Treatment-Resistant Depression and the Bipolar Spectrum: Recognition and Management

Professor of Psychiatry, University of Pittsburgh Medical Center

Introduction

Approximately 50% of those suffering from major depressive disorder (MDD) or dysthymia will not respond to an initial course of antidepressant medication and up to one fifth will remain depressed despite multiple therapeutic trials.¹ In practice, treatment often fails to produce the desired outcome because the course of antidepressant therapy was inadequate (ie, an insufficient dose and/or duration). Nevertheless, when it is clear that adequate courses of therapy have been prescribed and that the patient has adhered to the recommended therapy, the accuracy of the multi-axial diagnostic formulation should be carefully re-examined. Specifically, the treating or consulting psychiatrist must try to answer the following question: Is this episode of treatment-resistant depression (TRD) the result of an unrecognized or untreated complicating psychiatric or medical condition and, if so, could this patient actually be suffering from a subform of depression that is not responsive to antidepressant pharmacotherapy? Bipolar affective disorder represents one such differential diagnostic consideration, particularly the less clear-cut presentations that can be viewed as falling on the broad spectrum between the classic type I form of bipolar disorder and unipolar melancholia. Elsewhere in this issue, apparent continuum between recurrent (unipolar) MDD and bipolar affective disorders are reviewed.² The so-called bipolar spectrum extends from bipolar type II disorder to include patients with recurrent depression and antidepressant-induced hypomania, those with a family history of bipolar disorder, those who have experienced briefer upward mood swings that do not meet syndromal criteria for hypomania, and patients with cyclothymia (ie, a history of hypomanic episodes, coupled with only minor depressive episodes). This article explores the utility of the bipolar spectrum as an explanatory construct for understanding antidepressant nonresponse and as a prescriptive indicator for treatment selection.

Standardized Approaches to Treatment-Resistant Depression

Finding the right antidepressant for a patient with a difficult-to-treat depressive disorder is no easy matter. In fact, there are few reliable predictors of response to specific types of antidepressants and most of the replicated predictors of antidepressant response per se actually reflect correlates of placebo response, spontaneous remission, and other so-called nonspecific variables.¹ Moreover, clinicians tend to underestimate the magnitude of the variance in treatment response explained by placebo expectancy factors. In randomized placebo-controlled studies, the correlation between response rates in placebo and active antidepressant arms averages approximately .7, which indicates that approximately 50% of the “action” of an antidepressant is attributable to nonspecific factors.

Patients with TRD are, essentially by definition, unlikely to benefit from placebo-expectancy factors.³ Stripped of the placebo effect, the actually specific effect of antidepressants is relatively small. For example, in a series of meta-analyses of acute phase, double-blind, randomized controlled trials (RCTs) that led to the United States Food and Drug Administration’s approval of venlafaxine, bupropion sustained release (SR) and extended release (XL), and duloxetine, the absolute advantage of these drugs and selective serotonin reuptake inhibitors (SSRI) comparators over placebo ranged between 9% and 20%.^4-6 Thus, if no placebo response can be counted on for the patient with TRD, clinicians should be aware that the odds that a new trial of antidepressant medication will be successful are dwarfed by the odds of yet another therapeutic failure.

In the era of evidence-based medicine, various treatment strategies are typically arrayed in hierarchical algorithms, ranked according to criteria such as cost, ease of use, safety, and evidence of efficacy.¹ For MDD, a contemporary algorithm might recommend an initial course of therapy with a generic formulation of one of the SSRIs, which might be followed by treatment with a more expensive “branded” formulation of another potentially first-line medication (ie, escitalopram, bupropion XL, or venlafaxine extended release [ER]). If several courses of therapy with first-line medications prove ineffective, the algorithms emphasize use of various augmentation strategies, or the older tricyclic antidepressants (TCAs) or monoamine oxidase inhibitors (MAOIs). Nonresponse to these second- and third-line strategies, in turn, should lead to use of more invasive or expensive interventions such as electroconvulsive therapy (ECT) or vagus nerve stimulation (VNS).

The strength of such an approach, if broadly applied across practice settings, is that it helps to ensure that people receiving treatment for depression will, on average, be more likely to receive a higher level of care (ie, a series of adequate trials with interventions that are known to work).¹ However, there are several limitations to this approach, including the fact that the evidence base that underpins each of the options for TRD is inadequate. Also, off particular importance in clinical practice, it largely ignores the possibility that there may be clinically relevant predictors of differential response to the various algorithmic therapies.

An example is the case of John S., a 57-year-old widower who presents for treatment of a first lifetime episode of depression. There is no personal or family history of bipolarity, no relevant psychiatric comorbidity, and Mr. S. is in good health. Nevertheless, he is quite symptomatic and his condition meets criteria for a major depressive episode (MDE) with melancholic features. Without further information about his condition, appropriate first-line interventions might include any of the generic SSRIs or, if permitted by a formulary, “branded” escitalopram or venlafaxine ER (with the justification of potentially greater efficacy for patients with more severe depressions).^4,7 On occasion, a psychiatrist might even skip over the newer antidepressants and prescribe a TCA, arguing that this venerable class of medications has the strongest track record for the most severe depressive states.^8,9 However, it turns out that all of these considerations are likely to be ineffective because Mr. S. did not tell his psychiatrist that he is hearing the voice of his deceased wife and that he believes that God has cursed him because of past infidelities. With the knowledge that John S. is suffering from a severe MDE with psychotic features, the optimal strategies would include either the combination of an antipsychotic and antidepressant or ECT.¹⁰

The limitations of the algorithm approach to TRD are further illustrated by the results of the Sequenced Treatment Alternatives to Relieve Depression (STAR*D) study, which began to emerge in early 2006.^11-13 In this large, inclusive, multi-center trial, outpatients with MDD were first treated with citalopram (at doses up to 60 mg/day for up to 14 weeks) and, if unremitted, were eligible to participate in a series of randomized controlled trials of various alternate therapies, selected to represent various choice-points on a treatment algorithm. The seven therapeutic alternatives tested for citalopram nonresponders (ie, level-one TRD) represented theoretically distinct options. There were four “switching” strategies, which included one within-class switch to a second SSRI (ie, sertraline), two across-class switches to distinctly different antidepressants (ie, bupropion or venlafaxine), and cognitive therapy. There were also three augmentation strategies, in which bupropion, buspirone, or cognitive therapy were added to ongoing citalopram therapy. Results indicated that there were no unequivocal winners and none of the options were particularly effective, with remission rates ranging from 18% to 30%.^11,13 Even lower remission rates were observed in STAR*D’s randomized comparisons of second- and third-line strategies for TRD.^14-16 Clearly, when SSRI nonresponders are randomly assigned to various reasonable next-step interventions, there is much room for improvement.

