Available online at www.sciencedirect.com

Comprehensive Psychiatry xx (2012) xxx–xxxwww.elsevier.com/locate/comppsych

Methodological differences between pharmacological treatment guidelinesfor bipolar disorder: what to do for the clinicians?

Ludovic Samalina,⁎, Sebastien Guillaumeb, Philippe Courtetb, Mocrane Abbarc,Sylvie Lancrenond, Pierre-Michel Llorcaa

aDepartment of Adult Psychiatry B, CHU Clermont-Ferrand, 58 rue Montalembert, 63000 Clermont-Ferrand, FrancebCHU Montpellier, Psychiatric Emergency and Post Emergency Department, University of Montpellier, Inserm U1061, France

cCHU Caremeau, Department of Adult Psychiatry, Nîmes, FrancedSylia-Stat, Bourg-la-Reine, France

Abstract

Objective: Numerous guidelines for bipolar disorder have been published. The aim of this article is to underline the main differences betweenconsensus-based guidelines (CBG) and evidence-based guidelines (EBG) currently available for the management of bipolar disorder.Methods: A literature search for guidelines published since 2006 was performed. A qualitative analysis was then conducted to compare themethodologies and the guidelines contents.Results: Comparison between CBG and EBG found more similarities than differences. However, discordances were found in the first-linechoice of treatment (monotherapy or combination, use of lamotrigine or lithium in bipolar depression), time to reassessment and duration ofmaintenance treatment, introduction as from the acute phase a regimen compatible with long-term use and pharmacotherapy during pregnancy.Conclusions: The choice of policy, whatever the methodology used, is up to the authors and can, therefore, depend on their interpretation ofthe available scientific evidence. Combining both methodologies (CBG and EBG) enables us to meet the complete definition of evidence-based medicine.© 2012 Elsevier Inc. All rights reserved.

1. Introduction

In the last few decades an increasing number of drugshave been approved for the treatment of bipolar disorders.Making a relevant choice of medication represents a newchallenge for clinicians. Guidelines have been established tohelp them make their choice of appropriate care, evidence-based in specific clinical circumstances. However, someauthors have pointed out that the large number of guidelinesavailable makes their use complex and may lead toconflicting conclusions [1].

Evidence, on which recommendations are based, isnumerous and the interpretation of their significance canvary widely. For a given clinical situation, the therapeuticstrategies suggested may sometimes be quite different.

There are two types of methodology used for recom-mendations. In evidence-based guidelines (EBG), a task

⁎ Corresponding author. Tel.: +33 4 73 75 21 24; fax: +33 4 73 75 21 26.E-mail address: lsamalin@chu-clermontferrand.fr (L. Samalin).

0010-440X/$ – see front matter © 2012 Elsevier Inc. All rights reserved.http://dx.doi.org/10.1016/j.comppsych.2012.10.001

force draws up recommendations after critical analysis ofavailable data, which are selected and ranked according totheir level of evidence. First-line treatment is defined ac-cording to a higher level of evidence, such as that obtainedfrom randomized controlled trials, and second- or third-linetreatments, which are based on evidence from non-controlledor non-randomized studies or on expert advice. In consen-sus-based guidelines (CBG), recommendations are based onthe advice issued from the knowledge of the literature andthe practical experience of a panel of experts, who are askedto consider specific clinical questions or situations. Thismethodology enables the assessment of clinical situations forwhich evidence (from randomized controlled trials) arescarce or debated.

Each methodology has strengths and limitations.EBGs are developed from scientific evidence and not from

opinion or judgement. Their recommendations are ranked sothat the clinician knows on what level of evidence they arebased. However, EBG cannot establish recommendationsif there is no evidence available on which to base them. In

2 L. Samalin et al. / Comprehensive Psychiatry xx (2012) xxx–xxx

most cases, if there is no recommendation for a given treat-ment it is not because the treatment is not effective butbecause there are no sufficiently reliable studies of its action.

CBGs allow the clinician to be led by recommendationsthat bear closer relation to the patient characteristics follow-up in clinical practice than the restrictive inclusion criteriaof randomized controlled trials. Most guidelines for treat-ment of bipolar disorder are EBGs. However, two recentguidelines used CBG methodology: the International Con-sensus Group [2,3] and the Formalized Consensus Guide-lines from the French Association for Biological Psychiatryand Neuropsychopharmacology (a.k.a. AFPBN: Associa-tion Française de Psychiatrie Biologique et Neuropsycho-pharmacologie) [4,5].

We argue that while there are differences between EBGand CBG for bipolar disorder, both methodologies are usefulfor clinicians in practice.

The aim of this article is to underline the main differencesbetween EBG and CBG. We will discuss the respectiveinterests and conditions of use of these recommendations.

2. Materials and methods

2.1. Choice of guidelines

We reviewed the recently published guidelines in Englishfor the management of bipolar disorder. Electronic libraryand Web-based searches were performed using recognizedtools (MEDLINE, PubMED and EMBASE) to identifyguidelines that have been published since 2006. In addition,other sources of information known to us, such as nationalagency reports, were searched.

Because of recent developments in the management ofbipolar disorder, we limited our search to guidelines publishedin the last 5years. Searches were last updated onApril 26, 2012.

2.2. Comparison of the different guidelines

The aim of this work is not to rank the methodologies as wasdone in the assessment of treatment guidelines for schizophreniaon the basis of criteria from the Appraisal of GuidelinesREsearch andEvaluation (AGREE) project [6] but to carry out aqualitative comparison of the guidelines contents according totheir methodologies. We focused our analysis on some specificfeatures of the management of bipolar patients:

- Initial pharmacological treatment and electroconvul-sive therapy (ECT) during manic, depressive episodesand mixed state,

- Clinical features or factors that influence first-linetreatment choice during an acute episode,

- Decision to initiate a treatment in the acute phasecompatible with maintenance treatment,

- Time to reassessment and duration of long-termtreatment,

- Pharmacological treatment of bipolar disorder duringpregnancy.

3. Results

3.1. Guidelines available

Since 2006, seven sets of guidelines have been estab-lished or updated by official scientific societies (Table 1).CBG methodology was used in Expert Consensus Guide-lines from Sachs et al. [7] but was published in 2000. Ac-cording to the methodology of the review, this publicationwas considered out of date and therefore no longer relevant.

