e-mail: ns@mblcommunications.com


Dr. Sussman is editor of Primary Psychiatry and professor of psychiatry at the New York University School of Medicine in New York City.

Dr. Sussman is a consultant to and on the advisory boards of GlaxoSmithKline and Wyeth; and has received honoraria from AstraZeneca, Bristol-Myers Squibb, GlaxoSmithKline, and Wyeth.


This issue of Primary Psychiatry contains Part 2 of our review of emerging insights into the causes and management of suicidal patients. Eric A. Fertuck, PhD, and colleagues provide an overview of the nature of suicidality in borderline personality disorder (BPD). They note that this common condition is associated with one of the highest rates of healthcare utilization in both the psychiatric and primary care setting. BPD is characterized by suicidal ideation and behavior.

The approximately 10% rate for suicide completion in BPD is 400 times greater than the rate in the general population. This article is written with the needs of primary care physicians (PCPs) in mind, so that they become better able to recognize the clinical manifestations of BPD and the treatment options for those patients who are suicidal. The article emphasizes that primary care is often the best setting for both the identification of those at risk for suicide and for the prevention of suicidal behavior.

Kelly Posner, PhD, and colleagues expand on this theme. The authors describe PCPs as being in the “front line” of suicide prevention. They note that an estimated 45% of suicide victims see their physician in the month prior to their death. They also describe factors known to increase risk for suicide, such as depression and alcohol use disorders as well as serious and chronic medical illness. Improved detection of suicidal patients and management of suicidal risk in primary care settings have demonstrated significant improvement in practice.

Katharine A. Phillips, MD, underscores that suicidal ideation, suicide attempts, and completed suicide are common in individuals with body dysmorphic disorder (BDD). Remarkably, approximately 80% of individuals with BDD experience lifetime suicidal ideation, and 24% to 28% have attempted suicide. Nevertheless, BDD is under-recognized in clinical settings. The article reviews available evidence on suicidality in BDD and discusses how to recognize and diagnose this often-secret disorder.

Heidi Combs, MD, and Sharon Romm, MD, note that psychiatric inpatient suicide is rare, with a prevalence of between 0.1% and 0.4% of all psychiatric admissions. However, the authors note that such an event carries an especially powerful emotional charge, since a psychiatric inpatient unit is supposed to be a safe refuge from the destructive sequelae of mental illness. The authors continue that in-hospital suicides cause additional legal problems for the care delivery system and providers, and that the most frequent legal action involving a psychiatric service is the failure to protect patients from harming themselves. This article reviews the literature on inpatient suicide to see if such events can be predicted and forestalled. It identifies patients at especially high risk and explores risk factors in the care-delivery environment, such as staffing, length of stay, and physical surroundings. Articles were chosen that evaluated suicides occurring while patients were hospitalized in an acute psychiatric unit or soon after discharge. The authors excluded studies addressing outpatient suicide or suicide in correctional settings.

Unrelated to suicide, Martina L. Rodgers, MS, and colleagues examine the empirical evidence and personal accounts showing that many people with severe and persistent mental illness can lead satisfying, meaningful lives. That is, they recover. The authors describe the emergence of qualitative and quantitative measures of recovery, summarize available process and outcome definitions, describe current research methods utilized in the recovery literature, and provide a clinical model that integrates recovery with an evidence-based practice perspective. PP


Expert Roundtable Supplement


An expert panel review of clinical challenges in psychiatry and primary care


Funding for this activity has been provided by an educational grant from Bristol-Myers Squibb.


Accreditation Statement

This activity has been planned and implemented in accordance with the Essentials and Standards of the Accreditation Council for Continuing Medical Education (ACCME) through the joint sponsorship of the Mount Sinai School of Medicine and MBL Communications, Inc. The Mount Sinai School of Medicine is accredited by the ACCME to provide continuing medical education for physicians. 


Credit Designation

The Mount Sinai School of Medicine designates this educational activity for a maximum of 2 AMA PRA Category 1 Credit(s)TM. Physicians should only claim credit commensurate with the extent of their participation in the activity.


Faculty Disclosure Policy Statement

It is the policy of the Mount Sinai School of Medicine to ensure objectivity, balance, independence, transparency, and scientific rigor in all CME-sponsored educational activities. All faculty participating in the planning or implementation of a sponsored activity are expected to disclose to the audience any relevant financial relationships and to assist in resolving any conflict of interest that may arise from the relationship. Presenters must also make a meaningful disclosure to the audience of their discussions of unlabeled or unapproved drugs or devices. This information will be available as part of the course material.

This activity has been peer reviewed and approved by Eric Hollander, MD, Chair and Professor of Psychiatry at the Mount Sinai School of Medicine. Review Date: November 9, 2007.


Statement of Need and Purpose

Major depressive disorder (MDD) is the fourth largest contributor to the worldwide burden of disease and is expected to be second only to ischemic heart disease by the year 2020. Over 60% of suicide deaths in the United States are directly attributable to MDD, and >300,000 people successfully commit suicide in the US annually. Despite its prevalence, 50% of MDD cases go undetected, undiagnosed, and untreated. The impairment of depression can lead to decreased productivity, alcohol and substance abuse, and an increased risk of suicide.  Treatment resistant depression (TRD) is frequently defined as depressive illness that does not fully remit after a single initial treatment failure. Patients who only achieve partial response or continue to experience residual symptoms are likely to show reduced functioning and an increased risk of relapse. Up to 50% of patients do not show a full response to their first antidepressant treatment. This has led to a re-emergence of interest in treatment augmentation research. There is a higher frequency of suicide in patients with TRD as opposed to those with treatment responsive MDD. Although the results of several open-label trials suggest a potential role of second-generation antipsychotics (SGAs) in TRD, there has been a paucity of double-blind, placebo-controlled studies confirming whether this treatment strategy is truly effective. New data continue to emerge and it is important to determine how these findings apply to each of the SGAs. It is also important to report on the safety, tolerability, and efficacy of augmenting with SGAs versus other augmentation or switching strategies for TRD.

Target Audience

This activity is designed to meet the educational needs of primary care physicians and psychiatrists.

Learning Objectives

• Recognize the efficacy, safety, and tolerability of augmenting pharmacologic treatment of major depressive disorder (MDD) with atypical antipsychotics.
• Discuss the challenges of limited response to MDD treatment and its impact on the course of illness.


Faculty Disclosures

Michael E. Thase, MD, is a consultant to AstraZeneca, Bristol-Myers Squibb, Cephalon, Cyberonics, Eli Lilly, GlaxoSmithKline, Janssen, MedAvante, Neuronetics, Novartis, Organon, Sepracor, Shire, Supernus, and Wyeth; is on the speaker’s bureaus of AstraZeneca, Bristol-Myers Squibb, Cyberonics, Eli Lilly, GlaxoSmithKline, Organon, sanofi-aventis, and Wyeth; has equity in MedAvant; and receives book royalties from American Psychiatric Publishing, Guilford Publications, and Herald House. Dr. Thase discloses that he will discuss investigational uses of older pharmacologic agents for the treatment of major depressive disorder (MDD).

J. Craig Nelson, MD, is a consultant to and/or on the advisory boards of Abbott, Biovail, Bristol-Myers Squibb, Corcept, Eli Lilly, Forest, GlaxoSmithKline, Novartis, Orexigen, Organon, and Pfizer. Dr. Nelson discloses that he will discuss unapproved/investigational uses of pharmacologic agents for the treatment of MDD.

George I. Papakostas, MD, has served as a consultant to Aphios, Bristol-Myers Squibb, GlaxoSmithKline, Evotec, Inflabloc, Jazz, PAMLAB, Pfizer, and Wyeth; has received honoraria from Bristol-Myers Squibb, Evotec, GlaxoSmithKline, Inflabloc, Jazz, Lundbeck, PAMLAB, Pfizer, Titan, and Wyeth; and has received research support from Bristol-Myers Squibb, PAMLAB, and Pfizer. Dr. Papakostas discloses that he will discuss unapproved/investigational uses of aripiprazole, buspirone, olanzapine, pindolol, quetiapine, risperidone, triiodothyronine, and ziprasidone for the treatment of MDD.

Michael J. Gitlin, MD, has received honoraria from AstraZeneca, Bristol-Myers Squibb, Cephalon, Eli Lilly, GlaxoSmithKline, Pfizer, and Takeda. Dr. Gitlin discloses that he will discuss unapproved/investigational use of aripiprazole, clozapine, olanzapine, quetiapine, risperidone, and ziprasidone for the treatment of MDD.

Acknowledgment of Commercial Support

Funding for this activity has been provided by an educational grant from Bristol-Myers Squibb.


Peer Reviewers

David L. Ginsberg, MD, receives honoraria from AstraZeneca and GlaxoSmithKline.

Eric Hollander, MD, reports no affiliation with or financial interest in any organization that may pose a conflict of interest.


To Receive Credit for this Activity

Read this expert roundtable supplement, reflect on the information presented, and complete the CME posttest and evaluation. To obtain credit, you should score 70% or better. Early submission of this posttest is encouraged. Please submit this posttest by December 1, 2009 to be eligible for credit.

Release date: December 1, 2007
Termination date: December 31, 2009

The estimated time to complete this activity is 2 hours.


Under optimal circumstances, patients respond to treatments for major depression only 60% to 70% of the time. Therefore, there is a critical need for effective treatment strategies that augment available depression treatment. Currently, such strategies augment primary antidepressants with agents that increase the likelihood of treatment response. Augmentation agents include thyroid hormones, which are used to augment tricyclic antidepressants (TCAs); lithium, which also improves response to TCAs; and second-generation antipsychotics (SGAs), which are used to augment selective serotonin reuptake inhibitors and serotonin norepinephrine reuptake inhibitors. Other, less common, strategies include augmentation with stimulants, folate, and buspirone. Unfortunately, studies of augmentation efficacy are often limited or equivocal. Studies may overestimate the magnitude of effect, as augmentation may be attempted while patients still experience an initial response. Prescribers must be sure treatment strategies are not undermined by safety or tolerability concerns. Lithium, in particular, is not well tolerated by patients, and SGAs pose the risk of tardive dyskinesia, metabolic syndrome, and extrapyramidal symptoms. Clinicians must weigh these issues against a relatively limited base of knowledge.   

In this Expert Roundtable Supplement, Michael E. Thase, MD, discusses the history of augmentation strategies for depression. J. Craig Nelson, MD, reviews recent findings on augmentation with thyroid hormone, lithium, buspirone, and modafinil. George I. Papakostas, MD, reviews the efficacy of augmentation with SGAs. Finally, Michael J. Gitlin, MD, provides an overview of safety and tolerability issues.



Augmentation Strategies for Depression: History and Concepts

By Michael E. Thase, MD — Moderator

Dr. Thase is professor of psychiatry at the University of Pennsylvania School of Medicine in Philadelphia, the Philadelphia Veterans Affairs Medical Center and the University of Pittsburgh Medical Center.

Disclosures: Dr. Thase is a consultant to AstraZeneca, Bristol-Myers Squibb, Cephalon, Cyberonics, Eli Lilly, GlaxoSmithKline, Janssen, MedAvante, Neuronetics, Novartis, Organon, Sepracor, Shire, Supernus, and Wyeth; is on the speaker’s bureaus of AstraZeneca, Bristol-Myers Squibb, Cyberonics, Eli Lilly, GlaxoSmithKline, Organon, sanofi-aventis, and Wyeth; has equity in MedAvant; and receives book royalties from American Psychiatric Publishing, Guilford Publications, and Herald House.


Introduction: Need for Augmentation Strategies

Depression is one of the world’s great public health problems. As there are no perfect or uniformly effective treatments for depression, it is not surprising that treatment-resistant depression (TRD) is likewise an important public health problem. Although the potential benefits of antidepressants are now well documented, no widely used antidepressant can be expected to be effective in more than half the patients who begin to take it. Even under optimal circumstances (ie, a patient who is fully adherent to 12 weeks of treatment), there is only a 60% to 70% chance that the first choice of medication will be effective. The need for effective alternate strategies for TRD, as well as the need for innovations in service delivery systems to ensure those strategies are implemented in a timely manner, are foremost to fully realizing the potential benefits of antidepressant therapies.

Over the years, hierarchies of treatment strategies for TRD have been based on the widespread use of particular treatments, their ease of use, and their safety or complexity. One of the strategies consistently used since its introduction 20 years ago has been augmentation of the ineffective antidepressant by a second medication. The second agent may or may not have antidepressant effects of its own, but when used in combination with a primary antidepressant the agent reliably increases a patient’s likelihood of response and symptom remission.

When augmentation strategies were first implemented, only three classes of medication—the tricyclic antidepressants (TCAs), the heterocyclic alternatives to the TCAs (eg, trazodone, amoxapine, and maprotiline), and the monoamine oxidase inhibitors (MAOIs)—were available. These original antidepressants have since declined in use, largely replaced by a raft of newer antidepressants that, while no more effective on average, have more favorable profiles with respect to ease of use, tolerability, and safety in overdose. The number of options available to patients who do not respond to first-line treatments has thus also increased, multiplying the potential combinations of antidepressant medications and augmenting agents. Ironically, some of the most gratifying responses to pharmacotherapy observed among patients with TRD are observed with “old school” medications such as the TCA clomipramine and the MAOI tranylcypromine.


Augmentation Strategies   

The earliest augmentation strategies were first employed in the 1960s, almost coincident with the introduction of the first antidepressants (Slide 1). For example, benzodiazepines were commonly used to enhance the anxiolytic or sedative hypnotic effects experienced by patients taking TCAs or MAOIs. Evidence from studies performed in the 1960s indicated that anxiolytic medications rapidly and reliably reduced anxiety symptoms and insomnia associated with depressive states and, when administered from the outset, might hasten treatment response.1-3 However, concomitant prescription of anxiolytics did not greatly increase the likelihood of patient response or symptom remission, and their longer-term utility (in combination with antidepressants) was never systematically confirmed. Although even greater use was limited by concerns about abuse liability, many experts believe that concerns about the risk of benzodiazepine addiction in this instance have been overstated4 and a large number of patients with difficult-to-treat forms of depression continue to receive palliative benefit from concomitant prescription of benzodiazapines.




First-Generation Antipsychotics
Combinations of antipsychotics and antidepressants were also fairly widely used in the 1960s and 1970s to treat more severe depressive episodes, particularly those characterized by agitation, anxiety, and of course psychosis (Slide 2).5,6 In fact, a proprietary combination of perphenazine and amitriptyline, the most prevalent such combination in the United States, remained in use well into the 1980s, although the fixed-dose combination formulation was more popular in primary care than psychiatric settings, perhaps because the specialists preferred to titrate the component medications separately. Other combinations were used both in the US and abroad, including the antipsychotic trifluperazine and the MAOI tranylcypromine.7 In retrospect, the amitriptyline plus perphenazine combination capitalized on pharmacokinetic interactions: perphenazine increased amitriptyline and, to a lesser extent, nortriptyline blood levels, functionally doubling the dose of the TCA for the average patient. However, with the growing recognition that first-generation antipsychotics conveyed the risk of tardive dyskinesia (TD), coupled with the still controversial observation that patients with mood disorders were even more likely to develop TD than patients with schizophrenia, these combinations were used with increasing reluctance. By the 1990s, this strategy was no longer widely used.



Psychiatrists thus have known for nearly 50 years that antipsychotics have the potential to enhance antidepressant effects, particularly for severely ill patients. This augmentation strategy was largely discontinued, partly because of the introduction of other options and partly for fear of TD. The availability of the second-generation (atypical) antipsychotics (SGAs) has, of course, changed the therapeutic landscape considerably. As is discussed in subsequent sections of this supplement, even though SGAs are less associated with TD in short- and intermediate-term studies of schizophrenia, there continues to be uncertainty about whether patients with resistant depression who respond to augmentation with SGAs will require long-term administration and, if so, whether they are indeed at greater risk for development of TD.


Thyroid Hormone Augmentation

Beginning in the 1960s, reports suggested that thyroid hormone, when added to a TCA at the beginning of a treatment regimen, could accelerate or increase the likelihood of response.8-10 Interestingly, the original observations suggested the strategy was more useful for depressed women, and efficacy specifically in depressed men was never established.11 As women are at higher risk for undetected thyroid disease, one parsimonious explanation for the therapeutic activity of adjunctive thyroid hormone was correction of “pre-clinical” hypothyroidism.12 Consistent with this view, the performance of thyroid augmentation has been disappointing in studies of euthyroid depressed patients.13

These observations reinforced a long-abiding interest in the role of the thyroid axis in the etiopathogenesis of mood disorders. One view is that the thyroid axis responds to the stress of an affective disorder with increased hypothalamic drive, ie, elevated levels of thyroid releasing hormone (TRH). This, over time, results in increased thyroid tone, ie, higher-than-usual levels of circulating thyroid hormone and blunted thyroid stimulating hormone response to TRH. As more recent observations suggest that even patients with low normal levels of thyroid hormone have greater rates of depression14 and slower or less robust responses to antidepressants (Slide 3),15,16 it has been speculated that depressed patients may require a higher-than-usual level of thyroid activity to benefit fully from antidepressant therapy. The efficacy of thyroid augmentation thus may be linked to induction of a higher-than-usual thyroid state, which in turn may selectively benefit patients with low-normal thyroid functions.



It remains unknown whether better results are obtained with triiodothyronine (T3), the centrally active form of thyroid hormone, or whether thyroxine (T4), the normal thyroid replacement hormone, is equally useful. More evidence exists indicating that T3 can be used to enhance an incompletely effective antidepressant, but only because T3 has been more extensively studied than T4.


Lithium Augmentation

Since the 1960s, lithium augmentation has been used to enhance the effects of antidepressants. One of the early case series reported on the utility of the combination of lithium with MAOIs in a group consisting predominantly of patients with bipolar disorder.17 On the basis of several studies in the early 1980s,18-20 lithium augmentation became the preferred augmentation strategy for unipolar patients not adequately responding to TCAs. Several initial reports suggested dramatic antidepressant effects within 24–72 hours.18,19 Results of subsequent studies, however, reported that this kind of rapid and dramatic augmentation response is less common than a slower emerging response over 4–6 weeks (Slide 4),21,22 which is certainly more suggestive of a primary antidepressant effect.22 As lithium salts have antidepressant effects for a subset of depressed people,23 it is difficult to say whether lithium’s therapeutic effects are actually due to augmentation of the antidepressant or a primary antidepressant effect.24 Unfortunately, despite more than 20 years of research, the proper experiment to answer this question still has not been undertaken.




The challenging problem of treating people who do not benefit from first-line antidepressant medications has not been solved by the availability of newer classes of drugs. Over the years one strategy used by physicians treating depressed people are not benefiting from antidepressant monotherapy has been to add a second medication thought to augment or enhance the actions of the primary medication. The concepts that guided selection of the first strategies used for this purpose—anxiolytics, antipsychotics, thyroid hormone, and lithium—continue to be relevant to the treatment of depression and guide clinicians’ choices of medications in an effort to enhance the effects of newer-generation antidepressants.



1. Blackman B. The adjunctive role of diazepam in the treatment of depression. Clin Med (Northfield Il). 1963;70:1495-500.
2. Hare HP Jr. Comparison of chlordiazepoxide-amitriptyline combination with amitriptyline alone in anxiety-depressive states. J Clin Pharmacol New Drugs. 1971;11(6):456-460.
3. Feighner JP, Brauzer B, Gelenberg AJ, et al. A placebo-controlled multicenter trial of Limbitrol versus its components (amitriptyline and chlordiazepoxide) in the symptomatic treatment of depressive illness. Psychopharmacology (Berl). 1979;61(2):217-225.
4. Shader RI, Greenblatt DJ. Benzodiazepine overuse-misuse. J Clin Psychopharmacol. 1984;4(3):123-124.
5. Robertson MM, Trimble MR. Major tranquillisers used as antidepressants. A review. J Affect Disord. 1982;4(3):173-193.
6. Spiker DG, Weiss JC, Dealy RS, et al. The pharmacological treatment of delusional depression. Am J Psychiatry. 1985;142(4):430-436.
7. Mena A, Heistad G, Schiele BC, Janecek J. A comparison of tranylcypromine alone with tranylcypromine plus trifluoperazine in the treatment of chronic outpatients: a double-blind controlled study. J Neuropsychiatr. 1964;5:542-550.
8. Prange AJ Jr, Wilson IC, Rabon AM, Lipton MA. Enhancement of imipramine antidepressant activity by thyroid hormone. Am J Psychiatry. 1969;126(4):457-469.
9. Prange AJ Jr, Wilson IC, Lipton MA, Rabon AM, McClae TK, Knox AE. Use of a thyroid hormone to accelerate the action of imipramine. Psychosomatics. 1970;11(5):442-444.
10. Wilson IC, Prange AJ Jr, McClane TK, Rabon AM, Lipton MA. Thyroid-hormone enhancement of imipramine in nonretarded depressions. N Engl J Med. 1970;282(19):1063-1067.
11. Altshuler LL, Bauer M, Frye MA, et al. Does thyroid supplementation accelerate tricyclic antidepressant response? A review and meta-analysis of the literature. Am J Psychiatry. 2001;158(10):1617-1622.
12. Haggerty JJ Jr, Stern RA, Mason GA, Beckwith J, Morey CE, Prange AJ Jr. Subclinical hypothyroidism: a modifiable risk factor for depression? Am J Psychiatry. 1993;150(3):508-510.
13. Thase ME, Kupfer DJ, Jarrett DB. Treatment of imipramine-resistant recurrent depression: I. An open clinical trial of adjunctive L-triiodothyronine. J Clin Psychiatry. 1989;50(10):385-388.
14. 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. 1999;156(12):1909-1914.
15. Cole DP, Thase ME, Mallinger AG, et al. Slower treatment response in bipolar depression predicted by lower pretreatment thyroid function. Am J Psychiatry. 2002;159(1):116-121.
16. Gitlin M, Altshuler LL, Frye MA, et al. Peripheral thyroid hormones and response to selective serotonin reuptake inhibitors. J Psychiatry Neurosci. 2004;29(5):383-386.
17. Himmelhoch JM, Detre T, Kupfer DJ, Swartzburg M, Byck R. Treatment of previously intractable depressions with tranylcypromine and lithium. J Nerv Ment Dis. 1972; 155(3):216-220.
18. de Montigny C, Grunberg F, Mayer A, Deschenes JP. Lithium induces rapid relief of depression in tricyclic antidepressant drug non-responders. Br J Psychiatry. 1981;138:252-256.
19. de Montigny C, Cournoyer G, Morissette R, Langlois R, Caille G. Lithium carbonate addition in tricyclic antidepressant-resistant unipolar depression. Correlations with the neurobiologic actions of tricyclic antidepressant drugs and lithium ion on the serotonin system. Arch Gen Psychiatry. 1983;40(12):1327-1334.
20. Heninger GR, Charney DS, Sternberg DE. Lithium carbonate augmentation of antidepressant treatment. An effective prescription for treatment-refractory depression. Arch Gen Psychiatry. 1983;40(12):1335-1342.
21. Price LH, Charney DS, Heninger GR. Variability of response to lithium augmentation in refractory depression. Am J Psychiatry. 1986;143(11):1387-1392.
22. Thase ME, Kupfer DJ, Frank E, Jarrett DB. Treatment of imipramine-resistant recurrent depression: II. An open clinical trial of lithium augmentation. J Clin Psychiatry. 1989b;50(11):413-417.
23. Kupfer DJ, Pickar D, Himmelhoch JM, Detre TP. Are there two types of unipolar depression? Arch Gen Psychiatry. 1975;32(7):866-871.
24. Thase ME, Howland RH, Friedman ES. Treating antidepressant nonresponders with augmentation strategies: an overview. J Clin Psychiatry. 1998;59(suppl 5):5-12.



Recent Findings and Current Status of Augmentation Strategies

By J. Craig Nelson, MD

Dr. Nelson is the Leon J. Epstein professor of psychiatry and director of geriatric psychiatry in the Department of Psychiatry at the University of California San Francisco.

Disclosures: Dr. Nelson is a consultant to and/or on the advisory boards of Abbott, Biovail, Bristol-Myers Squibb, Corcept, Eli Lilly, Forest, GlaxoSmithKline, Novartis, Orexigen, Organon, and Pfizer.



Augmentation strategies have become popular in patients with a partial response or residual symptoms. In a survey conducted by the American Society of Consultant Pharmacists, 65% of 169 psychiatrists indicated that they would augment treatment as the next step in a partial responder in contrast to non-responders in whom they would switch to another drug (J.C. Nelson, unpublished data). This decision is based on common sense and practicality rather than empirical evidence. Many clinicians and patients think that if a patient achieves at least a partial response to a medication, it makes sense to continue that drug and add a second.  To some extent, the patient preferences for treatment shown in the Systematic Treatment Alternatives to Relieve Depression (STAR*D) study  reflect this approach.1 This discussion will update clinicians on the use of augmentation agents for the treatment of major depressive disorder (MDD).


Lithium Augmentation   

Lithium augmentation has been used since it was first described by de Montigny and colleagues2 in 1981. This strategy is based on a rational neurochemical hypothesis that lithium has a synergistic effect when added to a tricyclic antidepressant (TCA). Lithium increases serotonin turnover and the TCA increases postsynaptic serotonin receptor sensitivity.3 Recent debates have focused on whether lithium is as effective combined with selective serotonin reuptake inhibitors (SSRIs) as it is combined with a TCA. The rapid effects sometimes observed with lithium augmentation of TCAs does not seem to occur with SSRIs, conceivably because SSRIs do not increase postsynaptic serotonin receptor sensitivity.

A recent meta-analysis by Crossley and Bauer4 found  10 placebo-controlled lithium augmentation studies.  Augmentation with lithium was significantly more effective than augmentation with placebo. Forty percent of study participants on lithium responded versus 17% on placebo. However, many of the studies included in this meta-analysis were quite small. While one study had 61 patients, the other samples had ≤35 patients. Usual doses of lithium during augmentation were 600–900 mg/day, which translates into a lithium level of ≥0.4 mEq/L. Most recently, lithium augmentation was compared with triiodothyronine (T3) augmentation as a Level 3 strategy in the STAR*D study.5 Patients achieved a reasonable final lithium dose of about 900 mg/day; however, remission rates, while not significantly different, appeared to favor T3 (25% versus 13% to 16%) (Slide 1).6  Lithium was significantly less well tolerated than T3, with more patients discontinuing treatment.


Placebo-controlled study of the use of lithium for patients with treatment-resistant depression (TRD) is limited. Several early studies added lithium after 4 weeks of limited response to the initial antidepressant, and it is not clear that these patients were really treatment resistant. A recent study by Nierenberg and colleagues6 found lithium augmentation was not  more effective than placebo in a group of 35 patients who had failed a 6-week prospective nortriptyline trial.  

Another possible explanation of the apparent reduced effectiveness of lithium augmentation is the increased recognition of bipolar spectrum illness. We found, in patients with psychotic depression, that lithium was particularly effective for augmentation in patients with first-degree relatives with bipolar illness or probable histories of hypomania.7 It is possible that as clinicians have become more alert to the diagnosis of bipolar disorder and exclude these patients from studies, the efficacy of lithium augmentation has declined. 


Thyroid Augmentation   

Thyroid augmentation is another commonly employed augmentation strategy. Several controlled studies have suggested this strategy may accelerate response. Altshuler and colleagues8 performed a meta-analysis of six studies using T3 to accelerate response. Five of these showed a significant effect in increasing the speed of response. The effect was particularly notable in women. Three placebo-controlled studies subsequent to the meta-analysis have examined acceleration of response with T3, with two finding an advantage for T3.9,10 

The evidence for the effectiveness of T3 in patients with TRD is less clear. Aronson and colleagues11 performed a meta-analysis of eight controlled studies of T3 augmentation (Slide 2). Four of the studies were placebo-controlled, and those that were not used other controls or comparisons. For example, one study compared T3 and T4, and three of the other studies used a historical control for comparison. Overall, the studies indicated that T3 had a significant effect, with an odds ratio of 2 in 292 patients. However, in the four placebo-controlled trials, T3 was not superior to placebo. The placebo-controlled studies included 75 patients, and the odds ratio was 1.5 with a wide confidence interval.


