most interesting and conclusive findings thus far come from the investigation of aberrant hunger and satiety in anorexia nervosa and bulimia nervosa. Halmi and Sunday (1991) compared eating disordered patients with healthy control subjects. They found that both patient groups had abnormalities in the degree of hunger and satiety experienced during and after a meal. Other investigators have found some specific differences in satiety responses of eating-disordered patients (Garfinkel et al. 1979; Heatherington and Rolls 1991; Pirke et al. 1994). For example, preloading with certain foods affected later food choices differently in anorectic patients compared with control subjects. Healthy subjects reduced intake of cottage cheese following a preloading of this same food. Anorectic patients did not reduce their intake of cottage cheese after a preload of cottage cheese, although in preloading with cheese and crackers, they did reduce their intake of this food. This seems to imply that the anorectic patient’s cognitions toward food can override his or her satiety impulses. This has important implications for treatment, in that even if more specific biological interventions can be fashioned, it may be extrapolated on the basis of this data that the cognitions also need to be addressed to make significant progress.

Factors Related to Refractory Illness Experts in the field of eating disorders refer to a wide range of phenomena that may contribute to the development of anorexia nervosa and bulimia nervosa. These phenomena include biological, psychosocial, and cultural factors. Other areas that may influence outcome are the type of therapy, skill of therapist in applying that therapy, and “fit” between a particular patient and a particular therapist. Therapists treating patients with anorexia nervosa or bulimia nervosa have encountered those patients who seem obsessively driven downward by a potentially lethal illness. No interventions seem to alter this course and ❉

259 ❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

disaster seems, and sometimes is, inevitable. However some of such patients who seem destined to have a fatal outcome do not. They may have almost died from complications of their eating disorder, but as they regain their health, they apparently grasp a connection to life that enables them to fight the eating-disorder impulses. They learn and use tools to help themselves in that fight and eventually get much better. These patients have counterparts who have also faced death because of their eating disorder, but this group of patients resume the deadly eating-disorder behaviors as soon as opportunity permits. How these groups differ is a compelling mystery. The intractable eating disorders puzzle and frustrate the public as well as the specialists. These are the cases that generate wide speculation, incredulity, and fear. Frequent targets of blame are “dysfunctional families”; severe limitations of character, intellect, and resourcefulness; traumatic childhood; and genetic and/or biological defect. One or many of these may have contributed to a particular patient developing the illness; however, until treatment is better informed by advances in the study of etiology, practicality and need must guide the clinician. In that respect, treatment is one of the factors presumed to influence success or failure that can be modified (even if it cannot perform magic). This is in contrast with genetic defects, childhood events, and character. Problems with treatment of eating-disordered patients fall into two broad categories: experience of the treatment team and type of therapy. Inexperience in treating eating-disordered patients can result in a failed treatment and/or a sicker patient. For example, the underweight patient is desperate to avoid weight gain. Toward that end, he or she will lie and cheat. The patient can be ingenious in finding ways to appear heavier, such as wearing multiple layers of clothing. He or she can also be ingenious in finding ways to falsely elevate his or her weight when the patient gets weighed by the doctor or nurse. The normal-weight bulimic, also terrified by the prospect of giving up his or her unhealthy methods of controlling his or her weight, may similarly exaggerate compliance with treatment recommendations or minimize his or her active symptoms. Inexperienced clinicians are more easily fooled. They are often reluctant to consider that the patient is capable of deception. In fact, this is one of the most common problems encountered in training clinicians to treat this population. It seems difficult for beginning clinicians who are committed to helping patients to mistrust the truthfulness of their patients. The consequences of conducting a “false” treatment are obvious. The patient stays ill longer, perhaps deteriorates, and will have to spend more time and money on a rigorous therapy if he or she is going to recover. ❉

260 ❉

❉

The Difficult-to-Treat Patient With Eating Disorder ❉

Psychiatrist inexperience with medicating patients with eating disorders affects the effectiveness of treatment as well. Medication noncompliance, although a problem with all psychiatric patients, can be especially tricky with eating-disordered patients. Those who fear that the medication may cause them to gain weight, or lose their obsessive edge, may refuse recommended medication or become noncompliant. Another problem is related to the act of purging. If a patient is purging regularly, although the patient is taking his or her medications, he or she may be vomiting them up before complete digestion. This frequently is the cause of nonresponse to medication—the patient is not absorbing enough of the medication for it to have an effect. In addition, many clinicians give medications to patients that are severely emaciated. Pharmacotherapy tends to be ineffective when the patient is significantly underweight, and many depressive symptoms will resolve with weight gain. Underrecognition of comorbid disorders can also delay improvement. Comorbid substance abuse is relatively common in bulimic patients and thus should be asked about thoroughly and on several occasions. Often, even when the substance use does not constitute abuse, it contributes to exacerbation of the eating-disorder symptoms. The choice of treatment can interfere with a patient’s progress toward recovery. For example, an anorectic patient who is not psychologically able to take advantage of a group therapy might be placed in a group in which he or she becomes competitive with the other anorectic patients in the group and lose weight in an attempt to be the “best” anorectic. The use of psychodynamically based therapy to treat eating disorders continues to be a problem. Understanding and exploring underlying psychodynamic issues, without concomitant behavior change, does not lead to resolution of the eating disorder. Overall, anorectic and bulimic patients will resist changing their behavior unless the behavior problem is directly addressed. Some eating-disordered patients regress in insight-oriented therapy, with increased rumination, confusion, and exacerbation of behavioral symptoms. Unfortunately, some clinicians will continue with a treatment approach because they know it well, without regard for its impact on the patient. Anorectic and bulimic patients have symptoms that can be objectively measured. In conjunction with monitoring weight, electrolytes, amylase, and so forth, reports from those who live with the patient (or spend a lot of time with him or her) can often confirm or contest the perceived progress. If the patient shows no measurable improvement in any of his or her symptoms, the validity of the treatment must be questioned. ❉

261 ❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

Options for Refractory Patients Alternative methods can be considered when patients have failed at appropriate treatments. Contingencies attached to behavioral goals may need to be more unusual or aggressive. For example, one therapist whose patient was deeply attached to her finally set up a contingency such that the patient had to meet weight expectations to see the therapist each week. Wilfley and Cohen (1997) emphasize the need for new “alternative treatment strategies” for CBT-resistant patients. Possibilities would include interpersonal therapy or a combination of sequenced treatments. Such treatments are currently under investigation at several sites around the country. For patients with bulimia nervosa or anorexia nervosa who are not responding to once weekly CBT treatment, one option is to increase the frequency of the treatment to twice or three times per week. In addition, group and family therapy, outside treatment of comorbid diagnoses such as Alcoholics Anonymous, or therapy specifically targeting borderline personality disorder such as Linehan’s Dialectical Behavior Therapy may be added to increase the intensity of the intervention. If the increased frequency of the CBT is ineffective, a partial hospital or intensive outpatient program can be added to the individual treatment. Atypical medication choices should be considered when the usual drugs have failed. Very low doses of neuroleptics, such as chlorpromazine, may enable a panicked, anxiety-ridden patient to calm down enough to understand what his or her treatment team is doing. The newer antipsychotics, such as risperidone and olanzapine, have not been used extensively, but their side effect profiles are better than the older drugs in this class, and they may be good choices for the types of patients described at this point. We have found olanzapine especially helpful in this regard, and it has the added benefit of inducing weight gain. This medication seems to be well tolerated by this population aside from the frequent side effect of sedation. Clearly there is a need for controlled trials using this medication with anorexia nervosa. Other options include raising the level of the antidepressant. Occasionally some cases of anorexia nervosa are so severe that they are resistant to any of the more traditional forms of treatment, leaving the patient at chronic risk of death. Uncontrolled studies of electroconvulsive therapy showed some effectiveness in symptom reduction in such cases (Laboucarie et al. 1966). Residential placement may be needed for some patients. Until recent years, inpatient treatment centers for eating disorders could offer long-term ❉

262 ❉

❉

The Difficult-to-Treat Patient With Eating Disorder ❉

care, which was often needed for the particularly refractive patient. In this setting, patients were carefully monitored until free of behavioral symptoms. During their stay they were nutritionally rehabilitated and then taught to maintain their normal weight and eating habits with less and less supervision. Since the onset of managed care, most patients cannot receive the kind of inpatient care just described. If they are approved for hospital treatment, it is typically so brief that they have to leave before they have even reached the point of normal weight with normal eating. In other words, they leave the hospital before they are ready to enter the “maintenance phase.” In such cases, it is not uncommon for the patient to relapse soon after discharge. It is sometimes possible to find intermediate settings or specialized eating-disorder treatment centers to refer eating-disordered inpatients who have had frequent relapses, are not ready to take responsibility for their recovery, or whose home environment is unable to support treatment. Residential facilities for children and halfway houses for ill adults can provide a daily structure, easy access to support staff, and careful monitoring. Clinical Vignette: Anorexia Nervosa Ms. G is a 25-year-old married woman with a diagnosis of anorexia nervosa, restricting type, and obsessive-compulsive personality disorder. Prior to hospitalization, she was in a psychodynamic, twice-weekly outpatient psychotherapy for 1 year, primarily for her problem with obsessive thinking. She was also given clomipramine to treat this problem. Her obsessive-compulsive disorder began shortly after she married her longtime boyfriend from college. She reported a prior history of some obsessional thinking that had not interfered with her functioning. After she married, the obsessive thinking increased until she had to quit her job because she could not focus on her work. Coincident with this exacerbation was a severe weight loss. At 5′8″ she normally weighed 140 lbs. Just prior to her wedding, her weight had dropped to 120 lbs. According to Ms. G, no one expressed concern that her weight loss might be indicative of an eating disorder or some other medical problem. Apparently, most of the people in her life thought she was just “nervous” about the wedding. When the patient agreed to be seen for her eating disorder problem, her weight had dropped to 110 lbs. She admitted that she was always concerned about her weight and shape. This concern increased when she left college and began working in the fashion industry in a major metropolitan area. She was surrounded by people who were constantly dieting and were intensely concerned about image and appearance. It seemed that success in her industry required, at the least, a fashionable body type, style, and manner. It was quite natural for her to join in this orientation. Ms. G’s anorexia nervosa was well established by the time she had

❉

263 ❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

been married for 6 months. It was during this time that her weight had dropped to 110 lbs. Her husband and other family members expressed serious concerns about her health and her treatment. They convinced her to consult an expert in “weight and food problems.” During the brief period between arranging the consultation and being admitted to the hospital she rapidly lost another 10 lbs. When the patient was brought to the hospital she was rageful toward her husband and abusive to other family members. Typical though this may be of emaciated patients, it was painful for the family to be with her yet equally painful for them to leave her at the hospital. Experienced hospital staff offered comforting educational information to the family members about the irrational nature of anorexia nervosa and the effects of the malnutrition on her attitude and behavior. The patient’s initial treatment consisted of refeeding with liquid supplements 6 times a day and no medications. She continued with severe agitation and resistance and was refusing the supplements often enough to prevent adequate weight gain. A trial of risperidone and sertraline were started and were titrated as tolerated. Once the medication began to take effect she was able to comply with the program and began to gain weight. She achieved a weight of 120 lbs. at the point of discharge and then began a cognitive-behavioral–oriented outpatient psychotherapy. She continued to gain weight until she stabilized at her preanorectic weight of 140 lbs. As she gained weight she also reported that her obsessive thoughts decreased significantly. At this level of improvement it seemed reasonable to taper her off the medications. She has remained stable without them for over 1 year.

Clinical Vignette: Bulimia Nervosa Ms. A is a 15-year-old young woman with a diagnosis of bulimia nervosa. Her disorder began 2 years ago following the divorce of her parents. She reported that she began to feel overweight not long before her parents separated. Her weight at that time was 130 lbs., which was at the higher end of the range for her height 5′4″. This weight was typical of her weight-toheight ratio for many years. The patient vividly described the first time she tried purging by sticking her finger down her throat to “get rid of” the pizza she had just eaten. She was amazed that it was so “easy.” She began to induce vomiting a couple times a week to eliminate high fat foods that her mother had prepared. At the same time she began to notice a slight weight loss. Delighted by the weight loss, she induced vomiting more frequently, expecting that this would enable her to lose weight more quickly. She also began to limit her intake of food, which quickly led to intense food cravings. By this time, Ms. A had followed the steps that inevitably lead to bulimic behavior. She began to binge eat and then vomit daily. From the point that she started to occasionally induce vomiting, within 3 months she was bingeing and vomiting multiple times a day. Now, quite symptomatic with bulimia nervosa, Ms. A was totally

❉

264 ❉

❉

The Difficult-to-Treat Patient With Eating Disorder ❉

caught up in the process. Her fleeting worries about damage to her health were easily dismissed. During this period her mother walked in on Ms. A making herself vomit. She described the absence of shame or even distress, but instead felt some satisfaction that her parents would be concerned about her rather than their divorce. Ms. A’s mother insisted that she receive psychotherapy for this problem. The therapist was skilled in treating eating-disordered patients with CBT. Even though the patient was appropriately weighed and had her blood work followed, she succeeded in lying about her symptoms. It was not until she began to vomit blood that she panicked and admitted to a far more extensive problem than the therapist believed. She was hospitalized almost immediately on an intensive CBT-oriented eating-disorder unit. The treatment included structure and supervision of her meals, use of the bathroom, and free time. She was forced to stop purging. She was so concerned about gaining weight that she stopped eating. Because of her restriction the clinical staff inserted a nasal gastric tube for 24 hours to replenish her liquids and begin feeding her. At that point the patient reported realizing that she would have to eat. She continued to believe that she could outsmart the staff and her parents, and began to eat small amounts of food at each meal. At this time she began to take fluoxetine, increased to 60 mg, for some dysphoria but mainly for her reports that she had severe cravings to binge, which made it impossible to eat at all. Because of her irregular eating habits and her bingeing prior to admission, her weight had actually increased to 145 lbs. Once she had successfully eaten and not purged for 10 days she was transferred to an intensive partial hospital program and individual psychotherapy with a CBT focus and continued medication therapy. She actually seemed to improve more with the individual attention of the twice-aweek therapy only after returning to her binge/purge cycle immediately following discharge. She began to admit to the purging and then to use the CBT techniques to reduce the purging. She kept food records and thought records regularly and later in the treatment began to challenge herself by eating small quantities of the high fat “fear foods.” She has not required rehospitalization.

Clinical Wisdom and Summary Treatment of anorexia nervosa and bulimia nervosa poses a clear challenge to even the most experienced clinician. There are a few basic elements of the treatment that we have found to be successful. First is being clear that eating-disordered patients are often terrified of giving up their illness, and thus their honesty must always be in question. Many times an anorectic patient will binge and purge but be elusive about his or her symptoms until well into the treatment. Those who present for treatment require a different amount of effort and treatment style than those who are forced into treatment by par❉

265 ❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

ents and loved ones. However, both will progress best in treatment if the clinician can be empathetic to the patients’ fear of weight gain while still carefully monitoring her behavior. Second, with the recent changes in availability and length of inpatient hospitalization, it is important to realize that outpatient treatment is more challenging than ever. Treatment may need to continue for several years before a “cure” is achieved. Many times the difficult-to-treat patient with eating disorder will require multiple hospitalizations before he or she realizes the cost of the illness upon his or her life. In our experience the patient may benefit from multiple treatments at the same facility. Because eating disorder patients tend to be intellectualized and controlling by nature, they may “doctor shop,” switching from one physician to another in an attempt to avoid actual treatment. Alternatively, they may engage a clinician in tantalizing dialogue for months or years, without changing their behaviors. Overall, therapists should remember that Theander’s (1985) 30-year follow-up study showed 75% of the patients were functioning in the good category after 30 years. Although progress may be slow, eventually significant improvement is possible for most patients. Never give up on the eatingdisordered patient.

References Abraham SF, Mira M, Llewellyn-Jones D: Bulimia: a study of outcome. Int J Eat Disord 2:175–180, 1983 Agras WS: Pharmacotherapy of bulimia nervosa and binge-eating disorder: longer term outcomes. Psychopharmacol Bull 33:433–436, 1997 Agras WS, Schneider JA, Arnow B, et al: Cognitive behavioral and response prevention treatments for bulimia nervosa. J Consult Clin Psychol 57:215–221, 1989 Agras WS, Rossiter EM, Arnow B, et al: Pharmacologic and cognitive-behavioral treatment for bulimia nervosa: a controlled comparision. Am J Psychiatry 149:82– 87, 1992 Agras WS, Elise MR, Arnow B, et al: One year follow up of psychosocial and pharmacologic treatments for bulimia nerovsa. J Clin Psychiatry 55:179–183, 1994 American Psychiatric Association: Diagnostic and Statistical Manual of Mental Disorders, 4th Edition. Washington, DC, American Psychiatric Association, 1994 Andersen A, Holman J: Males with eating disorders: challenges for treatment and research. Psychopharm Bull 33:391–397, 1997 Attia E, Haiman C, Walsh T, et al: Does fluoxetine augment the inpatient treatment of anorexia nervosa? Am J Psychiatry 155:546–551, 1998

❉

266 ❉

❉

The Difficult-to-Treat Patient With Eating Disorder ❉

Biederman J, Herzog DL, Rivinus TM, et al: Amitriptyline in the treatment of anorexia nervosa: a double blind, placebo controlled study. J Clin Pscyhopharmacol 5:10– 13, 1985 Braun D, Sunday S, Halmi K: Psychiatric comorbidity in patients with eating disorders. Psychol Med 24:859–867, 1994 Brotman AW, Herzog DB, Hamburg P: Long-term course in 14 bulimic patients treated with psychotherapy. J Clin Psychiatry 49:157–160, 1988 Casper RC, Sxhlemmer RG, Javaid JI: A placebo controlled crossover study of oral clonidine in acute anorexia nervosa. Psychiatry Res 20:249–260, 1986 Channon S, De Silva P, Helmsley D, et al: A controlled trial of cognitive behavioral and behavioral treatment of anorexia nervosa. Behav Res Ther 27:529–536, 1989 Collings S, King M: Ten year follow up of 50 patients with bulimia nervosa. Br J Psychiatry 164:80–87, 1994 Cooper PJ, Coker S, Fleming C: Self help for bulimia nervosa: a preliminary report. Int J Eat Disord 16:401–404, 1994 Cooper PJ, Steere J: A comparison of two psychological treatments for bulimia nervosa: implications for models of maintenance. Behav Res Ther 33:875–886, 1995 Cooper PJ, Coker S, Fleming C: An evaluation of the efficacy of supervised cognitive behavioral self help in bulimia nervosa. J Psychosom Res 40:281–287, 1996 Craighead LW, Agras WS: Mechanisms of action in cognitive-behavioral and pharmacologic interventions for obesity and bulimia nervosa. J Consult Clin Psychol 59:115–125, 1991 Crisp AH, Lacey JH, Crutchfield M: Clomipramine and “drive” in people with anorexia nervosa: an inpatient study. Br J Psychiatry 150:355–358, 1987 Damlouji NF, Fergusen JM: Trazodone induced delirium in bulimic patients. Am J Psychiatry 141:434–436, 1984 Eagles T, Johnston M, Hunto D, et al: Increasing incidence of anorexia nervosa in the female population of northeast Scotland. Am J Psychiatry 152:1266–1271, 1995 Eckert ED, Halmi KA, Marchi P, et al: Ten year follow up of anorexia nervosa: clinical course and outcome. Psychol Med 25:143–156, 1995 Enas GG, Pope HG, Levine LR: Fluoxetine in bulimia nervosa: a double blind study, in New Research Program and Abstracts, American Psychiatric Association 142nd Annual Meeting, San Francisco, CA, May 6–11, 1989. Washington, DC, American Psychiatric Association, p 204 Fahy TA, Eisler I, Russell GF: A placebo controlled trial of d-fenfluramine in bulimia nervosa. Br J Psychiatry 162:597–603, 1993 Fairburn CG, Beglin SJ: Studies of the epidemiology of bulimia nervosa. Am J Psychiatry 147:401–408, 1990 Fairburn CG, Marcus MD, Wilson GT: Cognitive behavioral therapy for binge eating and bulimia nervosa: a comprehensive treatment manual, in Binge Eating: Nature, Assessment, and Treatment. Edited by Fairburn CG, Wilson GT. New York, Guilford, 1993a, pp 361–404

❉

267 ❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

Fairburn CG, Jones R, Peveler RC, et al: Psychotherapy and bulimia nervosa: the longer term effects of interpersonal therapy, behavioral therapy, and cognitive behavioral therapy. Arch Gen Psychiatry 50:419–428, 1993b Fairburn CG, Norman PA, Welch SL, et al: A prospective study of outcome in bulimia nervosa and the long term effects of three psychological treatments. Arch Gen Psychiatry 52:304–312, 1995 Fallon BA, Walsh T, Sadik C, et al: Outcome and clinical course in inpatient bulimic women: a 2–9 year follow up study. J Clin Psychiatry 52:272–278, 1991 Fornari V, Kaplan M, Sandberg D, et al: Depressive and anxiety disorders in anorexia nervosa and bulimia nervosa. Int J Eat Disord 12:21–29, 1992 Garfinkel PE, Moldofsky H, Garner DM: The stability of perceptual disturbances in anorexia nervosa. Psychol Med 9:703–708, 1979 Garner DM, Rockert W, Davis R, et al: Comparison of cognitive behavioral therapy and supportive expressive therapy for bulimia nervosa. Am J Psychiatry 150:37– 46, 1993 Goldbloom DS, Olmsted MP: Pharmacotherapy of bulimia nervosa with fluoxetine: assessment of clinically significant attitudinal change. Am J Psychiatry 150:770– 774, 1993 Goldbloom DS, Olmsted M, Davis R, et al: A randomized controlled trial of fluoxetine and cognitive behavioral therapy for bulimia nervosa: short term outcome. Behav Res Ther 35:803–811, 1997 Gross HA, Ebert MH, Faden VB, et al: A double blind placebo controlled study of lithium carbonate in primary anorexia nervosa. J Clin Psychopharmacol 1:378– 381, 1981 Gwirtsman HE, Guze BH, Yager J, et al: Fluoxetine treatment of anorexia nervosa: an open trial. J Clin Psychiatry 51:378–382, 1990 Halmi KA: Behavioral management for anorexia, in Handbook of Psychotherapy for Anorexia and Bulimia, 5th Edition. Edited by Garner DM, Garfunkel PE. New York, Guilford, 1984, pp 147–159 Halmi KA, Sunday SR: Temporal patterns of hunger and satiety ratings and related cognitions in anorexia and bulimia. Appetite 16:214–223, 1991 Halmi KA, Eckert E, LaDu TJ, et al: Anorexia nervosa: treatment efficacy of cyproheptadine and amitriptyline. Arch Gen Psychiatry 43:177–181, 1986 Halmi KA, Eckert E, Marchi P, et al: Comorbidity of psychiatric diagnoses in anorexia nervosa. Arch Gen Psychiatry 48:712–718, 1991 Heatherington M, Rolls BJ: Eating behavior in eating disorders: response to preloads. Physiol Behav 50:101–108, 1991 Heebink D, Halmi KA: Biological markers in eating disorders. Annual Review of Psychiatry 13:227–251, 1994 Herzog D, Keller M, Sacks N, et al: Psychiatric comorbidity in treatment seeking anorexics and bulimics. J Am Acad Child Adolesc Psychiatry 32:835–842, 1993

❉

268 ❉

❉

The Difficult-to-Treat Patient With Eating Disorder ❉

Herzog W, Schellberg D, Deter HC: First recovery in anorexia nervosa patients in the long-term course: a discrete time survival analysis. J Consult Clin Psychol 65:169–177, 1997 Hoek HW: Review of the epidemiological studies of eating disorders. International Review of Psychiatry 5:61–74, 1993 Holderness C, Brooks-Gunn J, Warren M: Comorbidity of eating disorders and substance abuse. Review of the literature. Int J Eat Disord 16:1–35, 1994 Holland AJ, Sicolte N, Treasure J: AN: evidence for a genetic basis. J Psychosom Res 32:561–571, 1988 Horne RL, Ferguson JM, Pope HG, et al: Treatment of bulimia with buproprion: a multicenter controlled trial. J Clin Psychiatry 49:262–264, 1988 Hsu LKG: Outcome of anorexia nervosa: a review of the literature (1954–1978). Arch Gen Psychiatry 37:1041–1046, 1980 Hsu LKG, Crisp AH, Harding B: Outcome of anorexia nervosa. Lancet 1:61–65, 1979 Hsu LKG, Clement L, Santhouse R: Treatment of bulimia nervosa with lithium carbonate: a controlled study. J Nerv Ment Dis 179:351–353, 1991 Hudson JI, Pope HG, Jonas JM, et al: Treatment of anorexia nervosa with antidepressants. J Clin Psychopharmacol 5:17–23, 1995 Jones DJ, Fox MM, Barigian HM, et al: The epidemiology of anorexia nervosa. Pychosom Med 42:551–558, 1980 Kaplan AS, Garfinkel PE, Darby PL, et al: Carbemazepine in the treatment of bulimia. Am J Psychiatry 140:1225–1226, 1983 Kaye W, Weltzin T, Hsu J: An open trial of fluoxetine in patients with anorexia nervosa. J Clin Psychiatry 52:464–471, 1991 Kaye W, Gendall K, Strober M: Serotonin neuronal function and selective serotonin reuptake inhibitor treatment in anorexia and bulimia nervosa. Biol Psychiatry 44:825–838, 1998 Keel PK, Mitchell JE: Outcome in bulimia nervosa. Am J Psychiatry 154:3–8, 1997 Kirkly BG, Schneider JA, Agras WS, et al: Comparison of two group treatments for bulimia. J Consult Clin Psychol 53:43–48, 1985 Kleifield EI, Wagner S, Halmi KA: Cognitive-behavioral treatment of anorexia nervosa. Med Clin North Am 19:715–734, 1996 Laboucarie J, Rascol A, Karkous E, et al: Anorexia nervosa: results of a clinical and therapeutic study of 173 cases. Revue de Medicine de Toulouse 2:193–210, 1966 Lacey JH, Crisp AH: Hunger, food intake and weight: the impact of clomipramine on a refeeding anorexia nervosa population. Postgrad Med J 1 (suppl 56):79–85, 1980 Leitenberg H, Rossen JC, Wolf J, et al: Comparison of cognitive behavior therapy and desipramine in the treatment of bulimia nervosa. Behav Res Ther 32:37–45, 1994

❉

269 ❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

Mitchell JE, Raymond NC: Cognitive-behavioral therapy in treatment of bulimia nervosa, in The Psychobiology and Treatment of Anorexia Nervosa and Bulimia Nervosa. Edited by Halmi K. Washington, DC, American Psychiatric Press, 1992, pp 307–327 Mitchell JE, Christenson G, Jennings J, et al: A placebo controlled double blind crossover study of naltrexone hydrochloride in outpatients with normal weight bulimia. J Clin Pharmacol 9:94–97, 1989 Mitchell JE, Pyle RL, Eckert ED, et al: A comparison study of antidepressants and structured intensive group psychotherapy in the treatment of bulimia nervosa. Arch Gen Psychiatry 47:149–157, 1990 Moldofsky H, Jeuniewic N, Garfinkel PE: Preliminary report on metoclopramisde in anorexia nervosa, in Anorexia Nervosa. Edited by Vigersky RA. New York, Raven, 1977, pp 373–376 Morgan HG, Russell GFM: Value of family background and clinical features as predictors of long-term outcome in anorexia nervosa: four year follow-up study of 41 patients. Psychol Med 5:355–371, 1975 Morgan HG, Purgold J, Welbourne J: Management and outcome in anorexia nervosa: a standardized prognostic study. Br J Psychiatry 143:282–287, 1983 Nash ES, Colborn AL: Outcome of hospitalized anorexics and bulimics in Cape Town, 1979–1989. S Afr Med J 84:74–79, 1994 Olmsted MP, Davis R, Garner DM, et al: Efficacy of a brief group psychoeducational intervention for bulimia nervosa. Behav Res Ther 29:71–84, 1991 Pirke KM, Kellner MB, Frief E, et al: Satiety and clolecystokinin. Int J Eat Disord 15:63–69, 1994 Pope HG Jr, Hudson JI: Treatment of bulimia with antidepressants. Psychopharmacology 78:176–179, 1982 Pope HG, Hudson JI, Jonas JM, et al: Antidepressant treatment of bulimia: a two year follow up study. J Clin Psychopharmacol 5:320–327, 1985 Pyle RL, Mitchell JE, Eckert ED, et al: Maintenance treatment and 6-month outcome for bulimic patients who respond to initial treatment. Am J Psychiatry 147:871– 875, 1990 Ratnasuriya RH, Eisler I, Szmukler GI, et al: Anorexia nervosa: outcome and prognostic factors after 20 years. Br J Psychiatry 158:495–502, 1991 Rossiter EM, Agras WS, Telch CF, et al: Cluster B personality disorder characteristics predict outcome in treatment of bulimia nervosa. Int J Eat Disord 13:349–357, 1993 Russell GF: The prognosis of eating disorders: a clinician’s approach, in The Course of Eating Disorders: Long-Term Follow-Up Studies of Anorexia and Bulimia Nervosa. Edited by Herzog W, Deter HC, Vandereycken W. Berlin, Germany, Springer, 1992, pp 198–213 Russell GFM, Szmukler GI, Dare C, et al: An evaluation of family therapy in anorexia nervosa and bulimia nervosa. Arch Gen Psychiatry 44:1047–1056, 1987

❉

270 ❉

❉

The Difficult-to-Treat Patient With Eating Disorder ❉

Russell GM: BN: an ominous variant of anorexia nervosa. Psychol Med 9:392–448, 1979 Steinhausen HC, Glanville K: Follow up studies of anorexia nervosa: a review of research findings (editorial). Psychol Med 13:239–249, 1983 Steinhausen HC, Rauss-Mason C, Seidel R: Follow up studies of anorexia nervosa: a review of four decades of outcome research. Psychol Med 21:447–454, 1991 Strober M, Katz J: Depression in the eating disorders: a review and analysis of descriptive, family, and biological findings, in Diagnostic Issues in AN and BN. Edited by Garner D, Garfinkel P. New York, Brunner/Mazel, 1988, pp 54–63 Swift WJ, Andrews D, Barklage N: The relationship between affective disorder and eating disorders: a review of the literature. Am J Psychiatry 143:290–299, 1986 Theander S: Anorexia nervosa: a psychiatric investigation of 94 female cases. Acta Psychiatr Scand 214 (suppl):1–194, 1970 Theander S: Outcome and prognosis in anorexia nervosa and bulimia: some results of previous investigations, compared with those of a Swedish long-term study. J Psychiatr Res 19:493–508, 1985 Vandereyecken W, Pierloot R: Pimozide combined with behavioral therapy in the short term treatment of anorexia nervosa: a double blind, placebo controlled crossover study. Acta Psychiatr Scand 66:445–451, 1982 Walsh BT: Pharmacological treatment, in The Pychobiology and Treatment of Anorexia Nervosa and Bulimia Nervosa. Edited by Halmi K. Washington, DC, American Psychiatric Press, 1992, pp 329–340 Walsh BT, Hadigan CM, Devlin MJ, et al: Long term outcome of antidepressant treatment for bulimia nervosa. Am J Psychiatry 148:1206–1212, 1991 Walsh BT, Wilson GT, Loeb KL, et al: Medication and psychotherapy in the treatment of bulimia nervosa. Am J Psychiatry 154:523–531, 1997 Wilfley D, Cohen L: Psychological treatment of bulimia nervosa and binge-eating disorder. Psychopharm Bull 33:437–454, 1997 Wonderlich S, Mitchell J: Eating disorders and comorbidity: empirical, conceptual and clinical implications. Psychopharm Bull 33:381–390, 1997 Wulliemier F, Rossel F, Sinclair K: La therapie comortementale de l’anorexie nerveuse [Behavior therapy in anorexia nervosa]. J Psychosom Res 19:267–272, 1975

❉

271 ❉

This page intentionally left blank

10 The Difficult-to-Treat Patient With Substance Abuse Mark Albanese, M.D. Edward Khantzian, M.D.

Introduction Recent studies indicate that patients with psychiatric diagnoses have substance use disorders more frequently than people in the population as a whole; likewise, individuals with substance use disorders suffer from psychiatric illness at rates higher than the general population (Kessler et al. 1997; Regier et al. 1990). These patients with comorbid disorders are commonly referred to as dual-diagnosis patients. In this chapter, we describe our approach to the assessment and treatment of these especially challenging patients. To underscore our belief that psychiatrists must treat these disorders in an integrated fashion, we discuss both patients who present initially with psychiatric symptoms and those who come to caregiver attention first with substance abuse problems. We begin by addressing two conceptual issues. ❉

273 ❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

Ideologies The clinician’s ideology may become problematic when working with dualdiagnosis patients and collaborating with substance-abuse professionals. Alcohol counselors, for example, might tend to see all depression in their clients as long-term withdrawal from alcohol, and antidepressant treatment as “drug substitution.” Whereas a depressive syndrome can be part of early abstinence from alcohol, one must entertain the possibility that there is a cooccurring depressive disorder that also must be treated.

Defining Dual The term dual diagnosis can be confusing and even controversial. To some, the term refers to patients with co-occurring medical and mental illnesses; to others, it refers to mentally retarded patients with a concomitant mental illness. In addition, to some, the co-occurring mental illness in a substance abuser must be a “serious” Axis I disorder before the patient qualifies as having a dual diagnosis. Some would insist that the psychiatric and substance use disorders must be independent. Others would maintain that the substance use disorder causes the psychiatric difficulties. Finally, others would insist that the psychiatric disorder leads to substance abuse. Thus, it is tempting to get caught up in the “primary/secondary disorder” or “cause/ consequence” controversy. A common-sense approach is to use a broad definition, enveloping all patients with both a diagnosable (i.e., DSM-IV-TR) substance abuse/dependence disorder and any diagnosable non–substance use disorder (Axis I and II). As we will see, for example, substance use is neither negligible nor unimportant in patients with Axis II disorders such as antisocial personality disorder and borderline personality disorder. Likewise, it makes sense to treat both the psychiatric disorder and substance use disorder, without spending too much time trying to ascertain which came first, which is primary, or which may have caused the other. There are data supporting both possibilities: substance use disorder precedes mental illness, and mental illness precedes substance use disorder (e.g., Christie et al. 1988; Khantzian 1997; Vaillant 1993). As we will see, the presence of one illness adversely affects the course of the other, whichever came first. For example, we found that even minor depressive symptoms resulted in early termination of substance abuse treatment (Albanese et al. 1997). Similarly, Wilens and colleagues (1998) found that attention-deficit/hyperactivity disorder (ADHD) is associated with longer duration of substance use ❉

274 ❉

❉

The Difficult-to-Treat Patient With Substance Abuse ❉

disorder and slower remission rate. Furthermore, as we will also see, more aggressive treatment of one disorder improves outcomes in the other. Finally, even when it is clear that one disorder precedes another, removing the precedent disorder does not necessarily result in resolution of the second disorder. For example, Brady and colleagues (1998) found significant posttraumatic stress disorder (PTSD) prevalence among a group of primary cocaine users. In this group, the trauma was generally associated with the procurement and use of cocaine. Whereas one might expect the PTSD to be exacerbated by continued cocaine misuse, one would not reasonably expect the PTSD to resolve merely with remission of cocaine use. A final introductory note: the term dual probably underestimates many of our patients’ problems, because many of them have more than two diagnoses. For example, it is not uncommon to have both anxiety and depression in addition to alcohol dependence, or bipolar disorder and both alcohol and cocaine dependence. Furthermore, many dual-diagnosis patients have concomitant Axis III disorders, either directly related to substance use or to selfcare deficits stemming from mental illness and substance abuse. The National Comorbidity Survey (NCS) reported that 14% of the population had three or more lifetime psychiatric diagnoses (Kessler et al. 1994).

Epidemiology The DSM-IV-TR (American Psychiatric Association 2000) reveals a long list of substance-related disorders, including acute intoxication and withdrawal syndromes for both alcohol and other drugs. Included in this list are the more long-term dependence and abuse disorders. DSM-IV-TR defines the main features of substance dependence as “a cluster of cognitive, behavioral, and physiological symptoms indicating that the individual continues use of the substance despite significant substance-related problems” (p. 192). Substance abuse is “a maladaptive pattern of substance use manifested by recurrent and significant adverse consequences related to the repeated use of substances” (p. 198). We will focus on these substance-related disorders. Because, in clinical practice, abuse and dependence are frequently not distinguishable, we will consider them together as substance use disorders. Analysis of the Epidemiologic Catchment Area (ECA; Regier et al. 1988, 1990) study reveals that while the lifetime prevalence of a mental disorder in the general population is 32.2%, among individuals who present with alcohol abuse or dependence it is 36.6%. Among those with nonalcohol drug abuse or dependence it is 53.1%. The comparison is even more striking ❉

275 ❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

when one recalls that the general population prevalence for mental disorders includes substance use disorders. Likewise, the NCS (Kessler et al. 1994, 1997) has shown that the lifetime prevalence of a mental disorder in the general population is 48%. In contrast, the NCS found the lifetime prevalence of a mental disorder among alcohol-abusing men to be 56.8%, among alcohol-dependent men to be 78.3%, among alcohol-abusing women to be 72.4%, and among alcohol-dependent women to be 86.0%. Similarly, among individuals who present with a psychiatric diagnosis, rates of comorbid substance use disorder are higher than for the general population. For example, the ECA study found that schizophrenic individuals have a lifetime substance use disorder prevalence of 47%. The corresponding rate of substance use disorders among bipolar individuals is 56.1% and 83.6% among those with antisocial personality disorder. Other studies have shown a high prevalence of substance use disorders among individuals with PTSD (Brady 1997; Kessler et al. 1995).

Implications of Comorbidity The psychiatric and substance abuse comorbidity has serious implications. Studies have shown that psychiatric illness results in decreased response to substance abuse treatment and a worse outcome. In fact, the degree of a patient’s psychiatric impairment predicts outcome of drug abuse treatment better than any other pretreatment factor, including severity of substance abuse. Likewise, the presence of a comorbid substance use disorder has a negative effect on appropriate social behavior and also predicts poor outcome in patients with chronic mental illness (Rosenthal et al. 1992).

Schizophrenia Substance-abusing schizophrenic patients are frequent consumers of psychiatric and emergency services (Bartels et al. 1993). Additionally, substanceabusing schizophrenic patients have an increased rate of psychiatric rehospitalization (Drake et al. 1989). This higher readmission rate persists even in the presence of medication compliance (Gupta et al. 1996; Swofford et al. 1996). Whereas substance use disorder is associated with medication noncompliance (Owen et al. 1996), substance abuse disorders contribute to acute schizophrenic exacerbation even when patients are medication compliant (Lieberman et al. 1987). Substance abuse is associated with earlier onset of schizophrenia (Kovasznay et al. 1997) and use of higher doses of ❉

276 ❉

❉

The Difficult-to-Treat Patient With Substance Abuse ❉

antipsychotics (D’Mello et al. 1995). Substance abuse is also associated with poor antipsychotic treatment response, perhaps because the substances alter mesolimbic dopaminergic systems (Bowers et al. 1990). Substance use may also accelerate the appearance of tardive dyskinesia (Bailey et al. 1997; Zaretsky et al. 1993). In an interesting study from Finland, Rasanen and colleagues (1998) followed a birth cohort of 11,017 subjects for 26 years, collecting data on psychiatric disorders and crimes. Men with schizophrenia and alcohol abuse were 25.2 times more likely to commit a violent crime than men with no psychiatric disorder. The risk for schizophrenic patients without alcoholism was 3.6. None of the patients with schizophrenia who did not abuse alcohol were recidivists (>2 offenses), whereas the risk of recidivism increased 9.5fold for schizophrenic subjects abusing alcohol.

Bipolar Disorder and Unipolar Depression Substance abuse comorbidity predicts poor outcome in bipolar disorder (Black et al. 1988). For example, there are higher rates of mixed and rapidcycling mania in bipolar patients with alcohol disorder, and these patients are more resistant to lithium treatment and take longer to recover (Keller et al. 1986; Post et al. 1991). Also, bipolar patients with a substance use disorder have earlier onset of mood disorder, greater likelihood of an additional comorbid Axis I disorder, higher anxiety scores, and more psychiatric hospitalizations (Sonne et al. 1994; Young et al. 1993). Substance use disorder is also associated with both treatment noncompliance and delayed onset of remission in patients with affective psychosis (Strakowski et al. 1998). Of note, Winokur and colleagues (1995) found in one research cohort that as bipolar patients attained effective mood stabilization, active alcohol abuse diminished with time. Interestingly, whereas this study suggests that alcoholism associated with bipolar disorder is often a secondary complication, our field is clearly too young to discern which of these patients merely need mood stabilization and which need more aggressive substance use disorder treatment. Thus, the safer course is to err on the side of treating both disorders. One recent study (Greenfield et al. 1998) showed that a diagnosis of major depression at entry into inpatient alcohol treatment predicted shorter times to first drink and relapse. Similarly, we found that elevated depression scores at time of admission to a substance abuse treatment facility predicted early treatment termination (Albanese et al. 1997). Hasin and colleagues ❉

277 ❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

(1996) found that improvement in major depression increased the chances of remission and reduced the chances of alcoholism relapse in patients with both disorders. Similarly, Loosen and colleagues (1990) followed 29 alcoholic men over 2 years and found that length of abstinence was approximately 50% shorter in those with a history of depression. Suicidality is a serious complication of having comorbid psychiatric and substance use disorders. Clearly, substance abuse and psychiatric illness each individually increase the risk of suicidal behavior (Berglund 1984; Rich et al. 1988). In dually diagnosed individuals, the risk is greater. For example, suicide rate in major affective disorder is higher in the presence of comorbid alcoholism (Fawcett 1988), and alcoholics with affective lability, such as depression or dysphoria, also have higher risk of suicide (Berglund 1984).

Assessment and Diagnosis If one of the disorders in a dually diagnosed patient is not diagnosed, it goes untreated. As noted earlier, an untreated disorder has adverse consequences for the identified, comorbid disorder. As with any medical diagnosis, the first step is a complete and accurate assessment. In patients with an identified psychiatric illness, the diagnosis of a comorbid substance use disorder can be difficult. For example, studies have found underreporting of substance use disorder in schizophrenic patients (Shaner et al. 1993; Stone et al. 1993). A recent report by Weiss and colleagues (1998b), however, found that in nonpsychotic patients with either PTSD or bipolar disorder, self-reports of substance use were highly valid. In fact, in many cases a negative urine screen accompanied a positive report. Where self-reports are not reliable—for example, in the acute-care setting or in patients with psychotic disorders—brief screening tests and structured interviews enhance the ability to detect substance abuse problems in psychiatric patients. For example, the Structured Clinical Interview for DSM-III-R (SCID; Spitzer and Williams 1986) has been used reliably in the dual-diagnosis population (Albanese et al. 1994a; Kosten et al. 1991). Other instruments have also been used (Breakey et al. 1998; Rosenberg et al. 1998). Furthermore, studies have found that when other measures for detecting substance use are combined with a structured interview, detection is further enhanced. For example, Fernandez-Pol and colleagues (1988) used both a nonstandardized interview and urine toxicology analyses to detect substance abuse in consecutive psychiatric admissions. Similarly, Ananth ❉

278 ❉

❉

The Difficult-to-Treat Patient With Substance Abuse ❉

and colleagues (1989) used the Diagnostic Interview Schedule (DIS; Robins et al. 1981) and urine toxicology screens to determine the prevalence of substance use disorder in psychiatric inpatients. Albanese and colleagues (1994a) used the SCID, urine toxicology analyses, and admission and discharge assessments to detect substance use disorders in patients admitted to a psychiatric facility. This study found the SCID to be the most accurate measure. Urine toxic screens are of limited value in making lifetime diagnoses, and, although clearly suggestive of use, do not necessarily indicate abuse or dependence. In addition, as in the Weiss study noted previously, the majority of patients with a diagnosis of current alcohol or drug disorder had negative urine screens. Finally, toxic screens can yield false positive results, either because of the assay error or because of other substances (e.g., nonsteroidal anti-inflammatory drugs) mimicking substances of abuse. Nevertheless, Albanese and colleagues (1994a) found that the urine screens were highly sensitive and specific in confirming drug use diagnoses in the confusing subpopulation of patients who had been assigned the diagnosis of either “psychosis not otherwise specified” or “no Axis I disorder.” This raises the complex issue of acute presentations in psychiatric patients. A clinician should be suspicious of a substance-induced psychiatric presentation under certain circumstances: 1. The first occurrence of psychiatric symptoms 2. An abrupt onset of symptoms 3. No family history of psychiatric illness (conversely, a positive family history of psychiatric disorder is suggestive of the presence of a primary psychiatric diagnosis) 4. An unusual presentation in a known psychiatric patient 5. Resolution of symptoms with time 6. Presence of physiological signs such as autonomic instability and pupillary constriction or dilation In patients who present with a substance use disorder, a complete psychiatric assessment should not overlook certain diagnostic issues. First is “organicity.” The dual-diagnosis population is more prone to head trauma (e.g., due to motor vehicle accidents, fights) as well as to direct toxic effects of substances on the central nervous system. And because they do not care for themselves medically or nutritionally, dual-diagnosis patients are more prone to underlying medical illnesses that might mimic primary psychiatric ❉

279 ❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

illnesses (e.g., vitamin deficiencies, thyroid dysfunction). Therefore, it is important to ask about head trauma, seizure history, history of medical illness, and current medications. This history may lead to further diagnostic studies, such as biochemistry profile, brain magnetic resonance imaging, and electroencephalogram. Another important area is psychological trauma, which, as noted earlier, is more prevalent among substance abusers. This higher rate of comorbidity is probably in part due to the dangerous situations in which substance abusers place themselves and to early exposure to substance-abusing households, with physically abusive parents. A patient suffering from PTSD might initially present as depressed, irritable, and/or emotionally labile. Hypomania/mania is another important area to ask about. We have treated many patients who presented as depressed, and were even taking antidepressants, who after careful questioning were noted to have had a manic episode at some point in their lives (Albanese 2001). These patients should probably be started on a mood stabilizer first to avoid the risk of antidepressant-induced mania. We and others (Calabrese and Delucchi 1990) have reported that the mood stabilizer frequently is effective for the depression as well as the mania. Akiskal and colleagues (Akiskal 1996; Akiskal and Mallya 1987) have discussed the bipolar spectrum concept, whereas McElroy and colleagues (1995) have described the mixed mania spectrum. We have found these concepts especially pertinent in the dually diagnosed population, in which patients frequently present with symptom clusters consistent with bipolar II disorder and other variants of classical bipolar disorder (Albanese et al. 1999). Psychosis is usually obvious. Occasionally, however, it can go undetected. For example, one patient that we assessed in a detoxification unit initially appeared to suffer from a paraphilia. After careful assessment, however, it became clear that he had a long-standing belief, persisting during prolonged periods of abstinence, that he could communicate his feelings for women telepathically. Thus, he was treated with an antipsychotic rather than a selective serotonin reuptake inhibitor (SSRI), with resolution of the psychotic symptoms. Finally, when assessing a substance-abusing patient who presents with psychiatric symptoms, it is important to ask about periods of abstinence and the persistence of the symptoms during those periods. For example, does the depression or mood lability persist even during lengthy abstinence? One of our patients was in residential substance abuse treatment for 2 years, during ❉

280 ❉

❉

The Difficult-to-Treat Patient With Substance Abuse ❉

which he experienced a manic episode that switched to depression; he described it as “a period of euphoria that ended with suicide time.”

Treatment The remainder of this chapter will focus on treatment issues in treatmentrefractory patients with comorbid psychiatric and substance use disorders.

Treatment Models In the past, mental health professionals frequently underdiagnosed substance use disorders, whereas substance abuse counselors missed psychiatric diagnoses (Green 1996; Miller 1994; Ridgely et al. 1990). When the symptoms of another diagnosis were recognized, they were often disregarded. For example, addiction professionals may have believed the patient’s mental illness to be a manifestation of substance use, whereas mental health professionals might have viewed the patient’s substance abuse as a complication of mental illness (Sciacca 1991). Treatment consisted of treating one disorder, with the expectation that the other disorder would resolve as a result. This was rarely the case. Thus, dually diagnosed patients were not provided effective treatment. Those who did receive treatment for both disorders did not receive it at the same time, in the same place, or by the same clinicians (Drake et al. 1996). Most facilities were not expert in treating both sets of difficulties, so, for example, an alcoholic manic patient would undergo detoxification, stabilization on psychiatric medication, then referral to another service or facility for alcoholism treatment. A patient might be told, “We’ll help you with your depression when you’ve been sober for 6 months,” or “We’ll be happy to enroll you in our alcoholism counseling program when you have had 6 months free of suicide attempts and are off medication” (SAMHSA 1996). This model has been termed serial treatment. The next treatment model has been termed parallel treatment, in which two agencies work with the patient at the same time, each treating one disorder. Theoretically, the agencies exchange information about the patient, but experience shows that such information exchange is often inadequate. Another problem develops when one agency proposes treatment that contradicts or is incompatible with the treatment of the other agency. For example, the psychiatrist in the mental health agency might prescribe medication that is prohibited for patients in the substance abuse program (SAMHSA 1996). ❉

281 ❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

The general trend for treatment is toward an integrated model, in which treatment occurs not only concurrently, but also in the same location and by the same staff, with competency in treating both sets of difficulties. Experts agree on the advantages of such a treatment approach (Drake et al. 1991; Kline et al. 1991; Minkoff 1991; Osher and Kofoed 1989; Ridgely et al. 1990). Integrated treatment has produced positive outcomes (Drake et al. 1995). In a prospective study of 47 patients with schizophrenia and substance use disorders, those randomly assigned to integrated outpatient treatment were more likely to develop and maintain a connection to the program than those assigned to nonintegrated treatment. Integrated treatment was associated with fewer hospitalizations, increased sobriety, and decreased psychiatric symptoms (Hellerstein et al. 1995). Integrated treatment entails provision of services according to biopsychosocial principles. For example, McLellan and colleagues (1993) controlled the dose of biological treatment (methadone) by maintaining 92 study participants (male intravenous opiate users in methadone maintenance treatment) on 60–90 mg/day. They were then randomized to three levels of psychosocial treatment for a 6-month trial: 1) methadone alone (minimum methadone services; MMS), 2) methadone plus counseling (standard methadone services; SMS), and 3) methadone plus counseling and onsite medical/psychiatric, employment, and family therapy (enhanced methadone services; EMS). Whereas MMS patients exhibited reductions in opiate use, 69% of them met criteria for protective transfer to SMS (i.e., eight consecutive weekly urines positive for heroin or cocaine or three or more medical/psychiatric emergencies). This was significantly different from the 41% of SMS subjects and 19% of EMS subjects who met these criteria. Endof-treatment outcomes revealed that the SMS group showed significantly more and larger improvements than did the MMS group; the EMS group showed significantly better outcomes than did the SMS group. MMS subjects who had been protectively transferred to SMS showed significant reductions in opiate and cocaine use within 4 weeks. In summary, then, this study and others underscore the notion that combining adequate biological and psychosocial interventions results in better treatment outcomes for dually diagnosed patients. Interestingly, in an article describing the efficacy of clozapine in substance-abusing schizophrenic patients, Buckley (1998) proposed that the therapeutic efficacy of the medication is derived not only from its biological profile but also from the totality of the treatment regimen. The mandatory blood monitoring visits, required because of the risk of agranulocytosis, ❉

282 ❉

❉

The Difficult-to-Treat Patient With Substance Abuse ❉

have the additional benefits of forming a therapeutic alliance between the patient and health care providers and offering an important psychotherapeutic context to the pharmacotherapy. Recently, Ries and colleagues (2000) compared their dual-diagnosis schizophrenic patients treated with integrated inpatient services with their schizophrenic patients without substance dependence. Whereas the dualdiagnosis patients were more dangerous on admission, they actually exhibited shorter hospital stay, greater symptomatic improvement, and no increase in 18-month readmission rate. This may reflect the advantages of integrated treatment. Similarly, Weiss and colleagues (2000) have treated substance-dependent bipolar patients with integrated group therapy (IGT) and found that IGT patients had better outcomes on several measures, including percentage of months abstinent. Many integrated dual-diagnosis programs, both inpatient and outpatient, have been described (Drake et al. 1991; Kline et al. 1991; Mowbray et al. 1995; Osher and Kofoed 1989; Rosenthal et al. 1992; Sciacca 1991). Our state hospital’s inpatient dual-diagnosis treatment track is similar to these programs. It is comprised primarily of bipolar, borderline, and schizophrenic patients with a variety of substance use disorders, mostly alcohol, cannabis, and cocaine. Patients are stabilized pharmacologically by a psychiatrist working with a multidisciplinary treatment team. Patients also participate in a group program run by a multidisciplinary staff who have acquired expertise in treating both psychiatric and substance use disorders. The wide variety of groups includes traditional 12-step, psychotherapy, relapse prevention, and psychoeducation, which especially deals with the effects that each disorder has on the other. In addition, we also offer 12-step groups especially geared toward dually diagnosed clients. As patients get closer to discharge, they frequently transition to an outpatient dual-diagnosis treatment program funded by the state. Some patients are discharged to a dual-diagnosis residential facility. One especially helpful group experience is modified dynamic group therapy (Albanese and Khantzian, in press; Khantzian et al. 1990). These heterogenous groups are typically comprised of schizophrenic patients with negative symptoms and verbal but severely impaired borderline patients. The latter group are frequently quite active in drawing out the former. The group is rooted in Khantzian’s self-medication hypothesis of the addictions, which maintains that substance abuse is a self-regulation disorder of affect, self-care, self-esteem, and relationships. Much of the content of the group is devoted to discussion of the interplay between the disorders as they occur in ❉

283 ❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

each member. In essence, each patient, because of the co-occurrence of the two (or more) disorders, has a double dose of self-regulation deficits. The interplay between the disorders can be more devastating than either disorder alone. For example, relationships can be thorny, stress-laden undertakings for anybody. If a person suffers from either borderline personality disorder or cocaine dependence, a relationship can be especially difficult. In persons with both disorders, the undertaking can be a life-or-death struggle, with the danger that the patient will relapse to not one but at least two disorders. To compound the difficulties, the cocaine-dependent borderline patient inevitably attempts to negotiate a relationship with a person showing similar comorbidity. Of note, we also employ a variety of other psychosocial interventions especially with our more difficult patients, such as those with comorbid schizophrenia and cocaine dependence. As has been described elsewhere (Satel 1995; Shaner et al. 1995), because these patients may use entitlement benefits to purchase street drugs, we frequently apply for either a representative payee or a guardian of estate who can competently organize the patient’s finances. Short of this, we have also set up specialty groups to help patients with budgets and other financial issues.

