clinical overview pharmacological management of unipolar...

18
Clinical overview Pharmacological management of unipolar depression Malhi GS, Hitching R, Berk M, Boyce P, Porter R, Fritz K. Pharmacological management of unipolar depression. Objective: To be used in conjunction with ‘Psychological management of unipolar depression’ [Lampe et al. Acta Psychiatr Scand 2013;127(Suppl. 443):2437] and ‘Lifestyle management of unipolar depression’ [Berk et al. Acta Psychiatr Scand 2013;127(Suppl. 443):3854]. To provide clinically relevant recommendations for the use of pharmacological treatments in depression derived from a literature review. Method: Using our previous Clinical Practice Guidelines [Malhi et al. Clinical practice recommendations for bipolar disorder. Acta Psychiatr Scand 2009;119(Suppl. 439):2746] as a foundation, these clinician guidelines target key practical considerations when prescribing pharmacotherapy. A comprehensive review of the literature was conducted using electronic database searches (PubMed, MEDLINE), and the findings have been synthesized and integrated alongside clinical experience. Results: The pharmacotherapy of depression is an iterative process that often results in partial and non-response. Beyond the initiation of antidepressants, the options within widely used strategies, such as combining agents and switching between agents, are difficult to proscribe because of the paucity of pertinent research. However, there is some evidence for second-line strategies, and a non-prescriptive algorithm can be derived that is based broadly on principles rather than specific steps. Conclusion: Depression is by its very nature a heterogeneous illness that is consequently difficult to treat. Invariably, situation-specific factors often play a significant role and must be considered, especially in the case of partial and non-response. Consulting with colleagues and trialling alternate treatment paradigms are essential strategies in the management of depression. G. S. Malhi 1,2 , R. Hitching 1,2 , M. Berk 3,4,5,6,7 , P. Boyce 2 , R. Porter 8 , K. Fritz 1,2 1 CADE Clinic, Department of Psychiatry, Royal North Shore Hospital, Sydney, NSW, Australia, 2 Discipline of Psychiatry, Sydney Medical School, University of Sydney, Sydney, NSW, Australia, 3 Deakin University, Geelong, Vic., Australia, 4 Department of Psychiatry, The University of Melbourne, Melbourne, Vic., Australia, 5 Brain Sciences Institute, Swinburne University of Technology, Melbourne, Vic., Australia, 6 The Mental Health Research Institute, Parkville, Vic., Australia, 7 Orygen Youth Health Research Institute, Parkville, Vic., Australia and 8 Department of Psychological Medicine, University of Otago, Christchurch, New Zealand Key words: major depressive disorder; antidepressants; treatment algorithm; clinical practice guidelines; recommendations Gin Malhi, CADE Clinic, Department of Psychiatry, Level 5 Building 36, Royal North Shore Hospital, St. Leonards, Sydney, NSW 2065, Australia. E-mail: [email protected] Clinical recommendations Major depression is a heterogeneous disorder that requires both empirical evidence and clinical experience when treating. Patient adherence is essential and should be appreciated by the patient and monitored by the physician. For partial and non-response, physicians should re-evaluate the diagnosis, consider alternate treatment paradigms and, if necessary, consult colleagues. Additional comments The proposed algorithm and recommendations should be used in conjunction with other recognized sources to guide the management of depression. The recommendations set out in this supplement are intended for adults who suffer from unipolar depression. Special populations, such as adolescents, pregnant women and the elderly, are not discussed in detail. This review is not intended to be a comprehensive guide to antidepressant medication. 6 Acta Psychiatr Scand 2013: 127 (Suppl. 443): 6–23 © 2013 John Wiley & Sons A/S. Published by Blackwell Publishing Ltd All rights reserved DOI: 10.1111/acps.12122 ACTA PSYCHIATRICA SCANDINAVICA

Upload: others

Post on 11-Aug-2020

5 views

Category:

Documents


0 download

TRANSCRIPT

Page 1: Clinical overview Pharmacological management of unipolar ...dro.deakin.edu.au/...pharmacologicalmanagement... · Pharmacological management of unipolar depression Malhi GS, Hitching

Clinical overview

Pharmacological management of unipolardepression

Malhi GS, Hitching R, Berk M, Boyce P, Porter R, Fritz K.Pharmacological management of unipolar depression.

Objective: To be used in conjunction with ‘Psychological management ofunipolar depression’ [Lampe et al. Acta Psychiatr Scand 2013;127(Suppl.443):24–37] and ‘Lifestyle management of unipolar depression’ [Berket al. Acta Psychiatr Scand 2013;127(Suppl. 443):38–54]. To provideclinically relevant recommendations for the use of pharmacologicaltreatments in depression derived from a literature review.Method: Using our previous Clinical Practice Guidelines [Malhi et al.Clinical practice recommendations for bipolar disorder. Acta PsychiatrScand 2009;119(Suppl. 439):27–46] as a foundation, these clinicianguidelines target key practical considerations when prescribingpharmacotherapy. A comprehensive review of the literature wasconducted using electronic database searches (PubMed, MEDLINE),and the findings have been synthesized and integrated alongside clinicalexperience.Results: The pharmacotherapy of depression is an iterative process thatoften results in partial and non-response. Beyond the initiation ofantidepressants, the options within widely used strategies, such ascombining agents and switching between agents, are difficult toproscribe because of the paucity of pertinent research. However, there issome evidence for second-line strategies, and a non-prescriptivealgorithm can be derived that is based broadly on principles rather thanspecific steps.Conclusion: Depression is by its very nature a heterogeneous illness thatis consequently difficult to treat. Invariably, situation-specific factorsoften play a significant role and must be considered, especially in thecase of partial and non-response. Consulting with colleagues andtrialling alternate treatment paradigms are essential strategies in themanagement of depression.

G. S. Malhi1,2, R. Hitching1,2,M. Berk3,4,5,6,7, P. Boyce2,R. Porter8, K. Fritz1,21CADE Clinic, Department of Psychiatry, Royal NorthShore Hospital, Sydney, NSW, Australia, 2Discipline ofPsychiatry, Sydney Medical School, University of Sydney,Sydney, NSW, Australia, 3Deakin University, Geelong,Vic., Australia, 4Department of Psychiatry, The Universityof Melbourne, Melbourne, Vic., Australia, 5BrainSciences Institute, Swinburne University of Technology,Melbourne, Vic., Australia, 6The Mental Health ResearchInstitute, Parkville, Vic., Australia, 7Orygen Youth HealthResearch Institute, Parkville, Vic., Australia and8Department of Psychological Medicine, University ofOtago, Christchurch, New Zealand

Key words: major depressive disorder; antidepressants;treatment algorithm; clinical practice guidelines;recommendations

Gin Malhi, CADE Clinic, Department of Psychiatry, Level5 Building 36, Royal North Shore Hospital, St. Leonards,Sydney, NSW 2065, Australia.E-mail: [email protected]

Clinical recommendations

• Major depression is a heterogeneous disorder that requires both empirical evidence and clinicalexperience when treating.

• Patient adherence is essential and should be appreciated by the patient and monitored by the physician.

• For partial and non-response, physicians should re-evaluate the diagnosis, consider alternatetreatment paradigms and, if necessary, consult colleagues.

Additional comments

• The proposed algorithm and recommendations should be used in conjunction with other recognizedsources to guide the management of depression.

• The recommendations set out in this supplement are intended for adults who suffer from unipolardepression. Special populations, such as adolescents, pregnant women and the elderly, are notdiscussed in detail.

• This review is not intended to be a comprehensive guide to antidepressant medication.

6

Acta Psychiatr Scand 2013: 127 (Suppl. 443): 6–23 © 2013 John Wiley & Sons A/S. Published by Blackwell Publishing LtdAll rights reservedDOI: 10.1111/acps.12122

ACTA PSYCHIATRICA SCANDINAVICA

Page 2: Clinical overview Pharmacological management of unipolar ...dro.deakin.edu.au/...pharmacologicalmanagement... · Pharmacological management of unipolar depression Malhi GS, Hitching

Introduction

Pharmacotherapy is an important componentwithin the biopsychosocial and lifestyle approachto treating depression. It builds upon the ‘SET APACE’ approach (1), and in doing so, expands the5P Model, which is a widely used approach forformulation. The latter considers the presentingproblem alongside predisposing, precipitating, per-petuating and protective factors and adopts alongitudinal personalized perspective (2). In addi-tion to psychological interventions (3) and lifestylemanagement (4), which are addressed in separatearticles of this supplement, pharmacotherapy is animportant treatment option that can be the keyingredient for reaching remission. However, itrequires informed administration and monitoringand may not be effective for all individuals.

The practical issues of medication choice, partialand non-response to an initial antidepressant treat-ment, and pharmacological strategies beyondprescription of the initial medication are majorconcerns in clinical practice and form the focus ofthis article. Pharmacotherapy treatment strategiesfor achieving optimal response and managingpotential non-response are discussed andaddressed by drawing upon both research evidenceand clinical experience. The latter is particularlyimportant when considering complex managementdecisions in which research remains challenging.

Aims of the study

The study aims to provide clinically relevant,evidence-based recommendations for an individu-alized formulation for pharmacotherapeuticmanagement.

Material and methods

Using our previous Clinical Practice Guidelines (1)as a foundation, these clinician recommendationstarget key practical considerations when prescrib-ing pharmacotherapy. A comprehensive review ofthe literature was conducted using electronic data-base searches (PubMed, MEDLINE), and the find-ings have been synthesized and integrated alongwith clinical experience.