The nature of the thyroid trials varied widely, and there are limitations to the studies. A small placebo-controlled trial reported by Gitlin and colleagues12 used a crossover design after randomizing patients to T3 or placebo. Crossover designs in depression are limited by the lingering effects of the prior treatment. Goodwin and colleagues13 substituted T3 for placebo in a blinded fashion, but without randomization to a parallel comparison. Arguably the best designed study was conducted by Joffe and colleagues.14 This was a placebo-controlled parallel comparison of T3, lithium and placebo in 50 patients receiving TCAs. Both T3 and lithium were more effective than placebo but augmentation occurred after only 5 weeks of failure to respond to the first antidepressant. The use of T3 in patients we would now consider treatment resistant is limited. There are no placebo-controlled studies of T3 in SSRI-resistant patients. The best evidence for the efficacy of T3 augmentation of SSRIs in TRD comes from STAR*D which found that ~25% of patients who had failed two prior antidepressant trials, remitted with T3.5 Forty-seven of the 70 patients receiving T3 were taking citalopram or sertraline.  

The dose of T3 for thyroid augmentation is typically 25–50 μg/day. T3 has relatively few side effects. As is the case with almost all augmentation strategies, it is unknown how long thyroid augmentation should be continued. Although no significant problems have yet been identified with long-term treatment, this has not been well studied in depression.


Stimulant Augmentation   

Stimulants are also commonly used to augment antidepressants but this strategy is not as well studied as lithium or thyroid augmentation. Several open studies, previously reviewed,15 with <100 total patients reported that the addition of methylphenidate or dextroamphetamine to a TCA or MAOI could be useful in the addressing TRD. Stimulants were usually employed in patients who were anergic or fatigued.  A recent study by Patkar and colleagues16 examined methylphenidate augmentation in 60 patients with MDD who had been resistant to at least one trial with various antidepressants, primarily SSRIs (Slide 3). These patients were randomly assigned to extended-release methylphenidate, 18–54 mg/day, or to placebo for a duration of 4 weeks. The response rates were 40% with methylphenidate and 23% with placebo, but this difference was not significant. Of note, the sample of 60 patients may have been too small to  detect subtle differences.


Modafinil has also been used to improve alertness or energy in targeted populations of sedated or fatigued patients. Several open trials support the use of modafinil as an augmentation agent. A double-blind study by Fava and colleagues17 looked at patients with MDD who experienced a partial response to 8 weeks of SSRI monotherapy and had residual fatigue or sleepiness (Slide 4). The study also implemented scales for sleepiness and fatigue. The patients were randomized to modafinil or placebo for 8 weeks. Approximately 150 patients participated in each arm of the study. Patients receiving modafinil showed significantly greater improvement on the Clinical Global Impression (CGI) scale.



Thase and colleagues18 performed a 12-week, open-label, dose-titration extension study of the study by Fava and colleagues.17 Patients who had received placebo in the randomized phase had the option to switch to modafinil. The response was sustained in patients, and some patients who had failed to respond to modafinil during the initial 8-week trial responded during the 12-week treatment with modafinil.

The mechanism of action for modafinil is less clear than that of other stimulants, and some clinicians have limited experience with this agent. It has thus far been used to treat specific residual symptoms. The usual dose has been 200–300 mg/day. A randomized trial by Fava and colleagues17  is one of the largest augmentation studies conducted and showed modafinil to be quite effective in treating patients with specific residual symptoms. 


Buspirone Augmentation   

Buspirone was the first augmentation strategy proposed for use with the SSRIs in 1991.19,20 Several previously reviewed15 open studies, with samples ranging up to 30 patients, suggested efficacy. However, a controlled study by Landen and colleagues21 found that buspirone augmentation was not more effective than placebo after 4 weeks of SSRI treatment failure in 119 patients. The high placebo response rate of 47% suggested that patients were continuing to improve with the initial SSRI. 

Buspirone augmentation was employed in Level 2 of the STAR*D study.1 While the overall remission rate, 32.9%, looked promising, patients who  elected to receive augmentation had fewer symptoms at the start of Level 2 treatment than the patients who elected or were randomized to switch.  The ultimate efficacy of buspirone for augmentation has yet to be determined.


Other Augmentation Strategies

Although pindolol has shown efficacy when used to accelerate initial SSRI treatment,15 it has not been effective in patients with TRD. In two studies of TRD patients, particularly the larger study by Perez and colleagues,22 augmentation with pindolol was no more effective than augmentation with placebo.

An emerging body of evidence suggests lamotrigine may be useful as an augmentation strategy. In addition to case reports, three retrospective series of cases (N=14, 34, and 37) suggest the value of the addition of lamotrigine to an ongoing antidepressant in unipolar patients who have been resistant to treatment.23-25 In addition, one placebo-controlled trial in 23 patients, who had failed at least one prior antidepressant, randomly assigned patients to the addition of lamotrigine or placebo to fluoxetine 20 mg/day.26 The sample included 8 patients with bipolar II depression and 15 patients with unipolar MDD.  In both groups lamotrigine failed to show superiority on the primary outcome measure, the Hamilton Rating Scale for Depression, but did show an advantage on the CGI scale. Lamotrigine would appear to be a promising but not established strategy.  

Several studies have examined the use of testosterone in depressed men. A report of five cases suggested testosterone replacement therapy might be useful in hypogonadal men with SSRI-resistant depression.27 The results of placebo-controlled studies have been mixed. Pope and colleagues28 randomly assigned 22 depressed men who had failed 4 weeks of antidepressant treatment and had low or borderline testosterone levels, to the addition of either testosterone gel or placebo to their ongoing antidepressant and found this strategy effective. The authors noted that the effect was only observed in patients whose serum testosterone levels rose appreciably.  A second randomized controlled trial (RCT) in 18 hypogonadal depressed men who were partial responders to prior treatment added testosterone gel or placebo to the continuing antidepressant. The study employed a 12-week trial followed by a crossover. No advantage for testosterone was observed.29 The third RCT examined intramuscular testosterone or placebo in 26 depressed subjects who had failed two prior antidepressant trials and were now receiving an SSRI.30 Again testosterone augmentation was not found to be superior to placebo. At this point, the usefulness of testosterone in depressed men with either low or adequate testosterone levels remains unclear.

Estrogen has been examined as a potential augmentation agent in women. Various comparisons have been performed with monotherapy, augmentation, estrogen versus an antidepressant, and in peri- and post-menopausal women. Two large retrospective analyses of SSRI trials found that postmenopausal women receiving hormone replacement therapy were more likely to respond to the SSRI.31,32 However, results of prospective studies have been mixed.  A small (N=16) study of perimenopausal women, who were partial responders and who were experiencing other perimenopausal symptoms, found that adding estrogen to antidepressant treatment elicited a better response than adding placebo.33 Two other placebo-controlled studies, however, in postmenopausal women, found no advantage of augmenting sertraline with an estrogen patch (N=22)34 or adding an estrogen/progesterone combination to venlafaxine (N=56).35 These studies illustrate the need for a placebo control as all groups in both studies improved. The controlled data to date suggest that estrogen augmentation in postmenopausal women is not effective but may be useful in symptomatic perimenopausal women who are depressed. 

Several studies have indicated that high homocysteine levels and low B12 and folate levels are associated with depression. Papakostas and colleagues36 found that low folate serum levels were associated with reduced response to fluoxetine. In a 10-week RCT by Coppen and Bailey,37 127 subjects with MDD who were receiving fluoxetine but were not treatment resistant were randomly assigned to folate 500 μg or placebo (Slide 5). Female patients taking folate were more likely to respond than those taking placebo. This effect was not observed in men and it was suggested this dose may be too low in men. Although the value of folate has not been established in TRD, given the minimal side effects observed, this simple strategy ought to be considered in women. Alternatively it is not clear if this effect is limited to women with low folate or if additional folate would be useful to women already taking multi-vitamins that often contain about 400 μg of folic acid.




Currently several augmentation strategies have been described for use in patients with TRD. Unfortunately, with the exception of STAR*D, almost all of these studies have been conducted in samples of <100 subjects. Only recently have studies employed designs that more carefully assure that patients have received adequate prior treatment and are in fact treatment resistant. Seldom have two augmentation strategies been compared with each other under similar conditions, thus the relative magnitude of their effects is largely unknown.  The lack of large comparison studies also limits what is known about predictors of response to these treatments. Thus selection of a particular agent is based more on clinical experience than empirical data. Some strategies—augmentation with T3, estrogen, testosterone, and folate—would appear to make sense in patients deficient in that substance, but that rational hypothesis has not been confirmed. Finally, the following question remains: If the patient responds to augmentation, should the augmenting agent be continued? This question has seldom been studied.



1. Rush AJ, Trivedi MH, Wisniewski SR, et al. Acute and longer-term outcomes in depressed outpatients requiring one or several treatment steps: a STAR*D report. Am J Psychiatry. 2006;163:1905-1917.
2. Dé Montigny C, Grunberg F, Mayer A, Deschenes JP. Lithium induces rapid relief of depression in tricyclic antidepressant drug non-responders. Br J Psychiatry. 1981;138:252-256.
3. Nelson JC. Lithium augmentation in refractory depression. In: Roose SP, Glassman.AH. Treatment Strategies for Refractory Depression. Washington, DC: American Psychiatric Press, Inc; 1990:35-49.
4. Crossley NA, Bauer M. Acceleration and augmentation of antidepressants with lithium for depressive disorders: two meta-analyses of randomized, placebo-controlled trials. J Clin Psychiatry. 2007;68:935-940.
5. Nierenberg AA, Fava M, Trivedi MH, et al. A comparison of lithium and T(3) augmentation following two failed medication treatments for depression: a STAR*D report. Am J Psychiatry. 2006;163(9):1519-1530.
6. Nierenberg AA, Papakostas GI, Petersen T, et al. Lithium augmentation of nortriptyline for subjects resistant to multiple antidepressants. J Clin Psychopharmacol.2003;23:92-95.
7. Nelson J C, Mazure C. Lithium augmentation in psychotic depression refractory to combined drug treatment. Am J Psychiatry. 1986;143:363-366.
8. Altshuler LL, Bauer M, Frye MA, et al. Does thyroid supplementation accelerate tricyclic antidepressant response? A review and meta-analysis of the literature. Am J Psychiatry. 2001;158(10):1617-1622.
9. Posternak M, Novak S, Stern R, et al. A pilot effectiveness study: placebo-controlled trial of adjunctive L-triiodothyronine (T3) used to accelerate and potentiate the antidepressant response. Int J Neuropsychopharmacol. 2007;13:1-11.
10. Cooper-Kazaz R, Apter JT, Cohen R, et al. Combined treatment with sertraline and liothyronine in major depression: a randomized, double-blind, placebo-controlled trial. Arch Gen Psychiatry. 2007;64:679-688.
11. Aronson R, Offman HJ, Joffe RT, Naylor CD. Triiodothyronine augmentation in the treatment of refractory depression. A meta-analysis. Arch Gen Psychiatry. 1996;53(9):842-848.
12. Gitlin MJ, Weiner H, Fairbanks L, Hershman JM, Friedfeld N. Failure of T3 to potentiate tricyclic antidepressant response. J Affect Disord. 1987;13(3):267-272.
13. Goodwin FK, Prange AJ Jr, Post RM, Muscettola G, Lipton MA. Potentiation of antidepressant effects by L-triiodothyronine in tricyclic nonresponders. Am J Psychiatry. 1982;139(1):34-38.
14. Joffe RT, Singer W, Levitt AJ, MacDonald C. A placebo-controlled comparison of lithium and triiodothyronine augmentation of tricyclic antidepressants in unipolar refractory depression. Arch Gen Psychiatry. 1993;50:387-393.
15. Nelson JC. Augmentation strategies in depression 2000. J Clin Psychiatry. 2000;61(suppl 2):13-19.
16. Patkar AA, Masand PS, Pae CU, et al. A randomized, double-blind, placebo-controlled trial of augmentation with an extended release formulation of methylphenidate in outpatients with treatment-resistant depression. J Clin Psychopharmacol. 2006;26(6):653-656.
17. Fava M, Thase ME, DeBattista C. A multicenter, placebo-controlled study of modafinil augmentation in partial responders to selective serotonin reuptake inhibitors with persistent fatigue and sleepiness. J Clin Psychiatry. 2005;66(1):85-93.
18. Thase ME, Fava M, DeBattista C, Arora S, Hughes RJ. Modafinil augmentation of SSRI therapy in patients with major depressive disorder and excessive sleepiness and fatigue: a 12-week, open-label, extension study. CNS Spectr. 2006;11(2):93-102.
19. Jacobsen FM.  Possible augmentation of antidepressant response by buspirone. J Clin Psychiatry. 1991;52:217-220.
20. Bakish D. Fluoxetine potentiation by buspirone: Three case histories. Can J Psychiatry. 1991;36:749-750.
21. Landén M, Björling G, Agren H, Fahlén T. A randomized, double-blind, placebo-controlled trial of buspirone in combination with an SSRI in patients with treatment-refractory depression. J Clin Psychiatry. 1998;59(12):664-668.
22. Pérez V, Soler J, Puigdemont D, Alvarez E, Artigas F, Grup de Recerca en Trastorns Afectius. A double-blind, randomized, placebo-controlled trial of pindolol augmentation in depressive patients resistant to serotonin reuptake inhibitors. Arch Gen Psychiatry. 1999;56(4):375-379.
23. Gabriel A. Lamotrigine adjunctive treatment in resistant unipolar depression: an open, descriptive study. Depress Anxiety. 2006;23:485-488.
24. Gutierrez RL, McKercher RM, Galea J, Jamison KL. Lamotrigine augmentation strategy for patients with treatment-resistant depression. CNS Spectr. 2005;10:800-805.
25. Barbee JG, Jamhour NJ. Lamotrigine as an augmentation agent in treatment-resistant depression. J Clin Psychiatry. 2002;63:737-741.
26. Barbosa L, Berk M, Vorster M. A double-blind, randomized, placebo-controlled trial of augmentation with lamotrigine or placebo in patients concomitantly treated with fluoxetine for resistant major depressive episodes. J Clin Psychiatry. 2003;64:403-407.
27. Seidman SN, Rabkin JG. Testosterone replacement therapy for hypogonadal men with SSRI-refractory depression. J Affect Disord. 1998;48:157-161.
28. Pope HG, Cohane GH, Kanayama G, et al. Testosterone gel supplementation for men with refractory depression: a randomized, placebo-controlled trial. Am J Psychiatry. 2003;160:105-111.
29. Orengo CA, Fullerton L, Kunik ME. Safety and efficacy of testosterone gel 1% augmentation in depressed men with partial response to antidepressant therapy. J Geriatr Psychiatry Neurol. 2005;18:20-24.
30. Seidman SN, Miyazaki M, Roose SP. Intramuscular testosterone supplementation to selective serotonin reuptake inhibitor in treatment-resistant depressed men: randomized placebo-controlled clinical trial. J Clin Psychopharmacol. 2005;25:584-588.
31. Schneider LS, Small GW, Hamilton SH, et al. Estrogen replacement and response to fluoxetine in a multicenter geriatric depression trial. Fluoxetine Collaborative Study Group. Am J Geriatr Psychiatry. 1997;5:97-106.
32. Schneider LS, Small GW, Clary CM. Estrogen replacement therapy and antidepressant response to sertraline in older depressed women. Am J Geriatr Psychiatry. 2001;9:393-399.
33. Morgan ML,Cook IA, Rapkin AJ, Leuchter AF. Estrogen augmentation of antidepressant in perimenopausal depression: a pilot study. J Clin Psychiatry. 2005;66:774-780.
34. Rasgon NL, Altshuler LL, Fairbanks LA, et al. Estrogen replacement therapy in the treatment of major depressive disorder in perimenopausal women. J Clin Psychiatry. 2002;63(suppl 7):45-48.
35. Dias RS, Kerr-Correa F, Moreno RA, et al. Efficacy of hormone therapy with and without methyltestosterone augmentation ov venlafaxine in the treatment of postmenopausal depression: a double-blind controlled pilot study. Menopause. 2006;13:202-211.
36. Papakostas GI, Petersen T, Mischoulon D, et al. Serum folate, vitamin B12, and homocysteine in major depressive disorder, Part 1: predictors of clinical response in fluoxetine-reistant depression. J Clin Psychiatry. 2004;65:1090-1095.
37. Coppen A, Bailey J. Enhancement of the antidepressant action of fluoxetine by folic acid: a randomised, placebo controlled trial. J Affect Disord. 2000;60(2):121-130.


Examining the Evidence on Augmentation with Atypical Antipsychotics

By George I. Papakostas, MD

Dr. Papakostas is assistant professor of psychiatry at Harvard Medical School, and staff psychiatrist in the Department of Psychiatry at Massachusetts General Hospital in Boston.

Disclosures: Dr. Papakostas has served as a consultant to Aphios, Bristol-Myers Squibb, GlaxoSmithKline, Evotec, Inflabloc, Jazz, PAMLAB, Pfizer, and Wyeth; has received honoraria from Bristol-Myers Squibb, Evotec, GlaxoSmithKline, Inflabloc, Jazz, Lundbeck, PAMLAB, Pfizer, Titan, and Wyeth; and has received research support from Bristol-Myers Squibb, PAMLAB, and Pfizer.


Introduction: Efficacy of Current Treatment Strategies for Depression   

There is mounting evidence to suggest that the efficacy of all available antidepressants when used as monotherapy to treat major depressive disorder (MDD) is, at best, modest. For example, a meta-analysis1 of all double-blind placebo-controlled studies of antidepressants published since 1980 revealed response rates of 53% for antidepressants versus 36% for placebo (difference in response rate of 16.8%) (Slide 1). To make matters worse, if one is to assume that “negative trials” (ie, trials which do not demonstrate the superiority of a drug over placebo) are less likely to be published than “positive trials” (trials which demonstrate the superiority of a drug versus placebo), it is quite possible that the margin of efficacy of antidepressants when compared to placebo is <16.8%. Thus, if one were to include all unpublished along with published double-blind, placebo-controlled trials of antidepressants for MDD, this efficacy margin could be <10%.


At the present time, it is unclear to what extent conclusions drawn from randomized, double-blind, placebo-controlled trials (ie, efficacy studies) also apply to “real-world” treatment settings. However, preliminary studies focusing on the use of antidepressants in “real-world” clinical settings present discouraging results. For example, Petersen and colleagues2 report remission rates as low as 20% to 23% following each successive treatment among patients with MDD enrolled in one of two academically-affiliated, depression-specialty clinic (Slide 2). In fact, only ~50% of patients enrolled achieved full remission of their depression. Similarly, only about one in three patients with MDD experienced a remission of their depression following treatment with the selective serotonin reuptake inhibitor (SSRI) citalopram during the first-level of the large, multicenter, Sequenced Treatment Alternatives to Relieve Depression (STAR*D) trial.3 Clearly, there is an urgent need to develop safer, better tolerated, and more effective treatments for MDD.



Augmentation of Antidepressants with Second-Generation Antipsychotics: A Novel Approach

There are three major “paths” toward the development of novel pharmacotherapeutic strategies for MDD (Slide 3). The first approach involves developing new drugs. The second approach involves identifying subpopulations of depressed patients that are more likely to experience the benefits of a given treatment versus placebo. Attempts have been made to identify such “subpopulations” with the use of either biological markers (ie, genetic markers), or clinical markers (ie, the presence of symptoms, symptoms clusters, or comorbid disorders). Finally, a third approach involves combining two pharmacologic treatments: either two antidepressants or an antidepressant with a novel agent. The case of augmenting antidepressants with second-generation (atypical) antipsychotics (SGAs) for treatment-resistant depression (TRD) falls under the third “path” of treatment development. Other examples which also fall under this category include the combination of SSRIs or serotonin norepinephrine reuptake inhibitors (SNRI) with other antidepressants including mirtazapine (a 5-HT2 receptor antagonist and α2-adrenergic receptor antagonist), bupropion (a norepinephrine dopamine reuptake inhibitor), tricyclic antidepressants, or with non-antidepressant agents including lithium, triiodothyronine (T3), pindolol, or buspirone.4


The preclinical rationale for the use of the SGAs in MDD derives from their complex neuropharmacologic effects at various monoaminergic receptors and transporters.4 Specifically, all SGAs appear to be 5-HT2 receptor antagonists.4 SGAs also vary in terms of neuropharmacologic effects. Ziprasidone and aripiprazole possess affinity for the 5-HT1A receptor, while ziprasidone and risperidone possess affinity for the 5-HT1D receptor.4 Ziprasidone has also been shown to inhibit the reuptake of serotonin and norepinephrine, while aripiprazole has been shown to possess mixed agonist and antagonist effects at various dopamine receptors.4 These effects were thought to be suggestive of antidepressant activity.4

Early clinical studies focusing on augmentation with SGAs in depression demonstrated mixed findings. The first clinical report ever to be published focusing on this treatment strategy was a case series describing eight patients with SSRI-resistant depression who experienced remission of symptoms following the addition of low doses of risperidone (0.5–1 mg).5 What was also notable in that report was that all patients achieved remission of symptoms of depression quite rapidly (within 1 week of combined treatment). This preliminary report was soon followed by a double-blind placebo-controlled study that focused on adding olanzapine to fluoxetine among fluoxetine nonresponders.6 A greater improvement in depressive symptoms was reported among patients treated with the combination of these two agents than either agent alone.6 

However, soon thereafter, doubt was cast on the potential utility of this treatment strategy when two subsequent studies combining olanzapine with fluoxetine for nortriptyline- or venlafaxine-resistant MDD failed to show that combining olanzapine with fluoxetine was more effective than monotherapy with either venlafaxine, fluoxetine, or nortriptyline.7,8 More recently, however, several double-blind, placebo-controlled studies focusing on augmenting antidepressants with either risperidone,9,10 quetiapine,11-13 or olanzapine14 for TRD re-kindled clinicians interest in this treatment strategy. Specifically, six of these seven trials9,10,12-14 demonstrated greater efficacy in TRD among patients treated with adjunctive SGAs than placebo (Thase and colleagues14 reported two separate but identical trials of olanzapine augmentation of fluoxetine, and found olanzapine augmentation to be effective in the first but not the second study). 

In order to reconcile the discrepancy in results between “positive” and “negative” studies, we conducted a random-effects model meta-analysis pooling all 10 randomized, double-blind, placebo-controlled clinical trials focusing on augmentation of antidepressants with SGAs for TRD.15 In that meta-analysis, the difference in remission rates between the SGAs and placebo was found to be statistically significant, with a 47.4% remission rate for augmentation with SGAs versus a 37% remission rate for augmentation with placebo (Slide 4). Though the difference in efficacy in favor of this augmentation strategy over placebo was pronounced, tolerability appeared to be a considerable limitation. Specifically, the difference in the rates of discontinuation due to intolerance for patients treated with SGAs compared to placebo was also statistically significant (22.3% for the SGAs and 12% for placebo, respectively).




Second-Generation Antipsychotic Augmentation for MDD: Questions Unanswered

Although the results of our meta-analysis provided evidence in support of the use of SGAs as adjuncts for TRD, two major questions remain: (A) is this evidence generalizeable to all SGAs; and (B) is there evidence supporting the long-term efficacy for this treatment strategy as there is for standard antidepressants?

Regarding the issue of generalizability, it is important to note that at the time the meta-analysis was conducted, evidence supporting the use of adjunctive ziprasidone in MDD derived exclusively from open-label, but not placebo-controlled trials. In one such study, Papakostas and colleagues16 treated 20 patients with SSRI-resistant MDD with adjunctive ziprasidone. Twenty-five percent of patients remitted during the trial and 50% experienced a clinical response (Slide 5).


However, more recently, two positive, double-blind, placebo-controlled trials investigating the use of adjunctive aripiprazole in MDD were either been published17 or presented at major scientific meetings (Slide 6).18 In the first study, Berman and colleagues17 focused on the use of aripiprazole augmentation for patients resistant to up to 1–3 retrospective antidepressant trials. To confirm treatment resistance, those patients underwent an 8-week, open-label trial with an SSRI (fluoxetine, sertraline, paroxetine or escitalopram) or an SNRI (venlafaxine). The patients who made insufficient symptom improvement had either aripiprazole or placebo added to their SSRI or SNRI regimen, under double-blind conditions for 6 weeks. A statistically significant difference in remission rates was observed, with 26% remission for aripiprazole versus 15.7% remission for placebo (P<.05). This study also reported relatively low rates of discontinuation due to intolerance in the treatment groups (2% for aripiprazole and 1.7% for placebo [P>.05]). The results of a separate study of identical design also demonstrated greater remission rates for adjunctive aripiprazole- than placebo-treated patients.18


Much less is know regarding long-term efficacy.  Rapaport and colleagues19 examined 386 MDD patients who failed to experience sufficient symptom improvement following treatment with citalopram. These patients then received adjunctive treatment (open-label) with risperidone for 4–6 weeks. Of these patients, 241 (63%) experienced significant symptom improvement. These patients were randomized under double-blind conditions to continue to receive the combination of risperidone and citalopram or to continue with citalopram but to undergo a substitution of risperidone for placebo (double-blind) for 24 weeks. Relapse rates between the two groups were not statistically significant (Slide 7).





From the evidence available to date, it appears that augmentation of standard, first-line antidepressants with SGAs is effective in some cases of TRD, at least during the acute phase of treatment. However, there are limitations in the tolerability of this combination strategy, while the long-term efficacy, tolerability, and safety of this treatment are not yet understood. Further research is required into how this compares with other augmentation strategies and other strategies for addressing TRD.



1. Papakostas GI, Fava M. Does the probability of receiving placebo influence the likelihood of responding to placebo or clinical trial outcome? A meta-regression of double-blind, randomized clinical trials in MDD. Neuropsychopharmacology. 2006;31(suppl 1):s158.
2. Petersen T, Papakostas GI, Posternak MA, et al. Empirical testing of two models for staging antidepressant treatment resistance. J Clin Psychopharmacol. 2005;25(4):336-341.
3. Trivedi MH, Rush AJ, Wisniewski SR, et al; STAR*D Study Team. Evaluation of outcomes with citalopram for depression using measurement-based care in STAR*D: implications for clinical practice. Am J Psychiatry. 2006;163(1):28-40.
4. Papakostas GI. Augmentation of standard antidepressants with atypical antipsychotic agents for treatment-resistant major depressive disorder. Essent Psychopharmacol. 2005;6(4):209-220.
5. Ostroff RB, Nelson JC. Risperidone augmentation of selective serotonin reuptake inhibitors in major depression. J Clin Psychiatry. 1999;60(4):256-259.
6. Shelton RC, Tollefson GD, Tohen M, et al. A novel augmentation strategy for treating resistant major depression. Am J Psychiatry. 2001;158(1):131-134.
7. Shelton RC, Williamson DJ, Corya SA, et al. Olanzapine/fluoxetine combination for treatment-resistant depression: a controlled study of SSRI and nortriptyline resistance. J Clin Psychiatry. 2005;66(10):1289-1297.
8. Corya SA, Williamson D, Sanger TM, Briggs SD, Case M, Tollefson G. A randomized, double-blind comparison of olanzapine/fluoxetine combination, olanzapine, fluoxetine, and venlafaxine in treatment-resistant depression. Depress Anxiety. 2006;23(6):364-372.
9. Keitner GI, Garlow SJ, Ryan CE, et al. Risperidone augmentation for patients with difficult-to-treat major depression. Poster presented at: 159th Annual Meeting of the American Psychiatric Association; May 20-26, 2006; Toronto, Canada.
10. Gharabawi G, Canuso C, Pandina G, et al. A double-blind, placebo-controlled trial of adjunctive risperidone for treatment-resistant major depressive disorder. Poster presented at: 25th Collegium Internationale Neuropsychopharmacologicum; July 9-13, 2006; Chicago, Illinois.
11. Khullar A, Chokka P, Fullerton D, McKenna S, Blackman A. Quetiapine as treatment of non-psychotic unipolar depression with residual symptoms: double blind, randomized, placebo controlled study. Poster presented at: 159th Annual Meeting of the American Psychiatric Association; May 20-26, 2006; Toronto, Canada.
12. Mattingly G, Ilivicky H, Canale J, Anderson R. Quetiapine augmentation for treatment-resistant depression. Poster presented at: 159th Annual Meeting of the American Psychiatric Association; May 20-26, 2006; Toronto, Canada.
13. McIntyre AW, Gendron A, Mcintyre A. Quetiapine augmentation of SSRIs/SNRIs in major depression with anxiety. Poster presented at: 159th Annual Meeting of the American Psychiatric Association; May 20-26, 2006; Toronto, Canada.
14. Thase ME, Corya SA, Osuntokun O, et al. A randomized, double-blind comparison of olanzapine/fluoxetine combination, olanzapine, and fluoxetine in treatment-resistant major depressive disorder. J Clin Psychiatry. 2007;68(2):224-236.
15. Papakostas GI, Shelton RC, Smith J, Fava M. Augmentation of antidepressants with atypical antipsychotic medications for treatment-resistant major depressive disorder: a meta-analysis. J Clin Psychiatry. 2007;68(6):826-831.
16. Papakostas GI, Petersen TJ, Nierenberg AA, et al. Ziprasidone augmentation of selective serotonin reuptake inhibitors (SSRIs) for SSRI-resistant major depressive disorder. J Clin Psychiatry. 2004;65(2):217-221.
17. Berman RM, Marcus RN, Swanink R, et al. The efficacy and safety of aripiprazole as adjunctive therapy in major depressive disorder: a multicenter, randomized, double-blind, placebo-controlled study. J Clin Psychiatry. 2007;68(6):843-583.
18. Thase ME, Marcus RN, Hennicken D, et al. Efficacy and safety of aripiprazole as adjunctive therapy in major depressive disorder: a multicenter, randomized, double-blind, placebo-controlled study (Study CN 138-163). Poster presented at: 61st Annual Convention of the Society of Biological Psychiatry; May 18-20, 2006; San Diego, California.
19. Rapaport MH, Gharabawi GM, Canuso CM, et al. Effects of risperidone augmentation in patients with treatment-resistant depression: Results of open-label treatment followed by double-blind continuation. Neuropsychopharmacology. 2006;31(11):2505-2513.


