Joseph F. Goldberg
`
`Treatment Guidelines:
`Current and Future Management of Bipolar Disorder
`
`Joseph F. Goldberg, M.D.
`
`© C
`
`The emergence of new treatments for bipolar disorder has coincided with a proliferation of pub-
`lished treatment algorithm recommendations and practice guidelines. Several guidelines derive from
`critical appraisals of current treatment literature and, as such, may serve as a critical reference re-
`source to complement individual clinical judgment. This review describes points of overlap and dis-
`cordance across currently available treatment guidelines for bipolar disorder and presents common
`clinical situations in which the consultation of treatment guidelines may provide clinicians with useful
`information and a rationale for making sequential treatment decisions.
`(J Clin Psychiatry 2000;61[suppl 13]:12–18)
`
`opyright 2001 Physicians Postgraduate Press, Inc.
`
`RDespite the availability of new anticonvulsants with pos-
`
`ecent years have witnessed the growth of an array of
`treatment options for all phases of bipolar disorder.
`
`ods for guideline development.1 To the extent that guide-
`lines offer readers a critical distillation of published treat-
`ment studies, some authors have observed that the strength
`of their recommendations relies on how well they account
`for (1) study designs (i.e., randomized clinical trials versus
`observational studies), (2) heterogeneity of patients stud-
`ied (greater heterogeneity across studies weakens their
`comparability), and (3) reporting of nonoverlapping confi-
`dence intervals around effect sizes (stronger recommenda-
`tions are warranted when the smallest effect, or lower
`boundary of the confidence interval, remains above the
`threshold below which negative outcomes outweigh ben-
`efits).2 At the same time, as noted by Kahn et al.,3 pub-
`lished evidence for treatment outcomes may be incom-
`plete or poorly applicable to usual practice circumstances;
`guidelines drawn from consensus-based expert opinions
`may then complement those that are solely evidence based
`and partly compensate for gaps in the empirical database.
`While clinical decisions for an individual patient usu-
`ally defy generic or formulaic procedures and complex
`situations often lack either definitive or generalizable ad-
`vice, guidelines can augment individual clinical judgment
`by summarizing reasonable options for initial and succes-
`sive treatments. In this sense, guidelines may be regarded
`as a resource document to consult in the course of medical
`decision making (Table 1). A guideline may provide useful
`recommendations and rationales for managing difficult
`clinical problems, for example by (1) formulating a treat-
`ment plan for bipolar prophylaxis during pregnancy (re-
`vised Expert Consensus Guidelines4 advocate either con-
`ventional or atypical neuroleptics as first-line treatments
`during both conception and the first trimester), (2) advis-
`ing patients on the longevity of mood stabilizer use after
`a single hypomanic episode (a decision often based on
`the severity of the episode and family history), (3) consid-
`ering the role of antidepressants in mixed mania (Depart-
`
`One personal copy may be printed
`
`sible mood-stabilizing properties, antidepressants, atypi-
`cal antipsychotics, and diagnosis-specific psychothera-
`pies, complex forms of bipolar disorder remain prevalent,
`and suboptimal treatment responses often necessitate se-
`rial pharmacotherapy trials. In managing complex forms
`of illness, clinicians often select from among diverse treat-
`ment options with little guidance from established criteria
`or systematic methodologies. Because sequential random-
`ized drug trials for the same bipolar cohort have not, as
`yet, been reported in the literature, serial treatment strate-
`gies remain largely an area guided more by clinical judg-
`ment and opinion than empirical study. This article will
`review concepts about the use of current treatment guide-
`lines for bipolar disorder, drawing especially on their
`value as resource documents for issues related to complex
`clinical management.
`Treatment guidelines offer a frame of reference for
`choosing from among the myriad of clinical options now
`available for all phases of bipolar illness. Much as the
`strength of data to support specific treatments varies
`greatly—from anecdotal case reports to well-designed
`randomized controlled trials—so too do guidelines vary in
`the degree to which they specify and grade the evidence
`for their recommendations or follow other standard meth-
`
`From Weill Medical College of Cornell University and the
`Payne Whitney Clinic, New York Presbyterian Hospital, New
`York.