The Heterogeneity of Depression and the Bipolar Spectrum

There is, of course, no reason to expect that a heterogeneous condition like MDD will respond uniformly well to any single alternate treatment strategy. One method to improve response to antidepressant therapies is to identify better ways to match patients with specific therapies. For decades, the most useful approach to manage the heterogeneity of depressive syndromes has been to identify and validate clinically meaningful subforms of affective illness. Indeed, until the promise of a system of differential therapeutics based on genomic or pathophysiologic methods is realized, clinical subtyping depressive episodes will remain an essential element or aspect of differential therapeutics. Among the several more enduring subtypes of depression that have been evaluated, the bipolar/unipolar (or nonbipolar) dichotomy has been arguably the most influential.² The year 2006, in fact, marks the 40th anniversary of Perris’¹⁷ influential monograph that modernized the Kraepelinian approach to recurrent mood disorders and outlined the basic case for distinguishing between bipolar and unipolar forms of mood disorders.

The first step of this classification, identifying patients who have had a manic episode, is the easiest; mania by definition is psychotic, incapacitating, and/or so severe that hospitalization is required, and the diagnosis of type I bipolar affective disorder is made with very high reliability. For example, in a study of people applying to join a voluntary registry for bipolar affective disorder, those who reported having suffered a past manic episode almost always met criteria for type I bipolar affective disorder.¹⁸

In practice, it is the second and third steps of the differential diagnosis, namely, identification of those who have experienced only milder or briefer hypomanic episodes and declaring with confidence that the depressive disorder is truly “unipolar,” that prove more challenging. Specifically, reliability of the diagnosis of bipolar type II disorder is problematic even under optimal circumstances¹⁹ and the incidence of bipolar type II disorder may be grossly underestimated by nonexpert interviewers.²⁰ Part of the problem with the reliability of the bipolar type II diagnosis is definitional (there is ongoing debate regarding the minimum duration of a hypomanic episode, ranging from 2–7 days) and part pertains to ascertainment or case finding (people almost never seek treatment for the more productive, euphoric hypomanic episodes and generally view these periods as nonpathologic).

Despite this challenge, there are some very good reasons to look for unrecognized bipolarity when evaluating patients with TRD. First, as discussed by Ghaemi²¹ in this issue, bipolar depressions are not particularly responsive to antidepressants and, even when apparently effective in the short run, standard antidepressants may not convey much protection against relapse.²² Thus, within a hypothetical universe of antidepressant nonresponders, patients with as yet unrecognized bipolar disorder would be expected to be overrepresented.

Second, although it is true that mania per se is not difficult to diagnose, approximately 70% of patients with bipolar affective disorder report that they were misdiagnosed early in their illness course. Not surprisingly, by far the most common misdiagnosis is MDD.²³ As the initial episode of bipolar disorder is as likely to be depression as it is mania, misdiagnosis is to a great extent inevitable. It is nevertheless important to reduce the time that elapses between onset of the initial episode of depression and recognition of bipolarity; in one prospective study, the first lifetime depressive episode preceded the onset of mania by an average of 6.4 years.²⁴ Similarly, in a survey of the membership of the National Depressive and Manic Depressive Association (now known as the National Depression and Bipolar Support Alliance), for example, it took contact with an average of four different physicians, spanning across almost one decade, before the correct diagnosis was finally achieved.²³

Systematically screening for bipolar disorder in patients with TRD also directly pertains to treatment selection. Whereas mood stabilizers are generally considered third- or fourth-line strategies for patients with unipolar depression, they are the cornerstone of management of bipolar depressions in order to lessen the risk of treatment-emergent affective switches.²⁵ Thus, making the correct diagnosis earlier in the course of a patient’s treatment may reduce the risk of iatrogenic complications, including induction of rapid cycling.

How Common is “Pseudo-Unipolar” Depression?

It has long been known that some patients with a history of recurrent depressive episodes have family histories and patterns of treatment response that are highly suggestive of bipolar affective disorder. More than 30 years ago, for example, Kupfer and colleagues²⁶ described a subgroup of patients with recurrent depression who were lithium-responsive and shared a number of clinical characteristics with patients with bipolar disorder. For years it was taught that such “pseudo-unipolar” patients might account for approximately 10% to 20% of all depressions.²⁷ Indeed, in the early work of Perris,¹⁷ it appeared that once a patient had experienced three unipolar depressive episodes, there was only an approximately 5% chance that he or she subsequently would develop a manic episode. Recent research calls these statistics into doubt; it appears that bipolar affective disorder is much more common than once thought and that this secular trend is explained by increasing recognition of bipolar type II disorder. Specifically, whereas the risk of mania has not increased (lifetime incidence rates continue to cluster at approximately 1% across industrial cultures²⁸), the incidence of hypomania is substantially greater than previously appreciated. Several more recent studies have documented that at least 4% of adults have bipolar type II disorder.^29,30 The incidence may increase to as high as 6% when the diagnostic criteria are further broadened, and briefer treatment-emergent hypomanias are included in the tally.³¹ Thus, if one assumes that 10% to 12% of the adult US population will experience at least one lifetime episode of MDD, approximately 50% of these depressions actually can be classified within the bipolar spectrum.