Clinical Considerations with Atypical Antipsychotic Augmentation

By Michael J. Gitlin, MD

Dr. Gitlin is professor of clinical psychiatry at the University of California, Los Angeles (UCLA) School of Medicine and director of the Mood Disorders Clinic at the Neuropsychiatric Clinic at UCLA.

Disclosures: Dr. Gitlin has received honoraria from AstraZeneca, Bristol-Myers Squibb, Cephalon, Eli Lilly, GlaxoSmithKline, Pfizer, and Takeda.


Introduction: Practical Clinical Applications

In spite of the tremendous advances made in developing antidepressant treatments and exploring augmentation with second-generation antipsychotics (SGAs), significant obstacles remain for psychiatrists: How should clinicians make use of cutting-edge augmentation studies? How should they use the antipsychotics in their treatment algorithm? How should they think about dosing and side effects?

The first major problem is there have been no great algorithmic studies on adjunctive strategies. In fact, the most useful algorithmic study was the Sequenced Treatment Alternatives to Relieve Depression (STAR*D) study, from which SGAs were entirely absent. The second problem is that STAR*D showed a clear pattern in which the efficacy of an augmentation agent depended on its placement in the algorithm. Those agents occurring later in the algorithm, such as monoamine oxidase inhibitors (MAOIs), adjunctive triiodothyronine (T3), and adjunctive lithium, appeared to be less effective than adjunctive buspirone, which occurred earlier in the algorithm. This was not the result of a direct comparison. Because of these two problems, it is therefore difficult to determine how to rank the efficacy of SGAs among the agents investigated by the STAR*D study. Among clinical treatment strategies, SGAs might appear toward the end of first-line strategies, perhaps after combinations and stimulants but before buspirone.

As a field, we also know very little about optimal dosing of SGAs when used as adjunctive treatments for depression. In general, however, doses used in adjunctive strategies are lower than those used for either acute mania or acute psychotic states. These are usually dosed in the range of 2.5–10 mg for aripiprazole, 20–60 mg for ziprasidone, 2.5–10 mg for olanzapine, 0.5–3 mg for risperidone, and 50–200 mg for quetiapine. However, these doses are based more on clinical experience than evidence-based data.


Side-Effect Concerns

Considering what is currently known about augmentation with SGAs, clinicians must consider how to make decisions within a relatively limited base of knowledge. Another factor to consider in making these decisions are side effects. There are class concerns, such as tardive dyskinesia (TD), and concerns that differ across agents, such as weight gain and propensity for metabolic syndrome. Many of these side-effect concerns have caused hesitation among clinicians regarding the use of SGAs earlier in the algorithm of treatment-resistant depression (TRD).

It is clear that SGAs are less associated with TD than first-generation antipsychotics (FGAs). In the best meta-analysis conducted on the subject thus far, the occurence of TD with SGAs was one seventh its occurence with FGAs.1 However, this study was conducted several years ago, and therefore did not include more recent medications such as ziprasidone and aripiprazole. In addition, TD studies are conducted with schizophrenia patients who have received full doses of antipsychotics. It is not known how relevant those data are to low-dose strategies in mood disorders patients. While it seems likely that TD risk is reduced with low-dose strategies, it has not yet been proven. It is also unknown whether rates of TD vary across SGAs. In theory, quetiapine, with its fast dopamine (D)2 dissociation; aripiprazole, with its D2 partial agonism; and clozapine should have the lowest risks of TD. However, there is not enough data to show that these agents are less likely to be associated with TD than any other SGA.

Prolactin is another concern for clinicians. It is clear that risperidone has a unique propensity to increase prolactin levels, whereas other SGAs increase prolactin rarely, if at all. Clinicians prescribing SGAs should look for side effects such as menstrual abnormalities in women; galactorrhea and gynecomastia in both men and women; and certainly sexual dysfunction in men.

Risperidone is far more likely to cause extrapyramidal symptoms (EPS) than the other SGAs, for all symptoms except agitation. Quetiapine seldom causes EPS, and clozapine is not associated with EPS at all. Ziprasidone and aripiprazole are most linked to agitation. A restlessness syndrome, often diagnosed as agitation or akathisia, occurs in ~10% of people taking ziprasidone or aripiprazole.


Metabolic Syndrome and Weight Gain

Even though medications that increase weight have existed since the beginning of modern psychopharmacology, starting with chlorpromazine and amitriptyline, it has become a greater concern in recent years. This is especially true with SGAs. The definition of metabolic syndrome is polythetic; it must meet three of five criteria (Slide 1).2 No one criterion is weighted more than another. Other risk factors for cardiovascular disease, such as cigarette smoking, are not typically included in the definition of metabolic syndrome but are considered additive risk factors. Obesity is clearly a strong mediating variable for a number of these factors, but it is not the only one. Some patients have suddenly and dramatically developed diabetic ketoacidosis early in the course of treatment with an SGA (specifically clozapine, olanzapine, or quetiapine) without gaining any weight at all. Treatment with SGAs causes both obesity-mediated metabolic syndrome and non-obesity–related concerns.



Metabolic syndrome in psychiatrically ill patients mirrors epidemiologic trends in society at large, where rates of type 2 diabetes among children and adults have increased markedly over the past 20 years. It is widely presumed that this is related to dietary factors, although a concrete cause has not yet been established. Clinicians must also remember that chronically psychiatrically ill patients, whether afflicted with schizophrenia, bipolar disorder, or major depressive disorder, tend to have very unhealthy habits. These habits may contribute to, or even cause, patients’ metabolic disturbances. It is also significant to note that abdominal obesity, not simple weight gain, is the relevant criterion. Truncal weight gain correlates with cardiovascular risk more than fat deposits elsewhere, making this specific type of obesity of greater concern.

A study by Allison and colleagues3 showed differential weight gain across antipsychotics (Slide 2). Aripiprazole was not included in the study because data were not available at the time, though its weight gain effects might be similar to those of ziprasidone and fluphenazine. It is clear, in any case, that weight gain is not uniform across agents.



The Consensus Development Conference on Antipsychotic Drugs and Obesity and Diabetes investigated the differential concern for weight gain, risk for diabetes, and worsening lipid profile.4 These can be divided into three groups: the clozapine-olanzapine group, which clearly has the highest propensity toward metabolic syndrome; the risperidone-quetiapine group, which has a medium propensity; and the ziprasidone-aripiprazole group, which has the lowest propensity (Slide 3).



Analyses of the Clinical Antipsychotic Trials of Intervention Effectiveness (CATIE) study also demonstrated that olanzapine has the greatest weight gain concern (Slides 4 and 5) and is associated with the highest glucose levels (Slide 6).5 Clinicians might be disinclined to use agents with greater weight gain propensity and metabolic syndrome concerns. However, they must bear in mind that some patients do much better on olanzapine, and some patients experience the most symptom improvement with clozapine.




Recommendations for Monitoring

Clinicians must be vigilant about monitoring patients taking antipsychotics as adjunctive antidepressants. The Consensus Development Conference recommended monitoring for a number of risk factors, such as family history, waist circumference, fasting plasma glucose, and fasting lipid glucose (Slide 7).4 Unfortunately, many clinicians do not rigorously follow these guidelines. Clinicians in institutional settings, with ample assistance from nurses and ancillary personnel, may have less difficulty drawing blood and following these extensive guidelines. However, clinicians in smaller practices face obstacles to carrying out the recommendations, both because they often lack the infrastructure to obtain frequent lab tests and because because they simply do not have a tradition of doing them. The formal recommendations are lofty but burdensome, and because the majority of clinicians in the community may not follow them, those doctors are left without practicable guidelines.


Should clinicians receive a second set of guidelines? It seems that there are practical measures clinicians can take to monitor their patients without undue burden. Though they will miss some rare diabetic ketoacidoses without consistently monitoring weight gain or other laboratory parameters, clinicians can safely assume that patients who have experienced significant weight gain—≥7% of baseline body weight, or ≥10 pounds for a 150-pound person—belong to a higher-risk group. When using SGAs as adjunctive antidepressants, identifying high-risk groups might best involve monitoring weight gain and testing glucose, lipid profiles, and blood pressure for those patients who have gained ≥7% of their baseline body weight.

Clinicians should be reminded that using lower doses of the SGAs does not mean they can be dismissive about weight gain. While there are some data indicating that weight gain is dose-related, at least for olanzapine and quetiapine, there has been no conclusive evidence. However, adjunctive therapy with an SGA does utilize relatively low doses of SGAs, which may give clinicians some peace of mind. This is a topic that clinicians will have to address more in the future as SGAs are increasingly used as adjunctive antidepressants.



In summary, the use of SGAs as adjunctive antidepressants has increased substantially. At this point, the data has not kept up with the clinical interest but we can anticipate a number of controlled studies in this area emerging over the next few years. With the use of these medications for depression comes the obligation to carefully consider side effect issues in choosing specific agents.



1. Correll Cu, Leucht S, Kane JM. Lower risk for tardive dyskinesia associated with second-generation antipsychotics: a systematic review of 1-year studies. Am J Psychiatry. 2004;161(3):414-425.
2. Ford ES, Giles WH, Dietz WH. Prevalence of the metabolic syndrome among US adults: findings from the third National Health and Nutrition Examination Survey. JAMA. 2002;287(3):356-359
3. Allison DB, Mentore JL, Heo M, et al. Antipsychotic-induced weight gain: a comprehensive research synthesis. Am J Psychiatry. 1999;156(11):1686-1696.
4. American Diabetes Association; American Psychiatric Association; American Association of Clinical Endocrinologists; North American Association for the Study of Obesity. Consensus development conference on antipsychotic drugs and obesity and diabetes. Diabetes Care. 2004;27(2):596-601.
5. Lieberman JA, Stroup TS, McEvoy JP, et al. Effectiveness of antipsychotic drugs in patients with chronic schizophrenia. N Engl J Med. 2005;353(12):1209-1223.


Question-and-Answer Session

Q: I have long been interested in using thyroid hormone as adjunctive treatment for depression, but I am consistently disappointed with the results in my patients. What has your experience been?

Dr. Thase: In my clinical experience, thyroid augmentation is the treatment of choice for a depressed patient with an underfunctioning thyroid gland, but it is not high on my list for anyone else with refractory depression. Simply said, I believe this treatment works by enhancing thyroid function and patients who have high normal thyroid function to begin with do not benefit. In a study of bipolar depression,1 the cutoff for thyroid-stimulating hormone (TSH) was ~2.5 and a below 50 percentile value for the free thyroxin index (FTI). Patients with one or both of these were less likely to respond to antidepressants. Since that time I have since been targeting those patients with a slightly higher-than-average TSH and a slightly lower than average FTI for thyroid augmentation.

Dr. Nelson: Many studies have looked at augmentation very early in the course of treatment, at ~4 weeks. This may overestimate the effect of augmentation since patients may still be in the midst of responding to initial treatment. Not many studies have carefully selected patients who are truly refractory to initial treatment. A study by Thase and colleagues2 was one of the only studies that incorporated a 12-week period of treatment with imipramine and psychotherapy. Augmentation with thyroid did not seem to add efficacy. In my own experience, I have been similarly disappointed with thyroid augmentation for people who are really treatment resistant.


Q: Have the studies focusing on augmenting tricyclic antidepressants (TCAs)  with first-generation (conventional) antipsychotics (FGAs) for major depressive disorder largely evaluated them as first-line treatments?

Dr. Thase: The original set of studies looked at FGAs as first-line treatments. In that era, FGAs were often used as augmentation agents because psychiatrists did not want to use benzodiazepines. Lower doses of FGAs showed good anxiolytic effects as long as they did not induce akathisia. The evidence of FGA efficacy in depression primarily comes from studies of their first-line use, not their use as augmentation strategies.

Dr. Nelson: There are several positive studies showing that FGAs have an advantage over placebo in the treatment of depression, especially with severely ill patients. The reason they are not considered “antidepressants,” however, is because they were less effective than imipramine for treating loss of interest or motor retardation—symptoms the field considers to be central to the concept of depression.3 However, antipsychotics are useful for treating many other symptoms of depression such as feelings of guilt, anxiety-tension, sleep disturbance, and of course, depressive delusions.3 Hollister and Overall4 argued that some depressed patients have a complete response to antipsychotics alone.


Q: Consider a hypothetical patient with moderately severe nonpsychotic depression, with no clear family history or worrisome historical factors suggestive of bipolar disorder, who has not responded to a second-line antidepressant. The clinician has tried one monotherapy after another, and the patient has experienced a 25% to 30% symptom reduction. The clinician now has decided to augment rather than go to a third favorite monotherapy. Given the state of the evidence, which strategy would you pick?   

Dr. Nelson: That is a difficult question at this point. Should one pick an augmentation strategy or a combination treatment strategy? If the patient remained generally depressed, I would use an antidepressant combination. I would use augmentation agents for specific prominent residual symptoms such as lack of energy and fatigue, or insomnia, or anxiety and agitation and select the agent most likely to help with those symptoms.   
Dr. Gitlin: There is a reasonable amount of positive data for lithium augmentation, most of it with tricyclic antidepressants (TCAs), but clinician and patient acceptance of lithium augmentation is very poor. There are also stimulants for which we do not have any supportive double-blind evidence, yet stimulant augmentation is the single most common augmenting technique clinicians use. It is quite difficult to make research-informed decisions when the findings do not translate into practices acceptable to clinicians and patients.    

Dr. Nelson: It would be nice to have a study similar to the Sequenced Treatment Alternatives to Relieve Depression (STAR*D) study, with a more extensive comparison of common augmentation and combination strategies, as well as a placebo arm. This kind of study would allow psychiatrists to compare the efficacy of these strategies. 


Q: The findings of the meta-analysis by Crossley and Baur5 favor lithium augmentation. However, the authors do not highlight the discrepancies in study durations. Many of these durations are ≤2 weeks. Clinical effects of lithium appear very early, but are not very apparent later. Might this meta-analysis indicate an early acceleration effect which might not appear as robust at week 2?  

Dr. Nelson: There are two duration issues that are variable and limited. One is the duration of prior treatment, which often was only 4 weeks. The other is the duration of the augmentation phase, which in some studies was only 48 hours. The short duration of these studies was based on de Montigny’s6 report of rapid response. There may indeed have been a rapid effect (acceleration) independent of long-term efficacy. Alternatively, Dr. Thase’s historical comparison7 had a longer duration of treatment (6 weeks) during which patients continued to improve.   
Dr. Thase: It is important to keep in mind the source of the patients in the study. Ours was part of a study of highly recurrent depression, and our patients had an average of seven prior episodes. The clinic was strict in its definition of bipolar disorder, and many of our patients would today be said to fall in the bipolar spectrum. Of the lithium responders, 20% to 30% had relatively fast responses, while the other lithium responders experienced slow, gradual, 4–6-week responses.7 For patients with difficult-to-treat depression, lithium can enhance antidepressant effect. This seems more likely to occur with a TCA than with a selective serotonin reuptake inhibitor (SSRI) because TCAs are primarily noradrenergic medications and lithium is a relatively weak serotonergic enhancer.

Q: Is there evidence of an advantage for the use of second-generation (atypical) antipsychotics (SGAs) as an adjunct in more severely depressed patients? It is my impression that severely ill patients are usually more willing to tolerate side effects.

Dr. Papakostas: Unfortunately, in our meta-analysis8 we did not have data that would allow us to look at whether the treatment difference is more robust in patients with severe depression versus those with mild depression. It would be interesting as a post-hoc analysis. However, in my clinical practice, when a patient has very severe depression with suicidal ideation or with irritability, using SGAs certainly comes to mind sooner than it would for a patient who is nearly remitted, whose initial treatment had a low side-effect burden.
Dr. Gitlin: I agree. We have difficulty telling patients with significant residual symptoms, but without severe depression, that we want to put them on antipsychotics. That can be quite jarring for patients. In the less severe patients who presumably have less agitation, we must be careful about using more sedating agents, such as quetiapine and olanzapine. A moderately residually depressed patient might prefer to try aripiprazole or ziprasidone than more sedating and weight-gaining agents.    
Dr. Nelson: It is still quite difficult to determine which augmentation agents will work best with which patients. The data on predictors of response to augmentation are very limited. These decisions are currently based on clinical experience.  It does seem that depression severity interacts with ability to tolerate side effects. Early studies of lithium augmentation were primarily done in more severely ill inpatients who were started on lithium 900 mg/day (300 mg TID). In these hospitalized patients, lithium was relatively well tolerated. In the STAR*D sample, however, which consisted of outpatients who were less ill, lithium was considered quite problematic in terms of patient tolerability.   



1. Cole DP, Thase ME, Mallinger AG, et al. Slower treatment response in bipolar depression predicted by lower pretreatment thyroid function. Am J Psychiatry. 2002;159(1):116-121.
2. Thase ME, Kupfer DJ, Jarrett DB. Treatment of imipramine-resistant recurrent depression: I. An open clinical trial of adjunctive L-triiodothyronine. J Clin Psychiatry. 1989;50(10):385-388. 
3. Raskin A, Schulterbrandt JG, Reatig N, McKeon. Differential response to chlorpromazine, imipramine, and placebo. Arch Gen Psychiatry. 1970;23:164-173.
4. Hollister LE, Overall JE. Reflections on the specificity of action of anti-depressants. Psychosomatics. 1965;6(5):361-365.
5. Crossley NA, Bauer M. Acceleration and augmentation of antidepressants with lithium for depressive disorders: two meta-analyses of randomized, placebo-controlled trials. J Clin Psychiatry. 2007;68(6):935-940.
6. Dé Montigny C, Grunberg F, Mayer A, Deschenes JP. Lithium induces rapid relief of depression in tricyclic antidepressant drug non-responders. Br J Psychiatry. 1981;138:252-256.
7. Thase ME, Kupfer DJ, Frank E, Jarrett DB. Treatment of imipramine-resistant recurrent depression: II. An open clinical trial of lithium augmentation. J Clin Psychiatry. 1989;50(11):413-417.
8. Papakostas GI, Shelton RC, Smith J, Fava M. Augmentation of antidepressants with atypical antipsychotic medications for treatment-resistant major depressive disorder: a meta-analysis. J Clin Psychiatry. 2007;68(6):826-831.


Dr. Ginsberg is vice-chair of clinical affairs in the Department of Psychiatry at New York University Medical Center in New York City.


Disclosure: Dr. Ginsberg receives honoraria for lectures, papers, and/or teaching from AstraZeneca and GlaxoSmithKline; and receives research support from Cyberonics.



Lithium-Induced Brugada Syndrome

The July 2007 “Psychopharmacology Reviews”1 discussed a report of second-degree, type 2 sinoatrial block that resulted from toxic lithium levels.2 Other possible electrocardiogram (ECG) manifestations of lithium toxicity include prolonged QT interval, T-wave flattening and inversion, first-degree atrioventricular conduction delay, and, rarely, ventricular tachycardia and ventricular fibrillation resulting in death.3-6

Even at therapeutic levels, there are other potential cardiac consequences of chronic lithium therapy, such as sinus node suppression.7-9 The mechanism of action underlying this effect may be related to lithium’s competitive inhibition of calcium in the sodium-calcium exchange in cardiac cells.10

In an ion channel disease known as Brugada syndrome, transmural dispersion of repolarization is accentuated as a result of preferential abbreviation of the right ventricular epicardial action potential. Brugada syndrome is characterized by syncopes and sudden cardiac death due to polymorphic ventricular tachyarrhythmia. In approximately 20% of cases, it is caused by mutations in the SCN5A gene on chromosome 3p21-23, encoding the a-subunit of the cardiac sodium channel.11

In a healthy population, the point prevalence of the Brugada syndrome ECG pattern has been estimated at 1–5 per 10,000 individuals wordwide.12 Such a pattern has a QRS morphology similar to right bundle branch block, with ST-segment elevation in leads V1-V3. In addition, some Brugada syndrome patients display QT prolongation, though when it does occur it is usually limited to the right precordial leads V1-V3. The following is a published report of a patient who developed Brugada syndrome in association with use of lithium.13

A 42-year-old male with schizoaffective disorder experienced a syncopal episode. He had been receiving lithium for 8 years and had a history of recurrent syncopes during this period. His ECG showed typical signs of Brugada syndrome, with ST-segment elevation in the precordial leads V1-V3. Serum lithium level was within therapeutic range (0.75 mmoL/L), but lithium was discontinued after the diagnosis of Brugada syndrome.

Three weeks after lithium withdrawal, the patient developed a manic episode and was admitted to a hospital in Germany. Physical examination was normal. Family medical and psychiatric histories were normal. Treatment with olanzapine 30 mg/day resulted in complete symptom remission after 4 weeks. ECG showed resolution of ST abnormalities. Programmed electrical stimulation from the right ventricular outflow tract induced only nonsustained ventricular tachycardia. However, a provocation test with ajmaline demonstrated ST-segment elevations typical for Brugada syndrome. Subsequently, an automatic defibrillator was implanted to prevent future syncopes and sudden cardiac death.

The case described here is consistent with Brugada syndrome precipitated by lithium, which has been previously reported.14 It is also known that sodium channel blockers, such as carbamazepine, oxcarbazepine, valproate, lamotrigine, phenothiazines, clozapine, selective serotonin reuptake inhibitors, and tricyclic antidepressants may also increase the risk of developing symptomatic Brugada syndrome.11 Whether the risk increases when lithium is combined with any of these other agents has not been reported but, given the number of patients receiving such combinations, deserves further study. PP



1. Ginsberg DL. Lithium-induced sinoatrial block. Primary Psychiatry. 2007;14(7):37-38.
2. Goldberger ZD. Sinoatrial block in lithium toxicity. Am J Psychiatry. 2007;164(5):831-832.
3. Brady HR, Horgan JH. Lithium and the heart: unanswered questions. Chest. 1988;93(1):166-169.
4. Mitchell JE, Mackenzie TB. Cardiac effects of lithium therapy in man: a review. J Clin Psychiatry. 1982;43(2):47-51.
5. Tilkian AG, Schroeder JS, Kao JJ, Hultgren HN. The cardiovascular effects of lithium in man. A review of the literature. Am J Med. 1976;61(5):665-670.
6. Montalescot G, Levy Y, Farge D, et al. Lithium causing a serious sinus-node dysfunction at therapeutic doses. Clin Cardiol. 1984;7(11):617-620.
7. Wellens HJ, Cats VM, Duren DR. Symptomatic sinus node abnormalities following lithium carbonate therapy. Am J Med. 1975;59(2):285-287.
8. Wilson JR, Kraus ES, Bailas MM, Rakita L. Reversible sinus-node abnormalities due to lithium carbonate therapy. N Engl J Med. 1976;294(22):1223-1224.
9. Terao T, Abe H, Abe K. Irreversible sinus node dysfunction induced by resumption of lithium therapy. Acta Psychiatr Scand. 1996;93(5):407-408.
10. Lai CL, Chen WJ, Huang CH, et al. Sinus node dysfunction in a patient with lithium intoxication. J Formos Med Assoc. 2000;99(1):66-68.
11. Brugada R, Brugada J, Antzelevitch C, et al. Sodium channel blockers identify risk for sudden death in patients with ST-segment elevation and right bundle branch block but structurally normal hearts. Circulation. 2000;101(5):510-515.
12. Hermida JS, Lemoine JL, Aoun FB, Jarry G, Rey JL, Quiret JC. Prevalence of the brugada syndrome in an apparently healthy population. Am J Cardiol. 2000;86(1):91-94.
13. Laske C, Soekadar SR, Laszlo R, Plewnia C. Brugada syndrome in a patient treated with lithium. Am J Psychiatry. 2007;164(9):1440-1441.
14. Darbar D, Yang T, Churchwell K, Wilde AA, Roden DM. Unmasking of brugada syndrome by lithium. Circulation. 2005;112(11):1527-1531.


Neonatal Brain Infarcts Possibly Due to Electroconvulsive Therapy During Pregnancy

Electroconvulsive therapy (ECT) has been in use since the 1940s and is primarily indicated for the treatment of depression, acute mania, schizophrenia, and catatonia. The American Psychiatric Association (APA)-published text regarding recommendations related to ECT1 states that ECT is a treatment with low risk and high efficacy in the management of specific disorders in all three trimesters of pregnancy. The following is a report of a young woman who received ECT for bipolar depression every 2 weeks throughout her pregnancy and who subsequently delivered an infant with neurologic abnormalities associated with several brain infarcts.2

A 22-year-old primigravida with a long history of bipolar depression received bifrontal ECT every 2 weeks during her entire pregnancy. No complications from the ECT were reported. Prior to becoming pregnant, the patient had received maintenance ECT with good effect. After becoming pregnant, the patient continued to receive ECT to treat ongoing severe depression that in the past had only been partially responsive to medication treatment. Between 20 and 34 weeks of gestational age, seven documented ECT treatments occurred. Fetal heart rates were recorded after each treatment with no reported abnormalities. There is no evidence of a nonstress test having been performed. Further specifics about the maternal ECT were not recorded.

At 36.1 weeks of gestation, just prior to her scheduled ECT, the patient was noted to have an elevated blood pressure of 162/100 mm Hg. Twenty four-hour urinalysis showed mildly elevated protein. All other laboratory values were normal. The patient underwent induction of labor for preeclampsia and vaginally delivered a male infant weighing 2,550 grams (5.61 lbs) with 1- and 5-minute Apgar scores of 4 and 7, without complications. Throughout labor and delivery, the fetal heart rate tracing was unremarkable. Anatomic survey at 20 weeks was normal.

The infant was stimulated with no response, and his upper extremities were tonic with extension posturing. An umbilical cord blood gas analysis showed a pH of 7.24, PCO2 of 42, HCO3 of 18, and PO2 of 27. The infant’s infectious disease workup, toxicity screens, and other blood tests were all negative. Brain computed tomography and magnetic resonance imaging scans performed on the second and third day of life showed a small left cerebellar, bihemispheric deep white matter, and cortical infarcts. The infant was discharged home at age 12 days. A neurology consultant who evaluated the infant concluded that the prognosis was guarded with expected long-term motor control issues. Further follow-up was not available.

As pointed out by the authors of this report, whether ECT in this case led to titanic uterine contractions causing a fetal bradycardia or other type arrhythmia that then resulted in fetal brain injury is a possibility but unproven. There are no prospective or controlled studies of ECT in pregnancy. The APA ECT Practice Guidelines Recommendations1 derive principally from reviews of case reports. The largest such review evaluated 300 reports of ECT in pregnancy from 1942–1991.3 Twenty-eight cases (9.3%) were noted to have complications, including fetal cardiac arrhythmias (five), vaginal bleeding (five), uterine contractions (two), abdominal pain (three), premature labor (four), miscarriage (five), stillbirth and neonatal death (three), neonatal respiratory distress (one), and teratogenicity (five). Of note, 25 of these 28 patients were >30 years of age at the time of publication of the review. There have been an additional six cases in the literature with obstetric outcomes reported.4-7 Included among these are reports of fetal heart rate decelerations in three of six cases, one requiring tocolytic medication, and one case of first trimester bleeding with subsequent miscarriage. Among these reports, there is a good deal of heterogeneity in regard to details of diagnosis, treatments, pregnancy trimester, other potential contributing factors, and outcomes reported.