Pharmacotherapy: General Principles Little has been written about pharmacological treatment of patients with dual disorders. One reason is that substance use disorder is usually an exclusion criterion in psychopharmacological studies. Another reason is patient ambivalence. These patients, with two or more stigmatizing illnesses, frequently convince themselves that they do not have an illness by not taking prescribed medication. We are reminded of a common refrain from our substance-abusing patients: “I’m not crazy, doc.” Conversely, for some patients, the grass is greener in another diagnostic category. For example, there is the alcoholic patient who believes that the “real” (only) problem is underlying depression, which he is “self-medicating with alcohol”; or the schizophrenic patient who is convinced that the real problem is her cocaine use, which is causing her psychotic symptoms. In addition to patient ambivalence is clinician ambivalence, which comes in a variety of forms. For example, some feel that treating substance-abusing patients with medications is an enabling behavior, in which the clinician is exchanging one chemical dependence for another. In our experience, such beliefs are usually unfounded. Before we look at specifics of psychopharmacological treatment, there ❉

284 ❉

❉

The Difficult-to-Treat Patient With Substance Abuse ❉

are a few more general points. First, in addition to starting psychiatric medications in substance abusers, we have also had the experience of tapering patients off medications without adverse consequences. Typically, patients may have been started on a medication (e.g., antipsychotic) during detoxification. When they get beyond the acute withdrawal phase, the medication may no longer be necessary. Second, patients with a substance use disorder appear more sensitive to medication side effects. Frequently, the side effects resemble the uncomfortable feelings they may have experienced during intoxication with, or withdrawal from, a substance of abuse. For example, the stimulating side effect of an antidepressant may be reminiscent of the psychomotor agitation or anxiety that emerges during alcohol withdrawal. Along these lines, Weiss and colleagues (1998a) recently reported that side effects were the most common reason for lithium noncompliance in substance-abusing patients with bipolar disorder. Because of this, we start our patients on especially low doses, titrate very gradually, and sometimes aim for a lower-than-usual maintenance dose. On the other hand, we frequently have found it necessary to use more than one medication in especially refractory dual-diagnosis patients (Albanese, in press; Albanese et al. 1998, 2000). Finally, given the adverse consequences of psychiatric and substance abuse comorbidity (e.g., suicide and violence), and given the relative safety of most psychiatric medications, the safer course is often to err on the side of more liberal use of appropriate psychiatric medication in this population.

Pharmacotherapy: Specific Disorders Pharmacotherapy for substance use disorders can be divided into several categories: 1) agonists that substitute for a substance during detoxification or maintenance (e.g., methadone or nicotine replacement), 2) antagonists (e.g., naltrexone for opioid users) or partial antagonists (e.g., buprenorphine), 3) aversive agents (e.g., disulfiram), 4) anticraving agents (e.g., naltrexone in alcoholics), 5) anti–drug-seeking agents (other than anticraving), 6) agents for comorbid medical problems, 7) agents for comorbid psychiatric problems, and 8) agents targeting both the substance abuse and comorbid psychiatric problem (Wilkins 1997). The focus of this section will be on the last two categories while emphasizing agents for comorbid psychiatric disorders in treatment-refractory patients. Depression Treatment of depression in substance abusers has been controversial. Some have argued that these depressive symptoms are a consequence of substance ❉

285 ❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

abuse and will resolve with time. Clearly, alcohol can cause symptoms of anxiety and depression. And, in many patients, these symptoms do resolve spontaneously during abstinence; yet, as we have seen, there is an increased prevalence of comorbid depression in people with substance use disorders. These patients do not exhibit the same temporal resolution of depressive symptoms (Brown et al. 1995), and their depression responds to antidepressant treatment. Fluoxetine has recently been studied by Cornelius and colleagues (1997), who followed 51 depressed alcoholic patients for 12 weeks after a 2to 3-day detoxification and a subsequent 1-week washout period. Patients were then randomized in a double-blind fashion to either fluoxetine or placebo. A statistically significant improvement in depressive symptoms was seen in the group treated with fluoxetine versus those on placebo. Interestingly, these investigators also found that over the course of the trial total alcohol consumption was significantly lower in the fluoxetine group than in the placebo group. Tricyclic antidepressants have also been used effectively. Imipramine was prescribed initially in an open trial of 60 depressed alcoholic patients, and resulted in 45% showing improvement in both mood and drinking behavior. An additional 13% responded after further dosage increases or treatment with disulfiram. In a subsequent 6-month, randomized discontinuation trial, 4 of 13 subjects (31%) relapsed during imipramine treatment, whereas 7 of 10 (70%) relapsed while taking placebo (Nunes et al. 1993). Subsequently, McGrath and colleagues (1996) randomized 69 alcoholic patients with depression to either imipramine or placebo after a 1-week washout period and followed the patients for 12 weeks. They found a 52% response rate (much or very much improved in both depression and alcohol ratings) in the imipramine group versus 21% in those on placebo. Although there was no overall effect on drinking outcome, patients whose mood improved showed decreased alcohol consumption that was more marked in those treated with imipramine. Mason and colleagues (1996) randomized 71 alcohol-dependent depressed subjects who had been abstinent a median of 8 days to either desipramine or placebo and followed them for 6 months. Depressive symptoms decreased significantly more in those treated with desipramine than those on placebo, and patients were abstinent significantly longer when receiving desipramine. Imipramine has also been used in opiate-dependent patients with depression (Nunes et al. 1998). In that study, 137 depressed patients receiving methadone maintenance treatment were randomized to a 12-week, double❉

286 ❉

❉

The Difficult-to-Treat Patient With Substance Abuse ❉

blind, placebo-controlled trial of imipramine. A favorable response was defined as a Clinical Global Impression scale score for depression of “much” or “very much improved” and at least a 75% reduction in self-reported drug or alcohol use or abstinence. Of the 84 patients who completed an adequate trial (at least 6 weeks), 42 received imipramine. Twenty-four (57%) were rated as responders, compared with 3 of 42 (7%) receiving placebo. Whereas imipramine was superior to placebo on some self-reported measures of substance use and craving, and mood improvement was associated with improvement in self-reported substance use, this effect was less robust than mood effect. Few patients achieved urine-confirmed abstinence. As mentioned earlier, we have observed that some patients find certain antidepressant side effects especially uncomfortable. For example, sedating antidepressants can result in early morning sedation that is reminiscent of a hangover. Conversely, especially activating antidepressants might cause nervousness that mimics withdrawal. In an unpublished review, we found sertraline to be safe and effective for depression and anxiety in 31 patients. These patients had a variety of substance use disorders and comorbid depression and/or anxiety disorder. Only one patient reported side effects. Anxiety Treatment choices are similar to those for depression. Additionally, buspirone is useful for generalized (but not panic-type) anxiety. In our experience, higher-than-usual buspirone doses are required, and frequently it is used in combination with another anxiolytic. The benzodiazepines, of course, are also very effective anti-anxiety agents, but the abuse potential limits their usefulness in dual-diagnosis patients. As a last resort, when other treatment modalities have resulted in no documented improvement, benzodiazepines can be considered. Adinoff (1992), for example, has reported long-term administration of benzodiazepines in seven male patients, all over 50 years old and each with at least a 25-year history of alcohol dependence. All seven had improvement in symptoms and maintained abstinence, except for a brief alcohol relapse in two. Furthermore, there was no benzodiazepine misuse. Mania and Affective Lability The importance of comorbid substance abuse and bipolar disorder was outlined earlier. As Weiss and colleagues (1998a) have noted, because of side effects lifetime adherence to lithium treatment is very low in substanceabusing bipolar patients. Compliance with divalproex is significantly higher. ❉

287 ❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

Brady and colleagues (1995) have pointed out that the higher rates of mixed, dysphoric, and rapid-cycling mania in substance abusers suggest the usefulness of divalproex in these individuals, because data indicate that the anticonvulsants are more effective than lithium for these manic states (Keck and McElroy 1996). Furthermore, Brady and colleagues (1995) used divalproex in nine acutely manic patients with comorbid substance use disorders. They found good resolution of affective symptoms, no adverse effects, and a significant decrease in days of substance use during the follow-up period. Likewise, we (Albanese et al. 2000) treated 20 mood-disordered, substanceabusing patients with divalproex and found the medication both efficacious for the mood disorder and safe alone or in combination with other psychiatric medications. As we noted in a previous section, mood lability and irritability are frequently found in nonbipolar conditions, most notably PTSD and borderline personality disorder. Again, we have found divalproex a safe and effective medication in these populations. In addition, we have found the atypical antipsychotics (most notably risperidone) safe and effective for mood lability—even when combined with other psychiatric medications—in substance-abusing bipolar, borderline, or PTSD patients (Albanese, in press). We use low doses of atypicals, which do not cause the side effects that are so troublesome to these patients. Psychosis Although the typical antipsychotics are both safe and effective in substanceabusing psychotic patients, the side-effect profile is disconcerting and may contribute to the drive to misuse substances. For example, LeDuc and Mittleman (1995) reported an association between use of traditional antipsychotics and stimulant abuse in schizophrenic patients. Voruganti and colleagues (1997) reported that schizophrenic patients with a history of neuroleptic-induced dysphoria (ND) were four times more likely than those without ND to develop substance use disorder during the course of their maintenance antipsychotic treatment. As a result, we usually prescribe an atypical antipsychotic. Clozapine (Buckley et al. 1994), olanzapine (Conley et al. 1997), and risperidone (Albanese, in press) have demonstrated efficacy in psychotic patients with substance use disorders. In addition, lower incidence of extrapyramidal side effects results in less use of anticholinergic medications, which may be abused, especially by patients with neuroleptic dysphoria. Of note, Buckley and colleagues (1994) found, in a 6-month study, that clozapine was equally efficacious in treatment-refractory schizo❉

288 ❉

❉

The Difficult-to-Treat Patient With Substance Abuse ❉

phrenic patients with and without substance abuse. The results of this study indicate that not only is clozapine not contraindicated in substance-abusing patients but that this medication also may improve the poor clinical response typical in schizophrenia complicated by substance use disorder. Even more intriguing is the emerging data indicating that the newer antipsychotics may also decrease substance use in psychotic patients, who are among the most chronically relapsing patients that we treat. Albanese and colleagues (1994b) reported two case studies of treatment-resistant schizophrenic patients with comorbid alcohol and cocaine dependence. Clozapine decreased both positive and negative symptoms of schizophrenia, as well as substance abuse. We have now followed these and other patients over several years, observing continued abstinence from alcohol and cocaine. Preclinical data in rats and monkeys suggest that clozapine reduces cocaine-seeking behavior. Clinically, Yovell and Opler (1994) reported one case in which a treatment-resistant schizoaffective patient treated with clozapine experienced resolution of psychosis and decreased cocaine use. In the case vignette that follows, we describe similar results. We have also found that clozapine is effective in substance-abusing borderline patients who have frequent psychotic episodes, poor reality testing, affective lability, and behavioral dyscontrol. For example, one of our borderline patients with a history of alcohol and cocaine dependence had had multiple psychiatric admissions, detoxifications, and incarcerations. She also had had trials of many antidepressants, antipsychotics, and mood stabilizers. We ultimately discharged her on a regimen of clozapine, paroxetine, and valproic acid. She has had no hospital admissions or relapse to substance abuse for over 5 years. To our knowledge, there are no published reports on the efficacy of the other atypical antipsychotics in decreasing substance use in psychotic patients. Of note, risperidone has been used to treat methamphetamine psychosis (Misra and Kofoed 1997), alcohol hallucinosis (Soyka et al. 1997), and cocaine craving in eight withdrawn cocaine-dependent male patients (Roy et al. 1998). The pharmacological mechanisms underlying clozapine’s effect on substance use are unresolved but presumably involve the mesolimbic dopamine system (Buckley 1998). Wilkins (1997) presents a succinct review of this subject, which is beyond the scope of this chapter. Clozapine, and perhaps other atypical antipsychotics, may work in substance use disorder via several possible mechanisms, including 1) reduction of neuroleptic dysphoria, as mentioned earlier; 2) reduction in extrapyramidal side effects (Miller and ❉

289 ❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

Tanenbaum 1989), and 3) reduced need to “self-medicate” depression and negative symptoms of schizophrenia (Albanese et al. 1994b; Brady et al. 1990; Freed 1975; Khantzian 1985, 1997; Siris et al. 1993). In a recent review of the pharmacotherapy of schizophrenic patients with substance use disorders, Wilkins (1997) reported that both desipramine and imipramine have been used successfully to decrease cocaine use in schizophrenic patients. Likewise, disulfiram has been used effectively in schizophrenic patients with alcoholism, despite earlier reports that disulfiram use is correlated with psychosis (Kingsbury and Salzman 1990). Attention-Deficit/Hyperactivity Disorder Stimulants, namely methylphenidate and dextroamphetamine, are the most effective medication treatment for attention-deficit/hyperactivity disorder (ADHD). Unfortunately, they also have high abuse potential. In some cases, however, we have been able to treat substance-abusing ADHD patients with methylphenidate. We have encountered several patients who appeared to be self-medicating ADHD-related symptoms with cocaine, who when treated with methylphenidate had resolution of ADHD symptoms without experiencing drug relapse (Khantzian 1983; Khantzian et al. 1984). Other treatment options for ADHD patients with substance use disorders include the Schedule IV stimulant pemoline, which has less abuse potential (but some risk of hepatic toxicity); tricyclic antidepressants, such as imipramine and desipramine; bupropion; and clonidine. We will now present a clinical vignette that illustrates some of the issues that we have mentioned in this chapter. Case Vignette Ms. T is a 29-year-old unmarried white woman whose first psychiatric admission occurred at age 14 years for depression. She reports onset of substance abuse in childhood, starting with alcohol, progressing to marijuana at age 12, and subsequently, to cocaine. She has had multiple psychiatric admissions since age 14, usually for paranoia, mood lability, and assaultiveness. The admissions typically occurred in the context of treatment noncompliance and were sometimes complicated by substance use. The patient has been treated with many different antidepressants, antipsychotics, and mood stabilizers. We first met the patient during her first admission to our state hospital. Because she had been causing a public disturbance, the police were called, and the patient reported that “people were trying to get” her. She became assaultive toward the police and was arrested. Because of continued

❉

290 ❉

❉

The Difficult-to-Treat Patient With Substance Abuse ❉

unusual statements (e.g., “I murdered someone last week.”), she was assessed by a court psychologist, to whom she revealed that she had recently stopped taking her psychiatric regimen of sertraline, risperidone, and divalproex sodium. The patient was sent to our hospital for court-ordered psychiatric assessment. After admission, Ms. T revealed that, in addition to not taking her medication, she had also been using cocaine; this was confirmed by urine toxicology screen. After several days, because of persistent delusions, irritability, and hostility, she was restarted on the above-mentioned medication regimen. Although she remained somewhat isolative, she was in good control and no longer delusional. The patient returned to court at the expiration of her 20-day commitment and was discharged by the judge. Five days later, again in the context of cocaine use and medication noncompliance, she became delusional and agitated, and assaulted her mother. Once again, she was hospitalized for court-ordered assessment and subsequent treatment, which lasted 9 months. During this hospital stay, because of persistent positive and negative symptoms of schizophrenia, despite sobriety and adequate antipsychotic doses, Ms. T was started on clozapine and stabilized on 350 mg/day. The patient also participated in our inpatient dual-diagnosis psychosocial rehabilitation program and received weekly individual supportive psychotherapy from her psychiatrist. The remaining psychotic symptoms resolved. Emergence of persistent depressive symptoms was treated with sertraline 100 mg qd, with good result. The patient was discharged to a state-funded dual-diagnosis residential program, where she has been for 4 years. Since discharge, she had a brief substance abuse relapse, but has been sober for 3½ years. In addition, she has had two brief psychiatric admissions at private hospitals for psychotic exacerbations, despite medication compliance and sobriety. Olanzapine has been added to clozapine to augment the latter’s antipsychotic efficacy. During a recent interview, the patient reported that she has completed junior college and is in the process of applying to a liberal arts program. In addition, she and her sponsor have recently started a Dual Recovery Anonymous group. Her only medication side effect has been weight gain.

This case highlights many of the points made previously, including the adverse outcomes and treatment-resistance of dually diagnosed patients. After starting an atypical antipsychotic, and augmenting this with other treatments, however, the patient’s course has improved. Of note, the patient has also received integrated psychosocial rehabilitation over the past 4½ years.

Summary and Clinical Wisdom We have reviewed the epidemiology of comorbid psychiatric and substance use disorders, noting that the prevalence of alcohol and drug abuse is higher ❉

291 ❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

among patients with identified mental illness than among the general population. Similarly, psychiatric disorders are more prevalent among patients with substance use disorders than in the population as a whole. We have also discussed the implications of the comorbidity in terms of the poorer treatment response and outcome for each disorder in the presence of the other. This is most dramatically illustrated by the increased likelihood of suicide and homicide among dual-diagnosis patients. In addition, we have reviewed some of the issues involved in assessing and diagnosing the comorbid disorder in a patient who presents first with one disorder. Finally, we have highlighted some of the treatment principles for approaching dual-diagnosis patients, emphasizing the importance of integrated treatment, provision of psychosocial modalities, and updated biological interventions. The course of treatment and therapeutic outcome for this population can be discouraging. However, both our experience and evidence in the literature reveal that careful evaluation and an emphasis on the treatment principles outlined in this chapter may significantly improve treatment efficacy and outcome for this most difficult group of patients.

References Adinoff B: Long-term therapy with benzodiazepines despite alcohol dependence disorder: seven case reports. Am J Addict 1:288–293, 1992 Akiskal HS: The prevalent clinical spectrum of bipolar disorders: beyond DSM-IV. J Clin Psychopharmacol 16 (suppl 1):4S–14S, 1996 Akiskal HS, Mallya G: Criteria for the “soft” bipolar spectrum: treatment implications. Psychopharmacol Bull 23:68–73, 1987 Albanese MJ: Assessing and treating comorbid mood and substance use disorders. Psychiatric Times, 2001, Vol 18, No 4, pp 55–58 Albanese MJ: Safety and efficacy of risperidone in substance abusers with psychosis (letter). Am J Addict (in press) Albanese MJ, Khantzian EJ: Divalproex in substance abusers with mood disorder. Poster presented at The American Academy of Addiction Psychiatry Ninth Annual Meeting and Symposium, Amelia Island, FL, December 3–6, 1998 Albanese MJ, Clodfelter RC Jr, Khantzian EJ: Underdiagnosis of bipolar disorder in substance abusers. Poster presented at the American Academy of Addiction Psychiatry 10th Annual Meeting and Symposium, Nassau, Bahamas, December 2– 5, 1999 Albanese MJ, Clodfelter RC Jr, Khantzian EJ: Divalproex sodium in substance abusers with mood disorder. J Clin Psychiatry 61:916–921, 2000

❉

292 ❉

❉

The Difficult-to-Treat Patient With Substance Abuse ❉

Albanese MJ, Bartel RL, Bruno RF, et al: Comparison of measures used to determine substance abuse in an inpatient psychiatric population. Am J Psychiatry 51:1077– 1078, 1994a Albanese MJ, Khantzian EJ, Murphy SL, et al: Decreased substance use in chronically psychotic patients treated with clozapine (letter). Am J Psychiatry 151:780–781, 1994b Albanese MJ, Blair W, DiRocco D, et al: Depression as a predictor of compliance with substance abuse treatment. Poster presented at Fifth Annual Research Day, The Consolidated Department of Psychiatry, Harvard Medical School, Boston, MA, March 12, 1997 Albanese MJ, Graham-Brown A, Vanelli M, et al: Risperidone augmentation of clozapine. Poster presented at Sixth Annual Research Day, The Consolidated Department of Psychiatry, Harvard Medical School, Boston, MA, April 22, 1998 American Psychiatric Association: Diagnostic and Statistical Manual of Mental Disorders, 4th Edition, Text Revision. Washington, DC, American Psychiatric Association, 2000 Ananth J, Vanderwater S, Kamal M, et al: Missed diagnosis of substance abuse in psychiatric patients. Hosp Community Psychiatry 40:297–299, 1989 Bailey L, Maxwell S, Brandabur MM: Substance abuse as a risk factor for tardive dyskinesia: a retrospective analysis of 1,027 patients. Psychopharmacol Bull 3:177– 181, 1997 Bartels SJ, Teague GB, Drake RE, et al: Substance abuse in schizophrenia: service utilization and costs. J Nerv Ment Dis 181:227–232, 1993 Berglund M: Suicide in alcoholism: a prospective study of 88 suicides, I: the multidimensional diagnosis at first admission. Arch Gen Psychiatry 41:888–891, 1984 Black DW, Winokur G, Hulbert J, et al: Predictors of immediate response in the treatment of mania: the importance of comorbidity. Biol Psychiatry 24:191–198, 1988 Bowers MB Jr, Mazure CM, Nelson JC, et al: Psychotogenic drug use and neuroleptic response. Schizophr Bull 16:81–85, 1990 Brady KT: Posttraumatic stress disorder and comorbidity: recognizing the many faces of PTSD. J Clin Psychiatry 58 (suppl 9):12–15, 1997 Brady KT, Anton R, Ballenger JC, et al: Cocaine abuse among schizophrenic patients. Am J Psychiatry 147:1164–1167, 1990 Brady KT, Sonne SC, Anton R, et al: Valproate in the treatment of acute bipolar affective episodes complicated by substance abuse: a pilot study. J Clin Psychiatry 56:118– 121, 1995 Brady KT, Dansky BS, Sonne SC, et al: Posttraumatic stress disorder and cocaine dependence. Am J Addict 7:128–135, 1998 Breakey WR, Calabrese L, Rosenblatt A, et al: Detecting alcohol use disorders in the severely mentally ill. Community Ment Health J 34:165–174, 1998 Brown SA, Inaba RK, Gillin JC, et al: Alcoholism and affective disorder: clinical course of depressive symptoms. Am J Psychiatry 152:45–52, 1995

❉

293 ❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

Buckley PF: Substance abuse in schizophrenia: a review. J Clin Psychiatry 59 (suppl 3):26–30, 1998 Buckley PF, Thompson PA, Way L, et al: Substance abuse and clozapine treatment. J Clin Psychiatry 55 (suppl B):114–116, 1994 Calabrese JR, Delucchi GA: Spectrum of efficacy of valproate in 55 patients with rapidcycling bipolar disorder. Am J Psychiatry 147:431–434, 1990 Christie KA, Burke JD, Regier DA, et al: Epidemiologic evidence for early onset of mental disorders and higher risk of drug abuse in young adults. Am J Psychiatry 145:971–975, 1988 Conley R, Gale E, Hirsch K: Olanzapine response in therapy-refractory schizophrenia with substance abuse (SA) (abstract). Schizophr Res 24:190, 1997 Cornelius JR, Salloum IM, Ehler JG, et al: Fluoxetine in depressed alcoholics: a doubleblind, placebo-controlled trial. Arch Gen Psychiatry 54:700–705, 1997 D’Mello DA, Boltz MK, Msibi B: Relationship between concurrent substance abuse in psychiatric patients and neuroleptic dosage. Am J Drug Alcohol Abuse 21:257– 265, 1995 Drake RE, Osher FC, Wallach MA: Alcohol use and abuse in schizophrenia: a prospective community study. J Nerv Ment Dis 177:408–414, 1989 Drake RE, Antosca LM, Noordsy DL, et al: New Hampshire’s specialized services for the dually diagnosed. New Dir Ment Health Serv 50:57–67, 1991 Drake RE, Noordsy DL, Ackerson T: Integrating mental health and substance abuse treatments, in Double Jeopardy: Chronic Mental Illness and Substance Use Disorders. Edited by Lehman AF, Dickson L. Chur, Switzerland, Harwood Academic Publishers, 1995, pp 251–264 Drake RE, Mueser KT, Clark RE, et al: The course, treatment, and outcome of substance disorders in persons with severe mental illness. Am J Orthopsychiatry 66:42–51, 1996 Fawcett J: Predictors of early suicide: identification and appropriate intervention. J Clin Psychiatry 49 (suppl):7–8, 1988 Fernandez-Pol B, Bluestone H, Mizruchi MS: Inner-city substance abuse patterns: a study of psychiatric inpatients. Am J Drug Alcohol Abuse 14:41–50, 1988 Freed EX: Alcoholism and schizophrenia: the search for perspectives: a review. J Stud Alcohol 36:853–881, 1975 Green VL: The resurrection and the life. Am J Orthopsychiatry 66:12–16, 1996 Greenfield SF, Weiss RD, Muenz LR, et al: The effect of depression on return to drinking: a prospective study. Arch Gen Psychiatry 55:259–265, 1998 Gupta S, Hendricks S, Kenkel AM, et al: Relapse in schizophrenia: is there a relationship to substance abuse? Schizophr Res 20:153–156, 1996 Hasin DS, Tsai WY, Endicott J, et al: The effects of major depression on alcoholism: five-year course. Am J Addict 5:144–155, 1996

❉

294 ❉

❉

The Difficult-to-Treat Patient With Substance Abuse ❉

Hellerstein DJ, Rosenthal RN, Miner CR: A prospective study of integrated outpatient treatment for substance-abusing schizophrenic outpatients. Am J Addict 4:33– 42, 1995 Keck PE, McElroy SL: Outcome in the pharmacologic treatment of bipolar disorder. J Clin Psychopharmacol 16 (suppl 1):15S-23S, 1996 Keller MB, Lavori PW, Coryell W, et al: Differential outcome of pure manic, mixed/ cycling, and pure depressive episodes in patients with bipolar illness. JAMA 255:3138–3142, 1986 Kessler RC, McGonagle KA, Zhao S, et al: Lifetime and 12-month prevalence of DSMIII-R psychiatric disorders in the United States: results from the National Comorbidity Survey. Arch Gen Psychiatry 51:8–19, 1994 Kessler RC, Sonnega A, Bromet E, et al: Posttraumatic stress disorder in the National Comorbidity Survey. Arch Gen Psychiatry 52:1048–1060, 1995 Kessler RC, Crum RM, Warner LA, et al: Lifetime co-occurrence of DSM-III-R alcohol abuse and dependence with other psychiatric disorders in the National Comorbidity Survey. Arch Gen Psychiatry 54:313–321, 1997 Khantzian EJ: An extreme case of cocaine dependence and marked improvement with methylphenidate treatment. Am J Psychiatry 140:784–785, 1983 Khantzian EJ: The self-medication hypothesis of addictive disorders: focus on heroin and cocaine dependence. Am J Psychiatry 142:1259–1264, 1985 Khantzian EJ: The self-medication hypothesis of substance use disorders: a reconsideration and recent applications. Harv Rev Psychiatry 4:231–244, 1997 Khantzian EJ, Gawin F, Kleber HD, et al: Methylphenidate (Ritalin) treatment of cocaine dependence: a preliminary report. J Subst Abuse Treat 1:107–112, 1984 Khantzian EJ: Halliday KS, McAuliffe WE: Addiction and the Vulnerable Self: Modified Dynamic Group Therapy for Substance Abusers. New York, Guilford, 1990 Kingsbury SJ, Salzman C: Disulfiram in the treatment of alcoholic patients with schizophrenia. Hospital and Community Psychiatry 41:133–134, 1990 Kline J, Harris M, Bebout RR, et al: Contrasting integrated and linkage models of treatment for homeless, dually diagnosed adults. New Dir Ment Health Serv 50:95–106, 1991 Kosten TR, Bryant K, Rounsaville BJ: The SCID: a clinical instrument for assessing psychiatric disorders in substance abusers. NIDA Res Monogr 108:213–219, 1991 Kovasznay B, Fleischer J, Tanenberg-Karant M: Substance use disorder and the early course of illness in schizophrenia and affective psychosis. Schizophr Bull 23:195– 201, 1997 LeDuc PA, Mittleman G: Schizophrenia and psychostimulant abuse: a review and reanalysis of clinical evidence. Psychopharmacology 121:407–427, 1995 Lieberman JA, Kane JM, Alvir J: Provocative tests with psychostimulant drugs in schizophrenia. Psychopharmacology 91:415–433, 1987 Loosen PT, Dew BW, Prange AJ: Long-term predictors of outcome in abstinent alcoholic men. Am J Psychiatry 147:1662–1666, 1990

❉

295 ❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

Mason BJ, Kocsis JH, Ritvo EC, et al: A double-blind, placebo-controlled trial of desipramine for primary alcohol dependence stratified on the presence or absence of major depression. JAMA 275:761–767, 1996 McElroy SL, Strakowski SM, Keck PE, et al: Differences and similarities in mixed and pure mania. Compr Psychiatry 36:187–195, 1995 McGrath PJ, Nunes EV, Stewart JW, et al: Imipramine treatment of alcoholics with primary depression: a placebo-controlled clinical trial. Arch Gen Psychiatry 53:232–240, 1996 McLellan AT, Arndt IO, Metzger DS, et al: The effects of psychosocial services on substance abuse treatment. JAMA 269:1953–1959, 1993 Miller F, Tanenbaum J: Drug abuse in schizophrenia. Hospital and Community Psychiatry 40:847–849, 1989 Miller NS: Issues in the diagnosis and treatment of comorbid addictive and other psychiatric disorders. Directions in Psychiatry 14:1–8, 1994 Minkoff K: Program components of a comprehensive integrated care system for seriously mentally ill patients with substance disorders. New Dir Ment Health Serv 50:13–27, 1991 Misra L, Kofoed L: Risperidone treatment of methamphetamine psychosis (letter). Am J Psychiatry 154:1170, 1997 Mowbray CT, Solomon M, Ribisl KM, et al: Treatment for mental illness and substance abuse in a public psychiatric hospital: successful strategies and challenging problems. J Subst Abuse Treat 12:129–139, 1995 Nunes EV, McGrath PJ, Quitkin FM, et al: Imipramine treatment of alcoholism with comorbid depression. Am J Psychiatry 150:963–965, 1993 Nunes EV, Quitkin FM, Donovan SJ, et al: Imipramine treatment of opiate-dependent patients with depressive disorders: a placebo-controlled trial. Arch Gen Psychiatry 55:153–160, 1998 Osher FC, Kofoed LL: Treatment of patients with psychiatric and psychoactive substance abuse disorders. Hospital and Community Psychiatry 40:1025–1030, 1989 Owen RR, Fischer EP, Booth BM, et al: Medication noncompliance and substance abuse among patients with schizophrenia. Psychiatr Serv 47:853–858, 1996 Post RM, Altshuler LL, Ketter TA, et al: Antiepileptic drugs in affective illness: clinical and theoretical implications, in Advances in Neurology, Vol 55. Edited by Smith D, Treiman D, Trimble M. New York, Raven, 1991, pp 239–277 Rasanen P, Tiihonen J, Isohanni M, et al: Schizophrenia, alcohol abuse, and violent behavior: a 26-year follow-up study of an unselected birth cohort. Scizophr Bull 24:437–441, 1998 Regier DA, Boyd JH, Burke JD, et al: One-month prevalence of mental disorders in the United States: based on five epidemiologic catchment area sites. Arch Gen Psychiatry 45:977–986, 1988

❉

296 ❉

❉

The Difficult-to-Treat Patient With Substance Abuse ❉

Regier DA, Farmer ME, Rae DS, et al: Comorbidity of mental disorders with alcohol and other drug abuse: results from the Epidemiologic Catchment Area (ECA) study. JAMA 264:2511–2518, 1990 Rich CL, Ricketts JE, Fowler RC, et al: Some differences between men and women who commit suicide. Am J Psychiatry 145:718–722, 1988 Ridgely MS, Goldman HH, Willenbring M: Barriers to the care of persons with dual diagnoses: organizational and financing issues. Schizophr Bull 16:123–131, 1990 Ries RK, Russo J, Wingerson D, et al: Shorter hospital stays and more rapid improvement among patients with schizophrenia and substance disorders. Pschiatr Serv 51:210–215, 2000 Robins LN, Helzer JE, Croughan J, et al: NIMH Diagnostic Interview Schedule: Version III (May 1981). Rockville, MD, National Institute of Mental Health, 1981 Rosenberg SD, Drake RE, Wolford GL, et al: Dartmouth Assessment of Lifestyle Instrument (DALI): a substance use disorder screen for people with severe mental illness. Am J Psychiatry 155:232–238, 1998 Rosenthal RN, Hellerstein DJ, Miner CR: Integrated services for treatment of schizophrenic substance abusers: demographics, symptoms, and substance abuse patterns. Psychiatr Q 63:3–26, 1992 Roy A, Roy M, Smelson DA: Risperidone, ERG, and cocaine craving (letter). Am J Addict 7:90, 1998 SAMHSA Conference Report and National Strategy on Co-occurring Disorders: Improving services for individuals at risk of, or with, co-occurring substance-related and mental health disorders. Rockville, MD, Substance Abuse and Mental Health Services Administration, U.S. Department of Health and Human Services, 1996 Satel SL: When disability benefits make patients sicker. N Engl J Med 333:794–796, 1995 Sciacca K: An integrated treatment approach for severely mentally ill individuals with substance disorders. New Dir Ment Health Serv 50:69–84, 1991 Shaner A, Khalsa ME, Roberts L, et al: Unrecognized cocaine use among schizophrenic patients. Am J Psychiatry 150:758–762, 1993 Shaner A, Eckman TA, Roberts LJ, et al: Disability income, cocaine use, and repeated hospitalization among schizophrenic cocaine abusers: a government-sponsored revolving door? N Engl J Med 333:777–783, 1995 Siris SG, Mason SE, Bermanzohn PC, et al: Dual-diagnosis/psychiatric comorbidity of drug dependence: epidemiology and treatment. Psychopharmacol Bull 29:127–132, 1993 Sonne SC, Brady KT, Morton WA: Substance abuse and bipolar affective disorder. J Nerv Ment Dis 182:349–352, 1994 Soyka M, Wegner U, Moeller HJ: Risperidone in treatment-refractory chronic alcohol hallucinosis (letter). Pharmacopsychiatry 30:135–136, 1997

❉

297 ❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

Spitzer RL, Williams JBW: Structured Clinical Interview for DSM-III-R—Patient Version (SCID-P). New York, New York State Psychiatric Institute, Biometrics Research, 1986 Stone AM, Greenstein RA, Gamble G, et al: Cocaine use by schizophrenic outpatients who receive depot neuroleptic medication. Hospital and Community Psychiatry 44:176–177, 1993 Strakowski SM, Keck PE, McElroy SL, et al: Twelve-month outcome after a first hospitalization for affective psychosis. Arch Gen Psychiatry 55:49–55, 1998 Swofford CD, Kasckow JW, Scheller-Gilkey G, et al: Substance use: a powerful predictor of relapse in schizophrenia. Schizophr Res 20:145–151, 1996 Vaillant GE: Is alcoholism more often the cause or the result of depression? Harv Rev Psychiatry 1:94–99, 1993 Voruganti LNP, Heslegrave RJ, Awad AG: Neuroleptic dysphoria may be the missing link between schizophrenia and substance abuse. J Nerv Ment Dis 185:463–465, 1997 Weiss RD, Greenfield SF, Najavits LM, et al: Medication compliance among patients with bipolar disorder and substance use disorder. J Clin Psychiatry 59:172–174, 1998a Weiss RD, Najavits LM, Greenfield SF, et al: Validity of substance use self-reports in dually diagnosed outpatients. Am J Psychiatry 155:127–128, 1998b Weiss RD, Griffin ML, Greenfield SF, et al: Group therapy for patients with bipolar disorder and substance dependence: results of a pilot study. J Clin Psychiatry 61:361–367, 2000 Wilens TE, Biederman J, Mick E: Does ADHD affect the course of substance abuse? Am J Addict 7:156–163, 1998 Wilkins JN: Pharmacotherapy of schizophrenia patients with comorbid substance abuse. Schizophr Bull 23:215–228, 1997 Winokur G, Coryell W, Akiskal HS, et al: Alcoholism in manic-depressive (bipolar) illness: familial illness, course of illness, and the primary-secondary distinction. Am J Psychiatry 152:365–372, 1995 Young LT, Cooke RG, Robb JC, et al: Anxious and non-anxious bipolar disorder. J Affect Disord 29:49–52, 1993 Yovell Y, Opler LA: Clozapine reverses cocaine craving in a treatment-resistant mentally ill chemical abuser: a case report and a hypothesis. J Nerv Ment Dis 182:591– 592, 1994 Zaretsky A, Rector NA, Seeman MV, et al: Current cannabis use and tardive dyskinesia. Scizophr Res 1993 11:3–8, 1993

❉

298 ❉

11 The Difficult-to-Treat Patient With Dementia or Traumatic Brain Injury Jeffrey L. Cummings, M.D. Ronald E. Saul, M.D. Tomoko Nakawatase, M.D.

D

ementia, particularly Alzheimer’s disease (AD), is a common cause of behavioral disturbances in the elderly, and traumatic brain injury (TBI) is a frequent cause of difficult-to-control behavior in patients of all ages. The presence of a brain disorder may alter the response of these patients to conven-

This project was supported by a National Institute of Health, Alzheimer’s Disease Center grant AG10123, an Alzheimer’s Disease Research California grant, and the Sidell-Kagan Foundation.

❉

299 ❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

tional psychotropic medications and also may increase the likelihood of adverse events during the course of treatment. Furthermore, accompanying cognitive abnormalities make monitoring of the response and surveillance for side effects more challenging for the clinician. In AD, drugs unique to treatment of this condition such as cholinergic agents may have beneficial psychotropic effects that can be used to help ameliorate behavioral disturbances. In this chapter, we present an approach to treating difficult-to-manage patients with AD or TBI. Background information on AD and TBI is presented first, the use of cholinergic agents for the treatment of behavioral disturbances in AD is described, pharmacotherapy of acute TBI patients is discussed, and finally the use of psychotropics in AD and TBI patients with chronic behavioral disorders is summarized.

Alzheimer’s Disease AD is a progressive neurodegenerative syndrome with neuropsychological and neuropsychiatric manifestations. The latter include apathy, agitation, anxiety, depression, delusions, hallucinations, disinhibition, irritability, purposeless behavior (rummaging and pacing), wandering, and disturbances of sexual behavior. The diagnosis of AD requires the presence of a dementia syndrome, including memory abnormalities, that has been progressive over a minimum 6-month period and is not attributable to another disorder affecting brain function (McKhann et al. 1984). Accurate clinical diagnosis can be achieved in approximately 90% of cases if diagnostic guidelines are applied. Current therapeutic approaches to AD include administration of antioxidant compounds such as vitamin E, use of cholinesterase inhibitors, and estrogen therapy of postmenopausal women. A combination of these agents will likely be coadministered with psychotropic agents in behaviorally disturbed patients with AD.

Behavioral Disturbances in Alzheimer’s Disease Nearly all patients with AD exhibit behavioral abnormalities (Cummings and Victoroff 1990; Sultzer et al. 1973). In the initial phases of the illness, apathy, irritability, and depression are most apparent, and as the disease progresses, delusions and agitation become more frequent. Behavioral symptoms relapse and remit over the course of AD, but once present they have a high likelihood of recurring. Most patients exhibit multiple behavioral dis❉

300 ❉

❉

The Difficult-to-Treat Patient With Dementia or Traumatic Brain Injury ❉

turbances simultaneously, and pharmacotherapy must often be multidimensional. The most common behavioral disturbances are described here. Agitation Agitation includes a variety of behavioral abnormalities such as physical assault, verbal aggression, and psychomotor restlessness. The incidence of agitation appears high, but estimates vary partly because of the difficulty in defining the syndrome. The prevalence of agitation in AD ranges from about 48% to 70% and may occur independently of other major neuropsychiatric disorders. Agitation becomes more common as AD progresses (Merriam et al. 1988; Teri et al. 1989). Psychosis Delusions are false beliefs based on an incorrect perception or interpretation of external reality that cannot be modified by reasoning (American Psychiatric Association 1994). Wragg and Jeste (1989) reviewed 22 studies reporting the frequency of delusions in AD patients and found rates varying from 10% to 73%, with most studies reporting frequencies in the range of 30% to 38%. Variability of reported rates may be due to variable criteria for AD and delusions, differing methodologies for assessing the psychopathology, and differences in the referral populations. The occurrence of delusions in AD correlated with increasing age and a multicenter study of predictors of AD course found that delusional patients were older than nondelusional patients (Swearer 1988). Drevets and Rubin (1989) found no relationship between age at onset and delusions in AD. Further investigations are needed to assess the relationship between age, stage of dementia, and the occurrence of delusions. Delusions can occur in any stage of AD, although delusions are more frequent in the middle and later phases of the illness. AD patients with delusions are more cognitively impaired than patients without delusions. Deterioration in cognitive function is more rapid in delusional patients compared to nondelusional patients, but there is no difference in survival. Delusions of paranoia, with fear of personal harm, theft of one’s property, and abandonment are common in AD. Misidentification syndromes occur when patients believe that their residence is not their home, strangers are living in the house (phantom boarder), television characters are real or in the home (picture sign), and people are imposters (Capgras syndrome). One study (Reisberg et al. 1987) reported 48% of 33 delusional AD patients had delusions of theft, 21% believed their house was not their home, and 9% ❉

301 ❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

believed that their spouse was an imposter. Delusions of grandeur or of religion commonly observed in schizophrenia or mania are unusual in AD. Depression The prevalence of depression in dementia is variable, with investigators reporting depressed mood in 0–87% of AD patients, with most studies identifying depression in 40%–50% (Cummings et al. 1995). AD patients meeting DSM-IV criteria for major depression are infrequent, but elements of a depression syndrome are frequent, occurring in 20%–40% of AD patients (Cummings et al. 1995). Mega and colleagues (1996) investigated the range of behavioral abnormalities in patients with AD compared with normal agematched control subjects using the Neuropsychiatric Inventory (Cummings et al. 1994) and found dysphoria in 38% of AD patients. Thoughts of being a burden and feelings of helplessness are commonly expressed, but suicide is rare in AD. Depressed mood is more state-dependent and modifiable by environmental stimuli, with patients quickly relinquishing sadness in response to distractions (Cummings et al. 1995). Because some demented patients with depression do not complain of depression or perceive themselves to be depressed, clinicians must depend on statements of caregivers or behavioral observations. Depression in these patients may present in alternate ways such as chronic irritability, multiple somatic complaints, withdrawal, refusal to eat or take responsibility for daily care, and constant negativism. Anxiety Anxiety has been reported in approximately 48% of AD patients (Mega et al. 1996). The most common manifestation is excessive anticipatory concern regarding upcoming events (Mendez et al. 1990). Anxiety is more common in the earlier stages of dementia, when patients are more aware of their deficits. The incidence of anxiety in AD patients may be higher than reported, as it may be expressed in the form of agitation, insomnia, motor restlessness, or tearfulness. Apathy Apathy is reported to be the most common behavioral disturbance in AD, occurring in up to 90% of patients (Craig et al. 1996; Mega et al. 1996). Apathy occurs in all stages of the illness but appears to correlate with disease progression (Craig et al. 1996; Mega et al. 1996). Patients become indifferent ❉

302 ❉

❉

The Difficult-to-Treat Patient With Dementia or Traumatic Brain Injury ❉

toward their surroundings and personal relationships, lack initiative and motivation, and are withdrawn and passive. Apathy appears to be related to prefrontal and anterior temporal dysfunction (Craig et al. 1996).

Alzheimer’s Disease: Specific Pharmacotherapy The only psychotropic agents currently available and unique to the management of AD are cholinergic agents. These drugs produce cognitive benefit or slow the rate of cognitive decline in a majority of patients (Cummings 2000; Cummings et al. 1998b; Farlow et al. 1992; Rogers et al. 1996). Recently, these drugs have been shown to have beneficial behavioral effects. Early studies with physostigmine indicated that this agent could reduce delusions and agitation in AD (Gorman et al. 1993). Tacrine (80–120 mg) was shown to reduce psychosis, apathy, anxiety, purposeless behavior, and visual hallucinations in AD patients not selected for behavioral disturbances (Kaufer et al. 1996, 1998; Raskind et al. 1997). Metrifonate (50–80 mg) has been shown to ameliorate behavioral disturbances in AD, lowering total behavioral abnormalities scale scores and improving visual hallucinations (Morris et al. 1998). Rivastigmine (6–12 mg) and galantamine (16–24 mg) also have been shown to reduce behavioral abnormalities in AD (Cummings 2000; Tariot et al. 2000). Head-to-head data comparing cholinesterase inhibitors are not available. An AD-specific strategy for treatment of behavioral disturbances begins with the initiation of antioxidant therapy (vitamin E, 2000 IU/day) and a cholinesterase inhibitor. This regimen alone may avert further pharmacotherapy. Patients whose behavioral problems are too urgent to await the 2- to 3-week period required to observe behavioral changes following cholinesterase inhibition therapy—or who do not respond to this intervention—will require the use of additional psychotropic medication as described later in this chapter. In some cases, pharmacotherapy of behavioral disturbances may be avoided through educational programs for the caregiver.

Traumatic Brain Injury Neuropsychiatric disorders are common complications of TBI, ranging from mild personality changes to delirium, dementia, psychosis, and severe personality alterations. Problems with attention, memory, and executive dysfunction may accompany the behavioral changes. These problems are managed by means of behavioral, environmental, psychosocial, and pharma❉

303 ❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

cological strategies. Reviews have been provided by Saul (1993a, 1993b), Silver and colleagues (1994, 1997), Mysiw and Sandel (1997), Rosenthal and colleagues (1998), Dombovy (1998), and Hornstein and Seliger (1998). TBI occurs at a rate of 200 per 100,000 people. Patients with even mild TBI, however, may have disabling symptoms, at least initially. Approximately 10%–15% of the mild cases will develop a persistent postconcussive syndrome requiring neuropsychiatric assessment and treatment to achieve optimal functional recovery. Ten percent of TBI cases are fatal; an additional 20%–30% of patients have moderate to severe injuries and need hospitalbased rehabilitation. The condition of the brain at the time of head trauma is the most important factor in the development of subsequent neurobehavioral syndromes. Advanced age, previous head injuries and other structural and metabolic abnormalities, toxic exposure, and substance abuse also increase the patient’s vulnerability (Levin and Eisenberg 1991; Silver et al. 1997). Preexisting socioeconomic status, level of education, and support networks will affect behavioral symptoms and the course of recovery (Silver et al. 1997).