`Presented at the Bipolar Disorder Advisory Summit, which
`was held August 13–15, 1999, in New York, N.Y., and supported
`by Janssen Pharmaceutica, L.P.
`Reprint requests to: Joseph F. Goldberg, M.D., Payne
`Whitney Clinic, 525 E. 68th St., New York, NY 10021
`(e-mail: JFGoldbe@mail.med.cornell.edu).
`
`12
`
`J Clin Psychiatry 2000;61 (suppl 13)
`
`1 of 7
`
`Alkermes, Ex. 1068
`
`

`

`Treatment Guidelines for Bipolar Disorder
`
`© C
`
`Table 1. Common Clinical Problems as Addressed by Treatment Guidelines
`When should a mood stabilizer be used indefinitely?
`If bipolar I with > 2 manic episodes or 1 manic episode if especially severe or strong bipolar family history4,14,25
`What is the optimal duration of antidepressant use after remission from bipolar depression?
`2–6 months after euthymia, although 25% of experts recommended indefinite treatment4
`Should antidepressants be used as monotherapy for bipolar depression?
`Originally described as an “occasional consideration” for bipolar II depression,14 but revised edition advises against the use of antidepressants
`without a mood stabilizer4
`Is there a preferred mood stabilizer for rapid cycling?
`Divalproex as first line,4,24 although some guidelines advocate lithium with equal or greater endorsement5
`Is there a preferred mood stabilizer for mixed mania?
`Divalproex4 or carbamazepine24 may be a treatment of choice, although some guidelines advocate lithium as being at least comparable to
`divalproex, based on current data5
`When is ECT indicated in bipolar disorder?
`For depressions with psychotic or suicidal features,31 especially after nonresponse to a mood stabilizer and 2 antidepressant trials,4 or for
`depressions unresponsive to trials of (cid:42) 2 mood stabilizers + 2 antidepressants4
`For rapid response12
`For pure or mixed manias unresponsive to prior mood stabilizers4,11,24
`As a later intervention for rapid cycling with current depression unresponsive to other pharmacotherapies4
`When are 2 or more mood stabilizers indicated?
`As 2nd- or 3rd-line treatment for acute euphoric mania after nonresponse to a single agent mood stabilizer4,11,12,15,24 or after nonresponse to 2
`different mood stabilizers, including lithium5
`In mixed states, as next step if unresponsive to divalproex4 or carbamazepine11 or only partially responsive to a single mood stabilizer4
`As 2nd-line for rapid cycling if no response to single agent mood stabilizers4,5,12
`As next intervention for bipolar depression if nonresponse to single agent mood stabilizer,5 especially if a breakthrough episode while on lithium
`or divalproex monotherapy4
`When should atypical antipsychotic medications be used in bipolar disorder?
`For psychosis associated with mania or depression5,12,15,24,31
`As an alternative 2nd-line monotherapy for rapid cycling (after divalproex, lithium, or carbamazepine), especially for manic phase4
`During the first trimester of pregnancy4 (conventional antipsychotics considered first line4)
`What is the role in bipolar disorder for newer anticonvulsants such as lamotrigine, gabapentin, and topiramate?
`Lamotrigine is considered an acceptable 1st-line mood stabilizer for bipolar depression or as augmentation for lithium or divalproex during
`breakthrough depressions with or without rapid cycling4
`Lamotrigine and gabapentin are both viewed as reasonable experimental options for acute mania after nonresponses to standard mood stabilizers,
`atypical antipsychotics, and/or electroconvulsive therapy5,11; in revised Expert Consensus Guidelines,4 lamotrigine is not recommended as a later
`intervention for acute mania except among patients with rapid cycling
`Topiramate and gabapentin are both considered appropriate later options after nonresponses to lithium, divalproex, and/or carbamazepine for acute
`mania with or without rapid cycling4
`Topiramate is considered an acceptable 2nd-line augmentation to promote weight loss when necessary (after diet and exercise counseling)4
`
`One personal copy may be printed
`
`opyright 2001 Physicians Postgraduate Press, Inc.