3.2. Comparison between guidelines

3.2.1. First-line treatment choice during an acute episodeThe recommendations for initial first-line treatment

during an acute mania, bipolar depression and mixed statedare summarized in Table 2.

3.2.1.1. Manic episode. The different guidelines are inagreement on discontinuing antidepressants (AD) and on thechoice of drugs to be used as a first-line treatment for mania:lithium, valproate and the second-generation antipsychotics(SGA)—olanzapine, risperidone, aripiprazole, quetiapine andziprasidone. Several recent randomized controlled studies haveshown lithium or valproate in combination with an SGA to bemore effective than when used alone [8–12]. Depending onwhether these findings have been taken into account by theauthors, the recommendations issued from the guidelines on thechoice of monotherapy or immediate combination may differ.

EBG The World Federation of Societies of BiologicalPsychiatry (WFSBP), the British Association forPsychopharmacology (BAP) and the NationalInstitute for Health and Clinical Excellence(NICE) propose the use of a “first choicemedication” and give different grades of rec-ommendation for monotherapy and combination[13–15]. The authors recommend monotherapy asfirst-line treatment to minimize the risk of sideeffects. Combinations (lithium or valproate withSGA) are used only for severe mania or inadequateresponse to treatment.The Canadian Network for Mood and AnxietyTreatments (CANMAT) and the National Healthand Medical Research Council (NHMRC) guide-lines suggest the use of antimanic drugs alone orin combination possible in first-line [16,17].

CBG The expert panel from AFPBN guideline recom-mends indifferently monotherapy (with lithium orvalproate) or combinations as first-line choice [4,5].

3.2.1.2. Bipolar depressive episode. Most guidelines focuson the use of mood stabilisers (MS) and not AD, which arenot recommended in monotherapy in the management ofbipolar depressive episodes.

EBG The only drug recommended in monotherapy asfirst-line treatment by all guidelines (except the

Table 1Guidelines for biological treatment of bipolar disorder since 2006.

Authors and years of publication Established by Title Methodology

Calabrese and Kasper, 2008/2004;Frye, 2011 [2,3]

International Consensus Group (ICG) International Consensus Group: CBG/EBG- on the evidence-base pharmacologic treatmentof bipolar I and II depression- on depression prevention in bipolar disorder

Llorca et al., 2010 [4,5] Association Française de Psychiatrie Biologiqueet de Neuropsychopharmacologie (AFPBN)

Formalized Consensus Guidelines: screening andmanagement of bipolar disorder

CBG

Grunze et al., 2009–2010/2004–2002[14,18,19,21]

World Federation of Societies of biologicalPsychiatry (WFSBP)

World Federation of Societies of biologicalPsychiatry (WFSBP) guidelines for biologicaltreatment of bipolar disorders part I, II, III update

EBG

Malhi et al., 2009 [17] National Health and Medical ResearchCouncil (NHMRC)

Clinical practice recommendations forbipolar disorder

EBG

Goodwin, 2009/2003 [13] British Association for Psychopharmacology(BAP)

Evidence-based guidelines for treatingbipolar disorder

EBG

Yatham et al., 2009/2007/2005 [16] Canadian Network for Mood and AnxietyTreatments (CANMAT)

Guidelines for the management of patients withbipolar disorder

EBG

NICE, 2006 [15] National Institute for health and ClinicalExcellence (NICE)

NICE new guidance to improve the treatment ofbipolar disorder

EBG

Abbreviations: CBG, consensus-based guideline; EBG, evidence-based guideline.

3L. Samalin et al. / Comprehensive Psychiatry xx (2012) xxx–xxx

NICE guidelines) is quetiapine. The others arelithium [16,17], lamotrigine [13,16,17] and olanza-pine [17]. There are differences between authors, inthe level of evidence for lithium as monotherapy,which is recommended as first-line treatment[16,17], second-line [13,18], or not at all [15].Several EBGs recommend AD (serotonin selectivereuptake inhibitors—SSRI—or bupropion) in com-bination with MS among their first-line strategies[13,15,16].NHMRC and WFSBP guidelines are much morecritical about the AD efficacy in bipolar depression[17,18]. One specific combination (olanzapine–fluoxetine) has a sufficiently high level of evidence(as second-line treatment). There is no evidence forall the other combinations with AD. They should beused only as last-choice strategies.

CBG The first-line pharmacological treatments recom-mended by the AFPBN guideline are lamotrigine inmonotherapy and SSRI with MS or two MS incombination [4,5]. Quetiapine was approved byHealth Authorities in bipolar depression in Francein November 2011 and was not available in Francewhen this guideline was drawn up.The ICG guideline recommends three agents asfirst-line monotherapy: lithium, lamotrigine andquetiapine [2]. Olanzapine–fluoxetine is as second-line strategy and the other AD combinations withMS are recommended as last-choice strategies.

3.2.1.3. Mixed state.

EBG All guidelines recommend discontinuing AD. Incontrast, authors disagree on the first-line treatmentchoice in mixed state (Table 2). The BAP and NICEguidelines [13,15] make no distinction between

treatments of classic mania and that of mixed state.The other guidelines consider lithium to be lessefficacious [14,16,17].

CBG The AFBPN guideline does not recommend lithiummonotherapy as first-line treatment [4,5].

3.2.2. Clinical features or factors that influence first-linetreatment choice during an acute episode

Recommendations for clinical features, or factors thatinfluence first-line treatment choice during an acute episode,are summarized in Table 3.

EBG The different EBGs agree on the use of SGA formaniaor depression with psychotic symptoms [13,15–17].The other clinical features influencing first-linetreatment choice are predominant symptoms duringthe episode, severity, and risk of suicide. Someguidelines take into account the type of BD (withspecific recommendations for BD II), the presence ofrapid cycles, comorbidities, personality disorder andpatient's history of manic and depressive episodes.Characteristics of treatment (current and previoustreatment, safety profile), patients (preference, level ofinsight and adherence, women of childbearingpotential) and family, may also influence the choiceof first-line treatment.