Pathogenesis of TBI and Associated Clinical Syndromes There are many ways the brain may sustain traumatic injury. During falls or other low velocity accidents, contusions occur either directly at the site of impact (coup injury) or opposite the site (contracoup). Contusions occur most frequently in the orbitofrontal regions, or at the tips and inferior aspects of the temporal lobes, where the brain is in contact with the roughened bony prominences of the cranium. In high-velocity accidents, rotational movement of the brain within the skull produces diffuse axonal injury (DAI) and loss of consciousness. This primary damage (DAI or cerebral contusion) occurs at the time of injury. Secondary damage results from complications such as hematomas (epidural, subdural, or intracerebral), infections, hydrocephalus, cerebral edema, and the effects of compression or hypoxic-ischemic injury (Grafman and Salazar 1996; Saul 1993a). The best indices of DAI severity and long-term outcome are l) the early scores on the Glasgow coma scale (a 15-point function scale, that documents eye opening, verbal responsiveness, and motor response to stimuli), 2) coma duration, and 3) the duration of posttraumatic amnesia (the time period until restoration of continuous day-to-day memory). A posttraumatic amnesia (PTA) of less than 2 weeks indicates a relatively favorable outcome; ❉

304 ❉

❉

The Difficult-to-Treat Patient With Dementia or Traumatic Brain Injury ❉

with a PTA greater than 4 weeks, only a small percentage of cases achieve this result (Levin and Eisenberg 1991). Focal contusions predominantly injure prefrontal and anterior temporal regions, resulting in impaired behavior and higher order cognitive processes, such as emotion and self-awareness. More posteriorly represented language, perceptual, praxic, and sensorimotor functions are usually spared. Depending on the locus of the frontal lesion and the often associated DAI (which disrupts pathways to and from prefrontal cortex), various features of a frontal lobe syndrome are typically seen (e.g., one or more deficits in executive function, cognitive processing, selfmonitoring, and drive are predominant) (Saul 1993b). The orbitofrontal (pseudopsychopathic) syndrome is expressed behaviorally as disinhibition, affective lability and hyperactivity, impulsivity, distractibility, antisocial behavior, sexual preoccupation without overt sexual aggression, impaired insight, and poor interpersonal judgment. Frontal injury involving medial regions and anterior cingulate cortex or adjacent connections presents as abulia (lack of will), apathy, indifference, decreased initiative, psychomotor retardation, inflexibility, perseveration, and motor impersistence. Akinesia, sparse verbal output, and incontinence may occur (Cummings 1985). A mixture of these clinical syndromes is more common than any one of them, all characterized by inability to plan, limited insight into the nature of the disability, unrealistic ambition, and poor judgment. Other associated features are an inability to appreciate the effects of one’s remarks or behavior on others, inattention to personal appearance and hygiene, and a loss of social graces (Silver et al. 1997; Stuss 1987). The temporolimbic syndromes usually associated with temporal lobe contusions are even more variable than the frontal lobe syndromes. Lesions in different regions with specialized functions may produce corresponding deficits in language, memory, high-level sensory integration, emotional regulation, and expression. Personality disorders induced by temporal lobe injury may be primary or secondary to complex partial seizures. In its most extreme form, extensive bilateral temporal damage produces the KlüverBucy syndrome, identified in the coma recovery stage by apathy, visual and tactile agnosia, prosopagnosia (impaired recognition of familiar faces), oral exploration with bulimia-hyperphagia, hypersexuality or altered sexual behavior, aphasia, dementia, and amnesia (Lilly et al. 1983). These symptoms contrast with features of the personality disturbance associated with the later emergence of complex partial seizures, including excessive talkativeness, circumstantiality, viscosity, hypergraphia, hyposexuality, and uncharacteristic philosophical and religious interests (Eames 1990). ❉

305 ❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

Management in Early Stages of Recovery From TBI In acute rehabilitation settings, behavioral disturbances such as agitation, distractibility, irritability, and apathy can impede or even preclude traditional therapeutic interventions such as physical, occupational, and speech therapy. Timely and effective management will allow the patient to achieve a higher level of recovery and improved clinical outcome. There is a general consensus that behavioral techniques should be utilized before resorting to pharmacological interventions. If behavioral approaches are ineffective—or if the behavioral disturbance puts the patient or others at risk of injury—a trial of medication is warranted (Hornstein and Seliger 1998). TBI patients, emerging from coma in an intensive care unit, will initially exhibit cognitive unresponsiveness but maintain gross wakefulness and sleep-wake cycles. It is imperative that emerging responsiveness should not be compromised by the administration of sedatives or narcotics, often mistakenly given for treatment of pain or motor restlessness. Such agents may lead to emerging symptoms of confusion or posttraumatic delirium and a request for psychiatric consultation. Patients exhibit wandering attention and cannot reliably focus, sustain, or shift attention among external stimuli and internal mental processes. New learning and day-to-day recall are significantly impaired. Perseveration and denial of illness are frequently seen. As confusion clears, confabulation with a unique, grandiose, and preposterous quality may repeatedly appear (Alexander 1989b). There are often rapid transitions from an underaroused, abulic, and apathetic state to a hyperaroused and agitated state, accompanied by combative and paranoid behavior. Delusions, hallucinosis, and other perceptual distortions are consistent with those seen in any other delirium. Disturbed sleep-wake cycles and altered thirst and appetite are often encountered. Many, but not all, patients pass through a period of agitation (defined as episodic motor behavior interfering with care or requiring restraints) or motor restlessness (continuous behavior [i.e., pulling at a splint] of less severity than agitation). Younger patients are more likely to develop agitation, with a longer period of agitated behavior associated with alcohol use at the time of injury (Alexander 1989a; Saul 1993a). At this stage, behavioral and environmental interventions such as frequent verbal reorientation and provision of clocks, calendars, and familiar objects from home will improve the patient’s information processing and adaptation. Increasing stimulation and activity will promote spontaneity in the hypoaroused patient. Stimulant medication—amphetamine, methylpheni❉

306 ❉

❉

The Difficult-to-Treat Patient With Dementia or Traumatic Brain Injury ❉

date, or pemoline—may be beneficial (Gualtieri 1991b). Methylphenidate should be initiated at 5 mg bid (maximum daily dose 30 mg) and dextroamphetamine at 2.5 mg bid in the morning and early afternoon (maximum daily dose 20 mg). The use of stimulant drugs entails some risk of the patient becoming irritable or agitated. Because rebound depression may follow abrupt discontinuation, stimulants should always be tapered slowly. At usual doses, stimulants apparently do not lower seizure threshold (Cope 1990; Hornstein and Seliger 1998; Silver et al. 1997). Some TBI patients may be more abulic (lacking spontaneity) and akinetic than somnolent. A number of reports indicate that this state responds favorably to L-dopa/carbidopa (Sinemet) or dopamine agonists such as bromocriptine. The dosage of administered Sinemet is 10/100–25/250 mg qid. Bromocriptine, a dopamine receptor agonist at midrange doses, shows considerable promise for improving frontal lobe function but is of unproven efficacy. It can be initiated at a dose of 2.5 mg tid with cautious weekly small incremental increases up to a maximum dose of 30–60 mg/day. Nausea and orthostatic hypotension may occur early in treatment. All dopaminergic agents can produce agitation, confusion, hallucinations, and nightmares. Gradual withdrawal is recommended, because rapid dosage reduction can result in significant adverse effects (Mysiw and Sandel 1997; Silver et al. 1997). In moderate doses (100–200 mg/day), the mixed dopamine agonist amantadine is not sedating and promotes arousal, attention, and motor performance. It may reduce fatigue. The initial dose of amantadine should be 50 mg bid with 100 mg/week increases as tolerated. Reported side effects, such as seizures and paradoxical excitement, are rare; however, the disturbed sleep-wake cycle often associated with amantadine may result in excessive daytime sleep, confusion, or fatigue. Adequate doses of a sedative should be administered at bedtime to avoid the necessity of repeated doses, which can produce oversedation and hangovers (Alexander 1989a). Neuroleptics such as haloperidol, or novel antipsychotics, are most commonly prescribed for acute agitation not responsive to behavioral measures. These drugs provide reliable short-term sedation but are less effective long term. In addition, antipsychotics may cause a paradoxical increase in agitation. To prevent undesirable side effects (rigidity, akathisia, autonomic instability, neuroleptic malignant syndrome, and probable retardation of cognitive recovery), antipsychotics should probably be restricted to patients with delusions or hallucinations and given at the lowest possible dose for the shortest possible time. In one nonrandomized and noncontrolled study, ❉

307 ❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

brain-injured patients who received haloperidol and patients who did not had similar outcomes (Mysiw and Sandel 1997). Silver and colleagues (1997) recommend starting with haloperidol at 1 mg orally or 0.5 mg im, with repeated administration every hour for several hours until acute aggression is controlled. Should this prove ineffective, the hourly dose is increased until the patient is so sedated that he or she no longer exhibits agitation. After 48 hours of control, the daily dosage should be gradually decreased by 25% per day to determine if aggressive behavior reemerges. If so, options include an increased dose of haloperidol and/or initiation of treatment with a drug more specific for maintenance of behavioral control. (See treatment of chronic aggression later in this chapter.) Newer antipsychotics such as risperidone, clozapine, olanzapine, and quetiapine may be effective in the refractory patient. With these agents, there may be fewer extrapyramidal side effects, but some patients experience sedation, headaches, dizziness, and gastrointestinal symptoms. Clozapine increases the risk of seizures (Silver et al. 1997) and agranulocytosis. Clozapine-treated patients must have weekly white blood cell counts for the first 6 months, with white blood cell counts every other week thereafter. Warden (1997) suggests that benzodiazepines and anticholinergic antidepressants should generally be avoided in TBI patients. (Anticholinergic agents are the most common class of drugs inducing delirium.) The reported dyscontrol syndrome induced by benzodiazepines is rare, but preexisting memory, coordination, and balance impairment can be exacerbated by benzodiazepines. Nevertheless, benzodiazepines such as lorazepam are indicated when rapid control of violent behavior is required or when immediate safety is a concern. Medication for treatment of posttraumatic delirium has principally been targeted toward the dopaminergic and γ-aminobutyric acid systems. Recent findings of low 5-hydroxyindoleacetic acid in patients during PTA suggest that agents affecting central serotonin, such as trazodone in the 25–400 mg/ day range, may more effectively decrease agitation and normalize sleep-wake cycles in posttraumatic delirium (Cassidy 1990; Mysiw and Sandel 1997). For treatment of acute aggression, Silver and colleagues (1997) recommend that lorazepam l–2 mg be administered every hour orally or intramuscularly until sedation is achieved. Intravenous lorazepam is also effective but should be given in doses less than 1 cc (1 mg)/minute to avoid laryngospasm. After 48 hours of control, gradual tapering as with neuroleptics is advised. Oral or parenteral lorazepam in 1 mg or 2 mg doses may be given in combination with 2–5 mg of halperidol—an excellent first-line approach ❉

308 ❉

❉

The Difficult-to-Treat Patient With Dementia or Traumatic Brain Injury ❉

that is often more effective than the use of either agent alone. Other alternatives to neuroleptics include carbamazepine, valproic acid, lithium, verapamil, naltrexone, or a β-blocker. Overall, none of these has worked reliably or safely enough to be recommended for treatment of agitation or aggression in the subacute recovery period. Carbamazepine, however, has been used effectively in treating acute agitation, even in patients without electroencephalogram abnormalities (Chatham-Showalter 1996). No more than 40% of patients respond well to these six agents, but good results may be seen in individual patients when these drugs are prescribed in judicious doses and monitored closely (Cope 1990). With clearing of confusion and resolution of PTA, post-TBI patients generally become increasingly aware of their physical deficits and achieve a measure of independence in mobility and activities of daily living (ADL), which permits discharge from inpatient rehabilitation facilities. Their insight into co-existing behavioral and cognitive deficits with their imposed social and vocational limitations is, however, often severely compromised. Recovery of cognitive abilities, goal-directed behavior, and social skills may be incomplete and prolonged, up to 10 years in some cases of severe TBI (Katz and Alexander 1990).

Psychosocial Treatment of TBI in Later Stages of Recovery In patients with moderate to severe TBI, disabling and often persistent cognitive and behavioral problems develop or become apparent months or even a year posttrauma. These problems probably reflect unique pathophysiologic processes set in motion at the time of the injury. They are often insidious in onset, frequently overlooked, and untreated (Gualtieri 1991a). Impaired cognitive processes include attention, information processing, memory, and language. Aggression, psychosis, depression, mania, and anxiety disorders comprise significant late neurobehavioral complications. Nonpharmacological cognitive and behavioral measures in the management of these disorders are primarily reviewed in this section, to be followed by specific recommendations regarding the common drug treatment of both chronic TBI and dementia. Cognitive Disorders The role of medication and behavioral modification in treating attentional and memory disorders is unclear. Stimulants have been used to increase alertness and reduce distractibility. TBI frequently results in combined anter❉

309 ❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

ograde and retrograde memory deficits, which shrink over time. Recent memory and new learning on both verbal and visual tasks frequently remain impaired, despite adequate recovery of intellectual ability following moderate or severe closed head injury (Brooks 1990; Levin 1985; Levin and Goldstein 1989). Memory augmenting drugs, drills, and computer-driven exercises are of questionable therapeutic value, but diaries and other compensatory measures may be useful (Whyte 1988). Aggression TBI patients frequently exhibit an excess of emotional expression in a continuum from mild irritability to uncontrolled rage attacks. There also may be sudden mood changes (emotional lability) and unprovoked crying or laughing (emotional incontinence). After aphasia, such disturbances, in particular irritability and aggression, are the most common manifestation of temporal lobe injury with a 70% incidence in the first year. Irritability and loss of temper often persists for 10–15 years after severe injury (Gualtieri 1991a). Silver and colleagues (1997) summarize the specific critical features of organic aggressive syndrome following TBI, which allow its differentiation from longstanding personality and related psychiatric disorders. The persistent or recurrent outbursts, whether verbal or physical, are disproportionate to the precipitating stress or provocation. Their build-up is explosive and triggered by trivial or modest stimuli. The aggression is nonpurposeful and periodic; brief outbursts are separated by long calm periods. Furthermore, patients are upset and embarrassed, and do not blame others or try to justify their behavior (Silver et al. 1997). Any underlying medical illness or associated epileptic, attentional, cognitive, and affective disorder should be identified and treated before the use of specific measures for aggression. A method for documenting aggressive fluctuations over time in response to interventions is mandatory. The Overt Aggression Scale developed by Yudofsky and colleagues (Yudofsky et al. 1986) has proven validity and reliability (Silver et al. 1997). Behavioral therapy should be the first line of treatment. Provision of structure and regularity in all activities; a calm, nonthreatening environment; and alternation of activity and rest periods are helpful measures. Appropriate behaviors should be positively reinforced with credits or tokens; direct or indirect reinforcement of inappropriate behaviors should be removed. In refractory cases, mechanical restraints should be used cautiously and only for limited periods of time. ❉

310 ❉

❉

The Difficult-to-Treat Patient With Dementia or Traumatic Brain Injury ❉

Mood Disorders Mood disorders are common in the months following TBI. For depression following severe TBI, reported rates vary from 10% to 77%. Patients often appear sad as they describe the loss of their former self, reflecting awareness of cognitive and motor deficits. It is often necessary to differentiate such grief reactions from major depression. TBI patients with depressed mood may not experience the associated somatic symptoms required for the diagnosis of major depression. The injury itself at times may be the direct result of severe depression, leading to substance abuse and risk taking if not suicidal behavior. Depression has been associated with left anterior frontal and basal ganglia lesions (Jorge et al. 1993). Silver and colleagues (1991), furthermore, hypothesized that there may be two types of depressive syndromes after TBI. The first group with reactive neurotransmitter receptor changes are more irritable, exhibit positive symptoms, and respond to antidepressants. The second group have more severe injuries with decreased neurotransmitters, leading to anhedonia, withdrawal, and apathy. They preferentially respond to dopamine agonists such as stimulants or amantidine. Regional damage may mask the expression of affective disturbance complicating the diagnosis of depression. Dysprosodic speech associated with right temporal-parietal lesions may limit the expression of the expected affect (Ross and Rush 1981). Without the appearance of depression, the examiner may overlook neurovegetative symptoms. Aphasic patients cannot provide important historical information. Depression often must be inferred from poor cooperation with rehabilitation, an apparent deterioration of a previously stable neurological deficit, or unusually slow recovery. In the absence of pseudobulbar palsy, excessive crying may be a useful clue. Manic and bipolar disorders occur after TBI, although much less frequently than depression. Presenting symptoms include pressured speech, flight of ideas, assaultive behavior, sleeplessness, grandiosity, distractibility, hyperactivity, hypersexuality, and impaired insight and judgment. Generally irritability is more common than euphoric mood (Cassidy 1990). Starkstein and colleagues (1990, 1991) reported a series of patients who developed secondary mania after brain insults of varying etiology with bilateral pathology or lesions in the right orbitalfrontal and related limbic association areas. In a number of patients, positron emission tomography studies revealed hypometabolism in the right basotemporal region. Valproic acid is the agent of first choice for treating this condition. ❉

311 ❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

Psychosis Late-onset psychosis can be a severe and devastating outcome of TBI. Psychosis may develop in patients with a clear sensonium within months after recovery from coma but occasionally can erupt as late as 3 years after injury. There is an estimated prevalence rate for delayed onset psychosis as high as 5% following moderate and 10% after severe TBI. The schizophrenic-like symptoms after TBI may be indistinguishable from those of the naturally occurring psychosis. Up to 15% of schizophrenic inpatients have a history of previous brain injury (Silver et al. 1991). Novel antipsychotics are the first line therapy for this dysfunction. Posttraumatic epilepsy, especially of frontal-temporal lobe origin, is associated with psychosis in about 7% of cases. The psychosis can be brief or persistent, is most commonly paranoid in content, and often occurs in a state of clear consciousness. Atypical features of psychosis, characteristically seen in epilepsy, include precipitous onset, confusion and agitation, rapid fluctuations in mood with intense affect, and an intermittent course with symptom-free intervals. Suicidal ideation and suicide attempts are a major management risk. The diagnosis of epileptic psychosis should be considered when there is a fluctuating level of consciousness and when events occurred in close approximation to a well-defined epileptic attack. Patients often are frequently unaware of having seizures. Unusual spells of variable duration (minutes to days) and consisting of uncharacteristic behavior, accompanied by marked lability of affect, depression, mania, delusions, and hallucinations, may occur in repetitive stereotyped episodes. Whether these attacks are real seizures is unclear, but anticonvulsive treatment with carbamazepine, phenytoin, and valproate has been effective (Neppe and Tucker 1992; Trimble 1985). Anxiety Severe anxiety and panic disorders, persisting up to 5 years, have been reported after TBI. Estimates of the incidence of anxiety within 1–2 years of injury average 26%, generally less than the incidence of depression (25%– 50% range). Anxious depressed patients have significantly longer depressive episodes than patients with depression alone. TBI patients are also at increased risk of developing posttraumatic stress disorder with typical symptoms of social withdrawal, flashbacks, and autonomic hyperactivity (Goldstein and Levin 1989). Rare cases of posttraumatic obsessive-compulsive disorder have also been described (Drummond and Gravestock 1988; ❉

312 ❉

❉

The Difficult-to-Treat Patient With Dementia or Traumatic Brain Injury ❉

Jenike and Brandon 1988). Treatment of choice for anxiety-related disorders is lorazepam. Patients intolerant of or unresponsive to this agent may benefit from buspirone or propranolol.

Psychopharmacological Treatment of Behavioral Disturbances in AD and Chronic TBI The noncholinergic psychopharmacological management of neuropsychiatric syndromes in patients with AD and chronic TBI is syndrome specific rather than disease specific. Once disease-specific strategies (see previous sections) have been exhausted, psychopharmacological intervention addresses behavioral syndromes, and similar strategies are used regardless of etiology. The following section presents the management of common behavioral disturbances in TBI and AD.

Agitation A consensus guideline for the treatment of agitation in older persons with dementia provides useful treatment strategies (Alexopoulos et al. 1998). One hundred leading experts on agitation were asked to evaluate a questionnaire based on 33 clinical situations. Pharmacological agents available for use in dementia and TBI patients are summarized in Table 11–1. Treatments of choice for specific presentations of agitation are summarized in Table 11– 2 and Figure 11–1. A combination of an antipsychotic and an anticonvulsant may be beneficial in more refractory cases.

Psychosis Treatment of delusions includes conventional neuroleptic agents as well as other classes of psychopharmacological drugs. Effective therapy may decrease aggression, reduce agitation, and improve the quality of life of the patient and caregiver (Wright and Cummings 1996). Low doses of highpotency agents such as haloperidol usually produce a satisfactory response, although extrapyramidal symptoms are common side effects. In a comparative meta-analysis of literature concerning the use of neuroleptics in dementia, Schneider and colleagues (1990) found that these agents provided modest benefit over placebo but found no specific neuroleptic agent demonstrating superiority in ameliorating behavioral disturbances in dementia. Several new antipsychotic agents have emerged recently and may prove ❉

313 ❉

❉

Figure 11–1.

The Difficult-to-Treat Psychiatric Patient

❉

Treatment algorithm for agitation.

beneficial in treating psychosis in dementia patients. These new agents have not been used extensively in cognitively impaired patients, and their utility, the appropriate dose range, and the expected side effects in demented patients are under investigation. These drugs have complex receptor activity profiles and are reported to achieve antipsychotic effects through effects on serotonergic as well as dopaminergic receptors. Table 11–3 lists the major ❉

314 ❉

❉

The Difficult-to-Treat Patient With Dementia or Traumatic Brain Injury ❉

Table 11–1.

Drugs used to treat agitation

Class

Agent (trade name)

Usual daily dose mg (range)

Neuroleptics and novel antipsychotics

Haloperidol (Haldol)

1 (0.5–3)

Fluphenazine (Prolixin)

1 (1–5)

Thioridazine (Mellaril)

25 (10–100)

Thiothixene (Navane)

2 (1–10)

Molindone (Moban)

75 (5–125)

Risperidone (Risperdal)

1 (0.5–6)

Olanzapine (Zyprexa)

10 (5–15)

Clozapine (Clozaril)

50 (12.5–100)

Quetiapine (Seroquel)

200 (200–400)

Ziprasidone (Geodon)

10 (10–20)

Trazodone

100 (100–400)

Buspirone (Buspar)

12 (15–30)

Divalproex (Depakote)

425 (250–3000)

Carbamazepine (Tegretol)

400 (200–1200)

Propranolol (Inderal)

120 (80–240)

Lorazepam (Ativan)

1 (0.5–6)

Tacrine (Cognex) (AD only)

160 (80–160)

Donepezil (Aricept) (AD only)

10 (5–10)

Nonneuroleptic

Rivastigmine (Exelon) (AD only) 9 (6–12) Galantamine (Reminyl) (AD only) Note.

24 (16–24)

AD=Alzheimer’s disease.

antipsychotic agents to be considered for use in patients with dementia. In our experience, clozapine—while requiring substantial monitoring—is often particularly efficacious in this setting.

Anxiety Benzodiazepines are not the best initial choice for patients with dementia and anxiety and may cause confusion, disorientation, sedation, disinhibi❉

315 ❉

Table 11–2.

Pharmacological intervention for specific presentations of agitation

Antidepressant alone (sertraline or paroxetine) Antidepressant plus antipsychotic or electroconvulsive therapy Benzodiazepine (lorazepam; consider oxazepam) Trazodone, consider benzodiazepine (zolpidem, lorazepam) Trazodone, consider conventional high potency antipsychotic, risperidone, olanzapine

Same as acute therapy Same as acute therapy

Aggression or anger Severe

Conventional high-potency antipsychotic (risperidone)

Mild Nonpsychiatric

Trazodone Antipsychotic

Divalproex, risperidone, olanzapine, or conventional high potency antipsychotic Divalproex, SSRIs, trazodone, buspirone Anticonvulsants (carbamazepine, divalproex)

Depression Without psychosis With psychosis

❉ 316

Anxiety Insomnia

❉

“Sundowning”

Note.

SSRIs=selective serotonin reuptake inhibitors.

Buspirone Trazodone Trazodone, consider risperidone, olanzapine, or conventional high-potency antipsychotic

❉

Long-term intervention Not applicable Risperidone, olazapine, quetiapine, ziprasidone, conventional high-potency antipsychotic

The Difficult-to-Treat Psychiatric Patient

Acute intervention Conventional high-potency antipsychotic Conventional high-potency antipsychotic

❉

Behavior Delirium Psychosis

❉

The Difficult-to-Treat Patient With Dementia or Traumatic Brain Injury ❉

Table 11–3.

Antipsychotic agents commonly used in the treatment of delusions Usual oral dose mg (range)

Agent (trade name) Conventional neuroleptics Haloperidol (Haldol)

1 (0.5–3)

Thiothixene (Navane)

2 (1–10)

Molindone (Moban)

15 (5–40)

Atypical neuroleptics Clozapine (Clozaril)

50 (12.5–100)

Risperidone (Risperdal)

1 (0.5–2)

Olanzapine (Zyprexa)

10 (5–10)

Quetiapine (Seroquel)

50 (12.5–200)

Ziprasidone (Geodon)

10 (10–20)

tion, and worsening memory loss (Wright and Cummings 1996). Buspirone has anxiolytic effects and fewer cognitive side effects. If choosing a benzodiazepine, use shorter acting agents metabolized through conjugation with no or few active metabolites, such as lorazepam or oxazepam. Table 11–4 shows the drugs commonly used to treat anxiety in dementia patients.

Depression Selective serotonin reuptake inhibitors (SSRIs) are an excellent first-line therapy for depression in dementia due to their favorable side-effect profile (Wright and Cummings 1996). Of the available SSRIs, fluoxetine tends to be activating and therefore should be given in the morning. Paroxetine tends to be more sedating and may be more useful in depressed dementia patients who are anxious. However, all the SSRIs can cause anxiety and agitation— often temporary—during the first few weeks after starting the medication. Table 11–5 shows the available SSRIs and other commonly used antidepressants. For depressed, demented patients with insomnia, a sedating antidepressant such as trazodone or nefazodone may be useful. However, the clinician should keep in mind that the insomnia may resolve when the depression is adequately treated. If a tricyclic antidepressant is used, those with fewer anticholinergic side effects are preferable, such as nortriptyline or desipramine. ❉

317 ❉

❉ Table 11–4.

The Difficult-to-Treat Psychiatric Patient

Drugs used to treat anxiety in dementia

Agent (trade name)

Usual daily dose mg (range)

Oxazepam (Serax)

30 (20–60)

Lorazepam (Ativan)

1 (0.5–6)

Buspirone (Buspar)

30 (15–45)

Propranolol (Inderal)

120 (80–240)

Table 11–5.

❉

Antidepressant agents used in dementia patients

Agent (trade name)

Usual daily dose, mg

Fluoxetine (Prozac)

10–40

Citalopram (Celexa)

30

Paroxetine (Paxil)

10–50

Sertraline (Zoloft)

50–200

Fluvoxamine (Luvox)

50–300

Buproprion (Wellbutrin)

200–450

Trazodone (Desyrel)

150–400

Venlafaxine (Effexor)

75–225

Nefazadone (Serzone)

200–500

Mirtazepine (Remeron)

15–45

Nortriptyline (Pamelor)

10–100

Desipramine (Norpramin)

50–300

For more treatment-resistant depression, conventional tricyclic agents (e.g., nortriptyline) are often most useful.

Apathy Apathy can be a disabling state in patients with AD or TBI. Cholinesterase inhibitors reduce apathy in AD, whereas psychostimulants such as methylphenidate are most useful in posttraumatic apathetic states.

❉

318 ❉

❉

The Difficult-to-Treat Patient With Dementia or Traumatic Brain Injury ❉

Multiple Symptoms and Combined Therapy Patients with dementia secondary to AD or TBI often have several symptoms simultaneously. In such cases, the most emergent and disabling symptoms should be treated first. When these have been adequately ameliorated, combined pharmacotherapy may be considered for less emergent and/or residual symptoms. However, the likelihood of drug interactions and toxicity increases with polypharmacy. Similarly, treatment-resistant patients with multiple symptoms may require combined pharmacotherapy for specific symptoms. For example, an antipsychotic may reduce delusions without sufficient amelioration of agitation. Such patients may benefit from combination therapy of an antipsychotic with an anti-agitation agent such as carbamazepine or valproate. Combinations of antipsychotics and antidepressants, or of antipsychotics and anxiolytics, may also be beneficial in selected cases. Clinical Vignette Ms. S, a 76-year old woman presented with a 3-year history of gradually progressive memory loss, word-finding difficulties, and impaired ability to find her way in her own neighborhood. She was brought in for clinical assessment after calling her son and reporting that neighbors were stealing food from her kitchen. The patient’s son reported that his mother had developed a paranoid delusion regarding the next door neighbors and also was suspicious that he was not her “real” son (Capgras delusion). Evaluation revealed an elderly woman who was cooperative but without insight into her difficulties. Mini-Mental State Examination (MMSE) score was 17, and extended mental status examination revealed intact attention, poor word list learning and recall, mild anomia, difficulty copying constructions, impaired calculation, and impaired abstraction of proverbs. Neurological examination was unremarkable. Thyroid functions, serum B12 level, and syphilis serology were normal, and MRI revealed moderate brain atrophy most severely affecting the temporal lobes. The patient was treated with vitamin E (2,000 IU/day) and donepezil (5 mg/day). After one month of therapy, she was generally brighter but continued to have frequent periods of delusional beliefs and agitated behavior. Treatment with olanzapine (2.5 mg/day) was initiated and the dose increased to 5 mg/day after 1 week. On this regimen her psychosis was substantially ameliorated. Cognitive function as measured by the MMSE was unchanged.

This case illustrates the use of standard therapy for AD (vitamin E and donepezil), an observation period to determine the psychotropic effects of ❉

319 ❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

cholinesterase inhibition therapy, and combination treatment with a novel antipsychotic and a cholinesterase inhibitor to control behavioral disturbances.

Summary AD and TBI are frequently accompanied by disabling behavioral disturbances. Nonpharmacological interventions should be optimized in these patients but often fall short of producing an acceptable response. AD-specific or TBIspecific interventions may resolve the behavioral disturbances but, more commonly, patients may require combinations of disease-specific and conventional psychotropic medications. Agitation, psychosis, depression, and anxiety can usually be treated by investigating the effects of different classes of medication sequentially, using combinations of agents creatively, and working closely with the caregiver both to reduce caregiver distress and to limit environmental factors that may provoke behavioral disturbances.

References Alexander MP: Diagnosis and long-term management of severe traumatic brain injury, in Head Injury, Course Syllabus #420. Presented at the annual meeting of the American Academy of Neurology Annual, Chicago, IL, April 1989a Alexander MP: Some neurobehavioral aspects of closed head injury, in Neurology and Psychiatry: A Meeting of Minds. Edited by Mueller J. Basel, Germany, Karger, 1989b, pp 175–191 Alexopoulos GS, Silver JM, Kahn DA, et al: Treatment for agitation in older persons with dementia. A Postgraduate Medicine Special Report, Post Graduate Medicine, April 1998, pp 1–85 American Psychiatric Association: Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition. Washington, DC, American Psychiatric Association, 1994 Brooks DN. Cognitive deficits, in Rehabilitation of the Adult and Child With Traumatic Brain Injury, 2nd Edition. Edited by Rosenthal M, Griffith ER, Bond MR, et al. Philadelphia, PA, FA Davis, 1990, pp 163–178 Cassidy JW: Pharmacological treatment of post-traumatic behavioral disorders: aggression and disorders of mood, in Neurobehavioral Sequelae of Traumatic Brain Injury. Edited by Wood RLI. Mahwah, NJ, Lawrence Erlbaum, 1990, pp 227–229 Chatham-Showalter PE: Carbamazepine for combativeness in acute traumatic brain injury. J Neuropsychiatry Clin Neurosci 8:96–99, 1996

❉

320 ❉

❉

The Difficult-to-Treat Patient With Dementia or Traumatic Brain Injury ❉

Cope DN: Pharmacology for behavioral deficits: disorders of cognition and affect, in Neurobehavioral Sequelae of Traumatic Brain Injury. Edited by Wood RLI. Mahwah, NJ, Lawrence Erlbaum, 1990, pp 248–274 Craig AH, Cummings JL, Fairbanks L, et al: Cerebral blood flow correlates of apathy in Alzheimer’s disease. Arch Neurol 53:1116–1120, 1996 Cummings JL: Behavioral disorders associated with frontal lobe injury, in Clinical Neuropsychiatry. Edited by Cummings JL. Orlando, FL, Grune & Stratton, 1985, pp 57–67 Cummings JL: Cholinesterase inhibitors: a new class of psychotropic compounds. Am J Psychiatry 157:4–15, 2000 Cummings JL, Victoroff JI: Noncognitive neuropsychiatric syndromes in Alzheimer’s disease. Neuropsychiatry Neuropsychol Behav Neurol 3:140–158, 1990 Cummings JL, Mega M, Gray K, et al: The Neuropsychiatric Inventory: comprehensive assessment of psychopathology in dementia. Neurology 44:2308–2314, 1994 Cummings JL, Ross W, Absher J, et al: Depressive symptoms in Alzheimer’s disease: assessment and determinants. Alzheimer Dis Assoc Disord 9:87–93, 1995 Cummings JL, Vinters HV, Cole GM, et al: Alzheimer’s disease: etiologies, pathophysiology, cognitive reserve, and treatment opportunities. Neurology 5 (suppl 1):S2– S17, 1998a Cummings JL, Cyrus PA, Bieber F, et al: Metrifonate treatment of the cognitive deficits of Alzheimer’s disease. Neurology 50:1214–1221, 1998b Dombovy ML: Traumatic brain injury, in Principles of Neurologic Rehabilitation. Edited by Lazar RB. New York, McGraw-Hill, 1998, pp 79–103 Drevets WC, Rubin EH: Psychotic symptoms and the longitudinal course of senile dementia of the Alzheimer type. Biol Psychiatry 25:39–48, 1989 Drummond LM, Gravestock S: Delayed emergence of obsessive-compulsive neurosis following head injury: case report and review of its theoretical implications. Br J Psychiatry 153:839–842, 1988 Eames P: Organic bases of behavioral disorders after traumatic brain injury, in Neurobehavioral Sequelae of Traumatic Brain Injury. Edited by Wood RLI. Mahwah, NJ, Lawrence Erlbaum, 1990, pp 133–152 Farlow M, Gracon SI, Hershey LA, et al: A controlled trial of tacrine in Alzheimer’s disease. JAMA 68:2523–2529, 1992 Goldstein FC, Levin HS: Manifestations of personality change after closed head injury, in Integrating Theory and Practice in Clinical Neuropsychology. Edited by Pcrecman E. Mahwah, NJ, Lawrence Erlbaum, 1989, pp 217–243 Gorman DG, Read S, Cummings JL: Cholinergic therapy of behavioral disturbances in Alzheimer’s disease. Neuropsychiatr Neuropsychol Behav Neurol 229–234, 1993 Grafman J, Salazar A: Traumatic brain injury, in Neuropsychiatry. Edited by Fogel, BS, Schiffer RB. Baltimore, MD, Williams & Wilkins, 1996, pp 935–946

❉

321 ❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

Gualtieri CT: The delayed neurobehavioral sequelae of traumatic brain injury, in Neuropsychiatry and Behavioral Pharmacology. Edited by Gualtieri CT. New York, Springer-Verlag, 1991a, pp 31–46 Gualtieri CT: The neuropsychiatric sequelae of traumatic brain injury, in Neuropsychiatry and Behavioral Pharmacology. Edited by Gualtieri CT. New York, Springer-Verlag, 1991b, pp 37–88 Hornstein A, Seliger GM: Behavioral and pharmacological management of brain injury, in Principles of Neurologic Rehabilitation. Edited by Lazar RB. New York, McGraw-Hill, 1998, pp 589–595 Jenike MA, Brandon AD: Obsessive-compulsive disorder and head trauma: a rare association. J Anxiety Disord 2:353–359, 1988 Jorge RE, Robinson RG, Starkstein SE, et al: Depression and anxiety following traumatic brain injury. J Neuropsychiatry Clin Neurosci 5:369–374, 1993 Katz DI, Alexander MP: Neurologic diagnosis and treatment planning, in Innovations in Head Injury Rehabilitation. Edited by Deutsch PM, Fralish KB. New York, Bender, 1990, pp 117–131 Kaufer D, Cummings JL, Christine D: Effect of tacrine on behavioral symptoms in Alzheimer’s disease: an open-label study. Journal of Geriatric Psychiatry and Neurology 9:1–6, 1996 Kaufer D, Cummings JL, Christine D: Differential neuropsychiatric symptom responses to tacrine in Alzheimer’s disease: relationship to dementia severity. J Neuropsychiatry Clin Neurosci 10:55–63, 1998 Levin HS: Outcome after head injury, part 11: neurobehaviorul recovery, in Central Nervous System Trauma Report—1985. Edited by Becker DP, Povlishock JT. Washington, DC, National Institute of Neurological and Communicative Disorders and Stroke, National Institutes of Health, 1985, pp 281–302 Levin HS, Eisenberg HM: Management of head injury: neurobehavioral outcome. Neurosurgery Clin N Am 2:457–471, 1991 Levin HS, Goldstein FC: Neurobehavioral aspects of traumatic brain injury, in Traumatic Brain Injury. Edited by Bach Y Rita P. New York, Demos, 1989, pp 53–72 Lilly R, Cummings JL, Benson DF, et al: The human Klüver-Bucy syndrome. Neurology 33:1141–1145, 1983 McKhann G, Drachman DD, Folstein M, et al: Clinical diagnosis of Alzheimer’s disease: report of the NINCDS-ADRDA Work Group under the auspices of the Department of Health and Human Services Task Force on Alzheimer’s Disease. Neurology 34:939–944, 1984 Mega MS, Cummings JL, Fiorello T, et al: The spectrum of behavioral changes in Alzheimer’s disease. Neurology 46:130–135, 1996 Mendez MF, Martin RJ, Smyth KA, et al: Psychiatric symptoms associated with Alzheimer’s disease. J Neuropsychiatry Clin Neurosci 2:28–33, 1990 Merriam AE, Aronson MK, Gaston P, et al: The psychiatric symptoms of Alzheimer’s disease. J Am Geriatr Soc 36:7–12, 1988

❉

322 ❉

❉

The Difficult-to-Treat Patient With Dementia or Traumatic Brain Injury ❉

Morris JC, Cyrus PA, Orazem J, et al: Metrifonate benefits cognitive, behavioral, and global function in patients with Alzheimer’s disease. Neurology 50:1222–1230, 1998 Mysiw JW, Sandel ME: The agitated brain injured patient, part 2: pathophysiology and treatment. Arch Phys Med Rehabil 78:213–220, 1997 Neppe VM, Tucker GJ: Neuropsychiatric aspects of seizure disorders, in Textbook of Neuropsychiatry, 2nd Edition. Edited by Yudofsky SC, Hales RE. Washington, DC, American Psychiatric Press, 1992, pp 397–426 Raskind MA, Sadowsky CH, Sigmund WR, et al: Effect of tacrine on language, praxis, and noncognitive behavioral problems in Alzheimer disease. Arch Neurol 54:836–840, 1997 Reisberg B, Borenstein J, Salob SP, et al: Behavioral symptoms in Alzheimer’s disease: phenomenology and treatment. J Clin Psychiatry 48 (suppl 5):9–15, 1987 Rogers SL, Friedhoff LT, and the Donepezil Study Group: The efficacy and safety of donepezil in patients with Alzheimer’s disease: results of a U.S. multicentre, randomized, double-blind, placebo-controlled trial. Dementia 7:293–303, 1996 Rosenthal M, Christensen BK, Ross TP: Depression following traumatic brain injury. Arch Phys Med Rehabil 79:90–103, 1998 Ross E, Rush J: Diagnosis and neuroanatomical correlates of depression in brain-damaged patients. Arch Gen Psychiatry 38:1344–1354, 1981 Saul RE: Neurobehavioral disorders following traumatic brain injury, part I: neurobehavioral sequelae in the early stages of recovery, in Physical Medicine and Rehabilitation: State of the Art Reviews, Vol 7. Edited by Stone L. Philadelphia, PA, Hanley & Belfus, 1993a, pp 581–591 Saul RE: Neurobehavioral disorders following traumatic brain injury, part II: late neurobehavioral sequelae, in Physical Medicine and Rehabilitation: State of the Art Reviews, Vol 7. Edited by Stone L. Philadelphia, PA, Hanley & Belfus, 1993b, pp 592–602 Schneider LS, Pollock VE, Lyness SA: A meta-analysis of controlled trials of neuroleptic treatment in dementia. J Am Geriatr Soc 38:553–563, 1990 Silver JM, Hales RE, Yudofsky SC: Neuropsychiatric sequelae of traumatic brain injury, in Neurobehavioral Sequelae of Traumatic Brain Injury. Edited by Wood RLI. Mahwah, NJ, Lawrence Erlbaum, 1990, pp 219–247 Silver JM, Yudofsky SC, Hales RE: Depression in traumatic brain injury. Neuropsychiatry, Neuropsychol Behav Neurol 4:12–23, 1991 Silver JM, Yudofsky SC, Hales RE (eds): Neuropsychiatry of Traumatic Brain Injury. Washington, DC, American Psychiatric Press, 1994 Silver JM, Hales RE, Yudofsky SC: Neuropsychiatric aspects of traumatic brain injury, in The American Psychiatric Press Textbook of Neuropsychiatry, 3rd Edition. Edited by Yudofsky SC, Hales RE. Washington, DC, American Psychiatric Press, 1997, pp 521–560

❉

323 ❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

Starkstein SE, Mayberg HS, Berthier ML, et al: Mania after brain injury: neuroradiologic and metabolic findings. Ann Neurol 27:652–659, 1990 Starkstein SE, Fedoroff P, Berthier ML, et al: Manic-depressive and pure manic states after brain lesions. Biol Psychiatry 29:149–158, 1991 Stuss DT: Contributions of frontal lobe injury to cognitive impairment after closed head injury: methods of assessment and recent findings, in Neurobehavioral Recovery From Head Injury. Edited by Levin HS, Grafman J, Eisenberg HM. New York, Oxford, 1987, pp 166–177 Sultzer DL, Levin HS, Mahler ME, et al: A comparison of psychiatric symptoms in vascular dementia and Alzheimer’s disease. Am J Psychiatry 150:1806–1812, 1993 Swearer JM, Drachman DA, O’Donnell BF, et al: Troublesome and disruptive behaviors in dementia: Relationships to diagnosis and disease severity. J Am Geriatr Soc 36:784–790, 1988 Tariot PN, Solomon PR, Morris JC, et al: A 5-month randomized, placebo-controlled trial of galantamine in AD. Neurology 54:2269–2276, 2000 Teri L, Borson S, Kiyak HA, et al: Behavioral disturbance, cognitive dysfunction, and functional skill. Prevalence and relationship in Alzheimer’s disease. J Am Geriatr Soc 37:109–116, 1989 Trimble MR: The psychoses of epilepsy and their treatment, in The Psychopharmacology of Epilepsy. Edited by Trimble MR. Chichester, England, Wiley, 1985, pp 83–94 Warden DL: Neuropsychiatry/neurobehavior of TBI, in Traumatic Brain Injury, Course Syllabus #425. Presented at the annual meeting of the American Academy of Neurology, Boston, MA, 1997, pp 49–63 Whyte J, Rosenthal M: Rehabilitation of the patient with head injury, in Rehabilitation Medicine. Edited by Delisa JA. Philadelphia, PA, JB Lippincott, 1988, pp 599–611 Wragg RE, Jeste DV: Overview of depression and psychosis in Alzheimer’s disease. Am J Psychiatry 146:577–587, 1989 Wright MT, Cummings JL: Neuropsychiatric disturbances in Alzheimer’s disease and other dementias: recognition and management. Neurologist 2:207–218, 1996 Yudofsky SC, Silver JM, Jackson W, et al: The Overt Aggression Scale for the objective rating of verbal and physical aggression. Am J Psychiatry 143:35–39, 1986

❉

324 ❉

12 The Difficult-to-Treat Psychiatric Patient With Comorbid Medical Illness David S. Harnett, M.D.

Comorbid medical illness presents major diagnostic and treatment challenges in the treatment of refractory psychiatric illness. The outcome of major depression, for example, may be less favorable in the presence of significant medical illness (Keitner et al. 1991). At the same time, recent research has increasingly highlighted the cost and clinical consequences of psychiatric illness, especially depression, on the course of medical illness (Unutzer et al. 1997). This includes increased ambulatory medical utilization, longer medical inpatient length of stay, increased readmission rate, and greater funcThanks to Kathleen Fabiano (Secretary) and Terri Niland (Librarian) for their efforts in the preparation of this manuscript.

❉

325 ❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

tional disability (Penninx et al. 1998). It is this author’s belief that for the difficult-to-treat psychiatric patient with comorbid medical illness, accurate diagnosis and effective treatment will be enhanced by familiarity with specific comorbid medical conditions and their relationship to psychiatric illness. Therefore, in this chapter, I will focus on the difficult-to-treat psychiatric patient with three common comorbid conditions: cardiovascular disease, diabetes mellitus, and respiratory disease (especially chronic obstructive pulmonary disease). These choices reflect my long experience in a general community hospital setting of largely geriatric patients. Space limitations preclude discussion of other medical comorbidities, such as cancer and HIV illness (see Holland 1998); endocrine, gastrointestinal, renal, and rheumatological disease; organ transplantation; and pregnancy (see Robinson and Yates 1999; Rundell and Wise 1996; Stoudemire et al. 2000).

The Difficult-to-Treat Psychiatric Patient With Comorbid Cardiovascular Disease Epidemiology The difficult-to-treat psychiatric patient with comorbid cardiovascular disease has been a subject of increasing interest to psychiatrists (Glassman and Shapiro 1998). Epidemiological data are sometimes hard to interpret because cardiovascular symptoms may be confused with those of anxiety (Miniati et al. 1998) and depression (Harnett 1994). Moreover, the course of cardiovascular illness is influenced by these same emotional factors (Glassman and Shapiro 1998). In any case, research suggests that depression is associated with the development of ischemic heart disease (Glassman and Shapiro 1998; Penninx et al. 2000) and that for those with preexisting cardiovascular disease, depression is an independent risk factor for increased cardiac mortality (Frasure-Smith et al. 1993, 1995; Glassman and Shapiro 1998). Major depression is seen in approximately 15%–20% of coronary artery disease (CAD) patients (Musselman et al. 1998) and is more common in elderly congestive heart failure patients (36.5%) (Koenig 1998). However, depression may occasionally be overdiagnosed if lethargy (Harnett 1994) or apathy (Krupp and Fogel 1997) secondary to congestive heart failure is inappropriately “counted” as major depression. In my experience, mild overdiuresis and relative hypotension secondary to excessive cardiac medication are potentially reversible causes of lethargy and cognitive impairment in such patients. There is also a danger in overdiagnosing depression in ❉

326 ❉

❉ The Difficult-to-Treat Psychiatric Patient With Comorbid Medical Illness ❉ patients who become demoralized, though not depressed, about their numerous medical setbacks. Such demoralized patients might be described as having “minor depression” by some researchers.

Pathophysiology The pathophysiology linking depression, ischemic heart disease, and increased cardiac mortality is not well understood. Enhanced platelet activation (which may promote coronary artery occlusion) has been found in depressed medically well young adults versus nondepressed control subjects (Musselman et al. 1996) and in depressed cardiac patients versus both healthy control subjects and nondepressed cardiac patients (LaghrissiThode et al. 1997). Autonomic nervous system dysfunction, as reflected in reduced heart rate variability, is seen in CAD and congestive heart failure (Dalack and Roose 1990) and may contribute to arrhythmia-related sudden death. Compared to nondepressed CAD control subjects, depressed CAD patients show a greater reduction in heart rate variability (Krittayaphong et al. 1997), although some earlier reports differ (Musselman et al. 1998). Finally, decreased adherence to suggested lifestyle changes and medical regimens among depressed patients (e.g., dropping out of exercise programs, not complying with low-dose aspirin) may contribute to cardiac morbidity (O’Connor 1998).