Results

Understanding depression and its context

To develop an appropriate management plan, it isessential to understand the context in whichdepression emerges and this can be facilitated by

use of the 5P Model. The aim is essentially to builda detailed ‘picture of the problem’ and tailor treat-ment accordingly. When applying the 5P Model topharmacotherapy, predisposing factors include pasttreatments and prior response. With respect to thelatter, pharmacological sensitivity to specific medi-cations is likely to be determined by the uniquebiology of an individual. Pharmacogenomicsaddresses this very issue and although this field isin its infancy, it is likely to have mainstream appli-cation in the future (5). Precipitating factors aremore predictable and include non-compliance,medical disorders, especially inflammatory disor-ders, ineffective treatments and iatrogenic causes,for example, the side-effects of medications used totreat diseases (levodopa, a drug for Parkinson’sdisease or interferon). Further, it is important tonote that with long-term medication, antidepres-sant effects are intended to be protective, and anyassociated side-effects are likely to become morepronounced with remission and can seeminglyperpetuate symptoms. A detailed correctly mappedout 5P Model is essential for optimal treatment,and misidentifying the 5Ps can result in compro-mising and prolonging treatment.

Partial and non-response are exceedingly com-mon and as will be discussed later, response ratesto pharmacotherapy differ based on individualcharacteristics. This no doubt contributes to poorresponse, and although some patients respond wellafter only one treatment trial, most require multi-ple treatment trials before achieving remission. Forthose who respond only partially, or not at all, tomultiple pharmacotherapeutic trials, re-evaluationshould be considered alongside consultation with acolleague. Reviewing the 5P Model is usefulbecause there are often additional features (i.e. per-sonality issues, anxiety symptoms, persisting inter-personal or psychosocial problems) that may helpexplain why the patient is not recovering. Thisstudy aims to outline the pharmacotherapeuticoptions for the treatment of depression at eachstage of management including initial choice ofmedication, strategies for negotiating partial ornon-response to initial medication, and suggestionsfor when patients do not reach remission followingmultiple therapeutic trials. Therefore, evaluatingand monitoring treatment response is important,and standardized rating scales are useful for assess-ing change. In this regard, a unidimensional scale,such as the Bech–Rafaelsen Melancholia Scale (6),or non-unidimensional scales like the HamiltonDepression Scale (HAM-D) (7), the Montgomery–Asberg Depression Rating Scale (MADRS) andthe Inventory of Depressive Symptomatology areof particular benefit. Asking the patient to rate

7

Pharmacological management of depression

Page 3: Clinical overview Pharmacological management of unipolar ...dro.deakin.edu.au/...pharmacologicalmanagement... · Pharmacological management of unipolar depression Malhi GS, Hitching

their mood on a 0–10 scale is practical way ofcharting treatment response.

Pharmacotherapy

Our clinical practice recommendations (1) for themanagement of unipolar non-psychotic depressionhighlight the need to assess what factors may influ-ence treatment response and prognosis. It is impor-tant to be able to foresee the patient’s treatmentoutcome (prognosis) when initiating treatment.The features to consider can be recalled with themnemonic 4C, which represents the four actions,each of which begins with the letter ‘c’: characterize– identify the clinical symptoms and subtype;calibrate – gauge the severity and chronicity of thesymptoms; corroborate – identify any comorbidi-ties and the context that is contributing to theillness; and consider – identify coping styles andadaptive features. Identifying factors that are likelyto influence treatment outcome in combinationwith a detailed 5P Model helps guide which phar-macological treatment is most likely to be effectivefor a particular patient at a specific juncture andwhich patients would not benefit from antidepres-sant medication.

Selecting an antidepressant. There are a large vari-ety of antidepressant medications, all of whichhave similar efficacy when treating community-based depression (8); however, they differ in termsof tolerability and response to specific features ofdepression (e.g. anxiety, insomnia) (Table 1). Toassist in selecting which option would most likelybenefit the patient, our previous recommendationsthat evaluate antidepressants on the basis of risk,adherence, tolerability and efficacy (RATE) remainpertinent (1).

Risk: Tricyclics antidepressants (TCAs), mono-amine oxidase inhibitors (MAOIs) and theserotonin and noradrenaline reuptake inhibitor(SNRI) venlafaxine, in particular, carry highrisks of toxicity in overdose and interactionswith other medications. Therefore, alternativeantidepressants, in particular selective serotoninreuptake inhibitors (SSRIs), are a much saferand consequently a more popular treatmentoption (Table 2).

Adherence: Treatment adherence is essential,but remarkably difficult to achieve, particularlylonger term. In this regard, medication choice iscritical, and the dosing of medications should besimple and convenient, and side-effects need toremain minimal to facilitate adherence/compli-

ance. Patient preference should always be con-sidered, as it will facilitate adherence.

Tolerability: Tolerability is intertwined withadherence. Medication side-effects impact toler-ability, which, in turn, affects adherence. There-fore, common treatment side-effects should bediscussed with the patient at the outset of treat-ment and side-effects should be assessed rou-tinely. Starting on a low dose will help inminimizing side-effects early in treatment.

Efficacy: Efficacy remains the sine qua non ofany treatment option, but the clinical effects ofantidepressants can take 7–10 days to emerge,and therefore, to ensure adherence, patientsmust be made aware of the likely delay.

Risk, adherence and tolerability are prioritizedbecause most antidepressant options are equallyeffective when treating clinical depression of mild-to-moderate severity (Fig. 1). SSRIs are undoubt-edly better tolerated and less complicated toadminister than older antidepressants, and henceadherence is better. Contrasting SSRIs and olderantidepressants in terms of efficacy provides per-haps the only example of a significant gradient ordifference in potency among the antidepressantarmamentarium. However, this differential is onlydiscernible with melancholia and psychoticdepression (9).

Further, the first and foremost consideration inpharmacotherapy is safety. In this regard, antide-pressants, especially SSRIs, are relatively safe, yetadverse effects can never be fully ruled out. Forinstance, the potential risk of increasing self-harmin the initial stages of treatment, especially inyouth, continues to be a concern. Research sug-gests there is a small trend for increased suicidalityin young adults (18–24 years), but that equallythere is reduced risk and increased protectiveeffects of SSRIs and newer antidepressants on sui-cide attempts in adults (10, 11). As mentioned pre-viously, adherence is necessary for any medicationto be effective, and thus is essential to gain thera-peutic benefit. Early discontinuation rates are high(12, 13), even though guidelines recommend thatthe minimum duration of antidepressant treatmentfor depression should be 6–12 months. During thistime, approximately 30% of patients discontinuewithin the first 30 days and more than 40%discontinue within 90 days of treatment (13).Given the high discontinuation rates, it is impor-tant to optimize adherence to treatment, especiallywhen prescribing antidepressants that are likely toproduce significant side-effects. Strategies such as

8

Malhi et al.

Page 4: Clinical overview Pharmacological management of unipolar ...dro.deakin.edu.au/...pharmacologicalmanagement... · Pharmacological management of unipolar depression Malhi GS, Hitching

education, self-management by patients and col-laborative care by practitioners have been shownto enhance patient adherence (14). Side-effects areone of the main reasons cited for early discontinu-ation of antidepressants; thus, tolerability is essen-tial and this is the domain (as opposed to efficacy)in which most antidepressants differentiate fromeach other.

The SSRIs, SNRIs and some newer agents (suchas agomelatine) are recommended first-linebecause they have better safety and tolerabilityprofiles than older antidepressants, such as theTCAs and MAOIs (15–17). The variety of first-linemedications makes initiating treatment a processthat requires careful consideration. Meta-analyseshave revealed modest differences in efficacy

Table 1. Summary of antidepressant clinical use, principal mechanisms of action and key features of depression for which the medication is useful

Antidepressant class Generic name of medication Principal mechanism of actionFeatures of depression for

which medicine is particularly useful

1st line SSRI (120–128) Selective 5-HT uptakeblockade

Anxiety

NARI (129–131) Reboxetine, Atomoxetine Reuptake inhibitor fornorepinephrine andepinephrine

Reboxetine may be useful for activation.Atomoxetine is prescribed for ADHDand is therefore not recommended1st line for depression.

NaSSA Mirtazapine Blocks the reuptake ofserotonin via 5-HT2Aand 5-HT2C receptors

Insomnia

Agomelatine Melatonergic agonistand 5-HT2C antagonist

Sleep problems

NDRI Bupropion Blocks the action of thenorepinephrine transporterand dopamine transporter

Melancholia; severe depression

2nd line SNRI (132) Venlafaxine†,Desvenlafaxine, Milnacipranand Duloxetine

Similar to TCAs but lack broadspectrum properties

Melancholia; severe depressionVenlafaxine: anxietyDuloxetine: pain

TCA (121–125, 133–135) Amitriptyline, Clomipramine Block reuptake of multiplemonoamines

PainMelancholia

Nefazodone 5-HT2A antagonist3rd line MAOI Irreversibly inhibit the

mitochondrial enzymesMonoamine oxidase A(prefers 5-HT) & B (prefers NA & EPH)

MelancholiaAtypical symptoms*Treatment resistant depression

Adjunctive SARI Trazodone Blocks reuptake of serotonin,noradrenaline and/or dopamine

Used when patients do not respondwell to 1st line

SSRI, selective serotonin reuptake inhibitor; NARI, noradrenaline reuptake inhibitor; NaSSA, noradrenaline and specific serotonergic antidepressant; NDRI, norepinephrine-dopa-mine reuptake inhibitor; SNRI, serotonin and noradrenaline reuptake inhibitor; TCA, tricyclics antidepressant; MAOI, monoamine oxidase inhibitor; SARI, serotonin antagonist andreuptake inhibitors; NA, noradrenaline; EPH, epinephrine.*Evidence is equivocal.†Low dose.