`
`ment of Veterans Affairs [VA] Practice Guidelines5 recom-
`mend avoiding antidepressants in mixed states), (4) con-
`templating the safety and efficacy of an antidepressant
`when used unopposed by a mood stabilizer in bipolar II
`depression (revised Expert Consensus Guidelines4 recom-
`mend initial therapy with a mood stabilizer in all phases of
`bipolar illness).
`On a broader level, Rush and colleagues6 have noted
`that guidelines may facilitate clinical decision making, re-
`duce clinically inappropriate or cost-inefficient variation
`in practice patterns, provide consistent treatment across
`different environments, individualize treatment, and in-
`crease cost-efficiency of treatment. They further acknowl-
`edge risks associated with the use of guidelines. These in-
`clude the potential for recommendations to be formulated
`on the basis of insufficient evidence or biased opinions, in-
`creased costs and service utilization related to training cli-
`nicians, the possibility that guidelines could be misused to
`substitute for clinical judgment, and the reality that com-
`plex cases often defy generalization in their treatment.
`Potential barriers to the adoption of guidelines, as noted
`by Gilbert et al.,7 involve physicians’ perceptions about
`guidelines (e.g., being “told” what to do) coupled with ad-
`
`ditional training and work related to implementing guide-
`lines, the potential “static” nature of guidelines set against
`constant change and advancement in new treatments or
`applications, and the potential for patient nonadherence
`to guideline-directed treatment (although studies of de-
`pression treatment in primary care settings suggest that
`longer-term patient compliance with pharmacotherapy
`may be higher during guideline-based interventions than
`with treatment as usual8).
`Eddy9 distinguished treatment standards from guide-
`lines and options. Standards reflect recommendations that
`apply in nearly all instances and almost always result in
`the best possible outcome. In contrast, guidelines describe
`treatment interventions that produce the best outcome
`most of the time, but not in almost all instances. Options
`refer to multiple treatment alternatives that produce simi-
`lar outcomes, but lack evidence that one is clearly superior
`to another. The importance of these distinctions becomes
`apparent when one considers the potential for misuse of
`treatment guidelines in legal or administrative settings
`should they be misconstrued as defining the standard of
`care within the field or the limits of reimbursable care in
`specific clinical situations.
`
`J Clin Psychiatry 2000;61 (suppl 13)
`
`13
`
`2 of 7
`
`Alkermes, Ex. 1068
`
`

`

`Joseph F. Goldberg
`
`Clinicians often must extrapolate from published stud-
`ies when adapting their findings to ordinary practice con-
`ditions. For example, randomized clinical drug trials typi-
`cally exclude patients with comorbid substance abuse, yet
`in community-based samples, 60% or more of bipolar pa-
`tients have histories of a substance use disorder.10 Re-
`ported outcomes for special subpopulations (e.g., bipolar
`patients with mixed mania or rapid cycling) sometimes
`derive from post hoc analyses of previously collected data
`sets, rather than the a priori randomization of unique pa-
`tient groups to different treatment arms. Treatment studies
`also vary in the adequacy of statistical power and sample
`sizes and the inclusion of concomitant pharmacotherapies
`or psychosocial treatments. Individual treatment studies
`must further be contextualized by the scarcity of published
`negative results from clinical trials and the dearth of well-
`designed polypharmacotherapy trials in bipolar illness.
`Guidelines generally attempt to account for constraints
`such as these in their efforts to assimilate formal recom-
`mendations, although they may offer only a starting point
`for highly idiosyncratic clinical situations.
`
`© C
`
`EXISTING PRACTICE GUIDELINES
`
`Several published guidelines or algorithms have gained
`particular attention for their breadth and scope, the en-
`dorsement of major organizations (e.g., the American Psy-
`chiatric Association [APA]), their applicability to critical
`patient populations (e.g., individuals seen in the VA), or
`their implementation in empirical treatment studies (e.g.,
`the Texas Medication Algorithm Project [TMAP]6,7,11).
`The rationale and development of these guidelines may be
`summarized as follows:
`
`opyright 2001 Physicians Postgraduate Press, Inc.
`
`and general practitioners. Specific recommenda-
`tions are annotated by supporting literature.