Comorbid Cardiovascular Disease: Psychopharmacological Treatment Tricyclic Antidepressants Although tricyclic antidepressants (TCAs) are generally safe in the medically well, concerns persist regarding TCA side effects in patients with comorbid cardiovascular disease. Preexisting congestive heart failure (Glassman et al. 1983), cardiac conduction disease (Roose et al. 1987), and concurrent use of cardiac medications that may reduce blood pressure all increase the incidence of TCA-induced orthostatic hypotension. Nortriptyline may be the TCA least likely to cause orthostasis (Roose et al. 1986), although this advantage has been rigorously demonstrated only against imipramine. Cautious use of midodrine (Medical Letter 1997), an α1 agonist, may lessen TCA-related orthostatic hypotension (Biol Therapies Psychiatry 1998). TCAs have effects similar to the class 1A antiarrhythmics (e.g., quinidine), including suppression of some arrhythmias and slowing of cardiac conduction. Thus, patients with preexisting bundle-branch block (QRS duration on ❉

327 ❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

electrocardiogram>0.11 seconds) given TCAs may develop high-grade atrioventricular block (Roose et al. 1987). The Cardiac Arrhythmia Suppression Trial (CAST; CAST Investigators 1989)—which found that class 1C and 1A antiarrhythmics were associated with increased mortality—has raised further concerns about the use of TCAs in patients with myocardial ischemia (Glassman et al. 1993). An ischemic myocardium may promote antiarrhythmic-induced ventricular fibrillation and sudden death (Glassman et al. 1998). Furthermore, anticholinergic medications, including TCAs, may increase heart rate and diminish heart rate variability, thus increasing arrhythmia risk (Glassman 1998a). TCAs may well be more efficacious than selective serotonin reuptake inhibitors (SSRIs) in some depressed patients, especially in those with severe symptoms. However, this must be balanced against possibly dangerous and uncomfortable side effects. In short, TCAs should usually be avoided in depressed cardiac patients. Nontricyclic Antidepressants The SSRI antidepressants appear to be the safest, best-studied agents for use with patients with cardiac disease (Barbey and Roose 1998) and are generally the medications of first choice for comorbid depression. In medically well depressed patients, the only common cardiovascular effect of SSRIs is slightly reduced heart rate (Fisch 1985). Severe bradycardia and syncope are rarely seen (Favre et al. 1999) but may be slightly more common in the presence of a β-blocker (Walley et al. 1993). There are occasional reports of SSRIs increasing blood pressure in patients with vasovagal syncope and episodic orthostatic hypotension (Amsterdam et al. 1999), and there is one reported case of paroxetine-related orthostatic hypotension with rechallenge (Andrews and Pinner 1998). There are also rare case reports of atrial fibrillation related to fluoxetine in medically ill patients (Buff et al. 1991). Clinically significant SSRI-related vasoconstriction (especially of damaged vessels) has been theorized but seems to be uncommon in clinical practice (Skop and Brown 1996). Fatal SSRI overdoses appear to be rare (Glassman 1998a), and the most common serious complication after very large SSRI overdose is seizure, not cardiotoxicity (Barbey and Roose 1998). Studies of SSRIs in depressed cardiac patients have generally been favorable. In 27 depressed patients with serious stable cardiac disease (ejection fraction ≤ 50% or ≥ 10 premature ventricular contractions per hour or QRS ≥ 10 seconds or combination), 7 weeks of fluoxetine (average dose 50 mg/ day at week 6) was well-tolerated. Heart rate decreased by only 6%, and there were no adverse effects on cardiac conduction, ventricular rhythm, or ❉

328 ❉

❉ The Difficult-to-Treat Psychiatric Patient With Comorbid Medical Illness ❉ orthostatic blood pressure (Roose et al.1998a). Surprisingly, cardiac ejection fraction improved 7%. A double-blind randomized 6-week trial of paroxetine (versus nortriptyline) in depressed patients with moderate stable ischemic heart disease revealed similarly benign findings (Roose et al. 1998b). In contrast, nortriptyline was associated with an 11% increase in heart rate and diminished heart rate variability. Sertraline has also proved benign in 26 postacute myocardial infarction depressed patients over 16 weeks, with no effect on heart rate (Roose 1998). Finally, preliminary data suggest that SSRIs (vs. TCAs) may actually ameliorate abnormalities in platelet function and heart rate variability associated with psychiatric illness (Glassman 1998a). SSRI-mediated interference with platelet aggregation may explain occasional reports of ecchymoses or bruising, which may respond to vitamin C 500 mg/day (Tielens 1997). Whereas a review of drug-drug interactions involving SSRIs (Greenblatt et al. 1998) is beyond the scope of this chapter, the clinician should always be aware of this possibility. For example, fluvoxamine’s ability to inhibit cytochrome enzymes 2C9 and 1A2 may substantially increase plasma warfarin levels, potentially affecting coagulation function (Preskorn 1996). Inhibitors of CYP-2D6, like fluoxetine or paroxetine, may increase blood levels of some β-blockers (e.g., propranolol, metoprolol) and antiarrhythmic agents (e.g., flecainide, propafenone, mexiletine). Citalopram and sertraline, however, seem to have less inhibitory effect on the main cytochrome systems. Besides the SSRIs, other nontricyclics may be considered in depressed cardiac patients. Buproprion (mean daily dose 442 mg) has a relatively benign cardiac side-effect profile, without significant effects on heart rate, normal rhythm, cardiac conduction, or ejection fraction (Roose et al. 1991). Occasionally, bupropion may cause a modest increase in supine blood pressure and less commonly, orthostatic hypotension. Venlafaxine—which inhibits reuptake of both serotonin and norepinephrine—may cause sustained diastolic hypertension. Whereas this problem is usually limited to doses >200 mg/day (Thase 1998), significant blood pressure increase has also been reported in three older patients at venlafaxine doses ≤ 50 mg/day (Kaltsounis et al. 1998). Mirtazapine and nefazodone have not been systematically studied in cardiac patients, although initial clinical experience suggests a benign cardiac side-effect profile and relative safety in overdose. In our experience, mirtazapine is sometimes useful where TCAs have been previously employed. It is usually mildly sedating and appetite stimulating, yet unlike TCAs has no significant effect on cardiac conduction. Whereas mirtazapine-related ortho❉

329 ❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

static hypotension is unusual, somnolence and dizziness are common (Preskorn 1997) and merit conservative dosing in debilitated cardiac patients. Mirtazapine has also been found to decrease heart rate variability (though less so than imipramine) (Tulen et al. 1996), increase serum cholesterol (vs. amitriptyline and placebo), and increase appetite and weight. Nefazodone’s chief cardiac side effect is orthostatic hypotension. Our limited experience with nefazodone in cardiac patients has not been encouraging, perhaps owing to poor tolerance of adequate doses. There is little controlled research on treatment-refractory depressed patients with cardiac illness. In my experience, bupropion can be used to augment SSRI therapy, although fluoxetine-bupropion interactions (e.g., anxiety) have been reported (Hopkins 1996). Thus, bupropion dosing should be conservative (37.5–75 mg/day, initially). For cardiac patients whose depression is refractory to SSRIs and/or bupropion, the treatment algorithm becomes even less clear. Though data are limited, my colleagues and I sometimes find venlafaxine useful in such treatment-refractory depressions. Blood pressure is carefully monitored, and the recently available longacting preparation seems to be better tolerated. Monoamine oxidase inhibitors (MAOIs) may be useful in atypical or refractory cases of depression (Pies 1998a), but their hypo- and hypertensive effects render them impractical for most depressed cardiac patients In refractory cases, electroconvulsive therapy (ECT) should be seriously considered, with appropriate medical precautions (see Chapter 13 of this book). Although recent myocardial infarction is a relative contraindication for ECT, this life-saving treatment should not be withheld from severely depressed cardiac patients who otherwise would merit ECT. Finally, in addition to treatment of discrete depressive and anxiety disorders in the cardiac patient, serotonergic agents may be considered for treatment of anger and hostility syndromes. For example, buspirone (42 mg/ day) was found helpful and well tolerated in irritable, “Type A” CAD patients without Axis I psychiatric disorder (Littman et al. 1993). Antipsychotics In most patients with psychotic disorders, the new atypical antipsychotic agents have many advantages over standard neuroleptics. Unfortunately, many atypical antipsychotics (clozapine, risperidone, olanzapine, quetiapine, ziprasidone) may cause significant orthostatic hypotension, compared with high-potency neuroleptics such as haloperidol. However, symptomatic orthostatic hypotension may be mitigated by very gradual dosage increase. ❉

330 ❉

❉ The Difficult-to-Treat Psychiatric Patient With Comorbid Medical Illness ❉ Moreover, in my experience, the atypical antipsychotics’ reduced rate of parkinsonian effects leads to an overall reduction in falls versus standard neuroleptics. Sudden death, possibly related to antipsychotic agents, has been a longstanding though poorly documented, concern. A possible link is antipsychotic-related prolonged cardiac repolarization (as reflected in QTc interval prolongation; QTc is the QT interval corrected for heart rate), which in turn is associated with torsades de pointes, a potentially fatal ventricular arrhythmia (Shader and Greenblatt 1998). Torsades de pointes has been reported with thioridazine, pimozide (Krahenbuhl et al. 1995) and haloperidol (Jackson et al. 1993), including high-dose intravenous haloperidol (Metzger and Friedman 1993; Wilt et al. 1993). Concern with QTc prolongation has led to scrutiny of the newer antipsychotics and withdrawal of sertindole from the market, and new “black box label warning” for older conventional antipsychotics thiorizadine and mesoridazine. Thirty-one percent of 61 clozapine patients were found to have ECG abnormalities, generally benign, and QTc prolongation that was dose related (Kang et al. 2000). Newly released ziprasidone appears to prolong the QTc slightly more than the other atypical antipsychotic agents, resulting in a package insert warning not quite warranting the black box. The clinical significance of this difference remains to be seen. In general, QTc greater than 450 milliseconds merits concern (Labbate and Ayd 2001). Antipsychotic-related QTc prolongation and torsades de pointes are probably quite rare; nevertheless, excessive antipsychotic blood levels should be avoided. Patients at risk include those experiencing renal/ hepatic failure or taking concurrent enzyme inhibitors (e.g., SSRIs that inhibit CYP 2D6 may raise risperidone blood levels). Other conditions that may preclude use of intravenous haloperidol or repolarization-prolonging drugs include hypokalemia, hypothyroidism, and preexisting prolonged QTc interval. Monitoring of the QTc interval should be considered in selected patients receiving antipsychotics. Venous thromboembolism (e.g., venous thrombosis, pulmonary embolism) has been associated with conventional antipsychotics, especially lowpotency agents (Zornberg and Jick 2000) and with clozapine (Hagg et al. 2000; Ihde-Scholl et al. 2001). Myocarditis with eosinophilic infiltrate may be more common with clozapine than previously thought (Kilian et al. 1999). Case reports (Low et al. 1998) have described other major cardiac complications with clozapine (e.g., myocardial infarction, cardiomyopathy, congestive heart failure, and various atrial and ventricular arrhythmias). One report (Low et al. 1998) noted clozapine-related atrial fibrillation and ❉

331 ❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

congestive heart failure that remitted when clozapine was discontinued. Clozapine has also been associated with tachycardia and reduced heart rate variability (Rechlin 1994). Dehydration requires cautious use of antipsychotics because it accentuates medication-related orthostatic hypotension and may increase the risk of neuroleptic malignant syndrome. Mood Stabilizers Lithium-related T-wave changes on the electrocardiogram are common, often clinically insignificant, and usually remit shortly after lithium discontinuation (DasGupta and Jefferson 1990). Sinus node dysfunction (“sick sinus syndrome”), although uncommon, is associated with lithium and may be symptomatic. Associated low heart rate may be remedied by pacemaker insertion if continued lithium therapy is necessary (Shader 1994). Although lithium toxicity has been associated with cardiac conduction disturbance and arrhythmias, it is uncertain if lithium at therapeutic levels is associated with arrhythmias in vulnerable cardiac patients (e.g., after myocardial infarction [MI] or in the presence of advanced congestive heart failure). Lithium has been reported both to improve and worsen congestive heart failure (DasGupta and Jefferson 1990), perhaps reflecting fluctuating lithium levels caused by frequent changes in hydration/electrolyte status and use of highdose diuretics. The potential benefit of stopping lithium after acute MI must be balanced against the possible adverse cardiac effects of acute mania. Dosage reduction, rather than discontinuation, may be advisable. Thiazide diuretics predictably increase lithium levels, whereas potassium-sparing diuretics (e.g., amiloride) and loop diuretics (e.g., furosemide) generally do not affect lithium levels (DasGupta and Jefferson 1990; Shader 1994). In addition, thiazides and amiloride are helpful in decreasing lithium-related polyuria. Angiotensin-converting enzyme inhibitors (e.g., captopril, enalapril, lisinopril) have been reported to increase lithium levels (DasGupta and Jefferson 1990). However this finding is controversial, and I recommend that lithium levels simply be followed closely. When lithium and calcium channel blockers are used concurrently, enhanced neurological or cardiac toxicity is possible (DasGupta and Jefferson 1990). Initial enthusiasm for use of calcium channel blockers (e.g., verapamil, nimodipine) in acute mania has waned (Pies 1998a). The mood-stabilizing anticonvulsants appear reasonably well-tolerated in cardiac patients. Carbamazepine, structurally similar to TCAs, has quinidinelike effects on cardiac conduction and may also provoke syndrome of inappropriate antidiuretic hormone (SIADH)–related hyponatremia. Carbamazepine ❉

332 ❉

❉ The Difficult-to-Treat Psychiatric Patient With Comorbid Medical Illness ❉ is also a prominent enzyme inducer and may lower levels of warfarin. Carbamazepine levels may be increased by the calcium channel blockers diltiazem and verapamil. Recently released oxcarbazepine may be a safer alternative to carbamazepine, although neither agents carry FDA labeling for bipolar disorder. Oxcarbazepine is apparently not associated with bone marrow suppression, hepatoxicity, or Stevens-Johnson Syndrome, and is less likely to cause skin rash. Hyponatremia and lesser enzyme induction of P450-3A4 may still occur (Ghaemi 2001).Valproate has a benign cardiac side-effect profile. It is a mild enzyme-inhibitor and may be associated with increased levels of warfarin (Physician’s Desk Reference 2001). Occasionally, valproate is associated with thrombocytopenia and impairment of platelet function. Gabapentin’s popularity as a mood stabilizer has outstripped formal evidence of its efficacy in bipolar disorder. It is generally safe and well tolerated but can cause ataxia and sedation. Lamotrigine’s side-effect profile is less benign than gabapentin (i.e., occasionally severe rash), but lamotrigine may be useful in bipolar depression (Calabrese et al. 1999). Topiramate—a putative mood stabilizer—may sometimes cause palpitations, bradycardia, or conduction abnormalities (Physician’s Desk Reference 2001). Risk of Falls and Fractures Surprisingly, a recent retrospective study of 2,428 nursing home residents revealed little difference in the rate of falls between those treated with TCAs or with SSRIs (Thapa et al. 1998). Possible explanations include selection bias (e.g., clinicians may preferentially prescribe SSRIs for higher risk patients [Avorn 1998]), disproportionate risk factors among SSRI-treated patients (e.g., ambulatory, not physically restrained), and increased “body sway” with sertraline (Laghrissi-Thode et al. 1995). Notably, for patients taking two or more cardiac medications, the TCA users were more likely to fall than were SSRI users. Detecting and treating osteoporosis and vitamin D deficiency may decrease fall-related fractures (LeBoff et al. 1999). Because, in our experience, many patients admitted to a geriatric medical-psychiatric inpatient unit are vitamin D deficient (D.S. Harnett, unpublished observations, 1998), routine vitamin D supplementation (e.g., 800 IU/day) in vulnerable individuals may be advisable (Compston 1998). Other factors associated with decreased bone mineral density include past or current depression, cigarette smoking, and excessive thyroid or adrenocortical function, elevated prolactin (of the atypical antipsychotics, most commonly seen with risperidone), and anticonvulsant use (Casey 1997; Schweiger et al. 2000). ❉

333 ❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

Psychosocial Intervention in Comorbid Cardiovascular Disease The treatment of emotional disturbance in CAD patient can be problematic. The therapist may need to balance the potentially cathartic benefits of having patients explore their fantasies and fears against the risk of mobilizing too much affect and possibly causing silent myocardial ischemia (Gullette et al. 1997; Jiang et al. 1995). Similarly, the defense mechanism of denial may predict better short-term outcome for acutely hospitalized, unstable angina patients but compromise long-term outcome by promoting noncompliance (Littman and Ketterer 1996). Several recent multifaceted psychosocial intervention studies in CAD patients with and without psychiatric illness have yielded conflicting results. The Lifestyle Heart Trial (Ornish et al. 1990) found that intensive lifestyle changes (vegetarian diet, aerobic exercise, stress management training, smoking cessation, group psychosocial support) without lipid-lowering drugs were associated with decreased coronary stenosis and fewer cardiac events at 1 (Ornish et al. 1990) and 5 (Ornish et al. 1998) years, versus outcome with “usual care.” A 4-month program of stress management was found to be superior to exercise training in reducing ambulatory ischemia and ischemia induced by mental stress in CAD patients (Blumenthal et al. 1997). A meta-analysis of 23 randomized controlled trials suggested that the addition of psychosocial treatments to standard cardiac rehabilitation regimes reduces mortality, morbidity, and emotional symptoms in CAD patients (Linden et al. 1996). In contrast, a multicenter psychological rehabilitation program of 2,328 post-MI patients found that seven weekly group outpatient sessions (involving psychological therapy, counseling, relaxation training, stress management training) did not improve anxiety, depression, or cardiac outcome and yielded little difference from control subjects (Jones and West 1996). Similarly, a nurse-managed, home-based, multifactorial risk-reduction program was found to be no better than control in reducing psychological distress in post-MI patients (Taylor et al. 1997). Surprisingly, a program of monthly telephone monitoring of psychological distress and home nursing visits for 1 year after MI, previously found to improve cardiac outcome in men, was recently found to worsen cardiac outcome in women (Frasure-Smith et al. 1997). The authors speculated that telephone monitoring may have undermined coping mechanisms, such as denial, and that home visits might have highlighted familial or social strain. Given these uncertainties, I recommend cautious use of “uncovering” psychotherapy in depressed cardiac patients. The following vignette typifies my integrated approach to such patients. ❉

334 ❉

❉ The Difficult-to-Treat Psychiatric Patient With Comorbid Medical Illness ❉ Clinical Vignette Ms. C is a 75-year-old, recently widowed woman with a longstanding history of episodic depression and congestive heart failure. She was admitted to a geriatric medical-psychiatric unit because of escalating depression, suicidal ideation, and psychosis. Ms. C was accompanied by her daughter, who, since her father’s death, had temporarily moved in to Ms. C’s home. Treatment history included a remote course of ECT and use of “Elavil” in the distant past. Ms. C was uncertain how helpful these treatments were. Her primary care physician began paroxetine several years ago. This was apparently helpful at first, then “stopped working,” although the patient continued to take it at her husband’s urging. Paroxetine dosage was 30 mg/ day at the time of admission. Ms. C noted that her depression had been worsening over the 6 months since her husband died and complained of early morning awakening with guilty ruminations. Her appetite had diminished and she had begun to neglect her personal care. She stopped watching her “soap operas” on television and rarely left her apartment. She stated that she wished to die because she did not want to be a “burden” on her daughter. Ms. C also complained that her upstairs neighbor was harassing her and “always knows what room I’m in,” adding, “I think she doesn’t like me.” Later, Ms. C talked about her father as if he were still alive, then corrected herself. She also made reference to bilateral shin pain and wondered if she had “bone cancer.” She proudly noted that she has eliminated all dairy products from her diet for several years, fearing that excessive dietary cholesterol could worsen her heart disease. Cognitive testing revealed mild patchy deficits and Mini-Mental State Examination score was 21 out of 30. Laboratory testing was notable for a urinary tract infection, and a 25hydroxyvitamin D level of 5 ng/mL (reference range 9–47 ng/mL) with lownormal serum calcium. Medical consultants found only mild congestive heart failure and determined that the patient’s hypovitaminosis D was likely responsible for her bony discomfort. Diagnostic impression was recurrent major depression. Whether the delusional symptoms were a part of a psychotic depression or a separate “organic” psychosis related to low-grade delirium and/or dementia was not clear. The patient’s cardiac medications were adjusted, her urinary tract infection treated, and her hypovitaminosis D managed with vitamin D and calcium replacement. Paroxetine was replaced with venlafaxine-XR, 75 mg/ day, then 150 mg/day, with trazodone 25 mg added at bedtime to help with sleep. Because of persisting fears of an intruder rummaging through her belongings at night, risperidone 0.5 mg at bedtime was added, with careful monitoring of orthostatic blood pressure change. The dosage was later increased to 1 mg at bedtime. Ms. C also participated in supportive group therapy. She was discharged home after 12 days on venlafaxine-XR and risperidone and enrolled in a partial hospitalization program. Two months later, her depression was much improved, her concerns about intruders and “harassment” had diminished, and her bony pain had lessened.

❉

335 ❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

The Difficult-to-Treat Psychiatric Patient With Comorbid Diabetes Mellitus Depression in patients with diabetes mellitus is common, diagnostically vexing, often malevolent in its course, and frequently associated with poor blood glucose (glycemic) control and diabetic complications (Cohen et al. 1997; Lustman et al. 1988, 1997b). It has even been suggested that major depression increases the risk for non–insulin-dependent diabetes mellitus (NIDDM) (Eaton et al. 1996). Depression may impair glycemic control because of lack of adherence to dietary and exercise programs (Marcus et al. 1992) or via complex effects on cortisol and/or growth hormone (Holsboer 1995). The prevalence of major depression in diabetes—considering both Type I insulin-dependent diabetes mellitus [IDDM] and Type II NIDDM—is about 15%–20% (Lustman et al. 1997b). As with depression in the medically well, depression in diabetic cohorts is characterized by a positive family history of depression and typical depressive symptoms (Lustman et al. 1992). And, as in other medical illnesses, diabetic patients may report apparent “depressive” symptoms such as weight loss, fatigue, and hypersomnia that are actually a direct manifestation of diabetes. One clue to depression in diabetic patients is the patient’s amplification of somatic symptoms of physical illness and his or her difficulty habituating to these aversive symptoms (Katon 1996). Mild hyperglycemia—with associated polyuria, polyphagia, polydipsia, fatigue, blurred vision, or paresthesia—may mimic hypochondriasis (Kornstein and Gardner 1993) or cognitive disorders (Reed and Mooradion 1998). Hypoglycemia may present with adrenergically mediated anxiety symptoms (e.g., diaphoresis, tremor) or neuroglycopenic symptoms, such as dizziness, headache, weakness, blurred vision, and confusion (Reed and Mooradion 1998).

Comorbid Diabetes and Psychiatric Illness: Pharmacological Treatments Tricyclic Antidepressants TCA effects on blood glucose are complex and may vary as a function of time. Thus acute use of TCAs may be associated with falling blood glucose, followed by hyperglycemic responses weeks or months later (Goodnick et al. 1995; Shader et al. 1970). In a recent double-blind, placebo-controlled ❉

336 ❉

❉ The Difficult-to-Treat Psychiatric Patient With Comorbid Medical Illness ❉ study, 68 diabetic patients with poor glycemic control were given nortriptyline (at therapeutic level) or placebo for 8 weeks (Lustman et al. 1997a). Twenty-eight subjects had active major depression. In the depressed group, nortriptyline was statistically better than placebo in alleviating depression but not in reducing glycated hemoglobin (gHb). Indeed, path analysis indicated that the direct effect of nortriptyline was to worsen glycemic control, even as improvement in depression had an independent beneficial effect on gHb. The authors concluded that major depression in diabetic patients can be effectively treated with nortriptyline at the expense of a direct hyperglycemic effect and is thus not an ideal agent in this population. (In the nondepressed diabetic patients, nortriptyline exerted a direct hyperglycemic effect that did not seem to be attributable to weight gain.) Other worrisome TCA side effects in diabetic patients include weight gain and carbohydrate/sweet cravings—perhaps less with desipramine than with other TCAs (Garland et al. 1988). Anticholinergic effects on bowel motility may worsen diabetic gastroparesis and further disturb a neurologically impaired bladder. TCAs may also cause anticholinergic-induced relaxation of the gastroesophageal sphincter, with consequent acid reflux. This may further confound diagnosis of gastrointestinal distress in a depressed diabetic. On the other hand, TCAs may reduce gastric acid via H2-blocking effects. Central anticholinergic effects are especially worrisome in elderly, depressed diabetic patients, given the potential for preexisting cerebrovascular disease (Tun et al. 1990). Nontricyclic Antidepressants Before the advent of the SSRI antidepressants, MAOIs were used in depressed diabetic patients. Hydrazine MAOIs (i.e., phenelzine) and nonhydrazine MAOIs (i.e., tranylcypromine) have been associated with reduced blood sugar and hypoglycemia (Shader et al. 1970). This led some, though not all, researchers to suggest that MAOIs should not be combined with insulin or oral hypoglycemic (sulfonylureas) agents (Goodnick et al. 1995; Shader et al. 1970). Additional MAOI side effects include weight gain (Garland et al. 1988) and orthostatic hypotension. Weight gain seems to be greater with phenelzine than with tranylcypromine (Rabkin et al. 1985). Orthostatic hypotension from MAOIs may be exacerbated by diabetic neuropathy-related hypotension and concurrent cardiac medications (e.g., nitrates, calcium channel blockers, antihypertensives). Enthusiasm for serotonergic agents began with experimental evidence of hypoglycemic and insulin-sensitizing effects (Goodnick et al. 1995). Most ❉

337 ❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

human research has involved nondepressed diabetic subjects. For example, in nondepressed NIDDM patients, fluoxetine (60 mg/day) has been associated with improvement in weight and gHb (Goodnick 1997). In one study, improvement was more apparent at 3 and 6 months than at 9 and 12 months (O’Kane et al. 1994), reminiscent of the transient weight loss reported with fluoxetine in some studies (Medical Letter 1994). Such short-term benefits have been confirmed recently in nondepressed NIDDM patients in a doubleblind study: compared with placebo, fluoxetine 60 mg was associated with decreased weight, fasting blood glucose, gHb, insulin, and triglyceride levels at 8 weeks (Daubresse et al. 1996). In a similar group of patients, 4 weeks of fluoxetine (60 mg/day) improved insulin-mediated glucose disposal despite absence of weight loss, suggesting a direct effect of fluoxetine on insulin sensitivity (Maheux et al. 1997). Depressed diabetic subjects have received little systematic study with SSRIs. However, an open 10-week study (Goodnick et al. 1997) of depressed NIDDM patients given sertraline 50 mg/day showed significant improvement on two depression rating scales and dietary compliance. Modest improvement of gHb levels was also seen. Although SSRIs have advantages over TCAs in depressed diabetic patients, potential SSRI-related gastrointestinal distress and sexual dysfunction may be hard to discern in a multiproblem medical patient taking many medications. In addition, SSRI-related diarrhea may lead to hypoglycemia in IDDM patients (T. Korkosz, personal communication, 1998). In addition, diabetic patients receiving multiple medications must be monitored for the potential enzyme-inhibiting effects of the SSRIs. For example, the sulfonylurea agent tolbutamide, although now used less frequently, is a substrate for the hepatic cytochrome P450–2C9 isoenzyme, which is inhibited by fluoxetine and fluvoxamine (Harvey and Preskorn 1996a). Finally, the SSRIs may cause weight gain, especially when taken for many months, although not as much as the TCAs. Paroxetine may be the SSRI most likey to cause weight gain. SSRI-related weight gain may be stemmed by switching to buproprion (Sussman and Ginsberg 1998). There are few relevant studies of other antidepressants in depressed diabetic patients. Warnock and Biggs (1997) described a 54-year-old depressed woman with IDDM treated with nefazodone (200–400 mg/day) who showed both a reduced need for insulin and a good antidepressant response. Electroconvulsive Therapy ECT treatment of diabetes has been associated with both amelioration and worsening of hyperglycemia. Hypoglycemia has not been reported (Reddy ❉

338 ❉

❉ The Difficult-to-Treat Psychiatric Patient With Comorbid Medical Illness ❉ and Nobler 1996). One review (Finestone and Weiner 1984) found no significant effects or trends in mild diabetes but found potentially profound hyperglycemia in severe diabetes. This may occur via ECT-induced acute (but transient) elevation of epinephrine, growth hormone, glucagon, and adrenocorticotrophic hormone (ACTH). In mild diabetes, post-ECT improvement in glycemic control may be due to improved compliance and/or reduced hypercortisolemia. On the other hand, severely diabetic patients are sometimes not able to compensate for ECT-induced hyperglycemia. A report of severe hyperglycemia after ECT in a patient without known diabetes suggests that hyperglycemia be considered in the differential diagnosis of post-ECT acute confusion (Reddy and Nobler 1996). Antipsychotics There is little systematic evidence to guide psychopharmacological treatment of psychotic disorders in diabetic patients. Standard antipsychotic agents merit caution because diabetic subjects seem particularly vulnerable to the development of tardive dyskinesia (Schultz et al. 1999). Although there is evidence both supporting (Shader et al. 1970) and refuting (Wirshing et al. 1998) the occurrence of neuroleptic-related hyperglycemia, there are increasing reports of new-onset diabetes and worsening of stable diabetes (including ketoacidosis) and elevated blood lipids related to clozapine and olanzapine use (Lindenmayer et al. 1999; Melkersson et al. 1999; Wirshing et al. 1998). New-onset diabetes has also been associated with quetiapine treatment (Sobel et al. 1999; Procyshyn et al. 2000). Among the atypical antipsychotics, risperidone seems least likely to cause glucose dysregulation, (although one case report [Croarkin et al. 2000] described diabetic ketoacidosis in an HIV-infected man possibly related to risperidone). Ziprasidone has not been associated with diabetogenic effects, although general clinical experience is limited as of this writing (Labbate and Ayd 2001). Weight gain caused by atypical antipsychotics (possibly related to blockade of dopamine2, histamine-1, and serotonin-2C receptors [Baptista 1999; Stahl 1998]) may also be problematic and may contribute to decreased glucose tolerance. Clozapine and olanzapine appear to cause most weight gain and ziprasidone may cause the least (Allison et al. 1999; Ganguli 1999; Stahl 1999). The seldom-used neuroleptic molindone has been found to cause weight loss or at least no weight gain (Ganguli 1999). Novel pharmacological strategies to combat antipsychotic-related weight gain (e.g., use of the oral hypoglycemic metformin, estradiol, tamoxifen) have been proposed (Baptista 1999), but routine use is premature. ❉

339 ❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

Mood Stabilizers Carbamazepine is less likely to cause weight gain than lithium or valproate (Pies 1998a; Sussman and Ginsberg 1999) and may therefore be useful in selected diabetic patients with bipolar disorder who struggle with weight control. Lamotrigine seems to cause less weight gain than valproate (Sussman and Ginsberg 1999). Topiramate, a novel anticonvulsant, may cause timelimited weight loss or at least minimal weight gain compared with other available anticonvulsants. Despite favorable preliminary studies, topiramate’s efficacy and safety in bipolar patients have yet to be established (Ghaemi and Gaughan 1999). Topiramate may cause neurocognitive side effects and theoretically worsen diabetes control via its effect as a carbonic anhydrase inhibitor (Sussman and Ginsberg 1999). Anxiolytics Alprazolam, but not placebo, improved glucose regulation in a double-blind study of poorly controlled diabetic patients with generalized anxiety disorder (Lustman et al. 1995). This beneficial effect was not a function of enhanced compliance. Surprisingly, alprazolam-treated psychiatrically well diabetic patients also experienced improved glycemic control (Lustman et al. 1995). β-Blockers have been cautiously prescribed in brittle diabetic patients because of their potential to mask adrenergic symptoms of hypoglycemia. However, in a dated case report (Baker et al. 1969), propranolol was used therapeutically in two girls to reduce repeated episodes of presumed stressrelated diabetic ketoacidosis. Psychotropics as Analgesics Antidepressants may have analgesic effects in painful diabetic neuropathy, irrespective of the presence of depression. There is some evidence (Harnett 1994) to suggest that TCAs (even at modest dosages) are superior to SSRIs in lessening pain. One study (Max et al. 1992) found that amitriptyline and desipramine promoted analgesia in depressed and nondepressed diabetic patients, whereas fluoxetine provided pain relief only in the depressed group. Sertraline and paroxetine may also be useful in diabetic neuropathy (Mendoza et al. 1998; Sindrup et al. 1990). Gabapentin, an anticonvulsant also being considered as a mood stabilizer, has recently been reported to alleviate diabetic neuropathy pain (Backonja et al. 1998). However, excessive sedation may occur with fixed-dose protocols. ❉

340 ❉

❉ The Difficult-to-Treat Psychiatric Patient With Comorbid Medical Illness ❉

Comorbid Diabetes: Psychosocial Intervention Psychosocial factors have long been suspected in diabetes, although it is often hard to tell whether psychosocial problems are causally related to or the consequence of the illness (Helz and Templeton 1990). Reports of psychotherapy helping emotional disturbance and glucose control have been essentially anecdotal. Menninger (1935) reported that individual psychotherapy reduced glycosuria related to emotional turmoil. Similarly, group and family therapy may increase compliance in diabetic adolescents and decrease hospitalization rates in diabetic children (Minuchin et al. 1975). The effects of stress reduction techniques (such as hypnosis and biofeedback) on diabetic control are unclear (Helz and Templeton 1990). A recent controlled study suggested efficacy of cognitive-behavioral therapy for major depression in diabetes, especially in those compliant with blood glucose monitoring (Lustman et al. 1998). Attention to the therapeutic relationship between patient and health care team was recently emphasized in a practical review of psychological care of IDDM patients (Jacobson 1999). Diabetic subjects may also suffer from comorbid eating disorders (Copeland and Anderson 1995; Crow et al.1998). Bulimic subjects, for example, may intentionally omit or reduce insulin doses as a method of purging. The consequent glucosuria and large urine volume may result in transient weight loss (from fluid and caloric loss). Unfortunately, such behavior may also worsen glycemic control and increase rates of diabetic complications. Moreover, the usual strategy of encouraging dietary restraint and weight loss in diabetic patients may backfire in those with eating disorders because the latter often engage in destructive cycles of excessive dietary restraint and disinhibition. Cognitive-behavioral therapy specifically directed at the eating disorder may be helpful.

The Difficult-to-Treat Psychiatric Patient With Comorbid Respiratory Disease Demographic studies have suggested an increased prevalence (8%–37%) of panic disorder (Pollack et al. 1996) and mood disorders (over 20%) (Borson et al. 1992) in patients with chronic obstructive pulmonary disease (COPD). Conversely, there may be an increased lifetime prevalence (47%) of respiratory disease in panic disorder (Zandbergen et al. 1991). The COPD patient demonstrates the complex interactions of comorbid respiratory and psychiatric symptoms, especially dyspnea and anxiety. For ❉

341 ❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

example, anxiety and dyspnea can occur simultaneously and can escalate each other. Thus, distinguishing a panic attack from an exacerbation of lung disease—an important clinical determination—can be quite challenging. Moreover, catastrophic misinterpretation of physical sensations, rather than severity of lung disease or response to bronchodilators, predicts panic disorder in COPD patients (Carr et al. 1994). Dyspnea and ventilatory control are also influenced by depression. Ventilatory response to inhaled CO 2 may be blunted in clinical depression (Greenberg and Kradin 1996), and depression is predictive of self-rated breathlessness in COPD patients (Kellner et al. 1992). Whereas the nature of this association remains unclear, depression may magnify the perception and distress of dyspnea. Alternatively, dyspnea may act as a psychosocial stressor that induces depression (Kellner et al. 1992). COPD-related anergia, psychomotor retardation, anorexia/weight loss, and sleep disturbance related to nocturnal coughing may confound the diagnosis of depression. Pervasive pessimism, diurnal mood variation, early morning awakening, and the sense of activity as “effortful” may be clues to depression in COPD patients (Greenberg and Kradin 1996), as may guilt and self-deprecatory ideation, in my experience.

Comorbid Respiratory Illness: Psychopharmacological Treatment Tricyclic Antidepressants TCAs may be of particular benefit for COPD patients with panic and/or depressive symptoms. Earlier case reports suggested additional TCA benefits for respiratory symptoms (Burns and Howell 1969; Yellowlees 1987), especially in asthma (Smoller et al. 1996; Yellowlees and Kalucy 1990); however, such benefits for COPD patients have not been verified. Thus, in one 12week, placebo-controlled trial of 30 depressed COPD patients (Borson et al. 1992), nortriptyline was associated with marked improvement of depression and anxiety, diminution of a variety of somatic symptoms, and enhanced functioning, but not with improved respiratory function. Nortriptyline also was associated with a moderate increase in heart rate (supine increase from 82.0 to 92.6 beats per minute). Patients with advanced COPD may have trouble tolerating side effects of higher dose TCA therapy (Smoller et al. 1996). Doxepin, at a mean dose of 105 mg/day, was poorly tolerated in 25% of COPD patients and ineffective in alleviating depression and anxiety in a small placebo-controlled series of COPD patients (Smoller et al. 1996). Anticholinergic “drying” effects in the ❉

342 ❉

❉ The Difficult-to-Treat Psychiatric Patient With Comorbid Medical Illness ❉ respiratory tract can sometimes be a disadvantage with TCAs. Some reports suggest that protriptyline improves diurnal and nocturnal hypoxemia in COPD patients and enhances ventilation in some sleep apnea patients (Series and Cormier 1990); however, use of protriptyline for sleep apnea or affective illness has substantially declined in recent years. Nontricyclic Antidepressants SSRIs have become increasingly popular in the treatment of panic disorder in medically well patients. Vulnerability to panic, possibly secondary to abnormal carbon dioxide sensitivity, is thought to be dampened by serotonin activation (Klein 1993). In one study (Papp et al. 1995), six COPD patients, only three with anxiety/mood diagnoses, all reported improved sense of well-being, and five out of six improved on activities of daily living after 6 weeks of sertraline 100 mg/day. There were no significant objective changes noted in pulmonary function. Finally, in a recent series of seven sertralinetreated (25–100 mg/day) COPD patients—three of whom had anxiety or mood disorders—all reported subjective improvement of dyspnea. Most reported improved exercise tolerance, and there was little change in pulmonary functions (Smoller et al. 1998). SSRIs should be started at low dosage to lessen the possibility of medication-related jitteriness and serotonin-induced diminution of CO2 sensitivity (Smoller et al. 1998). The clinician should also watch for covert SSRIrelated gastrointestinal distress and not attribute this to medications such as theophylline or antibiotics. Fluvoxamine can substantially inhibit the metabolism of theophylline (via inhibition of CYP1A2), causing elevated, and possibly toxic, theophylline levels (Rasmussen et al. 1995). Similarly, sudden cessation of tobacco use may also cause theophylline toxicity, as cigarette smoke-related induction of CYP1A2 enzymes ceases (Pies 1998b). Owing to the risk of benzodiazepine-related respiratory depression in COPD patients, clinicians should be aware that alprazolam and triazolam levels may be increased by CYP3A4 inhibitors like nefazodone, fluoxetine, and fluvoxamine (Harvey and Preskorn 1996b). Similarly, diazepam levels may be increased by CYP2C19 inhibitors like fluoxetine and fluvoxamine (Harvey and Preskorn 1996b). Although not absolutely contraindicated, MAOIs are infrequently prescribed to COPD and asthma patients. In theory, hypertensive crisis could follow concurrent use of MAOIs and sympathomimetic agents, such as β-2adrenergic bronchodilators (e.g., albuterol, terbutaline), epinephrine, and decongestants (e.g., pseudoephedrine, phenylpropanolamine). However, ❉

343 ❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

there has been little clinical documentation of toxicity with concurrent use of MAOIs and β-inhalers. One COPD patient receiving phenelzine became hypomanic, though not hypertensive, when isoetharine inhaler was added (Goldman and Tiller 1987). Helping COPD patients quit smoking is an important goal of treatment. A history of major depression is associated with cigarette smoking and predicts smoking cessation failure (Glassman 1993, 1998b). Bupropion (Hurt et al. 1997; Jorenby et al. 1999) and nortriptyline (Hall et al. 1998) but not SSRIs (Glassman 1998b) have been found to aide smoking cessation, even in the absence of a history of major depression. Antipsychotics Whereas short-term use of antipsychotics can lessen agitation and psychosis in delirious COPD patients, no medication is a substitute for correction of blood gas derangements, underlying infection, and so forth. And whereas conventional antipsychotics (e.g., haloperidol) are usually employed in delirium, several reports suggest that risperidone (Sipahimalani et al. 1997) and olanzapine (Sipahimalani and Masand 1998) may be beneficial—although not specifically in COPD patients. Olanzapine also appeared to reduce steroid-related mood changes in one asthma patient (Brown et al. 1999). Antipsychotics should be used cautiously, if at all, in those COPD patients with dehydration—a risk factor in neuroleptic malignant syndrome. Laryngeal dystonia and respiratory dyskinesia secondary to neuroleptics are also risks. Finally, several antimicrobials often used by COPD patients (e.g., ciprofloxacin, erythromycin, rifampin) may affect levels of olanzapine, clozapine, and other psychotropics (Pies 1999). Mood Stabilizers Mood stabilizers can generally be used safely in COPD patients, with cautious dosing. Lithium may actually attenuate steroid-related psychotic symptoms (Falk et al. 1979; Siegel 1978). Theophylline and aminophylline may lead to decreased lithium levels by inhibiting proximal tubular reabsorption and thereby increasing excretion of lithium. Smaller lithium dosages should probably be used in those COPD patients who are already tremulous (perhaps because of sympathomimetic pulmonary medications) or who are vulnerable to confusion. Although valproate, a mild cytochrome enzyme inhibitor, poses some risk in medically ill patients, it is not specifically problematic in COPD patients. However, valproate can be quite sedating and cause ataxia, especially in some debilitated patients. Valproate❉

344 ❉

❉ The Difficult-to-Treat Psychiatric Patient With Comorbid Medical Illness ❉ related gastrointestinal distress, which may be mistakenly attributed to theophylline or antibiotics (as with SSRIs, see earlier in this chapter), may be lessened by use of divalproex sprinkle capsules or concurrent use of histamine-2 blockers. Valproate (Abbos and Styra 1994) and lamotrigine (Preda et al. 1999) have been reported to help prevent steroid-related mania. Carbamazepine, now available in an extended-release preparation, is a potent enzyme-inducer and may reduce levels of numerous medications, including theophylline, alprazolam, clonazepam, and clozapine. Anxiolytics Benzodiazepine therapy has been controversial in COPD patients, owing to concerns about dose-related respiratory depression, especially with longer acting agents. Nevertheless, clinical observation suggests that benzodiazepines may be useful in COPD when used cautiously. Thus, Mitchell-Heggs and colleagues (1980) found that diazepam reduced dyspnea and improved exercise tolerance in “pink puffer” (emphysematous) syndrome. However, further attempts to differentiate the benzodiazepine response of “blue bloaters” (little dyspnea despite chronic hypercarbia) from “pink puffers” (more dyspnea, little CO2 retention) have not been fruitful, and benzodiazepine use has been discouraged (Smoller et al. 1996). In medically well patients, alprazolam may diminish panic anxiety by reducing sensitivity to CO2 (Woods et al. 1986). In COPD patients, this effect could be beneficial (i.e., by decreasing anxiety) or detrimental (i.e., via CO2 retention). In anxious, mechanically ventilated patients, benzodiazepine-related respiratory depression may be of less concern. Benzodiazepine therapy generally should be avoided in sleep apnea (Greenberg and Kradin 1996) and may exacerbate panic in such patients (Smoller et al. 1996). The anxiolytic buspirone may have a centrally mediated stimulating effect on respiration and may increase CO2 sensitivity (Mendelson et al. 1990). In theory, this may worsen panic. However, in case reports of patients with severe lung disease, buspirone (15–45 mg/day) seems to diminish anxiety without significant side effects or deterioration of respiratory function (Alderman et al. 1996). Two controlled trials of buspirone in anxious COPD patients have yielded conflicting results: improvement in anxiety, dyspnea, and exercise tolerance after 14 days at 20 mg/day (Argyropoulou et al. 1993) versus no effect on mild anxiety or exercise tolerance after 6 weeks at 30–60 mg/day (Singh et al. 1993). Buspirone is nonsedating, does not cause physical dependence, and can be readily coprescribed with bronchodilators (Kiev and Domantay 1988). However, buspirone’s disadvantages (vs. benzodiaz❉

345 ❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

epines) include delayed therapeutic response, less certain efficacy (especially in true panic disorder), and occasional gastrointestinal distress or stimulation.

Comorbid Respiratory Illness: Cognitive-Behavioral Treatment Treatment of psychiatric disorder in the COPD patient has emphasized cognitive-behavioral and pulmonary rehabilitative approaches (Celli 1998), as well as optimizing treatment of the underlying respiratory illness (Smoller et al. 1996). The rationale for cognitive-behavioral therapy in anxious COPD patients is based in part on the finding that catastrophic cognitions (e.g., “I’m dying!”) predict that COPD patients will panic (Carr et al. 1994). Behavioral therapists have traditionally emphasized in vivo exposure to fearful situations as critical in reducing phobic avoidance and panic attacks (Marks 1987). Cognitive-behavioral therapists (Barlow and Cerny 1988) have added the technique of interoceptive exposure, whereby fearful bodily sensations (“interoceptive” stimuli) are elicited in the office setting by such manipulations as voluntary hyperventilation or spinning in place. The therapist then helps desensitize the patient to these internal physiological cues. For the COPD patient, however, the therapist faces the additional challenge of helping the patient distinguish catastrophic thinking from a realistic recognition of worsening respiratory illness and the need for prompt medical intervention. In addition to interoceptive exposure, specific cognitivebehavioral techniques include relaxation/imagery training (Smoller et al. 1996), cognitive restructuring, breathing retraining, (Celli 1998; Clark 1985), and “homework” in the form of keeping daily records. A key ingredient of pulmonary rehabilitation—exercise training—encourages patients with phobic avoidance of activity and fear of dyspnea to repeatedly and increasingly exercise in the presence of reassuring medical personal (Agle and Baum 1977; Celli 1998). Finally, in a randomized trial, asthma patients who wrote about stressful illness experiences were found to have improved lung function after 4 months, compared with control subjects (Smyth et al. 1999).

Conclusion Comorbid medical disorders create many difficulties in the treatment of refractory psychiatric illness, and well-designed treatment outcome studies are scant. Psychiatric illness in the medically ill may differ from that in the ❉

346 ❉

❉ The Difficult-to-Treat Psychiatric Patient With Comorbid Medical Illness ❉ medically well in terms of epidemiological characteristics, underlying pathophysiology, and treatment response. The need for multiple medications in this population, along with the potential for drug-drug interactions, further complicates the clinician’s job. Future research may eventually support detailed algorithms for accurate diagnosis and effective treatment of psychiatric illness complicated by particular medical comorbidities. In the mean time, the clinician’s familiarity with a particular medical comorbidity will enhance diagnostic accuracy and safe, effective treatment.

References Abbos A, Styra R: Valproate prophylaxis against steroid-induced psychosis (letter). Can J Psychiatry 39:188–189, 1994 Agle DP, Baum GL: Psychological aspects of chronic pulmonary disease. Med Clin North Am 61:749–758, 1977 Alderman CP, Frith PA, Ben-Tovim DI: Buspirone for treatment of anxiety in patients with chronic obstructive airways disease. J Clin Psychopharmacol 16:410–411, 1996 Allison DB, Mentore JL, Heo M, et al: Antipsychotic-induced weight gain: a comprehensive research synthesis. Am J Psychiatry 156:1686–1696, 1999 Amsterdam JD, Garcia-Espana F, Fawcett J, et al: Blood pressure changes during shortterm fluoxetine treatment. J Clin Psychopharmacol 19:9–14, 1999 Andrews C, Pinner G: Postural hypotension induced by paroxetine (letter). BMJ 316:595, 1998 Argyropoulou P, Patakas D, Koukou A, et al: Buspirone effects on breathlessness and exercise performance in patients with chronic obstructive pulmonary disease. Respiration 60:216–220, 1993 Avorn J: Depression in the elderly falls and pit falls (editorial). N Engl J Med 339:918– 920, 1998 Backonja M, Beydoun A, Edwards KR, et al: Gabapentin for the symptomatic treatment of painful neuropathy in patients with diabetes mellitus. JAMA 280:1831–1836, 1998 Baker L, Baccai A, Kaye R, et al: Beta adrenergic blockade and juvenile diabetes: acute studies and long-term therapeutic trial. J Pediatr 75:19–29, 1969 Baptista T: Body weight induced by antipsychotic drugs: mechanisms and management. Acta Psychiatr Scand 100:3–16, 1999 Barbey JT, Roose SP: SSRI safety in overdose. J Clin Psychiatry 59 (suppl 15): 42–48, 1998 Barlow DH, Cerny JA: Psychological Treatment of Panic. New York, Guilford, 1988

❉

347 ❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

Blumenthal JA, Jiang W, Babyak MA, et al: Stress management and exercise training in cardiac patients with myocardial ischemia. Arch Intern Med 157:2213–2223, 1997 Borson S, McDonald G, Gayle T, et al: Improvement in mood, physical symptoms and function with nortriptyline for depression in patients with chronic obstructive pulmonary disease. Psychosomatics 33:190–201, 1992 Brown ES, Khan DA, Suppes T: Treatment of corticosteroid-induced mood changes with olanzapine (letter). Am J Psychiatry 156:968, 1999 Buff DD, Brenner R, Kirtane SS, et al: Dysrhythmia associated with fluoxetine treatment in an elderly patient with cardiac disease. J Clin Psychiatry 52:174–176, 1991 Burns BH, Howell JBL: Disproportionately severe breathlessness in chronic bronchitis. QJM 38:277–294, 1969 Calabrese JR, Bowden CL, McElroy SL, et al: Spectrum of activity of lamotrigine in treatment-refractory bipolar disorder. Am J Psychiatry 156:1019–1023, 1999 Cardiac Arrhythmia Suppression Trial (CAST) Investigators: Preliminary report: effect of encainide and flecainide on mortality in a randomized trial of arrhythmia suppression after myocardial infarction. N Engl J Med 321:406–412, 1989 Carr RE, Lehrer PM, Rausch LL, et al: Anxiety sensitivity and panic attacks in an asthmatic population. Behav Res Ther 32:411–418, 1994 Casey DE: The relationship of pharmacology to side effects. J Clin Psychiatry 58 (suppl 10):55–62, 1997 Celli BR: Pulmonary rehabilitation for COPD. Postgraduate Medicine 103:159–176, 1998 Clark DM: Respiratory control as a treatment for panic attacks. J Behav Ther Exp Psychiatry 16:23–30, 1985 Cohen ST, Welch G, Jacobson AM, et al: The association of lifetime psychiatric illness and increased retinopathy in patients with type I diabetes mellitus. Psychosomatics 38:98–108, 1997 Compston JE: Vitamin D deficiency: time for action—evidence supports routine supplementation for elderly people and others at risk. BMJ 317:1466–1467, 1998 Copeland PM, Anderson B: Diabetes mellitus and eating disorders. Harv Rev Psychiatry 3:36–49, 1995 Croarkin PE, Jacobs KM, Bain BK: Diabetic ketoacidosis associated with risperidone treatment (letter)? Psychosomatics 41:369–370, 2000 Crow SJ, Keel PA, Kendall D: Eating disorders and insulin-dependent diabetes mellitus. Psychosomatics 39:233–243, 1998 Dalack GW, Roose SP: Perspectives on the relationship between cardiovascular disease and affective disorder. J Clin Psychiatry 51 (suppl):4–9, 1990 DasGupta K, Jefferson JW: The use of lithium in the medically ill. Gen Hosp Psychiatry 12:83–97, 1990 Daubresse JC, Kolanowski J, Krzentowski G: Usefulness of fluoxetine in obese noninsulin-dependent diabetics: a multicenter study. Obes Res 4:391–396, 1996

❉

348 ❉

❉ The Difficult-to-Treat Psychiatric Patient With Comorbid Medical Illness ❉ Eaton WW, Pratt L, Armenian H, et al: Depression and risk for onset of type II diabetes, a prospective population-based study. Diabetes Care 19:1097–1102, 1996 Falk WE, Mahnke MW, Poskanzer DC: Lithium prophylaxis of corticotropin-induced psychosis. JAMA 241:1011–1012, 1979 Favre M-P, Sztajzel J, Bertschy G: Bradycardia during citalopram treatment: a case report. Pharmacol Res 39:149–150, 1999 Finestone DH, Weiner RD: Effects of ECT on diabetes mellitus. Acta Psychiatr Scand 70:321–326, 1984 Fisch C: Effect of fluoxetine on the electrocardiogram. J Clin Psychiatry 46:42–44, 1985 Frasure-Smith N, Lesperance F, Talajic M: Depression following myocardial infarction: impact on 6-month survival. JAMA 270:1819–1825, 1993 Frasure-Smith N, Lesperance F, Talajic M: Depression and 18-month prognosis after myocardial infarction. Circulation 91:999–1005, 1995 Frasure-Smith N, Lesperance F, Prince RH, et al: Randomized trial of home-based psychosocial nursing intervention for patients recovering from myocardial infarction. Lancet 350:473–479, 1997 Ganguli R: Weight gain associated with antipsychotic drugs. J Clin Psychiatry 60 (suppl 21):20–24, 1999 Garland EJ, Remick RA, Zis AP: Weight gain with antidepressants and lithium. J Clin Psychopharmacol 8:323–328, 1988 Ghaemi SN: Oxcarbazepine: a mood stabilizer rediscovered. International Drug Therapy Newsletter 36:20–22, 2001 Ghaemi SN, Gaughan S: Novel anticonvulsants: new generation of mood stabilizers? Part II. International Drug Therapy Newsletter 34:35–37, 1999 Glassman AH: Cigarette smoking: implications for psychiatric illness. Am J Psychiatry 150:546- 553, 1993 Glassman AH: Cardiovascular effects of antidepressant drugs: updated. J Clin Psychiatry 59 (suppl 15):13–18, 1998a Glassman AH: Psychiatry and cigarettes (commentary). Arch Gen Psychiatry 55:692– 693, 1998b Glassman AH, Shapiro PA: Depression and the course of coronary artery disease. Am J Psychiatry 155:4–11, 1998 Glassman AH, Johnson LL, Giardina EGV, et al: The use of imipramine in depressed patients with congestive heart failure. JAMA 250:1997–2001, 1983 Glassman AH, Roose SP, Bigger JT Jr: The safety of tricyclic antidepressants in cardiac patients: risk-benefit reconsidered. JAMA 269:2673–2675, 1993 Glassman AH, Rodriquez AL, Shapiro PA: The use of antidepressant drugs in patients with heart disease. J Clin Psychiatry 59 (suppl 10):16–21, 1998 Goldman LS, Tiller JA: Hypomania related to phenelzine and isoetharine interaction in one patient (case report). J Clin Psychiatry 48:170, 1987

❉

349 ❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

Goodnick PJ: Practical considerations in the treatment of depression in the diabetic patients. Primary Psychiatry 4:37–40, 1997 Goodnick PJ, Henry JH, Buki VMV: Treatment of depression in patients with diabetes mellitus. J Clin Psychiatry 56:128–136, 1995 Goodnick PJ, Kumar A, Henry JH, et al: Sertraline in coexisting major depression and diabetes mellitus. Psychopharmacology Bulletin 33:261–264, 1997 Greenberg DB, Kradin RL: Lung disease, in Textbook of Consultation-Liaison Psychiatry. Edited by Rundell JR, Wise MG. Washington, DC, American Psychiatric Press, 1996, pp 561–566 Greenblatt DJ, von Moltke LL, Harmatz JS, et al: Drug interactions with new antidepressants: role of human cytochromes P450. J Clin Psychiatry 59 (suppl 15):19– 27, 1998 Gullette ECD, Blumenthal JA, Babyak M, et al: Effects of mental stress on myocardial ischemia during daily life. JAMA 277:1521–1526, 1997 Hagg S, Spigset O, Soderstrom TG: Association of venous thromboembolism and clozapine. Lancet 355:1155–1156, 2000 Hall SM, Reus VI, Munoz RF, et al: Nortriptyline and cognitive-behavioral therapy in the treatment of cigarette smoking. Arch Gen Psychiatry 55:683–690, 1998 Harnett DS: Psychopharmacologic treatment of depression in the medical setting. Psychiatric Annals 24:545–551, 1994 Harvey AT, Preskorn SH: Cytochrome P450 enzymes: interpretation of their interactions with selective serotonin reuptake inhibitors: I. J Clin Psychopharmacol 16:273–285, 1996a Harvey AT, Preskorn SH: Cytochrome P450 enzymes: interpretation of their interactions with selective serotonin reuptake inhibitors: II. J Clin Psychopharmacol 16:345–355, 1996b Helz JW, Templeton B: Evidence of the role of of psychosocial factors in diabetes mellitus: a review. Am J Psychiatry 147:1275–1282, 1990 Holland JC (ed): Psycho-Oncology, 1st Edition, New York, Oxford University Press, 1998 Holsboer F: Neuroendocrinology of mood disorders, in Psychopharmacology: The Fourth Generation of Progress. Edited by Bloom FE, Kupfer DJ. New York, Raven, 1995, pp 957–969 Hopkins HS: Fluoxetine-bupropion interaction. Biological Therapies in Psychiatry Newsletter 19:31–32, 1996 Hurt RD, Sachs DPL, Glover ED: A comparison of sustained-release buproprion and placebo for smoking cessation. N Eng J Med 337:1195–1202, 1997 Ihde-Scholl T, Rolli ML, Jefferson JW: Clozapine and pulmonary embolus. Am J Psychiatry 158:499–500, 2001 Jackson T, Ditmanson L, Phibbs B: Torsades de pointes and low-dose oral haloperidol. Ann Intern Med 119:391–394, 1993