Table 2. Side-effects associated with common antidepressant groups. +, ++, +++: <10%, 11–29%, >30%

Weight gainSexual

dysfunctionCNS effect

(e.g. sedation, fatigue)Anticholinergic effect(e.g. dry mouth, tremor) GI distress

SSRI + +++ ++ ++ ++NARI ++ + + + +NaSSA ++ ++ +++ ++ +Agomelatine + + + + +NDRI ++ + ++ ++ +SNRI + ++ ++ ++ +++TCA ++ + + ++ +MAOI* + + + + +Nefazodone + + ++ ++ ++SARI + + +++ + ++

SSRI, selective serotonin reuptake inhibitor; NARI, noradrenaline reuptake inhibitor; NaSSA, noradrenaline and specific serotonergic antidepressant; NDRI, norepinephrine-dopa-mine reuptake inhibitor; SNRI, serotonin and noradrenaline reuptake inhibitor; TCA, tricyclics antidepressant; MAOI, monoamine oxidase inhibitor; SARI, serotonin antagonist andreuptake inhibitors.*MAOIs require dietary restrictions to prevent hypertensive crisis, some combinations with other drugs can be fatal.

9

Pharmacological management of depression

Page 5: Clinical overview Pharmacological management of unipolar ...dro.deakin.edu.au/...pharmacologicalmanagement... · Pharmacological management of unipolar depression Malhi GS, Hitching

between newer antidepressants, such as venlafaxineover SSRIs (8), escitalopram over comparators(18) and sertraline over other antidepressants (19).However, tolerability has shown to be inferior invenlafaxine compared with SSRIs (20). In trials, itis difficult to establish superior efficacy of one med-ication over another because of the large samplesizes this requires. Hence, most research groupshave not systematically compared the efficacy ofindividual agents but instead compared the effectsagainst placebo. Pharmaceutical companies haveconducted the majority of comparison studies, lar-gely for registration. Overall, escitalopram, mirt-azapine, sertraline and venlafaxine have somebenefits as compared to other antidepressantswhen considering dosage and severity of thedepressive episode, and reboxetine is the only anti-depressant to have a significantly lower responserate than other agents (21, 22).

Other factors to consider when choosing anantidepressant include the symptom profile, medi-cal and psychiatric comorbidity, previousresponse to antidepressants, patient preferenceand cost of treatment. Additionally, drug–druginteraction is an important consideration (partic-ularly when medical comorbidity exists), espe-cially because it can increase side-effects and lossof efficacy can result (23). Selection should alsotake into consideration the prevailing symptoms,such as insomnia and anxiety, because particularantidepressants may be better suited to treatingspecific symptoms because of their unique chemi-cal profile. For example, mirtazapine is often sed-ative (as a result of its effect on histaminicreceptors) and therefore may be of benefit where

insomnia is a prominent symptom (but balancedagainst weight gain), and reboxetine may be use-ful in drive-deficient ‘anergic’ states where sus-tained motivation is lacking. Depressed patientswho suffer from a disturbed sleep–wake cyclemay benefit from agomelatine because of its abil-ity to improve both night-time sleep and daytimefunctioning (24). Escitalopram is an antidepres-sant that is also used for anxiety disorders, thusis favourable for depressed patients where anxietyis a prominent symptom (Table 1).

Initializing treatment. Once it has been determinedthat the patient would benefit from antidepressantsand a specific medication has been chosen, prior tocommencing the medication the patient should beinformed about possible side-effects (in particular,the possibility of activating side-effects and agita-tion initially), the importance of adherence, thedelay prior to the antidepressant becoming effec-tive and self-management techniques, which isdesigned to foster personal empowerment andresponsibility. The patient should give consent totreatment and this should be recorded. Where pos-sible it is helpful to include a significant other inthe initiation of treatment. Planned follow-upshould be scheduled with regular monitoring ofside-effects and adjustment of treatment dosage asneeded.

The patient (and carer) may choose to keep amood diary to self-monitor their progress in whichthey can record their mood on a simple 0–10 scale(it is now possible to do this using a smart phoneapp).

Monitoring response. In clinical practice, responseis easily gauged as simply ‘getting better’. How-ever, to objectively assess the efficacy of medica-tion, the use of both observer-rated and self-ratedquestionnaires can be useful. The latter are com-monly used in research settings and in recent yearshave become more widely adopted in general prac-tice and primary care practice.

When measuring the efficacy of medication, a50% improvement on rating scales (e.g. HAM-Dor a self-report scale) is typically the point used todefine a clinical response, even though, in reality,this may not equate to substantive clinicalimprovement. Measuring and tracking efficacyrates in patients can be difficult, and the process isfurther complicated because the trajectory ofresponse is unique in every case. Ideally, both ini-tial (Fig. 2) and long-term responses (Fig. 3)should be charted to determine the likely outcomeand establish whether the current treatment is pro-viding sufficient benefit to the patient.

SSRIs

NARIs

NASSAs

Agomela

NDRI

SNRIs*

TCAs

MAOIs

Nefazodone

SARIs _ 0 +

First-line

Second-line

Clinical u lity depressants

LegendAcute efficacy

Tolerability difficul es

Fig. 1. The clinical utility of antidepressant agents ratedaccording to efficacy and tolerability. *Efficacy for SNRIs isplotted for severe depression. SSRI, selective serotonin reup-take inhibitor; NARI, noradrenaline reuptake inhibitor;NaSSA, noradrenaline and specific serotonergic antidepres-sant; NDRI, norepinephrine-dopamine reuptake inhibitor;SNRI, serotonin and noradrenaline reuptake inhibitor; TCA,tricyclics antidepressant; MAOI, monoamine oxidase inhibitor;SARI, serotonin antagonist and reuptake inhibitors. Adaptedfrom Clinical practice recommendations for depression (1).

10

Malhi et al.

Page 6: Clinical overview Pharmacological management of unipolar ...dro.deakin.edu.au/...pharmacologicalmanagement... · Pharmacological management of unipolar depression Malhi GS, Hitching

After commencing antidepressant treatment,there is usually a delay of 7–10 days before a dis-cernible response emerges (25, 26) and, in practice,evidence of an antidepressant effect is most likelyto occur within the first 2 weeks of treatment (25–27). Figure 2 shows the initial patterns of responsethat are common following the commencement oftreatment. Remission following treatment of thefirst antidepressant trial, as illustrated by curve A,only occurs in a quarter to a third of patients withmajor depression (28).

In some patients, there may be considerabledelay before symptoms improve (curve A2). There-fore, it is important to administer an adequate dos-age during a trial of 2–4 weeks, while ensuringcompliance.

However, approximately two-thirds of patientswho begin treatment for depression experienceeither partial response (curve B and B2) or noresponse (curve C), as demonstrated by a 20–50%reduction on a depression rating scale. Bothgroups will be discussed in further detail below.When such a scenario occurs, key medication strat-egies include i) switching, ii) augmentation, iii)combination treatment, and iv) alternative options.There are a range of pharmacological interventions

that are more suitable at each stage of iteration,and as discussed in more detail later, there is con-siderable evidence to support the efficacy of eachstrategy. However, there is little information as tohow the strategies compare against each other orsequencing specificities, especially in the case ofpartial response. Therefore, the following sugges-tions are based on expert opinion.

Following the initial response to an antidepres-sant, the long-term response is equally variable(Fig. 3). Once a patient shows short-term improve-ment, the symptoms either continue to improveand reach remission, or the response plateaus.When the latter scenario occurs, a treatmentstrategy change may be necessary. The STAR*Dstudy (28) showed that with successive trials of an-tidepressants, a cumulative response rate of 60%can be achieved, although it may require 2–3 anti-depressant trials before remission is reached. Long-term response patterns also show that despiteswitching to an alternative medication or adding anaugmenting agent, a significant proportion ofpatients with depression still only achieve a partialresponse (B, C) or may remain refractory (D).These patients require a more extensive re-evalua-tion of both the diagnosis and causal factors, and aparadigm shift is warranted. The following sectionsoutline pharmacotherapy processes for patientswith partial response and non-response patterns.

Partial response. If remission is still not achievedafter an appropriate time, an iterative process isnecessary to achieve further symptom remissionand a better outcome. First, there should be areview of the diagnosis to determine whether anti-depressant medication is necessary and whetherthere are any factors that could be maintaining thedepression.

Second, make sure that the dosage was opti-mized with the current antidepressant [TCAs, ven-lafaxine and escitalopram generally haveantidepressant activity across a broader dose range(29–31)] and, finally, adherence was appropriate. Ifthe dose was not optimized, increase the dose. If,after assessing the necessary factors, it transpiresthat further improvement is unlikely on the currentantidepressant, the first treatment strategy thatshould be considered is that of augmentation (referto the section ‘Combination and augmentation’).Augmentation is the preferred course of actionover switching treatments because when some levelof improvement has already occurred, enhance-ment of an ongoing treatment is ideal (28). Fur-ther, switching is likely to introduce the added riskof relapse because of withdrawal of the primaryantidepressant (Fig. 4).