The APA Practice Guidelines1 for ECT include recommendations for managing a pregnant patient undergoing ECT. These recommendations are designed to minimize potential complications such as aspiration and altered uteroplacental blood flow. Even with their implementation however, serious neonatal complications may result. The exact incidence of such complications, and the attribution of them to ECT, is difficult to determine in the absence of controlled studies. Risk-benefit assessments, both of the mother and the neonate, need to account for the severity of the underlying psychiatric condition and the safety and efficacy not only of ECT, but of alternate medication therapies as well. PP



1. Committee on Electroconvulsive Therapy. The Practice of Electroconvulsive Therapy, Recommendations for Treatment, Training, and Privileging: A Task Force Report of the American Psychiatric Association. 2nd ed. Washington, DC: American Psychiatric Association; 2001.
2. Pinette MG, Santarpio C, Wax JR, Blackstone J. Electroconvulsive therapy in pregnancy. Obstet Gynecol. 2007;110(2 Pt 2):465-466.
3. DeBattista C, Cochran M, Barry JJ, Brock-Utne JG. Fetal heart rate decelerations during ECT-induced seizures: is it important? Acta Anaesthesiol Scand. 2003;47(1):101-103.
4. Bhatia SC, Baldwin SA, Bhatia SK. Electroconvulsive therapy during the third trimester of pregnancy. J ECT. 1999;15(4):270-274.
5. Polster DS, Wisner KL. ECT-induced premature labor: a case report. J Clin Psychiatry. 1999;60(1):53-54.
6. Echevarria Moreno M, Martin Munoz J, Sanchez Valderrabanos J, Vazquez Gutierrez T. Electroconvulsive therapy in the first trimester of pregnancy. J ECT. 1998;14(4):251-254.
7. Livingston JC, Johnstone WM Jr, Hadi HA. Electroconvulsive therapy in a twin pregnancy: a case report. Am J Perinatol. 1994;11(2):116-118.


Gabapentin-Induced Rhabdomyolysis

Rhabdomyolysis is characterized by a 5-fold increase in the upper normal limit of blood creatine kinase (CK) levels, which usually is associated with acute renal failure.1 Accumulation of intracellular calcium, activation of proteases and lipases, production of free radicals, and infiltration by inflammatory cells are some of the mechanisms responsible for muscular necrosis. The diagnosis of rhabdomyolysis relies on the elevation of CK and myoglobin blood levels. The main therapeutic goals focus on removal of precipitating factors, correction of biochemical alterations, and treatment of acute renal failure by hemodialysis when necessary. Conservative interventions, such as early and vigorous hydration associated with forced alkaline diuresis, can improve the prognosis of this adverse event.

Rhabdomyolysis usually occurs following traumatic injuries, excessive physical activity, seizures, infections, and adverse drug reactions.1 Besides statins, recognized as the drugs most commonly associated with severe myopathy, neuroleptics—typically in the setting of neuroleptic malignant syndrome2,3—and proton pump inhibitors4 have been suggested as other possible causative agents of somatic muscular alterations.

Gabapentin is a γ-aminobutyric acid (GABA)ergic anticonvulsant indicated for adjunctive therapy in the treatment of partial seizures with and without secondary generalization, in adults with epilepsy. Despite the lack of placebo-controlled trials affirming its use, in recent years gabapentin has been used to treat numerous other neuropsychiatric conditions including bipolar disorder,5 anxiety disorders,6-9 somatization disorder,10 behavioral dyscontrol,11-13 cocaine and alcohol withdrawal,14,15 antipsychotic-induced movement disorders,16 neuropathic pain,17,18 ciguatera poisoning,19 nicotine20 and benzodiazepine dependence,21 tinnitus,22 and hot flashes in postmenopausal women and in men taking hormone therapy for prostate cancer.23,24 Gabapentin is well absorbed, renally excreted, does not bind to plasma proteins, and has few, if any, drug interactions. The most frequent adverse effects are drowsiness, dizziness, ataxia, and gastrointestinal upset.25 Postmarketing reports of gabapentin-associated sexual dysfunction in both men and women have also appeared,26-31 as well as a report of gabapentin-associated worsening renal function in two patients previously treated with lithium.32 The following is a report of gabapentin-induced rhabdomyolysis.33

An 85-year-old diabetic female was hospitalized for severe pain in her lower limbs and difficulty in walking, compromising her daily activities. On admission, the woman’s laboratory parameters, including CK and myoglobin, were in the normal range. Neurologic evaluation suggested a diagnosis of diabetic neuropathic pain, and therapy with gabapentin 150 mg TID was started. On the same day, the patient developed psychomotor agitation and gastric pain, which were treated with haloperidol 10 mg and lansoprazole 30 mg, respectively. Over the next few hours, the severity of muscular pain increased, with the patient developing myopathy with acute renal failure. Laboratory values were as follows: CK 459 U/L (normal 24–190 U/L), myoglobin 11,437 ng/mL (normal 10–90 ng/mL), and creatinine 4.59 mg/dL (normal 0.30–1.10 mg/dL). During the next 2 days, despite supportive treatment and discontinuation of haloperidol and lansoprazole, the patient’s condition progressively worsened as manifested by the following laboratory indices: CK 3095 U/L, myoglobin 17,000 mg/dL, and creatinine 4.77 mg/dL. No signs of trauma or edema were detected; consequently, compartmental or crush syndrome was excluded as a possibility. Gabapentin was then withdrawn. The patient’s condition rapidly improved, with a reduction in pain; progressive normalization of CK, myoglobin, and creatinine; and overall complete recovery after approximately 10 days.

The temporal sequence of events described above, with worsening of the clinical picture despite discontinuation of haloperidol and lansoprazole but rapid improvement after withdrawal of gabapentin, strongly suggests an association between gabapentin and rhabdomyolysis. On the Naranjo probability scale, the association is rated as probable.34 Gabapentin-associated myopathy has been previously reported in two patients, both of whom had end-stage renal disease.35

The mechanism of action underlying this association is not known, but may involve the action of gabapentin on voltage-gated calcium and sodium channels in muscle cells. Regardless of the pathophysiology, the appearance of the syndrome soon after the initiation of gabapentin at therapeutic doses is compatible with an idiosyncratic adverse event. Given the widespread use of gabapentin, including for off-label uses, clinical caution is advised. PP



1. Singh D, Chander V, Chopra K. Rhabdomyolysis. Methods Find Exp Clin Pharmacol. 2005;27(1):39-48.
2. Yoshikawa H, Watanabe T, Abe T, Oda Y, Ozawa K. Haloperidol-induced rhabdomyolysis without neuroleptic malignant syndrome in a handicapped child. Brain Dev. 2000;22(4):256-258.
3. Marinella MA. Rhabdomyolysis associated with haloperidol without evidence of NMS. Ann Pharmacother. 1997;31(7-8):927-928.
4. Clark DW, Strandell J. Myopathy including polymyositis: a likely class adverse effect of proton pump inhibitors? Eur J Clin Pharmacol. 2006;62(6):473-479.
5. Stanton SP, Keck PE Jr, McElroy SL. Treatment of acute mania with gabapentin. Am J Psychiatry. 1997;154(2):287.
6. Chouinard G, Beauclair L, Belanger MC. Gabapentin: long-term antianxiety and hypnotic effects in psychiatric patients with comorbid anxiety-related disorders. Can J Psychiatry. 1998;43(3):305.
7. Pollack MH, Matthews J, Scott EL. Gabapentin as a potential treatment for anxiety disorders. Am J Psychiatry. 1998;155(7):992-993.
8. Pande AC, Davidson JR, Jefferson JW, et al. Treatment of social phobia with gabapentin: A placebo-controlled study. J Clin Psychopharmacol. 1999;19(4):341-348.
9. Brannon N, Labbate L, Huber M. Gabapentin treatment for posttraumatic stress disorder. Can J Psychiatry. 2000;45(1):84.
10. Garcia-Campayo J, Sanz-Carrillo C. Gabapentin for the treatment of patients with somatization disorder. J Clin Psychiatry. 2001;62(6):474.
11. Ryback R, Ryback L. Gabapentin for behavioral dyscontrol. Am J Psychiatry. 1995;152(9):1399.
12. Herrmann N, Lanctot K, Myszak M. Effectiveness of gabapentin for the treatment of behavioral disorders in dementia. J Clin Psychopharmacol. 2000;20(1):90-93.
13. Low RA Jr, Brandes M. Gabapentin for the management of agitation. J Clin Psychopharmacol. 1999;19(5):482-483.
14. Markowitz JS, Finkenbine R, Myrick H, King L, Carson WH. Gabapentin abuse in a cocaine user: implications for treatment ? J Clin Psychopharmacol. 1997;17(5):423-424.
15. Myrick H, Malcolm R, Brady KT. Gabapentin treatment of alcohol withdrawal (letter). Am J Psychiatry. 1998;155(11):1632.
16. Hardoy MC, Hardoy MJ, Carta MG, Cabras PL. Gabapentin as a promising treatment for antipsychotic-induced movement disorders in schizoaffective and bipolar disorder patients. J Affect Disord. 1999;54(3):315-317.
17. Morris GL. Gabapentin. Epilepsia. 1999;40(suppl 5):S63-S70.
18. Laird MA, Gidal BE. Use of gabapentin in the treatment of neuropathic pain. Ann Pharmacother. 2000;34(6):802-807.
19. Perez CM, Vasquez PA, Perret CF. Treatment of ciguatera poisoning with gabapentin. N Engl J Med. 2001;344(9):692-693.
20. Myrick H, Malcolm R, Henderson S, McCormick K. Gabapentin for misuse of homemade nicotine nasal spray. Am J Psychiatry. 2001;158(3):498.
21. Crockford D, White WD, Campbell B. Gabapentin use in benzodiazepine dependence and detoxification. Can J Psychiatry. 2001;46(3):287.
22. Zapp JJ. Gabapentin for the treatment of tinnitus: a case report. Ear Nose Throat J. 2001;80(2):114-116.
23. Guttuso TJ Jr. Gabapentin’s effects on hot flashes and hypothermia. Neurology. 2000;54(11):2161-2163.
24. Jeffery SM, Pepe JJ, Popovich LM, Vitagliano G. Gabapentin for hot flashes in prostate cancer. Ann Pharmacother. 2002;36(3):433-436.
25. Dichter MA, Brodie J. New antiepileptic drugs. N Engl J Med. 1996;334(24):1583-1590.
26. Clark JD, Elliott J. Gabapentin-induced anorgasmia. Neurology. 1999;53(9):2209.
27. Labbate LA, Rubey RN. Gabapentin-induced ejaculatory failure and anorgasmia. Am J Psychiatry. 1999;156(6):972.
28. Brannon GE, Rolland PD. Anorgasmia in a patient with bipolar disorder type 1 treated with gabapentin. J Clin Psychopharmacol. 2000;20(3):379-381.
29. Husain AM, Carwile ST, Miller PP, Radtke RA. Improved sexual function in three men taking lamotrigine for epilepsy. South Med J. 2000;93(3):335-336.
30. Montes JM, Ferrando L. Gabapentin-induced anorgasmia as a cause of noncompliance in a bipolar patient. Bipolar Disord. 2001;3(1):52.
31. Grant AC, Oh H. Gabapentin-induced anorgasmia in women. Am J Psychiatry. 2002;159(7):1247.
32. Silvia RJ, Spitznas AL. Gabapentin-related changes in renal function: two case reports. J Clin Psychopharmacol. 2007;27(1):117-119.
33. Tuccori M, Lombardo G, Lapi F, Vannacci A, Blandizzi C, Del Tacca M. Gabapentin-induced severe myopathy. Ann Pharmacother. 2007;41(7):1301-1305.
34.  Naranjo CA, Busto U, Sellers EM, et al. A method for estimating the probability of adverse drug reactions. Clin Pharmacol Ther. 1981;30(2):239-245.
35. Lipson J, Lavoie S, Zimmerman D. Gabapentin-induced myopathy in 2 patients on short daily hemodialysis. Am J Kidney Dis. 2005;45(6):e100-e104.


Dr. Wu is associate director, Mr. Durkin is senior director of outcomes research, and Dr. Canuso is senior director of clinical development at Ortho-McNeil Janssen Scientific Affairs, LLC, in Titusville, New Jersey. Dr. Dickson is professor and past-chairman of the Pharmaceutical and Health Outcomes Sciences Department at the University of South Carolina in Columbia.

Disclosures: Drs. Wu and Canuso own stock in Johnson and Johnson and are employees of Ortho-McNeil Janssen Scientific Affairs, LLC. Dr. Dickson is consultant to Ortho-McNeil Janssen Scientific Affairs, LLC. Mr. Durkin is an employee of Ortho-McNeil Janssen Scientific Affairs, LLC.  This article was funded by Ortho-McNeil Janssen Scientific Affairs, LLC.

Acknowledgments: The authors thank Dr. Victor Navarro for his assistance. Data in this study were presented at the Annual Meeting of the American Psychiatric Association in San Diego, CA, May 19-24, 2007.

Please direct all correspondence to: Jasmanda H. Wu, MPH, PhD, Associate Director, Outcomes Research, Ortho-McNeil Janssen Scientific Affairs, LLC, 1125 Trenton-Harbourton Rd, Titusville, NJ 08560-0200; Tel: 609-730-7718; Fax: 609-730-2411; E-mail: jwu2@omjus.jnj.com.



Focus Points

• There is a high occurrence of liver diseases observed in patients with mental illness.
• True prevalence of liver diseases may be even higher than that observed.
• Clinicians should consider hepatic issues when treating patients with mental illness. 



Objective:  To evaluate the percentage of patients with treated liver diseases in patients with schizophrenia or bipolar disorder, compared to a comparison group without mental illness.
Methods: South Carolina State Medicaid program data were analyzed. Patients <65 years of age, with a schizophrenia or bipolar disorder diagnosis from 2002–2003 were identified. A 4–1 frequency-matching algorithm (4 without diagnosis to 1 with diagnosis) was applied. Treated liver diseases were assessed from ambulatory or hospital claims over the 24-month period and compared between the two groups. Direct standardization was used to account for the differences in demographics between the two groups.  
Results: A total of 5,211 patients with schizophrenia and 4,553 patients with bipolar disorder were identified. After standardizing for the differences in age, gender, and race, the percentages of patients with treated liver diseases in patients with schizophrenia (or bipolar disorder) was statistically significantly higher than the non-mentally ill comparison group (schizophrenia versus comparison=4.64% versus 4.14%,
P<.0001; bipolar disorder versus comparison=6.96% versus 2.66%, P<.0001). The percentages observed in the disease group were also higher than the reported prevalence of liver disease from the historic general population (1.3%).
Conclusion: Medicaid recipients with schizophrenia or bipolar disorder have higher percentages of patients with treated liver diseases than non-mentally ill Medicaid recipients. The percentages observed were also higher than the prevalence rate reported for the general population. Clinicians should consider hepatic issues in the management of their patients.



Schizophrenia and bipolar disorder are severe disorders that usually manifest in late adolescence or early adulthood.1-5 Both conditions involve substantial direct healthcare burden due to early onset, chronicity, and frequent need for repeated hospitalizations.6,7 In addition, patients experience impairments in occupational and social functioning that can result in increased family and societal burden.8,9

It is common for patients with schizophrenia or bipolar disorder to exhibit comorbidities such as substance abuse and hepatic viral infection.10-12 According to the literature, up to 60% of patients with bipolar disorder meet diagnostic criteria for substance abuse.13 Comorbid substance abuse disorder is reported to occur in almost 50% of patients with schizophrenia.13 Patients with alcohol abuse are at particular risk for hepatic impairment.14,15 Increased rates of drug abuse, particularly injection drug use, are likely to contribute to increased risk of hepatitis B virus (HBV) and hepatitis C virus (HCV) in patients with mental illness.16

It has been found that HBV and HCV occur more frequently in patients with severe mental illness than in the general population.17 A 2001 study revealed that the prevalence of HBV (23.4%) and HCV (19.6%), respectively, was 5 and 11 times higher for patients with severe mental illness than in the general population.16 A retrospective database study showed that 9.9% of patients with schizophrenia were infected with HCV.18 Both HBV and HCV are major causes of cirrhosis, hepatic decompensation, and hepatocellular carcinoma.19,20

Patients with schizophrenia and bipolar disorder may be at higher risk for acquiring another spectrum of liver diseases known as non-alcoholic fatty liver disease (NAFLD).21 Among the most common risk factors associated with NAFLD21 are those observed in metabolic syndrome.22 According to the literature, prevalence of metabolic syndrome was 36% for males and 51% for females in patients with schizophrenia. These rates are higher than in the general population (20% males; 25% females).23

Patients with schizophrenia or bipolar disorder commonly receive antipsychotic polypharmacy and pharmacologic add-on therapy for various aspects of their illness.24-27 Unintended interactions between medications may be associated with liver damage. Previous studies have evaluated viral hepatitis in people with severe mental illness.16,17 However, few research articles have been published that address the issue of overall liver diseases in psychiatric patients.

In this study, the authors examined insurance claims data from a State Medicaid program to evaluate the percentages of patients with treated liver diseases in Medicaid recipients with a schizophrenia or bipolar disorder diagnosis, compared to those without mental illness.



Study Cohort

Claims data were analyzed from the South Carolina State Medicaid program database. Approximately 500,000 eligible beneficiaries were identified in 2004 and 2005. Patient identifiers were removed to ensure patient confidentiality. The South Carolina Medicaid Program Committee and the University of South Carolina Institutional Review Board approved the study and did not require informed consent.

Patients <65 years of age who had at least one hospital claim or two ambulatory claims with a diagnosis of schizophrenia or bipolar disorder (International Classification of Diseases, Ninth Revision, Clinical Modification [ICD-9-CM],28 codes 295.xx, 296.0, 296.1, and 296.4-296.8) during 2002 and 2003 were selected. This method has been used previously for selecting patients with schizophrenia, schizoaffective disorder, or bipolar disorder for examining comorbid conditions, adverse events, and medication compliance.29-32 Diagnosis of schizophrenia included schizoaffective disorder with schizophrenia-spectrum illnesses (ICD-9-CM codes 295.7x). Patients who had skilled nursing home claims during 2002 and 2003 were excluded due to the absence of pharmacy claims records. Those without continuous eligibility during 2004 and 2005 were excluded. The enrollment period (January 1, 2002 to December 31, 2003) differed from the follow-up period (January 1, 2004 to December 31, 2005) to allow focus on liver conditions for existing rather than new schizophrenia or bipolar patients. Temporal sequence between the occurrence of liver disease and schizophrenia/bipolar disorder diagnoses was difficult to establish using claims data and was not required for inclusion. 

The authors of this study separated patients into two cohorts, including those with schizophrenia and those with bipolar disorder. Patients with both conditions were excluded. For each patient in either cohort, four individuals were randomly selected using frequency matching on the age category (age groups of 1–20, 21–40, and 41–64 years) to create the comparison group without mental illness. The authors used medical and pharmacy claims to exclude comparison group subjects who exhibited a diagnosis of mental illness (ICD-9-CM codes 290.xx to 319.xx) or who had received antipsychotics between January 1, 2002 through December 31, 2005. National Drug Codes33 were used to identify the presence of antipsychotics within the comparison group cohort, including aripiprazole, chlorpromazine, clozapine, fluphenazine, haloperidol, loxapine, mesoridazine, molindone, olanzapine, perphenazine, pimozide, promazine, quetiapine, risperidone, thioridazine, thiothixene, trifluoperazine, and ziprasidone.

The four cohorts (schizophrenia and bipolar disorder, and their corresponding comparison group without mental illness) were reviewed for new and existing liver diseases by searching claims for related ICD-9-CM codes occurring between January 1, 2004 and December 31, 2005.



Liver Diseases
Individuals were identified as having liver diseases (acute and chronic), viral hepatitis, or chronic cirrhosis if they had at least one hospital claim or two ambulatory claims for the ICD-9-CM codes listed in Table 1.28 The method using two separate outpatient visits to help increase the diagnostic specificity has been used previously to examine medical comorbidities in patients with bipolar disorder.29 Patients were identified as having NALFD if they had at least one hospital claim or two ambulatory claims for the ICD-9-CM codes listed in Table 1 and had a diagnosis of diabetes (the ICD-9-CM codes 250.xx), obesity (278.0), hyperlipidemia (272.0, 272.1, 272.2, 272.3, and 272.4), or hypertension (401.xx–405.xx). NAFLD patient selection also required the absence of specific ICD-9-CM codes in their claim history, including 571.0, 571.1, 571.2, 571.3, 573.1, 070, and V02.6. Ascertaining diagnosis of NAFLD generally is to rule out co-existent or alternate liver diseases and to evaluate the presence of metabolic syndrome and insulin resistance in patients.21 Determination of specific ICD-9-CM codes was based on hepatologist consultation and published literature.34 Absence of clinical details in the claims forms and different coding practice in different clinics may have compromised classification of specific types of liver diseases from these claims. The authors of this study intended to focus on the overall liver diseases as the primary outcome of interest.



Comorbid Conditions
The four study cohorts were evaluated for comorbid conditions (January 1, 2002–December 2003), which included hypertension (ICD-9-CM codes 401.xx–405.xx), hyperlipidemia (272.0, 272.1, 272.2, 272.3, and 272.4), diabetes (250.xx), obesity (278.0), human immunodeficiency virus (HIV) infection (042, 795.71, and V08), and substance abuse (291.xx, 292.xx, 303.xx–304.xx, and 305.xx).

Percentage of Patients with Treated Liver Diseases
The percentage of patients with treated liver diseases for 2004–2005 were computed using the number of new and existing liver disease cases divided by the total number of individuals in the schizophrenia, bipolar disorder, and comparison (individuals without mental illness) groups.

Direct Standardization
The adjusted percentages of patients with treated liver diseases were computed using direct standardization to account for the differences in age, gender, and race between the schizophrenia (or bipolar disorder) and the comparison without mental illness groups. The adjusted percentages were computed for the comparison group based on the age, gender, and race distribution of the schizophrenia (or bipolar disorder) cohort. The detailed formula for the computation is displayed in Table 2. For illustration purpose, only the age group between 36 and 50 is included in Table 2.




Sub-Analysis on Schizoaffective Disorder

The study authors also performed a subgroup analysis to evaluate demographic characteristics and outcomes of interest for schizoaffective disorder patients with schizophrenia-spectrum illnesses (ICD-9-CM codes 295.7x). 


Statistical Methods

Statistical analyses were performed using Statistical Analysis System (version 9.1.3). Standard descriptive statistics summarized the data in the schizophrenia, bipolar disorder, schizoaffective disorder with schizophrenia-spectrum illnesses, and comparison groups. The mean, median, standard deviation, minimum, and maximum were provided for continuous variables. Frequency distributions were provided for categorical variables, including gender, race, and comobid conditions. The differences in the categorical variables between the schizophrenia (bipolar disorder or schizoaffective disorder with schizophrenia-spectrum illnesses) and the comparison groups were assessed using a Chi-square test for independent samples. The differences in the continuous variables were assessed using a t-test. A two-sided 5% significance level was used for all tests.



Baseline Characteristics

Patients with schizophrenia were identified (n=5,211; n=20,844 comparison group without mental illness). Baseline characteristics (Table 3) indicated that age distribution was similar between the two groups as expected using frequency matching. There was a higher proportion of male patients in the schizophrenia group compared to the comparison group without mental illness. By race, black patients represented the highest proportion of patients in the schizophrenia cohort as compared to the comparison group.


For comorbid conditions, 19.2% of schizophrenia patients had substance abuse. Patients with schizophrenia were more likely to have HIV infection, hypertension, diabetes, hyperlipidemia, and obesity than the comparison group without mental illness.

The mean age of patients with bipolar disorder was 33.0. By race, white patients represented the highest proportion of patients in the bipolar cohort as compared to the comparison group without mental illness. Approximately 29.3% of patients with bipolar disorder had substance abuse, and patients with bipolar disorder were more likely to have HIV infection, hypertension, diabetes, hyperlipidemia, and obesity than the comparison group without mental illness.


Treated Liver Diseases

The percentage of patients with treated liver diseases in patients with schizophrenia were statistically significantly higher than the comparison group after standardizing the age, gender, and race distribution of the two groups (in schizophrenia versus comparison without mental illness, overall liver diseases was 4.64% versus 4.14%, P<.0001) (Table 4). The majority (80%) of these cases were coded using the liver disease-related ICD-9-CM codes other than abnormal liver function test (ICD-9-CM codes 790.4 and 794.8). Schizophrenia patients also had higher percentages of various liver diseases, except for NAFLD.



The differences in the percentages of patients with treated liver diseases were even greater between patients with bipolar disorder and the comparison group without mental illness after standardizing the age, gender, and race distribution of the two groups. In bipolar disorder versus comparison without mental illness, the percentage of overall liver diseases was 6.96% versus 2.66%, P<.0001; acute liver disease was 5.47% versus 2.00%, P<.0001; chronic liver disease was 6.94% versus 2.66%, P<.0001; viral hepatitis was 2.99% versus 0.73%, P<.0001; chronic cirrhosis was 2.09% versus 0.84%, P<.0001; and NAFLD was 1.89% versus 1.05%, P<.0001.

For various subgroup analyses, the percentage of patients with treated liver diseases were higher in females than males (eg, bipolar disorder, male [4.53%] versus female [8.21%], P<.0001). “Other” race had the highest prevalence of overall liver diseases, followed by white, then black (eg, bipolar disorder, white [7.59%] versus black [4.33%] versus “other” [8.67%], P=.0007). The group 31–50 years of age had the highest percentage of patients with treated liver diseases (eg, bipolar disorder, age <30 [2.97%] versus 31–50 years of age [10.76%], versus >50 [7.38%], P<.0001).


Schizoaffective Disorder Group

A total of 2,056 schizoaffective disorder patients with schizophrenia-spectrum illnesses were identified, along with 8,224 subjects in the comparison group without mental illness. The mean age was 42.1 years in these patients and a higher percentage of male patients in the schizoaffective disorder group were observed. In the schizoaffective versus comparison group, the male percentage was 42.7% versus 25.3%. Approximately 19% of schizoaffective disorder patients with schizophrenia-spectrum illnesses had substance abuse, and these patients were more likely to have hypertension, diabetes, hyperlipidemia, and obesity than the comparison group without mental illness. In the schizoaffective versus comparison without mental illness, hypertension was 35.5% versus 32.8%, P=.0214; diabetes was 20.5% versus 18.6%, P=.0515; hyperlipidemia was 17.7% versus 15.4%, P=.0141; and obesity was 8.6% versus 6.0%, P<.0001.

Schizoaffective disorder patients with schizophrenia-spectrum illnesses had higher percentages of patients with various treated liver diseases, except for NAFLD, compared to the comparison group without mental illness after standardizing the age, gender, and race distribution of the two groups. In the schizoaffective versus comparison without mental illness, overall liver disease was 4.62% versus 4.12%, P<.0001; acute liver disease was 3.84% versus 3.37%, P<.0001; chronic liver disease was 4.57% versus 4.11%, P<.0001; viral hepatitis was 1.70% versus 1.27%, P<.0001; chronic cirrhosis was 1.12% versus 1.06%, P<.0001; and NAFLD was 1.31% versus 1.55%, P<.0001.



In this study, claims data from the South Carolina State Medicaid program were used to evaluate the percentages of patients with treated liver diseases in Medicaid recipients with the diagnosis of schizophrenia (including schizoaffective disorder patients with schizophrenia-spectrum illnesses) or bipolar disorder as compared to those without diagnosis of mental illness. Results indicate that 4.64% of patients with schizophrenia and 6.96% with bipolar disorder had a diagnosis of liver disease, which is statistically significantly higher than Medicaid recipients without mental illness. Although the absolute difference in the prevalence rates observed between Medicaid recipients with schizophrenia and Medicaid recipients without mental illness is small, bias may be introduced when using Medicaid recipients without mental illness as the comparison group. The percentages of patients with treated liver diseases observed in Medicaid recipients with schizophrenia or bipolar disorder are approximately three to five times higher than the 1.3% reported prevalence of liver disease in the general population (National Health Interview Survey)35 and are more than 10 times higher than the 0.2% reported in the non-Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition,36 disorder control group in a study by Carney and Jones.29 The percentages of patients with treated liver diseases found in this study are slightly <7.4%, as reported from a claim-based study which evaluated comorbid conditions among patients with schizophrenia in the Veterans Health Administration database.30

The authors of this study also found that the differences in the percentages of patients with treated liver diseases between the disease group and the comparison group without mental illness were larger for patients with bipolar disorder than those with schizophrenia. Higher percentages of patients with substance abuse and viral hepatitis observed in patients with bipolar disorder may account for this finding. The percentages of patients with NAFLD were also higher in the bipolar disorder group than the comparison group. Most patients with NAFLD are female and obesity is present in >50% of these patients.21 The bipolar cohort reflected higher percentages of female patients and obesity and may partially explain the observation.