❉

350 ❉

❉ The Difficult-to-Treat Psychiatric Patient With Comorbid Medical Illness ❉ Jacobson AM: The psychological care of patients with insulin-dependent diabetes mellitus. N Engl J Med 334:1249–1253, 1999 Jiang W, Babyak M, Krantz DS, et al: Mental stress-induced myocardial ischemia and cardiac events. JAMA 21:1651–1656, 1995 Jones DA, West RR: Psychological rehabilitation after myocardial infarction: multicentre randomized controlled trial. BMJ 313:1517–1521, 1996 Jorenby DE, Leischow SJ, Nides MA, et al: A controlled trial of sustained-release bupropion, a nicotine patch, or both for smoking cessation. N Engl J Med 340:685– 91, 1999 Kaltsounis J, Alfaro CL, Trott JC: Venlafaxine-associated blood pressure elevation at low doses: a report of three cases. International Journal of Geriatric Psychopharmacology 1:95–97, 1998 Kang UG, Kwon JS, Ahn YM, et al: Electrocardiographic abnormalities in patients treated with clozapine. J Clin Psychiatry 61:441–446, 2000 Katon W: The impact of major depression on chronic medical illness. Gen Hosp Psychiatry 18:215–219, 1996 Keitner GI, Ryan CE, Miller IW, et al: Twelve-month outcome of patients with major depression and comorbid psychiatric or medical illness (compound depression). Am J Psychiatry 148:345–350, 1991 Kellner R, Samet J, Pathak D: Dyspnea, anxiety, and depression in chronic respiratory impairment. Gen Hosp Psychiatry 14:20–28, 1992 Kiev A, Domantay A: A study of buspirone coprescribed with bronchodilators in 32 anxious ambulatory patients. J Asthma 25:281–284, 1988 Kilian JG, Kerr K, Lawrence C, et al: Myocarditis and cardiomyopathy associated with clozapine. Lancet 354:1841–1845, 1999 Klein DF: False suffocations alarms, spontaneous panics, and related conditions: an integrative hypothesis. Arch Gen Psychiatry 50:306–317, 1993 Koenig HG: Depression in hospitalized older patients with congestive heart failure. Gen Hosp Psychiatry 20:29–43, 1998 Kornstein SG, Gardner DF: Endocrine disorders, in Psychiatric Care of the Medical Patient. Edited by Stoudemire A, Fogel BS. New York, Oxford University Press, 1993, pp 657–679 Krahenbuhl S, Sauter B, Kupferschmidt H, et al: Case report: Reversible QT prolongation with torsodes de pointes in a patient with pimozide intoxication. AM J Med Sci 309:315–316, 1995 Krittayaphong R, Cascio WE, Light KC, et al: Heart rate variability in patients with coronary artery disease: differences in patients with higher and lower depression scores. Psychosom Med 59:231–235, 1997 Krupp BH, Fogel BS: Motivational impairment in primary psychiatric and medical illness. Psychiatric Annals 27:34–38, 1997 Labbate LA, Ayd FJ: Ziprasidone: the latest FDA-approved atypical antipsychotic. International Drug Therapy Newsletter 36:25–32, 2001

❉

351 ❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

Laghrissi-Thode F, Pollock BG, Miller M, et al: Comparative effects of sertraline and nortriptyline on body sway in older depressed patients. Am J Geriatr Psychiatry 3:217–218, 1995 Laghrissi-Thode F, Wagner WR, Pollock BG, et al: Elevated platelet factor 4 and betathromboglobulin plasma levels in depressed patients with ischemic heart disease. Biol Psychiatry 42:290–295, 1997 LeBoff MS, Kohlmeier L, Hurwitz S, et al: Occult vitamin D deficiency in postmenopausal U.S. women with acute hip fracture. JAMA 281:1505–1511, 1999 Linden W, Stossel C, Maurice J: Psychosocial interventions for patients with coronary artery disease: a meta-analysis. Arch Intern Med 156:745–752, 1996 Lindenmayer J-P, Patel R: Olanzapine-induced ketoacidosis with diabetes mellitus (letter). Am J Psychiatry 156:1471, 1999 Littman AB, Ketterer MW: Behavioral medicine, in Textbook of Consultative-Liaison Psychiatry. Edited by Rundell JR, Wise MG. Washington, DC, American Psychiatric Press, 1996, pp 1080–1109 Littman AB, Fava M, McKool K, et al: Buspirone therapy for type A behavior, hostility and perceived stress in cardiac patients. Psychother Psychosom 59:107–110, 1993 Low RA, Fuller MA, Popli A: Clozapine induced atrial fibrillation (letter to editor). J Clinical Psychopharmacol 18:170, 1998 Lustman PJ, Griffith LS, Clouse RE: Depression in adults with diabetes: results of 5year follow-up study. Diabetes Care 2:605–12, 1988 Lustman PJ, Griffith LS, Gavard JA, et al: Depression in adults with diabetes. Diabetes Care 15:1631–1639, 1992 Lustman PJ, Eisen SA, Griffith LS: Effects of alprazolam on glucose regulation in diabetes, results of a double-blind, placebo-controlled trial. Diabetes Care 18:1133–1139, 1995 Lustman PJ, Griffith LS, Clouse RE, et al: Effects of nortriptyline on depression and glycemic control in diabetes: results of a double-blind, placebo-controlled trial. Psychosom Med 59:241–250, 1997a Lustman PJ, Griffith LS, Freedland KE, et al: The course of major depression in diabetes. Gen Hosp Psychiatry 19:138–143, 1997b Lustman PJ, Freedland KE, Griffith LS, et al: Predicting response to cognitive therapy of depression in type 2 diabetes. Gen Hosp Psychiatry 20:302–306, 1998 Maheux P, Ducros F, Bourque J: Fluoxetine improves insulin sensitivity in obese patients with non–insulin-dependent diabetes mellitus independently of weight loss. International Journal of Obesity 21:97–102, 1997 Marcus MD, Wing RR, Guare J: Lifetime prevalence of major depression and its effect on treatment outcome in obese type II diabetes patients. Diabetes Care 15:253– 255, 1992 Marks IM: Behavioral aspects of panic disorder. Am J Psychiatry 144:1160–1165, 1987

❉

352 ❉

❉ The Difficult-to-Treat Psychiatric Patient With Comorbid Medical Illness ❉ Max MB, Lynch SA, Muir J, et al: Effects of desipramine, amitriptyline and fluoxetine on pain in diabetic neuropathy. N Eng J Med 1992:1250–1256, 1992 Medical Letter: Fluoxetine (Prozac) and other drugs for treatment of obesity. Med Lett Drugs Ther 36:107–108, 1994 Medical Letter: Midodrine for orthostatic hypotension. Med Lett Drugs Ther 39:59– 60, 1997 Melkersson KI, Hulting A-L, Brismar KE: Different influences on classical APs and clozapine on glucose-insulin homeostasis in patients with schizophrenia or related psychoses. J Clin Psychiatry 60:783–791, 1999 Mendelson WB, Martin JV, Rapaport DM: Effects of buspirone on sleep and respiration. American Review of Respiratory Disease 141:1527–1530, 1990 Mendoza L, Goodnick PJ, Kumor A: Sertraline in diabetic neuropathy. New Research Report 116 presented at the annual meeting of the American Psychiatric Association, Toronto, Canada, May–June 1998 Menninger WC: Psychologic factors in the etiology of diabetes. J Nerv Ment Dis 8:1– 13, 1935 Metzger E, Friedman R: Prolongation of the corrected QT and torsades de pointes cardiac arrthymias associated with intravenous haloperidol in the medically ill. J Clin Psychopharmacol 13:128–132, 1993 Miniati M, Mauri M, Dell’Osso L, et al: Panic-agoraphobia spectrum and cardiovascular disease. CNS Spectrums 3:58–62, 1998 Minuchin S, Baker L, Rosman B, et al: A conceptual model of psychosomatic illness in children. Arch Gen Psychiatry 32:1031–1038, 1975 Mitchell-Heggs PK, Murphy K, Minty A, et al: Diazepam in the treatment of dyspnea in the “pink puffer” syndrome. QJM 49:9–20, 1980 Musselman DL, Tomer A, Manatunga AK, et al: Exaggerated platelet reactivity in major depression. Am J Psychiatry 153:1313–1317, 1996 Musselman DL, Evans DL, Nemeroff CB, et al: The relationship of depression to cardiovascular disease: epidemiology, biology, and treatment. Arch Gen Psychiatry 55:580–592, 1998 O’Connor CM: The clinical course and medical management of depressed patients with ischemic heart disease. Article presented at the annual meeting of the American Psychiatric Association, Toronto, Canada, May–June 1998 O’Kane M, Wiles PG, Wales JK: Fluoxetine in the treatment of obese type II diabetic patients. Diabetic Medicine 2:105–110, 1994 Ornish D, Brown SE, Scherwitz LW, et al: Can lifestyle changes reverse coronary heart disease: The Lifestyle Heart Trial. Lancet 336:129–133, 1990 Ornish D, Scherwitz LW, Billings JH, et al: Intensive lifestyle changes for reversal of coronary heart disease. JAMA 280:2001–2007, 1998 Papp LA, Weiss JR, Greenberg HE, et al: Sertraline for chronic obstructive pulmonary disease and comorbid anxiety and mood disorders (letter) Am J Psychiatry 152:1531, 1995

❉

353 ❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

Penninx BJH, Guralnik JM, Ferrucci L, et al: Depressive symptoms and physical decline in community-dwelling older persons. JAMA 279:1720–1726, 1998 Physicians’ Desk Reference, 55th Edition. Montvale, NJ, Medical Economics, 2001 Pies RW: Handbook of Essential Psychopharmacology, Washington, DC, American Psychiatric Press, 1998a Pies RW: Smoke, schizophrenia and cytochromes. Psychiatric Times Monograph, December 1998b, pp 22–23 Pies RW: Antimicrobial and psychiatry, part II: pharmacokinetic considerations. International Drug Therapy Newsletter 34:17–18, 1999 Pollack MH, Kradin R, Otto MW, et al: Prevalence of panic in patients referred for pulmonary function testing at a major medical center. Am J Psychiatry 153:110– 113, 1996 Preda A, Fazeli A, MacKay BG, et al: Lamotrigine as prophylaxis against steroidinduced mania (letter). J Clin Psychiatry 60:708–709, 1999 Preskorn SH: Clinical Pharmacology of Selective Serotonin Reuptake Inhibitors. Caddo, OK, Professional Communications, 1996 Preskorn SH: Selection of an antidepressant: mirtazapine. J Clin Psychiatry 58 (suppl 6):3–8, 1997 Procyshyn RM, Pande S, Tse G: New-onset diabetes mellitus associated with quetiapine. Can J Psychiatry 45:668–669, 2000 Rabkin J, Quitkin FM, McGrath, et al: Adverse reactions to monoamine oxidase inhibitors, part II: treatment correlates and clinical management. J Clin Psychopharmacol 5:2–9, 1985 Rasmussen B, Maenpoa J, Pelkonen O, et al: Selective serotonin reuptake inhibitors and theophylline metabolism in human liver microsomes: potent inhibition by fluvoxamine. Br J Clin Pharmacol 39:151–159, 1995 Rechlin T: The effect of amitriptyline, doxepin, fluvoxamine, and paroxetine treatment on heart rate variability. J Clin Psychopharmacol 14:392–395, 1994 Reddy S, Nobler MS: Dangerous hyperglycemia associated with electroconvulsive therapy. Convulsive Therapy 12:99–103, 1996 Reed RL, Mooradion AD: Management of diabetes mellitus in the nursing home. Annals of Long-Term Care 6:100–107, 1998 Robinson RG, Yates WR (eds): Psychiatric Treatment of the Medically Ill. New York, Marcel Dekker, 1999 Roose SP: Considerations for the use of antidepressants in patients with ischemic heart disease. Article presented at the annual meeting of the American Psychiatric Association, Toronto, Canada, May–June 1998 Roose SP, Glassman AH, Giardina EGV, et al: Nortriptyline in depressed patients with left ventricular impairment. JAMA 256:3253–3257, 1986 Roose SP, Glassman AH, Giardina EGV, et al: Tricyclic antidepressants in patients with cardiac conduction disease. Arch Gen Psychiatry 44:273–275, 1987

❉

354 ❉

❉ The Difficult-to-Treat Psychiatric Patient With Comorbid Medical Illness ❉ Roose SP, Dalack GW, Glassman AH, et al: Cardiovascular effects of bupropion in depressed patients with heart disease. Am J Psychiatry 148:512–516, 1991 Roose SP, Glassman AH, Attia E, et al: Cardiovascular effects of fluoxetine in depressed patients with heart disease. Am J Psychiatry 155:660–665, 1998a Roose SP, Laghrissi-Thode F, Kennedy JS, et al: Comparison of paroxetine and nortriptyline in depressed patients with ischemic heart disease. JAMA 279:287–291, 1998b Rundell JR, Wise MG (eds): Textbook of Consultation-Liaison Psychiatry. Washington DC, American Psychiatric Press, 1996 Schultz SK, Arndt S, Ho B-C, et al: Impaired glucose tolerance and abnormal movements in patients with schizophrenia. Am J Psychiatry 156:640–642, 1999 Schweiger U, Weber B, Deuschle M, et al: Lumbar bone mineral density in patients with major depression: evidence of increased bone loss at follow-up. Am J Psychiatry 157:118–120, 2000 Series F, Cormier Y: Effects of protriptyline on diurnal and nocturnal oxygenation in patients with chronic obstructive pulmonary disease. Ann Intern Med 113:507– 511, 1990 Siegel FP: Lithium for steroid-induced psychosis (letter). N Engl J Med 299:155–156, 1978 Shader RI: Approaches to the treatment of manic-depressive states, in Manual of Psychiatric Therapeutics, 2nd Edition. Edited by Shader RI. Boston, MA, Little, Brown, 1994, pp 247–258 Shader RI, Greenblatt DJ: Potassium, antipsychotic agents, arrhythmias, and sudden death (editorial). J Clin Psychopharmacol 18:427–428, 1998 Shader RI, Belfer ML, DiMascio A: Glucose metabolism, in Psychotropic Drug Side Effects, Clinical and Theoretical Perspectives. Edited by Shader RI, DiMascio A. Baltimore, MD, Williams & Wilkins, 1970, pp 46–62 Sindrup SH, Gram LF, Brosen K, et al: The selective serotonin reuptake inhibitor paroxetine is effective in the treatment of diabetic neuropathy symptoms. Pain 42:135–144, 1990 Singh NP, Despars JA, Stansbury DW, et al: Effects of buspirone on breathlessness and exercise tolerance in patients with chronic airflow obstruction and mild anxiety. Chest 103:800–804, 1993 Sipahimalani A, Masand PS: Olanzapine in the treatment of delirium. Psychosomatics 39:422–430, 1998 Sipahimalani A, Sime RM, Masand PS: Treatment of delirium with risperidone. International Journal of Geriatric Psychopharmacology 1:24–26, 1997 Skop BP, Brown TM: Potential vascular and bleeding complications of treatment with selective serotonin reuptake inhibitors. Psychosomatics 37:12–16, 1996 Smoller JW, Pollack MH, Otto MW, et al: Panic anxiety, dyspnea, and respiratory disease. Am J Respir Crit Care Med 154:6–17, 1996

❉

355 ❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

Smoller JW, Pollack MH, Systrom D, et al: Sertraline effects on dyspnea in patients with obstructive airways disease. Psychosomatics 39:24–29, 1998 Smyth JM, Stone AA, Hurewitz A, et al: Effects of writing about stressful experiences on symptom reduction in patients with asthma or rheumatoid arthritis: a randomized trial. JAMA 281:1304–1309, 1999 Sobel M, Jaggers ED, Franz MA: New-onset diabetes mellitus associated with the initiation of quetiapine treatment (letter). J Clin Psychiatry 60:556–557, 1999 Stahl SM: How to appease the appetite of psychotropic drugs. J Clin Psychiatry 59:500–501, 1998 Stahl SM: Dr. Stahl replies (letter). Antipsychotic drugs. J Clin Psychiatry 60 (suppl 21): 20–24, 1999 Stoudemire A, Fogel BS, Greenberg D (eds): Psychiatric Care of the Medical Patient, 2nd Edition. New York, Oxford University Press, 2000 Sussman N, Ginsberg D: Weight gain associated with SSRIs. Primary Psychiatry 5:28– 39, 1998 Sussman N, Ginsberg D: Effects of psychotropic drugs on weight. Psychiatric Annals 29:580–594, 1999 Taylor CB, Miller NH, Smith PM, et al: The effect of a home-based case-managed, multifactorial risk-reduction program on reducing psychological distress in patients with cardiovascular disease. J Cardiopulm Rehabil 17:157–162, 1997 Thapa PB, Gideon P, Cost TW, et al: Antidepressants and the risk of falls among nursing home residents. N Engl J Med 339:875–882, 1998 Thase ME: Effects of venlafaxine on blood pressure: a meta-analysis of original data from 3744 depressed patients. J Clin Psychiatry 59:502–508, 1998 Tielens JAE: Vitamin C for paroxetine-and fluvoxamine-associated bleeding (letter). Am J Psychiatry 153:883–884, 1997 Tulen JHM, Bruijn JA, deMan KJ, et al: Cardiovascular variability in major depressive disorder and effects of imipramine or mirtazapine. J Clin Psychopharmacol 16:135–145, 1996 Tun PA, Nathan DM, Perlmutter LC: Cognitive and affective disorders in elderly diabetes. Clin Geriatr Med 6:731–746, 1990 Unutzer J, Patrick DL, Simon G, et al: Depressive symptoms and the cost of health services in HMO patients aged 65 years and older. JAMA 277:1618–1623, 1997 Walley T, Pirmohamed, Proudlove C, et al: Interaction of metoprolol and fluoxetine. Lancet 341:967–968, 1993 Warnock JK, Biggs F: Nefazodone-induced hypoglycemia in a diabetic patient with major depression. Am J Psychiatry 154:288–289, 1997 Wilt JL, Minnema AM, Johnson RF, et al: Torsades de pointes associated with the use of intravenous haloperidol. Ann Intern Med 119:391–394, 1993 Wirshing DA, Spellberg BJ, Erhart SM, et al: Novel APs and new onset diabetes. Biol Psychiatry 44:778–783, 1998

❉

356 ❉

❉ The Difficult-to-Treat Psychiatric Patient With Comorbid Medical Illness ❉ Woods SW, Charney DS, Loke J, et al: Carbon dioxide sensitivity in panic anxiety. Arch Gen Psychiatry 43:900–909, 1986 Yellowlees PM, Kalucy RS: Psychobiological aspects of asthma and the consequent research implications. Chest 97:628–634, 1990 Yellowlees PM: The treatment of psychiatric disorders in patients with chronic airways obstruction. Med J Aust 147:349–352, 1987 Zandbergen J, Bright M, Pols H, et al: Higher lifetime prevalence of respiratory diseases in panic disorder? Am J Psychiatry 148:1583–1585, 1991 Zornberg GL, Jick H: Antipsychotic drug use and risk of first-time idiopathic venous thromboembolism: a case-control study. Lancet 356:1219–1223, 2000

❉

357 ❉

This page intentionally left blank

13 Special Somatic Treatments D.P. Devanand, M.D. Sarah H. Lisanby, M.D. Harold A. Sackeim, Ph.D.

In this chapter on somatic therapies, we will focus on electroconvulsive therapy (ECT) because of its longstanding history and its rich clinical and research database. Patients who are difficult to treat with medications and/ or psychotherapy are frequently referred for treatment with ECT. We will also discuss functional neurosurgery, vagus nerve stimulation, and transcranial magnetic stimulation as treatment strategies.

Supported in part by a NARSAD Independent Investigator Award to Dr. Devanand and NIMH grants R37 MH35636 and R10MH57009 to Dr. Sackeim

❉

359 ❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

Electroconvulsive Therapy ECT is the somatic treatment with the longest history of continuous use in psychiatry. Because ECT is rarely used as a first-line treatment, it could be argued that nearly all patients referred for ECT are difficult to treat. Patients are referred for ECT for several reasons: 1. Nonresponse or incomplete response to medication trials 2. Severity of illness necessitating rapid response in conditions that are known to respond well to ECT (e.g., psychotic depression with suicidality, severe depression accompanied by starvation, or catatonia) 3. Medical comorbidity that precludes the use of psychotropic medications that are otherwise indicated (e.g., nonresponse to selective serotonin reuptake inhibitors (SSRIs) followed by inability to use tricyclic antidepressants (TCAs) because of cardiac conduction disease) 4. Prior good response to ECT In addition, a large proportion of patients referred for ECT have failed to respond to medication trials with or without psychotherapy (Prudic et al. 1990, 1996). These difficult-to-treat patients have a range of diagnoses, with the most common being major depression, schizophrenia, and mania (discussion of these follows in later sections in this chapter). Other difficult-to-treat categories include patients with concomitant Axis II diagnoses, concomitant alcohol and substance abuse or dependence, double depression (major depression plus dysthymia), schizophrenia with predominantly negative symptoms and depression, secondary depression (secondary refers to the sequence in time of a nondepressive diagnosis followed by a depressive diagnosis), and severe medical comorbidity. Unfortunately, the literature on ECT in these areas is relatively sparse and largely uncontrolled (Black et al. 1988, 1993).

Prevalence of Difficult-to-Treat Patients Among Patients Referred for ECT It is commonly believed that the vast majority of patients referred for ECT are medication resistant. However, in patients with major depression, when strict criteria for an adequate medication trial are used, barely half the patients referred for ECT are rated as medication resistant (Prudic et al. 1990, 1996). This is especially true for patients with psychotic or delusional depression, who rarely reach adequate dosage levels of antipsychotic medica❉

360 ❉

❉

Special Somatic Treatments

❉

tions combined with antidpressants before treatment with ECT (Mulsant et al. 1997). These findings were obtained in tertiary academic settings, and it is likely that the prevalence of medication-resistant patients receiving ECT is even lower in other settings.

Cognitive Effects of ECT The cognitive side effects of ECT predominantly affect memory and are of two types: retrograde amnesia (memory of events before the treatment) and anterograde amnesia (retention of newly learned information). Whereas both deficits can be prominent during or immediately following the ECT course, these abate and often disappear in 2–3 months. Even patients who have received multiple courses of ECT do not show significant cognitive deficits on long-term follow-up (Devanand et al. 1991), and a comprehensive review of the literature concluded that there is no firm evidence indicating that ECT causes structural brain damage (Devanand et al. 1994). Nonetheless, there can be permanent, patchy memory loss for events that occurred in the few weeks to months preceding and during the treatment and shortly following the end of the ECT course (McElhiney et al. 1995; Sobin et al. 1995; Squire 1986; Weiner et al. 1986). Rarely, some patients continue to report persistent retrograde amnesia for events that occurred several months to years before the ECT course (Sackeim 1992; Weiner et al. 2001). In contrast, patients’ perceptions of their memory function is often markedly improved shortly following ECT (Coleman et al. 1996) and, in a variety of neuropsychological domains, objective performance is better on cognitive testing post-ECT compared with pre-ECT (Sackeim 1992; Sackeim et al. 1993). This improvement results from the enhancement of attention and concentration in depressed patients associated with clinical response. Prediction of cognitive impairment as a result of ECT has rarely been attempted. Patients with diminished global cognitive status (i.e., low MiniMental State Exam (MMSE) scores; Folstein et al. 1975) show improved global cognitive status following ECT but greater short- and long-term retrograde amnesia (Sobin et al. 1995). In addition, the time to recover full orientation following the first ECT treatment is a strong predictor of cognitive impairment, including persistent retrograde amnesia, following the ECT course (Sobin et al. 1995). Therefore, assessing the patient’s cognitive status carefully before and following early treatments in the ECT course can provide useful information about whether treatment parameters need to be adjusted. These adjustments to diminish the adverse cognitive effects should be weighed against the need to ensure that the treatment parameters are suf❉

361 ❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

ficient for clinical response (e.g., low-dose unilateral ECT given close to seizure threshold is not recommended because of its low response rate).

Technical Issues The minimum electrical dosage (seizure threshold) needed to elicit a grand mal seizure varies up to 50-fold across individuals (Coffey et al. 1995; Sackeim et al. 1987b, 1993, 2000). These differences imply that if all patients are treated with a fixed dose, as was commonly practiced in many centers, some patients may receive an electrical stimulus just above their seizure threshold and others may receive a stimulus at 10–20 times (or even greater) their seizure threshold, with consequent severe cognitive side effects. Therefore, to optimize the trade-off between efficacy and side effects, determination of seizure threshold at the first treatment is recommended, followed by an informed decision as to what dosage needs to be given at subsequent treatments. Empirical titration with stepwise increments in electrical dosage until a seizure occurs is currently the most accurate approach to determine seizure threshold (Weiner et al. 2001). Alternative methods take into account the evidence that seizure threshold is higher in males, in older people, and in those receiving bilateral compared with unilateral ECT (Coffey et al. 1995; Sackeim et al. 1987b). The older sine wave ECT devices have given way to brief pulse devices that administer a total electrical charge that is much lower and that leads to fewer and less severe cognitive side effects (Weiner et al. 1986). If, in addition, seizure threshold is assessed at the first treatment and an informed decision on dosage is made, the likelihood of profound disorientation with an organic brain syndrome caused by ECT is remote (Sackeim et al. 1993).

Comorbid Medical Conditions The death rate due to ECT is low, about 1 per 10,000 patients treated (Weiner et al. 2001). Nonetheless, close monitoring of elderly patients is required (Alexopoulos et al. 1984; Cattan et al. 1990). Depressed patients with cardiovascular disease who fail to respond to trials of SSRIs and newer agents can be treated with TCAs, but there are risks associated with TCA use if the patient has a cardiac conduction defect or is prone to orthostatic hypotension. These patients tend to be referred for ECT. There are no absolute contraindications to ECT (Weiner et al. 2001). The main medical risk of ECT is in patients with significant cardiovascular disease (Alexopoulos et al. 1984; Zielinski et al. 1993). During seizure elic❉

362 ❉

❉

Special Somatic Treatments

❉

itation there is a vagally mediated, parasympathetic surge with bradycardia, followed by a sympathetic surge with tachycardia and hypertension (Pitts 1982). These changes can manifest as pronounced bradycardia or asystole, or hypertension with tachycardia that can be quite dramatic (e.g., BP 260/ 140 mm Hg, heart rate 160–180 beats per minute). Fortunately, in most cases these changes are transient, and vital signs spontaneously return to normal within a few minutes. In a comparison of 40 depressed patients with severe cardiac disease and 40 depressed patients without cardiac disease, there were no deaths reported. The type of preexisting cardiac abnormality predicted the type of complication that occurred during the ECT course (Zielinski et al. 1993). Patients with ischemic heart disease, including a history of myocardial infarction, were prone to ischemic complications. Patients with preexisting atrial or ventricular arrhythmias were prone to arrhythmic complications acutely post-ECT. These findings suggest that besides the monitoring of vital signs and a three- or five-lead electrocardiogram, anticipatory interventions (e.g., applying nitropaste to the skin before ECT in patients with known ischemic heart disease) may be useful. Whereas some clinicians have advocated the routine use of β-blockers or calcium channel blockers (Figiel et al. 1993; Stoudemire et al. 1990), one concern is that the blocking of sympathetic response with its associated tachycardia and hypertension may leave an unopposed parasympathetic response that could lead to bradycardia and sinus arrest (Decina et al. 1984). Patients with pacemakers and implanted ventricular defibrillators have been safely treated with ECT (Lisanby et al. 1996), but a magnet needs to be available to externally turn the demand pacemaker or defibrillator on or off as required. In such cases, consultation by a cardiologist is strongly advised. The old adage that raised intracranial pressure is an absolute contraindication to ECT is no longer valid. There are now several reports of patients with intracranial neoplasms and both intracranial and extracranial aneurysms successfully treated with ECT (Devanand et al. 1990; Husum et al. 1983). It appears that the rise in intracranial pressure is secondary to the peripheral rise in blood pressure and the transient breakdown of the bloodbrain barrier that occurs during seizure elicitation (Laursen et al. 1991; Petito et al. 1977), with consequent extravasation of fluid into the extracellular space. Therefore, using antihypertensive medications to prevent the peripheral increase in blood pressure and heart rate during ECT may reduce the secondary rise in intracranial pressure. An additional factor that has made it possible to treat patients with intracranial lesions is the availability of computed tomography (CT) and magnetic resonance imaging (MRI), which ❉

363 ❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

decrease the likelihood of missing these lesions and not preparing adequately for the ECT procedure. Less medically dangerous conditions include a poor dentition that sometimes requires the use of special mouth guards to avoid a loose tooth from breaking or falling out with the consequent risk of aspiration and the use of concomitant medications that could complicate ECT. Theophylline, which is proconvulsant, has been associated with the development of status epilepticus during ECT even when plasma levels are in the therapeutic range for treatment of pulmonary disease (Devanand et al. 1987; Weiner et al. 2001). Therefore, the use of theophylline to increase seizure length, as some have advocated (Swartz and Lewis 1991) is fraught with risk, especially because the therapeutic advantage gained by lengthening seizures is highly questionable (Sackeim et al. 1993). Anticoagulants like warfarin can be used safely provided prothrombin time is monitored (Petrides and Fink 1996). Potential gastrointestinal reflux in obese or pregnant patients usually requires pretreatment with antacids (Miller 1994). In general, ECT is remarkably safe, and the presence of comorbid medical conditions rarely poses undue risk, with the possible exception of severe cardiac disease.

Major Depression In major depression, the first-line medication is usually an SSRI. If a severely depressed inpatient fails to respond to an SSRI, there are two main options: another medication trial (e.g., TCA) or an augmentation (e.g., lithium) strategy, or ECT. SSRI nonresponse leading to ECT treatment is quite common in clinical practice (Prudic et al. 1996). This trend may be caused by the time pressure in an inpatient hospital stay, which leads to ECT becoming the second-line treatment in preference to a second-line medication trial. An alternative explanation is that familiarity and comfort with the use of TCAs has declined among psychiatrists in recent years, partly because of the complex cardiovascular side effects of these drugs, which require careful monitoring. Some of these SSRI-resistant patients who are referred directly to ECT might have responded to an adequate trial of a TCA, with or without lithium augmentation, because there is some evidence that melancholic patients may preferentially respond to a TCA compared with an SSRI (Roose et al. 1994). The Difficult-to-Treat Patient With Major Depression Major depression is the most common diagnosis (80%) among patients who receive ECT (Thompson and Blaine 1987; Thompson et al. 1994). After a ❉

364 ❉

❉

Special Somatic Treatments

❉

number of early open trials showed response rates of 80%–100% in depressive disorders, several studies comparing “real” ECT to “sham” ECT (anesthesia without delivering an electrical stimulus) consistently showed a significant advantage for real ECT (Brandon et al. 1984; Freeman et al. 1978; Gregory et al. 1985; Johnstone et al. 1980). There is also some evidence that ECT is superior to antidepressant medications in the treatment of major depression, although the few published studies have had methodological limitations (Dinan and Barry 1989; Gangadhar et al. 1982; Greenblatt et al. 1964; Medical Research Council 1965). Perhaps the most convincing evidence for the efficacy of ECT in major depression is that varying the technique of administering ECT markedly alters the likelihood of treatment response. In an initial randomized double-blind study of 52 patients with major depression, administering ECT at or just above an individual’s threshold led to a 70% response rate for bilateral ECT compared with a 28% response rate for unilateral ECT (Sackeim et al. 1987b). In a subsequent study of 96 patients that employed the same two “low-dose” conditions and two higher dose conditions (“high-dose” unilateral and bilateral ECT, each administered at 150% above the individual’s seizure threshold), response rates were 17% for low-dose unilateral ECT, 43% for high-dose unilateral ECT, 65% for low-dose bilateral ECT, and 63% for high-dose bilateral ECT (Sackeim et al. 1993). A more recent study (Sackeim et al. 2000) found that increasing unilateral ECT dose to 500% above seizure threshold led to comparable efficacy to bilateral ECT at 150% above seizure threshold while retaining an advantage with respect to cognitive side effects. These data suggest that treatment using unilateral ECT at several times the patient’s seizure threshold achieves the optimal clinical trade-off between efficacy and cognitive side effects. ECT for Patients With Severe Depressive Illness ECT is particularly effective in patients with severe compared with mild depressive illness (DeCarolis et al. 1964). Patients with psychotic depression also have a robust response to ECT (Nobler and Sackeim 1996), perhaps because they are referred early for ECT and many are not medication resistant, a factor associated with relatively poor response (Mulsant et al. 1997; Prudic et al. 1996). For patients who are severely suicidal or on the brink of starvation, the response to ECT can be rapid and dramatic. Clinically, there can be difficulty in obtaining the severely ill patient’s informed consent and preventing dropout during the ECT course. The feeling of hopelessness, which can reach delusional proportions, sometimes makes it difficult for the ❉

365 ❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

patient to consider or accept treatment of any type, be it medications or ECT. In all these near-emergency or emergency situations, moderately suprathreshold bilateral ECT (150% above seizure threshold) or very highdose unilateral ECT (at least five times seizure threshold) is indicated. These recommendations are based primarily on clinical experience because these severely ill patients do not present in large enough numbers to conduct randomized clinical trials, and comparing active treatment to placebo/sham treatments in severely ill patients could raise ethical concerns. Medication Resistance in Major Depression Optimistic estimates of the efficacy of ECT in major depression have been tempered by new findings showing that medication-resistant patients are less likely to respond, and more likely to relapse, than patients who did not receive an adequate medication trial before ECT (therefore their medication resistance status is unknown). In 53 depressed patients treated with bilateral ECT, the response rate was only 50% in patients with established medication resistance compared with 86% in patients who had not received an adequate medication trial before ECT (Prudic et al. 1990). These data have been replicated in a larger, prospective study (Prudic et al. 1996). In an overlapping series of 58 patients, following clinical response to ECT, only 32% of nonresistant patients relapsed in the first year of follow-up, compared with 64% of patients who had been identified as medication-resistant before the ECT course (Sackeim et al. 1990). Astonishingly, the type of medication trial post-ECT was only marginally related to the risk of relapse; it was the medication resistance before ECT that was the main predictor. Recent data support the view that if left untreated, most patients relapse during the first 6 months post-ECT (Sackeim et al. 2001). In a multicenter trial of depressed patients who responded to ECT, patients randomized postECT to placebo had a relapse rate of 84%, those randomized to nortriptyline had a relapse rate of 60%, and those randomized to nortriptyline plus lithium had a relapse rate of 39% (Sackeim et al. 2001). The combination of nortriptyline and lithium is recommended at this stage of knowledge; ongoing trials comparing continuation ECT with medication treatment should help clarify the optimal approach to treating the medication-resistant patient following reponse to ECT. The high rate of relapse following ECT has raked up an old debate about whether antidepressant medications should be stopped before and during the ECT course. This issue remains to be resolved empirically (Sackeim et al. 1990). Theoretically, it is possible that starting the right antidepressant ❉

366 ❉

❉

Special Somatic Treatments

❉

medication during the ECT course will decrease the likelihood of subsequent relapse. Because there is generally no increase in medical risk when combining antidepressant medications with ECT, it is reasonable to continue antidepressants during the ECT course or to begin antidepressant treatment before ending the ECT course, especially if rapid post-ECT relapse is anticipated. Continuation or Maintenance ECT in Patients With Major Depression There are depressed patients who respond to ECT and ECT alone. These patients suffer repeated relapses on continuation pharmacotherapy, and continuation ECT is sometimes their only effective option. This approach involves continuing ECT at a lower frequency than used during the acute course and can range from once a week to once every 1–2 months. The frequency depends on clinical status, with increasing frequency when the patient worsens and decreasing frequency when the patient maintains remission. A large number of patients have been treated in this manner with reported good results, but published reports have been based on uncontrolled studies (Geoghegan and Stevenson 1949; Petrides et al. 1994).

Schizophrenia Although ECT was initially developed as a treatment for schizophrenia, its use in this context has declined during the last few decades. The advent of typical antipsychotics, and more recently atypical antipsychotics, has led to a decreased need for ECT. The literature on ECT in schizophrenia has been the subject of two major reviews in recent years (Fink and Sackeim 1996; Krueger and Sackeim 1995). The extant data support the use of ECT, where variable response rates have rivaled those of traditional neuroleptics in some studies (Krueger and Sackeim 1995; May 1968; Small 1985). In schizophrenia, controlled comparisons of the relative efficacy of ECT and atypical neuroleptics remain to be conducted. Another question is whether ECT can serve as an augmenting agent to improve clinical response to neuroleptic treatment. After a preliminary report (Gujavarty et al. 1987), other studies have confirmed an augmenting effect for ECT in patients who receive traditional neuroleptics (Chanpattana et al. 1999; Small et al. 1982). It is important to keep in mind that schizophrenia is a chronic illness and that after acute response to ECT, continuation treatment strategies are critical. ECT is indicated in the following three situations in patients with schizophrenia: prominent affective symptomatology, catatonia, and history ❉

367 ❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

of prior good response to ECT (Weiner et al. 2001). Patients with schizoaffective disorder appear to have a response rate intermediate between affective disorder and schizophrenia (Dodwell and Goldberg 1989), but controlled studies are lacking. Catatonia of all types is responsive to ECT, suggesting that ECT may reverse the neurobiological abnormalities, presumably involving the basal ganglia and other motor centers that result in catatonia. As with any other psychiatric diagnostic category, history of good response to a treatment, in this case ECT, should lead to preference for the same efficacious treatment in the index episode or index exacerbation. One category that is notably absent from the American Psychiatric Association Task Force (Weiner et al. 2001) list of recommendations is the schizophrenic patient with prominent negative symptoms. In this situation, the data for the use of ECT are not very convincing, although differentiating this type of patient from a schizophrenic patient with secondary depression can present a diagnostic dilemma. The Difficult-to-Treat Schizophrenic Patient Nowadays, virtually all patients with schizophrenia who are referred for ECT are difficult to treat and medication resistant. In contrast, many of the early studies used ECT as a first- or second-line treatment in schizophrenia (Pippard and Ellam 1981; Royal College of Psychiatrists 1995). It is likely that these early studies included a far broader group of patients under the “schizophrenia” label because of the diagnostic systems that were in use at that time. To help answer the question about the efficacy of ECT in schizophrenia, two studies evaluated real versus sham ECT (Brandon et al. 1985; Taylor and Fleminger 1980). This research found superior efficacy for real ECT, but this difference diminished by several months post-ECT. By that time, of course, most patients had ceased to receive ECT and were receiving various pharmacological regimens. These findings raise questions about the persistence of clinical improvement following ECT in patients with schizophrenia. Interestingly, in the Brandon and colleagues (1985) study, patients who showed the best response came under the category of “acute” schizophrenia, a finding consistent with the early literature (Brill et al. 1957; Ulett et al. 1956). Recently, Chanpattana and colleagues (1999) conducted a controlled trial of continuation ECT in 114 treatment-resistant patients with schizophrenia. To enter the study, patients had to receive and fail one or more adequate trials with antipsychotic medication, including a subgroup that had failed to respond to clozapine. Patients then received a standardized course ❉

368 ❉

❉

Special Somatic Treatments

❉

of treatment with flupenthixol (12–24 mg/day) and bilateral ECT; 58 (50.9%) responded. After a stabilization period of a few weeks, during which response had to be maintained, 51 patients were randomized to continuation treatment with the neuroleptic alone, ECT alone, or the combination treatment. Relapse over 6 months was significantly greater with either form of monotherapy (93%) compared with the combination treatment (40% relapse). This study, which is the largest of its type, suggests that combination treatment with ECT and antipsychotics is effective in the management of both acute psychotic symptoms and relapse prevention in patients with schizophrenia.

Mania Although ECT was widely used for mania in the past (Black et al. 1987b; McCabe and Norris 1977), and is still used for medication-resistant and severely ill manic patients, controlled data from clinical trials are fairly limited. This is primarily because of the practical difficulties in conducting randomized trials in manic patients, both with respect to consent for a research study and the severity of illness that makes it difficult to maintain patients in a placebo or sham ECT condition. In the largest controlled study, 34 patients were randomized to receive either bilateral ECT (average nine treatments) or lithium with blood-level monitoring for 8 weeks (Small et al. 1988). Both groups showed comparable response rates (lithium 81.2% and ECT 94.9%), but the use of concomitant neuroleptics was required and a high dropout rate limited interpretation of the findings. Nonetheless, ECT resulted in quicker symptomatic improvement. Other studies with smaller sample sizes support the efficacy of ECT in mania (Milstein et al. 1987; Mukherjee et al. 1988; Sikdar et al. 1994). Given the paucity of controlled data, a comprehensive review of both uncontrolled and controlled studies was conducted (Mukherjee et al. 1994). This review revealed that a total of 589 manic patients treated with ECT averaged 80% response across these different studies. In summary, ECT is clearly efficacious in the treatment of acute mania. The Difficult-to-Treat Manic Patient After the advent of lithium and neuroleptics, and more recently anticonvulsants, the use of ECT in mania has declined considerably. However, patients whose manic symptoms do not respond, or do not respond sufficiently, to these medications are sometimes referred for ECT. The literature suggests ❉

369 ❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

that all subtypes of mania (e.g., psychotic or nonpsychotic) respond well to ECT. One limitation in the use of ECT in mania pertains to the use of concomitant medications during the ECT course. Neuroleptics have a marginal effect in lowering seizure threshold, usually do not need to be discontinued, and may be partly synergistic (Krueger and Sackeim 1995). Anticonvulsants, by definition, need to be tapered and stopped before the ECT course because the elicitation of a grand mal seizure is central to the treatment. Lithium also needs to be preferably stopped, or tapered to a very low dose, because the interaction between lithium and ECT can result in the development of severe cognitive impairment characterized by memory loss and a delirious state and increased risk of status epilepticus (Weiner et al. 2001). Medication Resistance in Mania Medication resistance is perhaps the most common reason for referral of a manic patient for ECT. Clinical reports indicate good response to ECT in medication-resistant manic patients, but this has not been studied systematically. There is controversy concerning the efficacy of right unilateral ECT in mania (Mukherjee et al. 1994), and bilateral ECT is usually preferred because of its well-established therapeutic profile. However, high-dose unilateral ECT (e.g., five times threshold) can be administered if the potential cognitive deficits associated with bilateral ECT are a concern. Response to ECT may be particularly rapid in mania (Perris and d’Elia 1971), with evidence of symptom improvement after the first or second ECT treatment and complete response after as few as five to six treatments. As with other major psychiatric disorders, maintaining remission is a critical task. For continuation treatment after acute response, this may involve the administration of different antimanic medications from those used before the ECT course. In some patients, maintenance or continuation ECT may be required. Controlled data are not available, but clinical experience suggests that these alternatives are feasible and often successful in preventing relapse. Because most manic episodes are intrinsically time limited, the need for a continuation ECT regimen needs to be assessed on an ongoing basis. As with major depression, one strategy is to gradually increase the interval between ECT treatments during the continuation phase (e.g., once a week to once every 2 weeks to once a month). If the patient maintains a sustained remission, then continuation ECT can be discontinued. However, patients often will still require maintenance treatment with mood stabilizers to prevent relapse. ❉

370 ❉

❉

Special Somatic Treatments

❉

ECT in Difficult-to-Treat Patients With Other Disorders Patients with other disorders are occasionally referred for ECT. The limited research database on the use of ECT in some relatively common disorders will be summarized. Chart review studies suggest that in schizoaffective disorder, response rates to ECT are approximately 50% (Tsuang et al. 1979). Similar findings have been reported for the use of ECT in secondary depression, where depression follows (in time) another Axis I or II disorder, including alcohol or substance abuse (Black et al. 1987a; Zorumski et al. 1986). Therefore, depression with comorbid personality disorder is not a contraindication to ECT. Indeed, one report suggests that short-term response to ECT is not diminished in depressed patients with Axis II disorder, but these patients may be at higher risk for relapse (Zimmerman et al. 1986). Patients with double depression (i.e., major depression superimposed on an underlying dysthymia) do respond well to ECT, but in some cases the dysthymia persists (Prudic et al. 1993). Conversely, some patients with double depression remit completely with ECT, raising questions about the diagnostic accuracy of a prior chronic condition like dysthymia. Obsessive-compulsive disorder (OCD) was considered particularly resistant to ECT treatment, but more recent data have called this dogma into question (Khanna et al. 1988). Clinically, it is reasonable to consider ECT in patients with concomitant depression and OCD, but its efficacy in pure refractory OCD is not established. For all chronic psychiatric disorders, adequate forethought should be given to the treatment strategy following the ECT course if the patient’s condition responds (e.g., new medication regimens or continuation ECT may be required). Parkinson’s disease patients who become depressed do show good response to ECT in many cases, both in their depressive and motor symptomatology. Case reports and open trials have reported efficacy for ECT in treating the movement disorder of Parkinson’s disease (Abrams 1989; Douyon et al. 1989). However, in the only double-blind, controlled, real-versussham ECT study to date, real ECT showed an advantage in only one of the parameters studied (on-off phenomenon) in 11 patients with Parkinson’s disease (Andersen et al. 1987). Two potential pitfalls should be considered before ECT is given to treat the motor deficits of Parkinson’s disease: ECT is short acting, whereas Parkinson’s disease is a lifelong condition, and there is preliminary evidence that depressed patients with basal ganglia lesions (not all were Parkinson’s patients) are prone to develop severe cognitive impair❉

371 ❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

ment following ECT (Figiel et al. 1990). Unfortunately, the effects of ECT on cognitive performance remain to be studied systematically in Parkinson’s patients. There is a case literature indicating that the beneficial effects of ECT in Parkinson’s disease can be sustained by continuation/maintenance ECT (Pridmore et al. 1995). In tardive dyskinesia, evidence for the efficacy of ECT is based on a few case reports and remains equivocal (Flaherty et al. 1984; Mukherjee and Debsikdar 1994). In depressed patients with profound cognitive impairment, the so-called pseudodementia syndrome, there are case reports of the successful use of ECT with remission of both depressive and cognitive deficits (Devanand and Krueger 1994). However, in such a situation an accurate diagnosis is essential before ECT is considered, because there is the risk of worsening cognition if the patient actually suffers from early dementia. There is no upper age limit for ECT, and the assessment should depend on comorbid illnesses and medications rather than age itself. In children and adolescents, there are few data on efficacy in major depression, and there is some concern about the negative cognitive impact of ECT in this vulnerable population (Weiner et al. 2001). However, case reports suggest strong efficacy for ECT in major depression and catatonia in this age group (Griesemer et al. 1997). Pregnancy presents unusual risks: teratogenic effects of repeated barbiturate anesthesia on the fetus and gastrointestinal and cardiovascular risks for the mother. In reality, several hundred pregnant patients have been treated with ECT, and complications, primarily abruptio placentae, have been rarely reported (Miller 1994). Clinical Vignette: A Medication-Resistant Patient With Cardiac Complications Mr. A was an 85-year-old man with a history of recurrent major depression, coronary artery disease status post myocardial infarction, ischemic cardiomyopathy with an ejection fraction of 13% on cardiac catheterization, and complete heart block managed with a permanent pacemaker for the past 10 years. Two years before this psychiatric admission, he began to experience recurrent ventricular tachycardia, including an episode of sustained ventricular tachycardia accompanied by severe hypotension with systolic blood pressure of 40 mm Hg. He was successfully cardioverted, and an automatic intracardiac defibrillator (ICD) was implanted. Mr. A had two prior episodes of depression, with onset of the first episode at age 60. Each episode was treated with ECT. His most recent depressive episode began shortly after the implantation of his ICD and had lasted for nearly 2 years before admission. His depressive symptoms remained unresponsive to adequate medication trials, which included various combina-

❉

372 ❉

❉

Special Somatic Treatments

❉

tions of fluoxetine, lithium, buproprion, methylphenidate (15 mg/day), and a monoamine oxidase inhibitor. He was admitted, and after washout of his prior medications, he failed to respond to an adequate trial of fluoxetine 60 mg/day. Mr. A then signed informed consent for a course of ECT. After medical and cardiology consultations and a 7-day washout from fluoxetine he received a course of bilateral ECT. His medications during the ECT course included lorazepam as needed (0.5 mg qhs), betoptic drops 0.5% bid, captopril 12.5 mg bid, furosemide 20 mg qd, KCl 10 mEq qd, and warfarin 2.5 mg and 5 mg on alternate days. Bilateral ECT was administered three times per week. Atropine (0.4 mg iv) was given 2 minutes before anesthesia induction. Methohexital (50 mg, 0.75 mg/kg) and succinylcholine (60 mg) were used as the intravenous anesthetic agents. The patient was oxygenated from anesthetic administration until the return of spontaneous respiration. Seizure duration was monitored by bifrontal electroencephalogram (EEG) and motor manifestations using the blood pressure cuff technique (inflating cuff to well above systolic to prevent succinylcholine from entering the limb). Seizure threshold was determined at the first and last treatments by an empirical titration procedure involving the administration of subconvulsive stimuli with predetermined increases in stimulus intensity until a generalized seizure longer than 25 seconds by motor or EEG duration was elicited (Sackeim et al. 1987b). Initial seizure threshold was 896 mC (140 Hz frequency, 2.0 ms pulse width, 2.0 sec train duration, 0.8 A current). After 12 treatments at 1092 mC (22% above initial seizure threshold), retitration revealed that Mr. A’s threshold had risen to 1092 mC (140 Hz frequency, 1.5 ms pulse width, 3.25 sec train duration, 0.8 A current). Including the two titration sessions, he received a total of 14 treatments. The cardiology consultant was present for the first several treatments to monitor the manipulations of the ICD and pacemaker. An external defibrillator was charged and kept readily accessible during all treatments. In the event an arrhythmia occurred, an external defibrillator would have been needed because of the time delay inherent in reactivating the ICD. At each treatment, the ICD was deactivated with a magnet immediately preECT and then reactivated postictally. A second magnet was taped in place over the pacemaker immediately pre-ECT. While in place, the magnet converted the VVI (demand) pacemaker to a DD (fixed rate) pacemaker with a rate of 100. The magnet was removed posttreatment, and the pacemaker returned to VVI mode with a rate of 80/minute. The treatments were generally well tolerated and there were no adverse sequelae. During the immediate postictal period of two ECT treatments Mr. A developed a brief run of multifocal premature ventricular contractions and bigeminy that remitted spontaneously. At the eighth treatment, he developed a brief run of nonsustained ventricular tachycardia in the postictal period that self-corrected with no further management. During the ECT course, the patient experienced two episodes of ICD firing while on the inpatient unit with no significant side effects or medical

❉

373 ❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

sequelae. The patient’s ICD and pacemaker were evaluated in the cardiac electrophysiology lab on a weekly basis and were found to be in working order with no adverse effects from the treatments. Mr. A’s depressive symptoms began to improve after the fifth treatment, and this improvement continued gradually throughout the ECT course. His modified MMSE score midtreatment was 43 (out of 57) and increased to 49 at the end of treatment when his depression had remitted. Maintenance therapy with nortriptyline and fluoxetine was initiated before discharge, and he maintained improvement during the first few weeks of follow-up. Long-term follow-up was not conducted.