Non response (C)

Delayed response*(A2)

Par al response (B)

Delayed & Par lresponse (B2)

Remission (A)

TimeWeek 4a on oftreatment

Week 12

Return to func oning

Highimpairment

Fig. 2. Schematic representation of potential response pat-terns following an initial antidepressant treatment. *Delayedresponse followed by remission.

Poor response (C)

High impairment

Successive responsewith eventualremission (A)

Plateau (B)

No response (D)

Return to func oning

Fig. 3. Long-term response patterns to treatment.

11

Pharmacological management of depression

Page 7: Clinical overview Pharmacological management of unipolar ...dro.deakin.edu.au/...pharmacologicalmanagement... · Pharmacological management of unipolar depression Malhi GS, Hitching

Once the patient has been on the augmentingagent for a month, evaluate the progress using avalidated scale. If marked improvement is present,continue with treatment and assess again in amonth with the goal of reaching remission. Ifremission is achieved, the augmenting agent shouldbe tapered down after a month of remission. If thesymptoms plateau with the augmenting agent, con-sider stopping the augmenting agent and switchingthe primary agent. When switching antidepres-sants, carefully choose a new antidepressant withevidence for superiority to the initial antidepres-sant. If remission is not achieved by 3 months, aparadigm shift is warranted. Identify and investi-gate any unique factors that may be contributingto the patient’s depression, including any social,lifestyle or other variables that may influence thedelay in treatment progression. Also consider thatthe treatment algorithm (Fig. 5) and clinical man-agement and therapeutic strategies that are com-monly used may not be appropriate for thepatient.

Treatment resistance/non-response. Despite a clini-cian’s best effort to subtype and tailor a patient’s

treatment, some patients do not respond to treat-ment as anticipated. When there has been noimprovement following an optimized dose of anantidepressant, the first step should be a para-digm shift. While taking into consideration thatthe treatment algorithm may not be appropriatefor this patient, re-evaluate the diagnosis, takingnote that the patient may have developed newsymptoms, the depression is more complex thanoriginally thought or an alternative diagnosisbetter fits the patient’s symptoms. During thisassessment, other unique and individualized fac-tors should be taken into consideration, such asthe patient’s personality structure and lifestyle,social factors or substance abuse. Seek advicefrom colleagues if necessary. If it has been estab-lished that the original diagnosis was accurate,switch the patient to a new antidepressant. Whenchoosing a new antidepressant, consider one thathas shown evidence for superiority to the origi-nal antidepressant (refer back to the section,‘Selecting an antidepressant’). Before switchingthe patient’s medication, consider which switch-ing strategy (refer to the section ‘Substitute/switching treatments’) will be utilized and

Fig. 4. Treatment algorithm for managing varying response following first-line antidepressant. Prior to commencing a new treat-ment strategy, it is important to re-evaluate symptoms to ensure the diagnosis is correct and that pharmacotherapy is a suitableoption.

12

Malhi et al.

Page 8: Clinical overview Pharmacological management of unipolar ...dro.deakin.edu.au/...pharmacologicalmanagement... · Pharmacological management of unipolar depression Malhi GS, Hitching

educate the patient about the strategy and thenew medication. Once the patient has been onthe new antidepressant for a month, evaluate thepatient’s progress using a validated scale. If thereis a decline in symptoms, continue with the treat-ment and assess again in a month with the goalof reaching remission. If the symptoms plateau,consider adding an augmenting agent or combin-ing with another antidepressant (refer to the sec-tion ‘Combination and augmentation’). If, afterfurther evaluation of the patient, it is determinedthat the patient is still not responsive to antide-pressant medication, the patient is considered tohave treatment resistant depression (TRD),defined as a lack of improvement following ade-quate trials of two or more antidepressants. WithTRD, research supports the addition of an evi-dence-based psychotherapy (32), switching to aneurostimulation treatment such as ECT (33),and continuing with pharmacological strategies.These strategies are addressed below and expandon our previous treatment guidelines (1).

Strategies

Increase dose. As mentioned previously, beforealtering any treatment, allow a trial of appropriateduration, usually 2–6 weeks, at adequate dosage(34, 35). Many antidepressants have a relativelynarrow therapeutic range in which the agent isconsidered effective and safe, and research showsthat increasing the dose on these medications isnot any more effective (36). However, others, suchas venlafaxine and TCAs (other than nortrypti-line), have very broad dose ranges with up to aten-fold increase in oral dosage. For instance, ven-lafaxine can be safely administered at effectivedoses from 37.5 to 375 mg (37). However, clinicalmonitoring at high doses is especially important asside-effects and therapy discontinuations usuallyincrease with dosage.

It is difficult to be exact regarding duration ofmedication at any particular stage of treatmentbecause both clinical improvement and lack ofresponse are susceptible to many factors that canalter outcome. For example, relationship and

occupational factors can often impact clinicalsymptoms and either accelerate or hinder improve-ment. Therefore, duration recommendations areonly a guide. If an adequate dose has been admin-istered for 6 weeks and there is non-response,switch medications, but if there is partial response,continue the current medication for a further6 weeks (at the same dose; but a future doseincrease can be considered if remission is notachieved). If after 12 weeks the patient does notimprove, follow the guidelines for partial ornon-response strategies.

Combination and augmentation. Combining antide-pressants (38), which entails either adding anantidepressant to another antidepressant, or aug-mentation (14), which includes adding a ‘boost-ing’ agent to an antidepressant or increasing thedose, are both seemingly effective strategies. Bothare commonly employed to enhance the effect ofongoing antidepressant treatment; however, evi-dence for either is limited, and that which isavailable is constrained by the small sample sizesof most studies and the lack of comparisons toplacebo. Clinically, combining two or moreantidepressants to enhance therapeutic effects ordecrease side-effects is common practice (39), yetresearch on the efficacy of specific antidepressantcombinations is more sparse than augmentationstrategies. Both combining and augmentationshare problems concerning safety and tolerabil-ity. Therefore, it is important to monitor thepatient carefully for side-effects and potentialtoxicity.

Combination. Research has shown that combina-tions of antidepressants can be more effective thanmonotherapy without compromising tolerance (40,41). However, long-term placebo-controlledcomparator studies are needed to substantiate arecommendation of polypharmacy (42). Whendetermining which antidepressants to combine, arationale based on pharmacokinetic advantage orsynergistic action should inform the decision, inaddition to past experience and knowledge. Forexample, venlafaxine and mirtazapine are pharma-cologically synergistic or complementary becauseof their distinct receptor profiles (43).

A double-blind randomized trial where patientswere treated with combinations of mirtazapine andfluoxetine, or venlafaxine or buproprion showedthat these were equally effective for depression asfluoxetine monotherapy and that they were welltolerated (40). A recent meta-analysis found thatmirtazapine and tricyclic antidepressants incombination with SSRIs are better than SSRI

Reviewadherence &

dose

Re-assess for

Seekconsul on

Re-evaluatediagnosis

ECT

Augment/Combine

ClinicalM

anagement

Therapeuc

Strategies

Fig. 5. Managing partial or non-response with clinicalmanagement and therapeutic strategies (1).

13

Pharmacological management of depression

Page 9: Clinical overview Pharmacological management of unipolar ...dro.deakin.edu.au/...pharmacologicalmanagement... · Pharmacological management of unipolar depression Malhi GS, Hitching

monotherapy in achieving remission (44). Addi-tionally, of the studies that reported drop-out ratesand side-effects, there was no difference betweenthe combined and monotherapy groups. Similarly,a double-blind study that compared mirtazapinemonotherapy and paroxetine monotherapy againsta combination of both antidepressants reportedthat remission rates after 6 weeks were highest forthe latter (45). However, not all research has sup-ported the use of combinations and not all combi-nations are pharmacologically sound (34).

Augmentation. Similar to combining antidepres-sants, augmenting agents, such as lithium, T3 (46–48) and some antipsychotics, can enhance the effectof an antidepressant, especially in treatment strate-gies for TRD. T3 is beneficial for patients withsubclinical hypothyroidism who suffer fromdepression as a consequence. However, T3 is alsoused as an augmentation agent for depressedpatients without hypothyroidism. Patients aug-mented with T3 should be monitored in the sameway as patients with hypothyroidism, with TSH,free T4 and T3 levels regularly checked (48).

Lithium, a widely supported and used augmen-tation agent, is found to be more effective than pla-cebo in augmentation of TCAs, SSRIs and otherantidepressants (49, 50). It is recommended thatlithium be administered once daily at an oral dosethat achieves plasma levels within the therapeuticrange (0.5–1.0 mEq/l) (51). If there is no responseto lithium within 7–10 days, alternative strategiesshould be considered. Care should be taken whendiscontinuing the use of lithium as research showsthat antidepressant augmentation by lithiumresults in significantly higher relapse rates afterabrupt withdrawal (52).

Atypical antipsychotics are widely used clini-cally as add-on agents for TRD, and in additionto empirical evidence, research studies supportthis strategy (53, 54). Placebo-controlled studieshave found that aripiprazole, olanzapine andrisperidone can be effective as augmentationagents. These are generally administered at muchlower doses than those recommended for schizo-phrenia and bipolar disorder (54–56), but thispractice is off-label and these agents are not for-mally indicated for augmentation. Further,adverse effects of these agents must be closelymonitored, as weight gain, potential metabolicsyndrome and extrapyramidal side-effects are ofgreat concern, especially in the context of long-term therapy.