Improving Recognition of Depressions within the Bipolar Spectrum

Perhaps the most controversial aspect of the differential diagnosis of bipolar affective disorder involves the classification of treatment-emergent manic or hypomanic episodes. In the Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition, Text Revision (DSM-IV-TR),^10,32 for example, manic or hypomanic episodes that develop during antidepressant therapy are classified in the same way as depressive episodes that develop after initiation of oral contraceptives or interferon. As such, antidepressant-induced affective switches are not permitted to count toward the diagnosis of bipolar affective disorder. Akiskal and Benazzi³³ have pointed out the dubious logic underpinning this decision. Specifically, it assumes that antidepressants have the capacity to cause mania or hypomania in people who were not otherwise vulnerable to bipolarity and negates strong evidence from longitudinal studies that treatment-emergent affective switches are one of the most robust predictors of subsequent “spontaneous” bipolar episodes.^22,34,35 Indeed, treatment-emergent hypomanic episodes appear to be indistinguishable from spontaneously occurring episodes.^31,36 Although the DSM-IV-TR may not permit a patient with a history of a treatment-emergent hypomanic episode to receive the formal diagnosis of bipolar disorder, the nomenclature does not prevent modifying the treatment plan to take this information into account.

Several other aspects of the depressed patient’s history and state may aid in recognition of bipolarity even in the absence of treatment-emergent affective switches. Patients with bipolar depression tend to have a relatively early onset of illness when compared to those with nonbipolar depression,²⁷ and there is a particularly high risk of subsequent bipolarity among those who experience an initial episode of depression at <20 years of age.^24,37 In this regard, a positive family history of bipolar disorder is associated with both an early age of onset as well as a higher risk of subsequent treatment-emergent hypomanic episodes. According to Akiskal,³⁸ the likelihood of switching from a diagnosis of MDD to a bipolar diagnosis was greatest when there was a history of manic depression in three first-degree family members or across three consecutive family members.

Among those with early-onset depression, psychotic and catatonic features also have been associated with a greater likelihood of bipolar disorder.^37,39,40 Although these are relatively uncommon features of depression in ambulatory settings, they are not rare among more severely depressed hospitalized patients and, when evident, they should greatly increase the index of suspicion of bipolar affective disorder.

By contrast, the so-called atypical or reversed neurovegetative features of depression, including anergia or leaden paralysis, hypersomnia, and weight gain or increase in appetite, are extremely common in ambulatory populations. Associated with an earlier age of onset, atypical depressive features also are over-represented among patients with bipolar disorder.^41-43 For example, in one large prospective study, 45% of outpatients with bipolar type II disorder manifest multiple reverse neurovegetative features, as compared to only 25% of patients with MDD.⁴⁴ In a second study involving an overlapping group of patients with MDD, 24% of the patients with atypical features had a family history of bipolar disorder, as compared to only 13% of those who did not have atypical features.⁴¹

As psychomotor agitation is traditionally considered one of the hallmarks of more severe or typical (ie, melancholic) depressions, it may be surprising to note that several studies also have demonstrated a greater risk of bipolarity in patients with agitated depression.^45,46 This apparent paradox may be attributable to unrecognized mixed states, the most difficult-to-treat presentation of bipolar affective disorder. In evaluating psychomotorically agitated patients with TRD, it is thus worthwhile to look for other signs and symptoms of dysphoric activation during antidepressant therapy, including decreased need for sleep, increased libido, grandiose ideas, or racing thoughts despite the persistence of the depressive syndrome. The onset of suicidal ideation during antidepressant therapy also may be an indicator of mixed states and has been proposed as one possible explanation of recent controversy of treatment-emergent suicidality in teenagers.⁴⁷ Of note, recognition of mixed states in bipolar type II patients is not helped by the DSM-IV-TR, which does not include this possibility as a formal classification.

Even more controversial are the proposed temperamental and behavioral correlates of bipolar spectrum diagnoses. Akiskal³⁸ has argued that, at least among individuals who have not experienced a full manic episode, behavioral indicators of hypomania are more valuable than subjective reports of elation or other mood changes. Taking a biographic (rather than descriptive) approach to the phenomenology of hypomania, he identified numerous provisional indicators, outlined in the Table.³⁸ Akiskal further emphasized that although many of these behavioral indicators have low sensitivity for bipolarity (ie, they occur in only a small minority of patients), they have high specificity (ie, they rarely occur in individuals with nonbipolar depression).

Of particular interest in the evaluation of patients with long histories of nonproductive psychiatric treatment, Akiskal³⁸ also suggested three triads of comorbidities that were linked to bipolarity. Namely, a history of three different anxiety disorder diagnoses, abuse/dependence with three different substances, and/or three different cluster B personality disorders (ie, borderline, antisocial, and histrionic personality disorders).

When aspects of the history are in doubt, it is helpful to talk to the spouse or romantic partner, relatives, or close friends, as these people may be more objective than the patient or at least more accurate judges of the impact of the patient’s hypomanic behavior on others. Siblings and parents also may provide additional family history, including more subtle, transgenerational behavioral indicators of bipolarity, such as promiscuity, flamboyance, eminence, episodic creativity, impulsive financial undertakings, or marked irritability.

Treating Bipolar Spectrum Depressions

There have been relatively few controlled studies of TRD, even fewer studies of bipolar type I depression, and only a handful of studies of treatment-resistant bipolar depression. This unfortunate state of affairs largely reflects two facts. First, the vast majority of research on antidepressant medication is funded by the pharmaceutical industry (which is understandably focused on registration studies of novel drugs for treatment of MDEs). Second, studies of bipolar depression or TRD are much more difficult to conduct than studies of uncomplicated MDEs.

With such a scanty amount of evidence from RCTs to inform practice, clinicians must base treatment decisions for patients with treatment-resistant bipolar spectrum depressions on expert opinion, first hand experience, and word-of-mouth. Although the sources of information are no doubt better than no evidence at all, none are particularly reliable or valid. Drawing upon the meager data that are available, some suggestions are outlined below.

Mood Stabilizers

There is essentially universal consensus in contemporary practice guidelines that patients with bipolar type I depressions should received mood stabilizers, either alone or in combination with antidepressants (see, for example, Thase²² for a review of the recommendations of practice guidelines). Although the mood stabilizers—including lithium and the anticonvulsants valproate, carbamazepine, and lamotrigine—have only modest antidepressant effects, they do not provoke switches or cycling and, even if ineffective, will be indicated subsequently for prophylaxis.