CBG The expert panel from AFPBN recommends SGA formania or depression with psychotic symptoms [4,5].As EBG, AFBPN and ICG guidelines take intoaccount other clinical features and factors in the first-line treatment choice during an acute episode(Table 3).

3.2.3. Choice of treatment in the acute phase consistent withlong-term treatment

Whatever the guideline methodology, management of theacute phase is always considered separately from long-term

Table 2Recommendations for initial first-line treatment during an acute mania, bipolar depression and mixed state.

Evidence-based guidelines Consensus-based guidelines NICE BAP WFSBP NHRMC CANMAT AFBPN ICG

Categories/levels of evidence or

method to obtain consensus

for first-line treatment

Level 1 (++, +, -):Meta-analysis,systematic reviews of RCTs or RCTs (with very low to high risk of bias)

Level I: Meta-analysis of RCTs, at least 1 large, good quality, RCT or replicated, smaller RCTs

Category A: 2 or more positive RCTs-DB with placebo and 1 or more positive RCT with active comparator negative studies must be outweighed by at least 2 more positive studies or meta-analysis with placebo or active comparator

Level I: Systematic review of all relevant RCTs Level II: 1 or more properly designed RCT

Level 1: Meta-analysis or replicated RCTs with placebo Level 2: 1 or more RCTs with placebo or active comparator

A categorical of first-line was designed if at least 50% of the experts rated a treatment as 7-9 range on a 9-points scale. The range 7-9 was rated by expert when treatment was extremely or usually appropriate

Category 1: Evidence of efficacy in RCTs with placebo in treatment of acute bipolar depression and long-term treatment for both poles of illness Method to obtain consensus from a group of international experts was not presented

Discontinue AD Monotherapy:- Severe:SGAs, ECT

- Not severe: Li, VPA

Discontinue AD Monotherapy: - Severe:SGAs, VPA

- Not severe: Li, AC (VPA, CBZ), SGAs

Discontinue AD Monotherapy: Li, AC (VPA, CBZ), SGAs (ARP, ASN, QTP, OLZ, RSP, ZPD), HAL

Discontinue AD Monotherapy: Li, VPA, SGAs (ARP, QTP, OLZ, RSP, ZPD)

Combination: Li/VPA + SGAs

Discontinue AD Monotherapy: Li, VPA, SGAs (ARP, QTP, QTP xr, OLZ, RSP, ZPD)

Combination: Li/VPA + SGAs (ARP, QTP, OLZ, RSP)

Discontinue AD Monotherapy: Li, AC (VPA, VPM), SGAs (ARP, OLZ, RSP)*, HAL, ECT

Combination:Li/VPA + SGAs, Li + AC (VPA, VPM)

No RG (not the purpose of the guideline) Initial first-line

treatment during a manic

episode

Monotherapy Li, AC, SGAs « First choice medication » Combination Li/VPA + SGAs possible use as first line treatment Monotherapy: QTP

Monotherapy: Li, LMT, SGAs (OLZ, QTP) Initial first-line

treatment during a bipolar

depressive episode

Monotherapy : - Severe:ECT

Combination: - Severe/moderate:Li/AC/SGAs +SSRI, Li/AC + QTP, + LMT

Monotherapy : - Severe:ECT - Moderate/mild:QTP, LMT

Combination: - Moderate: Li/VPA/SGAs + SSRI

Insufficient level of evidence to usecombination MS + SSRI in first line

Monotherapy: Li, LMT, QTP

Combination: Li/VPA + SSRI OLZ+ bupropion Li + VPA

Monotherapy: LMT*, ECT

Combination:Li/AC/SGAs + SSRI

Monotherapy: Li, LMT, QTP

Discontinue AD Initial first-line treatment

during a mixed state

Consider treating patients as if they had anacute manic episode

Discontinue AD No RG (cited VPA, CBZ, OLZ, RSP, ARI, ZPD)

Discontinue AD VPA, OLZ, OFC, VPA+OLZ

Discontinue AD No RG (cited VPA, CBZ, SGAs, ECT)

Discontinue AD VPA, SGAs (ARP, OLZ, RSP), Li/VPA + SGAs

No RG (not the purpose of the guideline)

Abbreviations: AC, anticonvulsants; AD, antidepressants; ARP, aripiprazole; ASN, asenapine; CBZ, carbamazepine; ECT, electroconvulsive therapy; HAL, haloperidol; Li, lithium; LMT, lamotrigine; OLZ,olanzapine; QTP, quetiapine; RCT, randomized controlled trial; RG, recommendation grade; RSP, risperidone; SGA, second-generation antipsychotic; SSRI, selective serotonin reuptake inhibitor; VPA,valproate; VPM, valpromide; ZPD, ziprasidone. * QTP was not available in France when guideline was developed.

4L.Sam

alinet

al./Com

prehensivePsychiatry

xx(2012)

xxx–xxx

Table 3Recommendations for clinical features or factors that influence first-line treatment choice during an acute episode.

Consensus-based guidelinesEvidence-based guidelinesNICE BAP WFSBP NHRMC CANMAT AFBPN ICG

First-line treatment

during a mood episode with

psychotic features

Mania

Depression

Mania

Depression

Mania:

Depression

Mania:

Depression

Mania

Depression*

Others clinical features

influencing first-line treatment

choice

Others factors influencing first-

line treatment choice

: no RG

:add SGAs, ECT

:(add) SGAs (if psychosis is not congruent with severe affective symptoms)

:(add) SGAs

No RG add SGAs

:add SGAs

no distinction with mania without psychotic features

:SGAs + SSRI, ECT

:SGAs (ARP, OLZ, RSP)Li/VPA + SGAs

:SGAs + AD (SSRI)

No RG

Symptoms of episode, severity, suicidality, presence of rapid cycles

NB: Type of BD; recommendations for BD I should be cautiously applied to treating BD II

Symptoms of episode, severity, suicidality (for depression), patient’s history of manic and depressive episodes

Symptoms of episode, severity, suicidality (for depression), type of BD I or II (with specific RG for depression)

Symptoms of episode, severity, suicidality, risk to others, presence of rapid cycles

Symptoms of episode, severity, suicidality, risk to others, type of BD I or II (with specific RG), presence of rapid cycles, multiple prior mood episodes, patient’s history of manic and depressive episodes, presence of comorbidities and personality disorder