Substitution/switching medication. There are twomain considerations when switching antidepres-

sants: first, which antidepressant to trial next andsecond, which strategy to employ when switching.

Switching antidepressants. Randomized control tri-als have found that switching to a different antide-pressant improves response and remission rateswhen switching for both non-response (57) andintolerability reasons. Switching within a class (e.g.from one SSRI to another) is no more effectivewhen compared to switching out of class (e.g. froman SSRI to a non-SSRI) (28, 58, 59), but if treat-ment has been curtailed (e.g. medication has onlybeen taken for a few days) because of side-effects,then switching within a class to another agent maybe a worthwhile option.

The majority of patients begin with a first-lineantidepressant, such as an SSRI. Patients who areintolerant to the initial SSRI often benefit from asecond SSRI. It is recommended that patients usea lower dose of the second SSRI initially, and itmay be necessary to taper the first SSRI for alonger period (e.g. switching from citalopram toescitalopram can be done immediately; switchingfrom fluoxetine to another antidepressant, wash-out period of at least a week is suggested beforecommencing the second agent at a lower dose).However, if the patient was non-responsive to theinitial SSRI, switching out of class to a SNRI (i.e.venlafaxine), norepinephrine-dopamine reuptakeinhibitor (NDRI) or TCA is recommended (58,60). Switching to a TCA carries the benefit ofhigher response rates but is disadvantaged bygreater risk of side-effects. If a patient begins witha TCA, which is uncommon, switching to anotherTCA has the advantage that a switch can occurwithout a wash-out period, and the dose on thenew TCA can remain the same. As SSRIs are typi-cally the starting point for antidepressant treat-ment, the majority of research examines the switchfrom SSRIs to another antidepressant. Therefore,there is a lack of research evidence for othercombinations but most combinations have beensupported.

Switching strategies. There are three main strate-gies that should be considered when deciding thebest way to switch medications: i) overlap medica-tions, ii) taper or stop/start medications; and iii)have a washout period. The most common strategyutilized in practice is to introduce and withdrawmedication gradually to minimize any un-medi-cated period and avoid the risk of the serotoninsyndrome. However, the preferred switching strat-egy chosen should depend on the degree ofresponse achieved from the initial antidepressanttrial (Fig. 6).

14

Malhi et al.

Page 10: Clinical overview Pharmacological management of unipolar ...dro.deakin.edu.au/...pharmacologicalmanagement... · Pharmacological management of unipolar depression Malhi GS, Hitching

(A) Concurrent switch: Changes in dose of bothmedications are implemented simultaneously, thus,overlapping medications. Although this optionoffers the advantage that the individual is alwaysmedicated, there is an increased likelihood of inter-actions and side-effects because the medicationsare administered jointly.

(B) Overlapping switch: Dose changes are onlyimplemented in one medication at a time. The ini-tial medication is continued at full dose while grad-ually commencing the new medication. Begintapering the medication that is being substitutedonce optimal dose has been met on the new medi-cation. This strategy is preferred for partialresponse because the cross-over is important tohelp retain any benefit achieved from the initialmedication thus far. Similar to concurrent switch,there is an increased risk of interactions and side-effects because the medications are overlapped.However, iatrogenic side-effects are more easilyidentified because only one medication is modifiedat a time.

(C) Sequential switch: Taper the initial medicationand once this has been fully withdrawn, graduallyintroduce and titrate the new medication to opti-mal dosage. This option is the cleanest way ofsubstituting one medication for another but takesmuch longer, especially if it also includes a wash-out period. This option also increases the chanceof worsening because there is a considerable period

when medication is at a subtherapeutic dose.Therefore, this strategy is preferable for those whodid not respond to the initial medication, becausethis switch can be done quickly without a washoutperiod and has the advantage of not overlappingmedications, which increases side-effects. However,switching to, or from, an MAOI requires a clearwashout period of at least a week to avoid the riskof severe drug–drug interactions.

Electroconvulsive therapy (ECT)

Electroconvulsive therapy involves applying elec-trical current to the brain to induce a convulsion(seizure) and is an effective alternative treatmentoption in cases of marked severity, risk or ongoingnon-response to medication or psychological treat-ments (33, 61–63). The recommended frequency istwo to three times per week under monitored con-ditions to achieve rapid and effective antidepres-sant effects (64, 65). Of note, the speed of responseis usually faster with three administrations perweek; however, the degree of cognitive impairmentis likely to be greater even if the overall number ofECT treatments is the same (64). ECT should beused in conjunction with ongoing antidepressanttreatment (66). The advantage of this strategy isthat it decreases the frequency of early post-ECTrelapse, which can be quite high in the first6 months if ECT is used on its own (67). If concur-rent antidepressant treatment is not used withECT, an antidepressant trial should be reintro-duced following ECT treatment as antidepressantsare often effective after ECT, even if they were notparticularly beneficial prior.

Overall, ECT is very effective, with responserates of up to 80–90% (68), but is considerablylower for patients who have failed to respond toadequate antidepressant medication trials (69).However, the lower response rates in theseinstances may reflect the greater severity of the ill-ness, as ECT is commonly used as a last resortonce antidepressants have failed. Of note, ECT ismore effective than antidepressant monotherapyfor patients with melancholia (70) and especiallythose with psychotic depression (68). Additionally,ECT is also recommended for treating severedepression during pregnancy as it poses less of arisk to the foetus and mother (71) compared withantidepressant use (72, 73).

Novel treatments

Other options that include neurostimulation arerepetitive transcranial magnetic stimulation(rTMS), vagus nerve stimulation (VNS) and deep

*Non- response

al response

(b) Overlapping switch (a) Concurrent switch

(c) Sequential switch

Fig. 6. Schematic representation of strategies for switchingantidepressant medication. The preferred switching strategychosen should depend on the degree of response achieved fromthe initial antidepressant trial: (a) changes in dose of both med-ications are implemented simultaneously, thus, beneficial forpartial response; (b) dose changes are only implemented in onemedication at a time, while holding the initial medication con-stant at the original dose until the second medication hasreached optimal dose; thus, this strategy is suitable for partialresponse; (c) this option is the safest way (least likely to causeany interactions) of substituting one medication for anotherand is preferable for those who did not respond to the initialmedication.

15

Pharmacological management of depression

Page 11: Clinical overview Pharmacological management of unipolar ...dro.deakin.edu.au/...pharmacologicalmanagement... · Pharmacological management of unipolar depression Malhi GS, Hitching

brain stimulation (DBS), the latter two of whichare more invasive than ECT. rTMS involves gener-ating a superficial current by way of electromag-netic induction in the dorsolateral prefrontalcortex. Unlike ECT, it has shown minimal adverseneurocognitive effects (74). The VNS device has agood safety profile but the efficacy is still beingevaluated. DBS involves implanting a stimulatingelectrode in a specified area of the brain. Earlyresults in open-label trials of DBS use for depres-sion treatment are promising (75–77), but needRCT evidence before it can be recommended forclinical practice.

Substantial resources are being devoted to thegenetic epidemiology of depression and areanother source for novel treatment options fordepression. Although the overall genetic contribu-tion to depression is likely to be minor (78),preliminary research suggests that efficacy ofantidepressants may be predicted by genetic mark-ers (79), yet more research is necessary to replicatethese findings. Escitalopram response was pre-dicted by a marker in the gene encoding interleu-kin-11 and IL-6 gene, in which the latter has beenestablished as a candidate gene for depression (80).Additionally, response to nortriptyline has beenshown to be associated with the UST gene,although this gene has yet to be an establishedgenetic marker for depression. Brain-derived neu-rotrophic factor (BDNF) has been examined as asusceptibility locus for the development of depres-sion, but the results have been inconclusive as thereare contradictory findings in recent research (81–83).Additionally, the gene–environment hypothesisfocused on the link between the serotonin trans-porter-linked polymorphic region (5-HTTLPR)and stressful life events on increased risks ofdepression. Although the results suggest that per-sistent depression increases with this specific gene xenvironment interaction, further research hasfailed to replicate these findings (84, 85).

Importance of re-evaluating diagnosis

It is unlikely that core features of major depres-sion, such as depressed mood and feelings ofhopelessness, occur in isolation from symptoms ofanxiety, substance misuse, personality disordersor eating disorders. In fact, a large proportion(estimated 50–70%) of patients with depressionreport anxiety symptoms (86–88). These com-monly occurring comorbidities complicate diagno-sis and treatment and contribute to a poorerprognosis (89, 90). Therefore, re-evaluating thediagnosis following partial and non-response toantidepressant treatment is crucial, keeping in

mind that depression may not be the primary orsole diagnosis.

Bipolar depression is another complex layer thatshould be explored when diagnosing and treating,as 10–20% of patients diagnosed with unipolardepression will experience hypomania or mania inthe course of their illness, resulting in a revision ofdiagnosis to bipolar disorder (91, 92). There is alsogrowing evidence to suggest that hypomanic symp-toms commonly occur in a subset of patients withdepression and these contribute to treatment non-response (93). Predictive factors for bipolar disor-der that have been robust across studies includeearly age at illness onset (94–96), presence of psy-chosis (94, 97, 98) and family history of mania (94,95, 97–99). Even if manic symptoms were previ-ously explored, critically examine the patient’s pastand recent history specifically targeting symptomsof mania or hypomania.