Among the mood stabilizers, lithium has been studied most extensively and there is unequivocal evidence of efficacy for both bipolar type I depression⁴⁸ and in combination with antidepressants for nonbipolar forms of TRD.⁴⁹ An additional advantage of lithium is that it is the only therapy that has been shown to reduce the risk of suicidal behavior for patients with bipolar disorder.^50,51

Despite these advantages, lithium augmentation is no longer widely used in contemporary practice. There are several reasons for this. First, lithium therapy requires blood tests and serial monitoring, which may be a disincentive. Second, as most of the evidence in TRD comes from studies of TCAs,¹ efficacy has not been established convincingly with the newer antidepressants. Third, some data suggest that tolerability in combination with SSRIs or serotonin norepinephrine reuptake inhibitors (SNRIs) can be more problematic than commonly appreciated.⁵² Last, lithium has never been a preferred treatment for patients with rapid-cycling forms of bipolar disorder, and its utility for bipolar type II depression has not been convincingly established.^22,48 Thus, there are legitimate questions about both the utility and tolerability of lithium augmentation for patients with bipolar spectrum depressions that have not responded to SSRIs or SNRIs.

Among the anticonvulsant mood stabilizers, lamotrigine receives high marks in more recent practice guidelines for bipolar depression²² and is now widely prescribed by mood disorder specialists for treatment of both bipolar types I and II depressive episodes.⁵³ Despite such current enthusiasm, it is important to keep in mind that the basis of the FDA’s approval of lamotrigine for treatment of bipolar disorder is from relapse-prevention studies, not acute-phase studies. In fact, the performance of lamotrigine in placebo-controlled studies of bipolar depression has been mixed.^54,55 Moreover, there are no positive data from placebo-controlled studies of lamotrigine therapy of nonbipolar TRD.

With respect to treatment-resistant bipolar depression, monotherapy with lamotrigine was significantly more effective than gabapentin in a small study using a crossover design.⁵⁶ In another small study using a parallel-group design for patients who had failed to respond to lithium or valproate in combination with SSRIs and/or bupropion, trends favored add-on therapy with lamotrigine over two active comparators (risperidone or inositol), although few of the differences were statistically significant and the absolute remission rate was only 25%.⁵⁷ In a third, larger study of outpatients with rapid-cycling subtype of bipolar disorder,⁵⁸ the antidepressant effects of lamotrigine monotherapy were significant among the subset of patients with bipolar type II depressions but not among those with bipolar type I depression.

In addition to the limited evidence of efficacy, concerns regarding the risk of Stevens Johnson syndrome and related severe dermatologic reactions necessarily limit even wider use of lamotrigine for bipolar spectrum depressions. Precautionary strategies (initiating therapy at 12.5–25 mg/day, slow upward titration, and instructing patients to stop treatment immediately at the first sign of a rash) reduce the risk of severe rash to <1 per 1,000 patients treated, a level that is similar to that observed during therapy with carbamazapine or valproate.⁵⁹ Nevertheless, because milder, relatively benign allergic reactions are still common (ie, 5% to 10% during the first 2 months of therapy), a significant minority of clinicians continue to have a high index of concern about lamotrigine therapy.

Atypical Antipsychotics

Arguably also classified as mood stabilizers, atypical antipsychotics are increasingly used by psychiatrists for treatment of bipolar disorder and nonbipolar TRD.^1,22 To date, all of the atypical antipsychotics (except clozapine) have been approved by the FDA for treatment of mania, and data on their utility for prevention of relapse following successful acute-phase therapy are slowly emerging. Olanzapine, the first atypical antipsychotic to be approved for treatment of mania, is the best studied; there is evidence of efficacy in double-blind RCTs of dysphoric mania,⁶⁰ rapid-cycling bipolar type I disorder,⁶¹ bipolar type I depression (both alone and in combination with fluoxetine),⁶² and refractory unipolar depression.⁶³ It is noteworthy that, in the study of bipolar type I depression, the combination of olanzapine and fluoxetine was significantly more effective than olanzapine monotherapy, yet was associated with no greater risk of mania than inert placebo.⁶² The proprietary combination of olanzapine and fluoxetine is the only treatment currently approved for bipolar depression, and an application for the approval of the compound for TRD is currently under FDA review.

Results of two large, placebo-controlled RCTs likewise have established the efficacy of quetiapine monotherapy for bipolar depression.^64,65 In both studies, two doses of quetiapine (300 and 600 mg/day) were significantly more effective than placebo; there was no hint of greater benefit for the higher dose in either study. The efficacy of quetiapine in these trials was evident in patients with bipolar type I depression and among patients with a history of rapid cycling. In the second study, as well as a pooled analysis of the two studies, quetiapine was effective in patients with bipolar type II depressive episodes. Although quetiapine has not yet been systematically studied in TRD, it is likely to become the first single therapy specifically approved for treatment of bipolar depression.

Studies of the other atypical antipsychotics for treatment of bipolar depression and TRD are currently underway. As there is evidence from RCTs that each of these medications are efficacious treatments of dysphoric mania, it seems almost certain that the entire class of atypical antipsychotics will be proven useful for bipolar spectrum depressions. Nevertheless, in terms of the strength of the evidence, olanzapine and quetiapine should receive the highest marks.