The percentages of patients with treated liver diseases found in the two different samples of the comparison groups without mental illness were different (eg, the adjusted percentages of patients with treated liver diseases were 4.14% versus 2.66%). Since the percentages in the comparison groups were adjusted based on the demographic distribution of the disease groups, the differences observed are likely related to the differences in demographic distribution of these two samples.

It is surprising that the percentages of patients with substance abuse in schizophrenia patients (19.2%) and bipolar disorder patients (29.3%) was much lower than those reported in the literature.13  However, substance abuse is often underreported and underdiagnosed and may not be fully represented in claims data since treatment may be bundled and not readily recognized in the claims history.

The authors of this study found a higher proportion of male and black patients in the schizophrenia group, and a higher proportion of female and white patients in the bipolar disorder group. Some evidence suggests that clinicians may have a tendency to overdiagnose schizophrenia in some ethnic groups.36 It is unknown whether the differences in the diagnosis of schizophrenia and bipolar disorder observed in our study represent true differences among gender and race groups or if it results from clinician bias or cultural insensitivity. In addition, since the Medicaid population has a higher proportion of minority groups compared to the general population, the differences observed among gender and race groups in this study may be larger than other studies due to overrepresentation of minority population.

The true prevalence of liver diseases may be higher due to failure of asymptomatic cases to present for care. For example, HBV and HCV do not have subjective symptoms in most instances and many people may be unaware that they are infected with HBV and HCV.37 Cases with symptoms may be under-recognized because common complaints are often nonspecific (eg, fatigue, muscle ache, headache, abdominal discomfort).38 Seroprevalence rates of HBV and HCV have been reported16 and the prevalence of HBV (23.4%) and HCV (19.6%) found in that study was nearly 10 times higher than those observed in this study.

The findings of this research may have important implications for the management of mentally ill patients. Clinicians should consider hepatic issues when treating patients with mental illness who are at higher risk for liver disease. In addition to addressing treatment issues such as psychotic symptoms, clinicians could better serve their patients by evaluating and monitoring risk factors and clinical indications of liver disease. Clinicians should be aware that hepatic disease may impair the liver’s synthetic function, resulting in alterations in the protein binding and the bioactivity of various treatments. Further, when severe, hepatic compromise may also affect drug metabolism. As a result, dose adjustment of most antipsychotics is often necessary in these patients. Liver diseases may reduce antipsychotic treatment options for these patients and clinical consideration of medications that place a lower metabolic burden on the liver may present a viable treatment option for some patients.

This study has some limitations. Claims database are not well suited to address epidemiology questions, and the true prevalence rate of liver diseases was not evaluated in the study. Medicaid data are confounded by people moving in and out of the system who may not be fully captured. In addition, some individuals at high risk for medical comorbidities, such as those who are homeless, may not have acquired Medicaid coverage. Further, the findings of this study were based on data from the South Carolina State Medicaid program and may not be generalizable. Similar studies in different populations or study settings would be useful to provide context for these findings. Because the data were collected for reimbursement rather than clinical purposes, it is possible that some diseases experienced by the members may be misclassified or omitted from the database. The authors’ strategy of using at least one hospital claim or two ambulatory claims for liver disease-related ICD-9-CM codes to identify study participants was aimed toward minimizing misclassification of liver conditions. Finally, claims databases generally lack clinical details, and the classification of various types of liver diseases may not be accurate.



In this study, Medicaid recipients with schizophrenia or bipolar disorder showed higher percentages of treated liver diseases than non-mentally ill Medicaid recipients, and the percentages observed were also higher than the prevalence rate reported for the general population. The differences in the percentages of patients with treated liver diseases between the disease group and the comparison group without mental illness were larger for patients with bipolar disorder than with schizophrenia. The true prevalence of liver diseases may be higher than that observed due to failure of asymptomatic cases to present for care. Clinicians should consider hepatic issues when treating patients with mental illness who are at higher risk for liver disease. PP



1. Practice Guideline for the Treatment of Patients with Schizophrenia. 2nd ed. Washington, DC: American Psychiatric Association; 2004.
2. Lieberman JA. Prediction of outcome in first-episode schizophrenia. J Clin Psychiatry. 1999;V54(suppl 3):13-17.
3. Regier DA, Narrow WE, Rae DS, Manderscheid RW, Locke BZ, Goodwin FK. The de facto US mental and addictive disorders service system. Epidemiologic catchment area prospective 1-year prevalence rates of disorders and services. Arch Gen Psychiatry. 1993;50(2):85-94.
4. Akiskal HS. The prevalent clinical spectrum of bipolar disorders. J Clin Psychopharmacol. 1996;16(suppl 1):4S-14S.
5. Grant BF, Stinson FS, Hasin DS, et al. Prevalence, correlates, and comorbidity of Bipolar I Disorder and Axis I and II Disorders: results from the national epidemiologic survey on alcohol and related conditions. J Clin Psychiatry. 2005;66(10):1205-1215.
6. Robinson D, Woerner MG, Alvir JM, et al. Predictors of relapse following response from a first episode of schizophrenia or schizoaffective disorder. Arch Gen Psychiatry. 1999;56(3):241-247.
7. Patel NC, Crismon ML, Pondrom M. Rehospitalization rates of patients with bipolar disorder discharged on a mood stabilizer versus a mood stabilizer plus an atypical or typical antipsychotic. J Behav Health Serv Res. 2005;32(4):438-445.
8. Wu EQ, Birnbaum HG, Shi L, et al. The economic burden of schizophrenia in the United States in 2002. J Clin Psychiatry. 2005;66(9):1122-1129.
9. Wyatt RJ, Henter I. An economic evaluation of manic-depressive illness–1991. Soc Psychiatry Psychiatr Epidemiol. 1995;30(5):213-219.
10. Green AL. Treatment of schizophrenia and comorbid substance abuse: pharmacologic approaches. J Clin Psychiatry. 2006;67(suppl 7):31-35.
11. Cournos F, McKinnon K, Sullivan G. Schizophrenia and comorbid human immunodeficiency virus or hepatitis C virus. J Clin Psychiatry. 2005;66(suppl 6):27-33.
12. Dinwiddie SH, Shicker L, Newman T. Prevalence of hepatitis C among psychiatric patients in the public sector. Am J Psychiatry. 2003;160(1):172-174.
13. 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. 1990;264(19):2511-2518.
14. Meyerhoff DJ, Bode C, Nixon SJ, de Bruin EA, Bode JC, Seitz HK. Health risks of chronic moderate and heavy alcohol consumption: how much is too much? Alcohol Clin Exp Res. 2005;29(7):1334-1340.
15. Ereshefsky L. Pharmacokinetics and drug interactions: update for new antipsychotics. J Clin Psychiatry. 1996:57(suppl 11):12-25.
16. Rosenberg SD, Goodman LA, Osher FC, et al. Prevalence of HIV, hepatitis B, and hepatitis C in people with severe mental illness. Am J Public Health. 2001;91(1):31-37.
17. Goldberg RW. Hepatitis and HIV screening, education and treatment for adults with serious mental illness. Gen Hosp Psychiatry. 2004;26(2):167-168.
18. Huckans MS, Blackwell AD, Harms TA, Hauser P. Management of hepatitis C disease among VA patients with schizophrenia and substance use disorders. Psychiatr Serv. 2006;57(3):403-406.
19. Strader DB, Seeff LB. Hepatitis C: a brief clinical overview. Ilar J. 2001;42(2):107-116.
20. Han SH. Natural course, therapeutic options and economic evaluation of therapies for chronic hepatitis B. Drugs. 2006;66(14):1831-1851.
21. Cortez-Pinto H, Camilo ME. Non-alcoholic fatty liver disease/non-alcoholic steatohepatitis (NAFLD/NASH): diagnosis and clinical course. Best Pract Res Clin Gastroenterol. 2004;18(6):1089-1104.
22. Executive Summary of The Third Report of the National Cholesterol Education Program (NCEP) Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol In Adults (Adult Treatment Panel III). JAMA. 2001;285(19):2486-2497.
23. McEvoy JP, Meyer JM, Goff DC, et al. Prevalence of the metabolic syndrome in patients with schizophrenia: baseline results from the Clinical Antipsychotic Trials of Intervention Effectiveness (CATIE) schizophrenia trial and comparison with national estimates from NHANES III. Schizophr Res. 2005;80(1):19-32.
24. Botts S, Hines H, Littrell R. Antipsychotic polypharmacy in the ambulatory care setting, 1993-2000. Psychiatr Serv. 2003;54(8):1086.
25. Stahl SM, Grady MM. High-cost use of second-generation antipsychotic under California’s Medicaid Program. Psychiatr Serv. 2006;57(1):127-129.
26. McCue RE, Waheed R, Urcuyo L. Polypharmacy in patients with schizophrenia. J Clin Psychiatry. 2003;64(9):984-989.
27. Fleischhacker WW. New developments in the pharmacotherapy of schizophrenia. J Neural Transm Suppl. 2003;(64):105-17.
28. ICD-9-CM: International Classification of Diseases. 9th Rev, Clinical Modification, 6th ed (CD-ROM). Baltimore, Maryland: Centers for Medicare and Medicaid Services; 2005.
29. Carney CP, Jones LE. Medical comorbidities in women and men with bipolar disorders: a population-based controlled study. Psychosom Med. 2006;68(5):684-691.
30. Kilbourne AM, Cornelius JR, Han X, et al. General-medical conditions in older patients with serious mental illness. Am J Geriatr Psychiatry. 2005;13(3):250-254.
31. Guo JJ, Keck PE Jr, Corey-Lisle PK, et al. Risk of diabetes mellitus associated with atypical antipsychotic use among patients with bipolar disorder: a retrospective, population-based, case-control study. J Clin Psychiatry. 2006;67(7):1055-1061.
32. Weiden PJ, Kozma C, Grogg A, Locklear J. Partial compliance and risk of rehospitalization among California Medicaid Patients with schizophrenia. Psychiatr Serv. 2004;55(8):886-891.
33. US Food and Drug Administration. The National Drug Code Directory. Available at: www.fda.gov/cder/ndc. Accessed September 12, 2007.
34. Chan KA, Truman A, Gurwitz JH, et al. A cohort study of the incidence of serious acute liver injury in diabetic patients treated with hypoglycemic agents. Arch Intern Med. 2003;163(6):782-734.
35. Centers for Disease Control and Prevention. National Center for Health Statistics. Summary Health Statistics for US Adults: National Health Interview Survey, 2004. Vital and Health Statistics. May 2006; series 10, number 228.
36. Diagnostic and Statistical Manual of Mental Disorders. 4th ed. Washington, DC: American Psychiatic Association; 1994.
37. Iino S. Natural history of hepatitis B and C virus infections. Oncology. 2002;62(suppl 1):18-23.
38. Ware JE, Bayliss MS, Mannocchia M, Davis GL. Health-related quality of life in chronic hepatitis C: impact of disease and treatment response. The Interventional Therapy Group. Hepatology. 1999;30(2):550-555.


Dr. Robinson is a consultant with Worldwide Drug Development in Burlington, Vermont.

Disclosure: Dr. Robinson has served as a consultant to Bristol-Myers Squibb, CeNeRx, Epix, Genaissance, Medicinova, Ono Pharmaceuticals, Pfizer, and Somerset.



The introduction of several new atypical antipsychotics over the past decade poses uncertainties about what constitutes optimal drug therapy for an individual patient, both in choosing initial drug therapy and ongoing management. The advent of the second-generation antipsychotics (SGAs) bred a new category of drugs with some advantages over traditional agents, although the extent of superiority in efficacy and tolerability over first-generation antipsychotics (FGAs) remains unclear. SGAs are a pharmacologically heterogeneous group of drugs with differing liabilities, depending on the agent. Two recent monographs review the pharmacology and therapeutics of the antipsychotics and summarize available data to provide the most current information to guide clinical practice.1,2

Therapeutic indications for antipsychotics include short-term management of acute psychoses and agitated states as well as long-term treatment of chronic psychotic disorders such as schizophrenia and delusional disorders. SGAs have largely replaced FGAs in clinical practice, although the limited data from comparative trials do not show SGAs to be clearly superior. Purported therapeutic advantages of SGAs, which vary depending on the specific agent, appear relatively modest compared with traditional antipsychotics employed in moderate dosage and with adjunctive antiparkinson medication, if needed.1,3,4 As noted in a recent editorial,5 emerging data on SGAs make it apparent that the significant side-effect burden of antipsychotic drug therapy has shifted, not disappeared. 


Mechanism of Action of Antipsychotics

Although antipsychotics emerged in the 1950s with the introduction of chlorpromazine (the result of a serendipitous discovery), the principal mechanism of action of antipsychotics remained unclear for nearly a decade thereafter. In a milestone publication reviewing pharmacologic evidence elucidating the neuropathology of schizophrenia, Matthysse6 pointed out that dopamine agonists precipitate or worsen psychosis while dopamine antagonists (eg, antipsychotics) effectively treat chronic psychosis and mania. In ensuing years, the focus of drug discovery centered on dopamine-receptor antagonists, primarily dopamine (D)2-receptor blockers. Potent antidopaminergic drugs, while highly effective, also carried a significant risk of extrapyramidal symptoms (EPS) and tardive dyskinesia.

A major advance in understanding the pharmacology of antipsychotics derives from a clinical trial involving clozapine, the first of the SGAs.7 This trial compared clozapine and chlorpromazine in schizophrenic patients who failed a prior course of high-dose haloperidol and found clozapine superior to chlorpromazine, while devoid of the adverse neurologic effects of other antipsychotics.7 Clozapine was labeled “atypical” because it combined superior efficacy with low risk of EPS. This term subsequently was applied broadly (and uncritically) to all of the SGAs despite having significant pharmacologic heterogeneity—chemically, pharmacologically, and clinically.1


Neuropharmacology of Antipsychotics

The weight of the evidence indicates that dopamine-receptor blockade is essential to clinical antipsychotic activity, especially for controlling hallucinations and delusions, but emerging data suggest this may not be the only mechanism of action. Despite the advances in cloning neuroreceptor subtypes and defining specific receptor-binding characteristics of individual antipsychotics, a unifying theory of the mechanism of action of antipsychotics has not emerged. The ideal pharmacologic profile for an antipsychotic remains an unanswered question.1

Given its low propensity for EPS and unexcelled clinical efficacy, the pharmacologic profile of clozapine became the subject of intense scrutiny in designing new antipsychotics. The pharmacology of clozapine is especially complex; it binds loosely (and transiently) to D2 receptors as well as other dopamine receptor subtypes, and also binds with high affinity to histamine (H1), acetylcholine muscarinic (M1), α-adrenergic (α1), and multiple serotonin (5-HT2A, 5-HT2C, 5-HT7) receptors. This complexity leaves the minimal EPS risk and unexcelled antipsychotic effectiveness of clozapine unexplained.1 However, emerging data suggest that significant binding to 5-HT2A receptors in addition to D2 receptors confers lower risk of EPS.

The receptor-binding profiles of the atypical antipsychotics exhibit significant differences amongst the SGAs. Understanding and appreciating these distinctions can guide in selecting pharmacotherapy for the individual patient, a task that can be daunting. Clinicians are confronted with many options that span drug choice, dose optimization, adjunctive and augmentation therapy, and the decision to switch drugs. Given unclear evidence of therapeutic superiority of SGAs3-5 and their heterogeneous pharmacology, a panel of psychopharmacologists recently convened to address unresolved therapeutic issues and to offer guidance in clinical decision making.2 One factor deemed important by the panel was acquiring a thorough understanding of the differing pharmacologic profiles of the SGAs.


Receptor Affinities and Choice of Antipsychotic Drug

While comparative trials, if available, are useful in choosing pharmacotherapy, the panel stressed that knowledge of the pharmacodynamic profiles of the SGAs is helpful, given the modest differences in antipsychotic efficacy but clear-cut safety differences. The pharmacodynamic effects of an antipsychotic are dictated by its receptor-binding properties, both with respect to efficacy and adverse effects. While traditional antipsychotics bind principally and most avidly to D2 receptors, SGAs exhibit high affinities for other receptors equal to or exceeding that for the D2 receptor.

SGAs can be grouped according to patterns of receptor affinities and dissociation. For example, clozapine, olanzapine, and ziprasidone have higher affinity for 5-HT2A receptors relative to D2 receptors, a binding profile differing somewhat from other SGAs.1 Positron emission tomography (PET) studies show that the SGAs clozapine, olanzapine, quetiapine, and ziprasidone (but not aripiprazole and risperidone) have moderate affinity and low avidity (ie, rapid dissociation) for D2 receptors in the basal ganglia, a pattern consistent with low EPS liability. The anticholinergic effects of clozapine and olanzapine also may limit risk of EPS and avoid a need to add an antimuscarinic antiparkinson agent, as often required with conventional antipsychotics to rebalance critical dopamine-cholinergic functions in the basal ganglia.1 Clozapine and olanzapine (which resembles clozapine structurally and pharmacologically) have high affinities relative to the D2 receptor for M1, H1, 5-HT2C, and 5-HT2A receptors, in that order respectively. This profile of receptor interactions informs about both the therapeutic and side-effect profiles of these two SGAs, eg, well-established antipsychotic efficacy with minimal risk of EPS and significant liability for excessive weight gain and sedation (Table).2



Quetiapine exhibits high affinities relative to the D2 receptor for H1 and α1-adrenergic receptors, reflecting a propensity for causing sedation and postural symptoms. Risperidone binds significantly to 5-HT2A, α1, and D2 receptors, and may have a somewhat higher propensity for EPS than other SGAs. Ziprasidone, like risperidone, has high affinity for the 5-HT2A receptor relative to D2 receptors but differs from risperidone by binding to multiple dopamine-receptor subtypes as well as to 5-HT2C and 5-HT1A receptors.


Drugs Acting as Partial Agonists

Binding of a drug to a receptor can theoretically result in either an agonist or antagonist action, or in “partial agonism,” as discussed in a previous column.8 A drug acting as an antagonist blocks receptor activation and takes a receptor “out of play,” while an agonist functions like an endogenous neurotransmitter to fully activate a receptor.2 A partial agonist exhibits affinity for a receptor but lacks the full effect of the endogenous neurotransmitter. It competes with the natural transmitter and thereby produces an attenuated response. Aripiprazole has high-binding affinity for D2 receptors (as well as 5-HT1A and 5-HT2A receptors) and PET studies demonstrate high occupancy of D2 receptors (>90%) at therapeutic doses. However, its pharmacologic action as a partial agonist attenuates full response and diminishes receptor activation by the endogenous neurotransmitter dopamine.

Ziprasidone exhibits high affinity for 5-HT1A receptors where it functions as a partial agonist. In addition, it binds with moderate affinity to the serotonin transporter in brain tissues. It is postulated that these pharmacologic properties potentially convey efficacy of ziprasidone in anxiety and depressive states in addition to psychosis.1



On combining drugs or switching antipsychotic therapy, the additive receptor-binding properties of two concurrent SGAs should be considered. In general, D2-receptor blockade of ≥50% is needed for antipsychotic efficacy, while the threshold for risk of EPS is considered to be ≥80% occupancy (aripiprazole is an exception because it possesses high affinity for D2 receptors but acts as a partial dopamine agonist). Combined use of antipsychotics may have a greater propensity for side effects than either drug alone, depending on similar receptor-binding properties (Table).2

Another pharmacologic consideration is the fact that the brain adapts to psychotropic medications through a series of compensatory mechanisms, eg, upregulation of receptors in response to a drug acting as an antagonist, and downregulation of receptors in response to treatment with an agonist drug.2 The possibility of pharmacologic adaptation needs to be considered when switching drugs, since withdrawal effects may ensue from the withdrawn drug but be mistakenly attributed to the new medication. For example, abrupt discontinuation of a sedative SGA on switching to aripiprazole or ziprasidone, which do not block histamine receptors, may cause transient “activation” due to a pharmacologic withdrawal syndrome. As another example, on switching from an antipsychotic with potent muscarinic anticholinergic properties (M1 antagonist) to one with little anticholinergic activity, abrupt discontinuation may cause transient “cholinergic rebound.” This may be erroneously attributed to the new antipsychotic rather than to withdrawal of the prior drug, and lead to premature changes in therapy or drug dose.



SGAs do not differ substantially in efficacy from traditional antipsychotics but have differing side-effect burdens. Because of the pharmacologic heterogeneity of this group of drugs, an understanding of the specific receptor-binding properties of each antipsychotic can be helpful in selecting appropriate drug therapy for the individual patient. PP



1. Gardner DM, Baldessarini RJ, Waraich P. Modern antipsychotic drugs: a critical review. CMAJ. 2005;172(13):1703-1711.
2. Weiden PJ, Preskorn SH, Fahnestock PA, et al. Translating the psychopharmacology of antipsychotics to individualized treatment of severe mental illness: a roadmap. J Clin Psychiatry. 2007;68(suppl 7):6-46.
3. Lieberman JA, Stroup TS, McEvoy JP, et al. Effectiveness if antipsychotic drugs in patients with chronic schizophrenia. N Eng J Med. 2005;353(12):1209-1233.
4. Rosenheck RA, Leslie DL, Sindelar J, et al. Cost-effectiveness of second generation antipsychotics and perphenazine in a randomized trial of treatment of chronic schizophrenia. Am J Psychiatry. 2006;163(12):2080-2095.
5. Freedman R, Carpenter WT, Davis JM, Goldman HH, Tamminga CA, Thomas M. The cost of drugs for schizophrenia. Am J Psychiatry. 2006;163(12):2029-2031.
6. Matthysse S. Antipsychotic drug actions: a clue to the neuropathology of schizophrenia? Fed Proc. 1973;32(2):200-205.
7. Kane J, Honigfeld G, Singer J, Meltzer HY. Clozapine for the treatment resistant schizophrenic. A double-blind comparison with chlorpromazine. Arch Gen Psychiatry. 1988;45(9):789-796.
8. Robinson DS. CNS receptor partial agonists: a new approach to drug discovery. Primary Psychiatry. 2007;14(8):22-24.


Dr. Kim is director of the Hennepin Women’s Mental Health Program at Hennepin County Medical Center in Minneapolis, Minnesota, and clinical assistant professor of psychiatry at the University of Minnesota Medical School in Minneapolis. Dr. Kolpe is staff psychiatrist at the Hennepin Women’s Mental Health Program and clinical assistant professor of psychiatry at the University of Minnesota Medical School.

Disclosures: The authors report no affiliation with or financial interest in any organization that may pose a conflict of interest.
Please direct all correspondence to: Helen Kim, MD, Hennepin Women’s Mental Health Program, S110 Clinic, Hennepin County Medical Center, 900 S 8th St, Minneapolis, MN 55404; Tel: 612-347-6851; Fax: 612-373-1859; E-mail: kimxx237@umn.edu.





What should a clinician consider in selecting a psychotropic medication for pregnant and lactating women with psychiatric symptoms? While postpartum depression is increasingly recognized as a public health priority, vulnerability to depression and anxiety begin in pregnancy. Clinicians are often called upon to counsel pregnant and postpartum patients about the risks and benefits of psychotropic medication. Patients generally overestimate the risks of psychotropic drugs and underestimate the impact of untreated psychiatric illness on themselves and their families. This article reviews recent studies and treatment considerations in selecting psychotropic medications for perinatal women with mood and anxiety symptoms.



The perinatal period can be a high-risk time for worsening mood and anxiety symptoms. While postpartum depression has been identified as a major public health concern,1 vulnerability to depression clearly begins before delivery and continues into the postpartum.2-6 A systematic review of perinatal depression found that the point of prevalence of major depressive disorder (MDD) and minor depression during pregnancy ranged from 6.5% to 12.9%, while another study of >3,000 pregnant patients found that 20% had significant depressive symptoms.7 Many factors have been implicated in worsening mood and anxiety symptoms during pregnancy, including psychosocial risk factors as well as physiologic changes associated with the perinatal period.8-11 Disrupting a maintenance antidepressant may also lead to worsening symptoms in pregnancy as demonstrated in a recent study by Cohen and colleagues,12 in which 68% of women with a history of recurrent MDD relapsed with depression after discontinuing antidepressants compared to 26% who maintained treatment.

With the high prevalence of MDD among women of reproductive age, clinicians are often called upon to help patients weigh the benefits and risks of psychotropic medications during pregnancy and lactation. Recent studies have highlighted both the potential risks of antidepressants in pregnancy as well as the effects of untreated psychiatric symptoms during and after pregnancy. Several studies from the renowned Motherisk Program in Toronto have examined the tendency of pregnant patients to underestimate this risk of untreated depression and anxiety13,14 and overestimate the risk of medication. For example, though the baseline rate of malformations in the general population is approximately 3% to 4%, women exposed to known nonteratogens assigned themselves a risk of 24% for major malformation before hearing about relevant medical studies, and then 14.5% thereafter.15 This misperception of risk can lead patients and physicians to avoid or terminate otherwise wanted pregnancies or avoid needed pharmacotherapy. Clinicians can help put the risks of medication into context by reminding patients that pregnancy itself carries many risks including spontaneous abortion and congenital defects. Many studies have documented the fact that untreated psychiatric illness can compound these risks by contributing to poor self care, decreased prenatal compliance, increased nicotine and substance misuse,16,17 poor obstetrical outcomes,18,19 and increased risk of postpartum depression.20 Given the potential risks of medication during pregnancy, patients often feel like they must not put their own need for symptom relief ahead of the well-being of the pregnancy. Clinicians can help patients realize that these two priorities are inextricably tied since emotional distress also impacts the pregnancy.

Routine formalized screening for major Axis I diagnoses both prenatally and postnatally can assist clinicians in identifying patients with psychiatric symptoms.21,22 Risk factors for depression and anxiety during and after pregnancy are also readily identifiable and include prior history of depression, young age, poverty, stressful life events, and limited social support.23,24 After screening for these risk factors, evaluating current symptoms, and ruling out possible medical conditions (eg, anemia or thyroid dysfunction) that may be contributing to mood, anxiety, and neurovegetative symptoms, clinicians must then direct patients toward the most appropriate course of treatment. This process involves careful review of the risks of untreated illness versus the risks and benefits of available treatment including medication, therapy, and supportive interventions. As Stowe and colleagues25 eloquently articulated, perinatal patients must be guided toward minimizing not only exposure to medication but exposure to illness as well. This article reviews recent studies and focuses on treatment considerations in selecting psychotropic medications for perinatal women with mood and anxiety symptoms.


Pharmacologic Considerations in Pregnancy


Congenital Malformations
Medications taken during pregnancy are considered teratogenic if they increase the risk of congenital malformations above the baseline risk of 3% to 4%. Most studies of tricyclic antidepressants (TCAs) and selective serotonin reuptake inhibitors (SSRIs) are reassuring and show no increased risk of major congenital malformations.26,27 However, while previous studies have shown no association between paroxetine and congenital malformations,28-30 the manufacturer of paroxetine issued a warning in 2005 that two studies found a possible association between first-trimester paroxetine exposure and increased risk for cardiac defects, particularly atrial and ventral septal in nature.31 Subsequently, the Food and Drug Administration issued a similar warning that first-trimester paroxetine use was associated with an increased risk of major malformations (4% versus 3%), particularly cardiac defects (2% versus 1%), and changed the pregnancy labeling from category C to D, which indicates that controlled or observational studies in pregnant women have demonstrated a risk to the fetus. In addition, the American College of Obstetricians and Gynecologists has recommended avoiding paroxetine use in pregnant women unless the benefits outweigh the risks of discontinuing the medication.32

While there have been fewer reported cases of prenatal exposures to non-SSRIs, the limited data available have not shown an increased risk of congenital malformations with venlafaxine,33 mirtazapine,34 nefazodone, or trazodone.35 In an admirable attempt to prospectively assess reproductive risk, the drug manufacturer of bupropion established a postmarketing epidemiologic surveillance registry in 1997. This registry reported a preliminary finding of a possible association between bupropion exposure and increased risk of birth defects involving the heart and great vessels. In the most recent report from this registry spanning September 1997 through August 2006, 1,443 prospectively registered pregnancies involved exposure to bupropion.36 Among this group, 483 were lost to follow up and 134 pregnancies were pending. Among the remaining 833 prospectively reported pregnancy outcomes, the Bupropion Pregnancy Registry Advisory Committee found no evidence of increased risk of birth defects; however, they warned that the relatively small sample size and large percentage of patients lost to follow up make it impossible to draw definitive conclusions about the possible teratogenic risk of bupropion. These findings are consistent with other studies including a prospective study of 136 women exposed to bupropion in the first trimester of pregnancy that found no increased risk of major malformations,37 and a claims-based, retrospective cohort study using the United Healthcare database which found no consistent pattern of birth defects associated with prenatal exposure to bupropion.38 To register pregnancies in women on bupropion, clinicians can call the GlaxoSmithKline Pregnancy Registry at 800-336-2176.