Paradigm shift

In the event that the iterative treatment optionshave been explored and the patient is still non-responsive, consider individual factors, uniquecharacteristics and other variables that may beinfluencing treatment resistance. Additionally,consider other treatment options outside of thetreatment algorithm. The patient may have devel-oped new symptoms; the depression is more com-plex than originally thought; or there are social/lifestyle factors that may be a contributing factor.For example, since the commencement of the ini-tial antidepressant, the patient may have lost hisjob, which in turn negatively affects his relation-ship and housing situation. In this instance, anoverall reassessment is needed while taking intoconsideration the patient’s lifestyle, psychologicaltreatments, substance abuse and/or more aggres-sive treatment options (e.g. ECT). It is possiblethat the original diagnosis may be wrong. Return-ing to the 5P Model and reassessing the patient’sproblems are in the best interest for the patient andmay open up new possibilities for treatment. If ithas been established that the original diagnosis iscorrect and the patient’s symptoms are non-responsive to the previously suggested treatmentoptions, novel treatment options may be warranted.

Remission/recovery

Duration of treatment. Long-term antidepressanttreatment should be considered on an individualbasis, while taking into consideration the risks ofcontinued medication use and benefits of preventa-tive care. Research recommendations suggest that

16

Malhi et al.

Page 12: Clinical overview Pharmacological management of unipolar ...dro.deakin.edu.au/...pharmacologicalmanagement... · Pharmacological management of unipolar depression Malhi GS, Hitching

antidepressant treatment should continue for atleast 1 year following the onset of symptoms foran initial episode and at least 2 years if the patientcarries any risk factors, such as having many priordepressive episodes, particularly if they are severe,chronic and difficult to treat with residual symp-toms, significant comorbidity or the patient is ofolder age (100–103). If psychotic features are pres-ent, then continue treatment for up to 3 years(104). Meta-analysis results suggest that to main-tain remission, the treating dose should be contin-ued; decreasing the dose significantly increasesrelapse/recurrence rates (105).

During antidepressant maintenance, thepatient’s condition should continually be assessedas the risk of relapse is relatively high and isincreased by a variety of factors (Fig. 7). Relapserates are highest immediately following remissionand diminish with time (106). In addition to anti-depressant treatment, cognitive behavioural ther-apy (CBT) has shown long-term effects inpreventing relapse and recurrences. Thus, integrat-ing CBT with antidepressant treatment maydecrease both relapse rates and the need for long-term antidepressant maintenance (32).

Ensuring adherence. Patient improvement withpharmacotherapy heavily relies on adherence;thus, it is crucial that adherence is activelyaddressed and monitored. Patients who aredepressed are unmotivated and feel down, soengaging in an unnatural routine of taking medica-tion, which is an anomalous habit for healthy indi-viduals, is even more difficult for those who aredepressed. To increase adherence, educationalinterventions should be used in conjunction withbehavioural changes in order to gain maximumbenefit. Complex behavioural changes are the mosteffective in improving antidepressant medication

adherence (107). Prior to starting antidepressanttreatment, both the physician and patient mustunderstand what the ultimate goal of treatment isand why medication is important. The patientshould be clear of what is necessary to achieve thisgoal, including what should be done if a dose ismissed. Missing a dose should be activelyaddressed so the patient understands that missinga dose does not result in failed treatment, butrather, repeatedly neglecting the prescribed dosagecauses ineffective treatment. Patients should beadvised not to ‘double-up’ their medication shouldthey miss a dose. When assessing the patient ateach visit, physicians should include questionsregarding adherence, which should become routineduring each visit. Although side-effects and toler-ance are often viewed as negative, they are anotherindicator of patient adherence. Thus, if the patientreports no side-effects, the patient’s adherenceshould be questioned. When side-effects such asweight gain or sexual dysfunction compromiseadherence to medication, it is worthwhile switchingto an antidepressant with a more favourable side-effect profile, such as agomelatine that does notcarry these side-effects (108, 109), to ensure main-tenance of treatment.

Titrating treatment. Treatment guidelines recom-mend upward dose titration during the initialstage of antidepressant treatment (110–112).Titration upwards normally involves increasingthe antidepressant dose gradually while monitor-ing for efficacy, and as mentioned, it is an iterativeprocess that may involve many dose increasesdepending on the antidepressant being prescribed.However, the starting dose needs to be taken intoconsideration when weighing the advantages anddisadvantages of titration. The initial antidepres-sant dose is likely to vary depending on whetherthe clinician has planned titration a priori or hasmade a decision to increase dosage post hoc. Thelatter is usually the consequence of a partial ornon-response. In general, titration is a usefulapproach for reducing side-effects, but it delaysthe achievement of a therapeutic dose, and thelikelihood of a clinical response. However,research shows that patient adherence is signifi-cantly greater when patients begin on a low dosewith future titration compared to patients whocommence treatment at therapeutic levels.Patients who commenced antidepressant treat-ment at a dose less than or equal to the dose rec-ommended in the APA guidelines and whotitrated their antidepressant dose within 60 dayswere 2.6 times more likely to adhere to treatmentthan patients who began at a dose recommended

Presence of psychosis

Treatment resistance

Time

Concurrent factorsGender – FemaleLife events/social stressComorbid medical illness

Residual symptoms

on of episode

Severity ofdepression

Fig. 7. Factors increasing the risk of acute relapse in depres-sion. Concurrent factors should be taken into considerationwhen choosing treatment options. Note. Specific factors thatincrease the risk of relapse are labelled at the point of wherethey are likely to impact treatment.

17

Pharmacological management of depression

Page 13: Clinical overview Pharmacological management of unipolar ...dro.deakin.edu.au/...pharmacologicalmanagement... · Pharmacological management of unipolar depression Malhi GS, Hitching

by APA guidelines. These findings apply to arange of SSRI and SNRI medications (113).

Commencing a patient on antidepressant levelsat, or less than, the recommended dose with theintention to titrate treatment is particularly usefulfor patients who are frail (e.g. elderly, those withcomorbid medical illness) (114, 115) or whencommencing certain classes of antidepressants. Forexample, tricyclic antidepressants need to betitrated because of potential cardiac toxicity associ-ated with peak plasma levels (116). Additionally, inpractice, patients on SSRIs and SNRIs candecrease their risk of transitory side-effects by com-mencing treatment on low doses and titrating thedose slowly to reach therapeutic levels (115, 117).

Although this initial titration of treatmentincreases the complexity of management, it doesfoster closer clinician follow-up and ensures adher-ence. Therefore, the combined effect of titrationand regular clinician follow-up should significantlyimprove patient outcome.

Discussion

Psychological/social/lifestyle factors

Depression is a complex neuropsychiatric disorderthat is unlikely caused solely by biological factors,but rather includes psychological, personality andsocial components (118). Therefore, the election oftreatment for depression should incorporate phar-macological, psychological and lifestyle aspectsand be individually tailored for the patient. Ourcompanion paper on ‘Psychological Managementof Unipolar Depression’ (3) explains in more detailvarious therapeutic approaches and appropriatetreatment appraisal based on presenting symptomsand patient characteristics/preferences and history.The likely outcome from treatment also needs toconsider the patient’s lifestyle and personalchoices. Our companion paper on the ‘LifestyleManagement of Unipolar Depression’ (4) will alsoassist clinicians in diminishing the likelihood ofpartial and non-response. Moreover, the inclusionof all three recommendations is ideal in the mainte-nance phase of treatment.

Integration of care

Role of primary care physician and general practi-tioner. Primary care physicians are pivotal tomental health care, especially if they have beentreating the patient long term. The amount ofinvolvement the primary physician has with treat-ing the patient’s depression depends on the com-plexity of the patient’s illness and the level of

involvement of the psychiatrist. Patients who aretreated by a psychiatrist on a regular basis willhave less contact with their primary physician forthe depression treatment, yet the GP should con-tinue to be aware of the treatment in case of emer-gency and the patient needs medical assistance onshort notice. Patients with mild cases of depressionare likely to continue seeing their primary physi-cian for monitoring and treatment maintenance asthey are less likely to engage with a psychiatrist fulltime. Regardless of the patient’s severity, the psy-chiatrist and primary physician should communi-cate with each other regarding changes in mentalor physical state and changes in treatment.

Family and friends. Depressed patients are likely toexperience stress associated with their family andother relationships, which is often a factor contrib-uting to the patient’s illness (119). However, posi-tive social networks are fundamental for patientimprovement, and these relationships should bemaximized throughout treatment.

In conclusion, the pharmacotherapy of depres-sion is a complex process because it is a heteroge-neous disorder and each individual, when affected,requires consideration of unique factors. In addi-tion, effective treatment relies on both empiricalevidence and clinical experience. In practice, firstand foremost, adherence is essential, and this needsto be fully appreciated by the patient and regularlymonitored by the physician. Response rates to ini-tial antidepressant treatment are lower than thosereported in clinical trials and as such follow-upwith subsequent treatments is often necessary.Careful and comprehensive clinical assessment ofthe symptomatic profile is critical and shouldinclude the use of a standardized rating scale (e.g.HAM-D) to gauge severity and track change.Although there are many strategies and withinthese there are multiple options, the extantresearch suggests that adding an augmentationagent and switching to a new antidepressant areeffective tactics and that these are best suited forpartial and non-responders respectively. In casesof TRD, it is important to consult with colleaguesto gain a different perspective and consider alter-nate treatment paradigms. It is also useful to thor-oughly re-evaluate the diagnosis noting that theremay be additional contributing factors as the ill-ness has most likely evolved from the time of initialpresentation.

Acknowledgements

This publication has been made possible through infrastruc-ture support from NHMRC Program Grant (510135) and

18

Malhi et al.