`• The Texas Medication Algorithm Project6,7,11 in-
`volved a Rand-style survey of academicians and
`clinicians, followed by a consensus conference
`that led to the formulation of a multistep algorithm,
`subsequently implemented at 16 sites.
`• The Canadian Network for Mood and Anxiety
`Treatments (CANMAT)15 was developed by a
`group of clinicians and clinical researchers from
`across Canada and methodologically incorporated
`a large-scale literature review and classification of
`the quality of existing evidence. Initial algorithm
`recommendations were reviewed by 206 psychia-
`trists and 91 family practitioners, with further cri-
`tique by additional clinicians in both Canada and
`the United States.
`
`Since the introduction of treatment guidelines, both
`within psychiatry and elsewhere in medicine, questions re-
`main about how their availability affects clinicians’ actual
`practice patterns. What factors affect clinician adherence
`to guidelines, and how do patients treated according to
`guideline recommendations differ in their treatment out-
`comes from those who receive treatment as usual?
`Regarding clinicians’ reactions to practice guidelines,
`Cabana et al.16 identified several potential barriers to the
`use of practice guidelines for primary care medicine. Over
`half of physician survey respondents cited a number
`of obstacles to guideline use, including a lack of awareness
`of guidelines, a lack of familiarity with their use, and
`disagreement with guideline recommendations. Less fre-
`quently cited potential obstacles to guideline use included
`a lack of physician self-efficacy, lack of outcome expec-
`tancy, and external factors (e.g., guidelines perceived as in-
`convenient, cumbersome, or confusing). Extensive data on
`the outcome of patients treated by guideline recommen-
`dations versus treatment as usual are not yet available,
`although as noted previously, antidepressant pharmaco-
`therapy compliance was found to be higher when guide-
`line-based approaches for depression were used in primary
`care medical settings.8 In addition, an open trial17 of guide-
`line-based treatment for severely and persistently mentally
`ill bipolar outpatients found that at least a 30% improve-
`ment from baseline levels of psychopathology was evident
`after 4 months in over half of patients.
`
`One personal copy may be printed
`
`• The Practice Guideline for the Treatment of Pa-
`tients With Bipolar Disorder12 was developed by
`the APA in 1994. As described by Zarin et al.,13 the
`initial draft for this document was created by an
`expert work group, combined with a literature
`review and subsequent review by 120 individuals
`and 40 organizations.
`• In 1996, The Expert Consensus Guideline Series14
`reported the aggregate opinions from survey re-
`sults among 68 identified experts in the treatment
`of bipolar illness in response to specific clinical
`situations. A revision of these guidelines, involv-
`ing a different expert cohort of 65 clinical investi-
`gators, was recently published.4
`• The Clinical Practice Guidelines for Bipolar Dis-
`order From the Department of Veterans Affairs5
`was developed by an initial literature review, fol-
`lowed by consumer input via focus groups. A
`14-member work group summarized the initial rec-
`ommendations, which were then critiqued by 10
`non-VA experts along with input from other experts
`
`SPECIFIC CLINICAL SITUATIONS
`
`Acute Mania
`As summarized in Table 2, first-line interventions for
`acute mania in most current practice guidelines involve
`the use of a mood stabilizer as monotherapy, typically ei-
`ther lithium or divalproex sodium. Guidelines vary in their
`elaboration of additional points for management consider-
`
`14
`
`J Clin Psychiatry 2000;61 (suppl 13)
`
`3 of 7
`
`Alkermes, Ex. 1068
`
`

`

`Treatment Guidelines for Bipolar Disorder
`
`Later Interventions
`
`ECT; gabapentin; topiramate
`
`Consider clozapine, lamotrigine,
`gabapentin
`
`ECT; lamotrigine; gabapentin
`
`Expert Consensus4
`
`Lithium or divalproex
`
`VA
`
`TMAP
`
`Lithium; discontinue antidepressants;
`benzodiazepine for insomnia/
`agitation; neuroleptic if psychotic
`Divalproex or lithium
`
`© C
`
`Table 2. Treatments for Acute Euphoric/Classic Mania Across Practice Guidelinesa
`Guideline
`1st-Line Treatment
`Next Interventions
`APA
`Lithium, divalproex, or carbamazepine;
`If no response by 2–3 weeks, add 2nd
`ECT for rapid response
`mood stabilizer; ECT; adjunctive
`benzodiazepines or neuroleptics
`if needed
`Benzodiazepine; atypical neuroleptic;
`divalproex + lithium; change atypical
`neuroleptic; lithium + divalproex +
`carbamazepine
`If no response by 3 weeks, change mood
`stabilizers; combine 2 mood stabilizers
`if partial response
`Anticonvulsant + lithium; different
`anticonvulsant plus lithium;
`divalproex + carbamazepine; atypical
`neuroleptic + mood stabilizer
`2 mood stabilizers or switch to different
`mood stabilizer if partial or
`nonresponse
`
`opyright 2001 Physicians Postgraduate Press, Inc.