The major factors limiting the use of atypical antipsychotics for patients with all forms of TRD are cost and concerns about metabolic side effects (eg, weight gain, dyslipidemia, and risk of development of diabetes and other metabolic complications). The former issue is, of course, time-limited, as generic formulations of at least one atypical, risperidone, will soon be available. With respect to concerns about metabolic side effects, although there is the greatest concern with olanzapine, all patients treated with atypical antipsychotics should be monitored closely, including documenting weight, waist/belt size, fasting blood glucose, and lipid profiles prior to instituting therapy.⁶⁶ There is sufficient evidence of differences in the metabolic profiles of the various atypical antipsychotics that if tolerability proves problematic with olanzapine or quetiapine, another member of the class may still be successful.^66-68

Antidepressants: Special Considerations

As TRD patients with bipolar spectrum depressions have not responded to a number of different antidepressants, the overall approach to treatment necessarily points to other therapeutic options. Indeed, antidepressants should be used judiciously, if at al, in patients who have developed treatment-emergent affective switches into mania or hypomania, mixed states, or rapid cycling. Generally, practice guidelines emphasize that such patients should be treated more conservatively with mood stabilizers or atypical antipsychotics, either singly or in combination.²²

When therapeutic alternatives to antidepressants have been exhausted, there are several pertinent considerations for the patient with a history of nonresponse to several SSRIs. First, TCAs should be avoided, both because they are the most likely to be associated with treatment-emergent affective switches and because they have the greatest lethality in overdose.²² Second, among the remaining newer antidepressants, bupropion receives the highest marks for patients with bipolar depression, both because of its low incidence of sexual side effects⁶ and relatively lower risk of treatment-emergent affective switches.^69,70 Third, although the SNRI venlafaxine has a particularly strong track record in TRD,⁷¹ it appears to be associated with a higher risk of treatment-emergent affective switches than SSRIs^70,72 or bupropion.⁷⁰ At the higher doses often used for patients with TRD, venlafaxine therapy also is associated with an approximately 3% to 5% risk of high blood pressure.⁷³ Thus, venlafaxine should be considered a third-line option for bipolar spectrum patients with TRD. It is not yet clear if the second SNRI to be introduced to the US market, duloxetine, will have the same liabilities as venlafaxine. One likely advantage over venlafaxine is a lower incidence of treatment-emergent high blood pressure; in placebo-controlled registration studies, duloxetine therapy was associated with a <1% incidence.⁷³

For patients who do not respond to bupropion or an SNRI, the MAOIs have the best track record in bipolar depression.^48,74 Although these data are derived from studies of older MAOIs such as tranylcypromine,^75-77 safety and tolerability concerns now would lead many clinicians to favor an initial trial with one of the newer compounds (ie, the selegiline patch or, outside of the US, moclobemide). Nevertheless, it is important to keep in mind that, whereas these medications do not require the dietary restrictions necessary with the older agents, they are not free of the risk of drug-drug interactions when used in close proximity to SSRIs or SNRIs. Thus, patients must be withdrawn from the ineffective antidepressant and washed-out for at least 7–14 days (28–35 days for fluoxetine) before MAOI therapy can be started.⁷⁸

The selegiline skin patch, the only antidepressant medication to be introduced in the US in the past 2 years, is of interest for several reasons beyond the novel mode of administration.⁷⁹ Already available for a number of years in oral form for treatment of Parkinson’s disease, at low doses selegiline is a selective, irreversible inhibitor of the B subtype of monoamine oxidase (MAO). However, in these “B-selective” doses selegiline is not an effective antidepressant and, as higher doses begin to inhibit the A subform of MAO in the gut (ie, the mechanism that initiates the so-called “cheese effect”), oral antidepressant therapy with selegiline did not convey a significant advantage compared to the older MAOIs. Transdermal delivery avoids direct inhibition of MAO A in the gut as well as first-pass metabolism in the liver, permitting relatively higher levels of selegiline to reach the brain and hence permitting antidepressant activity without dietary restriction, at least at the minimum therapeutic dose (ie, 6 mg/24 hours). The basis of approval comes exclusively from studies of MDD. To date, there is only a limited amount of experience in treatment of bipolar depression.

Other Treatment Options

If these more conventional treatment strategies fail to relieve a resistant bipolar spectrum depression, a number of alternatives are available.²² Possible ambulatory strategies include adding thyroid hormone⁸⁰ or psychostimulants and other dopamine agonists to ongoing antidepressant therapy,^81-83 phototherapy,⁸⁴ and sleep deprivation.⁸⁵ If the depression is not too severe, intensive psychotherapy also might be employed as an add-on therapy, although the evidence for these interventions largely comes from longer-term relapse prevention studies.^86-88 For the most severely ill patients, there is no more effective strategy than ECT. Whereas the efficacy of ECT, for both bipolar depression and TRD, is indisputable, the risk of relapse following successful acute phase treatment is staggeringly high. Evidence from one study of patients with unipolar TRD suggests that risk can be lessened significantly by combining lithium and an antidepressant medication for preventive therapy.⁸⁹ However, it is unclear if other mood stabilizers or atypical antipsychotics may convey the same protection for patients with lithium-resistant bipolar depression. Other factors that restrict the use of ECT include cost, stigma, and cognitive side effects. Unlike ECT, the less predictable therapeutic effects of VNS therapy appear to be more durable but they may take weeks or even months to emerge.⁹⁰

Conclusion

Part of the notorious heterogeneity of MDD is attributable to the overlap with bipolar affective disorder. For patients with a history of hypomania or more attenuated expressions of the illness within the bipolar spectrum, depression may be more chronic and more difficult to treat with antidepressant medications. As recent evidence indicates that up to 50% of those seeking treatment for depression may have conditions that fall within this softer end of the bipolar spectrum, clinicians evaluating patients with a history of nonresponse to antidepressants should carefully screen for indicators of bipolarity. When a bipolar spectrum depression is recognized, better outcomes may be obtained by judiciously limiting the use of antidepressants and by greater use of mood stabilizers and atypical antipsychotics.

Disclosure: Dr. Thase is a consultant to and/or on the advisory boards of AstraZeneca, Bristol-Myers Squibb, Cephalon, Cyberonics, Eli Lilly, Forest, GlaxoSmithKline, Janssen, Novartis, Organon, Sepracor, Shire, and Wyeth; and is on the speaker’s bureaus of AstraZeneca, Bristol-Myers Squibb, Cyberonics, Eli Lilly, GlaxoSmithKline, Organon, sanofi-aventis, and Wyeth.

References

1. Thase ME. Therapeutic alternatives for difficult-to-treat depression: a narrative review of the state of the evidence. CNS Spectr. 2004;9(11):808-816, 818-821.