Perinatal Effects Following Late-Pregnancy Antidepressant Exposure
In utero exposure to antidepressants has been associated with transient symptoms of possible medication withdrawal or toxicity in neonates. These symptoms have been described with many SSRIs and serotonin norepinephrine reuptake inhibitors (SNRIs) and include irritability, tremulousness, insomnia, poor feeding, temperature dysregulation, increased or decreased muscle tone, and/or respiratory distress.39,40 Using a formal screening tool, one study found that 30% of newborns exposed to SSRIs in late pregnancy developed neonatal abstinence symptoms.41 Although these neonatal syndromes are generally transient and not life threatening, in 2004 antidepressant manufacturers and the FDA decided to modify their drug labeling to include a recommendation to consider tapering antidepressants in the last part of pregnancy. For women with recurrent depression or anxiety conditions, this strategy may not be prudent since it would withdraw treatment just as patients are transitioning to the postpartum, a time of increased risk for affective instability.

Another recent study highlighted another concerning finding regarding antidepressant use in pregnancy. In 2006, Chambers and colleagues42 published a study that found a possible association between SSRI exposure after 20 weeks’ gestation and persistent pulmonary hypertension of the newborn (PPHN), a serious and rare condition with a baseline rate of 1–2 out of 1,000 live births and mortality rate of 10% to 20%. Although this study was retrospective in design and involved only a small number of affected infants, its findings are concerning and highlight the uncertainty that patients and providers must assume in choosing to use antidepressants or any psychotropic medications during pregnancy.

Long-Term Neurobehavioral Effects

It remains to be seen whether in utero antidepressant exposure is associated with any long-term neurologic or behavioral effects. In a cohort of children 4–5 years of age exposed in utero to SSRIs, levels of internalizing and externalizing behavior did not differ significantly between children who were (N=22) or were not (N=14) exposed prenatally to SSRIs.43,44 In another study, Nulman and colleagues45 compared a cohort of mother-child pairs exposed throughout gestation to TCAs (N=46) or fluoxetine (N=40) to a nondepressed comparison group (N=36). Children between 15 and 71 months of age were assessed and compared in terms of intelligence quotient (IQ), language, behavior, and temperament, with adjustment for severity of maternal depression, maternal IQ, socioeconomic status, maternal smoking, and alcohol history. Children exposed in utero to TCAs or fluoxetine were found to have no difference in temperament, language, or cognitive development compared to nonexposed children. In fact, it was exposure to maternal depression and not medication itself that was associated with less language and cognitive achievement.


Mood Stabilizers

Lithium use during pregnancy has been associated with perinatal toxicity, including case reports of lethargy, hypotonia, cyanosis, respiratory distress, and diabetes insipidus.46 In addition, lithium use during the first trimester has been associated with an increased risk of a serious congenital heart defect known as Ebstein’s anomaly and occurs in approximately 1 out of 1,000 live births. For women with moderate-to-severe bipolar disorder with recurrent episodes of mania or depression, the relatively small risk of these adverse pregnancy outcomes may be far overshadowed by the much greater risk of relapse. Maintenance lithium treatment for these patients during pregnancy may be the most prudent treatment option. For women with less severe bipolar disorder who have had significant periods of stability, slowly tapering off lithium and reintroducing it after the first trimester or just after delivery may help lower medication exposure while also minimizing risk of relapse during the postpartum, a time of known high risk in women with bipolar disorder.47

Lamotrigine is another highly effective mood stabilizer, particularly for bipolar depression. As with bupropion, the manufacturer of lamotrigine took a very proactive step in 1992 and established a pregnancy registry to assess the reproductive safety risk of lamotrigine. In January 2007, the Lamotrigine Pregnancy Registry reported its most recent update spanning the time period from September 1992 through September 2006 during which there were 1,539 prospectively registered pregnancies involving lamotrigine exposure.48 Among this group 908 outcomes involved first-trimester lamotrigine exposure with 26 (2.9%) major birth defects. This is similar to the baseline frequency of malformations reported in cohorts of women using antiepileptic monotherapy. Of note, in the 133 pregnancy outcomes involving polytherapy with lamotrigine and valproate, plus or minus another anticonvulsant, there was an alarming 11.3% rate of major congenital defects. Overall, the findings for lamotrigine monotherapy have been reassuring and consistent with other reports.49 However, the North American Antiepileptic Drug Pregnancy Registry recently found that infants who are exposed to lamotrigine as monotherapy during pregnancy have a much higher risk of oral cleft defects.50 In this study, 564 children exposed to lamotrigine monotherapy had a prevalence rate of major malformations of 2.7%; however, five infants had oral cleft lip/palate, yielding a prevalence rate of 9 out of 1,000 births compared to a baseline prevalence of 0.5–2.0 per 1,000 in unexposed infants. A larger sample size is needed to confirm this finding. Even if future prospective studies also find this association between first-trimester lamotrigine use and oral cleft lip and palate, the overall risk appears to be low and may be overshadowed by the high risk of recurrent illness in women with moderate-to-severe forms of bipolar disorder. Clinicians can help expand the available database of lamotrigine exposures by registering all pregnant patients on lamotrigine with the Lamotrigine Pregnancy Registry by calling 800-336-2176, or having patients enroll themselves in the North American AED Pregnancy Registry by calling 888-233-2334.

Valproic acid and carbamazepine have well-established risks of neural tube defects of 1.0% to 5.0% and 0.5% to 1.0%, respectively.51,52 In utero valproate exposure has also been associated with a higher frequency of major anomalies53,54 as well as neurodevelopmental delay in exposed children.55,56 Neonatal complications associated with valproate include irritability, jitteriness, and feeding problems.57 Liver toxicity has also been described following valproate and carbamazepine use during pregnancy.58,59 Less is known about the reproductive safety of newer anticonvulsants such as gabapentin, oxcarbazepine, and topiramate. While these newer anticonvulsants would not be the first choice in pregnancy, they may be indicated for pregnant women with refractory bipolar illness and a history of good response to these medications. Providers should encourage pregnant women who elect to continue any mood stabilizer to take high-dose folate (4 mg/day) for the theoretical benefit of reducing risk of neural tube defects. In addition, pregnant women should undergo a second trimester Level II ultrasound to screen for major congenital anomalies.



Untreated psychosis can be harmful in pregnancy, as demonstrated by one study that found a two-fold increase in adverse pregnancy outcomes (eg, stillbirth, preterm delivery, low birth weight) in women diagnosed with schizophrenia during pregnancy.60 Psychotic symptoms can also lead to disorganized behavior, increased paranoia, avoidance of prenatal care, increased drug and alcohol use, and other high-risk behaviors. First-generation antipsychotics with high and midpotency (eg, haloperidol, perphenazine) are generally considered to be the antipsychotics of choice during pregnancy since they have not consistently demonstrated teratogenic risk.57 Low-potency phenothiazines (eg, chlorpromazine) have shown increased risk of congenital malformations and should be avoided.61 Case reports of neonatal toxicity following in utero exposure to typical antipsychotics have been described and include motor restlessness, tremor, feeding difficulties, increased muscle tone, and abnormal motor movements.62,63

Studies of the reproductive safety of atypical antipsychotics have been limited to small case reports and case series. Olanzapine was not associated with major malformations in a manufacturer-sponsored study of 23 prospectively ascertained pregnancies.64 McKenna and colleagues65 reported on 151 pregnancy outcomes following exposure to olanzapine (N=60), risperidone (N=49), quetiapine (N=36), and clozapine (N=6). Among this total group there were 110 (approximately 73%) live births, 22 (approximately 15%) spontaneous abortions, 15 (approximately 10%) therapeutic abortions, four (approximately 3%) stillbirths, and one (approximately 1%) infant with major malformations. There was no statistically significant difference in pregnancy outcomes between this group and a comparison group of pregnant women except for a higher rate of low birth weight in exposed babies (approximately 10% versus 2%). Given the small sample sizes and the notable differences between the exposed and comparison group (eg, differences in socioeconomic background and rates of elective abortion), no definitive conclusions can be drawn about the reproductive safety of atypical antipsychotics in general. However, one theoretical concern with these medications is their propensity to cause significant weight gain that in pregnancy can lead to or exacerbate pre-existing hypertension and diabetes. This must be considered along with the potentially hazardous consequences of discontinuing an antipsychotic that is effectively treating a pregnant woman with a history of psychotic symptoms.

Based on the available evidence, experts have supported the use of typical antipsychotics for acute treatment of mania or psychosis during pregnancy.57 Others have suggested that the risk with high-potency typical antipsychotics may be less than the risks associated with available mood stabilizers,61 and that they offer a reasonable treatment alternative for women with bipolar disorder with recurrent symptoms during pregnancy.57



Benzodiazepine use during pregnancy has been associated with case reports of perinatal toxicity, including temperature dysregulation, apnea, depressed APGAR scores, hypotonia, and poor feeding. In addition, early studies revealed an elevated risk of oral cleft palate defects compared to the baseline risk in the general population.66 However, more recent studies have shown that the overall risk of cleft lip and palate with benzodiazepine use in pregnancy is likely quite low.67,68 In considering the risks and benefits of benzodiazepines, clinicians should also consider the risks of untreated insomnia and anxiety in pregnancy, which may lead to physiologic effects as well as diminished self care, worsening mood, and impaired functioning. Given the consequences of untreated psychiatric symptoms and the limited and controversial risks associated with benzodiazepine use, some women with overwhelming anxiety symptoms or sleep disturbance may find that the benefits outweigh any theoretical risks.


Pharmacologic Considerations in the Postpartum

Women can experience a range of emotional and psychological symptoms following delivery. Heightened emotional sensitivity, anxiety, and sleep disturbance can affect 80% of women after delivery in the form of the postpartum blues, a normal and self-limited syndrome that usually begins 2–3 days postpartum and resolves in 7–14 days. Approximately 10% to 20% of all postpartum women experience a more serious condition called postpartum depression (PPD). The Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition–Text Revision,69 classifies PPD as MDD that occurs within 4 weeks postpartum; however, many new mothers develop symptoms more than 2–3 months after delivery. Postpartum anxiety syndromes are also extremely common and can include panic attacks, intense anxiety about not getting enough sleep, obsessive worry about the baby’s health or safety, and intrusive thoughts or mental images of hurting the baby.70 New mothers with intrusive thoughts are often ashamed of these ego-dystonic images and may develop behaviors to diffuse some of their anxiety and fear, such as avoiding sharp objects or compulsively checking their infants’ breathing. Postpartum patients can be reassured that though intrusive thoughts are common, violently acting them out is very uncommon. Postpartum psychosis is a serious and rare condition that occurs in 1–2 of every 1,000 postpartum women. Postpartum psychosis can begin acutely within the first 48–72 hours after delivery and may include delirium, agitation, irritability, and mood lability. Psychosis can also occur in new mothers with a history of a chronic psychotic disorder or as part of a major depressive or manic episode. Psychotic symptoms in any postpartum woman require immediate intervention in order to protect both mother and infant from harm.


Evaluation of Postpartum Women with Psychiatric Risk Factors or Symptoms

Evaluation of postpartum psychiatric symptoms includes assessment for medical conditions such as anemia or thyroid disorders that can contribute to low mood, anxiety, fatigue, and sleep disturbance. Systematic screening for psychosocial risk factors as well as current symptoms facilitates prevention, detection, and treatment of postpartum psychiatric syndromes. The Edinburgh Postnatal Depression Scale (EPDS) is a widely recommended, cost-effective means of screening for PPD.71 The EPDS is a 10-item self-rating scale that has been validated in Spanish and English and asks about depressive symptoms in the preceding week, including crying spells, decreased interest and pleasure, increased guilt, anxiety, sleeping problems, and thoughts of self harm.72 To improve prevention interventions, clinicians can also screen for risk factors both during and after pregnancy. These risk factors include a personal history of mood or anxiety disorder, psychiatric symptoms during pregnancy, limited social support, interpersonal conflicts, and negative life events during and after pregnancy.73-75 In addition, a history of extreme premenstrual irritability may be a possible marker for increased vulnerability during times of hormone fluctuation, such as the dramatic hormone shifts after delivery.76



Despite the high prevalence of postpartum psychiatric illness, there are only a limited number of medication treatment studies. In small open studies, venlafaxine (N=15),77 bupropion (N=8),78 fluvoxamine (N=6),79 and sertraline (N=26),80 have shown efficacy in treating PPD. Paroxetine alone (N=16) or in combination with 12 sessions of cognitive-behavioral therapy (CBT; N=19) improved mood and anxiety symptoms in a group of women with PPD and anxiety symptoms.81 In small randomized trials, sertraline has shown benefit in both preventing82 and treating83 PPD. In another controlled study of 61 women with PPD, fluoxetine was significantly more effective than placebo in treating PPD.84 Combining fluoxetine with six sessions of CBT did not produce additional improvement. Of note, 101 out of 188 eligible patients refused to enter the trial, mainly because of ambivalence about being randomized to take an antidepressant. Furthermore, the trial excluded nursing mothers as well as women with a history of severe or chronic depression. All these factors limit the generalizability of the findings and underscore the difficulty in carrying out randomized trials in this population. Clearly, more studies are needed with larger sample sizes and longer follow-up periods to compare different antidepressants and other psychosocial interventions.85

The choice of an antidepressant is largely guided by a patient’s depressive symptoms and past history of medication response. If a patient plans to breastfeed, clinicians must facilitate a careful risk-benefit discussion about taking psychotropic medication during lactation (see below). In addition, patients must consider the risks of untreated maternal depression which can lead to impaired mother-infant attachment.86-88 and higher risk of cognitive and behavioral problems.89,90 Women at increased risk for postpartum depression should consider initiating prophylactic antidepressants either in late pregnancy or early postpartum. Alternatively, women may elect a wait-and-see approach; however, patients, their loved ones, and psychiatrists should be vigilant for early signs of recurrence in order to institute prompt treatment.


Hormone Therapy

To temper the tremendous changes in estrogen and progesterone following birth, some studies have looked at hormone therapy. In one study, women with PPD who received transdermal estrogen had significantly lower mean EPDS scores at 4 weeks of treatment compared to placebo.91 However, in another study a single dose of synthetic progestogen administered within 48 hours after delivery was significantly more likely than placebo to be associated with increased depressive symptoms at 6 weeks postpartum.92 Given the limited data available, no specific recommendations can be made regarding the use of hormone therapy for PPD.93,94



All psychotropic medications are secreted into breast milk. Mothers on psychotropic medications require a thorough discussion of the risks and benefits of breastfeeding while taking medication to treat psychiatric symptoms.

A recent analysis of the available studies of antidepressants during lactation revealed that sertraline, paroxetine, and nortriptyline are the least likely to lead to accumulation in the infant.95 Other studies of TCAs and SSRIs have been reassuring and show no consistent association between any particular antidepressant and problems in nursing newborns. There have been isolated case reports of elevated infant levels and toxicity with breastfeeding mothers taking doxepin or fluoxetine.96 Little is known about the safety of other antidepressants during lactation, such as venlafaxine, bupropion, or mirtazapine.

Mood Stabilizers
For women with bipolar disorder, the choice to nurse while on a mood stabilizer is even less clear cut than with antidepressants. Lithium can quickly accumulate in the nursing infant and lead to levels exceeding 50% of the maternal level. Given this risk of lithium toxicity in the nursing infant, breastfeeding while on lithium is generally not recommended,97,98 however, a recent case series of 10 mother-infant pairs noted low infant lithium levels in nursing infants.99 The authors of this article encourage reassessment of general recommendations against lithium during breastfeeding and suggest that nursing on lithium may be appropriate for mothers with bipolar disorder who have healthy infants and who can reliably work with a pediatrician to monitor the infant and obtain infant lithium level, thyroid-stimulating hormone, blood urea nitrogen, and creatinine in the immediate postpartum, at 4–6 weeks postpartum, and then every 8–12 weeks thereafter.

Although not absolutely contraindicated in nursing mothers, valproate has been associated with infant anemia and thrombocytopenia, and carbamazepine has been associated with transient hepatic dysfunction.97 Mothers should be informed of signs of hepatic dysfunction or anemia, such as sedation or poor feeding. Among the few available reports, breastfed infants of mothers on lamotrigine have been shown to have serum levels that are approximately 30% of the mothers’ level.100,101 Though no adverse effects were noted in these infants, the severe life-threatening rash associated with lamotrigine in children and adults may also be a concern for infants exposed to lamotrigine through breastmilk.97

In general, clinicians should advise nursing women on psychotropic medications to monitor infants for behavioral changes, such as excessive sedation, jitteriness, or inconsolable crying. Infants who develop these symptoms should be evaluated by their pediatricians for possible drug toxicity. In the meantime, mothers can consider temporarily pumping and storing/discarding their breastmilk and using formula to see if their infants’ symptoms resolve. In infants who are preterm or have any medical problems, mothers on psychotropic medication could also consider pumping and storing/discarding breastmilk and introducing nursing later when the infant is healthy and can presumably metabolize medication more efficiently.


Supportive Interventions

Although a detailed discussion is beyond the scope of this article, it is important to note that nonpharmacologic interventions such as education, support services, and psychotherapy can be invaluable alternatives or adjuncts to medication for women with perinatal psychiatric symptoms. Establishing a therapeutic alliance; exploring feelings about the pregnancy and motherhood; and taking inventory of strengths, supports, and stressors can help target interventions. Studies have demonstrated that depressed pregnant and postpartum women can also benefit greatly from interpersonal psychotherapy, a time-limited therapy focused on certain areas of particular relevance to pregnant women such as role transitions and interpersonal disputes.102-105 Supportive therapy and CBT have also been shown anecdotally to help with perinatal depression and anxiety as well as referrals to mothers’ groups, childcare resources, and financial assistance agencies. Clinicians can also refer women to Postpartum Support International, an organization with many local chapters that offers support and resources for women with postpartum psychiatric illness.



The perinatal period can be a high-risk time for significant psychiatric symptoms. In treating pregnant and postpartum patients, clinicians must adopt an individualized treatment approach that incorporates an up-to-date discussion of the risks and benefits of medication and psychosocial interventions as well as the impact of untreated psychiatric symptoms on mothers and families. Several recent studies have highlighted the fact that reproductive psychiatry is a dynamic field with treatment recommendations that continue to evolve and currently can only serve as general guides rather than absolute mandates. Ultimately, clinicians must help patients reflect on these guidelines in the context of their own experience of illness as well as their own perception of the risks and benefits of different treatment options. PP