This case illustrates several points. First, patients with even severe cardiac disease can be successfully treated with ECT provided cardiology consultation and appropriate monitoring are used (Zielinski et al. 1993). In essence, there is no absolute contraindication to ECT (Weiner et al. 2001). Second, this patient had an extremely high seizure threshold. This is common in elderly males, and the electrical stimulus intensity usually needs to be high in such cases to elicit a seizure. Third, his cognitive performance as measured by the MMSE actually improved during treatment. Although this may seem odd, this outcome is more the norm than the exception when modern ECT is conducted with a brief pulse device with adequate anesthesia and oxygenation. This is because of an interaction between depression and cognitive impairment in patients who receive ECT: global cognitive performance may improve secondary to lifting of the depression, but specific anterograde and retrograde memory deficits may develop that are largely shortterm in duration (Sackeim et al. 1993). These types of deficits can be elicited by neuropsychological testing, and the recall item (three out of three objects at 5 minutes) of the MMSE may provide a more useful indicator than the total MMSE score. Finally, this patient, who was medication resistant, showed good clinical response to ECT. As discussed earlier, such patients can be difficult to treat, and when they do respond to ECT, the importance of preventing relapse by using medications or continuation ECT cannot be overemphasized.

Functional Neurosurgery The aim of functional neurosurgery in psychiatry is to alter the physiology of specific neural systems to treat specific disorders (Diering and Bell 1991). Previously referred to as psychosurgery, functional neurosurgery is currently conducted in a very small number of patients with treatment-resistant mood, anxiety, and other psychiatric disorders (Yudofsky and Ovsiew 1990). ❉

374 ❉

❉

Special Somatic Treatments

❉

Although originally used extensively in schizophrenia, functional neurosurgery has generally been found to be more effective in major depression and OCD (Ballantine and Giriunas 1979).

Current Stereotactic Neurosurgical Procedures for Psychiatric Disorders Modern stereotactic approaches have reduced morbidity and produce only transient effects on attention, without reported long-term effects on neurological status, higher brain functions, or personality (Corkin et al. 1979; Maxwell 1993). Nonetheless, a significant percentage of patients may have a new onset seizure disorder following these surgical interventions. Three of the more common procedures (cingulotomy, subcaudate tractotomy, anterior capsulotomy) are described in this section (see Mindus and Jenike 1992 for a review). Other procedures include amygdalotomy (for aggression), hypothalamotomy, thalamotomy, and surgery for various combinations of limbic structures (Ballantine and Giriunas 1979). Whereas modern stereotactic technique and standardized psychiatric diagnosis have improved the generalizability of these studies, most remain uncontrolled and retrospective. Anterior Cingulotomy Anterior cingulotomy refers to the bilateral severing of the anterior supracallosal fibers of the anterior cingulate, altering connections within the limbic system. This procedure has been reported to be efficacious in the treatment of mood disorders (Ballantine et al. 1987), anxiety disorders (Ballantine et al. 1977), and intractable pain (Foltz and White 1962). Therapeutic results are generally better in the treatment of major depression (about 60%) than in OCD and, importantly, appear only after some considerable postsurgery delay. Ballantine and colleagues (1977) reported that 75% of 154 patients with mood disorder improved after bilateral cingulotomy. A retrospective review by Jenike and colleagues (1991) found that 25%–30% of OCD patients benefited. The therapeutic effect in OCD is thought to be caused by interruption of abnormal fronto-striatal-thalamic activity, as suggested by PET imaging studies (Martuza et al. 1990). The claim that destruction of the anterior cingulate is associated with antidepressant effects is compatible with some findings from imaging studies of sleep deprivation (Wu et al. 1992), ECT (Scott et al. 1994), and pharmacological treatment (Bench et al. 1995), indicating reduced functional activity in this region concomitant with antidepressant response. ❉

375 ❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

Stereotactic Subcaudate Tractotomy A small literature describes the use of stereotactic subcaudate tractotomy (SST) (Bridges et al. 1994). SST transects the subcaudate region of the orbital gyrus, thereby transecting thalamo-anterior cingulate fibers (Broseta et al. 1979). The procedure of Knight (1965) involved the use of implanted radioactive yttrium (90y) beads. About 70%–80% of patients with major depression were reported as improved, with lower success rates of about 50% in anxiety disorders (Broseta et al. 1979; Martuza et al. 1990). Bridges and colleagues (1994) described their experience in performing over 1,300 SST procedures and suggested that episodic unipolar depression and history of good response to ECT were associated with good SST outcome. Anterior Capsulotomy Anterior capsulotomy entails the destruction of the anterior limb of the internal capsule, which severs thalamo-orbitofrontal fibers (Hay 1993). This may be accomplished via radiofrequency heat lesions (Bingley et al. 1973) or gamma irradiation using a stereotactic gamma unit (Leksell and Backlund 1979; Rylander 1979). This procedure has been reported to have efficacy in OCD with about a 70% success rate (Martuza et al. 1990). There have been some reports of personality change, lack of initiative, and mood elevation following capsulotomy (Hay 1993; Martuza et al. 1990) and concern about long-term distal effects in cortex, suggestive of edema, following large lesions produced with gamma irradiation.

Summary The literature on the use of surgical ablation techniques to treat psychiatric disorders is small, and because all procedures other than gamma irradiation involve breaching the skull, there has been little opportunity for sham-controlled investigation. The retrospective nature of the clinical reports and the gravity of the intervention limit certainty about the claims of therapeutic properties. Patients who receive these procedures almost invariably have chronic mood disorders or OCD and have virtually exhausted their options for other forms of treatment. Modern criteria for use of functional neurosurgery are especially rigorous in selecting for highly treatment-resistant patients. Therefore, reports of significant improvement in a substantial percentage of patients deserve attention. The literature on functional neurosurgery in mood disorders supports the role of limbic structures in the regulation of emotion and other psychiatric syndromes. Whereas the specific anatomical targets for the various neurosur❉

376 ❉

❉

Special Somatic Treatments

❉

gical approaches have varied, all the procedures affect limbic networks, directly or indirectly. This may result in alterations in the relative activity (e.g., neurophysiological or neurochemical) of various nodes in this network. In addition, transection of frontolimbic and thalamo-cingulate connections may alter serotonergic and dopaminergic transmission at sites remote from the lesion (Corkin et al. 1979). For example, the association between the time course of mood changes and the appearance of thalamic atrophy following frontal leukotomy supports the importance of remote effects. Neuroimaging studies may help establish links between the modulation of disordered networks and the efficacy of stereotactic functional neurosurgery for psychiatric disorders, as has been demonstrated in the case of pallidotomy for Parkinson’s disease (Eidelberg et al. 1997). Such work may provide another means to isolate the relevant circuits dysregulated in specific psychiatric syndromes and may help select patients likely to respond to particular surgical interventions.

Vagus Nerve Stimulation Recently, chronically indwelling electrical stimulators have been used in lieu of neurosurgery to achieve functional alteration of networks for therapeutic benefit in central pain syndromes and movement disorders. This approach has been applied to psychiatric disorders in the past (Escobedo et al. 1973) and may have future applications. Vagus nerve stimulation (VNS), with an implanted stimulator in the chest wall and stimulating leads placed over the left vagus nerve, has been approved for the treatment of treatment-resistant epilepsy. VNS has shown notable anticonvulsant effects in controlled trials (DeGiorgio et al. 1997; Schachter and Saper 1998) and affects central nervous system monoamine systems in ways that may be of benefit in mood disorder (Ben-Menachem et al. 1995; George et al. 2000). Its safety and efficacy in treatment-resistant major depression is currently under investigation, and the findings from the first open pilot study suggest that approximately 40%– 50% of treatment-resistant patients show sustained improvement with chronic VNS (Rush et al. 2000). Absence of even transient clinical improvement with ECT may predict lack of response to VNS (Rush et al. 2000).

Transcranial Magnetic Stimulation Transcranial magnetic stimulation (TMS) holds promise as a new paradigm to study brain-behavior relations and as an interventional or therapeutic ❉

377 ❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

tool. When TMS pulses are delivered rhythmically and repetitively, the technique is called repetitive TMS (rTMS). The ability of rTMS to stimulate brain areas noninvasively presents a significant advance beyond techniques that require the invasive method of direct cortical or transcranial electrical stimulation, as in ECT. rTMS appears to have focal and selective excitatory or inhibitory cortical effects, as a function of stimulation parameters, thereby offering the capacity to probe both the anatomic localization and the neurophysiological alterations that result in psychiatric symptoms or clinical changes. Research with this new tool has contributed to our understanding of a variety of clinical and basic issues (George et al. 1999) and as a therapeutic intervention has received special attention in the treatment of mood disorders (George et al. 1999; Lisanby and Sackeim 2000).

Description of the Technique Using the principle of electromagnetic induction, magnetic stimulators capitalize on the ability of time-varying magnetic fields to induce eddy currents in biological tissue. The magnetic stimulator stores electrical charge, which is discharged in brief pulses through a stimulating coil. Each pulse produces a magnetic field around the coil. When the magnetic field is adjacent to a conducting medium, such as the brain, an electrical current is induced in that medium. At sufficient magnetic field intensity, the induced current will result in the depolarization of neurons. This technique has been used to stimulate peripheral nerves and more recently the central nervous system. The scalp and skull are transparent to the magnetic field and do not present the impedance encountered by direct electrical stimulation. This allows far better control over the site and intensity of stimulation with TMS than with transcranial electrical stimulation. The focality of rTMS depends on coil design, geometry, and orientation relative to neuronal fibers (Amassian et al. 1992; Cohen et al. 1991). The most focal coils have a resolution of about 0.5 cm, as demonstrated by the selective stimulation of cortical representations of neighboring muscle groups in the motor strip (Brasil-Neto et al. 1992). The strength of the magnetic field falls off logarithmically with increasing distance from the coil. Depth of stimulation, even at high intensity, is estimated to be 2 cm below the scalp, reflecting stimulation of the underlying cortex near the gray-white junction (Rudiak and Marg 1994). This does not mean that rTMS does not have remote or transsynaptic effects. In the first simultaneous 15O-PET study of rTMS effects, Paus and colleagues (1997) reported that rTMS delivered to the frontal eye field re❉

378 ❉

❉

Special Somatic Treatments

❉

sulted in dose-dependent distal effects in superior parietal and medial parieto-occipital regions, consistent with connectivity within the visual system. The first magnetic stimulators delivered single magnetic pulses at slow repetition rates (0.3 Hz). Referred to as single-pulse TMS or slow TMS, this type of stimulation has been used to map sensorimotor cortex and to selectively interrupt visual perception. However, faster repetition rates are generally required to alter higher cortical function and emotional processes (Ojemann and Silbergeld 1993). New high-frequency repetitive magnetic stimulators achieve repetition rates up to 60 Hz. Stimulation frequencies above 1 Hz are referred to as fast rTMS.

Clinical Effects of rTMS in Mood Disorders Recent trials suggest that rTMS has therapeutic properties in major depression. Some reports found that single-pulse TMS, administered at the vertex and various locations, reduced depressive symptoms (Grisaru et al. 1994; Höflich et al. 1993). Whereas left dorsolateral prefrontal cortex (DLPFC) rTMS is reported to induce transient sadness in normal volunteers (George et al. 1997; Pascual-Leone et al. 1996a), recent studies show notable antidepressant effects when rTMS is delivered to the left DLPFC in clinical samples. An open trial found left prefrontal rTMS exerted antidepressant effects in 2 of 6 medication-resistant patients (George et al. 1995). Catalá and colleagues (1996) found daily rTMS applied to the left DLPFC improved symptoms of depression in 7 medication-resistant patients with psychotic depression, whereas right DLPFC rTMS had no effect. Figiel and colleagues (1998) administered fast left DLPFC rTMS to 56 largely medication-resistant depressed patients referred for ECT. After 5 days of rTMS, they reported a 42% response rate, with a substantial number of patients not requiring ECT. Conca and colleagues (1996) openly treated a cohort of depressed patients with an SSRI alone or with an SSRI and rTMS augmentation. The rTMS group had a faster antidepressant response, suggesting that rTMS may have potential as an augmentation strategy. Grunhaus and colleagues (1998) conducted the first study in which patients were randomized to rTMS or ECT. In this nonblinded comparison, 40 patients were studied with daily fast left DLPFC rTMS for 4 weeks or ECT. Among nonpsychotic patients, no difference was found in efficacy or speed of response between the two treatments. Both interventions exerted therapeutic effects in psychotic depression, but ECT was superior. Like the Figiel and colleagues’ (1998) results, this study suggests that some patients may ❉

379 ❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

show the same magnitude of short-term response to rTMS as ECT. Because this use of rTMS is nonconvulsive, does not involve anesthesia, and has no obvious detrimental cognitive side effects, the initial clinical findings are encouraging. Nonetheless, the efficacy of any potential antidepressant treatment must be established with placebo- or sham-controlled procedures under doubleblind conditions. This presents a particular problem for TMS. TMS necessitates that an investigator administer the treatment, usually in sessions lasting about 20 minutes. This individual is not blind to treatment conditions and can have extensive, repeated contact with patients. In addition, there has been variability in the nature of the sham contrast used with active rTMS conditions. Sham rTMS usually involves angulating the electromagnetic coil relative to the head so as to mimic the peripheral effects of active rTMS (stimulation of peripheral muscles in the face and scalp, acoustic artifact) while minimizing induced current in brain. Different methods of angulation have been used, with the possibility that some supposed sham conditions result in significant induced current in brain (Lisanby et al. 2001). There have been a few controlled trials under quasiblinded conditions. Pascual-Leone and colleagues (1996b) reported that 5 days of left DLPFC rTMS exerted marked antidepressant effects in 11 of 17 medication-resistant patients with psychotic depression, whereas right DLPFC rTMS, vertex rTMS, and left and right DLPFC sham rTMS produced no change. The magnitude of therapeutic effect in this study was remarkable, with 50% reductions in Hamilton Rating Scale for Depression scores obtained after 5 days of fast left DLPFC rTMS. In contrast, in another blinded, sham-controlled, cross-over trial, George and colleagues (1997) found daily left DLPFC rTMS for 2 weeks had significant but only modest antidepressant activity in outpatients with major depression. The reason for this discrepancy is unknown but may relate to the specific stimulation parameters used. Other groups have failed to replicate the large effect size found by Pascual-Leone and colleagues (1996b). Loo and colleagues (1999) randomized 18 nonpsychotic depressed patients to fast left DLPFC or sham rTMS for 10 days. They used stimulus parameters similar to the Pascual-Leone and colleagues study, but found no difference between active and sham treatment. Another recent sham-controlled study of slow rTMS delivered with a nonfocal, round coil over the right hemisphere in 71 patients with major depression found robust antidepressant effects with the active condition (Klein et al. 1999). This is of particular importance because the major safety risk of fast rTMS is the induction of a seizure (Wassermann 1998). Slow rTMS is likely devoid of this risk. ❉

380 ❉

❉

Special Somatic Treatments

❉

Relapse Following TMS Studies suggest that the effects of TMS may not persist for an extended period and that relapse may occur within the first few weeks following the end of daily sessions (e.g., Pascual-Leone et al. 1996b). However, many patients in this research were not receiving other treatment. Early relapse after discontinuation of an antidepressant treatment is not surprising and is frequently seen after ECT if a maintenance medication is not immediately instituted (Prudic et al. 1996; Sackeim et al. 1990). Ongoing studies are now looking at TMS/medication combination strategies or longer TMS courses aimed at achieving more persistent therapeutic effects.

TMS in Other Psychiatric Conditions To date, one study has examined the efficacy of rTMS in the treatment of acute mania. Grisauru and colleagues (1998) randomized 17 patients with acute mania to fast left or right DLPFC rTMS while maintaining ongoing pharmacological regimens. In contrast to the findings in major depression, fast left DLPFC rTMS was ineffective, whereas fast right DLPFC rTMS exerted a marked antimanic effect. This raises the possibility that the optimal site for rTMS therapeutic effects may depend on the type of psychiatric disorder. Treatment studies in other psychiatric disorders are beginning to emerge. Four published studies in schizophrenia have produced disparate results. In eight patients with schizophrenia and prominent negative symptoms, Nahas and colleagues (1999) found that compared with sham stimulation, a single 20-minute session of fast rTMS to the left DLPFC resulted in slightly improved negative symptoms and improved scores on an attentional task. Rollnick et al. (2000) reported reductions in psychotic symptoms following a 2-week course of fast rTMS in another blinded crossover trial. Klein and colleagues (1999) found that 10 days of 1-Hz rTMS to the right prefontal cortex was ineffective in schizophrenia. In a double-blind crossover trial, Hoffman and colleagues (2000) reported that 1-Hz rTMS to the left temporoparietal region decreased reports of auditory hallucinations in 12 patients. It remains unclear how long such effects may last, an issue of considerable import for a chronic disorder like schizophrenia. In a randomized trial of left and right prefrontal and mid-occipital rTMS in 12 patients with OCD, Greenberg and colleagues (1997a) found that a single session of right prefrontal rTMS decreased compulsive urges for 8 hours. Mood was also transiently improved, but there was no effect on ❉

381 ❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

anxiety or obsessions. McCann and colleagues (1998) reported that two patients with posttraumatic stress disorder (PTSD) showed improvement during open treatment with 1-Hz rTMS over the right frontal cortex. Grisaru and colleagues (1998) stimulated 10 PTSD patients over motor cortex and reported decreased anxiety.

Summary There are preliminary indications that rTMS may have antidepressant and antimanic properties, even in treatment-resistant samples. However, the use of rTMS remains investigational and the ultimate therapeutic role of rTMS is still uncertain. Also, the persistence of therapeutic change following rTMS has yet to be investigated. It is possible that rTMS will turn out to be like sleep deprivation, an interesting manipulation that results in rapid antidepressant effects, but with little clinical utility because of rapid relapse. Considerable further research is needed to clarify the therapeutic potential of rTMS, which currently cannot be recommended for routine clinical use.

Conclusions From this review, it should be clear that nonpharmacological somatic treatments have an important role to play in the difficult-to-treat patient. The most widely used somatic treatment is ECT, and an extensive research literature supports its use in a variety of difficult-to-treat patients. The efficacy of ECT has been well established in major depression and to a lesser degree in mania and schizophrenia. Medication-resistant depressed patients do respond well to ECT but at considerably lower rates than patients who have not received an adequate medication trial pre-ECT. This also suggests some overlap in the mechanism of action of ECT and commonly used antidepressant medications. The research data on the use of ECT in medication-resistant manic patients is not extensive, but clinical series do suggest efficacy. In schizophrenia, there is more evidence for the utility of ECT as an augmenting agent for antipsychotics rather than as an independent treatment modality. The efficacy of ECT in other diagnostic categories is not well established, and similarly its utility in difficult-to-treat patients with other diagnoses is based primarily on case series and anecdotal evidence. If ECT is used for conditions that are inherently chronic (e.g., Parkinson’s disease) or for disorders characterized by many recurrences, a long-term treatment ❉

382 ❉

❉

Special Somatic Treatments

❉

plan, which may need to include continuation and maintenance ECT, should be formulated in advance. In neurologically vulnerable populations, the cognitive side effects of ECT may be magnified and close monitoring of cognitive status during the treatment course (e.g., weekly MMSE examination) is advised. Although ECT is a very safe procedure with few medical complications, patients with cardiovascular disease may need special precautions, as do patients with intracranial lesions. Functional neurosurgery is rarely used, but its availability may be life saving and greatly improve the quality of life in the rare, severely ill, medication-resistant patient. The main indications, intractable mood or OCD, are based on case series and anecdotal evidence without adequate controlled data. Preliminary data from ongoing studies suggest that VNS may have potential in the care of patients with treatment-resistant major depression. TMS is an exciting new treatment modality, but the research database on its clinical utility is still fairly sparse. It remains unclear if subconvulsive stimulation with rTMS can be as effective as grand mal seizures elicited during ECT. From a neurophysiologic perspective, the ability to selectively stimulate or inhibit specific brain regions with TMS opens up a range of possibilities that cannot be achieved with a treatment like ECT. The epidemiologic evidence indicates that the use of ECT is again on the rise in the United States, and the advent of TMS as a potential therapeutic modality has raised new hopes for patients and practitioners alike. These changes confirm that the practicing psychiatrist needs to be aware of the pros and cons of using the specific somatic treatments described in this chapter in the difficult-to-treat patient.

References Abrams R: ECT for Parkinson’s disease (editorial). Am J Psychiatry 146:1391–1393, 1989 Alexopoulos GS, Shamoian CJ, Lucas J, et al: Medical problems of geriatric psychiatric patients and younger controls during electroconvulsive therapy. J Am Geriatr Soc 32:651–654, 1984 Amassian VE, Eberle L, Maccabee PJ, et al: Modeling magnetic coil excitation of human cerebral cortex with a peripheral nerve immersed in a brain-shaped volume conductor: the significance of fiber bending in excitation. Electroencephalogr Clin Neurophysiol Suppl 85:291–301, 1992 Andersen K, Balldin J, Gottfries CG, et al: A double-blind evaluation of electroconvulsive therapy in Parkinson’s disease with on-off phenomena. Acta Neurol Scand 76:191–199, 1987

❉

383 ❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

Ballantine HT, Giriunas IE: Advances in psychiatric surgery, in Functional Neurosurgery. Edited by Rasmussen Tand M. New York, Raven, 1979 Ballantine HT, Levy BS, Dagi T, et al: Cingulotomy for psychiatric illness: report of 13 years’ experience, in Neurosurgical Treatment in Psychiatry, Pain, and Epilepsy. Edited by Sweet W, Obrador Alcalde S, Martin-Rodriguez J, et al. Baltimore, MD, University Park Press, 1977, pp 333–354 Ballantine HT, Bouckoms AJ, Thomas EK, et al: Treatment of psychiatric illness by stereotactic cingulotomy. Biol Psychiatry 22:807–819, 1987 Ben-Menachem E, Hamberger A, Hedner T, et al: Effects of vagus nerve stimulation on amino acids and other metabolites in the CSF of patients with partial seizures. Epilepsy Research 20:221–227, 1995 Bench CJ, Frackowiak RS, Dolan RJ: Changes in regional cerebral blood flow on recovery from depression. Psychol Med 25:247–261, 1995 Bingley T, Leksell L, Meyerson BA, et al: Stereotactic anterior capsulotomy in anxiety and obsessive-compulsive states, in Surgical Approaches in Psychiatry. Edited by Laitinen LV, Livingston KE. Baltimore, MD, University Park Press, 1973, pp 159– 164 Black DW, Winokur G, Nasrallah A: Treatment and outcome in secondary depression: a naturalistic study of 1087 patients. J Clin Psychiatry 48:438–441, 1987a Black DW, Winokur G, Nasrallah A: Treatment of mania: a naturalistic study of electroconvulsive therapy versus lithium in 438 patients. J Clin Psychiatry 48:132– 139, 1987b Black DW, Winokur G, Hulbert J, et al: Predictors of immediate response in the treatment of mania: the importance of comorbidity. Biol Psychiatry 24:191–198, 1988 Black DW, Winokur G, Nasrallah A: A multivariate analysis of the experience of 423 depressed inpatients treated with electroconvulsive therapy. Convulsive Therapy 9:112–120, 1993 Brandon S, Cowley P, McDonald C, et al: Electroconvulsive therapy: results in depressive illness from the Leicestershire trial. BMJ 288:22–25, 1984 Brandon S, Cowley P, McDonald C, et al: Leicester ECT trial: results in schizophrenia. Br J Psychiatry 146:177–183, 1985 Brasil-Neto JP, Cohen LG, Panizza M, et al: Optimal focal transcranial magnetic activation of the human motor cortex: effects of coil orientation, shape of the induced current pulse, and stimulus intensity. J Clin Neurophysiol 9:132–136, 1992 Bridges PK, Barlett JR, Hale AS, et al: Psychosurgery: stereotactic subcaudate tractomy: an indespensable treatment. Br J Psychiatry 165:599–611, 1994 Brill NQ, Crumpton E, Eiduson S, et al: Investigation of the therapeutic components and various factors associated with improvement with electroconvulsive treatment: a preliminary report. Am J Psychiatry 113:997–1008, 1957

❉

384 ❉

❉

Special Somatic Treatments

❉

Broseta J, Barcia-Salorio JL, Roldan P, et al: Stereotactic subcaudate tractotomy: longterm results and measuring of effects on psychiatric symptoms, in Modern Concepts in Psychiatric Surgery. Edited by Hitchcock ER, Ballantine HT. New York, Elsevier/North-Holland Biomedical Press, 1979, pp 241–252 Catalá MD, Rubio B, Pascual-Leone A: Lateralized effect of rapid-rate transcranial magnetic stimulation of dorsal prefrontal cortex on depression. Neurology 46:A327, 1996 Cattan RA, Barry PP, Mead G, et al: Electroconvulsive therapy in octogenarians. J Am Geriatr Soc 38:753–758, 1990 Chanpattana W, Somchai Chakrabhand ML, Sackeim HA, et al: Continuation ECT in treatment-resistant schizophrenia: a controlled study. J ECT 15:178–192, 1999 Coffey CE, Lucke J, Weiner RD, et al: Seizure threshold in electroconvulsive therapy, I: initial seizure threshold. Biol Psychiatry 37:713–720, 1995 Cohen MR, Swartz CM: Absence of nimodipine premedication effect on memory after electroconvulsive therapy. Neuropsychobiology 24:165–168, 1991 Coleman EA, Sackeim HA, Prudick J et al: Subjective memory complaints prior to and following electroconvulsive therapy. Biol Psychiatry 39:346–356, 1996 Conca A, Koppi S, Konig P, et al: Transcranial magnetic stimulation: a novel antidepressive strategy? Neuropsychobiol 43:204–207, 1996 Corkin S, Twitchell TE, Sullivan EV: Safety and efficacy of cingulotomy for pain and psychiatric disorders, in Modern Concepts in Psychiatric Surgery. Edited by Hitchcock ER, Ballantine HT. New York, Elsevier/North-Holland Biomedical Pres, 1979, pp 253–272 DeCarolis V, Gibertz F, Roccatagliata G, et al: Imipramine and electroshock in the treatment of depression: a clinical statistical analysis of 437 cases. Diseases of the Nervous System 16:29–42, 1964 Decina P, Malitz S, Sackeim HA, et al: Cardiac arrest during ECT modified by βadrenergic blockade. Am J Psychiatry 141:298–300, 1984 DeGiorgio C, Handforth A, Schacter S, et al.: Multicenter, double-blind, controlled trial of vagus nerve stimulation (NCP System) for medically intractable partialonset seizures: initial report of the United States E05 Study Group (abstract). Epilepsia 38 (suppl 8):133, 1997 Devanand DP, Krueger RB: Electroconvulsive therapy in the elderly. Current Opinion in Psychiatry 7:359–364, 1994 Devanand DP, Decina P, Sackeim HA, et al: Serial dexamethasone suppression tests in initial suppressors and nonsuppressors treated with ECT. Biol Psychiatry 22:463– 472, 1987 Devanand DP, Malitz S, Sackeim HA: ECT in a patient with aortic aneurysm. J Clin Psychiatry 51:255–256, 1990 Devanand DP, Verma AK, Tirumalasetti F, et al: Absence of cognitive impairment after more than 100 lifetime ECT treatments. Am J Psychiatry 148:929–932, 1991

❉

385 ❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

Devanand DP, Dwork AJ, Hutchinson ER, et al: Does electroconvulsive therapy alter brain structure? Am J Psychiatry 151:957–970, 1994 Diering SL, Bell WO: Functional neurosurgery for psychiatric disorders: a historical perspective. Stereotact Funct Neurosurg 57:175–194, 1991 Dinan TG, Barry S: A comparison of ECT with a combined lithium and tricyclic combination among depressed tricyclic nonresponders. Acta Psychiatr Scand 80:97–100, 1989 Dodwell D, Goldberg D: A study of factors associated with response to electroconvulsive therapy in patients with schizpophrenic symptoms. Br J Psychiatry 154:635– 639, 1989 Douyon R, Serby M, Klutchko B, et al: ECT and Parkinson’s disease revisited: a “naturalistic” study. Am J Psychiatry 146:1451–1455, 1989 Eidelberg D, Moeller JR, Kazumata, et al: Metabolic correlates of pallidal neuronal activity in Parkinson’s disease. Brain 120:1315–1324, 1997 Escobedo R, Fernandez-Guardiola R, Solis G: Chronic stimulation of the cingulum in humans with behavioral disorders, in Surgical Approaches in Psychiatry. Edited by Laitinen LV, Livingston KE. Baltimore, MD, University Park Press, 1973, pp 65–68 Figiel GS, Coffey CE, Djang WT, et al: Brain magnetic resonance imaging findings in ECT-induced delirium. J Neuropsychiatry Clin Neurosci 2:53–58, 1990 Figiel GS, DeLeo B, Zorumski CF, et al: Combined use of labetolol and nifedipine in controlling the cardiovascular response from ECT. J Geriatr Psychiatry Neurol 6:20–24, 1993 Figiel GS, Epstein C, McDonald WM, et al: The use of rapid rate transcranial magnetic stimulation (rTMS) in refractory depression patients. J Neuropsychiatry Clin Neurosci 10:20–25, 1998 Fink M, Sackeim HA: Convulsive therapy in schizophrenia? Schizophr Bull 22:27– 39, 1996 Flaherty JA, Naidu J, Dysken M: ECT, emergent dyskinesia, and depression. Am J Psychiatry 141:808–809, 1984 Folstein MF, Folstein SE, McHugh PR: “Mini-Mental State.” J Psychiatric Res 12:189– 198, 1975 Foltz EL, White LE: Pain “relief” by frontal cingulotomy. J Neurosurg 19:89–100, 1962 Freeman CP, Basson JV, Crighton A: Double-blind controlled trail of electroconvulsive therapy (ECT) and simulated ECT in depressive illness. Lancet 1:738–740, 1978 Gangadhar BN, Kapur RL, Kalyanasundaram S: Comparison of ECT with imipramine in endogenous depression: a double blind study. Br J Psychiatry 141:367–371, 1982 Geoghegan JJ, Stevenson GH: Prophylactic electroshock. Am J Psychiatry 105:494– 496, 1949

❉

386 ❉

❉

Special Somatic Treatments

❉

George MS, Wasserman EM, Williams WA, et al: Daily repetitive transcranial magnetic stimulation (rTMS) improves mood in depression. Neuroreport 6:1853–1856, 1995 George MS, Wassermann EM, Kimbrell TA: Mood improvement following daily left prefrontal repetitive transcranial magnetic stimulation in patients with depression: a placebo-controlled crossover trial. Am J Psychiatry 154:1752–1756, 1997 George MS, Sackeim HA, Rush AJ, et al: Vagus nerve stimulation: a new tool for brain research and therapy. Biol Psychiatry 47:287–295, 2000 George MS, Lisanby SH, Sackeim HA: Transcranial magnetic stimulation: applications in neuropsychiatry. Arch Gen Psychiatry 56:300–311, 1999 Greenberg BD, George MS, Dearing J, et al: Effects of prefrontal transcranial magnetic stimulation (rTMS) in OCD: a preliminary study. Am J Psychiatry 154:867–869, 1997a Greenberg BD, Hallett M, Post RM: Mood improvement following daily left prefrontal repetitive transcranial magnetic stimulation in patients with depression: a placebo-controlled crossover trial. Am J Psychiatry 154:1752–1756, 1997b Greenblatt M, Grooser GH, Wechsler HA: Differential response of hospitalized depressed patients in somatic therapy. Am J Psychiatry 120:935–943, 1964 Gregory S, Shawcross CR, Gill D: The Nottingham ECT Study: a double-blind comparison of bilateral, unilateral and simulated ECT in depressive illness. Br J Psychiatry 146:520–524, 1985 Griesemer DA, Kellner CH, Beale MD, et al: Electroconvulsive therapy for treatment of intractable seizures: initial findings in two children. Neurology 49:1389–1392, 1997 Grisaru N, Yarovslavsky U, Abarbanel J: Transcranial magnetic stimulation in depression and schizophrenia. Eur Neuropsychopharmacol 4:287–288, 1994 Grisaru N, Amir M, Cohen H, et al: Effect of transcranial magnetic stimulation in posttraumatic stress disorder: a preliminary study. Biol Psychiatry 44:52–55, 1998 Grisaru N, Chudakov B, Yaroslavsky Y, et al: TMS in mania: a controlled study. Am J Psychiatry 155:1608–1610, 1998 Grunhaus L, Dannon P, Schrieber S: Effects of transcranial magnetic stimulation on severe depression: similarities with ECT (abstract #254). Biol Psychiatry 43:76S, 1998 Gujavarty K, Greenberg LB, Fink M: Electroconvulsive therapy and neuroleptic medication in therapy-resistant positive-symptom psychosis. Convulsive Therapy 3:185–195, 1987 Hay P, Sachdev P, Cumming S, et al: Treatment of obsessive-compulsive disorder by psychosurgery. Acta Psychiatr Scand 87:197–207, 1993 Hoffman RE, Boutros NN, Hu S, et al: Transcranial magnetic stimulation and auditory hallucinations in schizophrenia. Lancet 355:1073–1075, 2000

❉

387 ❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

Höflich G, Kasper S, Hufnagel A, et al: Application of transcranial magnetic stimulation in treatment of drug-resistant major depression: a report of two cases. Human Psychopharmacology 8:361–365, 1993 Husum B, Vester-Andersen T, Buchmann G, et al: Electroconvulsive therapy and intracranial aneurysm: prevention of blood pressure elevation in a normotensive patient by hydralazine and propranolol. Anaesthesia 38:1205–1207, 1983 Jenike MA, Baer L, Ballantine HT, et al: Cingulotomy for refractory obsessive-compulsive disorder. Arch Gen Psychiatry 48:548–555, 1991 Johnstone EC, Deakin JF, Lawler P, et al: The Northwick Park electroconvulsive therapy trial. Lancet 2:1317–1320, 1980 Khanna S, Gangadhar BN, Sinha V, et al: Electroconvulsive therapy in obsessive-compulsive disorder. Convulsive Therapy 4:314–320, 1988 Klein E, Puyerovsky M, Koren D, et al: Right prefrontal slow repetitive transcranial magnetic stimulation in schizophrenia: a double-blind sham-controlled pilot study. Biol Psychiatry 46:1451–1454, 1999 Klein E, Kreinin I, Chistyakov A, et al: Therapeutic efficacy of right prefrontal slow repetitive transcranial magnetic stimulation in major depression: a double-blind controlled study. Arch Gen Psychiatry 56:315–320, 1999 Knight G: Stereotactic tractotomy in the surgical treatment of mental illness. J Neurol Neurosurg Psychiatry 28:304, 1965 Krueger RB, Sackeim HA: Electroconvulsive therapy and schizophrenia, in Schizophrenia. Edited by Hirsch SR, Weinberger D. Oxford, England, Blackwell, 1995, pp 503–545 Laursen H, Gjerris A, Bolwig TG, et al: Cerebral edema and vascular permeability to serum proteins following electroconvulsive shock in rats. Convulsive Therapy 7:237–244, 1991 Leksell L, Backlund EO: Stereotactic gammacapsulotomy, in Modern Concepts in Psychiatric Surgery. Edited by Hitchcock ER, Ballantine HT. New York, Elsevier/ North-Holland Biomedical Press, 1979, pp 213–216 Lisanby SH, Devanand DP, Nobler MS, et al: Exceptionally high seizure threshold: ECT device limitations. Convulsive Therapy 12:156–164, 1996 Lisanby SH, Sackeim HA: Therapeutic brain interventions and the nature of emotion, in The Neuropsychology of Emotion. Edited by Borod J. New York, Oxford University Press, 2000, pp 456–491 Lisanby SH, Gutman D, Luber B, et al: Sham TMS: intracerebral measurement of the induced electrical field and the induction of motor-evoked potentials. Biol Psychiatry 49:460–463, 2001 Loo C, Mitchell P, Sachdev P, et al: Double-blind controlled investigation of transcranial magnetic stimulation for the treatment of resistant major depression. Am J Psychiatry 156:946–948, 1999

❉

388 ❉

❉

Special Somatic Treatments

❉

Martuza RL, Chiocca EA, Jenike M.A, et al: Stereotactic radiofrequency thermal cingulotomy for obsessive-compulsive disorder. J Neuropsychiatry Clin Neurosci 2:331–336, 1990 Maxwell RE: Behavioral modification, in Brain Surgery: Complication Avoidance and Management. Edited by Apuzzo MLJ. New York, Churchill Livingstone, 1993, pp 1557–1565 May PR: Treatment of Schizophrenia: A Comparative Study of Five Treatment Methods. New York, Science House, 1968 McCabe MS, Norris B: ECT versus chlorpromazine in mania. Biol Psychiatry 12:245– 254, 1977 McCann UD, Kimbrell TA, Morgan CM, et al: Repetitive transcranial magnetic stimulation for posttraumatic stress disorder (letter). Arch Gen Psychiatry 55:276– 279, 1998 McElhiney MC, Moody BJ, Steif BL, et al: Autobiographical memory and mood: effects of electroconvulsive therapy. Neuropsychology 9:501–517, 1995 Medical Research Council: Clinical trial of the treatment of depressive illness. BMJ 5439:881–886, 1965 Miller LJ: Use of electroconvulsive therapy during pregnancy. Hospital and Community Psychiatry 45:444–450, 1994 Milstein V, Small JG, Klapper MH, et al: Uni- versus bilateral ECT in the treatment of mania. Convulsive Therapy 3:1–9, 1987 Mindus P, Jenike MA: Neurosurgical treatment of malignant obsessive compulsive disorder. Psychiatr Clin North Am 15:921–938, 1992 Mukherjee S, Debsikdar V: Absence of neuroletpic-induced parkinsonism in psychotic patients receiving adjunctive electroconvulsive therapy. Convulsive Therapy 10:53–58, 1994 Mukherjee S, Sackeim HA, Lee C: Unilateral ECT in the treatment of manic episodes. Convulsive Therapy 4:74–80, 1988 Mukherjee S, Sackeim HA, Schnur DB: Electroconvulsive therapy of acute manic episodes: a review of 50 years’ experience. Am J Psychiatry 151:169–176, 1994 Mulsant BH, Haskett RF, Prudic J, et al: Low use of neuroleptic drugs in the treatment of psychotic depression. Am J Psychiatry 154:559–561, 1997 Nahas Z, McConnell K, Collins S, et al: Could left prefrontal rTMS modify negative symptoms and attention in schizophrenia? Biol Psychiatry 45:37S, 1999 Nobler MS, Sackeim HA: Electroconvulsive therapy: Clinical and biological aspects, in Prediction of Response to Antidepressant Treatments. Edited by Goodnick PJ. Washington, DC, American Psychiatric Press, 1996, pp 177–198 Ojemann GA, Silbergeld DL: Approaches to epilepsy surgery. Neurosurg Clin N Am 4:183–191, 1993 Pascual-Leone A, Catala MD, Pascual-Leone Pascual A: Lateralized effect of rapidrate transcranial magnetic stimulation of the prefrontal cortex on mood. Neurology 46:499–502, 1996a

❉

389 ❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

Pascual-Leone A, Rubio B, Pallardo F, et al: Rapid-rate transcranial magnetic stimulation of left dorsolateral prefrontal cortex in drug-resistant depression. Lancet 348:233–237, 1996b Paus T, Jech R, Thompson CJ, et al: Transcranial magnetic stimulation during positron emission tomography. J Neurosci 17:3178–3184, 1997 Perris C, d’Elia G: A study of bipolar (manic-depressive) and unipolar recurrent depressive psychoses, IX: therapy and prognosis. Acta Psychiatr Scand (suppl) 42:153–171, 1966 Petito CK, Schaefer JA, Plum F: Ultrastructural characteristics of the brain and bloodbrain barrier in experimental seizures. Brain Res 127:251–267, 1977 Petrides G, Fink M: Atrial fibrillation, anticoagulation, and electroconvulsive therapy. Convulsive Therapy 12:91–98, 1996 Petrides G, Dhossche D, Fink M, et al: Continuation ECT: relapse prevention in affective disorders. Convulsive Therapy 10:189–194, 1994 Pippard J, Ellam L: Electroconvulsive Treatment in Great Britain. Ashford, England, Gaskell Books, 1981 Pitts FJ: Medical physiology of ECT, in Electroconvulsive Therapy: Biological Foundations and Clinical Applications. Edited by Abrams R, Essman W, New York, Spectrum, 1982, pp 57–90 Pridmore S, Yeo PT, Pasha MI: ECT for the physical signs of Parkinson’s disease without depressive disorder (letter). J Neurol Neurosurg Psychiatry 58:641–642, 1995 Prudic J, Sackeim H, Devanand D: Medication resistance and clinical response to electroconvulsive therapy. Psychiatry Res 31:287–296, 1990 Prudic J, Sackeim HA, Devanand DP, et al: The efficacy of ECT in double depression. Depression 1:38–44, 1993 Prudic J, Haskett RF, Mulsant B, et al: Resistance to antidepressant medications and short-term clinical response to ECT. Am J Psychiatry 153:985–992, 1996 Rollnick JD, Huber TJ, Mogk H, et al: High frequency repetitive transcranial magnetic stimulation (rTMS) of the dorsolateral prefrontal cortex in schizophrenia patients. Neuroreport 11:4013–4015, 2000 Roose SP, Glassman AH, Attia E, et al: Comparative efficacy of selective serotonin reuptake inhibitors and tricyclics in the treatment of melancholia. Am J Psychiatry 151:1735–1739, 1994 Royal College of Psychiatrists: The Second Report of the Royal College of Psychiatrists’ Special Committee on ECT. London, Royal College of Psychiatrists, 1995 Rudiak D, Marg E: Finding depth of magnetic brain stimulation: a reevaluation. Electroencephalogr Clin Neurophysiol Suppl 93:358–371, 1994 Rush AJ, George MS, Sackeim HA, et al : Vagus nerve stimulation (VNSTM) for treatment-resistant depressions: a multicenter study. Biol Psychiatry 47:276–286, 2000

❉

390 ❉

❉

Special Somatic Treatments

❉

Rylander G: Stereotactic radiosurgery in anxiety and obsessive-compulsive states: psychiatric surgery, in Modern Concepts in Psychiatric Surgery. Edited by Hitchcock ER, Ballantine HT. New York, Elsevier/North-Holland Biomedical Press, 1979, pp 253–240 Sackeim HA: The cognitive effects of electroconvulsive therapy, in Cognitive Disorders: Pathophysiology and Treatment. Edited by Moos WH, Gamzu ER, Thal LJ. New York, Marcel Dekker, 1992, pp 183–228 Sackeim HA, Stern Y: The neuropsychiatry of memory and amnesia, in The American Psychiatric Press Textbook of Neuropsychiatry, 3rd Edition. Edited by Yudofsky SC, Hales RE. Washington, DC, American Psychiatric Press, 1997, pp 501–518 Sackeim HA, Decina P, Kanzler M, et al: Effects of electrode placement on the efficacy of titrated, low-dose ECT. Am J Psychiatry 144:1449–1455, 1987a Sackeim HA, Decina P, Prohovnik I, et al: Seizure threshold in electroconvulsive therapy: effects of sex, age, electrode placement, and number of treatments. Arch Gen Psychiatry 44:355–360, 1987b Sackeim HA, Prudic J, Devanand DP, et al: The impact of medication resistance and continuation pharmacotherapy on relapse following response to electroconvulsive therapy in major depression. J Clin Psychopharmacol 10:96–104, 1990 Sackeim HA, Freeman J, McElhiney M, et al: Effects of major depression on estimates of intelligence. J Clin Exp Neuropsychol 14:268–288, 1992 Sackeim HA, Prudic J, Devanand DP, et al: Effects of stimulus intensity and electrode placement on the efficacy and cognitive effects of ECT. N Engl J Med 328:839– 846, 1993 Sackeim HA, Prudic J, Devanand DP, et al: A prospective, randomized, double-blind comparison of bilateral and right unilateral ECT at different stimulus intensities. Arch Gen Psychiatry 57:581–590, 2000 Sackeim HA, Haskett RF, Mulsant BH, et al: Continuation pharmacotherapy in the prevention of relapse following electroconvulsive therapy: a randomized controlled trial. JAMA 285:1299–1307, 2001 Schachter SC, Saper CB: Vagus nerve stimulation. Epilepsia 39:677–686, 1998 Scott AI, Dougall N, Ross M, et al: Short-term effects of electroconvulsive treatment on the uptake of 99mTc-exametazime into brain in major depression shown with single photon emission tomography. J Affect Disord 30:27–34, 1994 Sikdar S, Kulhara P, Avasthi A, et al: Combined chlorpromazine and electroconvulsive therapy in mania. Br J Psychiatry 164:806–810, 1994 Small JG: Efficacy of electroconvulsive therapy in schizophrenia, mania, and other disorders, I: schizophrenia. Convulsive Therapy 1:263–270, 1985 Small JG, Milstein V, Klapper M, et al: ECT combined with neuroleptics in the treatment of schizophrenia. Psychopharm Bull 18:34–35, 1982 Small JG, Klapper MH, Kellams JJ, et al: Electroconvulsive treatment compared with lithium in the management of manic states. Arch Gen Psychiatry 45:727–732, 1988

❉

391 ❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

Sobin C, Sackeim HA, Prudic J, et al: Predictors of retrograde amnesia following ECT. Am J Psychiatry 152:995–1001, 1995 Squire LR: Memory functions as affected by electroconvulsive therapy. Ann N Y Acad Sci 462:307–314, 1986 Stoudemire A, Knos G, Gladson M, et al: Labetalol in the control of cardiovascular responses to electroconvulsive therapy in high-risk depressed medical patients. J Clin Psychiatry 51:508–512, 1990 Swartz CM, Lewis RK: Theophylline reversal of electroconvulsive therapy (ECT) seizure inhibition. Psychosomatics 32:47–51, 1991 Taylor P, Fleminger JJ: ECT for schizophrenia. Lancet 1:1380–1382, 1980 Thompson JW, Blaine JD: Use of ECT in the United States in 1975 and 1980. Am J Psychiatry 144:557–562, 1987 Thompson JW, Weiner RD, Myers CP: Use of ECT in the United States in 1975, 1980, and 1986. Am J Psychiatry 151:1657–1661, 1994 Tsuang MT, Dempsey GM: Long-term outcome of major psychoses, II: schizoaffective disorder compared with schizophrenia, affective disorder, and surgical control group. Arch Gen Psychiatry 36:1302–1304, 1979 Ulett G, Smith K, Gleser G: Evaluation of convulsive and subconvulsive shock therapies utilizing a control group. Am J Psychiatry 112:795–802, 1956 Wassermann EM: Risk and safety of repetitive transcranial magnetic stimulation: Electroencephalogr Clin Neurophysiol Suppl 108:1–16, 1998 Weiner RD, Rogers HJ, Davidson JR, et al: Effects of stimulus parameters on cognitive side effects. Ann N Y Acad Sci 462:315–325, 1986 Weiner RD, Fochmann L, Kellner C, et al: The Practice of ECT: Recommendations for Treatment, Training, and Privileging, 2nd Edition. Washington, DC, American Psychiatric Press, 2001 Wu JC, Gillin JC, Buchsbaum MS, et al: Effect of sleep deprivation on brain metabolism of depressed patients. Am J Psychiatry 149:538–543, 1992 Yudofsky S, Ovsiew F: Neurosurgical and related interventions for the treatment of patients with psychiatric disorders. J Neuropsychiatry 2:253–255, 1990 Zielinski RJ, Roose SP, Devanand DP, et al: Cardiovascular complications of ECT in depressed patients with cardiac disease. Am J Psychiatry 150:904–909, 1993 Zimmerman M, Coryell W, Pfohl B, et al: ECT response in depressed patients with and without a DSM-III personality disorder. Am J Psychiatry 143:1030–1032, 1986 Zorumski CF, Rutherford JL, Burke WJ, et al: ECT in primary and secondary depression. J Clin Psychiatry 47:298–300, 1986

❉

392 ❉

14 Summary Clinical Wisdom in Psychiatry Mantosh J. Dewan, M.D. Ronald W. Pies, M.D.

E

ach chapter in this book addresses the science, the art, and the clinical wisdom needed to treat very ill psychiatric patients. There is an impressive and rapidly growing corpus of knowledge pertinent to the initial treatment of mentally ill patients. However, after the first one or two steps, we are very quickly across the boundaries of scientific, evidence-based data and firmly into the realm of the art and poetry of psychiatry. For instance, the initial treatment of a depressed patient may be an antidepressant, but it is not clear which specific agent would be the best choice (Dewan and Anand 1999). If there is an inadequate response, empirical data suggest that the second step is to increase the dose (Fava et al. 1994). If this too does not work satisfactorily, there are no data that clearly point to a third step. Does the clinician add psychotherapy, and if so, what kind? Or an augmenting agent, such as lithium or thyroid medication? Or switch to another antidepressant, and if so, which one? Or combine the first antidepressant with another antidepressant? ❉

393 ❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

Need for More Research It is discouraging that in many important areas the evidence base is quickly exhausted. As Drs. Goodwin and Ghaemi pointed out in Chapter 2, despite many decades of debate, it is still unclear whether a bipolar depressed patient should be treated with a mood stabilizer alone, an antidepressant alone, or a combination of the two. They strongly recommend a trial of mood stabilizer alone, based on a well-reasoned position that militates against usual current practice. Similarly, in Chapter 13 Devanand and colleagues tell us that although a large number of depressed patients have been treated with maintenance ECT, with the frequency adjusted according to the varying clinical symptomatology, with good results, controlled trials are lacking. However, there are now ongoing studies to clarify the effectiveness of medications and ECT in long-term maintenance of difficult-to-treat patients. Every chapter of this book is replete with examples of deficits in outcome data. The majority of both the psychopharmacology and psychotherapy research literature focuses on short-term (6- to 12-week) studies with “clean” patients. These “efficacy” studies are essentially irrelevant to the clinical treatment of severely ill, complex, difficult-to-treat patients; longterm and “effectiveness” data from a broad sampling of patients are needed. All of the chapters in this book are rich in specific suggestions for future research, and we certainly echo the need for effectiveness research with the difficult-to-treat population. It is encouraging that a number of our authors—and others—are in fact engaged in just these kinds of studies.

Beyond Evidence-Based Psychiatry: Clinical Wisdom In this chapter, we will summarize some common threads that were found in working with almost all kinds of difficult-to-treat patients. Some authors identified evidence-based treatments that were not carefully applied in the clinical situation; other treatments are more from the realm of poetry.

Diagnosis, Comorbidity, and Treatment Resistance Several authors in this book emphasized the need for a more careful and complete diagnosis that could lead to more complete and therefore better treatment. Each comorbid disorder must be treated separately and fully. In Chapter 11, Cummings and colleagues report that in traumatic brain ❉

394 ❉

❉

Summary: Clinical Wisdom in Psychiatry

❉

injury focal contusions predominantly injure prefrontal and anterior temporal lobes. They describe the more familiar frontal lobe syndrome and also the lesser known syndromes associated with orbitofrontal (pseudopsychopathic), medial frontal/anterior cingulate cortex, and dorsolateral convexity lesions. They inform us that a mix of these clinical syndromes is more common than any one of them, and all are characterized by inability to plan, limited insight, unrealistic ambition, and poor judgment. Associated features include inattention to personal appearance and hygiene and a loss of social graces. An accurate diagnosis prevents dismissing these patients as psychopathic, malingering, or willfully noncompliant, and may lead to greater empathy and improved outcome. Substance and psychiatric comorbidity are so universal that they must be specifically looked for, identified, and treated separately in every difficultto-treat patient. For instance, in Chapter 10, Albanese and Khantzian remind us that “the term dual probably underestimates many of our patients’ problems, because many of them have more than two diagnoses[!]...The National Comorbidity Survey (NCS) reported that 14% of the population had three or more lifetime psychiatric diagnoses” (p. 275). Treatment-resistant patients sometimes are not! Devanand and colleagues report that about half of “treatment-resistant” depressed patients referred for ECT were found to be missed diagnoses of psychotic depression. These patients had not been treated with a combination of antipsychotic and antidepressant medications, which would likely be effective. Therefore, these patients were not necessarily treatment resistant.