Symptoms of episode, severity, suicidality, type of BD I or II, predominant polarity of illness, presence of rapid cycles, presence of comorbidities and personality disorders (with specific RG)

Symptoms of episode, course of illness, type of BD I or II (with specific RG), presence of rapid cycles

Current treatment, previous treatment

Patients preference

Women of child-bearing potential,individual risk of side effects

Current long-term treatment, previous treatment

Patients and family preference/experience,

Level of treatment adherence

Women of child-bearing potential, safety profile

Previous treatment

Patients preference

Route and ease of administration

Safety profile, modifying medical factors

Status (in/outpatient, voluntary or not) Previous treatment

Patients and family preferences

Level of treatment adherence

Safety profile

Current treatment, previous treatment

Degree of insight, ability to adhere to treatment

Safety profile

Current long-term treatment

Level of treatment adherence

Safety profile

Patient and family history of response

Safety profile

Abbreviations: ARP, aripiprazole; BD, bipolar disorder; ECT, electroconvulsive therapy; Li, lithium; OLZ, olanzapine; RG, recommendation grade; RSP, risperidone; SGA, second-generation antipsychotic;SSRI, selective serotonin reuptake inhibitor; VPA, valproate. * QTP was not available in France when guideline was developed.

5L.Sam

alinet

al./Com

prehensivePsychiatry

xx(2012)

xxx–xxx

Table4

Recom

mendatio

nsfortreatm

entin

theacutephaseconsistent

with

long-term

treatm

ent.

Evidence-basedgu

idelines

Con

sensus-based

guidelines

NICE

BAP

WFSB

PNHRMC

CANMAT

AFBPN

ICG

Mania

anddepression:

Cho

osemedication

taking

into

account

side

effect

andfuture

prop

hylaxis.

Norecommendatio

n.Differentiate

short-term

andlong

-term

treatm

ent

strategies

(Choose

medicationin

theacute

phasewith

outtaking

into

accoun

tfuture

prop

hylaxis)

Mania

anddepression:

Choosemedication

consideringevidence

forefficacy

asmaintenance

treatm

ent

ifappropriate.

Man

ia:Con

sidermedication

efficacy

andtolerabilityand

also

factor

thelik

elihood

ofcontinuing

acutetreatm

ent

into

maintenance

phase.

Mania:Patientstreated

successfully

with

acute

treatm

entcancontinue

ifappropriateinto

maintenance

phase

Man

iaan

ddepression

:Cho

osemedicationtaking

into

accoun

tpredom

inant

polarity

ofillness

and

presence

ofrapidcycles.

Depression:

Levelof

evidence

forfirst-lin

etreatm

enttaking

into

accoun

tdrug

s'efficacy

inacute

bipo

lardepression

andlong

-term

treatm

ent

Depression:

No

recommendatio

n.Depression:

Taper

antid

epressantsin

the

maintenance

phase

(ifused),with

the

exceptionof

thosewith

high

lyrecurrentbipo

lar

Idepressive

episod

es.

Maintenan

cetreatment:

Con

tinuing

thetreatm

ent

initiated

intheacute

phase(ifithas

acknow

ledged

preventiv

eeffect)andtaking

into

accountthepredom

inant

polarity

ofthepatient.

Depressionprophylaxis:

Contin

uesuccessful

medication

from

theacuteph

ase.Taper

antid

epressantsin

the

maintenance

phase(ifused),

unless

thepatient

hasahistory

ofrelapseafterdiscontin

uing

antid

epressants.

Maintenan

cetreatment:

Contin

uetheindex

medicationused

for

theacuteepisod

e.

6 L. Samalin et al. / Comprehensive Psychiatry xx (2012) xxx–xxx

treatment. Recommendations for treatment in the acutephase consistent with long-term treatment are summarized inTable 4.

EBG Some guidelines [14,15,17,18], recommend start-ing treatment in the acute phase when possible witha drug that is effective in maintenance treatment(Table 4). The others guidelines [13,16,19] consid-er only essential points such as clinical features orpredominant polarity of illness when choosingprophylaxis medication.

CBG For long-term management, the AFPBN and ICGguidelines [3–5] suggest continuing the treatmentinitiated in the acute phase (if it has an acknowl-edged preventive effect). The AFBPN guidelinetakes into consideration the predominant thymicpolarity of the patient or the occurrence of rapidcycling as from the acute phase in the choice oftreatment. The ICG guideline ranks medication formanagement of bipolar depression taking intoaccount evidence of efficacy in acute bipolardepression and long-term treatment.

3.2.4. Time to reassessment and length oflong-term treatment

Recommendations for the time to reassessment and theglobal duration of long-term treatment are summarized inTable 5.

EBG CANMAT, WFSBP and NHMRC guidelinessuggest a time to reassessment depending on thetype of episode [14,17,18]. WFSBP guideline takesinto account the level of response. The continuationor discontinuation of initial medication will bedecided on the basis the level of response (full,partial or no response).Only NICE and WFSBP guidelines give recommen-dations for the duration of long-term treatment[15,19].

CBG The AFPBN guideline [4,5] proposes a time toreassessment depending on the type of MS used. Thepanel of experts also considers that the benefit-riskratio is in favor of continuing maintenance therapy aslong as possible, even for life if the patient presentsfactors of poor prognosis.

3.2.5. Specific populations: biological treatment of bipolardisorder during pregnancy

Recommendations for biological treatment of bipolardisorder during pregnancy are summarized in Table 6.

EBG Only some guidelines propose recommendations inthis specific population [15,16]. Low dose of first-generation antipsychotic (FGA) or SGA is recom-mended. Anticonvulsants (valproate and carbamaz-epine) and lithium are not considered to be used.Use of lamotrigine is not consensual. If an AD isneeded, SSRIs (excluding paroxetine) can be

Table 5Recommendations for the time to reassessment and duration of long-term treatment.

Evidence-based guidelines Consensus-based guidelines

NICE BAP WFSBP NHRMC CANMAT AFBPN ICG

Time to reassessment No recommendation No recommendation Mania: Mania and depression:If after a suitable period(7days for mania,4–6weeks for depression)there is no significantimprovement, consideralternative treatmentoptions.