2. Saggese et al, The Role of Recurrence and Cyclicity in Differentiating Mood Disorder Diagnoses. Prim Psych. 2006;13(11):46-51.

3. Rush AJ, Thase ME, Dube S. Research issues in the study of difficult-to-treat depression. Biol Psychiatry. 2003;53(8):743-753.

4. Thase ME, Entsuah AR, Rudolph RL. Remission rates during treatment with venlafaxine or selective serotonin reuptake inhibitors. Br J Psychiatry. 2001;178:234-241.

5. Thase ME. Comparing the efficacy of the newer antidepressants. In: Gilaberte I, ed. Nuevas Perspectivas en la Fepresión. Madrid, Spain: Aula Médica Endiciones; 2004:253-286.

6. Thase ME, Haight BR, Richard N, et al. Remission rates following antidepressant therapy with bupropion or selective serotonin reuptake inhibitors: a meta-analysis of original data from 7 randomized controlled trials. J Clin Psychiatry. 2005;66(8):974-981.

7. Thase ME. Managing depressive and anxiety disorders with escitalopram. Expert Opin Pharmacother. 2006;7(4):429-440.

8. Barbui C, Guaiana G, Hotopf M. Amitriptyline for inpatients and SSRIs for outpatients with depression? Systematic review and meta-regression analysis. Pharmacopsychiatry. 2004;37(3):93-97.

9. Anderson IM. Meta-analytical studies on new antidepressants. Br Med Bull. 2001;57:161-178.

10. American Psychiatric Association. Practice guideline for the treatment of patients with major depressive disorder (revision). Am J Psychiatry. 2000;157(suppl 4):1-45.

11. Rush AJ, Trivedi MH, Wisniewski SR, et al. Bupropion-SR, sertraline, or venlafaxine-XR after failure of SSRIs for depression. N Engl J Med. 2006;354(12):1231-1242.

12. Trivedi MH, Fava M, Wisniewski SR, et al. Medication augmentation after the failure of SSRIs for depression. N Engl J Med. 2006;354(12):1243-1252.

13. Trivedi MH, Rush AJ, Wisniewski SR, et al. 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.

14. Fava M, Rush AJ, Wisniewski SR, et al. A comparison of mirtazapine and nortriptyline following two consecutive failed medication treatments for depressed outpatients: a STAR*D report. Am J Psychiatry. 2006;163(7):1161-1172.

15. McGrath PJ, Stewart JW, Fava M, et al. A comparison of tranylcypromine to the combination of venlafaxine-XR plus mirtazapine following three failed antidepressant medication trials for depression: a STAR*D report. Am J Psychiatry. In press.

16. Nierenberg AA, Fava M, Trivedi MH, et al. A Comparison of Lithium and T3 Augmentation Following Two Failed Medication Treatments for Depression: A STAR*D Report. Am J Psychiatry. 2006;163(9):1519-1530.

17. Perris C. A study of bipolar (manic-depressive) and unipolar recurrent depressive psychoses. Acta Psychiatr Scand Suppl. 1966;194:1-188.

18. Cluss PA, Marcus SC, Kelleher KJ, Thase ME, Arvay LA, Kupfer DJ. Diagnostic certainty of a voluntary bipolar disorder case registry. J Affect Disord. 1999;52(1-3):93-99.

19. Coryell W. Bipolar II disorder: a progress report. J Affect Disord. 1996;41(3):159-162.

20. Dunner DL, Tay LK. Diagnostic reliability of the history of hypomania in bipolar II patients and patients with major depression. Compr Psychiatry. 1993;34(5):303-307.

21. Ghaemi [REFERENCE IN THIS ISSUE TK]

22. Thase ME. Bipolar depression: issues in diagnosis and treatment. Harv Rev Psychiatry. 2005;13(5):257-271.

23. Hirschfeld RM, Lewis L, Vornik LA. Perceptions and impact of bipolar disorder: how far have we really come? Results of the national depressive and manic-depressive association 2000 survey of individuals with bipolar disorder. J Clin Psychiatry. 2003;64(2):161-174.

24. Akiskal HS, Walker P, Puzantian VR, King D, Rosenthal TL, Dranon M. Bipolar outcome in the course of depressive illness. Phenomenologic, familial, and pharmacologic predictors. J Affect Disord. 1983;5(2):115-128.

25. American Psychiatric Association. Practice guideline for the treatment of patients with bipolar disorder (revision). Am J Psychiatry. 2002;159(suppl 4):1-50.

26. Kupfer DJ, Pickar D, Himmelhoch JM, Detre TP. Are there two types of unipolar depression? Arch Gen Psychiatry. 1975;32(7):866-871.

27. Goodwin FK, Jamison KR. Manic-Depressive Illness. London, England: Oxford University Press; 1990.

28. Weissman MM, Bland RC, Camino GJ, et al. Cross-national epidemiology of major depression and bipolar disorder. JAMA. 1996;276(4):293-299.

29. Angst J. Bipolar disorder--a seriously underestimated health burden. Eur Arch Psychiatry Clin Neurosci. 2004;254(2):59-60.

30. Judd LL, Akiskal HS. The prevalence and disability of bipolar spectrum disorders in the US population: re-analysis of the ECA database taking into account subthreshold cases. J Affect Disord. 2003;73(1-2):123-131.

31. Akiskal HS, Hantouche EG, Allilaire JF, et al. Validating antidepressant-associated hypomania (bipolar III): a systematic comparison with spontaneous hypomania (bipolar II). J Affect Disord. 2003;73(1-2):65-74.

32. Diagnostic and Statistical Manual of Mental Disorders. 4th ed text rev. Washington, DC: American Psychiatric Association; 2000.

33. Akiskal HS, Benazzi F. The DSM-IV and ICD-10 categories of recurrent [major] depressive and bipolar II disorders: evidence that they lie on a dimensional spectrum. J Affect Disord. 2006;92(1):45-54.

34. Ghaemi SN, Boiman EE, Goodwin FK. Diagnosing bipolar disorder and the effect of antidepressants: a naturalistic study. J Clin Psychiatry. 2000;61(10):804-808.