Page 14: Clinical overview Pharmacological management of unipolar ...dro.deakin.edu.au/...pharmacologicalmanagement... · Pharmacological management of unipolar depression Malhi GS, Hitching

CADE Clinic, Department of Psychiatry, The University ofSydney, and an unrestricted educational grant from EliLilly.

Declaration of interest

In the past 3 years, Professor Gin Malhi has served on a num-

ber of international and national pharmaceutical advisory

boards, received funding for research and has been in receipt

of honoraria for talks at sponsored meetings worldwide involv-

ing the following companies: AstraZeneca, Lundbeck, Sanofi-

Aventis, Eli Lilly, Janssen-Cilag, Organon, Pfizer, Servier and

Wyeth. Professor Michael Berk has received funding for

research from Stanley Medical Research Foundation, MBF,

NHMRC, Beyond Blue, Geelong Medical Research Founda-

tion, Bristol Myers Squibb, Eli Lilly, Glaxo Smith K-line,

Organon, Novartis, Mayne Pharma, Servier, AstraZeneca, has

received honoraria for speaking engagements from AstraZene-

ca, Bristol Myers Squibb, Eli Lilly, Glaxo Smith K-line, Jans-

sen-Cilag, Lundbeck, Organon, Pfizer, Sanofi Synthelabo,

Solvay, Wyeth and served as a consultant to AstraZeneca,

Bristol Myers Squibb, Eli Lilly, Glaxo Smith K-line, Janssen-

Cilag, Lundbeck, Pfizer. In the last 5 years, Professor Richard

Porter has received honoraria for speaking engagements from

Janssen-Cilag and Sanofi-Aventis. Professor Philip Boyce has

served on a number of international and national pharmaceuti-

cal advisory boards, received funding for research and has

been in receipt of honoraria for talks at sponsored meetings

worldwide involving the following companies: AstraZeneca,

Lundbeck, Eli Lilly, Servier, Janssen-Cilag and Boehringer In-

gelheim. Kristina Fritz and Rita Hitching have no competing

interests to report.

References

1. Malhi GS, Adams D, Porter R et al. Clinical practice rec-ommendations for depression. Acta Psychiatr Scand2009;119(Suppl. 439):8–26.

2. Malhi GS, Hitching R, Coulston CM, Boyce P, Porter R,Fritz K. Preface: individualized management of unipolardepression. Acta Psychiatr Scand 2013;127(Suppl. 443):1–5.

3. Lampe L, Coulston CM, Berk L. Psychological manage-ment of unipolar depression. Acta Psychiatr Scand2013;127(Suppl. 443):24–37 .

4. Berk M, Sarris J, Coulson CE, Jacka FN. Lifestyle man-agement of unipolar depression. Acta Psychiatr Scand2013;127(Suppl. 443):38–54.

5. Serretti A, Drago A, Spina E. Pharmacogenomics ofDepression. Adv Biol Psychiatry 2010;25:58–74.

6. Bech P. The Bech–Rafaelsen Melancholia Scale (MES)in clinical trials of therapies in depressive disorders: a20-year review of its use as outcome measure. ActaPsychiatr Scand 2002;106:252–264.

7. McIntyre RS, Konarski JZ, Kennedy SH, Dickens SE,Bagby RM. Measuring depressive symptoms in the natu-ralistic primary-care setting. Int J Clin Pract2007;61:1278–1282.

8. Nemeroff CB, Entsuah R, Benattia I, Demitrack M, SloanDM, Thase ME. Comprehensive analysis of remission(COMPARE) with venlafaxine versus SSRIs. BiolPsychiatry 2008;63:424–434.

9. Taylor MA, Fink M. Restoring melancholia in theclassification of mood disorders. J Affect Disord2008;105:1–14.

10. Gunnell D, Saperia J, Ashby D. Selective serotonin reup-take inhibitors (SSRIs) and suicide in adults: meta-anal-ysis of drug company data from placebo controlled,randomised controlled trials submitted to the MHRA’ssafety review. BMJ 2005;330:385.

11. Hammad TA, Laughren TP, Racoosin JA. Suicide rates inshort-term randomized controlled trials of newer antide-pressants. J Clin Psychopharmacol 2006;26:203–207.

12. Hansen DG, Vach W, Rosholm J, Sondergaard J, GramLF, Kragstrup J. Early discontinuation of antidepres-sants in general practice: association with patient andprescriber characteristics. Fam Pract 2004;21:623–629.

13. Olfson M, Marcus SC, Tedeschi M, Wan GJ. Continuityof antidepressant treatment for adults with depressionint he United States. Am J Psychiatry 2006;163:101–108.

14. Trivedi MH, Rush AJ, Gaynes BN et al. Maximizing theadequacy of medication treatment in controlled trialsand clinical practice: STAR(*)D measurement-basedcare. Neuropsychopharmacology 2007;32:2479–2489.

15. Anderson IM. Selective serotonin reuptake inhibitorsversus tricyclic antidepressants: a meta-analysis of effi-cacy and tolerability. J Affect Disord 2000;58:19–36.

16. No€el PH, Aguilar C, Cornell J. A systematic review ofnewer pharmacotherapies for depression in adults:evidence report summary. Ann Intern Med 2000;132:743–756.

17. Arroll B, Macgillivray S, Ogston S et al. Efficacy andtolerability of tricyclic antidepressants and SSRIs com-pared with placebo for treatment of depression inprimary care: a meta-analysis. Ann Fam Med 2005;3:449–456.

18. Kennedy SH, Andersen HF, Thase ME. Escitalopramin the treatment of major depressive disorder: a meta-analysis. Curr Med Res Opin 2009;25:161–175.

19. Cipriani A, Furukawa TA, Geddes JR et al. Doesrandomized evidence support sertraline as first-line anti-depressant for adults with acute major depression?A systematic review and meta-analysis. J Clin Psychiatry2008;69:1732–1742.

20. Schueler YB, Koesters M, Wieseler B et al. A systematicreview of duloxetine and venlafaxine in major depres-sion, including unpublished data. Acta Psychiatr Scand2011;123:247–265.

21. Cipriani A, Furukawa TA, Salanti G et al. Comparativeefficacy and acceptability of 12 new-generation antide-pressants: a multiple-treatments meta-analysis. Lancet2009;373:746–758.

22. Eyding D, Lelgemann M, Grouven U et al. Reboxetinefor acute treatment of major depression: systematicreview and meta-analysis of published and unpublishedplacebo and selective serotonin reuptake inhibitor con-trolled trials. Br Med J 2010;341:c4737.

23. Preskorn SH, Borges-Gonzalez S, Flockhart D. Clini-cally relevant pharmacology of neuropsychiatric drugsapproved over the last three years: part II. J PsychiatrPract 2006;12:312–316.

24. Quera-Salva MA, Lemoine P, Guilleminault C. Impact ofthe novel antidepressant agomelatine on disturbed sleep-wake cycles in depressed patients. Hum Psychopharma-col 2010;25:222–229.

25. Posternak MA, Zimmerman M. Is there a delay in theantidepressant effect? A meta-analysis. J Clin Psychiatry2005;66:148–158.

26. Stassen HH, Kuny S, Hell D. The speech analysisapproach to determining onset of improvement underantidepressants. Eur Neuropsychopharmacol 1998;8:303–310.

19

Pharmacological management of depression

Page 15: Clinical overview Pharmacological management of unipolar ...dro.deakin.edu.au/...pharmacologicalmanagement... · Pharmacological management of unipolar depression Malhi GS, Hitching

27. Stassen HH, Angst J, Delini-Stula A. Fluoxetine versusmoclobemide: cross-comparison between the timecourses of improvement. Pharmacopsychiatry 1999;32:56–60.

28. Rush AJ. STAR*D: what have we learned? Am J Psychi-atry 2007;164:201–204.

29. Adli M, Baethge C, Heinz A, Langlitz N, Bauer M. Isdose escalation of antidepressants a rational strategyafter a medium-dose treatment has failed? A systematicreview. Eur Arch Psychiatry Clin Neurosci 2005;255:387–400.

30. Burke WJ, Gergel I, Bose A. Fixed-dose trial of the sin-gle isomer SSRI escitalopram in depressed outpatients.J Clin Psychiatry 2002;63:331–336.

31. Rudolph RL, Derivan AT. The safety and tolerability ofvenlafaxine hydrochloride: analysis of the clinical trialsdatabase. J Clin Psychopharmacol 1996;16:54S–61S.

32. Parikh SV, Segal ZV, Grigoriadis S et al. Canadian Net-work for Mood and Anxiety Treatments (CANMAT)clinical guidelines for the management of major depres-sive disorder in adults. II. Psychotherapy alone or incombination with antidepressant medication. J Clin Psy-chopharmacol 2009;117:S15–S25.

33. Kennedy SH, Milev R, Giacobbe P et al. Canadian Net-work for Mood and Anxiety Treatments (CANMAT)clinical guidelines for the management of depressive dis-order in adults. IV. Neurostimulation therapies. J AffectDisord 2009;117:S44–S53.

34. Licht RW, Qvitzau S. Treatment strategies in patientswith major depression not responding to first-line sertra-line treatment. A randomised study of extended durationof treatment, dose increase or mianserin augmentation.Psychopharmacology 2002;161:143–151.

35. Henricus GR, Huyser J, Swinkles JA, Schene AH. Doseescalation for insufficient response to standard-doseselective serotonin reuptake inhibitors in major depres-sive disorder. Br J Psychiatry 2006;189:309–316.