Acute mania: Treatmentbe tried at least 2weeksat adequate doses beforeconcluding that the patientis unlikely to respond.

Mania and depression:If partial or no response,the therapeutic strategyshould be modified aftera duration of treatmentthat will be definedaccording to the MS used:

No recommendation- No response after 2weeks,switch to another first choicemedication.

- 4weeks for Li

- Partial response after2weeks, continues treatment,optimize dosage and if noimprovement over the next3weeks, consider add-ontreatment.

- 3weeks for AC, SGA,ECT

Depression: - 2weeks for FGA- No response after 4weeks,consider switch or combination- Partial responseafter 4weeks, continuetreatment, optimize dosageand if no improvement overthe next 4weeks, consideradd-on treatment

Duration of long-termtreatment

- At least 2years afteran episode of bipolardisorder

- Discontinuation oflong-term treatmentis not indicated whenthere is a good clinicalcontrol of the illness.

Long-term treatment shouldbe continued life longwhenever possible(WFSBP 2004).

No recommendation - Long-term treatment isrecommended for allpatients with bipolardisorder.

- As long as possible inboth BD II and I.

No recommendation

- Up to 5years if theperson has risk factorsfor relapse, such as ahistory of frequentrelapses or severepsychotic episodes,comorbid substancemisuse, ongoingstressful life eventsor poor social support

- When it is necessary,it should be tapered.

- For patients who refuse it,the effective acute-phasedosages should be continuedfor at least 3–6months.

- Lifelong continuoustreatment should beconsidered, accordingto the following factors:family history of BD,severity and high numberof episode, suicidalbehavior, comorbidities,residual symptoms, earlyage of onset, short timeintervals between episodes.

Abbreviations: AC, anticonvulsants; BD, bipolar disorder; ECT, electroconvulsive therapy; FGA, first-generation antipsychotic; Li, lithium; MS, mood stabilizer; SGA, second-generation antipsychotic.

7L.Sam

alinet

al./Com

prehensivePsychiatry

xx(2012)

xxx–xxx

Table 6Recommendations for biological treatment of bipolar disorder during pregnancy.

Evidence-based guidelines Consensus-based guidelines

NICE BAP WFSBP NHRMC CANMAT AFBPN ICG

Biological treatmentof bipolar disorderduring pregnancy

Discontinue AC(VPA, CBZ, LMT)and Li

Risk of teratogenicityfrom medications:AC (VPANCBZNLMT)NLiNSGA

No recommendation Gives references ofspecific guidelinesduring the perinatalperiod (NICE antenataland postnatal mentalhealth or ACOGcommittee on practicebulletins obstetrics)

LMT, SGAs, ECT Discontinue AC Review available evidence

Mania: low-doseSGAs or FGAs Avoid VPA, CBZ

Continue Li (after to haveinformed patient aboutthe benefit-risk ratio andprovided closely-spacedand multidisciplinarymonitoring)

- Severe and noresponse: ECT, Li

Severe mania: ECT

Avoid Li, VPA, CBZ

Mania: SGAs

Not VPA in first-line

Depression:Discontinuation orswitching medicinesrisks precipitating relapse Depression:- Mild/moderate:

psychotherapy Management decisionsmust be the results ofan individualizedassessment andconsiderations of thebalance of risks and benefits.

Medications should beavoided or used asmonotherapy in minimallyeffective doses, especiallyduring the first trimester.Risks and benefits ofcontinuing medicationshould be based onseverity and past responseto treatment.

- Severe: SGAs+SSRI(not paroxetine), ECT

Experts recommend closeclinical monitoring and toconsider the risks and benefitsassociated with treatmentcontinuation or discontinuation.

- Severe: QTP orSSRIs (not paroxetine)+prophylactic medication,

- Mild/moderate:psychotherapy

Abbreviations: AC, anticonvulsants; CBZ, carbamazepine; ECT, electroconvulsive therapy; FGA, first-generation antipsychotic; Li, lithium; LMT, lamotrigine; QTP, quetiapine; SGA, second-generationantipsychotic; SSRI, selective serotonin reuptake inhibitor; VPA, valproate.

8L.Sam

alinet

al./Com

prehensivePsychiatry

xx(2012)

xxx–xxx

9L. Samalin et al. / Comprehensive Psychiatry xx (2012) xxx–xxx

prescribed. ECT is an option if the episode is severeor without any response to treatment.BAP guideline reviews available evidence andproposes recommendations with a grade of terato-genic risk related to the use of MS [13].NHMRC and WFSBP guidelines do not cover thisfield.

CBG AFBPN guideline recommends using an SGA infirst-line treatment during pregnancy [4,5]. If thepatient presents a severe depressive episode,combined SGA and SSRI (avoiding paroxetine)or ECT may be considered. Anticonvulsant treat-ment must be discontinued; lithium can be main-tained after the patient has been informed about thebenefit-risk ratio and provided with closely-spacedand multidisciplinary monitoring.The ICG guideline describes the available evidenceduring pregnancy [2]. Valproate is not consideredas a first-line treatment option.

4. Discussion

Numerous sets of guidelines for the treatment of bipolardisorder have been developed in response to the wideningvariety of drugs now used and to the large number of studiespublished on the subject. Our discussion will focus on thedifferences between EBG and CBG issued since 2006 basedon the aspects previously described.

4.1. Methodology

There are intra- and intergroup variations in methodologybetween EBG and CBG.

Recommendations from EBG may vary because ofdifferences in the ranking of the level of evidence. Therecent WFSBP guideline update modified the criteria ofranking by including studies with negative results. Thus, onthe basis of the study by Young et al. [20], that demonstratesno significant difference in efficacy between lithium inmonotherapy and placebo in the treatment of bipolar de-pression, lithium is relegated to level D of evidence on a scaleof A–F [18,20,21]. In contrast, the CANMAT guideline,which does not take into consideration negative studies,ranked lithium at level 1 of evidence on a scale of 1–4 [16].Inclusion criteria of scientific data may vary across the dif-ferent guidelines depending also on their date of publication.The use of lamotrigine as an add-on treatment to lithium inbipolar depression has recently demonstrated its efficacy in arandomized placebo-controlled trial [22]. This combinationis not yet recommended as a first-line treatment, because, thepublication is more recent than the last published guidelines.