35. Chun BJ, Dunner DL. A review of antidepressant-induced hypomania in major depression: suggestions for DSM-V. Bipolar Disord. 2004;6(1):32-42.

36. Stoll AL, Mayer PV, Kolbrener M, et al. Antidepressant-associated mania: a controlled comparison with spontaneous mania. Am J Psychiatry. 1994;151(11):1642-1645.

37. Strober M, Carlson G. Predictors of bipolar illness in adolescents with major depression: a follow-up investigation. Adolesc Psychiatry. 1982;10:299-319.

38. Akiskal HS. Searching for behavioral indicators of bipolar II in patients presenting with major depressive episodes: the “red sign,” the “rule of three” and other biographic signs of temperamental extravagance, activation and hypomania. J Affect Disord. 2005;84(2-3):279-290.

39. Akiskal HS, Maser JD, Zeller PJ, et al. Switching from ‘unipolar’ to bipolar II. An 11-year prospective study of clinical and temperamental predictors in 559 patients. Arch Gen Psychiatry. 1995;52(2):114-123.

40. Braunig P, Kruger S, Shugar G. Prevalence and clinical significance of catatonic symptoms in mania. Compr Psychiatry. 1998;39(1):35-46.

41. Akiskal HS, Benazzi F. Atypical depression: a variant of bipolar II or a bridge between unipolar and bipolar II? J Affect Disord. 2005;84(2-3):209-217.

42. Benazzi F. Symptoms of depression as possible markers of bipolar II disorder. Prog Neuropsychopharmacol Biol Psychiatry. 2006;30(3):471-477.

43. Perugi G, Akiskal HS, Lattanzi L, et al. The high prevalence of “soft” bipolar (II) features in atypical depression. Compr Psychiatry. 1998;39(2):63-71.

44. Benazzi F. Depression with DSM-IV atypical features: a marker for bipolar II disorder. Eur Arch Psychiatry Clin Neurosci. 2000;250(1):53-55.

45. Benazzi F, Koukopoulos A, Akiskal HS. Toward a validation of a new definition of agitated depression as a bipolar mixed state (mixed depression). Eur Psychiatry. 2004;19(2):85-90.

46. Maj M, Pirozzi R, Magliano L, Fiorillo A, Bartoli L. Agitated “unipolar” major depression: prevalence, phenomenology, and outcome. J Clin Psychiatry. 2006;67(5):712-719.

47. Akiskal HS, Benazzi F, Perugi G, Rihmer Z. Agitated “unipolar” depression re-conceptualized as a depressive mixed state: implications for the antidepressant-suicide controversy. J Affect Disord. 2005;85(3):245-258.

48. Thase ME, Sachs GS. Bipolar depression: pharmacotherapy and related therapeutic strategies. Biol Psychiatry. 2000;48(6):558-572.

49. Bauer M, Dopfmer S. Lithium augmentation in treatment-resistant depression: Meta-analysis of placebo-controlled studies. J Clin Psychopharmacol. 1999;19(5):427-434.

50. Baldessarini RJ, Tondo L, Hennen J. Lithium treatment and suicide risk in major affective disorders: update and new findings. J Clin Psychiatry. 2003;64(suppl 5):44-52.

51. Goodwin FK, Fireman B, Simon GE, Hunkeler EM, Lee J, Revicki D. Suicide risk in bipolar disorder during treatment with lithium and divalproex. JAMA. 2003;290(11):1467-1473.

52. Fagiolini A, Buysse DJ, Frank E, Houck PR, Luther JF, Kupfer DJ. Tolerability of combined treatment with lithium and paroxetine in patients with bipolar disorder and depression. J Clin Psychopharmacol. 2001;21(5):474-478.

53. Marangell LB, Martinez JM, Ketter TA, et al. Lamotrigine treatment of bipolar disorder: data from the first 500 patients in STEP-BD. Bipolar Disord. 2004;6(2):139-143.

54. Calabrese JR, Bowden CL, Sachs GS, Ascher JA, Monaghan E, Rudd GD. A double-blind placebo-controlled study of lamotrigine monotherapy in outpatients with bipolar I depression. Lamictal 602 Study Group. J Clin Psychiatry. 1999;60(2):79-88.

55. Muzina DJ, Calabrese JR. Pharmacological profile and clinical utility of lamotrigine in bipolar disorders. In: Akiskal HS, Tohen M, eds. Bipolar Psychopharmacotherapy: Caring for the Patient. Hoboken, NJ: John Wiley & Sons; 2006:44-62.

56. Frye MA, Ketter TA, Kimbrell TA, et al. A placebo-controlled study of lamotrigine and gabapentin monotherapy in refractory mood disorders. J Clin Psychopharmacol. 2000;20(6):607-614.

57. Nierenberg AA, Ostacher MJ, Calabrese JR, K, et al. Treatment-resistant bipolar depression: a STEP-BD equipoise randomized effectiveness trial of antidepressant augmentation with lamotrigine, inositol, or risperidone. Am J Psychiatry. 2006;163(2):210-216.

58. Calabrese JR, Suppes T, Bowden CL, et al. A double-blind, placebo-controlled, prophylaxis study of lamotrigine in rapid-cycling bipolar disorder. Lamictal 614 Study Group. J Clin Psychiatry. 2000;61(11):841-850.

59. Bowden CL, Asnis GM, Ginsberg LD, Bentley B, Leadbetter R, White R. Safety and tolerability of lamotrigine for bipolar disorder. Drug Saf. 2004;27(3):173-184.

60. Baker RW, Tohen M, Fawcett J, et al. Acute dysphoric mania: treatment response to olanzapine versus placebo. J Clin Psychopharmacol. 2003;23(2):132-137.

61. Vieta E, Calabrese JR, Hennen J, et al. Comparison of rapid-cycling and non-rapid-cycling bipolar I manic patients during treatment with olanzapine: analysis of pooled data. J Clin Psychiatry. 2004;65(10):1420-1428.