1. Wisner KL, Chambers C, Sit DK. Postpartum depression: a major public health problem. JAMA. 2006;296(21):2616-2618.
2. Stowe ZN, Hostetter AL, Newport DJ. The onset of postpartum depression: implications for clinical screening in obstetrical and primary care. Am J Obstet Gynecol. 2005;192(2):522-526.
3. Evans J, Heron J, Francomb H, Oke S, Golding J. Cohort study of depressed mood during pregnancy and after childbirth. BMJ. 2001;323(7307):257-260.
4. Andersson L, Sundstrom-Poromaa I, Wulff M, Astrom M, Bixo M. Depression and anxiety during pregnancy and six months postpartum: a follow-up study. Acta Obstet Gynecol Scand. 2006;85(8):937-944.
5. Eberhard-Gran M, Tambs K, Opjordsmoen S, Skrondal A, Eskild A. Depression during pregnancy and after delivery: a repeated measurement study. J Psychosom Obstet Gynaecol. 2004;25(1):15-21.
6. Gavin NI, Gaynes BN, Lohr KN, Meltzer-Brody S, Gartlehner G, Swinson T. Perinatal depression: a systematic review of prevalence and incidence. Obstet Gynecol. 2005;106(5 Pt 1):1071-1083.
7. Marcus SM, Flynn HA, Blow FC, Barry KL. Depressive symptoms among pregnant women screened in obstetric settings. J Womens Health (Larchmt). 2003;12(4):373-380.
8. Hendrick V, Altshuler LL, Suri R. Hormonal changes in the postpartum and implications for postpartum depression. Psychosomatics. 1998;39(2):93-101.
9. Parry BL, Sorenson DL, Meliska CJ, et al. Hormonal basis of mood and postpartum disorders. Curr Womens Health Rep. 2003;3(3):230-235.
10. Kammerer M, Taylor A, Glover V. The HPA axis and perinatal depression: a hypothesis. Arch Womens Ment Health. 2003;9(4):187-196.
11. Bloch M, Schmidt PJ, Danaceau M, Murphy J, Nieman L, Rubinow DR. Effects of gonadal steroids in women with a history of postpartum depression. Am J Psychiatry. 2000;157(6):924-930.
12. Cohen LS, Altshuler LL, Harlow BL, et al. Relapse of major depression during pregnancy in women who maintain or discontinue antidepressant treatment. JAMA. 2006;295(5):499-507.
13. Bonari L, Koren G, Einarson TR, Jasper JD, Taddio A, Einarson A. Use of antidepressants by pregnant women: evaluation of perception of risk, efficacy of evidence based counseling and determinants of decision making. Arch Womens Men Health. 2005;8(4):214-220.
14. Koren G. The way women perceive teratogenic risk. Can J Clin Pharmacol. 2007;14(1):e10-6.
15. Koren G, Bologa M, Long D, Feldman Y, Shear NH. Perception of teratogenic risk by pregnant women exposed to drugs and chemicals during the first trimester. Am J Obstet Gynecol. 1989;160(5 Pt 1):1190-1194.
16. Zhu SH, Valbo A. Depression and smoking during pregnancy. Addic Behav. 2002;27(4):649-658.
17. Hanna EZ, Faden VB, Dufour MC. The motivational correlates of drinking, smoking, and illicit drug use during pregnancy. J Substance Abuse. 1994;6(2):155-167.
18. Kurki T, Hiilesmaa V, Raitasalo R, Mattila H, Ylikorkala O. Depression and anxiety in early pregnancy and risk for preeclampsia. Obstet Gynecol. 2000;95(4):487-490.
19. Chung TK, Lau TK, Yip AS, Chiu HF, Lee DT. Antepartum depressive symptomatology is associated with adverse obstetric and neonatal outcomes. Psychosom Med. 2001;63(5):830-834.
20. Beck CT. Postpartum depression predictors inventory–revised. Adv Neonatal Care. 2003;3(1):47-48.
21. Stowe ZN, Hostetter AL, Newport DJ. The onset of postpartum depression: implications for clinical screening in obstetrical and primary care. Am J Obstet Gynecol. 2005;192(2):522-526.
22. Spitzer RL, Williams JB, Kroenke K, Hornyak R, McMurray J. Validity and utility of the PRIME-MD patient health questionnaire in assessment of 3000 obstetric/gynecologic patients: the PRIME-MD Patient Health Questionnaire Obstetric-Gynecology Study. Am J Obstet Gynecol. 2002;183(3):759-769.
23. Robertson E, Grace S, Wallington T, Stewart DE. Antenatal risk factors for postpartum depression: a synthesis of recent literature. Gen Hosp Psychiatry. 2004;26(4):289-295.
24. Altshuler LL, Hendrick V, Cohen LS. An update on mood and anxiety disorders during pregnancy and the postpartum period. Prim Care Companion J Clin Psychiatry. 2000;2(6):217-222.
25. Stowe ZN, Calhoun K, Ramsey C, Sadek N, Newport DJ. Mood disorders during pregnancy and lactation: defining issues of exposure and treatment. CNS Spectr. 2001;6(2):150-166.
26. Hendrick V, Smith LM, Suri R, Hwang S, Haynes D, Altshuler L. Birth outcomes after prenatal exposure to antidepressant medication. Am J Obstet Gynecol. 2003;188(3):812-815.
27.  Malm H, Klaukka T, Neuvonen PJ. Risks associated with selective serotonin reuptake inhibitors in pregnancy. Obstet Gynecol. 2005;106(6):1289-1296.
28. Ericson A, Kallen B, Wiholm B. Delivery outcome after the use of antidepressants in early pregnancy. Eur J Clin Pharmacol. 1999;55(7):503-508.
29. Kulin NA, Pastuszak A, Sage S, et al. Pregnancy outcome following maternal use of the new selective serotonin reuptake inhibitors: a prospective controlled multicenter study. JAMA. 1998;279(8):609-610.
30. Einarson TR, Einarson A. Newer antidepressants in pregnancy and rates of major malformations: a meta-analysis of prospective comparative studies. Pharmacoepidemiol Drug Saf. 2005;14(12):823-827.
31. December 08, 2005. GSK changes Paroxetine Prescribing Information. Available at: www.gsk.com/media/archive.htm#nolink. Accessed August 27, 2007.
32. ACOG Committee on Obstetric Practice. ACOG Committee Opinion No. 354: Treatment with selective serotonin reuptake inhibitors during pregnancy. Obstet Gynecol. 2006;108(6):1601-1603.
33. Einarson A, Fatoye B, Sarkar M, et al. Pregnancy outcome following gestational exposure to venlafaxine: a multicenter prospective controlled study. Am J Psychiatry. 2001;158(10):1728-1730.
34. Djulus J, Koren G, Einarson TR, et al. Exposure to mirtazapine during pregnancy: a prospective, comparative study of birth outcomes. J Clin Psychiatry. 2006;67(8):1280-1284.
35. Einarson A, Bonari L, Voyer-Lavigne S, et al. A multicentre prospective controlled study to determine the safety of trazadone and nefazadone use during pregnancy. Can J Psychiatry. 2003;48(2):106-110.
36. GlaxoSmithKline GlaxoSmithKline Pregnancy Registries. Bupropion Pregnancy Registry Interim Report. Available at: http://pregnancyregistry.gsk.com/documents/bup_report_spring2007.pdf. Accessed August 27, 2007.
37. Chun-Fai-Chan B, Koren G, Fayez I, et al. Pregnancy outcome of women exposed to bupropion during pregnancy: a prospective comparative study. Am J Obstet Gynecol. 2005;192(3):932-936.
38. Cole JA, Modell JG, Haight BR, Cosmatos IS, Stoler JM, Walker AM. Bupropion in pregnancy and the prevalence of congenital malformations. Pharmacoepidemiol and Drug Saf. 2007;16(5):474-484.
39. Moses-Kolko EL, Bogen D, Perel J, et al. Neonatal signs after late in utero exposure to serotonin reuptake inhibitors: literature review and implications for clinical applications. JAMA. 2005;293(19):2372-2383.
40. Kallen B. Neonate characteristics after maternal use of antidepressants late in pregnancy. Arch Pediatr Adolesc Med. 2004;158(4):312-316.
41. Levinson-Castiel R, Merlob P, Linder N, Sirota L, Klinger G. Neonatal abstinence syndrome after in utero exposure to selective serotonin reuptake inhibitors in term infants. Arch Pediatr Adolesc Med. 2006;160(2):173-176.
42. Chambers CD, Hernandez-Diaz S, Van Marter LJ, et al. Selective serotonin-reuptake inhibitors and risk of persistent pulmonary hypertension of the newborn. N Eng J Med. 2006;354(6):579-587.
43. Misri S, Reebye P, Kendrick K, et al. Internalizing behaviors in 4-year-old children exposed in utero to psychotropic medications. Am J Psychiatry. 2006;163(6):1026-1032.
44. Oberlander TF, Reebye P, Misri S, Papsdorf M, Kim J, Grunau RE. Externalizing and attentional behaviors in children of depressed mothers treated with a selective serotonin reuptake inhibitor antidepressant during pregnancy. Arch Pediatr Adolesc Med. 2007;161(1):22-29.
45. Nulman I, Rovet J, Stewart DE, et al. Child development following exposure to tricyclic antidepressants or fluoxetine throughout fetal life: a prospective, controlled study. Am J Psychiatry. 2002;159(11):1889-1895
46. Newport DJ, Viguera AC, Beach AJ, Ritchie JC, Cohen LS, Stowe ZN. Lithium placental passage and obstetrical outcome: implications for clinical management during late pregnancy. Am J Psychiatry. 2005;162(11):2162-2170.
47. Viguera AC, Nonacs R, Cohen LS, Tondo L, Murray A, Baldessarini RJ. Risk of recurrence of bipolar disorder in pregnant and nonpregnant women after discontinuing lithium maintenance. Am J Psychiatry. 2000;157(2):179-184.
48. The Lamotrigine Pregnancy Registry. Interim Report. 1 September 1992 through 31 March 2007. Available at: http://pregnancyregistry.gsk.com/documents/lam_report_spring2007.pdf. Accessed September 3, 2007.
49. Cunnington M, Tennis P, International lamotrigine pregnancy registry scientific advisory committee. Lamotrigine and the risk of malformations in pregnancy. Neurology. 2005;64(6):955-960.
50. Holmes LB, Wyszynski DF, Baldwin EJ, et al. Increased risk for non-syndromic cleft palate among infants exposed to lamotrigine during pregnancy. Birth Defect Research Part A: Clinical and Molecular Teratology. 2006;76(5):318.
51. Rosa FW. Spina bifida in infants of women treated with carbamazepine during pregnancy. N Engl J Med. 1991;324:674-677.
52. Perucca E. Birth Defects after prenatal exposure to antiepileptic drugs. Lancet Neurol. 2005;4(11):781-786.
53. Kaplan PW. Reproductive health effects and teratogenicity of antiepileptic drugs. Neurology. 2004;63(10 suppl 4):S13-S23.
54. Kini U, Adab N, Vinten J, Fryer A, Clayton-Smith J; Liverpool and Manchester Neurodevelopmental Study Group. Dysmorphic features: an important clue to the diagnosis and severity of fetal anticonvulsant syndromes. Arch Dis Child Fetal Neonatal Ed. 2006;91(2):F90-95.
55. Vinten J, Adab N, Kini U, et al. Neuropsychological effects of exposure to anticonvulsant medication in utero. Neurology. 2005;64(6):949-954.
56. Eriksson K, Viinikainen K, Monkkonen A, et al. Children exposed to valproate in utero: population based evaluation of risks and confounding factors for long-term neurocognitive development. Epilepsy Res. 2005;65(3):189-200.
57. Yonkers KA, Wisner KL, Stowe Z, et al. Management of bipolar disorder during pregnancy and the postpartum period. Am J Psychiatry. 2004;161(4):608-620.
58. Frey B, Schubiger G, Musy JP. Transient cholestatic hepatitis in a neonate associated with carbamazepine exposure during pregnancy and breastfeeding. Eur J Pediatr. 1990;150(2):136-138.
59. Felding I, Rane A. Congenital liver damage after treatment of mother with valproic acid and phenytoin. Acta Paediatr Scand. 1984;73(4):565-568.
60. Nilsson E, Lichtenstein P, Cnattingius S, Murray RM, Hultman CM. Women with schizophrenia: pregnancy outcome and infant death among their offspring. Schizophre Res. 2002;58(2-3):221-229.
61. Altshuler LL, Cohen L, Szuba MP, Burt VK, Gitlin M, Mintz J. Pharmacologic management of psychiatric illness during pregnancy: dilemmas and guidelines. Am J Psychiatry. 1996;153(5):592-606.
62. Tamer A, McKey R, Arias D, Worley L, Fogel BJ. Phenothiazine-induced extrapyramidal dysfunction in the neonate. J Pediatr. 1969;75(3):479-480.
63. Auerbach JG, Hans SL, Marcus J, Maeir S. Maternal psychotropic medication and neonatal behavior. Neurotoxicol Teratol. 1992;14(6):399-406.
64. Goldstein D. Olanzapine-exposed pregnancies and lactation: early experience. J Clin Psychopharmacol. 2000;20(4):399-403.
65. McKenna K, Koren G, Tetelbaum M, et al. Pregnancy outcome of women using atypical antipsychotic drugs: a prospective comparative study. J Clin Psychiatry. 2005;66(4):444-449.
66. Dolovich LR, Addis A, Vaillancourt JM, Power JD, Koren G, Einarson TR. Benzodiazepine use in pregnancy and major malformations or oral cleft: meta-analysis of cohort and case-control studies. Br Med J Clin Res Ed. 1998;317(7162):839-843.
67. Lin AE, Peller AJ, Westgate MN, Houde K, Franz A, Holmes LB. Clonazepam use in pregnancy and the risk of malformations. Birth Defects Res A Clin Mol Teratol. 2004;70(8):534-536.
68. Iqbal MM, Sobhan T, Ryals T. Effects of commonly used benzodiazepines on the fetus, the neonate, and the nursing infant. Psychiatr Serv. 2002;53(1):39-49.
69. Diagnostic and Statistical Manual of Mental Disorders. 4th ed, text rev. Washington, DC: American Psychiatric Association; 2000.
70. Brandes M, Soares CN, Cohen LS. Postpartum onset obsessive-compulsive disorder: diagnosis and management. Arch Womens Ment Health. 2004;7(2):99-110.
71. Wisner KL, Parry BL, Piontek CM. Clinical practice. Postpartum depression. N Engl J Med. 2002;347:194-199.
72. Cox JL, Holden JM, Sagovsky R. Detection of postnatal depression. Development of the 10-item Edinburgh Postnatal Depression Scale. Br J Psychiatry. 1987;150:782-786.
73. Robertson E, Grace S, Wallington T, Stewart DE. Antenatal risk factors for postpartum depression: A synthesis of recent literature. Gen Hosp Psychiatry. 2004;26(4):289-295.
74. O’Hara MW, Schlechte JA, Lewis DA, Varner MW. Controlled prospective study of postpartum mood disorders: psychological, environmental and hormonal variables. J Abnorm Psychol. 1991;100(1):63-73.
75. Stowe ZN, Nemeroff CB. Women at risk for postpartum-onset major depression. Am J Obstet Gynecol. 1995;173(2):639-645.
76. Stewart DE, Boydell KM. Psychologic distress during menopause: associations across the reproductive life cycle. Int J Psychiatry Med. 1993;23(2):157-162.
77. Cohen LS, Viguera AC, Bouffard SM, et al. Venlafaxine in the treatment of postpartum depression. J Clin Psychiatry. 2001;62(8):529-596.
78. Nonacs RM, Soares CN, Viguera AC, Pearson K, Poitras JR, Cohen LS. Bupropion SR for the treatment of postpartum depression: a pilot study. Int J Neuropsychopharmacol. 2005;8(3):445-449.
79. Suri R, Burt VK, Altshuler LL, Zuckerbrow-Miller J, Fairbanks L. Fluvoxamine for postpartum depression [letter]. Am J Psychiatry. 2001;158(10):1739-1740.
80. Stowe ZN, Casarella J, Landry J, Nemeroff CB. Sertraline in the treatment of women with postpartum onset major depression. Depression. 1995;3:49-55.
81. Misri S, Reebye P, Corral M, Milis L. The use of paroxetine and cognitive-behavioral therapy in postpartum depression and anxiety: a randomized controlled trial. J Clin Psychiatry. 2004;65(9):1236-1241.
82. Wisner KL, Perel JM, Peindl KS, et al. Prevention of postpartum depression: a pilot randomized clinical trial. Am J Psychiatry. 2004;161(7):1290-1292.
83. Wisner KL, Hanusa BH, Perel JM, et al. Postpartum depression: a randomized trial of sertraline versus nortriptyline. J Clin Psychopharmacol. 2006;26(4):353-360.
84. Appleby L, Warner R, Whitton A, et al. A controlled study of fluoxetine and cognitive-behavioural counseling in the treatment of postnatal depression. BMJ. 1997;314(7085):932-936.
85. Hoffbrand S, Howard L, Crawley H. Antidepressant treatment for post-natal depression. Cochrane Database Syst Rev. 2001;(2):CD002018.
86. Brockington IF, Oates J, George S, et al. A screening questionnaire for mother-infant bonding disorders. Arch Womens Ment Health. 2001;3(4):133-140.
87. Moehler E, Brunner R, Wiebel A, Reck C, Resch F. Maternal depressive symptoms in the postnatal period are associated with long-term impairment of mother-child bonding. Arch Womens Ment Health. 2006;9(5):273-278.
88. Righetti-Veltema M, Conne-Perreard E, Bousquet A, Manzano J. Postpartum depression and mother-infant relationship at 3 months old. J Affect Disord. 2002;70(3):291-306.
89. Brennan PA, Hammen C, Andersen MJ, et al. Chronicity, severity, and timing of maternal depressive symptoms: relationships with child outcomes at age 5. Dev Psychol. 2000;36(6):759-766.
90. Murray L, Hipwell A, Hooper R, Stein A, Cooper P. The cognitive development of 5-year-old children of postnatally depressed mothers. J Child Psychol Psychiatry. 1996;37(8):927-935.
91. Gregoire AJ, Kumar R, Everitt B, Henderson AF, Studd JW. Transdermal estrogen for the treatment of severe postnatal depression. Lancet. 1996;347(9006):930-933.
92. Lawrie T, Hofmeyr G, De Jager M, et al. A double-blind randomised placebo controlled trial of postnatal norethisterone enanthate: the effect on postnatal depression and serum hormones. Br J Obstet Gynaecology. 1998;105(10):1082-1090.
93. Lawrie TA, Herxheimer A, Dalton K. Oestrogens and progestins for preventing and treating postpartum depression. Cochrane Database Syst Rev. 2000;(2):CD001690.
94. Gentile S. The role of estrogen therapy in postpartum psychiatric disorders: an update. CNS Spectr. 2005;10(12):944-952.
95. Weissman AM, Levy BT, Hartz AJ, et al. Pooled analysis of antidepressant levels in lactating mothers, breast milk, and nursing infants. Am J Psychiatry. 2004;161(6):1066-1078.
96. Lester BM, Cucca J, Andreozzi L, Flanagan P, Oh W. Possible association between fluoxetine hydrochloride and colic in an infant. J Am Acad Child Adolesc Psychiatry. 1993;32(6):1253-1255.
97. Chaudron LH, Jefferson JW. Mood stabilizers during breastfeeding: a review. J Clin Psychiatry. 2000;61(2):79-90.
98. American Academy of Pediatrics Committee on Drugs. Transfer of drugs and chemicals into human milk. Pediatrics. 2001;108(3):776-789.
99. Viguera AC, Newport J, Ritchie J, et al. Lithium in breast milk and nursing infants: clinical implications. Am J Psychiatry. 2007;164(2):342-345.
100. Tomson T, Ohman I, Vitols S. Lamotrigine in pregnancy and lactation: a case report. Epilepsia. 1997;38(9):1039-1041.
101. Ohman I, Tomson T, Vitols S. Lamotrigine levels in plasma and breast milk in nursing women and their infants. Epilepsia. 1998;39(suppl 2):21.
102. O’Hara MW, Stuart S, Gorman LL, Wenzel A. Efficacy of interpersonal psychotherapy for postpartum depression. Arch Gen Psychiatry. 2000;57(11):1039-1045.
103. Zlotnick C, Johnson SL, Miller IW, Pearlstein T, Howard M. Postpartum depression in women receiving public assistance: pilot study of an interpersonal-therapy-oriented group intervention. Am J Psychiatry. 2001;158(4):638-640.
104. Spinelli MG, Endicott J. Controlled clinical trial of interpersonal psychotherapy versus parenting education program for depressed pregnant women. Am J Psychiatry. 2003;160(3):555-562.
105. Reay R, Fisher Y, Robertson M, et al. Group interpersonal psychotherapy for postnatal depression: a pilot study. Arch Womens Ment Health. 2006;9(1):31-39.


Levels of Active Brain Response to Facial-Expressed Emotion in Alcoholism

Many social and physiologic factors can lead recovering alcoholics to relapse, including anxiety disorders, anger, and social pressure. According to a new study, the brains of alcoholics may have a significantly diminished ability to recognize and process certain emotions such as fear or disgust, which may explain a propensity for alcohol relapse.

The study used functional magnetic resonance imaging (fMRI) to examine the blood-oxygen level dependent (BOLD) responses of 11 male alcoholics who were subjected to a variety of facial emotions in emotion-decoding tests. Lead author, Jasmin B. Salloum, PhD, of the National Institute on Alcohol Abuse and Alcoholism, has published other studies on fMRI imaging and emotional-recognition.

The alcoholic patients’ responses to facial expressions of happiness, sadness, anger, fear, and disgust were compared to the responses of 11 male, non-alcoholic controls. Alcoholic patients and non-alcoholic controls both identified the intensity of emotions accurately during a non-complicated emotional decoding task. The two groups, however, showed significantly different BOLD responses during a facial emotion recognition task. Brain activation for the alcoholic patients was generally lower than that of the non-alcoholic group and varied most during the decoding of facial-expressed emotions of fear and disgust. For example, the affective division of the anterior cingulate cortex (ACC), which has been associated with autonomic cognitive processing and emotional processing, showed decreased activity in alcoholic patients. Anger was the only facial-expressed emotion that caused significant ACC activation in alcoholic patients; those BOLD results were not significantly different than those of controls.

According to the authors of this study, the “blunted” ACC activation demonstrated by alcoholic patients may affect their ability to recognize dangerous situations, such as bars or social scenes where drinking behavior occurs. In turn, there may be an increased possibility of alcohol relapse. In addition, the failure to properly interpret facial-expressed emotions of others can lead to social disregard, or to indifference to interventions by clinicians or loved ones.

Several limitations to this study include a small sample size as well as comorbid pathology and substance use in the alcoholic patient group. The authors acknowledge that alcoholic patients with no serious comorbidities could show different results.

Funding for this research was provided by the National Institute on Alcohol Abuse and Alcoholism and the National Institute of Mental Health. (Alcohol Clin Exp Res. 2007;31(9):1490-504). —LS


Coronary Artery Disease is More Significant in Patients Suffering from Incident MDD

Coronary artery disease (CAD) is the world’s most common form of heart disease. It can result in heart attack and angina and can contribute to heart failure and arrhythmias. Recent research has found high rates of recurrent and incident major depressive disorder (MDD) in this patient population.

Karina Davidson, PhD, from Columbia University College of Physicians and Surgeons in New York City, and colleagues, investigated 88 patients to assess CAD severity in patients suffering from incident and recurrent MDD. The authors assessed history of depression and current depressive symptoms through patient interviews. They evaluated CAD severity via coronary angiography within 1 month of the aforementioned interviews. CAD severity was defined as: 0=no vessels significantly affected; 1=only one vessel affected; 2=two vessels affected; and 3=three vessels affected and/or left main obstruction.

Davidson and colleagues found that approximately 24% of acute coronary syndrome patients have comorbid MDD. Of these patients, approximately 67% had recurrent MDD and 33% had incident MDD. However, the authors found that the patients suffering from incident MDD had more severe types of CAD compared to patients with recurrent MDD patients. The mean CAD severity scores for incident MDD patients was 2.4 while the mean CAD severity scores for recurrent MDD patients was 1.6 (P=.043).

Due to these findings, Davidson and colleagues believe that patients with incident MDD should be differentiated from patients with recurrent MDD via two distinct subtypes. They propose using angiograms as a means of determining CAD severity. Future studies should explore the mechanisms underlying these differences. (J Psychiatr Res. In Press.) —CN


Effects of Mania and Depression on Functional Impairment and Disability

Bipolar disorder is a disabling condition that affects both moods and daily functioning. Gregory E. Simon, MD, MPH, at the Center for Health Studies in Seattle, Washington, and colleagues, reviewed data from 441 patients in a randomized trial of a care management and psychoeducational intervention in order to measure the relationship between changes in mood symptoms and changes in functioning in patients treated for bipolar disorder.

Symptom severity was assessed using the Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition, structured clinical interview, and functional status was measured with the social subscale of the 36-item Short-form Health Survey (SF-36). Patients were enrolled between August 1999 and October 2000, and follow-up data were collected until October 2001. Patients were assessed at baseline and every 3 months during the 12-month follow-up period.

Four measures of impairment and disability were utilized and included the SF-36 Role-Emotional score, SF-36 Social Function score, days unable to perform household duties, and days disabled from performing other activities (P<.001 for all comparisons). Severity of depressive symptoms were strongly and consistently associated with all measures, even after adjustment for co-ocurring manic symptoms. Compared to patients in remission, patients with a depressive episode scored approximately 60 points lower on the SF-36 scales and were unable to participate in daily activities for an additional 15 days. Severity of mania and hypomania symptoms showed significant association with all measures as well (P<.001 for all comparisons); however, associations were weaker after factoring in co-ocurring depressive symptoms. Compared with patients in remission, patients with mania or hypomania scored approximately 30 points lower on the SF-36 scale and reported 9 additional days of disability.

“Our study shows that symptoms of bipolar disorder have a significant impact on daily functioning and disability,” Dr. Simon said. “When symptoms of bipolar disorder improve, daily functioning improves as well.”

Dr. Simon added that this relationship was stronger and more consistent for symptoms of depression than for symptoms of mania or agitation.

The study was limited in that it was an observational study as opposed to a randomized trial.

“We can show that improvement in depression is followed by improvement in disability,” Dr. Simon said, “but we cannot absolutely prove that the improvement in depression caused the improvement in disability.”

Funding for this research was provided by the National Institute of Mental Health. (J Clin Psychiatry. 2007;68:1237-1245) —DC


Possible Genetic Link Found for Premenstrual Dysphoric Disorder

Premenstrual dysphoric disorder (PMDD) is more than just a severe case of premenstrual syndrome. Affecting approximately 8% of women, PMDD is characterized by a noticeably depressed mood, anxiety, and/or irritability in the week leading up to a woman’s menstrual cycle. In order to receive an official diagnosis, the symptoms must also be severe enough to cause impairment in daily life. Though not included in the main text of the Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition, Text Revision, PMDD is listed as an area requiring further study.

Though researchers have suspected that the underlying cause of PMDD is hormonal, a new study by Liang Huo, MD, at the National Institutes of Mental Health, and colleagues, was the first to link PMDD to an estrogen receptor gene. The study included 91 women with PMDD and 56 women without PMDD or any history of mood disorders linked to their menstrual cycle. The women had an average age of 40 years and were Caucasian with similar demographic and socioeconomic backgrounds. Blood samples from each woman were analyzed. Researchers looked specifically at numerous single nucleotide polymorphisms (SNPs) in three specific genes: ESR1, ESR2 and COMT.

ESR1 and ESR2 are two types of estrogen receptors while COMT is involved in the process of metabolizing estrogen. The results showed that four SNPs located on the ESR1 were significantly more prominent in the PMDD group. Even after removing 29 women with a history of major depressive disorder from the PMDD group, the results were still significant. David R. Rubinow, MD, a study co-author, noted that while this study suggests the involvement of specific genes linked to estrogen, the patients had normal levels of estrogen. The problem, therefore, may lie in the abnormal response to a normal level of hormones.

Huo and colleagues also note that the absence of the SNPs in the control group could be as telling as their presence in the PMDD group. The control group may have some sort of protective factor against menstrual-related mood disorders. However, further research is needed given the relatively small sample size of the current study.

Funding for this research was provided by the Intramural Research Program at the National Institute of Mental Health.

(Biol Psychiatry. 2007; June 26; Epub ahead of print.) —VJ

Psychiatric Dispatches is written by Dena Croog, Virginia Jackson, Christopher Naccari, and Lonnie Stoltzfoos.


Dr. Kornstein is professor of Psychiatry and Obstetrics/Gynecology, executive director of the Institute for Women’s Health, and executive director of the Mood Disorders Institute at Virginia Commonwealth University in Richmond.

Disclosure: Dr. Kornstein is on the advisory boards of or receives honoraria from Biovail, Bristol-Myers Squibb, Eli Lilly, Forest, Neurocrine, Pfizer, Sepracor, and Wyeth; and receives research support from AstraZeneca Boehringer-Ingelheim, Bristol-Myers Squibb, the Department of Health and Human Services, Eli Lilly, Forest, the National Institute of Mental Health,  Novartis, Sanofi-Synthelabo, Sepracor, and Wyeth.

Please direct all correspondence to: Susan G. Kornstein, MD, Dept of Psychiatry, Virginia Commonwealth University, PO Box 980710, Richmond, VA 23298-0710; Tel: 804-828-5637; Fax: 804-828-5644; E-mail: skornste@vcu.edu.


In March 2008, clinicians, researchers, policymakers, and advocates from across the world will convene in Melbourne, Australia for the 3 rd International Congress on Women’s Mental Health. This occasion will celebrate the remarkable progress that has been made in the field of women’s mental health since the last Congress in 2004. This progress includes advances in our understanding of sex and gender differences in mental illness and treatment efficacy, our knowledge of the etiology and treatment of psychiatric conditions related to the reporoductive cycle, and our appreciation of the role of not only biologic but social and cultural influences on women’s mental health and well-being.

This issue of Primary Psychiatry provides a sampling of topics from the field of women’s mental health across the life span. Nancy C. Raymond, MD, and Jennifer B. Beldon, MD, present an overview of the evaluation and management of eating disorders, which are among the most serious illnesses affecting young women. Diagnostic classification of anorexia nervosa, bulimia nervosa, and binge-eating disorder is reviewed. The authors underscore the importance of clinicians’ awareness of the warning signs of these disorders, since patients themselves are typically not forthcoming about their symptoms and often hide their abnormal eating and purging behaviors from family, friends, and healthcare professionals. In addition, patients with eating disorders may develop medical complications that can become life threatening. The authors highlight the effectiveness of a multidisciplinary approach to the treatment of eating disorders, including medication, psychotherapy, nutritional evaluation, and general medical monitoring.

One aspect of treating female patients that many clinicians find challenging is decisions regarding the use of psychotropic medications during pregnancy and the postpartum period. Recent studies suggesting possible associations of paroxetine and lamotrigine with congenital malformations have heightened the controversies in this arena. Helen G. Kim, MD, and Manasi Kolpe, MD, discuss the risks and benefits of pharmacologic treatment of perinatal women with mood and anxiety disorders. They emphasize that clinicians and patients must weigh in their decision-making process the risks of medications against the risks of untreated illness on the health and safety of both mother and infant.

Another area that has shown great progress in research over the last several years is psychiatric aspects of the menopausal transition. There is now improved clarity in defining the phases of the transition with the development of the Stages of Reproductive Aging Workshop criteria. In addition, several major studies have recently been published showing that the menopausal transition is associated with an increased risk for depression in women both with and without a previous history of mood disorder. Susan G. Kornstein, MD, and Larry Culpepper, MD, MPH, review the biologic, psychologic, and social factors that may contribute to depression during the menopausal transition along with strategies to evaluate and treat depression in midlife women.

The field of women’s mental health is advancing at a rapid pace. In addition to progress in research and clinical practice, there is an increasing recognition of the public health importance of women’s mental health. There is more discussion of the need to improve access to services, screening, and prevention, and there is a greater awareness of the impact of women’s health on children, families, and communities.

It is hoped that the articles in this issue will enable clinicians to better understand, evaluate, and manage mental health problems in their female patients. Perhaps some of the readers will be inspired to come to Melbourne to learn more. PP


Dr. Raymond is professor in the Departments of Psychiatry and Family Medicine and the director of the Center of Excellence in Women’s Health at the University of Minnesota Medical School in Minneapolis. Dr. Beldon is staff psychiatrist at Boynton Health Service at the University of Minnesota.

Disclosures: Dr. Raymond has received grant support from the Department of Health and Human Services and the National Institute of Mental Health. Dr. Beldon reports no affiliation with or financial interest in any organization that may pose a conflict of interest.

Please direct all correspondence to: Nancy C. Raymond, MD, Department of Psychiatry, University of Minnesota Medical School, F282/2A West, 2450 Riverside Ave, Minneapolis, MN 55454; Tel: 612-273-9808; Fax: 612-273-9779; E-mail: raymo002@umn.edu; Web site:  www.womenshealth.umn.edu.



Eating disorders are potentially severe and life threatening, affecting primarily adolescents and young women. Anorexia nervosa and bulimia nervosa have their onset in young adults; illness during this period can significantly affect both physical and psychological development. Since these women are fearful of having their disorders discovered, they are unlikely to seek psychiatric care. Therefore, it is important for primary care physicians, obstetricians, gynecologists, and even dentists or any clinician who works with young women to be able to identify, assess, refer, or treat young women with eating disorders. This article identifies common clinical presentations of women with eating disorders, how to assess these patients, and the best settings and methods for treatment.



Eating disorders are potentially severe or life threatening and predominately affect young women. These disorders can take an irreversible toll on both the physical and mental health of young women. Onset of bulimia nervosa and anorexia nervosa often occurs at a critical time of development, during the early adolescent to young adult years. Both physical and psychological development can be adversely affected. The onset of binge-eating disorder occurs somewhat later, but the obesity that is typical of the disorder can also have negative physical and psychological effects on those who suffer from it.

Approximately 33% of women with anorexia and 6% of those with bulimia are treated by mental health professionals.1 Therefore, it is essential that primary care physicians, obstetrician/gynecologists, and all other healthcare providers that offer care to young women screen their patients for eating disorders. They must be able to determine if an eating disorder is present, the severity of the problem, whether the patient is medically and psychiatrically stable, what they can provide in the way of treatment, and when to refer the patient to a comprehensive care team.


Prevalence, Incidence, and Mortality

It is estimated that the incidence rate of anorexia is 8 per 100,000 people and the rate for bulimia is 13 per 100,000.1 However, in young women the prevalence rates are 0.3% for anorexia and 1% for bulimia.1 Binge-eating disorder was found in approximately 2% of a community sample and 30% of subjects in a weight loss sample.2 Approximately 95% to 97% of those with anorexia and 80% to 85% of those with bulimia are women.3 However, only approximately 66% of those suffering from binge-eating disorder are women.2,4

A large number of women do not meet Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition (DSM-IV),3 criteria for an eating disorder but still have significant morbidity related to disordered eating behavior and cognitions. These women would be diagnosed with eating disorder not otherwise specified (ED NOS). Women in this category are thought to represent approximately 60% of patients presenting for clinical treatment of eating disorders compared to 14.5% presenting with anorexia and 25.5% with bulimia.5 The true prevalence of ED NOS is uncertain. Studies indicate that the morbidity and mortality associated with these subthreshold cases of eating disorders are, in many cases, equivalent to the morbidity associated with anorexia or bulimia; however, little research has been done on the prevalence, course of illness, symptomatology, or treatment of ED NOS.

Among patients with anorexia, mortality is estimated to be 0.56% per year.6 The suicide mortality rate for those with eating disorders is higher than for any of the other major psychiatric disorders including major depressive disorder (MDD) and bipolar disorder, and the rate is approximately three times greater than the suicide rate for those with opiate use, alcohol abuse, and schizophrenia.7 The mortality rate from all natural and unnatural causes for those with eating disorders is second only to opiate abuse, and those with eating disorders have a higher rate than those with alcohol abuse, schizophrenia, MDD, and bipolar disorder.8



The assessment of women with eating disorders consists of a physical exam as well as a psychiatric and mental-status exam. A simple set of screening questions has been suggested by Powers (Table 1).9 In addition to reviewing the specific criteria for eating disorders (Tables 2–4),10-12 it is important to obtain a weight, eating behavior, and nutritional history. Menstrual history is critical since absence of menses is one of the diagnostic criteria for anorexics and menstrual irregularity often occurs in women with bulimia. While obesity is not one of the diagnostic criteria for binge-eating disorder, it is common in women with binge-eating disorder and, therefore, they too can suffer from menstrual irregularity. Regardless of the particular eating disorder in question, all patients must be assessed for excessive use of diuretics, laxatives, diet pills, and ipecac. A dental history to assess for sensitive teeth is important since recurrent vomiting can cause thinning of tooth enamel. Patients often complain of feeling bloated after eating, a symptom which may be due to delayed gastric emptying. Patients may also complain of muscle weakness that can be caused by metabolic abnormalities and palpitations that can be caused by hypokalemia. Patients with severe anorexia often complain of cold intolerance.13





Women with anorexia and binge-eating disorder are predisposed to other psychiatric disorders and substance abuse disorders, so it is important to assess patients for these disorders. Psychiatric disorders that are often associated with eating disorders include depression, anxiety disorders, and obsessive-compulsive disorders.14 Even though the body image disturbance and odd beliefs about food and weight may seem to be of psychotic proportions, psychotic disorders are rare in patients with eating disorders.

The review of systems needs to include questions to help rule out other potential medical problems that can present with similar symptoms to eating disorders or complicate the treatment of the eating disorder, such as thyroid abnormalities, diabetes mellitus, cystic fibrosis, and inflammatory bowel disease.9

On physical exam, special emphasis should be placed on height, weight, and state of hydration. The latter can be evaluated by examining for orthostatic hypotension. Patients with anorexia are often bradycardic but can be tachycardiac due to dehydrations. Thorough cardiac and abdominal exams are indicated. An examination of the teeth to look for signs of erosion of the enamel secondary to vomiting is important. Hair loss, lanugo-like hair, and easy bruising due to possible thrombocytopenia may be seen in later stages of anorexia.13

Screening laboratory tests that should be administered include complete blood count and differential (to screen for leukopenia, anemia, and thrombocytopenia); basic electrolytes (to screen for hypokalemia, hyponatremia, and metabolic alkalosis with a high sodium bicarbonate level in patients who vomit or metabolic acidosis and low sodium bicarbonate levels in patients who abuse laxatives)15; calcium, magnesium, and phosphorous levels (to screen for deficiencies); and blood urea nitrogen and creatinine (to assess kidney function).13 If abnormalities in electrolytes are not present at baseline, these levels can become abnormal during refeeding. Particularly during refeeding, patients can experience profound or even lethal hypophosphatemia. Electrolytes often initially need to be monitored daily during the refeeding process. An electrocardiogram should be obtained because of the possibility of arrythmias and corrected Q-T interval prolongation in both anorexia and bulimia. In addition, especially in low-weight anorexic patients with long-standing disease, a bone-mineral density should be obtained to determine the level of osteopenia or osteoporosis.