`
`CANMAT
`
`Add carbamazepine to lithium
`Lithium or divalproex; ECT for
`or divalproex if no response;
`severe behavior disturbance; add
`reconsider ECT; add lamotrigine,
`neuroleptic (cid:40) benzodiazepine if
`gabapentin, risperidone, or
`psychotic; add benzodiazepine (cid:40)
`calcium channel blocker;
`neuroleptic for marked behavior
`consider clozapine
`disturbance
`aAbbreviations: APA = American Psychiatric Association Practice Guideline for the Treatment of Patients With Bipolar Disorder,12
`CANMAT = Canadian Network for Mood and Anxiety Treatments,15,24 ECT = electroconvulsive therapy, TMAP = Texas Medication Algorithm
`Project,6,7,11 VA = Department of Veterans Affairs.5
`
`One personal copy may be printed
`
`manias, largely in the context of a literature that describes
`a differential treatment response to anticonvulsant mood
`stabilizers such as divalproex20,21 or carbamazepine22 as
`compared with lithium, as well as a different course of ill-
`ness and longer time to recovery in mixed versus pure ma-
`nia.23 As outlined in Table 3, some guidelines embrace this
`literature by recommending divalproex4,14 and/or carba-
`mazepine11,24 as the initial treatment of choice for dyspho-
`ric mania, while others regard existing data as more provi-
`sional and recommend lithium as the first-line treatment
`for both pure and mixed manias.5
`Evidence-based information about continuation ther-
`apy and long-term prophylaxis in bipolar disorder is not
`extensive. As described in Table 4, most guidelines advo-
`cate the long-term or indefinite use of a mood stabilizer
`for all bipolar I patients who have had 2 or more manias or
`1 “severe” mania, especially those with a family history of
`bipolar disorder.4,11,15,25
`
`Bipolar Depression
`Controversy persists regarding the use of antidepres-
`sants in patients with bipolar disorder on the basis of lit-
`erature which suggests that antidepressants may induce
`manias in at least one third of bipolar patients26,27 and may
`hasten cycle accelerations via a kindling mechanism in ap-
`proximately one quarter of bipolar patients who develop
`rapid cycling.27 At the same time, some of the literature
`supports the safety and efficacy of standard antidepres-
`sants as monotherapies for depression in patients with bi-
`polar II disorder, including fluoxetine28 and venlafaxine.29
`Few empirical data substantiate recommendations about
`the relative merits of treating bipolar depression with mul-
`
`ation. For example, when a rapid response is crucial, some
`guidelines favor the use of divalproex14 or electroconvul-
`sive therapy (ECT).12 VA practice guidelines5 emphasize
`the elimination of antidepressants in the management of
`acute mania.
`Despite the broadening use of polypharmacology regi-
`mens for both the acute and long-term treatment of bipolar
`disorder,18 existing guidelines generally regard mood-
`stabilizer monotherapy as an optimal initial strategy, al-
`though they vary about whether to augment with a second
`mood stabilizer, if necessary, as a next step4,12,14 or to
`switch to a different mood stabilizer altogether if no re-
`sponse occurs.4,5 In the TMAP,11 sequential approaches to
`nonresponse are described in which dual therapy with lith-
`ium plus an anticonvulsant mood stabilizer is recom-
`mended after nonresponse to monotherapy with either
`agent, followed by dual anticonvulsants, the introduction
`of atypical neuroleptics, then ECT, then more experi-
`mental agents (e.g., lamotrigine, gabapentin). At present,
`antipsychotic medications (either conventional or atypi-
`cal) are not recommended as first-line monotherapies or
`adjunctive agents to treat euphoric mania, unless psycho-
`sis clearly is present4 or management of behavioral agita-
`tion is needed.24 However, a newly emerging database on
`the efficacy of atypical antipsychotics such as olanzapine
`as monotherapy for acute mania19 may prompt the reas-
`sessment of their role in future guidelines. Similarly, the
`adjunctive use of benzodiazepines for agitation is usually
`described as appropriate augmentation if and when clini-
`cally necessary.