62. Tohen M, Vieta E, Calabrese J, et al. Efficacy of olanzapine and olanzapine-fluoxetine combination in the treatment of bipolar I depression. Arch Gen Psychiatry. 2003;60(11):1079-1088.

63. 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.

64. Calabrese JR, Keck PE Jr, Macfadden W, et al. A randomized, double-blind, placebo-controlled trial of quetiapine in the treatment of bipolar I or II depression. Am J Psychiatry. 2005;162(7):1351-1360.

65. Thase ME, MacFadden W, Weisler RH, et al. Efficacy of quetiapine monotherapy in bipolar I and II depression: a double-blind, placebo-controlled study (the bolder II study). J Clin Psychopharmacol. In press.

66. Newcomer JW, Haupt DW, Flavin KS, et al. Atypical antipsychotic treatment effects on metabolic outcome. Schizophr Bull. 2005;31(2):566.

67. Melkersson K, Dahl ML. Adverse metabolic effects associated with atypical antipsychotics: literature review and clinical implications. Drugs. 2004;64(7):701-723.

68. Stroup TS, Lieberman JA, McEvoy JP, et al. Effectiveness of olanzapine, quetiapine, risperidone, and ziprasidone in patients with chronic schizophrenia following discontinuation of a previous atypical antipsychotic. Am J Psychiatry. 2006;163(4):611-622.

69. Sachs GS, Lafer B, Stoll AL, et al. A double-blind trial of bupropion versus desipramine for bipolar depression. J Clin Psychiatry. 1994;55(9):391-393.

70. Leverich GS, Altshuler LL, Frye MA, et al. Risk of switch in mood polarity to hypomania or mania in patients with bipolar depression during acute and continuation trials of venlafaxine, sertraline, and bupropion as adjuncts to mood stabilizers. Am J Psychiatry. 2006;163(2):232-239.

71. Thase ME, Shelton RC, Khan A. Treatment with venlafaxine extended release after SSRI nonresponse or intolerance: a randomized comparison of standard- and higher-dosing strategies. J Clin Psychopharmacol. 2006;26(3):250-258.

72. Vieta E, Martinez-Aran A, Goikolea JM, et al. A randomized trial comparing paroxetine and venlafaxine in the treatment of bipolar depressed patients taking mood stabilizers. J Clin Psychiatry. 2002;63(6):508-512.

73. Thase ME. Effects of venlafaxine on blood pressure: a meta-analysis of original data from 3744 depressed patients. J Clin Psychiatry. 1998;59(10):502-508.

74. Thase ME, Trivedi MH, Rush AJ. MAOIs in the contemporary treatment of depression. Neuropsychopharmacology. 1995;12(3):185-219.

75. Himmelhoch JM, Fuchs CZ, Symons BJ. A double-blind study of tranylcypromine treatment of major anergic depression. J Nerv Ment Dis. 1982;170(10):628-634.

76. Himmelhoch JM, Thase ME, Mallinger AG, Houck P. Tranylcypromine versus imipramine in anergic bipolar depression. Am J Psychiatry. 1991;148(7):910-916.

77. Thase ME, Mallinger AG, McKnight D, Himmelhoch JM. Treatment of imipramine-resistant recurrent depression, IV: A double-blind crossover study of tranylcypromine for anergic bipolar depression. Am J Psychiatry. 1992;149(2):195-198.

78. Beasley CM Jr, Masica DN, Heiligenstein JH, Wheadon DE, Zerbe RL. Possible monoamine oxidase inhibitor-serotonin reuptake inhibitor interaction: fluoxetine clinical data and preclinical findings. J Clin Psychopharmacol. 1993;13(5):312-320.

79. Thase ME. Novel transdermal delivery formulation of the monoamine oxidase inhibitor selegiline nearing release for treatment of depression. J Clin Psychiatry. 2006;67(4):671-672.

80. Bauer M, London ED, Rasgon N, et al. Supraphysiological doses of levothyroxine alter regional cerebral metabolism and improve mood in bipolar depression. Mol Psychiatry. 2005;10(5):456-469.

81. Carlson PJ, Merlock MC, Suppes T. Adjunctive stimulant use in patients with bipolar disorder: treatment of residual depression and sedation. Bipolar Disord. 2004;6(5):416-420.

82. Goldberg JF, Burdick KE, Endick CJ. Preliminary randomized, double-blind, placebo-controlled trial of pramipexole added to mood stabilizers for treatment-resistant bipolar depression. Am J Psychiatry. 2004;161(3):564-566.

83. Cassano P, Lattanzi L, Soldani F, et al. Pramipexole in treatment-resistant depression: an extended follow-up. Depress Anxiety. 2004;20(3):131-138.

84. Sohn CH, Lam RW. Treatment of seasonal affective disorder: unipolar versus bipolar differences. Curr Psychiatry Rep. 2004;6(6):478-485.

85. Altshuler LL, Frye MA, Gitlin MJ. Acceleration and augmentation strategies for treating bipolar depression. Biol Psychiatry. 2003;53(8):691-700.

86. Lam DH, Watkins ER, Hayward P, et al. A randomized controlled study of cognitive therapy for relapse prevention for bipolar affective disorder: outcome of the first year. Arch Gen Psychiatry. 2003;60(2):145-152.

87. Frank E, Kupfer DJ, Thase ME, et al. Two-year outcomes for interpersonal and social rhythm therapy in individuals with bipolar I disorder. Arch Gen Psychiatry. 2005;62(9):996-1004.

88. Miklowitz DJ, George EL, Richards JA, Simoneau TL, Suddath RL. A randomized study of family-focused psychoeducation and pharmacotherapy in the outpatient management of bipolar disorder. Arch Gen Psychiatry. 2003;60(9):904-912.

89. Sackeim HA, Haskett RF, Mulsant BH, et al. Continuation pharmacotherapy in the prevention of relapse following electroconvulsive therapy: a randomized controlled trial. JAMA. 2001;285(10):1299-1307.

90. Nahas Z, Marangell LB, Husain MM, et al. Two-year outcome of vagus nerve stimulation (VNS) for treatment of major depressive episodes. J Clin Psychiatry. 2005;66(9):1097-1104.