Treatment Alliance and a Deep Respect for the Patient When little empirically derived evidence was available to these master clinicians, we were impressed at the power of the poetry that comes into play as they describe the heroic manner in which their patients deal with their affliction and how treatment requires the treatment team to go many an extra mile. It becomes clear that the fragility of the treatment alliance and the tenuousness of the basic trust require the therapist to be the one to make the greater effort in an empathic yet technically sophisticated way. Several authors offer helpful technical advice regarding fostering a working alliance. In Chapter 9, Harris and colleagues suggest that At the start of treatment in either setting, the patient’s attitude toward therapy will always be ambivalent (if not entirely rejecting). Nevertheless, patients should be encouraged to join in a positive alliance with their

❉

395 ❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

treatment team. The most obvious and universal source of anorectic patients’ treatment resistance is their awareness that treatment will, if successful, interfere with their allegiance to thinness. (p. 254)

They also suggest that clinicians engage the patient in his or her own treatment[, which] may help the patient develop a somewhat positive alliance. One useful technique is to ask the patient to first list the benefits and then list costs to him or her in maintaining his or her anorectic disorder. In being asked to list these benefits, the patient is validated in his or her experience of the anorexia nervosa as important in his or her life. The patient’s list of the costs in maintaining the illness provides a foundation of personal reasons from which questions can be raised about the desirability of this disorder. This same list can also enable the patient to see the value of treatment. (p. 254)

In Chapter 6, Hembree and colleagues provide clear guidelines for engaging patients with posttraumatic stress disorder (PTSD). They also lay strong emphasis on the importance of the treatment alliance with these abused, naturally nontrusting patients, and note that there is a need for “frequent, grounding-type comments (e.g., ‘you are safe here...the memory can not hurt you...I’m here to help you’)” (p. 166). In Chapter 7, Zanarini and Silk suggest reframing the self-mutilation, devaluing, and rage that is so difficult to deal with in patients suffering from borderline personality disorder: Rather than wasting time feeling upset or even horrified, therapists wishing to work with severely disturbed borderline patients need to remember that, in many cases, this type of behavior began in childhood. At that time, it had nothing to do with therapy and was secretive in nature. Rather than seeing self-harm as an iatrogenic behavior with all its meaning tied to treatment, it is probably more accurate to see it as 1) a long-standing form of selfsoothing, 2) a protective reaction in which the patient’s volcanic rage is turned back upon him- or herself, and 3) as an addiction, with all the power that implies. (p. 184)

This insight frees the therapist to empathize with a hateful patient and to respect the enormity of the patient’s heroic struggle against the demons that haunt him or her.

A Comprehensive and Sophisticated Treatment Plan In this book, there was repeated emphasis on the need for an accurate and complete diagnosis, which includes a careful search for medical, psychiatric, ❉

396 ❉

❉

Summary: Clinical Wisdom in Psychiatry

❉

and substance comorbid conditions. The comprehensive treatment plan then includes the best available treatments for each of these diagnostic conditions. There was less clarity and greater variability on whether these comorbid conditions could be treated concurrently or sequentially. Another concern was that clinicians sometimes do not apply what we know works. For instance, antipsychotics are a must for the maintenance treatment of schizophrenia but still have a very high relapse rate. The second most effective treatment is family therapy, which reduces the relapse rate by half if added to medication management. However, the schizophrenia Patient Outcomes Research Team (PORT) study found that antipsychotics were misused half of the time and family therapy was available to only a minority of patients (Lehman and Steinwach 1998). In Chapter 2, Ghaemi and Goodwin advocate for the primacy of longterm outcome over short-term gain. They state that, with the treatment-resistant bipolar patient, there are many ways in which the treatment can fail but few in which it can succeed.... The gravest error in treating patients with bipolar disorder is to focus on the immediate clinical situation at the expense of the longitudinal course of illness. Thus, when the patient is depressed, the clinician focuses on treating the depression, frequently with antidepressant agents; and when the patient is manic, the focus shifts to treating mania, frequently with neuroleptic agents. Instead, the focus should be on the fact that bipolar disorder represents an ongoing process, and it requires a long-term perspective for optimal treatment. Simply put, the only treatments that have been shown to work in both the short and long term for bipolar disorder are mood-stabilizing agents (lithium and some anticonvulsants). Thus, in treating a patient with bipolar disorder, the first priority is always mood stabilization. Although this may seem unnecessary to point out, a review of the evidence makes it clear that all too often mood-stabilizing agents are not used aggressively enough in the treatment of bipolar disorder, whereas antidepressants and typical neuroleptics are used all too aggressively. (pp. 8–9)

There is often poor agreement among the numerous practice guidelines, expert consensus statements, and algorithm projects. This may be one explanation for why clinical practice often does not match guidelines (e.g., in the PORT study). Another reason why an effective, proven treatment is not used appropriately is to be found in the controversy surrounding ECT. Misguided and even mischievous political pressures, plus a history of misuse, are responsible for the underutilization of ECT in difficult-to-treat depressed patients and perhaps in difficult-to-treat schizophrenic patients as well. ❉

397 ❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

Monotherapy and Combined Treatments Reviews of the empirical literature indicate that medications are the primary treatment modality for bipolar disorder and schizophrenia; psychotherapy is essential for the treatment of dissociative identity disorder (DID), PTSD, and personality disorders. Disorders such as OCD, panic, and mild to moderate depression respond equally to medications or therapy. Although not always supported by the literature, difficult-to-treat patients, irrespective of diagnosis, are generally treated with combined or even multiple modalities. This is emphasized by the case vignettes in this book, which without exception employed multimodal treatment. When combined treatment is more effective, it appears that a very specific psychotherapeutic technique is needed. For instance, generic supportive therapy adds little to medication treatment in schizophrenic patients; however, combining with two highly specific approaches, personal therapy (Hogarty et al. 1997) and family psychoeducation (McFarlane 1994), is more effective than medication alone. Similarly, Keller and colleagues (2000) found that the combination of an antidepressant and a very specific therapy, the cognitive-behavioral analysis system, is markedly superior to either treatment alone. Given the influx of excellent new psychotropic medications, Zanarini and Silk in Chapter 7 wisely warn us that medications are sometimes poorly used or overused to eliminate symptoms while missing the more serious underlying pathology. Similarly, there is the danger that difficult-to-treat patients may sometimes continue in ineffective psychotherapy for too long, thereby denying the patient a chance at effective biological, or combined, treatment.

Consultation and Supervision In Chapter 8, Kluft suggests that supervision is an essential ingredient for therapists who are relatively inexperienced in treating difficult-to-treat DID patients. This recommendation can certainly be strongly endorsed for clinicians treating very difficult patients, whether they are borderline, have DID, or have an eating disorder. Although many clinicians have an informal network that allows a “Can I run a case by you?” consult, there probably is a paucity of formal consultation and supervision. As Kluft writes, The treatment of DID raises so many issues and complexities that only a minority of psychotherapists, regardless of their general competence and experience, are either “naturals” or adapt smoothly and rapidly to this

❉

398 ❉

❉

Summary: Clinical Wisdom in Psychiatry

❉

work.... Countertransference errors are extremely common in work with DID patients because they are so complex, their crises can prove difficult to manage, their traumatic material is so affectively charged, their attachment and dependency issues are often stressful to address, the issues surrounding their treatment are so fraught with controversy, and the therapist so often finds him- or herself feeling exasperated or deskilled. (pp. 217–218)

He urges getting supervision with an experienced, skilled clinician. In Chapter 9, Harris and colleagues describe the problems of treating eating-disordered patients as falling into two categories: experience of the treatment team and type of therapy.... Inexperienced clinicians are more easily fooled. They are often reluctant to consider that the patient is capable of deception. In fact, this is one of the most common problems encountered in training clinicians to treat this population.

Further, The choice of treatment can interfere with a patient’s progress toward recovery.... The use of psychodynamically based therapy to treat eating disorders continues to be a problem. Understanding and exploring underlying psychodynamic issues, without concomitant behavior change, does not lead to resolution of the eating disorder.... Unfortunately, some clinicians will continue with a treatment approach because they know it well, without regard for its impact on the patient. (pp. 260–261)

Both these issues would be ideally addressed in consultation or supervision with an experienced expert.

Don’t Give Up! Several authors point out that even diagnoses that are synonymous with the “difficult patient” have reasonable outcomes when treated effectively. For instance, as Harris states, although “[e]ating disorders seem to have earned the reputation for difficult if not impossible to treat among the general public as well as the medical and psychiatric communities..., 75% of the patients were functioning in the good category after 30 years.... Never give up on the eating-disordered patient” (pp. 243, 266). Our experts wove a rich tapestry with these common threads: When confronted with a difficult-to-treat patient, check and recheck the basics. Forge an alliance, no matter if you have to initially work harder than the ❉

399 ❉

❉

The Difficult-to-Treat Psychiatric Patient

❉

patient. Reevaluate the diagnosis and especially look for comorbidity. Use specific therapies with clear goals. There are algorithms that can be derived, based partly on empirical evidence and substantially on clinical wisdom. Don’t give up. When in doubt or, worse, in trouble, consult an expert colleague. Properly treated, outcomes with even some notoriously difficult-totreat patients are favorable and the process indeed rewarding. Begin treatment based on the current science but with a healthy helping of art; when that does not prove enough, add more science, art, intuition, and poetry... I’ve set aside my prescription pad and analytic calm— dropped all pretense of science: it’s you and me now, pressed cold against death’s ribs. I used what tricks I know to keep you living through another bony night, another flurry of final phone calls. And you, as always, refuting life: denaturing love, companions, sex. Well, you leave my office alive. That’s as close to certainty as our work gets. “Crisis,” Ron Pies (1992)

References Dewan M, Anand V: Evaluating the tolerability of the newer antidepressants. J Nerv Ment Dis 187:96–101, 1999 Fava M, Rosenbaum J, McGrath P, et al: Lithium and tricyclic augmentaton of fluoxetine treatment for resistant major depression. Am J Psychiatry 151:1372–1374, 1994

❉

400 ❉

❉

Summary: Clinical Wisdom in Psychiatry

❉

Hogarty G, Kornblith S, Greenwald B, et al: Three year trial of personal therapy among schizophrenic patients living with or independent of family, 1: description of study and effects on relapse rates. Am J Psychiatry 154:1504–1513, 1997 Keller M, McCullough J, Klein D, et al: A comparison of nefazodone, the cognitive behavioral analysis system of psychotherapy, and their combination for the treatment of chronic depression. N Engl J Med 342:1462–1470, 2000 Lehman A, Steinwach S: Translating research into practice: the Schizophrenia Patient Outcomes Research Team (PORT) treatment recommendations. Schizophr Bull 24:1–10, 1998 McFarlane WR: Multiple-family groups and psychoeducation in the treatment of schizophrenia. New Dir Ment Health Serv 62:13–22, 1994 Pies R: Crisis: In Riding Down Dark. Troy, ME, Nightshade Press, 1992

❉

401 ❉

This page intentionally left blank

Index Page numbers in boldface type refer to figures or tables.

AA (Alcoholics Anonymous), 168, 262 AAPs. See Antipsychotics, atypical Abecarnil, for generalized anxiety disorder, 136 Acetophenazine, 48 ACTH. See Adrenocorticotropic hormone Acupuncture, for depression, 102 AD. See Alzheimer’s disease Adaptationalism, for dissociative identity disorder, 230 S-Adenosylmethionine (SAMe), 87, 102 ADHD. See Attention-deficit/ hyperactivity disorder Adrenocorticotropic hormone (ACTH) depression induced by, 88 electroconvulsive therapy-induced elevation of, 339 Agency for Health Care Policy and Research, 96, 98 Aggressive behavior alcohol abuse, schizophrenia and, 277 threats by dissociative identity disorder patients, 219–220 traumatic brain injury and, 310 pharmacotherapy for, 308–309 Agitation Alzheimer’s disease and, 301

❉

pharmacotherapy for, 307–309, 313, 314–316 traumatic brain injury and, 306–308 Agoraphobia, 141 Agranulocytosis, 59, 282 Akathisia, 32, 43, 50, 59, 72 Albuterol, 343 Alcohol abuse and psychiatric disorders, 275. See also Dualdiagnosis patients anorexia nervosa, 248 bipolar disorder, 12, 277 borderline personality disorder, 186 depression, 86, 88, 286 posttraumatic stress disorder, 150, 158, 167, 168 prevalence of, 275–276 schizophrenia, 51, 277 social phobia, 119 suicidality, 278 Alcoholics Anonymous (AA), 168, 262 Alexander technique, for generalized anxiety disorder, 137 Algorithms for treatment of agitation, 314 of bipolar disorder, 32–33, 33 of depression, 106–108 of dissociative identity disorder, 234–238, 235–237 of schizophrenia, 52–53, 69–72, 70–71

403 ❉

❉

The Difficult-to-Treat Psychiatric Patient

Alprazolam, 103 for borderline personality disorder, 201 for generalized anxiety disorder, 134, 137 interaction with antidepressants, 343 for panic disorder, 138, 139, 142 for posttraumatic stress disorder, 152 for schizophrenia, 60–61 for social phobia, 116 use in chronic obstructive pulmonary disease, 345 use in diabetes mellitus, 340 Alter personalities. See Dissociative identity disorder Alternative medicine therapies for depression, 102 for generalized anxiety disorder, 137 Alzheimer’s disease (AD), 299–303, 313–320 behavioral disturbances in, 300–303 agitation, 301 anxiety, 302 apathy, 302–303 depression, 302 psychosis, 301–302 clinical vignette of, 319 diagnosis of, 300 pharmacotherapy for, 303, 313–320 agitation, 313, 314–316 anxiety, 315, 317, 318 apathy, 318 depression, 317–318, 318 multiple symptoms and combined treatment, 319– 320 psychosis, 313–315, 317 therapeutic approaches to, 300 Amantadine for depression, 93

❉

❉

for traumatic brain injury patients, 307 American Psychiatric Association (APA) Practice Guidelines for schizophrenia, 44, 54, 60, 62 Amiloride, interaction with lithium, 332 Aminophylline, 344 Amitriptyline for anorexia nervosa, 248 for borderline personality disorder, 200 for bulimia nervosa, 249 for generalized anxiety disorder, 135 for posttraumatic stress disorder, 151, 156 use as analgesic in diabetes mellitus, 340 Amnesia dissociative, 210 drug-induced, 11 electroconvulsive therapy-induced, 361 noncompliance and, 11 posttraumatic, 304–305, 308–310 “tip of the tongue” phenomenon, 11 Amoxapine, for psychotic depression, 83 Amphetamine depression induced by, 88 for traumatic brain injury patients, 306 Amygdalotomy, 375 Anabolic steroids, depression induced by, 88 Aneurysms, electroconvulsive therapy and, 363 Anger, as obstacle to treatment of posttraumatic stress disorder, 164–165, 168, 173

404 ❉

❉

Index ❉

Angiotensin-converting enzyme inhibitors, interaction with lithium, 332 Anhedonia, 60, 123 Anorexia nervosa aberrant hunger and satiety in, 259 age distribution of, 245 chronic course of, 246, 253 clinical vignette of, 263–264 comorbidity with, 248, 256 diagnosis of, 244 electroconvulsive therapy for, 262 factors associated with outcome of, 258–261 biological factors, 258–259 prognostic factors, 258 refractory illness, 259–261 genetic studies of, 258–259 mortality from, 246 options for refractory patients, 262– 263 outcome studies of, 246 pharmacotherapy for, 247–249, 256, 262 prevalence of, 245 psychotherapy for, 250, 253–256 cognitive-behavioral therapy, 250, 255 contingency planning, 255 inpatient programs, 253, 255– 256, 262–263 team approach, 254 therapeutic alliance, 254, 395– 396 response to treatment for, 247–250 subtypes of, 244 treatment recommendations for, 253–256 treatment-resistant features of, 243, 244 Anterior capsulotomy, 376 Anterior cingulotomy, 375

❉

for depression, 102 Antiarrhythmic agents, interaction with selective serotonin reuptake inhibitors, 329 Anticholinergic effects of antipsychotics, 50 of tricyclic antidepressants in diabetic patients, 337 Anticonvulsants. See also specific drugs for agitation, 314–316 for bipolar disorder, 10, 19–23, 33 acute mania, 20–21 acute mixed episodes, 22 combining with lithium, 21–23 indications for, 21 for borderline personality disorder, 200–201 for bulimia nervosa, 250 for depression, 93 depression induced by, 88 discontinuing before electroconvulsive therapy, 370 for dual-diagnosis patients, 287–288 for generalized anxiety disorder, 131 for panic disorder, 141 for posttraumatic epilepsy, 312 use in cardiovascular disease, 332– 333 use in chronic obstructive pulmonary disease, 344–345 Antidepressants, 18. See also specific drugs and classes for agitation, 314–316 for anorexia nervosa, 248 for bipolar depression, 10, 22, 27– 29, 33, 84 bipolar II disorder, 14 duration of treatment with, 29 mood-destabilizing effects of, 3, 13–15, 27 range of responses to, 8

405 ❉

❉

The Difficult-to-Treat Psychiatric Patient

Antidepressants (continued) for borderline personality disorder, 200 for bulimia nervosa, 249–250, 257 classifying response to, 2 for cognitively impaired patients, 308, 317–318, 318 diagnostic accuracy and response to, 3 dosage of, 89–90 drug interactions with between antidepressant classes, 94 carbamazepine, 21 for dual-diagnosis patients, 285–287 for dysthymic disorder, 85 combining with psychotherapy, 85–86 for generalized anxiety disorder, 131, 133, 135, 137 inadequate treatment with, 3 managing side effects of, 89 mania induced by, 8, 13–15, 27, 84, 280 novel drugs, 91–92 for obsessive-compulsive disorder, 124–125 for panic disorder, 138–140 for posttraumatic stress disorder, 150–151, 158–159 for schizophrenia, 56, 57 for social phobia, 116, 117, 119, 121–123 for unipolar depression, 89–92 S-adenosylmethionine, 87, 102 augmentation strategies, 92–93, 93 combinations of, 93–94 combining with psychotherapy, 97–101 decision to augment or switch drugs, 90

❉

❉

factors associated with poor response to, 82–83 methylfolate and, 87 noncompliance with, 89 nonresponse vs. partial response to, 90 optimization of initial treatment with, 89–90 premature discontinuation of, 82 psychotic depression, 83 rate of response to, 81 switching between, 90–92 use in cardiovascular disease, 327– 330 use in chronic obstructive pulmonary disease, 342–344 use in diabetes mellitus, 336–338 as analgesics, 340 Antidepressants, tricyclic (TCAs). See also specific drugs augmentation of, 92 for bipolar depression, 10 mood-destabilizing effects of, 13–14 blood levels and response to, 89 for borderline personality disorder, 200 for bulimia nervosa, 249–250 for cognitively impaired patients, 317–318, 318 combining with other antidepressants, 94 combining with stimulants, 94 for dual-diagnosis patients, 286–287 for dysthymic disorder, 85 falls and fractures associated with, 333 for generalized anxiety disorder, 133, 135, 137 orthostatic hypotension induced by, 327 for panic disorder, 139

406 ❉

❉ for posttraumatic stress disorder, 151, 159 for social phobia, 119 switching between other antidepressants and, 91, 364 use in cardiovascular disease, 327– 328 use in chronic obstructive pulmonary disease, 342–343 use in diabetes mellitus, 336–337 as analgesics, 340 Antihypertensive agents, depression induced by, 88 Antipsychotics. See also specific drugs for agitation, 314–316 for anorexia nervosa, 247–248, 262 bioavailability of, 3–4 for bipolar disorder, 10, 20, 28–29, 33 atypical agents, 25–26 for borderline personality disorder, 199–200 classifying response to, 2 combinations of, 58–60 depot neuroleptics, 55 for dissociative identity disorder, 228 drug interactions with, 4 antimicrobials, 344 carbamazepine, 4, 21, 58 lithium, 58 selective serotonin reuptake inhibitors, 57 for generalized anxiety disorder, 131, 135–136 inadequate treatment with, 3 for psychosis in cognitively impaired patients, 313–315, 317 for psychotic depression, 83 for schizophrenia algorithms for, 52–53, 70–71

❉

Index ❉ atypical agents, 1, 43, 52–53, 55–56, 69, 73 augmentation of, 56–61 changing between classes of, 54–55 combinations of, 58–60, 72 combining with electroconvulsive therapy, 369 depot neuroleptics, 55 dosages and therapeutic levels of, 46, 47–49, 54 noncompliance with, 42, 46, 50– 51, 68–69 phases of response to, 45 plasma levels of, 45, 46, 69 in pregnancy and lactation, 68 premature discontinuation of, 45–46 resistance to, 45–46 response to conventional agents, 42 substance abuse and response to, 277 side effects of, 4–5, 25, 50, 307, 331 extrapyramidal symptoms, 4, 25, 26, 28, 32, 55, 59, 72 weight gain, 50, 339 somewhat typical (SANs), 54 for traumatic brain injury patients, 307–308 use in cardiovascular disease, 330– 332 use in chronic obstructive pulmonary disease, 344 use in diabetes mellitus, 339 Antipsychotics, atypical (AAPs), 1 for agitation, 315, 316 “awakening” experiences with, 50 for bipolar disorder, 25–26, 33 for borderline personality disorder, 200

407 ❉

❉

The Difficult-to-Treat Psychiatric Patient

Antipsychotics, atypical (continued) changing between, 56 combining with other antipsychotics, 58–60 depot preparations of, 55 for dissociative identity disorder, 228 dosage and therapeutic levels of, 47 for dual-diagnosis patients, 288–290 electroconvulsive therapy and, 61–62, 72 extrapyramidal side effects of, 43 for generalized anxiety disorder, 135–136 noncompliance with, 50–51 for obsessive-compulsive disorder, 126–128, 131 for posttraumatic stress disorder, 159 in pregnancy and lactation, 68 for schizophrenia, 43, 52–53, 55–56, 69, 73 switching from typical antipsychotic to, 56 use in cardiovascular disease, 330–332 use in cognitively impaired patients, 317 use in diabetes mellitus, 339 weight gain induced by, 339 Antisocial personality disorder, 274 Anxiety disorders, 115–143. See also specific disorders comorbidity with Alzheimer’s disease, 302 bipolar disorder, 20 chronic obstructive pulmonary disease, 341–342 depression, 88 eating disorders, 248 posttraumatic stress disorder, 150, 158

❉

❉

substance abuse, 287 traumatic brain injury, 312–313 functional neurosurgery for, 375 generalized anxiety disorder, 131–138 obsessive-compulsive disorder, 124–131 panic disorder, 138–143 pharmacotherapy in cognitively impaired patients, 315, 317, 318 social phobia, 115–124 Anxiety intolerance, as obstacle to treatment of posttraumatic stress disorder, 164–166 Anxiety management training, 132, 133, 134 for posttraumatic stress disorder, 153, 161 APA (American Psychiatric Association) Practice Guidelines for schizophrenia, 44, 54, 60, 62 Apathy Alzheimer’s disease and, 302–303 pharmacotherapy in cognitively impaired patients, 318 traumatic brain injury and, 306 Aphasia, 310, 311 Aplastic anemia, carbamazepineinduced, 21 Aricept. See Donepezil Aromatherapy, for generalized anxiety disorder, 137 Aspirin, interaction with valproate, 21 Assertive community treatment programs, for schizophrenia, 67, 68 Asthma. See Chronic obstructive pulmonary disease Atenolol, for social phobia, 117 Ativan. See Lorazepam Atrial fibrillation, drug-induced

408 ❉

❉

Index ❉

clozapine, 331 fluoxetine, 328 Atropine, for electroconvulsive therapy, 373 Attention-deficit/hyperactivity disorder (ADHD) bipolar disorder and, 20 substance abuse and, 274–275, 290 Avoidant personality disorder obsessive-compulsive disorder and, 126 social phobia and, 119 Beck Depression Inventory, 168 Behavior therapy (BT). See also Cognitive-behavioral therapy for bulimia nervosa, 251 for depression, 81, 95–97 for dysthymic disorder, 85 for obsessive-compulsive disorder, 125, 128–129 for posttraumatic stress disorder, 152–155, 160–166 (See also Exposure therapy) for schizophrenia, 67 for traumatic brain injury patients, 306, 310 Behavioral family management for bipolar disorder (BFM-BD), 31 Benzodiazepines (BZDs). See also specific drugs addiction potential of, 132, 137, 141, 142 for bipolar disorder, 26–27 for borderline personality disorder, 201 for catatonia, 61 for dual-diagnosis patients, 287 for generalized anxiety disorder, 131, 132, 134, 135, 137 for panic disorder, 138, 141, 142

❉

for posttraumatic stress disorder, 152 respiratory depression induced by, 345 for schizophrenia, 56, 60–61, 64 for social phobia, 116, 121–122 use in chronic obstructive pulmonary disease, 345 use in cognitively impaired patients, 308, 315, 316, 317 use in diabetes mellitus, 340 Betoptic drops, during electroconvulsive therapy, 373 BFM-BD (behavioral family management for bipolar disorder), 31 Binge-eating disorder, 244–245 Bioavailability of drug, 3–4 Bipolar disorder, 1, 7–34 age at onset of, 11, 15, 20 anticonvulsants for, 10, 19–23, 33 antidepressants for, 10, 22, 27–29, 33 bipolar II disorder, 14 duration of treatment with, 29 mood-destabilizing effects of, 3, 13–15, 27 range of responses to, 8 antipsychotics for, 10, 20, 25–26, 33 for breakthrough depression, 28–29 approaches to treatment resistance in, 8, 20–29 acute and breakthrough depression, 27–29 acute mania, 20–21 acute mixed episode, 21–22 breakthrough manic/mixed episodes and rapid cycling, 22–27 bipolar II disorder, 18 antidepressants for, 14

409 ❉

❉

The Difficult-to-Treat Psychiatric Patient

Bipolar disorder (continued) bipolar II disorder (continued) diagnosis of, 16 prevalence of, 9 clinical complexity of, 7–8 clinical vignette of, 32 combining mood stabilizers for, 22– 25 comorbidity with, 12–13, 20 dissociative identity disorder, 214 medical conditions, 20 substance abuse, 12–13, 276, 277, 280, 285, 287–288 traumatic brain injury, 311 defining treatment resistance in, 10– 11 depression in, 18, 83–84 differential diagnosis of, 18–19, 19 electroconvulsive therapy for, 29, 369–370 emphasizing long-term outcome in treatment of, 397 goal of treatment for, 8 medication noncompliance in, 11– 12 misdiagnosis of, 9, 15–19 DIGFAST mnemonic for mania, 16, 17 manic-depressive spectrum, 16– 18, 18 mixed episode, 19, 21–22 not otherwise specified (NOS), 10, 16, 18 prevalence of, 9–10, 15 prevalence of treatment resistance in, 1 psychosis and, 20, 32 psychotherapy for, 30–31 rapid cycling, 19–20, 22–27 routine treatment for, 10 substance abuse and, 12–13

❉

❉

treatment algorithm for, 32–33, 33 Bleeding, valproate-induced, 21 β−Blockers. See also specific drugs for electroconvulsive therapy, 363 for generalized anxiety disorder, 133, 136 interaction with selective serotonin reuptake inhibitors, 329 for social phobia, 117, 122 for traumatic brain injury patients, 309 for tremor, 11 use in diabetes mellitus, 340 Blood-brain barrier, 4, 363 Body therapies, for generalized anxiety disorder, 137 Bone mineral density, 333 Borderline personality disorder (BPD), 179–204 clinical vignette of, 203–204 comorbidity with, 182 obsessive-compulsive disorder, 126 panic disorder, 139 social phobia, 119 substance abuse, 274, 284 differentiation from atypical depression, 84 hospitalization for, 184–185, 201– 202 long-term course and outcome of, 202 pharmacotherapy for, 198–201, 203–204 prevalence of, 180 problematic psychopathology in, 181–196 cognitive problems, 182–184 demandingness, 189–190 dependency and counterdependency, 193– 194

410 ❉

❉

Index ❉

devaluation, 187–188 distortions of truth, 195 dysphoria, 181–182 entitlement, 190–191 manipulation, 188–189 sadomasochistic tendencies, 195–196 self-mutilation, 184–185 special relationships, 192–193 suicidality, 185–187 treatment regressions, 191–192 psychotherapy for, 196–198 Dialectical Behavior Therapy, 139, 262 therapeutic alliance, 396 severity subtypes of, 179–181 BPD. See Borderline personality disorder BPRS (Brief Psychiatric Rating Scale), 57–58 Brain injury. See Traumatic brain injury Brain lesions, electroconvulsive therapy and, 363–364 Breast-feeding, antipsychotics and, 68 Breathing retraining, for posttraumatic stress disorder, 161 Brief dynamic therapy, for depression, 95 Brief Psychiatric Rating Scale (BPRS), 57–58 Bromocriptine for depression, 93 for traumatic brain injury patients, 307 Bronchodilators, 343–345 BT. See Behavior therapy Bulimia nervosa aberrant hunger and satiety in, 259 age at onset of, 245 clinical vignette of, 264–265 comorbidity with, 249, 258, 261 diagnosis of, 244

❉

factors associated with outcome of, 258–261 biological factors, 258–259 prognostic factors, 258 refractory illness, 259–261 medical complications of, 257 mortality from, 247 options for refractory patients, 262–263 outcome studies of, 246–247 pharmacotherapy for, 249–250, 257 prevalence of, 245 psychotherapy for, 250–252 cognitive-behavioral therapy, 251–252, 257 contingency planning, 257 efficacy vs. pharmacotherapy, 252–253 interpersonal therapy, 252, 257 purging and nonpurging types of, 244 relapses of, 247 response to treatment of, 249–253 treatment recommendations for, 256–257 Buprenorphine, 285 Bupropion, 2, 103 for bipolar disorder, 10, 14, 29 for bulimia nervosa, 249 for cognitively impaired patients, 318 for depression, 94 interaction with fluoxetine, 330 for obsessive-compulsive disorder, 130, 131 for posttraumatic stress disorder, 159 for social phobia, 123 use in cardiovascular disease, 329, 330, 373 use in diabetes mellitus, 338 Buspar. See Buspirone

411 ❉

❉

The Difficult-to-Treat Psychiatric Patient

Buspirone for agitation, 315, 316 combining with selective serotonin reuptake inhibitors, 94 for depression, 93 for dual-diagnosis patients, 287 for generalized anxiety disorder, 131–133 for obsessive-compulsive disorder, 128 for social phobia, 117, 122 use in cardiovascular disease, 330 use in chronic obstructive pulmonary disease, 345–346 use in cognitively impaired patients, 317, 318 BZDs. See Benzodiazepines CAD (coronary artery disease). See Cardiovascular disease Caffeine intake, panic disorder and, 140–141 Calcium channel blockers for bipolar disorder, 26 drug interactions with carbamazepine, 333 lithium, 332 for electroconvulsive therapy, 363 Cannabis abuse, schizophrenia and, 51 Capgras syndrome, 301, 319 Capsulotomy, anterior, 376 Captopril during electroconvulsive therapy, 373 interaction with lithium, 332 Carbamazepine for agitation, 315, 316 for bipolar disorder, 21–24 acute mania, 21 combining with lithium, 21–23 combining with valproate, 22–24

❉

❉

for borderline personality disorder, 200 for bulimia nervosa, 250 for depression, 93 drug interactions with, 4, 21, 23–24 calcium channel blockers, 333 neuroleptics, 58 warfarin, 21, 333 effects on thyroid hormones, 28 for posttraumatic epilepsy, 312 for posttraumatic stress disorder, 152, 159 for schizoaffective disorder, 58 for schizophrenia, 56, 57–58, 64 side effects of, 21 for traumatic brain injury patients, 309 use in cardiovascular disease, 332– 333 use in chronic obstructive pulmonary disease, 345 use in diabetes mellitus, 340 Carbohydrate craving, lithiuminduced, 11–12 Cardiac Arrhythmia Suppression Trial (CAST), 328 Cardiac arrhythmias drug-induced antipsychotics, 59, 60, 331 clozapine, 331–332 lithium, 332 selective serotonin reuptake inhibitors, 328 tricyclic antidepressants, 327– 328 electroconvulsive therapy-induced, 363, 373 Cardiomyopathy, clozapine-induced, 331 Cardiovascular disease, comorbid, 326–335 clinical vignette of, 335

412 ❉

❉

Index ❉

with depression, 326–327, 335 electroconvulsive therapy in, 330, 362–363, 372–374 epidemiology of, 326–327 pathophysiology of, 327 pharmacotherapy in, 327–333 antipsychotics, 330–332 falls and fractures related to, 333 mood stabilizers, 332–333 nontricyclic antidepressants, 328–330 tricyclic antidepressants, 327–328 psychosocial interventions in, 334 Case management, for schizophrenia, 68 CAST (Cardiac Arrhythmia Suppression Trial), 328 Catatonia benzodiazepines for, 61 electroconvulsive therapy for, 61, 62, 368, 372 CBASP. See Cognitive behavioral analysis system of psychotherapy CBT. See Cognitive-behavioral therapy Celexa. See Citalopram Chlorpromazine, 72 for anorexia nervosa, 247, 262 for bipolar disorder, 25 dosage and therapeutic levels of, 46, 48 Chlorprothixene, 48 Cholinesterase inhibitors, for Alzheimer’s disease, 300, 303, 318–320 Chronic obstructive pulmonary disease (COPD), comorbid, 341–346 cognitive-behavioral therapy for, 346 with depression, 342 with panic disorder, 341 pharmacotherapy in, 342–346

❉

antipsychotics, 344 anxiolytics, 345–346 mood stabilizers, 344–345 nontricyclic antidepressants, 343–344 tricyclic antidepressants, 342–343 Cimetidine, depression induced by, 88 Cingulotomy, anterior, 375 for depression, 102 Ciprofloxacin, 344 Citalopram for cognitively impaired patients, 318 for obsessive-compulsive disorder, 125 for panic disorder, 140 for schizophrenia, 57 Classification of treatment response, 2 Clinical Global Impression scale, 23, 287 Clinical vignettes Alzheimer’s disease, 319 anorexia nervosa, 263–264 bipolar disorder, 32 borderline personality disorder, 203–204 bulimia nervosa, 264–265 cardiovascular disease, 335 depression, 102–104 dissociative identity disorder, 232–233 dual diagnosis, 290–292 generalized anxiety disorder, 137 obsessive-compulsive disorder, 130–131 panic disorder, 142 posttraumatic stress disorder, 167–174 schizophrenia, 72–73 social phobia, 123–124

413 ❉

❉

The Difficult-to-Treat Psychiatric Patient

Clomipramine for anorexia nervosa, 248 for obsessive-compulsive disorder, 124–125, 127, 130, 131 intravenous, 129 for panic disorder, 139 Clonazepam for avoidant personality disorder, 126 for bipolar disorder, 27, 32 depression induced by, 88 for obsessive-compulsive disorder, 128, 130, 131 for panic disorder, 138 for schizophrenia, 61 for social phobia, 116, 119, 121–124 Clonidine for anorexia nervosa, 248 depression induced by, 88 for posttraumatic stress disorder, 152, 159 Clozapine, 1 adequate trial of, 56 for agitation, 315, 315 for bipolar disorder, 25 for borderline personality disorder, 200 combining with other drugs benzodiazepines, 61 dopaminergic agents, 60 lamotrigine, 58 ondansetron, 63 risperidone, 59 typical antipsychotics, 59 dosage and therapeutic levels of, 47, 56 for dual-diagnosis patients, 282– 283, 288–291 electroconvulsive therapy and, 61– 62, 72 for obsessive-compulsive disorder, 127

❉

❉

orthostatic hypotension induced by, 330 in pregnancy, 68 for schizophrenia, 43, 53, 55–56, 68, 71, 73 time to full response to, 46 use in cardiovascular disease, 330–332 use in cognitively impaired patients, 317 use in diabetes mellitus, 340 weight gain induced by, 339 Clozaril. See Clozapine Cocaine use, 284, 289–291. See also Dual-diagnosis patients bipolar disorder and, 11, 12 depression induced by, 88 posttraumatic stress disorder and, 275 schizophrenia and, 51 Cognex. See Tacrine Cognitive abnormalities in Alzheimer’s disease, 299–303, 313–320 borderline personality disorder and, 182–184 dissociative identity disorder and, 219 due to traumatic brain injury, 299– 300, 303–320 electroconvulsive therapy-induced, 361–362 mood stabilizer-induced, 11 Cognitive behavioral analysis system of psychotherapy (CBASP) for depression, 97, 99 for dysthymic disorder, 85 Cognitive-behavioral therapy (CBT) for anorexia nervosa, 250, 255, 262 for bipolar disorder, 31 for borderline personality disorder, 198

414 ❉

❉ for bulimia nervosa, 251–252, 257, 262, 265 compared with pharmacotherapy, 252–253 in chronic obstructive pulmonary disease, 346 in diabetes mellitus, 341 for dysthymic disorder, 85 for generalized anxiety disorder, 132, 133 for obsessive-compulsive disorder, 130 for panic disorder, 138, 139, 141 for posttraumatic stress disorder, 152–155 for schizophrenia, 44, 64, 67 for social phobia, 117–118, 120– 121, 124 compared with benzodiazepines, 116 with depressive symptoms, 119 Cognitive enhancement therapy, for schizophrenia, 64 Cognitive therapy (CT) for depression, 81, 82, 95–97 atypical depression, 84 combining with pharmacotherapy, 97–101 for dysthymic disorder, 85 for obsessive-compulsive disorder, 128, 129 for posttraumatic stress disorder, 153–154 for social phobia, 118 Combat veterans. See Posttraumatic stress disorder Comorbidity, 3, 395 with anorexia nervosa, 248, 256 with bipolar disorder, 12–13, 20 with borderline personality disorder, 182

❉

Index ❉ with bulimia nervosa, 249, 258, 261 with depression, 86–89, 87, 88 with dissociative identity disorder, 214, 218–219, 228 dual diagnosis, 3, 273–292 with generalized anxiety disorder, 131 with medical illness, 325–347 National Comorbidity Survey, 9, 275, 276 with obsessive-compulsive disorder, 126–127 with panic disorder, 138–140 with posttraumatic stress disorder, 150, 158, 166, 168, 275, 276 with schizophrenia, 51 with social phobia, 118–119 Compliance. See Noncompliance with treatment Comprehensive treatment plan, 396–397 Computed tomography, 363 Computerized assessment, for social phobia, 123 Congestive heart failure. See Cardiovascular disease Constipation, antipsychotic-induced, 50 Consultation, 217, 398–399 Contingency planning for eating disorders, 262 anorexia nervosa, 255 bulimia nervosa, 257 Contusions, cerebral, 304–305 COPD. See Chronic obstructive pulmonary disease Coronary artery disease (CAD). See Cardiovascular disease Corticosteroids, depression induced by, 88

415 ❉

❉

The Difficult-to-Treat Psychiatric Patient

Costs of medications, 5 of treatment-resistant psychiatric disorders, 2 Counterdependency of borderline patients, 193–194 Countertransference reactions with bipolar patients, 30 with dissociative identity disorder patients, 217–218, 398 Creativity, mood stabilizer-induced loss of, 4, 11 Cruelty of borderline patients, 195–196 CT. See Cognitive therapy CX516, for schizophrenia, 62 Cycloserine depression induced by, 88 for schizophrenia, 62, 63, 64 Cyclothymia, 16, 18 Cyproheptadine, for anorexia nervosa, 248–249, 256 Cytochrome P450-related drug interactions with carbamazepine, 21, 23–24 between quetiapine and other atypical antipsychotics, 59–60 with selective serotonin reuptake inhibitors, 329, 331, 338, 343 smoking and, 343 with valproate, 21, 23–24 DAI (diffuse axonal injury), 304, 305 Day treatment programs, for schizophrenia, 51 DDIS (Dissociative Disorders Interview Schedule), 210 Delirium, posttraumatic, 308 Delusions Alzheimer’s disease and, 301–302, 319 pharmacotherapy for, 313–315, 317

❉

❉

borderline personality disorder and, 183 schizophrenia and, 42 traumatic brain injury and, 306 Demandingness of borderline patients, 189–190 Dementia. See Alzheimer’s disease Dentition, electroconvulsive therapy and, 364 Depakote. See Valproate Dependency of borderline patients, 193–194 of dissociative identity disorder patients, 222–223 Depersonalization disorder, 211 Depression, 1, 81–105 alternative medicine therapies for, 102 antidepressants for, 89–94, 106–107 augmentation of, 92–93, 93 for cognitively impaired patients, 317–318, 318 combinations of, 93–94 combining with psychotherapy, 97–101 decision to augment or switch drugs, 90 in dual-diagnosis patients, 285–287 noncompliance with, 89 nonresponse vs. partial response to, 90 optimization of initial treatment with, 89–90 options for treatment-refractory disease, 89–94 premature discontinuation of, 82 rate of response to, 81 switching between, 90–92 atypical, 84 bipolar, 18, 83–84

416 ❉

❉ compared with unipolar major depression, 84 management of, 27–29 classifying treatment response in, 2 clinical vignette of, 102–104 combined pharmacotherapypsychotherapy for, 97–101, 103, 104 acute phase, 98–99 in elderly patients, 100–101 maintenance phase, 99–100 sequential, 97–98 simultaneous, 98–100 comorbidity with, 86–89 Alzheimer’s disease, 302 borderline personality disorder, 182, 186, 197 cardiovascular disease, 326–327, 335 chronic obstructive pulmonary disease, 342 diabetes mellitus, 336 eating disorders, 248, 249 generalized anxiety disorder, 88, 131 medical illness, 86–88, 87, 88, 325 obsessive-compulsive disorder, 126 panic disorder, 88, 138, 139 posttraumatic stress disorder, 150, 158, 168 psychiatric disorders, 88–89 social phobia, 118–119 substance abuse, 86, 277–278, 285–286 traumatic brain injury, 311 differential diagnosis of, 18–19, 19 double, 85, 371 dysthymia, 84–86

❉

Index ❉ electroconvulsive therapy for, 83, 94, 101, 103, 360–361, 364–367, 372–374 continuous or maintenance, 367 in double depression, 371 efficacy studies of, 365 in medication-resistant patients, 366–367 relapses after, 366–367 in secondary depression, 371 in severe depressive illness, 365–366 factors associated with treatment resistance in, 82–83 folate deficiency and, 87, 88 functional neurosurgery for, 102, 374–377 medication-induced, 88 phototherapy for, 102 positive predictive factors in, 104 prevalence of treatment resistance in, 1 psychotherapy for, 82, 94–97 augmentation of, 96 combining psychotherapies, 97 combining with pharmacotherapy, 97–101 duration of, 82–83, 96 optimization of, 94–96 rate of response to, 81 switching to another technique, 96 psychotic, 83 schizophrenia and, 51, 57 seasonal, 86, 102 secondary, 18–19, 371 smoking and, 344 subtypes of, 83–86 transcranial magnetic stimulation for, 101–102, 379–380 treatment algorithm for, 106–108

417 ❉

❉

The Difficult-to-Treat Psychiatric Patient

Depression (continued) vagal nerve stimulation for, 102, 377 vitamin B12 deficiency and, 88 DES (Dissociative Experiences Scale), 210 Desipramine blood levels and response to, 89 for bulimia nervosa, 249–250 efficacy compared with psychotherapy, 252 for cognitively impaired patients, 317, 318 for dual-diagnosis patients, 286 for fluoxetine nonresponders, 92 for generalized anxiety disorder, 137 mood-destabilizing effects in bipolar disorder, 13, 14 for posttraumatic stress disorder, 151 use in diabetes mellitus, 337 as analgesic, 340 Desyrel. See Trazodone Determinants of Therapeutic Movement Instrument (DTMI), 214–215 Devaluation, borderline personality disorder and, 187–188 Dexamethasone suppression test, 82, 88 Dextroamphetamine abuse potential of, 290 for depression, 94 for schizophrenia, 60 for traumatic brain injury patients, 307 Diabetes mellitus, comorbid, 336–341 with depression, 336 electroconvulsive therapy in, 338–339 pharmacotherapy in, 336–340 antipsychotics, 339

❉

❉

anxiolytics, 340 mood stabilizers, 340 nontricyclic antidepressants, 337–338 psychotropics as analgesics, 340 tricyclic antidepressants, 336– 337 psychosocial interventions in, 341 Diagnostic accuracy, 3, 394–395 Diagnostic Interview Schedule (DIS), 9, 279 Dialectical Behavior Therapy, for borderline personality disorder, 139, 262 Diarrhea, generalized anxiety disorder and, 137 Diazepam for generalized anxiety disorder, 133 for schizophrenia, 61 use in chronic obstructive pulmonary disease, 345 DID. See Dissociative identity disorder Dietary factors depression and, 87, 88 lithium and, 11–12 Diffuse axonal injury (DAI), 304, 305 Diltiazem, interaction with carbamazepine, 333 DIS (Diagnostic Interview Schedule), 9, 279 Dissociative disorders, 209–238 classification of, 209 courses of, 210 depersonalization disorder, 211 dissociative amnesia, 210 dissociative disorder not otherwise specified, 211 dissociative fugue, 210 dissociative identity disorder, 210–211 dissociative trance disorder, 211 prevalence of, 210

418 ❉

❉ Dissociative Disorders Interview Schedule (DDIS), 210 Dissociative Experiences Scale (DES), 210 Dissociative identity disorder (DID), 209–238, 398 age at onset of, 211 alters’ fears in, 221 attachment issues in, 222 causes of poor treatment response in, 215–224 external factors, 223–224 patient variables, 218–223 process/interaction variables, 215–217 therapist variables, 217–218 clinical vignette of, 232–233 comorbidity with, 214, 218–219, 228 conventional treatment for, 212–213 course of, 210–211 ego strength and, 219 fantasy proneness and, 220 loyalty issues and, 222 media accounts of, 223–224 pharmacotherapy for, 212–213, 228 previous life experiences and treatment of, 222 psychotherapy for, 212, 224–231 adaptationalism, 230 expected response to, 213–215 factors affecting response to, 215–224 hypnosis, 212, 231, 233 logistics of, 223 minimization, 230 modalities of, 212, 229–231 optimization of, 229 personality-oriented therapy, 229–230 stages of, 212

❉

Index ❉ strategic integrationalism, 229 tactical integrationalism, 229 working with alters, 225–227 revictimization of persons with, 224 “third reality” of persons with, 221 transference reactions and, 217, 220, 222 treatment algorithm for, 234–238, 235–237 violence threats and, 219–220 Disulfiram, 285 Diuretics, interaction with lithium, 332 Divalproex. See Valproate Donepezil, for Alzheimer’s disease, 315, 319 Dopamine agonists for schizophrenia, 60 for traumatic brain injury patients, 307 Dosing schedule, 12 Doxepin, 32 use in chronic obstructive pulmonary disease, 342 Drug interactions, 4 with antidepressants between classes, 94 selective serotonin reuptake inhibitors, 57, 328, 329, 331 between stimulants and monoamine oxidase inhibitors, 94 with antipsychotics carbamazepine, 4, 21, 58 lithium, 58 between quetiapine and other atypical antipsychotics, 59–60 selective serotonin reuptake inhibitors, 57, 331 with carbamazepine, 4, 21, 23–24 with valproate, 21, 23–24

419 ❉

❉

The Difficult-to-Treat Psychiatric Patient

Drug sensitivity, 4 Dry mouth, antipsychotic-induced, 50 DTMI (Determinants of Therapeutic Movement Instrument), 214–215 Dual-diagnosis patients, 3, 273–292, 395 assessment and diagnosis of, 278–281 abstinent periods, 280–281 clinical indicators, 279 mania, 280 medical evaluation, 279–280 psychological trauma, 280 psychosis, 280 self-reports, 278 structured interviews, 278–279 urine toxicology, 278–279 clinical vignette of, 290–292 clinician’s ideology and treatment of, 274 comorbidity in, 276–278 anorexia nervosa, 248 anxiety, 287 attention-deficit/hyperactivity disorder, 274–275, 290 bipolar disorder, 12–13, 276, 277, 280, 285, 287–288 borderline personality disorder, 182, 186 bulimia nervosa, 258, 261 depression, 86, 277–278, 285–287 dissociative identity disorder, 214, 218, 228 dysthymic disorder, 85 generalized anxiety disorder, 134 panic disorder, 138, 139, 140 posttraumatic stress disorder, 150, 158, 167, 168, 275, 276, 280 schizophrenia, 51, 276–277, 282–283, 288–290

❉

❉

social phobia, 119 suicidality, 278 definition of, 274–275 definitions of substance abuse and dependence, 275 epidemiology of, 275–276 models of treatment for, 281–284 integrated treatment, 282–284 parallel treatment, 281 serial treatment, 281 pharmacotherapy for, 284–290 anxiety disorders, 287 attention-deficit/hyperactivity disorder, 290 categories of, 285 depression, 285–287 mania and affective lability, 287–288 principles of, 284–285 psychosis, 288–290 prevalence of, 275–276 Duke Brief Social Phobia Scale, 120 Dysphoria antidepressant-induced, 3 borderline personality disorder and, 181–182 neuroleptic-induced, 288 Dyspnea, depression and, 342 Dysthymic disorder, 18, 84–86 comorbidity with, 85 double depression, 85, 371 prognosis for, 85 psychotherapy for, 85–86 response to antidepressants in, 85 underdiagnosis and undertreatment of, 84–85 Eating disorders, 243–266, 399. See also Anorexia nervosa; Bingeeating disorder; Bulimia nervosa aberrant hunger and satiety in, 259 clinical vignettes of, 263–265

420 ❉

❉

Index ❉

comorbidity with borderline personality disorder, 182 depression, 248 diabetes mellitus, 341 dissociative identity disorder, 214 diagnosis of, 244–245 factors associated with outcome of, 258–261 biological factors, 258–259 prognostic factors, 258 refractory illness, 259–261 gender and, 245 options for refractory patients, 262–263 outcome studies of, 245–247 prevalence of, 245 response to treatment for, 247–253 pharmacotherapy, 247–250 psychotherapy, 250–252 relative efficacy of pharmacotherapy vs. psychotherapy, 252–253 treatment recommendations for, 253–257 ECA (Epidemiological Catchment Area) study, 9, 12, 275, 276 ECT. See Electroconvulsive therapy EEG (electroencephalogram), sleep, as predictor of treatment response in depression, 82, 83 Effexor. See Venlafaxine Ego strength, dissociative identity disorder and, 219 Elderly persons combined pharmacotherapypsychotherapy for depression in, 100–101 generalized anxiety disorder in, 134, 135

❉

Electroconvulsive therapy (ECT), 2, 360–374, 382–383 age and, 372 for children and adolescents, 372 clinical vignette of, 372–374 cognitive effects of, 361–362, 383 adjusting treatment parameters for, 361–362 amnesia, 361 prediction of, 361 comorbid medical disorders and, 362–364 cardiovascular disease, 330, 362–363, 372–374 diabetes mellitus, 338–339 intracranial lesions, 363–364 less dangerous conditions, 364 indications for, 360, 364–372 anorexia nervosa, 262 catatonia, 61, 62, 368, 372 depression, 83, 94, 101, 103, 360–361, 364–367, 372– 374 double depression, 371 mania, 29, 369–370 obsessive-compulsive disorder, 371 Parkinson’s disease, 371–372 pseudodementia syndrome, 372 schizoaffective disorder, 368, 371 schizophrenia, 61–62, 64, 72, 367–369 tardive dyskinesia, 372 mania induced by, 29 medications for, 373 mortality from, 362 mouth guards for, 364 in pregnancy, 68, 372 reasons for referral to, 360 prevalence of difficult-to-treat patients, 360–361