`In several guidelines, mixed states or dysphoric manias
`have been accorded separate commentary from euphoric
`
`J Clin Psychiatry 2000;61 (suppl 13)
`
`15
`
`4 of 7
`
`Alkermes, Ex. 1068
`
`

`

`Joseph F. Goldberg
`
`Table 3. Treatments for Mixed/Dysphoric Mania Across Practice Guidelinesa
`Guideline
`1st-Line Treatment
`Next Interventions
`APA
`Same as for pure mania;
`Same as for pure mania
`discontinue antidepressants
`Divalproexb
`Lithium; discontinue antidepressants
`Divalproex or carbamazepine
`
`Expert Consensus4
`VA
`TMAP
`
`Same as for pure mania
`Same as for pure mania
`Carbamazepine + lithium or
`divalproex + lithium;
`divalproex + carbamazepine
`2 mood stabilizers or switch to
`different mood stabilizer if
`partial or nonresponse
`
`CANMAT
`
`Divalproex or carbamazepine;
`+ neuroleptic if mood-incongruent
`psychosis; (cid:40) benzodiazepine for
`behavior disturbance
`
`© C
`
` aAbbreviations are explained in the first footnote to Table 2.
`bDivalproex considered treatment of choice.
`
`Table 4. Continuation and Maintenance Treatment Across Practice Guidelinesa
`Guideline
`1st-Line Treatment
`Next Interventions
`APA
`Divalproex or carbamazepine
`
`Lithium
`
`Lithium and divalproex
`
`Carbamazepine: 2nd line
`
`Later Interventions
`
`Add atypical neuroleptic; consider
`ECT; lamotrigine; gabapentin
`
`Add carbamazepine to lithium or
`divalproex if no response;
`consider ECT; add lamotrigine,
`gabapentin, risperidone, or
`calcium channel blocker;
`consider clozapine
`
`Additional Comments
`Longevity of treatment based
`on “individual risks/benefits”
`Lifetime prophylaxis after 2 episodes
`of mania or 1 episode of severe
`mania; bipolar II after 3 episodes of
`hypomania or antidepressant-
`induced mania
`Taper neuroleptics or benzodiazepines;
`psychosocial rehabilitation emphasized
`Prophylaxis after 2 episodes of mania
`or 1 episode of mania with positive
`family history; use lowest doses to
`achieve therapeutic blood levels,
`taper adjunctive medications
`Indefinite prophylaxis if history of
`recurrent episodes, especially if
`severe or with positive family
`history of bipolar disorder; after a
`single episode of low severity, may
`taper off pharmacotherapy after 6–12
`months over a 1–3 month period, and
`monitor annually
`
`tiple mood stabilizers versus a single mood stabilizer plus
`an antidepressant, although a recent report30 suggested a
`superior response to lithium or valproate plus paroxetine
`as compared with lithium plus valproate. Unlike the case
`for unipolar depression, sequential trials of antidepres-
`sants are shunned by some guidelines31 after nonresponse
`to an initial antidepressant, in favor of other serial inter-
`ventions (Table 5). Data also are scant regarding the opti-
`mal duration of antidepressant use after the remission of
`depressive symptoms, although some guidelines4,25 advise
`tapering off antidepressants as soon as 6 to 12 weeks after
`remission.