In CBG, the number of options and specific clinicalsituations, the assessment scale used and the data analysemay differ between guidelines. In the expert consensusguidelines: Treatment of bipolar disorder [7], the surveyasked about 1276 options in 48 specific clinical situations

versus 3017 options and 238 clinical situations in AFPBNguideline [4,5].

The International Consensus Group on Bipolar depres-sion used a mixed methodology to elaborate guidelines. Itproposes a first part given pharmacologic treatments basedon three categories of evidence and a second part onprinciples of treatment, special populations and futureclinical research based on experts consensus [2]. However,the consensus part of this guideline does not describe specificmethodology to obtain consensus.

From a methodological point of view, the combined useof EBG methodology to rank treatment and of CBGmethodology to specify management of specific populationsor other cases where evidence is not available would be thebest. Combining both methodologies enables us to meet thecomplete definition of evidence-based medicine that “meansintegrating individual clinical expertise with the bestavailable external clinical evidence from systematic re-search” [23].

4.2. Therapeutic discrepancies on the management ofbipolar disorder

4.2.1. Choice of initial pharmacological treatment duringan acute episode

The efficacy of lithium or valproate in combination withSGA in the management of manic episodes has made theseassociations first-line choices in different guidelines. However,choosing a combination from the outset for the management ofmania will depend on whether the Task Force that developedthe guidelines gives priority to efficacy or tolerance and not onthe guidelines methodology, which is used.

In contrast with unipolar depression, few studies havebeen conducted in the last 20years about the specific interestof AD in the treatment of bipolar depression [24]. Thecurrent EBGs are not concordant about their use incombination with MS in bipolar depression (without rapidcycle) because available evidence is limited and divergent[25,26]. The WFSBP and NHRMC guidelines specify thatnot all SSRIs have been studied in bipolar depression andsome (notably paroxetine) have multiple published RCTswith negative results [1]. This insufficient level of evidenceand the risks of mood switching to mania and of inductionof rapid cycling justify that all SSRIs in combination withMS in bipolar depression could not be recommended.However, these combinations are often prescribed inclinical practice in bipolar depression and their use seemsto be correlated with depression severity [27].

Conversely, taper and discontinue any AD is a consensualtherapeutic recommendation from all EBG and CBG duringmanic or mixed state.

If quetiapine in monotherapy is a consensual recommen-dation for first-line treatment in bipolar depression in allguidelines, the ranking of lamotrigine and lithium in mono-therapy merits some commentary. Five and three of theseven reported EBG and CBG recommend respectively

10 L. Samalin et al. / Comprehensive Psychiatry xx (2012) xxx–xxx

lamotrigine and lithium monotherapy as first-line treatmentin acute bipolar depression. However, if lamotrigine mono-therapy demonstrated a prophylactic role in bipolar depres-sion [28], it did not demonstrate an efficacy in acute bipolardepression in five placebo-controlled clinical studies [29].The EBG recommendations for the use of lamotriginemonotherapy as first-line treatment outweigh the positiveresults of efficacy from a large placebo-controlled study [30],a small crossover trial (in refractory depression) [31] and arecent meta-analysis (overall pool effect of lamotrigine wasmodest and larger in more severely depressed patients) [32].Only the WFSBP guideline specifies that this compoundshould have been, strictly speaking, ranked at the lowestlevel of evidence due to the five negatives studies. However,according to the positive results previously described andtaking into account clinical experience, the WFSBP TaskForce decided that it was more appropriate to recommendlamotrigine as a second-line treatment.

As for lithium (and the consideration or not of the negativestudy from Young et al. [20]), the recommendation grade oflamotrigine depends on how the results of the studies areinterpreted by the Task Force elaborating the EBG.

In mixed state, recommendations still differ about the useof lithium, owing to the lack of reliable scientific evidence,as indicated by the BAP [13]. In this type of situation, inwhich scientific evidence is weak, CBGs are able to proposerecommendations according to clinical experience.

4.2.2. Choice of treatment based on the clinical features orother factors during an acute episode

The current classifications of bipolar disorder distinguishdifferent clinical forms of mood episode and several dimen-sional features (psychotic symptoms, severity). In everydaypractice, taking these dimensional aspects into account seemsto be a criterion of choice for therapeutic strategy.

Despite a low level of evidence, there is wide agreement onthe use of SGA in certain clinical forms such asmania or bipolardepression with psychotic features. Severity of mood episodeprioritizes therapeutics considered to be quickly efficacious(SGA, valproate or ECT) or drug combinations. When risk ofsuicide is high, lithium is an option. The presence of rapid cycleimposes to avoid AD and to use a MS combination.

Some guidelines [2,4,5,16,18] propose a specific recom-mendation grade for bipolar disorder II. There are somerecommendations (with less rigorous evidence) for the use ofAD monotherapy in bipolar II depression. According to themeta-analysis by Bond et al. [33], the risk of treatmentemergent affective switches during acute treatment with ADmay be intermediate in patients with bipolar disorder IIdepression, between the risk of switch observed in patientswith bipolar disorder I depression (which is the highest) andthe risk in patients with unipolar depression.

Other factors that have to be taken into account, asprevious and current treatment, patients' choice and families'advice are cited in most guidelines. Partial adherence can bean argument to use an SGA depot.

The assessment of safety profile is always recommendedbefore treatment choice. Some guidelines recommend avoid-ing lithium and some anticonvulsants for women of child-bearing potential.

For all those specific clinical situations, there is no realdiscrepancy between EBG and CBG recommendations.Lack of evidence (i.e. lack of RCT) led to recommendationsdeveloped by the Task Force and most of the time close toclinical practice.

4.2.3. Choice of treatment in the acute phase consistent withlong-term treatment

The current dual approach of most guidelines, whichrecommend therapeutic strategies for the acute phase and formaintenance, seems to be giving way to a single overalltherapeutic model [34]. This artificial separation has beenessential for therapeutic investigation but application inclinical practice is difficult.

Future guidelines would recommend “main” drug treat-ments of proven efficacy in the acute and preventive phases,and suggest “sequential” treatments in combination with themain treatment for recurrence.