421 ❉

❉

The Difficult-to-Treat Psychiatric Patient

Electroconvulsive therapy (ECT) (continued) seizures induced by, 362, 373 technical issues in, 362 underutilization of, 397 Electroencephalogram (EEG), sleep, as predictor of treatment response in depression, 82, 83 EMDR. See Eye movement desensitization and reprocessing Emphysema. See Chronic obstructive pulmonary disease Enalapril, interaction with lithium, 332 Energy loss, drug-induced, 11 Entitlement, borderline personality disorder and, 190–191 Environmental interventions, for traumatic brain injury patients, 306 Epidemiological Catchment Area (ECA) study, 9, 12, 275, 276 Epilepsy, posttraumatic, 312 Epinephrine electroconvulsive therapy-induced elevation of, 339 interaction with monoamine oxidase inhibitors, 343 EPS. See Extrapyramidal symptoms ERP (exposure and response prevention), for obsessivecompulsive disorder, 125, 128–131 Erythromycin, 344 Estradiol, 339 Estrogen for Alzheimer’s disease, 300 for depression, 93 for schizophrenia, 63 Ethambutol, depression induced by, 88 Euthymia, 8 Exelon. See Rivastigmine Exercise

❉

❉

for chronic obstructive pulmonary disease, 346 for depression, 102 for generalized anxiety disorder, 137 Expert Consensus Guidelines for schizophrenia, 51, 55, 60 Exposure and response prevention (ERP), for obsessive-compulsive disorder, 125, 128–131 Exposure therapy interoceptive exposure in chronic obstructive pulmonary disease, 346 for obsessive-compulsive disorder, 125, 128–131 for posttraumatic stress disorder, 152–153, 160–166, 170–173 anger, numbing, and anxiety intolerance as obstacles to, 164–166 educating patient about rationale for, 161, 164 imaginal exposure, 162–163 in vivo exposure, 161–162 prolonged treatment program, 160–161 therapeutic alliance and, 163– 164, 169 for social phobia, 117–118, 121 Extrapyramidal symptoms (EPS), 4, 25, 26, 28, 32, 72 antipsychotic combinations and, 59 atypical antipsychotics and, 43 depot neuroleptics and, 55 selective serotonin reuptake inhibitors and, 57 Eye movement desensitization and reprocessing (EMDR) for depression, 104 for dissociative identity disorder, 212

422 ❉

❉

Index ❉

for posttraumatic stress disorder, 154 Falls, 333 Family therapy for anorexia nervosa, 250 for bipolar disorder, 31 for bulimia nervosa, 257 for depression, 97 combining with pharmacotherapy, 100 in diabetes mellitus, 341 for schizophrenia, 44, 51, 72, 73, 397 Famotidine, for schizophrenia, 63, 64 Fantasy proneness, 220 Fast metabolizers, 3 Flashbacks in dissociative identity disorder, 228 in posttraumatic stress disorder, 167, 168, 170–172 Flecainide, interaction with selective serotonin reuptake inhibitors, 329 Fluoxetine, 103 for anorexia nervosa, 248 atrial fibrillation induced by, 328 augmentation of, 92 olanzapine, 93 for bipolar disorder, 14, 32 for borderline personality disorder, 200 for bulimia nervosa, 249, 265 compared with psychotherapy, 252–253 for cognitively impaired patients, 317, 318 combining with tricyclic antidepressants, 94 dosage of, 89–90 drug interactions with alprazolam, 343 β-blockers, 329

❉

bupropion, 330 tolbutamide, 338 triazolam, 343 for dual-diagnosis patients, 286 for obsessive-compulsive disorder, 125 for panic disorder, 140 for posttraumatic stress disorder, 150–151 relapse after initial response to, 89–90 for schizophrenia, 57 switching between sertraline and, 91 use in cardiovascular disease, 328–329, 373 use in diabetes mellitus, 338 as analgesic, 340 Flupenthixol, 369 Fluphenazine, 55, 72 for agitation, 315 dosage and therapeutic levels of, 48 Fluphenazine decanoate, 55, 72 Flurazepam, 103 Fluvoxamine for cognitively impaired patients, 318 drug interactions with alprazolam, 343 diazepam, 343 theophylline, 343 tolbutamide, 338 triazolam, 343 warfarin, 329 for obsessive-compulsive disorder, 125, 130, 131 for panic disorder, 138, 139, 140 for psychotic depression, 83 for schizophrenia, 57 for social phobia, 116 Folate, 11 depression and deficiency of, 87, 88

423 ❉

❉

The Difficult-to-Treat Psychiatric Patient

Fractures, 333 Frontal lobe syndrome, 305, 395 Fugue, dissociative, 210 Functional neurosurgery, 374–377, 383 current stereotactic procedures, 375–376 anterior capsulotomy, 376 anterior cingulotomy, 375 subcaudate tractotomy, 376 for depression, 102 for obsessive-compulsive disorder, 129 Furosemide during electroconvulsive therapy, 373 interaction with lithium, 332 Gabapentin for avoidant personality disorder, 126 for bipolar disorder, 15, 24–25, 32 for panic disorder, 141 for posttraumatic stress disorder, 159 for schizophrenia, 58 side effects of, 333 for social phobia, 117, 119, 122 use as analgesic in diabetes mellitus, 340 Galantamine, for Alzheimer’s disease, 303, 315 Gamma knife surgery, for obsessivecompulsive disorder, 129 Gastrointestinal distress drug-induced selective serotonin reuptake inhibitors, 128, 338, 343 tricyclic antidepressants, 337 valproate, 345 electroconvulsive therapy and, 364

❉

❉

generalized anxiety disorder and, 137 Generalized anxiety disorder, 131–138 clinical vignette of, 137 complicating factors in treatment of, 134 definition of, 131 depression and, 88, 131 in elderly persons, 134, 135 nonpharmacological interventions for, 133 pharmacotherapy for, 131–136 prevalence of, 131 treatment recommendations for, 134–137 Geodon. See Ziprasidone Glasgow coma scale, 304 Glucagon, electroconvulsive therapyinduced elevation of, 339 Glucocorticoids, depression induced by, 88 L-Glutamine, 12 Glycine, for schizophrenia, 62 Group therapy for bulimia nervosa, 251 for depression, 95 combining with pharmacotherapy, 100 in diabetes mellitus, 341 modified dynamic group therapy for dual-diagnosis patients, 283–284 for schizophrenia, 44, 51, 64–65, 72 for social phobia, 120–121 Growth hormone, electroconvulsive therapy-induced elevation of, 339 Guanfacine, for posttraumatic stress disorder, 152, 159 Hair loss, valproate-induced, 11 Haldol. See Haloperidol Hallucinations

424 ❉

❉

Index ❉

in borderline personality disorder, 183 in schizophrenia, 42 Haloperidol, 55, 72 for agitation, 315 arrhythmia induced by, 331 for bipolar disorder, 25 for borderline personality disorder, 199 combining with atypical antipsychotic, 59 dosage and therapeutic levels of, 46, 47 interaction with carbamazepine, 4 use in cognitively impaired patients, 307–308, 313, 317 Haloperidol decanoate, 55 Hamilton Rating Scale for Depression, 24, 91, 96, 98 Head injury. See Traumatic brain injury Heart disease. See Cardiovascular disease Hematologic side effects of carbamazepine, 21 of valproate, 21 Hepatotoxicity, valproate-induced, 21 Herbal remedies for depression, 102 for generalized anxiety disorder, 136 Histamine blockers, 88, 345 Hospitalization. See Inpatient treatment HPA (hypothalamic-pituitary axis) abnormalities, 28 5-Hydroxyindoleacetic acid, 308 Hydroxyzine, for generalized anxiety disorder, 132, 133, 135 Hyperglycemia, 336. See also Diabetes mellitus electroconvulsive therapy-induced, 338–339 neuroleptic-related, 339

❉

Hypericum perforatum for depression, 102 for generalized anxiety disorder, 136 Hyperprolactinemia, 59, 333 Hypersalivation, 59 Hypertension electroconvulsive therapy-induced, 363 venlafaxine-induced, 329 Hypnosis, for dissociative identity disorder, 212, 231, 233 Hypoglycemia, 336, 338 lithium-induced, 11 monoamine oxidase inhibitorinduced, 337 Hypomania, 12, 16 antidepressant-induced, 8 diagnosis of, 9, 15 electroconvulsive therapy-induced, 29 patient enjoyment of, 30 seasonal depression and, 86 substance abuse and, 280 Hyponatremia, drug-induced carbamazepine, 332 oxcarbazepine, 333 Hypotension, drug-induced antipsychotics, 60, 330 bupropion, 329 monoamine oxidase inhibitors, 337 selective serotonin reuptake inhibitors, 328 tranylcypromine, 90 tricyclic antidepressants, 327 Hypothalamic-pituitary axis (HPA) abnormalities, 28 Hypothalamotomy, 375 Hypothyroidism, 3, 27–28, 93 ICD (intracardiac defibrillator), electroconvulsive therapy and, 363, 372–374

425 ❉

❉

The Difficult-to-Treat Psychiatric Patient

Imipramine, 104 for bipolar disorder, 13, 14, 32 blood levels and response to, 89 for bulimia nervosa, 249 compared with psychotherapy, 252 combining with psychotherapy, 97–98, 100 for dual-diagnosis patients, 286–287 for generalized anxiety disorder, 133, 135, 137 for panic disorder, 139 for posttraumatic stress disorder, 151 switching to other antidepressants from, 91 Impulse cluster disorders, obsessivecompulsive disorder and, 127 Inderal. See Propranolol Inositol, for posttraumatic stress disorder, 152 Inpatient treatment for borderline personality disorder, 184–185, 201–203 for dissociative identity disorder, 220 for eating disorders, 262–263 anorexia nervosa, 253, 255–256 Insight-oriented psychotherapy, for schizophrenia, 44, 51 Insulin, interaction with monoamine oxidase inhibitors, 337 Integrated treatment of dual-diagnosis patients, 282–284 Interpersonal and social rhythm therapy (IP/SRT) for bipolar disorder, 31 Interpersonal therapy (IPT) for bulimia nervosa, 252, 257 for depression, 81, 82, 95–97

❉

❉

combining with pharmacotherapy, 98, 100–101 for dysthymic disorder, 85 for obsessive-compulsive disorder, 129 Intracardiac defibrillator (ICD), electroconvulsive therapy and, 363, 372–374 Intracranial pressure elevation, electroconvulsive therapy and, 363–364 Iowa 500 study, 9 IP/SRT (interpersonal and social rhythm therapy) for bipolar disorder, 31 IPT. See Interpersonal therapy Irritability, traumatic brain injury and, 310 Kava kava, for generalized anxiety disorder, 136 Klüver-Bucy syndrome, 305 Kundalini yoga, for obsessivecompulsive disorder, 129 Lamotrigine for bipolar disorder, 24 for borderline personality disorder, 201 combining with clozapine, 58 dosage of, 24 interaction with valproate, 24 for panic disorder, 141 for posttraumatic stress disorder, 159 for schizophrenia, 58 side effects of, 24, 333 use in chronic obstructive pulmonary disease, 345 use in diabetes mellitus, 340 Laryngeal dystonia, 344

426 ❉

❉ Leukopenia, carbamazepine-induced, 21 Levodopa, depression induced by, 88 Levodopa/carbidopa, for traumatic brain injury patients, 307 Liebowitz Social Anxiety Scale, 120 Lifestyle Heart Trial, 334 Lisinopril, interaction with lithium, 332 Lithium for anorexia nervosa, 248 anti-insulin effects of, 11 for bipolar disorder, 10, 18, 32, 33 acute mania, 20–21 acute mixed episodes, 21–22 combining with anticonvulsants, 21–23 combining with antipsychotics, 25 rate of response to, 10 for bulimia nervosa, 250 for depression, 92, 93, 101 psychotic, 83 drug interactions with aminophylline, 344 angiotensin-converting enzyme inhibitors, 332 calcium channel blockers, 332 diuretics, 332 neuroleptics, 58 theophylline, 344 for dual-diagnosis patients, 287 electroconvulsive therapy and, 369, 370 noncompliance with side effects and, 4, 11 substance abuse and, 285 for obsessive-compulsive disorder, 128 for posttraumatic stress disorder, 152, 159 for schizophrenia, 56, 57–58, 64

❉

Index ❉ for traumatic brain injury patients, 309 use in cardiovascular disease, 332, 373 use in chronic obstructive pulmonary disease, 344 use in diabetes mellitus, 340 Lorazepam for aggressive behavior, 308 for agitation, 315, 316 for cognitively impaired patients, 308, 313, 317, 318 during electroconvulsive therapy, 373 for generalized anxiety disorder, 134, 135, 137 intravenous, 308 for panic disorder, 138 for schizophrenia, 61 Loxapine, 47, 54 Loxitane. See Loxapine Luvox. See Fluvoxamine Lying, by borderline patients, 195 Magnetic resonance imaging (MRI), 363 Mania. See also Bipolar disorder acute, 20–21 antidepressant-induced, 8, 13–15, 27, 84, 280 beakthrough manic/mixed episodes and rapid cycling, 22–27 antipsychotics for, 25–26 benzodiazepines for, 26–27 calcium channel blockers for, 26 combining mood stabilizers for, 22–25 omega-3 fatty acids for, 27 diagnosis of, 15, 16 DIGFAST mnemonic, 16, 17 manic-depressive spectrum, 16–18, 18

427 ❉

❉

The Difficult-to-Treat Psychiatric Patient

Mania (continued) electroconvulsive therapy for, 29, 369–370 in difficult-to-treat patients, 369–370 discontinuing medications before, 370 maintenance or continuation treatment, 370 in medication-resistant patients, 370 electroconvulsive therapy-induced, 29 lack of insight in, 16 medication-resistant, 370 prevalence of, 9 psychosis and, 20 in relatives of probands, 9 response to lithium for, 10 substance abuse and, 12, 280, 287– 288 transcranial magnetic stimulation for, 381 traumatic brain injury and, 311 Manipulation, by borderline patients, 188–189 MAOIs. See Monoamine oxidase inhibitors Marital therapy, for depression, 95, 97, 104 Medical illness, comorbid, 325–347 with bipolar disorder, 20 with bulimia nervosa, 257 cardiovascular disease, 326–335 chronic obstructive pulmonary disease, 341–346 with depression, 86–88, 87, 88, 326 diabetes mellitus, 336–341 with dissociative identity disorder, 214, 233 with panic disorder, 138 with schizophrenia, 51

❉

❉

Medication-related factors in treatment resistance, 3–4 bioavailability, 3–4 compliance, 4–5 costs, 5 cultural beliefs, 12 drug interactions, 4 inadequate treatment, 3 side effects, 4 Meditation techniques for generalized anxiety disorder, 137 for obsessive-compulsive disorder, 129 Mellaril. See Thioridazine Memory problems. See Amnesia Mesoridazine, 54 arrhythmia induced by, 331 dosage of, 48 Metformin, 339 Methadone maintenance treatment, 282, 285 for schizophrenia, 63, 64 Methohexital, for electroconvulsive therapy, 373 N-Methyl-D-aspartate (NMDA) receptors, in schizophrenia, 62 α−Methyldopa, depression induced by, 88 Methylfolate, 87 Methylphenidate abuse potential of, 290 for depression, 94 for dual-diagnosis patients, 290 for obsessive-compulsive disorder, 130 for traumatic brain injury patients, 306–307, 318 use in cardiovascular disease, 373 Metoclopramide, for anorexia nervosa, 248 Metoprolol, interaction with selective serotonin reuptake inhibitors, 329

428 ❉

❉

Index ❉

Metrifonate, for Alzheimer’s disease, 303 Mexiletine, interaction with selective serotonin reuptake inhibitors, 329 Midodrine, 327 Mini-Mental State Examination (MMSE), 319, 335, 361, 374, 383 Minimization, for dissociative identity disorder, 230 Mirtazapine for bipolar disorder, 10 for cognitively impaired patients, 318 for depression, 91–92, 94 for generalized anxiety disorder, 135 for posttraumatic stress disorder, 159 use in cardiovascular disease, 329– 330 MMSE (Mini-Mental State Examination), 319, 335, 361, 374, 383 Moban. See Molindone Modafanil, for depression, 93 Modified dynamic group therapy, for dual-diagnosis patients, 283–284 Molindone, 72 for agitation, 315 dosage of, 47 for psychosis in cognitively impaired patients, 317 use in diabetes mellitus, 339 Monoamine oxidase inhibitors (MAOIs) for anorexia nervosa, 248 for atypical depression, 84 augmentation of, 93 for borderline personality disorder, 200 combining with stimulants, 94 combining with tricyclic antidepressants, 94

❉

for dysthymic disorder, 85 interaction with β-adrenergic bronchodilators, 343ñ344 for panic disorder, 139 for posttraumatic stress disorder, 151, 159 side effects of, 337 for social phobia, 116, 123 switching from tricyclic antidepressants to, 91 use in cardiovascular disease, 330, 373 use in chronic obstructive pulmonary disease, 343–344 use in diabetes mellitus, 337 Mood charting, 31 Mood stabilizers, 8, 10. See also specific drugs for bipolar disorder, 20–27 acute mania, 20–21 acute mixed episodes, 21–22 breakthrough manic/mixed episodes and rapid cycling, 22–27 combinations of, 22–25 depression, 84 dosing schedule for, 12 underutilization of, 20 for borderline personality disorder, 126, 200–201 for bulimia nervosa, 250 for depression, 93 dosing schedule for, 12 for dual-diagnosis patients, 287–288 for posttraumatic stress disorder, 152, 159 for schizophrenia, 57–58, 64 side effects and noncompliance with, 4, 11–12 use in cardiovascular disease, 332–333

429 ❉

❉

The Difficult-to-Treat Psychiatric Patient

Mood stabilizers (continued) use in chronic obstructive pulmonary disease, 344–345 use in diabetes mellitus, 340 MRI (magnetic resonance imaging), 363 Multiple personalities. See Dissociative identity disorder Myocardial infarction. See also Cardiovascular disease clozapine and, 331 electroconvulsive therapy and, 330, 363, 372 lithium and, 332 psychosocial interventions and, 334 Myocardial ischemia, 328, 334 Myocarditis, clozapine-induced, 331 Naltrexone, 285 for borderline personality disorder, 201 for bulimia nervosa, 250 for traumatic brain injury patients, 309 National Comorbidity Survey (NCS), 9, 275, 276 Nausea, selective serotonin reuptake inhibitor-induced, 128 Navane. See Thiothixene NCS (National Comorbidity Survey), 9, 275, 276 Nefazodone for cognitively impaired patients, 317, 318 combining with selective serotonin reuptake inhibitors, 94 for dysthymic disorder, 85 for generalized anxiety disorder, 133, 135, 137 interaction with alprazolam and triazolam, 343 for panic disorder, 141

❉

❉

for posttraumatic stress disorder, 159 for social phobia, 117, 123 use in cardiovascular disease, 329 Neuroleptic malignant syndrome (NMS), 55, 344 Neuroleptics. See Antipsychotics Neurosurgery. See Functional neurosurgery Nimodipine for bipolar disorder, 26 interaction with lithium, 332 NMDA (N-methyl-D-aspartate) receptors, in schizophrenia, 62 NMS (neuroleptic malignant syndrome), 55, 344 Noncompliance with treatment, 4–5 for bipolar disorder, 11–12 for depression, 89 dosing schedule and, 12 drug side effects and, 4, 11 for eating disorders, 261 for schizophrenia, 42, 46, 50–51, 68–69 Nonresponders, 2 Norpramin. See Desipramine Nortriptyline, 103 for bulimia nervosa, 249 for cognitively impaired patients, 317–318, 318 combining with interpersonal therapy for elderly patients, 100–101 combining with lithium, 101 use in cardiovascular disease, 327, 329 use in chronic obstructive pulmonary disease, 342 use in diabetes mellitus, 337 Numbing, as obstacle to treatment of posttraumatic stress disorder, 164–165

430 ❉

❉

Index ❉

Obsessive-compulsive disorder (OCD), 124–131 clinical vignette of, 130–131 comorbidity with, 124, 126–127 anorexia nervosa, 263 schizophrenia, 51, 73 traumatic brain injury, 312 electroconvulsive therapy for, 371 factors associated with lack of treatment response in, 125–127 pharmacotherapy for, 124–125, 127–128 prevalence of, 124 psychotherapy for, 125, 128–129 behavior therapy, 125, 128–129 other approaches, 129 transcranial magnetic stimulation for, 381–382 treatment recommendations for, 127–129 functional neurosurgery, 129, 375 novel and experimental treatments, 129 pharmacotherapy, 127–128 psychotherapy, 128–129 Occupational therapy, for schizophrenia, 51 OCD. See Obsessive-compulsive disorder Olanzapine for agitation, 315, 316 “awakening” experiences with, 50 for bipolar disorder, 25–26, 28, 32 for borderline personality disorder, 200 combining with quetiapine, 60 combining with typical antipsychotic, 59 for depression, 93 dosage and therapeutic levels of, 47 for dual-diagnosis patients, 288

❉

for generalized anxiety disorder, 135 for obsessive-compulsive disorder, 127, 128, 131 orthostatic hypotension induced by, 330 for posttraumatic stress disorder, 159 for schizophrenia, 43 use in chronic obstructive pulmonary disease, 344 use in cognitively impaired patients, 317, 319 use in diabetes mellitus, 340 weight gain induced by, 339 Omega-3 fatty acids, for bipolar disorder, 27 Ondansetron, for schizophrenia, 63, 64 Oral hypoglycemic agents, interaction with monoamine oxidase inhibitors, 337 Orap. See Pimozide Orbitofrontal syndrome, 305, 395 Osteoporosis, 333 Overt Aggression Scale, 310 Oxazepam for cognitively impaired patients, 317, 318 for generalized anxiety disorder, 135 Oxcarbazepine, 333 Pacemakers, electroconvulsive therapy and, 363, 372–374 PACTs (Programs for Assertive Community Treatment), for schizophrenia, 67, 68 Pallidotomy, for Parkinson’s disease, 377 Pamelor. See Nortriptyline Panic disorder, 138–143 agoraphobia and, 141 caffeine intake and, 140–141 clinical vignette of, 142

431 ❉

❉

The Difficult-to-Treat Psychiatric Patient

Panic disorder (continued) comorbidity with, 138–140 chronic obstructive pulmonary disease, 341–343 depression, 88, 138, 139 generalized anxiety disorder, 136, 142 social phobia, 119, 140 traumatic brain injury, 312 factors associated with lack of treatment response in, 139–140 panic attacks in, 138 pharmacotherapy for, 138–139 prevalence of, 138 psychotherapy for, 138, 139 treatment recommendations for, 140–141 Parallel treatment of dual-diagnosis patients, 281 Parkinson’s disease electroconvulsive therapy for, 371 pallidotomy for, 377 Paroxetine, 103, 104 for bipolar disorder, 14, 29 combining with tricyclic antidepressants, 94 for generalized anxiety disorder, 135, 137 interaction with β-blockers, 329 for obsessive-compulsive disorder, 125 orthostatic hypotension induced by, 328 for panic disorder, 138, 139, 140, 142 for posttraumatic stress disorder, 150–151, 156 for psychotic depression, 83 for social phobia, 116, 121, 123 use in cardiovascular disease, 329, 335

❉

❉

use in cognitively impaired patients, 317, 318 use in diabetes mellitus, 338 as analgesic, 340 weight gain induced by, 338 Partial responders, 2 Passive-dependent personality disorder, 119 Patient Outcomes Research Team (PORT) study in schizophrenia, 56, 397 Paxil. See Paroxetine Pemoline for dual-diagnosis patients, 290 for traumatic brain injury patients, 307 Performance anxiety. See Social phobia Pergolide, for depression, 93 Permitil. See Fluphenazine Perphenazine, 48 Personality disorders borderline personality disorder, 179–204 comorbidity with depression, 88–89, 371 obsessive-compulsive disorder, 126 panic disorder, 138, 139 social phobia, 119 substance abuse, 274 traumatic brain injury, 305 Personality-oriented therapy, for dissociative identity disorder, 229–230 PET (positron emission tomography), 375 Pharmacotherapy. See also specific drugs and classes for agitation, 307–309, 313, 314–316 for Alzheimer’s disease, 303, 313–320, 314–318

432 ❉

❉

Index ❉

for bipolar disorder, 10–12, 20–29, 32–33, 33 for borderline personality disorder, 198–201, 203–204 in cardiovascular disease, 327–333 for depression, 83–86, 89–94, 93, 106–107 in diabetes mellitus, 336–340 for dissociative identity disorder, 212–213, 228 for dual-diagnosis patients, 284–290 for eating disorders, 262 anorexia nervosa, 247–249, 256 bulimia nervosa, 249–250, 257 compared with psychotherapy, 252–253 for generalized anxiety disorder, 131–136 monotherapy vs. combined treatments, 398 for obsessive-compulsive disorder, 124–125, 127–128 for panic disorder, 138–139 for posttraumatic stress disorder, 150–152, 158–160 for schizophrenia, 42–43, 45–46, 47–49, 54–61 for social phobia, 116–117, 121–123 for traumatic brain injury, 306–309, 313–319, 314–318 Phenelzine, 103 for borderline personality disorder, 200 for bulimia nervosa, 249 for posttraumatic stress disorder, 151 for social phobia, 116, 123 switching from imipramine to, 91 use in diabetes mellitus, 337 Phenobarbital, depression induced by, 88 Phenylpropanolamine, 343

❉

Phenytoin, for posttraumatic epilepsy, 312 Phototherapy, for depression, 86, 102 Physostigmine, for Alzheimer’s disease, 303 Pimozide, 54–55 for anorexia nervosa, 247 arrhythmia induced by, 331 combining with atypical antipsychotic, 59 dosage of, 47 Pindolol, for depression, 93 “Pink puffer” syndrome, 345 Piper methysticum, for generalized anxiety disorder, 136 Polarity, for generalized anxiety disorder, 137 PORT (Patient Outcomes Research Team) study in schizophrenia, 56, 397 Positron emission tomography (PET), 375 Possession trance, 210 Posttraumatic amnesia (PTA), 304–305, 308–310 Posttraumatic epilepsy, 312 Posttraumatic stress disorder (PTSD), 149–174 chronic, 150 clinical vignette of, 167–174 comorbidity with, 150, 158, 166, 168, 275, 276 borderline personality disorder, 182 traumatic brain injury, 312 diagnostic criteria for, 150 factors that interfere with treatment response in, 156–157 biological factors, 156 psychological and behavioral factors, 157

433 ❉

❉

The Difficult-to-Treat Psychiatric Patient

Posttraumatic stress disorder (PTSD) (continued) maintaining treatment focus for, 166, 172–173 pharmacotherapy for, 150–152, 158–160 atypical antipsychotics, 159 combining with psychotherapy, 159, 160 discontinuation of, 160 mood stabilizers, 159 other medications, 151–152 partial response to, 159 selective serotonin reuptake inhibitors, 150–151, 158 tricyclic antidepressants, 151, 159 prevalence of, 150 psychosocial interventions for, 152–155, 160–166 anger, numbing, and anxiety intolerance as obstacles to, 164–166, 168, 173 anxiety management/stress inoculation training, 153, 161 breathing retraining, 161 cognitive therapy, 153–154 education and treatment rationale, 161 exposure therapy, 152–153, 160–164, 170–173 eye movement desensitization and reprocessing, 154 outcome studies of cognitivebehavioral therapy, 154–155 therapeutic alliance, 163–164, 169, 396 strategies for treatment-resistant patients, 158–166, 160–163 combining pharmacotherapy and psychotherapy, 160

❉

❉

comorbid disorders, 158 pharmacotherapy, 158–160 psychosocial treatments, 160–166 transcranial magnetic stimulation for, 382 Pregnancy antipsychotics in, 68 electroconvulsive therapy in, 68, 372 Prevalence of treatment resistance, 1 Programs for Assertive Community Treatment (PACTs), for schizophrenia, 67, 68 Prolixin. See Fluphenazine Promazine, 48 Propafenone, interaction with selective serotonin reuptake inhibitors, 329 Propranolol for agitation, 315 for anxiety in cognitively impaired patients, 318 depression induced by, 88 for extrapyramidal symptoms, 11, 32 interaction with selective serotonin reuptake inhibitors, 329 for posttraumatic stress disorder, 152, 159 for schizophrenia, 56, 63, 64 for social phobia, 117 Protriptyline, use in chronic obstructive pulmonary disease, 343 Prozac. See Fluoxetine Pseudodementia syndrome, electroconvulsive therapy for, 372 Pseudoephedrine, 343 Psychoanalysis, for depression, 95 Psychodynamic psychotherapy for borderline personality disorder, 198

434 ❉

❉

Index ❉

for depression, 95, 96, 103 for dissociative identity disorder, 212 for eating disorders, 253, 261 Psychosis. See also Schizophrenia Alzheimer’s disease and, 301–302 pharmacotherapy for, 313–315, 317 bipolar disorder and, 20, 32 borderline personality disorder and, 183 depression and, 82 dissociative identity disorder and, 214 schizophrenia, 41–74 substance abuse and, 280 traumatic brain injury and, 306, 312 Psychosurgery. See Functional neurosurgery Psychotherapy/psychosocial interventions. See also specific psychotherapies for anorexia nervosa, 250, 253–256 cognitive-behavioral therapy, 255 contingency planning, 255 inpatient programs, 253, 255–256, 262–263 team approach, 254 therapeutic alliance, 254 for bipolar disorder, 30–31 for borderline personality disorder, 196–198 for bulimia nervosa, 250–252 efficacy vs. pharmacotherapy, 252–253 in cardiovascular disease, 334 in chronic obstructive pulmonary disease, 346 for depression, 82, 94–97, 103 augmentation of, 96 combining psychotherapies, 97

❉

combining with pharmacotherapy, 97–101 duration of, 82–83, 96 optimization of, 94–96 rate of response to, 81 switching to another technique, 96 in diabetes mellitus, 341 for dissociative identity disorder, 212, 224–231 adaptationalism, 230 causes of poor response to, 215–224 expected response to, 213–215 factors affecting response to, 215–224 hypnosis, 212, 231 logistics of, 223 minimization, 230 modalities of, 212, 229–231 optimization of, 229 personality-oriented therapy, 229–230 stages of, 212 strategic integrationalism, 229 strategies when therapy is not progressing well, 224–231 tactical integrationalism, 229 techniques of, 212 working with alters, 225–227 for dysthymic disorder, 85 combining with antidepressants, 85–86 for generalized anxiety disorder, 132, 133 inappropriate use of, 5 monotherapy vs. combined treatments, 398 for obsessive-compulsive disorder, 125, 128–129 for panic disorder, 138, 139, 141

435 ❉

❉

The Difficult-to-Treat Psychiatric Patient

Psychosis (continued) for posttraumatic stress disorder, 152–155, 160–166 anger, numbing, and anxiety intolerance as obstacles to, 164–166 anxiety management/stress inoculation training, 153, 161 breathing retraining, 161 cognitive therapy, 153–154 combining with pharmacotherapy, 159, 160 education and treatment rationale, 161 exposure therapy, 152–153, 160– 164 eye movement desensitization and reprocessing, 154 outcome studies of cognitivebehavioral therapy, 154–155 therapeutic alliance, 163–164, 169 for schizophrenia, 63–68 adverse effects of, 5, 51, 65–66 causes of poor response to, 51 cognitive-behavioral therapy, 44, 64, 67 expected response to conventional therapies, 43– 45 family therapy, 44, 51, 72, 73 group therapy, 44, 51, 64–65, 72 individual therapy, 44, 51, 63–64 Programs for Assertive Community Treatment, 67 social skills training, 44, 51, 67– 68 vocational rehabilitation, 67, 68 for social phobia, 117–118, 120–121 for traumatic brain injury patients, 309–313

❉

❉

aggression, 310 anxiety, 312–313 cognitive disorders, 309–310 mood disorders, 311 psychosis, 312 PTA (posttraumatic amnesia), 304–305, 308–310 PTSD. See Posttraumatic stress disorder Pulmonary embolism, antipsychoticinduced, 331 QT interval prolongation, antipsychotic-induced, 59, 331 Quetiapine, 56 for agitation, 315, 316 combining with another atypical antipsychotic, 59–60 dosage of, 47 for generalized anxiety disorder, 135–136 orthostatic hypotension induced by, 330 use in cognitively impaired patients, 317 use in diabetes mellitus, 340 Quinidine, 327 Ranitidine, depression induced by, 88 Rape victims, 153, 155, 156, 167. See also Posttraumatic stress disorder Rash, lamotrigine-induced, 24 Reboxetine, for depression, 92, 94 Relaxation techniques for chronic obstructive pulmonary disease, 346 for depression, 102 for generalized anxiety disorder, 134, 136–137 for obsessive-compulsive disorder, 129 for panic disorder, 139 for social phobia, 118

436 ❉

❉

Index ❉

Remeron. See Mirtazapine Reminyl. See Galantamine Research needs, 394 Reserpine depression induced by, 88 for schizophrenia, 63 Respiratory depression, benzodiazepine-induced, 345 Respiratory disease. See Chronic obstructive pulmonary disease Respiratory dyskinesia, 344 Respiratory suppression/arrest, induced by benzodiazepineclozapine combination, 61 Rifampin, 344 Risperdal. See Risperidone Risperidone, 291 for agitation, 315, 316 for bipolar disorder, 25, 26, 28, 32 combining with clozapine, 59 combining with quetiapine, 60 combining with typical antipsychotic, 59 dosage and therapeutic levels of, 47 for dual-diagnosis patients, 288, 289 electroconvulsive therapy and, 62 for generalized anxiety disorder, 136 interaction with selective serotonin reuptake inhibitors, 331 for obsessive-compulsive disorder, 127, 128 orthostatic hypotension induced by, 330 for posttraumatic stress disorder, 159 for schizophrenia, 43 use in cardiovascular disease, 335 use in chronic obstructive pulmonary disease, 344 use in cognitively impaired patients, 317 use in diabetes mellitus, 340

❉

Rivastigmine, for Alzheimer’s disease, 303, 315 Sadomasochistic tendencies of borderline patients, 195–196 SAMe (S-adenosylmethionine), 87, 102 SANs (somewhat typical antipsychotics), 54 Schizoaffective disorder bipolar type, 17, 18 carbamazepine for, 58 electroconvulsive therapy for, 368, 371 mania misdiagnosed as, 20 risperidone for, 28 valproate for, 58 Schizoid personality disorder, 126 Schizophrenia, 1–2, 41–74 algorithms for treatment-refractory disease, 52–53, 69–72, 70–71 American Psychiatric Association Practice Guidelines for, 44, 54, 60, 62 antipsychotics for algorithms for, 52–53, 70–71 atypical agents, 1, 43, 52–53, 55–56, 69, 73 augmentation of, 56–61 changing between classes of, 54–55 combinations of, 58–60, 72 depot neuroleptics, 55 dosages and therapeutic levels of, 46, 47–49, 54 noncompliance with, 42, 46, 50–51, 68–69 phases of response to, 45 plasma levels of, 45, 46, 69 in pregnancy and lactation, 68 premature discontinuation of, 45–46 resistance to, 45–46

437 ❉

❉

The Difficult-to-Treat Psychiatric Patient

Schizophrenia (continued) antipsychotics for (continued) response to conventional agents, 42, 72 side effects of, 4–5, 50 substance abuse and response to, 277 clinical course of, 42 clinical vignette of, 72–73 comorbidity with depression, 51, 57 dissociative identity disorder, 214 obsessive-compulsive disorder, 51, 73 substance abuse, 51, 276–277, 282–283 electroconvulsive therapy for, 61–62, 64, 72, 367–369 combining with antipsychotics, 369 in difficult-to-treat patients, 368–369 efficacy studies of, 368 indications for, 367–368 elements of successful treatment of, 73–74 Expert Consensus Guidelines for, 51, 55, 60 factors related to poor treatment response in, 45–51 comorbid conditions, 51 medication-related issues, 45–46, 47–49 neuroleptic resistance, 45 noncompliance, 46, 50–51 plasma neuroleptic levels, 45 hyperdopaminergic model of, 60 major symptom clusters in, 69 mania misdiagnosed as, 20 medical workup for patients with, 51

❉

❉

nonresponders and partial responders to treatment for, 45, 69, 70–71 Patient Outcomes Research Team study of, 56, 397 pharmacological options for treatment-refractory disease, 54–61 antidepressants, 57 atypical agents, 55–56 benzodiazepines, 56, 60–61, 64 changing to another class of typical agent, 54–55 combining antipsychotics, 58–60, 72 depot neuroleptics, 55 dopamine agonists/ psychostimulants, 60 dosage change, 54 miscellaneous agents, 62–63, 64 mood stabilizers, 57–58, 64 positive and negative symptoms of, 42 positive predictive factors in, 42 prevalence of, 42 prevalence of treatment resistance in, 1 psychosocial interventions for, 63–68 adverse effects of, 5, 51, 65–66 causes of poor response to, 51 cognitive-behavioral therapy, 44, 64, 67 expected response to conventional therapies, 43–45 family therapy, 44, 51, 72, 73, 397 group therapy, 44, 51, 64–65, 72 individual therapy, 44, 51, 63–64 Programs for Assertive Community Treatment, 67

438 ❉

❉

Index ❉

psychosocial interventions for (continued) social skills training, 44, 51, 67–68 vocational rehabilitation, 67, 68 relapses of, 42, 43, 72 transcranial magnetic stimulation for, 381 treatment response rates in, 42–45 Schizotypal personality disorder, 126 SCID (Structured Clinical Interview for DSM-III-R), 278, 279 Seasonal affective disorder, 86, 102 Sedation, antipsychotic-induced, 50, 59, 60 Seizures electroconvulsive therapy-induced, 362, 373 posttraumatic, 312 vagus nerve stimulation for, 377 Selective serotonin reuptake inhibitors (SSRIs). See also specific drugs for agitation, 316 for anorexia nervosa, 248 for atypical depression, 84 augmentation of, 2, 92, 93, 105 for bipolar disorder, 10, 14 for borderline personality disorder, 200 for bulimia nervosa, 249, 257 for cognitively impaired patients, 317, 318 combining with other antidepressants, 94 combining with stimulants, 94 for dissociative identity disorder, 228 dosage of, 105 drug interactions with, 328 S-adenosylmethionine, 87 antipsychotics, 57, 331 for dysthymic disorder, 85

❉

electroconvulsive therapy after nonresponse to, 364 falls and fractures associated with, 333 for generalized anxiety disorder, 133, 135 as initial treatment for depression, 104 for obsessive-compulsive disorder, 124–127 overdose of, 328 for panic disorder, 138–140, 142, 143 for posttraumatic stress disorder, 150–151, 158 for psychotic depression, 83 for schizophrenia, 57 for seasonal depression, 86 side effects of, 4, 73, 128 for social phobia, 116, 119, 121 switching to other antidepressants from, 91, 105, 364 use in cardiovascular disease, 328–329 use in chronic obstructive pulmonary disease, 343 use in diabetes mellitus, 338 as analgesics, 340 Selegiline, for schizophrenia, 60 Self-mutilation, in borderline personality disorder, 184–185 Serax. See Oxazepam Serentil. See Mesoridazine Serial treatment of dual-diagnosis patients, 281 Serlect. See Sertindole Seroquel. See Quetiapine Serotonin syndrome, 87, 94 Sertindole, 47, 331 Sertraline, 291 for bipolar disorder, 32 “body sway” associated with, 333

439 ❉

❉

The Difficult-to-Treat Psychiatric Patient

Sertraline (continued) for cognitively impaired patients, 318 for obsessive-compulsive disorder, 125 for panic disorder, 138–139, 140, 142 for posttraumatic stress disorder, 150–151, 158 for psychotic depression, 83 for schizophrenia, 73 for social phobia, 116 switching from other antidepressants to, 91 use in chronic obstructive pulmonary disease, 343 use in diabetes mellitus, 338 as analgesic, 340 Serzone. See Nefazodone SET (social effectiveness therapy), for social phobia, 119, 121 Sexual dysfunction, drug-induced, 4–5 selective serotonin reuptake inhibitors, 4, 73, 128, 130, 338 SIADH (syndrome of inappropriate antidiuretic hormone), carbamazepine-induced, 332 Sick sinus syndrome, lithium-induced, 332 Side effects of drugs, 4. See also specific drugs in substance-abusing patients, 285 treatment noncompliance due to, 4 Sinemet. See Levodopa/carbidopa SIT (stress inoculation training), for posttraumatic stress disorder, 153 Sleep electroencephalogram (EEG), as predictor of treatment response in depression, 82, 83 Slow metabolizers, 3 Smoking

❉

❉

cytochrome P450-related drug interactions and, 343 depression and, 344 Social effectiveness therapy (SET), for social phobia, 119, 121 Social phobia, 115–124 age at onset of, 115 chronicity of, 122 clinical vignette of, 123–124 comorbidity with, 118–119 depression, 118–119 panic disorder, 119, 140 posttraumatic stress disorder, 150, 158 factors associated with lack of treatment response in, 118–119 pharmacotherapy for, 116–117, 121–123 benzodiazepines, 116, 121–122 β-blockers, 117, 122 buspirone, 117, 122 gabapentin, 117, 119, 122 inadequate response to, 122 monoamine oxidase inhibitors, 116 nefazodone, 117, 123 selective serotonin reuptake inhibitors, 116, 121 venlafaxine, 117, 122–123 prevalence of, 115 psychotherapy for, 117–118, 120–121 subtypes of, 116 discrete, 116, 121 generalized, 116 therapeutic alliance for treatment of, 118 treatment recommendations for, 119–123 computerized assessment, 123 pharmacotherapy, 121–123 psychotherapy, 120–121

440 ❉

❉ Social skills training for schizophrenia, 44, 51, 67–68 for social phobia, 118, 119, 121 Somatic treatments, 359–383 classifying response to, 2 electroconvulsive therapy, 360–374, 382–383 functional neurosurgery, 374–377, 383 transcranial magnetic stimulation, 377–383 vagus nerve stimulation, 377, 383 Sparine. See Promazine SSRIs. See Selective serotonin reuptake inhibitors St. John’s Wort for depression, 102 for generalized anxiety disorder, 136 Status epilepticus, electroconvulsive therapy-induced, 364 Stelazine. See Trifluoperazine Stereotactic neurosurgery. See Functional neurosurgery Stevens-Johnson syndrome, 24, 333 Stimulants abuse potential of, 290 for depression, 94 for schizophrenia, 60 for traumatic brain injury patients, 306–307, 309, 318 Strategic integrationalism, for dissociative identity disorder, 229 Stress inoculation training (SIT), for posttraumatic stress disorder, 153 Structured Clinical Interview for DSM-III-R (SCID), 278, 279 Structured Clinical Interview for the Diagnosis of DSM-IV Dissociative Disorders, Revised, 232 Subcaudate tractotomy, stereotactic, 376

❉

Index ❉ Substance abuse, 3, 273–292, 395. See also Dual-diagnosis patients Succinylcholine, for electroconvulsive therapy, 373 Suicidality, 103 akathisia and, 50 Alzheimer’s disease and, 302 anxiety disorders and, 88 “awakening” experiences and, 50 borderline personality disorder and, 185–187, 203 dissociative identity disorder and, 232, 233 dual diagnosis and, 86, 278 posttraumatic stress disorder and, 166, 167, 172–173 traumatic brain injury and, 312 Sulfonylurea interactions with monoamine oxidase inhibitors, 337 with selective serotonin reuptake inhibitors, 338 Sulpiride, for anorexia nervosa, 247 Supervision, 398–399 Supportive psychotherapy for anorexia nervosa, 250 for bulimia nervosa, 251 for schizophrenia, 44, 51 Surgery. See Functional neurosurgery Syndrome of inappropriate antidiuretic hormone (SIADH), carbamazepine-induced, 332 Synthroid, 32 T3 (triiodothyronine), for depression, 92–93 Tacrine, for Alzheimer’s disease, 303, 315 Tactical integrationalism, for dissociative identity disorder, 229 Tamoxifen, 339 Taractan. See Chlorprothixene

441 ❉

❉

The Difficult-to-Treat Psychiatric Patient

Tardive dyskinesia, 26, 28, 55, 277 electroconvulsive therapy for, 372 TBI. See Traumatic brain injury TCAs. See Antidepressants, tricyclic Tegretol. See Carbamazepine Temazepam, for posttraumatic stress disorder, 152 Temporal lobe tumors, 51 Temporolimbic syndromes, 305 Terbutaline, 343 Thalamotomy, 375 Theophylline, 343–345 for electroconvulsive therapy, 364 Therapeutic alliance, 395–396 anorexia nervosa and, 254, 395–396 bipolar disorder and, 30 borderline personality disorder and, 396 dissociative identity disorder and, 215 obsessive-compulsive disorder and, 126 posttraumatic stress disorder and, 163–164, 169, 396 social phobia and, 118 Therapeutic massage, for generalized anxiety disorder, 137 Thioridazine, 50 for agitation, 315 arrhythmia induced by, 331 dosage of, 48 Thiothixene, 72 for agitation, 315 for cognitively impaired patients, 317 combining with atypical antipsychotic, 59 dosage and therapeutic levels of, 48 Thorazine. See Chlorpromazine Thrombocytopenia, valproate-induced, 333

❉

❉

Thromboembolic disease, antipsychotic-induced, 331 Thyroid dysfunction, 2, 93, 280 bipolar disorder and, 12, 20, 27–28, 32 Thyroid-stimulating hormone (TSH), 27–28, 32 Thyroid supplementation, for depression, 92–93 L-Thyroxine, 28 Tic disorders, 128 Tindal. See Acetophenazine “Tip of the tongue” phenomenon, 11 TMS. See Transcranial magnetic stimulation Token economy, for schizophrenia, 67 Tolbutamide, interaction with selective serotonin reuptake inhibitors, 338 Topiramate for bipolar disorder, 25 for posttraumatic stress disorder, 159 for schizophrenia, 58 use in cardiovascular disease, 333 use in diabetes mellitus, 340 Torsades de pointes, antipsychoticinduced, 331 Tourette syndrome, 128 Trance phenomena, 210 Transcranial magnetic stimulation (TMS), 377–383 for acute mania, 381 for depression, 101–102, 379–380 dorsolateral prefrontal cortex, 379–380 for obsessive-compulsive disorder, 129, 381–382 for posttraumatic stress disorder, 382 relapse after, 381 repetitive, 378 for schizophrenia, 381

442 ❉

❉ technique for, 378–379 Transference reactions, 5 of bipolar patients, 30 of borderline patients, 198 of dissociative identity disorder patients, 217, 220, 222 Tranylcypromine, 90 for borderline personality disorder, 200 use in diabetes mellitus, 337 Traumatic brain injury (TBI), 299–300, 303–319, 394–395 clinical syndromes associated with, 305 management in early stages of recovery from, 306–309 mortality from, 304 neuropsychiatric features of, 303 pathogenesis of, 304–305 pharmacotherapy for, 303, 313–319 agitation, 313, 314–316 anxiety, 315, 317, 318 apathy, 318 depression, 317–318, 318 multiple symptoms and combined treatment, 319 psychosis, 313–315, 317 prevalence of, 304 prognostic factors in, 304 psychosocial treatment in later stages of recovery from, 309– 313 aggression, 310 anxiety, 312–313 cognitive disorders, 309–310 mood disorders, 311 psychosis, 312 severity of, 304 therapeutic approaches to, 303–304 Trazodone, 103 for agitation, 315, 316 for anorexia nervosa, 248

❉

Index ❉ for bulimia nervosa, 249 for cognitively impaired patients, 317, 318 combining with selective serotonin reuptake inhibitors, 94 for generalized anxiety disorder, 133, 135 for obsessive-compulsive disorder, 128 for posttraumatic delirium, 308 Treatment-resistant patients, 1–5, 3 93–400 with anxiety disorder, 115–143 with bipolar disorder, 1, 7–34 with borderline personality disorder, 179–204 classifying treatment response of, 2 comorbidity and, 3, 395 medical illness, 325–347 compliance and, 4–5 comprehensive treatment plan for, 396–397 consultation and supervision for treatment of, 398–399 costs to, 2 with dementia or traumatic brain injury, 299–320, 394–395 with depression, 1, 81–105 diagnostic accuracy and, 3, 394–395 with dissociative disorder, 209–238 with eating disorder, 243–266 inappropriate use of psychosocial interventions for, 5 medication-related factors and, 3–4 monotherapy vs. combined treatments for, 398 not giving up on, 399–400 with posttraumatic stress disorder, 149–174 prevalence of, 1 with schizophrenia, 1–2, 41–74

443 ❉

❉

The Difficult-to-Treat Psychiatric Patient

Treatment-resistant patients (continued) special somatic treatments for, 359–383 with substance abuse, 3, 273–292 therapeutic alliance and respect for, 395–396 Tremor, lithium-induced, 11 Triazolam, interaction with antidepressants, 343 Trifluoperazine, 48, 50 Triflupromazine, 48 Triiodothyronine (T3), for depression, 92–93 Trilafon. See Perphenazine Truth distortions of borderline patients, 195 TSH (thyroid-stimulating hormone), 27–28, 32 Urine toxicology screening, 278–279 Vagus nerve stimulation (VNS), 377, 383 anticonvulsant effects of, 377 for depression, 102, 377 Valproate, 291 for agitation, 315, 316 for bipolar disorder, 21–24, 32, 33 acute mania, 21 acute mixed episodes, 22 combining with carbamazepine, 22–24 combining with lithium, 21–23 side effects and noncompliance with, 11 for borderline personality disorder, 200–201 for bulimia nervosa, 250 for depression, 93 drug interactions with, 21, 23–24 lamotrigine, 24

❉

❉

warfarin, 333 for dual-diagnosis patients, 287–288 effects on thyroid hormones, 28 for panic disorder, 141 for posttraumatic epilepsy, 312 for posttraumatic stress disorder, 152, 159 for schizophrenia, 58 side effects of, 21 for traumatic brain injury patients, 309, 311 use in cardiovascular disease, 333 use in children, 21 use in chronic obstructive pulmonary disease, 344–345 use in diabetes mellitus, 340 Venlafaxine for bipolar disorder, 10, 14 for cognitively impaired patients, 318 for depression, 91, 94 for generalized anxiety disorder, 133, 135 for panic disorder, 141 for posttraumatic stress disorder, 159 for social phobia, 117, 122–123 use in cardiovascular disease, 329, 335 Venous thrombosis, antipsychoticinduced, 331 Ventricular fibrillation, 328 Verapamil for bipolar disorder, 26 interaction with carbamazepine, 333 interaction with lithium, 332 for traumatic brain injury patients, 309 Vesprin. See Triflupromazine Vitamin B12 deficiency, depression and, 88

444 ❉

❉

Index ❉

Vitamin B preparations, for druginduced cognitive/memory problems, 11 Vitamin C, for selective serotonin reuptake inhibitor-induced bruising, 329 Vitamin D deficiency, 333, 335 Vitamin E, for Alzheimer’s disease, 300, 303, 319 VNS. See Vagus nerve stimulation Vocational rehabilitation, for schizophrenia, 67, 68

tricyclic antidepressants, 337 Wellbutrin. See Bupropion Wilson’s disease, 51

Warfarin drug interactions with carbamazepine, 21, 333 fluvoxamine, 329 valproate, 333 electroconvulsive therapy and, 364, 373 Weight gain, drug-induced antipsychotics, 50, 339 monoamine oxidase inhibitors, 337 mood stabilizers, 11, 340 selective serotonin reuptake inhibitors, 128, 338

Ziprasidone, 56, 59 for agitation, 315, 316 antidepressant effects of, 57 arrhythmia induced by, 331 dosage of, 47 orthostatic hypotension induced by, 330 use in cognitively impaired patients, 317 use in diabetes mellitus, 340 weight gain induced by, 339 Zoloft. See Sertraline Zyprexa. See Olanzapine

❉

Yale-Brown Obsessive-Compulsive Scale (Y-BOCS), 125, 126, 128 Yoga for generalized anxiety disorder, 137 for obsessive-compulsive disorder, 129 Young Mania Rating Scale, 24

445 ❉