Clinical Presentation

Anorexia Nervosa

By definition, patients with anorexia must be underweight. Their body weight must be less than what is expected for their height (the DSM-IV suggests <85% of normal weight) or they must be failing to make expected weight gain during a period of growth.10 According to the diagnostic criteria, they must have an intense fear of gaining weight or becoming fat even though they are underweight. Some patients may be knowledgeable enough about the diagnostic criteria to deny this symptom but behaviorally will show a great deal of difficulty in eating the food necessary to gain weight. Additionally, amenorrhea is a criterion for women who are postmenarchal and are not taking oral contraception or other hormones. Clinical presentation often includes an adamant denial of the illness. Patients often exhibit hyperactivity and like to be in constant motion as this burns calories. Often they prefer standing to sitting. Sleep disturbance is common. Irritability and social isolation are also very common. Patients often complain of constipation, though physical exam and x-rays may not indicate this as a problem. Instead, patients may be eating too little to have normal bowel movements. Cold intolerance, dizziness, and fainting spells are also quite common. On examination, patients will often present with low body temperature, bradycardia, hypotension, and dehydration; in end-stage disease, edema secondary to hypoproteinemia is not unusual. Patients will also be noted to have dry skin, brittle hair, brittle nails, and sometimes loss of hair.


Bulimia Nervosa

In contrast to anorexia, most women with bulimia are of normal weight. However, due to their episodes of binge eating and purging they may exhibit a number of physiologic abnormalities. Binge-eating episodes involve eating a large quantity of food in a short period of time with the sense that one cannot control what or how much one is eating. The most common method of purging to counteract the binge-eating behavior is self-induced vomiting, although this may be combined with or substituted by misuse of laxatives, diuretics, or enemas. Ipecac is a particularly dangerous way to induce vomiting as repeated use can cause severe cardiac damage. In the nonpurging subtype of bulimia, patients use behaviors such as fasting or excessive exercise to compensate for binge eating but do not regularly engage in the purging behavior. Other common symptoms are abdominal pain, constipation, and edema of the hands or feet. Important physical signs of bulimia are swollen parotid glands, erosion of the dental enamel on the inside of the teeth making the disorder diagnosable by dentists, and calluses on the back of the finger from the effects of gastric hydrochloride acid hitting the finger during self-induced vomiting. However, many patients with bulimia appear quite healthy during an initial inspection. Women with bulimia typically engage in the bulimic behaviors in secrecy. Because of this, the disorder can escape detection for many years. It is often 8–10 years after the initiation of the bulimic behavior before a patient seeks help and the diagnosis is made.


Binge-Eating Disorder

Patients with binge-eating disorder frequently present requesting diet advice and with weight concerns. Unless they are asked directly, patients will often not report that they have binge-eating behavior (ie, eating large quantities of food in a short period of time with a sense of loss of control). Since women with binge-eating disorder are more likely to fail traditional weight-loss programs and since they are more likely to regain weight after dieting, it is very important to screen all obese patients for binge-eating disorder.16 The majority of symptoms that patients with binge-eating disorder have are related to complications of obesity. Obese individuals with binge-eating disorder are more likely to have depression than non-binge eating obese women, so it is important to screen for other psychiatric disorders.



Anorexia Nervosa

The preferred approach to treating eating disorders is within the context of a multidisciplinary team, and this particularly holds true for anorexia. The gold standard is to involve the patient in a structured treatment program including a nutritionist, individual and family therapy, close medical monitoring by a physician, and, if needed, a psychiatrist. Ultimately, the best treatment for anorexia is restoring weight through adequate dietary intake. However, achieving this can be quite difficult and sometimes requires a combination of medications and intensive psychotherapy. A day treatment program or even inpatient hospitalization is often required to achieve weight gain. It is generally accepted that hospitalization is necessary in very low-weight patients who are experiencing cardiac sequelae (eg, hypotension, bradycardia, arrhythmia) or electrolyte abnormalities, and in patients who are deemed unlikely to improve without a closely monitored refeeding environment. Because of the deeply-held cognitive distortions about their appearance, and ambivalence about treatment, it is sometimes necessary to hospitalize patients against their wishes. Follow-up studies have reported that inpatient treatment is still beneficial in these cases, and patients often report that they later recognized the need for hospitalization.17

Few good medication trials for the treatment of anorexia exist, and many trials inadequately characterize the phase of the illness during which the patients were treated (ie, early or late), which can have an impact on outcome. Of note, the anorexic patient can be viewed as fitting into two subtypes. The first type is the acutely ill patient, suffering physical consequences of the disorder (eg, amenorrhea, osteoporosis) as well as other effects of starvation. The effects of starvation have been documented and include low mood, mental and physical apathy, difficulties with concentration, obsession with food, and anxiety.18 The treatment for this phase is weight restoration, which will often ameliorate or eliminate the comorbid mood or anxiety symptoms. There are no double-blind, placebo-controlled trials of any medication that demonstrate efficacy in promoting weight restoration in low-weight patients with anorexia. Two studies (one double-blind and one open-label) in which fluoxetine was administered at target doses of up to 60 mg showed no benefit.19,20 It is possible that the relative lack of dietary precursors necessary for the synthesis of catecholamines is part of the reason for the lack of efficacy during this phase of the illness.21 Recent case studies have shown atypical antipsychotics may help to reduce obsessive thoughts, improve body image, and improve anxiety in patients with anorexia. According to reports, risperidone 0.5–1.5 mg/day and olanzapine 2.5–15.0 mg/day have been suggested to be beneficial.21 Though double-blind, placebo-controlled studies need to be conducted, these reports appear promising. It is important that the distorted body image and odd beliefs about food, weight, and appetite found in patients with anorexia are nonpsychotic symptoms in all but the exceptional patient. There are no studies of benzodiazepines; however, it has been suggested that a small dose of lorazepam 0.5–1.0 mg just prior to meals can help lower anxiety surrounding the eating process. This should only be used on a short-term basis in the initial phases of treatment to decrease anticipatory anxiety.

For the second type of anorexic patient, the weight-restored patient, antidepressants may be helpful in maintaining weight. Most studies have been on the use of fluoxetine, and while there are conflicting outcomes in the literature, it appears some patients may benefit from the use of a selective serotonin reuptake inhibitor (SSRI) to help prevent relapse.21 In one such study dosages of fluoxetine 20–60 mg were administered with the average dose being approximately 40 mg.22 However, others have found no benefit from fluoxetine 60–80 mg over a 1-year period in weight-recovered anorexic patients.23 Patients that continue to struggle with depression, obsessions, or anxiety despite weight-restoration should be treated as with any other patient and may benefit from an antidepressant. There are no studies of the efficacy of antidepressants in treating comorbid diagnosis such as depression in patients with anorexia.

Certainly attention must be given to all the various medical complications. One of particular importance is early-onset osteoporosis seen in patients with anorexia. Again, adequate nutrition is the best treatment, but one must consider use of vitamin supplements to treat osteoporosis in young anorexic women.24 The Society for Adolescent Medicine recommends treating anorexic patients with 1,200 mg of calcium and 400 IU of vitamin D daily. Studies have been conducted to evaluate the use of hormone-replacement therapy in these patients, but as yet no clear benefit on bone density has been shown.25


Bulimia Nervosa

Bulimia is most effectively treated with a combination of outpatient psychotherapy and medications. Much research supports the use of cognitive-behavioral therapy (CBT) as an effective treatment for bulimia.26,27 There is some evidence that interpersonal therapy is a useful treatment.28 A combination of CBT and medications have been found to be more efficacious than medications alone.27,29 There is good evidence that SSRIs are helpful in reducing binge eating and purging behavior, and, in fact, fluoxetine has Food and Drug Administration approval for the treatment of bulimia. SSRIs appear to be effective in the treatment of bulimic symptoms regardless of the presence of comorbid mood or anxiety disorders. Patients with no reported depression or anxiety symptoms can still respond to fluoxetine 60 mg with a significant reduction in binge eating and purging episodes; however, 20 mg has been found to be ineffective.30,31 These doses are similar to those that would be given to patients with obsessive-compulsive disorder. Many SSRIs have been studied, and there is evidence that other SSRIs such as sertraline 100–200 mg/day32,33 and citalopram34 are helpful as well. The SSRIs seem to decrease preoccupation with food and weight, and to help with impulse control over urges to binge eat.

Bupropion is specifically contraindicated in patients with bulimia. There is an increased risk of seizures as a result of administration of bupropion in patients with electrolyte abnormalities, often found in patients with bulimia and anorexia.35 Some evidence supports the use of ondansetron in the treatment of bulimia.36,37 In published studies, patients were instructed to take up to 24 mg/day of ondansetron in 4 mg dosages when they felt the urge to binge or 30 minutes before they were to eat a meal that might lead to a binge.37 Topiramate in a dosage range of 25–400 mg/day has been shown in case studies and one small controlled trial to decrease the tendency to binge, with secondary decreases in compensatory behaviors (ie, purging).38 However, topiramate has side effects that sometimes make it intolerable, particularly fatigue, flu-like symptoms, and paresthesias.38 There is minimal but intriguing evidence that opiate antagonists such as naltrexone in doses of ≥100 mg might be helpful in treating bulimia.39,40 However, in these dosages liver function must be monitored and patients cannot take nonsteroidal anti-inflammatory medications while on naltrexone.41,42

Ultimately, the majority of patients treated with medications alone do not become completely abstinent from binge eating and purging. Long-term studies indicate that abstinence is necessary for recovery.43 Even studies of combination treatment with psychotherapy and medications show there is still room to improve upon the treatment of this disorder.


Binge-Eating Disorder

Fewer clinical trials study the treatment of binge-eating disorder. Medications alone are not found to be highly efficacious. However, there is some evidence that SSRIs help control bingeing behavior in binge-eating disorder as in bulimia. Doses of fluoxetine 60–80 mg/day and citalopram 60 mg/day in comparison to placebo were found to be effective in reducing binge frequency. Topiramate 25–600 mg/day (mean dose=250 mg/day) has also been used to treat binge-eating disorder and has been found to be efficacious in reducing binge-eating episodes.44 CBT has also been successfully used in the treatment of binge-eating disorder.45 Treatment tends to decrease binge eating but does not lead to weight loss. Fluoxetine may offer some advantage over behavioral modification alone. In one study, obese women with or without binge-eating disorder receiving behavioral treatment lost more weight when on fluoxetine (as opposed to placebo).46 Atomoxetine, a selective norepinephrine reuptake inhibitor that is prescribed to induce weight loss when given in doses of 40–120 mg/day (mean=106 mg/day) was found to reduce binge-eating episodes in women with binge-eating disorder in a double-blind, placebo-controlled study.47



Eating disorders represent one of the most serious illnesses among young women. The disorders can have serious long- and short-term mental and physical manifestations for young women who suffer from them. Aggressive treatment that leads to remission of symptoms is critical in order to minimize the long-term health risks of the disorders. Medications may help to reduce the symptoms of bulimia and binge-eating disorder but seldom are sufficient for treating anorexia. Even if medications are helpful in reducing symptoms, they often do not bring about complete remission of symptoms. Maintenance of a healthy weight and good nutrition are essential for recovery and prevention of long-term sequelae. Often an interprofessional team that includes a dietician, therapist, general physician, and/or psychiatrist is necessary for recovery. For some patients, structured self-help manuals may also be useful in helping patients with bulimia or binge-eating disorder to reduce binge eating.48 Additional information on the diagnosis, assessment, and management of eating disorders can be found in the American Psychiatric Association treatment guidelines, which were updated in 2006.49 PP



1. Hoek HW. Incidence, prevalence and mortality of anorexia nervosa and other eating disorders. Curr Opin Psychiatry. 2006;19(4):389-394.
2. Spitzer RL, Devlin MJ, Walsh BT, Hasin D, Wing R, Marcus MD. Binge eating disorder: a multisite field trial of the diagnostic criteria. Int J Eat Dis. 1992;11:191-203.
3. Diagnostic and Statistical Manual of Mental Disorders. 4th ed. Washington, DC: American Psychiatric Association; 1994.
4. Spitzer RL, Yanovski S, Wadden T, et al. Binge eating disorder: its further validation in a multisite study. Int J Eat Dis. 1993;13(2):137-153.
5.  Fairburn CG, Bohn K. Eating disorder (NOS (EDNOS): an example of the troublesome ‘not otherwise specified’ (NOS) category in DSM-IV. Behav Res Ther. 2005;43(6):691-701.
6. Sullivan PF. Mortality in anorexia nervosa. Am J Psychiatry. 1995;152(7):1073-1074.
7. Harris EC, Barraclough B. Suicide as an outcome for mental disorders. A meta-analysis. Br J Psychiatry. 1997;170:205-228.
8. Harris EC, Barraclough B. Excess mortality of mental disorder. Br J Psychiatry. 1998;173:11-53.
9. Powers PS. Initial assessment and early treatment options for anorexia nervosa and bulimia nervosa. Psychiatr Clin North Am. 1996;19(4):639-655.
10. Diagnostic and Statistical Manual of Mental Disorders. 4th ed rev. Washington, DC: American Psychiatric Association; 2000:589.
11. Diagnostic and Statistical Manual of Mental Disorders. 4th ed rev. Washington, DC: American Psychiatric Association; 2000:594.
12. Diagnostic and Statistical Manual of Mental Disorders. 4th ed rev. Washington, DC: American Psychiatric Association; 2000:787.
13. Carney CP, Andersen AE. Eating disorders, guide to medical evaluation and complications. Psychiatr Clin North Am. 1996;19(4):657-679.
14. Mitchell JE. Medical complications of anorexia nervosa and bulimia. Psychiat Med. 1983;1(3):229-255.
15. Powers PS, Coovert DL, Brightwell DR, Stevens BA. Other psychiatric disorders among bulimic patients. Compr Psychiatry. 1988;29(5):503-508.
16. Raymond NC, de Zwaan M, Mitchell JE, Ackard D, Thuras P. Effect of a very low calorie diet on the diagnostic category of individuals with binge eating disorder. Int J Eat Disord. 2002;31(1):49-56.
17. Guarda AS, Pinto AM, Coughlin JW, Hussain S, Haug NA, Heinberg LJ. Perceived coercion and change in perceived need for admission in patients hospitalized for eating disorders. Am J Psychiatry. 2007;164(1):108-114.
18. Keys A, Henschel A, Mickelsen O, Taylor HS, eds. Biology of Human Starvation. Vol I. Minneapolis, MN: University of Minnesota Press; 1950.
19. Attia E, Schroeder L. Pharmacologic treatment of anorexia nervosa: where do we go from here? Int J Eat Disord. 2005;37(Suppl):S60-S69.
20. Strober M, Pataki C, Freeman R, DeAntonio M. No effect of adjunctive fluoxetine on eating behavior or weight phobia during the inpatient treatment of anorexia nervosa: an historical case-control study. J Child Adolesc Psychopharmacol. 1999;9(3):195-201.
21. Steffen KJ, Roerig JL, Mitchell JE, Uppala S. Emerging drugs for eating disorder treatment. Expert Opin Emerg Drugs. 2006;11(2):315-336.
22. Kaye WH, Nagata T, Weltzin TE, et al. Double-blind placebo-controlled administration of fluoxetine in restricting- and restricting-purging-type anorexia nervosa. Biol Psychiatry. 2001;49(7):644-652.
23. Walsh BT, Kaplan AS, Attia E, et al. Fluoxetine after weight restoration in anorexia nervosa: a randomized controlled trial. JAMA. 2006;295(22):2605-2612. Erratum in: JAMA. 2006;296(8):934.
24. Golden NH, Lanzkowsky L, Schebendach J, Palestro CJ, Jacobson MS, Shenker IR. The effect of estrogen-progestin treatment on bone mineral density in anorexia nervosa. J Pediatr Adolesc Gynecol. 2002;15(3):135-143.
25. Klibanski A, Biller BM, Schoenfeld DA, Herzog DB, Saxe VC. The effects of estrogen administration on trabecular bone loss in young women with anorexia nervosa. J Clin Endocrinol Metab. 1995;80(3):898-904.
26. Fairburn CG, Jones R, Peveler RC, et al. Three psychological treatments for bulimia nervosa. A comparative trial. Arch Gen Psychiatry. 1991;48(5):463-469.
27. Walsh BT, Wilson GT, Loeb KL, et al. Medication and psychotherapy in the treatment of bulimia nervosa. Am J Psychiatry. 1997;154(4):523-531.
28. Fairburn CG, Jones R, Peveler RC, Hope RA, O’Connor M. Psychotherapy and bulimia nervosa. Longer-term effects of interpersonal psychotherapy, behavior therapy, and cognitive behavior therapy. Arch Gen Psychiatry. 1993;50(6):419-428.
29. Agras WS, Rossiter EM, Arnow B, et al. Pharmacologic and cognitive-behavioral treatment for bulimia nervosa: a controlled comparison. Am J Psychiatry. 1992;149(1):82-87.
30. Goldstein DJ, Wilson MG, Ascroft RC, al-Banna M. Effectiveness of fluoxetine therapy in bulimia nervosa regardless of comorbid depression. Int J Eat Disord. 1999;25(1):19-27.
31. Fluoxetine Bulimia Nervosa Collaborative Study Group. Fluoxetine in the treatment of bulimia nervosa. A multicenter, placebo-controlled, double-blind trial. Arch Gen Psychiatry. 1992;49(2):139-147.
32. Leombruni P, Amianto F, Delsedime N, Gramaglia C, Abbate-Daga G, Fassino S. Citalopram versus fluoxetine for the treatment of patients with bulimia nervosa: a single-blind randomized controlled trial. Adv Ther. 2006;23(3):481-494.
33. Milano W, Petrella C, Sabatino C, Capasso A. Treatment of bulimia nervosa with sertraline: a randomized controlled trial. Adv Ther. 2004;21(4):232-237.
34. Leombruni P, Piero A, Brustolin A, et al. A 12 to 24 weeks pilot study of sertraline treatment in obese women binge eaters. Hum Psychopharmacol. 2006;21(3):181-188.
35. Horne RL, Ferguson JM, Pope HG Jr, et al. Treatment of bulimia with bupropion: a multicenter controlled trial. J Clin Psychiatry. 1988;49(7):262-266.
36. Faris PL, Eckert ED, Kim SW, et al. Evidence for a vagal pathophysiology for bulimia nervosa and the accompanying depressive symptoms. J Affect Disord. 2006;92(1):79-90.
37. Faris PL, Kim SW, Meller WH, et al. Effect of decreasing afferent vagal activity with ondansetron on symptoms of bulimia nervosa: a randomised, double-blind trial. Lancet. 2000;355(9206):792-797.
38. Hedges DW, Reimherr FW, Hoopes SP, et al. Treatment of bulimia nervosa with topiramate in a randomized, double-blind, placebo-controlled trial, part 2: improvement in psychiatric measures. J Clin Psychiatry. 2003;64(12):1449-1454.
39. Mitchell JE, Laine DE, Morley JE, Levine AS. Naloxone but not CCK-8 may attenuate binge-eating behavior in patients with the bulimia syndrome. Biol Psychiatry. 1986;21(14):1399-1406.
40. Marrazzi MA, Bacon JP, Kinzie J, Luby ED. Naltrexone use in the treatment of anorexia nervosa and bulimia nervosa. Int Clin Psychopharmacol. 1995;10(3):163-172.
41. Kim SW, Grant JE, Adson DE, Remmel RP. A preliminary report on possible naltrexone and nonsteroidal analgesic interactions. J Clin Psychopharmacol. 2001;21(6):632-634.
42. Kim SW, Grant JE, Yoon G, Williams KA, Remmel RP. Safety of high-dose naltrexone treatment: hepatic transaminase profiles among outpatients. Clin Neuropharmacol. 2006;29(2):77-79.
43. Olmsted MP, Kaplan AS, Rockert W. Rate and prediction of relapse in bulimia nervosa. Am J Psychiatry. 1994;151(5):738-743.
44. McElroy SL, Hudson JI, Malhotra S, Welge JA, Nelson EB, Keck PE Jr. Citalopram in the treatment of binge-eating disorder: A placebo-controlled trial. J Clin Psychiatry. 2003;64(7):807-813.
45. Devlin MJ, Goldfein JA, Petkova E, et al. Cognitive behavioral therapy and fluoxetine as adjuncts to group behavioral therapy for binge eating disorder. Obes Res. 2005;13(6):1077-1088.
46. Marcus MD, Wing RR, Ewing L, Kern E, McDermott M, Gooding W. A double-blind, placebo-controlled trial of fluoxetine plus behavior modification in the treatment of obese binge-eaters and non-binge-eaters. Am J Psychiatry. 1990;147(7):876-881.
47. McElroy SL, Guerdjikova A, Kotwal R, et al. Atomoxetine in the treatment of binge-eating disorder: a randomized placebo-controlled trial. J Clin Psychiatry. 2007;68(3):390-398.
48. Stefano SC, Bacaltchuk J, Blay SL, Hay P. Self-help treatments for disorders of recurrent binge eating: a systematic review. Acta Psychiatr Scand. 2006;113(6):452-459.
49. Yager J, Devlin MJ, Halmi KA, et al. Practice Guideline for the Treatment of Patients with Eating Disorders. 3rd ed. Washington, DC: American Psychiatric Association; 2006.


e-mail: ns@mblcommunications.com


Dr. Sussman is editor of Primary Psychiatry and professor of psychiatry at the New York University School of Medicine in New York City.

Dr. Sussman is a consultant to and on the advisory boards of GlaxoSmithKline and Wyeth; and has received honoraria from AstraZeneca, Bristol-Myers Squibb, GlaxoSmithKline, and Wyeth.


In many respects, the diagnosis and treatment of psychiatric disorders represents one of the most daunting tasks in medical practice. There is diagnostic uncertainty and unpredictability of treatment response, even with some of the most frequently encountered disorders. This is particularly evident when it comes to defining and managing bipolar disorder. At the moment, for example, there are several areas of controversy involving the disorder. The first issue is the possibility that this disorder is being overdiagnosed. The second issue is whether antidepressants are helpful in treating patients with bipolar disorder, and more importantly, whether they carry a risk of aggravating the illness over time.

Compelling evidence of the dramatic change in diagnostic patterns describes a recent rapid increase in the outpatient diagnosis of youth bipolar disorder.1 Researchers compared rates of growth from 1994–1995 and 2002–2003 in visits with a bipolar disorder diagnosis by individuals 0–19 years of age versus those ≥20 years. While the diagnosis of bipolar disorder in adults increased nearly 2-fold during the 10-year study period, the diagnosis of bipolar disorder in youths increased approximately 40-fold during this period.

The study also compared the treatment provided to youths and adults during those visits. In terms of treatment, most youths (90.6%) and adults (86.4%) received a psychotropic medication during bipolar disorder visits, with comparable rates of mood stabilizers, antipsychotics, and antidepressants prescribed for both age groups. Youths and adults received a mood stabilizer in approximately 66% of the visits. Anticonvulsants were the mood stabilizers most frequently prescribed in both samples. A similar proportion of youths and adults received a prescription of an antidepressant. Approximately 33% of the visits with antidepressant prescriptions in both age groups did not include prescription of a mood stabilizer.

The impressive increase in the diagnosis of childhood and adolescent bipolar disorder in United States office-based practice indicates a shift in clinical diagnostic practices. In broad terms, either bipolar disorder was historically underdiagnosed in children and adolescents and that problem has now been rectified, or bipolar disorder is currently being overdiagnosed in this age group. Without independent systematic diagnostic assessments, we cannot confidently select between these competing hypotheses.

No doubt, there is increased clinical and public awareness of bipolar disorder, the result of numerous studies and books published in recent years as well as the marketing and medical education activities associated with the many new drugs approved as treatments for bipolar disorder.

Moreno and colleagues1 note that it is possible that pediatric bipolar disorder, previously underdiagnosed, is now being appropriately recognized at earlier ages. They note that the median age-of-onset of bipolar disorder has been located from 19–23 years of age, indicating that the illness starts at a younger age in approximately 50% of patients. It is also possible that bipolar disorder is being overdiagnosed in the pediatric population.

Only a small proportion of children and adolescents meet the full bipolar disorder criteria used to diagnose adults. Instead of, or along with, elation and grandiosity, youths frequently exhibit symptoms such as distractibility, pressured speech, and irritability. The latter symptoms overlap with attention-deficit/hyperactivity disorder (ADHD) and it is possible that some children being treated for ADHD are in fact bipolar. The growing use of second-generation antipsychotics and mood stabilizers may also be behind the shift toward earlier diagnosis and treatment of child and adolescent bipolar disorder.

Currently, there is no consensus concerning diagnostic criteria or preferred interventions for youths diagnosed with bipolar disorder. The aforementioned study found that they are treated the same as adults. A confounding element in diagnosing children and adolescents is the frequent absence of elated mood or grandiosity as symptoms. Instead,  irritability may be prominent, leading to a diagnosis of ADHD or oppositional defiant disorder.

A major question among experts in the treatment of bipolar disorder is whether and when to use antidepressants. Recently promulgated practice guidelines typically advise against the use of antidepressants during bipolar mixed states or dysphoric manias. However, few studies have examined the outcomes of patients with co-occurring manic and depressive symptoms who are treated with antidepressants in addition to mood-stabilizing drugs. This latter question is addressed in a newly published study2 that compares outcomes in patients with bipolar disorder who were treated with a mood-stabilizing agent versus without an antidepressant for a bipolar depressive episode during which they had two or more concurrent manic symptoms. The 335 participants were drawn from the first 2,000 enrollees in the National Institute of Mental Health Systematic Treatment Enhancement Program for Bipolar Disorder. The authors found that adjunctive antidepressant use significantly increased mania symptom severity at 3-month follow-up, higher baseline depression severity was associated with a lower probability of recovery at 3 months, and antidepressant use neither hastened nor prolonged time to recovery.

Thus, patients with bipolar depression accompanied by manic symptoms do not get well with antidepressants any more quickly relative to treatment with mood stabilizers alone, and treatment with antidepressants may lead to greater manic symptom severity.

In terms of children, the lack of research on the efficacy and long-term effects of antidepressants and mood stabilizers makes it difficult to make informed treatment decisions. As mentioned by Moreno and colleagues,1 physicians are practically generalizing pharmacologic treatment principles “developed from adult clinical trials to the treatment of children and adolescents.” If this is, in fact, the correct way to proceed at the moment, the results presented by Goldberg and colleagues2 argue that antidepressants should be used with caution.

In this issue of Primary Psychiatry, Raquel Choua, MD, and colleagues, describe a case report of priapism associated with multiple psychotropics. They also review the literature on this rare side effect. Priapism, a urologic emergency, has been reported to be associated with trazodone and with antipsychotics, with the most likely mechanism being a-adrenergic blockade. This case report serves to remind clinicians to, in turn, remind their patients about psychotropic-induced sexual side effects.

Jasmanda H. Wu, PhD, and colleagues, report on the percentages of patients with various types of treated liver diseases in Medicaid recipients with a diagnosis of schizophrenia or bipolar disorder, compared to Medicaid recipients without a diagnosis of mental illness. The authors also examine the percentages of patients with treated liver diseases in Medicaid recipients with a schizophrenia or bipolar disorder diagnosis, compared to the reported prevalence of liver disease in the general population from the literature. These findings are important because most antipsychotics are metabolized primarily in the liver, so hepatic impairment may affect metabolism. Clinicians should be aware of high occurrence of liver diseases in patients with mental illness and should consider hepatic issues when treating such patients. PP



1. Moreno C, Laje G, Blanco C, Jiang H, Schmidt AB, Olfson M. National trends in the outpatient diagnosis and treatment of bipolar disorder in youth. Arch Gen Psychiatry. 2007;64(9):1032-1039.
2. Goldberg JF, Perlis RH, Ghaemi SN, et al. Adjunctive antidepressant use and symptomatic recovery among bipolar depressed patients with concomitant manic symptoms: findings from the STEP-BD. Am J Psychiatry. 2007;164(9):1348-1355.