4.2.4. Time to reassessment and duration oflong-term treatment

There is no consensus on the duration of treatment.Differences between guidelines result from the methodologyused. CBG propose a duration that can be considered asbearing a closer relation to clinical reality.

4.2.5. Specific populations: biological treatment for bipolardisorder during pregnancy

Available evidence in this field is lacking, to proposeevidence-based recommendations. Most of the guidelines ontreatment of bipolar disorder during pregnancy are elabo-rated with a consensus-based methodology. With AFBPNguideline, other specific CBG for biological treatment duringpregnancy exist [35,36]. Guidelines elaborated from anexpert panel of obstetrician-gynaecologists' [35] recommendtreatment according to low risk of malformation. Guidelinesissued from psychiatrist's consensus focus on the balancebetween the risk of recurrence after treatment discontinua-tion and the risk of malformations with treatment continu-ation [3,36]. Use of SGA and exclusion of valproate are themost frequent recommendations during pregnancy. Lamo-trigine or lithium is discussed according to the clinicalsituation. However, lamotrigine due to controversial evi-dence is not differentiated from other anticonvulsants inNICE and AFBPN guidelines. In BAP guideline, thediscrimination among anticonvulsants, with respect to theirteratogenic risks during pregnancy, ranks lamotrigine as atlowest risk (valproateNcarbamazepineN lamotrigine).

Guidelines advise to minimize the exposure risk to mul-tiple medications to avoid combinations of drugs (especiallyduring the first trimester of pregnancy) and recommend onlypsychotherapy in mild/moderate bipolar depression.

11L. Samalin et al. / Comprehensive Psychiatry xx (2012) xxx–xxx

ECT is suggested as an alternative during pregnancy incases of severe episode, suicidal ideation or no response topreviousmedication. This physical treatment is considered as arelatively safe procedure without indications of teratogenesis.

5. Conclusion

Comparison between guidelines for treatment of bipolardisorder found more similarities than differences. However,new available clinical evidence (like superior efficacy ofMS in combination with SGA versus MS in monotherapy inmania) or lack of evidence in specific situations (like mixedstate, duration of long-term treatment or women duringpregnancy) generates inconsistent recommendations be-tween guidelines.

Combining EBG and CBG methodologies may helpclinicians to have a real evidence-based medicine practice,including both clinical expertise and scientific evidence.However, the choice of policy, whatever the methodologyused, is up to the authors who establish the guidelines andcan, therefore, depend on their interpretation of the availablescientific evidence and on treatment traditions betweenEurope, Asia and America [37].

References

[1] Nivoli AM, Colom F, Murru A, Pacchiarotti I, Castro-Loli P,Gonzalez-Pinto A, et al. New treatment guidelines for acute bipolardepression: a systematic review. J Affect Disord 2011;129(1–3):14-26.

[2] Calabrese JR, Kasper S. International Consensus Group on theevidence-based pharmacologic treatment of bipolar I and II depression.J Clin Psychiatry 2008;69(10):1632-46.

[3] Frye MA. International consensus group on depression prevention inbipolar disorder. J Clin Psychiatry 2011;72(10):1295-310.

[4] Llorca PM, Courtet P, Martin P, Abbar M, Gay C, Meynard JA, et al.French Association of Biological Psychiatry and Neuropsychophar-macology. Formalized Recommendations of Experts for screeningand management of bipolar disorder. Encéphale 2010;36(Suppl 4):S86-S102.

[5] Llorca PM, Courtet P, Martin P, Abbar M, Gay C, Meynard JA, et al.Formalized Consensus Guidelines: screening and management ofbipolar disorder. In: & Boutros N, editor. Yearbook of internationalpsychiatry and behavioral neurosciences: Nova Biomedical Books;2012 chapter 8.

[6] AGREE. Development and validation of an international appraisalinstrument for assessing the quality of clinical practice guidelines: theAGREE project. Qual Saf Health Care 2003;12(1):18-23.

[7] Sachs GS, Printz DJ, Kahn DA, Carpenter D, Docherty JP. The expertconsensus guideline series: medication treatment of bipolar disorder2000. Postgrad Med 2000;(Spec No):1-104.

[8] Sachs G, Chengappa KN, Suppes T, Mullen JA, Brecher M, DevineNA, et al. Quetiapine with lithium or divalproex for the treatment ofbipolar mania: a randomized, double-blind, placebo-controlled study.Bipolar Disord 2004;6(3):213-23.

[9] Sachs GS, Grossman F, Ghaemi SN, Okamoto A, Bowden CL.Combination of a mood stabilizer with risperidone or haloperidol fortreatment of acute mania: a double-blind, placebo-controlled compar-ison of efficacy and safety. Am J Psychiatry 2002;159(7):1146-54.

[10] Tohen M, Chengappa KN, Suppes T, Zarate Jr CA, Calabrese JR,Bowden CL, et al. Efficacy of olanzapine in combination with val-proate or lithium in the treatment of mania in patients partially non-

responsive to valproate or lithium monotherapy. Arch Gen Psychiatry2002;59(1):62-9.

[11] Vieta E, T'Joen C, McQuade RD, Carson Jr WH, Marcus RN, SanchezR, et al. Efficacy of adjunctive aripiprazole to either valproate orlithium in bipolar mania patients partially nonresponsive to valproate/lithium monotherapy: a placebo-controlled study. Am J Psychiatry2008;165(10):1316-25.

[12] Yatham LN, Grossman F, Augustyns I, Vieta E, Ravindran A. Moodstabilisers plus risperidone or placebo in the treatment of acute mania.International, double-blind, randomised controlled trial. Br J Psychi-atry 2003;182:141-7.

[13] Goodwin GM. Evidence-based guidelines for treating bipolar disorder:revised second edition–recommendations from the British Associationfor Psychopharmacology. J Psychopharmacol 2009;23(4):346-88.

[14] Grunze H, Vieta E, Goodwin GM, Bowden C, Licht RW, Moller HJ,et al. The World Federation of Societies of Biological Psychiatry(WFSBP) guidelines for the biological treatment of bipolar disorders:update 2009 on the treatment of acute mania. World J Biol Psychiatry2009;10(2):85-116.