`Despite these limitations, most treatment guidelines ad-
`vocate the use of a mood stabilizer at optimal doses as a
`first-line intervention for the treatment of pure depressed
`phases of bipolar disorder. Lithium is ranked as a first
`choice in some guidelines,5,12,31 with combinations of
`
`mood stabilizers and/or the addition of antidepressants re-
`served for nonresponders after several weeks.5,12 The si-
`multaneous initiation of a mood stabilizer and an antide-
`pressant is described as an appropriate first step in the
`TMAP.11 The CANMAT guidelines31 propose rapid con-
`sideration of ECT in the presence of either suicidality or
`psychosis.
`
`Specific Antidepressants
`The introduction of antidepressants is generally ad-
`vised after nonresponse to one (or more5) mood stabiliz-
`ers. Reflecting the small database of clinical trials using
`standard antidepressants for bipolar depression, guidelines
`that recommend particular antidepressants tend to favor
`bupropion or selective serotonin reuptake inhibitors
`(SSRIs; especially paroxetine, studied in one double-blind
`trial32) as first-line agents.4,11 In the revised Expert Con-
`
`16
`
`J Clin Psychiatry 2000;61 (suppl 13)
`
`opyright 2001 Physicians Postgraduate Press, Inc.
`
`Expert Consensus4
`
`VA
`
`TMAP
`
`CANMAT
`
`One personal copy may be printed
`
`Preferred agents and duration of
`prophylaxis not specified
`No specific agent(s) preferred
`
`Maintain mood stabilizer at optimal
`levels, taper off benzodiazepine
`(cid:40) neuroleptics once asymptomatic
`for 2–3 weeks; taper after 6–12
`weeks of euthymia
`
`aAbbreviations are explained in the first footnote to Table 2.
`
`5 of 7
`
`Alkermes, Ex. 1068
`
`

`

`Table 5. Treatment of Bipolar Depression Across Guidelinesa
`Guideline
`1st-Line Treatment
`APA
`Begin and optimize mood stabilizer (lithium preferred)
`Expert Consensus4
`Begin and optimize mood stabilizer
`VA
`Begin and optimize mood stabilizer (lithium 1st choice)
`
`Treatment Guidelines for Bipolar Disorder
`
`Next Interventions
`Add specific psychotherapy and/or antidepressant
`
`Add divalproex or carbamazepine to lithium if no response
`after 2–4 weeks; then, conservative use of antidepressants
`at “lowest doses for shortest possible times” (no specific
`agents preferred); ECT if no response
`Switch antidepressants (SSRI > bupropion or vice-versa;
`nefazodone; venlafaxine); then 2 antidepressants + mood
`stabilizer; then MAOI + mood stabilizer; then ECT; then
`experimental agents (eg, lamotrigine)
`CBT or IPT if mild severity; 2 mood stabilizers or mood
`stabilizer + antidepressant; neuroleptic if psychotic; if
`nonresponse: 3 mood stabilizers or clozapine, ECT, or
`novel treatments
`aAbbreviations: CBT = cognitive-behavioral therapy, IPT = interpersonal psychotherapy, MAOI = monoamine oxidase inhibitor,
`SSRI = selective serotonin reuptake inhibitor. Additional abbreviations are explained in the first footnote to Table 2.
`
`TMAP
`
`CANMAT
`
`Mood stabilizer plus antidepressant (SSRI or bupropion
`preferred)
`
`© C
`
`Begin and optimize mood stabilizer; ECT if marked
`suicidality or psychosis
`
`opyright 2001 Physicians Postgraduate Press, Inc.