[15] NICE. National Institute for Health and Clinical Excellence. In: &Alden Press UK, editor. The management of bipolar disorder in adults,children and adolescents, in primary and secondary care. London: TheBritish Psychological Society and the Royal College of Psychiatrists;2006. p. 1-32.

[16] Yatham LN, Kennedy SH, Schaffer A, Parikh SV, Beaulieu S,O'Donovan C, et al. Canadian Network for Mood and Anxiety Treat-ments (CANMAT) and International Society for Bipolar Disorders(ISBD) collaborative update of CANMAT guidelines for the manage-ment of patients with bipolar disorder: update 2009. Bipolar Disord2009;11(3):225-55.

[17] Malhi GS, Adams D, Lampe L, Paton M, O'Connor N, Newton LA, etal. Clinical practice recommendations for bipolar disorder. ActaPsychiatr Scand Suppl 2009;439:27-46.

[18] Grunze H, Vieta E, Goodwin GM, Bowden C, Licht RW, Moller HJ,et al. The World Federation of Societies of Biological Psychiatry(WFSBP) Guidelines for the Biological Treatment of Bipolar Dis-orders: update 2010 on the treatment of acute bipolar depression.World J Biol Psychiatry 2010;11(2):81-109.

[19] Grunze H, Kasper S, Goodwin G, Bowden C, Moller HJ. The WorldFederation of Societies of Biological Psychiatry (WFSBP) guidelinesfor the biological treatment of bipolar disorders, part III: maintenancetreatment. World J Biol Psychiatry 2004;5(3):120-35.

[20] Young AH, McElroy SL, Bauer M, Philips N, Chang W, Olausson B,et al. A double-blind, placebo-controlled study of quetiapine andlithium monotherapy in adults in the acute phase of bipolar depression(EMBOLDEN I). J Clin Psychiatry 2008;71(2):150-62.

[21] Grunze H, Kasper S, Goodwin G, Bowden C, Baldwin D, Licht R, et al.World Federation of Societies of Biological Psychiatry (WFSBP)guidelines for biological treatment of bipolar disorders. Part I: treatmentof bipolar depression. World J Biol Psychiatry 2002;3(3):115-24.

[22] Van der Loos ML, Mulder PG, Hartong EG, Blom MB, VergouwenAC, de Keyzer HJ, et al. Efficacy and safety of lamotrigine as add-ontreatment to lithium in bipolar depression: a multicenter, double-blind,placebo-controlled trial. J Clin Psychiatry 2009;70(2):223-31.

[23] Sackett DL, Rosenberg WM, Gray JA, Haynes RB, Richardson WS.Evidence based medicine: what it is and what it isn't. BMJ 1996;312(7023):71-2.

[24] Moller HJ, Grunze H, Broich K. Do recent efficacy data on the drugtreatment of acute bipolar depression support the position that drugsother than antidepressants are the treatment of choice? A conceptualreview. Eur Arch Psychiatry Clin Neurosci 2006;256(1):1-16.

[25] Gijsman HJ, Geddes JR, Rendell JM, Nolen WA, Goodwin GM.Antidepressants for bipolar depression: a systematic review ofrandomized, controlled trials. Am J Psychiatry 2004;161(9):1537-47.

[26] Sidor MM, Macqueen GM. Antidepressants for the acute treatment ofbipolar depression: a systematic review and meta-analysis. J ClinPsychiatry 2011;72(2):156-67.

12 L. Samalin et al. / Comprehensive Psychiatry xx (2012) xxx–xxx

[27] Goldberg JF, Brooks III JO, Kurita K, Hoblyn JC, Ghaemi SN, PerlisRH, et al. Depressive illness burden associated with complexpolypharmacy in patients with bipolar disorder: findings from theSTEP-BD. J Clin Psychiatry 2009;70(2):155-62.

[28] Smith LA, Cornelius V, Warnock A, Bell A, Young AH. Effectivenessof mood stabilizers and antipsychotics in the maintenance phase ofbipolar disorder: a systematic review of randomized controlled trials.Bipolar Disord 2007;9(4):394-412.

[29] Calabrese JR, Huffman RF, White RL, Edwards S, Thompson TR,Ascher JA, et al. Lamotrigine in the acute treatment of bipolardepression: results of five double-blind, placebo-controlled clinicaltrials. Bipolar Disord 2008;10(2):323-33.

[30] Calabrese JR, Bowden CL, Sachs GS, Ascher JA, Monaghan E, RuddGD. A double-blind placebo-controlled study of lamotrigine mono-therapy in outpatients with bipolar I depression. Lamictal 602 StudyGroup. J Clin Psychiatry 1999;60(2):79-88.

[31] Frye MA, Ketter TA, Kimbrell TA, Dunn RT, Speer AM, Osuch EA,et al. A placebo-controlled study of lamotrigine and gabapentinmonotherapy in refractory mood disorders. J Clin Psychopharmacol2000;20(6):607-14.

[32] Geddes JR, Calabrese JR, Goodwin GM. Lamotrigine for treatment ofbipolar depression: independent meta-analysis and meta-regression ofindividual patient data from five randomised trials. Br J Psychiatry2009;194(1):4-9.

[33] Bond DJ, Noronha MM, Kauer-Sant'Anna M, Lam RW, Yatham LN.Antidepressant-associated mood elevations in bipolar II disorder com-pared with bipolar I disorder and major depressive disorder: a systematicreview and meta-analysis. J Clin Psychiatry 2008;69(10):1589-601.

[34] Soreca I, Frank E, Kupfer DJ. The phenomenology of bipolar disorder:what drives the high rate of medical burden and determines long-termprognosis? Depress Anxiety 2009;26(1):73-82.

[35] Practice ACOG. Bulletin: clinical management guidelines forobstetrician-gynecologists. Use of psychiatric medications duringpregnancy and lactation. Obstet Gynecol 2008;111(4):1001-20.

[36] Yonkers KA, Wisner KL, Stowe Z, Leibenluft E, Cohen L, Miller L, etal. Management of bipolar disorder during pregnancy and thepostpartum period. Am J Psychiatry 2004;161(4):608-20.

[37] Vestergaard P. Guidelines for maintenance treatment of bipolardisorder: are there discrepancies between European and NorthAmerican recommendations? Bipolar Disord 2004;6(6):519-22.