`
`sensus Guideline,4 first-line treatments for melancholic
`depressions included venlafaxine, bupropion, or paroxe-
`tine, followed by other SSRIs (sertraline, citalopram, and
`fluoxetine). In the presence of atypical depressive fea-
`tures, bupropion was ranked as a leading first-line treat-
`ment, followed by paroxetine, sertraline, venlafaxine, and
`citalopram. Bupropion was considered the antidepressant
`of first choice for moderate depression, followed by
`paroxetine, sertraline, citalopram, fluoxetine, and venla-
`faxine. Other guidelines list a range of appropriate antide-
`pressant classes for bipolar depression, broadly including
`SSRIs, serotonergic-norepinephrine reuptake inhibitors
`(SNRIs), monoamine oxidase inhibitors (MAOIs), and bu-
`propion.31 Many recommend avoiding tricyclic antide-
`pressants (TCAs) because of the reported increased risk
`for inductions of mania and cycle acceleration.26,27
`
`Rapid Cycling
`Originally defined by Dunner and Fieve33 as a robust
`predictor of the failure of lithium prophylaxis, rapid cy-
`cling has been associated with a potentially better relative
`response to divalproex and possibly other anticonvulsant
`mood stabilizers such as lamotrigine.34 These observations
`are reflected by guidelines that recommend divalproex as
`a mood stabilizer of choice for rapid cycling in any given
`phase.4,24 Nonetheless, some guidelines question the re-
`puted differences in treatment outcome between lithium
`and anticonvulsant mood stabilizers on the basis of the
`limited data available and recommend either as an appro-
`priate mood stabilizer in patients with rapid cycling.5
`Subsequent treatment recommendations for partial or non-
`responses generally involve adding additional mood stabi-
`lizers or atypical antipsychotics,4 followed by ECT or
`more experimental treatments if necessary (e.g., lamotri-
`gine, gabapentin, calcium channel blockers, thyroid hor-
`mone).24 Some guidelines also focus less on the specific
`choice of mood stabilizer(s) for rapid cycling than on
`other aspects of treatment, such as optimization of thyroid
`
`function12 and/or the elimination of antidepressant medi-
`cations whenever possible.5
`Reflecting recent literature on bipolar depression and
`on rapid cycling,34,35 revised Expert Consensus Guide-
`lines4 recommend lamotrigine as a first-line option for cur-
`rent depression in patients with rapid cycling. Atypical
`antipsychotics are described as a second-line alternative
`monotherapy.
`
`SUMMARY
`
`In summary, most currently published guidelines rec-
`ommend using a single mood stabilizer as a first step for
`acute mania; combinations of mood stabilizers are de-
`scribed as appropriate second steps. In mixed states, sev-
`eral guidelines favor divalproex or carbamazepine as first-
`line mood stabilizers, although others advise using lithium
`no less often than anticonvulsants. Many urge discontinu-
`ing antidepressants in both pure manias and mixed states.
`Bipolar depression should initially be treated with an
`optimally dosed mood stabilizer, but guidelines vary in
`their subsequent recommendations: an antidepressant
`plus a mood stabilizer is often described as a desirable op-
`tion. Atypical neuroleptics are widely favored over con-
`ventional neuroleptics, although their role currently re-
`mains uncertain beyond the treatment of psychosis during
`mania or depression, as second-line agents for rapid cy-
`cling, or in affective episodes unresponsive to standard
`mood stabilizers. During long-term maintenance treat-
`ment, most guidelines favor the simplification of drug
`regimens and lithium monotherapy as a first choice for
`lifetime prophylaxis.
`
`One personal copy may be printed
`
`Drug names: bupropion (Wellbutrin), carbamazepine (Tegretol and oth-
`ers), citalopram (Celexa), clozapine (Clozaril and others), divalproex
`sodium (Depakote), fluoxetine (Prozac), gabapentin (Neurontin), lamo-
`trigine (Lamictal), nefazodone (Serzone), olanzapine (Zyprexa), parox-
`etine (Paxil), risperidone (Risperdal), sertraline (Zoloft), topiramate
`(Topamax), venlafaxine (Effexor).
`
`J Clin Psychiatry 2000;61 (suppl 13)
`
`17
`
`6 of 7
`
`Alkermes, Ex. 1068
`
`

`

`Joseph F. Goldberg
`
`REFERENCES
`
`© C
`
` 1. Shaneyfelt TM, Mayo-Smith MF, Rothwangl J. Are guidelines following
`guidelines? the methodological quality of clinical practice guidelines in
`the peer-reviewed medical literature. JAMA 1999;281:1900–1905
` 2. Wilson MC, Hayward RSA, Tunis SR, et al. Users’ guides to the medical
`literature, VIII: how to use clinical practice guidelines. B: what are the rec-
`ommendations and will they help you in caring for your patients? JAMA
`1995;274:1630–1632
` 3. Kahn DA, Docherty JP, Carpenter D, et al. Consensus methods in practice
`guideline development: a review and description of a new method.
`Psychopharmacol Bu