36. Ruhe HG, Huyser J, Swinkels JA, Schene AH. Doseescalation for insufficient response to standard-doseselective serotonin reuptake inhibitors in major depres-sive disorder: systematic review. Br J Psychiatry2006;189:309–316.

37. Debonnel G, Saint-Andre E, Hebert C, de Montigny C,Lavoie N, Blier P. Differential physiological effects of alow dose and high doses of venlafaxine in major depres-sion. Int J Neuropsychopharmacol 2007;10:51–61.

38. Corya SA, Williamson D, Sanger TM, Briggs SD, CaseM, Tollefson G. A randomized, double-blind compari-son of olanzapine/fluoxetine combination, olanzapine,fluoxetine, and venlafaxine in treatment-resistant depres-sion. Depress Anxiety 2006;23:364–372.

39. Horgan D, Dodd S, Berk M. A survey of combinationantidepressant use in Australia. Australas Psychiatry2007;15:26–29.

40. Blier P, Ward HE, Tremblay P, Laberge L, H�ebert C,Bergeron R. Combination of antidepressant medicationsfrom treatment initiation for major depressive disorder:a double-blind randomized study. Am J Psychiatry2010;167:281–288.

41. Bech P, Fava M, Trivedi MH, Wisniewski SR, Rush AJ.Outcomes on the pharmacopsychometric triangle in bu-propion-SR vs. buspirone augmentation of citalopramin the STAR*D trial. Acta Psychiatr Scand2012;125:342–348.

42. Rush AJ. Combining antidepressant medications: a goodidea? Am J Psychiatry 2010;167:241–243.

43. Malhi GS, Ng F, Berk M. Dual–dual action? Combiningvenlafaxine and mirtazapine in the treatment of depres-sion. Aust N Z J Psychiatry 2008;42:346–349.

44. Rocha FL, Fuzikawa C, Riera R, Hara C. Combinationof antidepressants in the treatment of major depressivedisorder: a systematic review and meta-analysis. J ClinPsychopharmacol 2012;32:278–281.

45. Blier P, Gobbi G, Turcotte JE et al. Mirtazapine andparoxetine in major depression: a comparison of mono-therapy versus their combination from treatment initia-tion. Eur Neuropsychopharmacol 2009;19:457–465.

46. Altshuler LL, Bauer M, Frye MA et al. Does thyroidsupplementation accelerate tricyclic antidepressantresponse? A review and meta-analysis of the literature.Am J Psychiatry 2001;158:1617–1622.

47. Aronson R, Offman HJ, Joffe RT, Naylor CD. Triiodo-thyronine augmentation in the treatment of refractorydepression. A meta-analysis. Arch Gen Psychiatry1996;53:842–848.

48. Rosenthal LJ, Goldner WS, O’Reardon JP. T3 augmen-tation in major depressive disorder: safety consider-ations. Am J Psychiatry 2011;168:1035–1040.

49. Crossley NA, Bauer M. Acceleration and augmentationof antidepressants with lithium for depressive disorders:two meta-analyses of randomized, placebo-controlledtrials. J Clin Psychiatry 2007;68:934–940.

50. Bauer M, Bschor T, Kunz D, Berghofer A, Strohle A,Muller-Oerlinghausen B. Double-blind, placebo-con-trolled trial of the use of lithium to augment antidepres-sant medication in continuation treatment of unipolarmajor depression. Am J Psychiatry 2000;157:1429–1435.

51. Berghofer A, Grof P, Muller-Oerlinghausen B. Recom-mendations for the safe use of lithium. In: Bauer M,Grof P, Muller-Oerlinghausen B, eds. Lithium in neuro-psychiatry: the comprehensive guide. London: Informahealthcare, 2006:443–464.

52. Bschor T, Kunz D, Bauer M. Successful lithium augmen-tation: how long should lithium be continued? XI WorldCongress of Psychiatry; 6-11 August 1999; Paper pre-sented at Hamburg, Germany, 2009.

53. Papakostas GI, Shelton RC, Smith J, Fava M. Augmenta-tion of antidepressants with atypical antipsychoticmedications for treatment-resistant major depressive dis-order: a meta-analysis. J Clin Psychiatry 2007;68:826–831.

54. Philip NS, Carpenter LL, Tyrka AR, Price LH. Augmen-tation of antidepressants with atypical antipsychotics: areview of the current literature. J Psychiatr Pract2008;14:34–44.

55. Berman RM, Marcus RN, Swanink R et al. The efficacyand safety of aripiprazole as adjunctive therapy in majordepressive disorder: a multicenter, randomized, double-blind, placebo-controlled study. J Clin Psychiatry2007;68:843–853.

56. Marcus RN, McQuade RD, Carson WH et al. The effi-cacy and safety of aripiprazole as adjunctive therapy inmajor depressive disorder: a second multicenter,randomized, double-blind, placebo-controlled study.J Clin Psychopharmacol 2008;28:156–165.

57. Nakajima S, Uchida H, Suzuki T et al. Is switching antide-pressants following early nonresponse more beneficial inacute-phase treatment of depression?: A randomizedopen-label trial. Prog Neuropsychopharmacol Biol Psy-chiatry 2011;35:1983–1989.

58. Ruhe HG, Huyser J, Swinkels JA, Schene AH. Switchingantidepressants after a first selective serotonin reuptake

20

Malhi et al.

Page 16: Clinical overview Pharmacological management of unipolar ...dro.deakin.edu.au/...pharmacologicalmanagement... · Pharmacological management of unipolar depression Malhi GS, Hitching

inhibitor in major depressive disorder: a systematicreview. J Clin Psychiatry 2006;67:1836–1855.

59. Souery D, Serretti A, Calati R et al. Switching antide-pressant class does not improve response or remission intreatment-resistant depression. J Clin Psychopharmacol2011;31:512–516.

60. Rush AJ, Trivedi MH, Wisniewski SR et al. Bupropion-SR, sertraline, or venlafaxine-XR after failure of SSRIsfor depression. N Engl J Med 2006;354:1231–1242.

61. Pagnin D, Queiroz V, Pini S, Cassano GB. Efficacy ofECT in depression: a meta-analytic review. J ECT2004;20:13–20.

62. Lisanby SH. Electroconvulsive therapy for depression.N Engl J Med 2007;357:1939–1945.

63. UK ECT Review Group. Efficacy and safety of electro-convulsive therapy in depressive disorders: a systematicreview and meta-analysis. Lancet 2003;361:799–808.

64. Shapira B, Tubi N, Lerer B. Balancing speed of responseto ECT in major depression and adverse cognitiveeffects: role of treatment schedule. J ECT 2000;16:97–109.

65. Ottosson J-O, Odeberg H. Evidence-based electroconvul-sive therapy. Acta Psychiatr Scand 2012;125:177–184.

66. Sackeim HA, Dillingham EM, Prudic J et al. Effect ofconcomitant pharmacotherapy on electroconvulsivetherapy outcomes: short-term efficacy and adverseeffects. Arch Gen Psychiatry 2009;66:729–737.

67. Kellner CH, Knapp RG, Petrides G et al. Continuationelectroconvulsive therapy vs pharmacotherapy forrelapse prevention in major depression: a multisite studyfrom the Consortium for Research in ElectroconvulsiveTherapy (CORE). Arch Gen Psychiatry 2006;63:1337–1344.

68. Petrides G, Fink M, Husain MM et al. ECT remissionrates in psychotic versus nonpsychotic depressedpatients: a report from CORE. J ECT 2001;17:244–253.

69. Prudic J, Haskett RF, Mulsant B et al. Resistance toantidepressant medications and short-term clinicalresponse to ECT. Am J Psychiatry 1996;153:985–992.

70. Rasmussen KG, Stevens SR, Kung S, Mohan A. Melan-cholic symptoms as assessed by the Hamilton Depres-sion Rating Scale and outcomes with and withoutelectroconvulsive therapy on an in-patient mood disor-ders unit. Acta Neuropsychiatr 2010;22:21–25.

71. Miller LJ. Use of electroconvulsive therapy duringpregnancy. Hosp Community Psychiatry 1994;45:444–450.

72. Yonkers KA, Wisner KL, Stewart DE et al. The man-agement of depression during pregnancy: a report fromthe American Psychiatric Association and the AmericanCollege of Obstetricians and Gynecologists. Gen HospPsychiatry 2009;31:403–413.

73. Udechuku A, Nguyen T, Hill R, Szego K. Antidepres-sants in pregnancy: a systematic review. Aust N Z J Psy-chiatry 2010;44:978–996.

74. Huang M-L, Luo B-Y, Hu J-B et al. Repetitive transcranialmagnetic stimulation in combination with citalopram inyoung patients with first-episode major depressive disor-der: a double-blind, randomized, sham-controlled trial.Aust N Z J Psychiatry 2012;46:257–264.

75. Lozano AM, Mayberg HS, Giacobbe P, Hamani C,Craddock RC, Kennedy SH. Subcallosal cingulate gyrusdeep brain stimulation for treatment-resistant depres-sion. Biol Psychiatry 2008;64:461–467.

76. Kennedy SH, Giacobbe P, Rizvi SJ et al. Deep brain stim-ulation for treatment-resistant depression: follow-upafter 3 to 6 years. Am J Psychiatry 2011;168:502–510.

77. Blomstedt P, Sj€oberg RL, Hansson M, Bodlund O, HarizMI. Deep brain stimulation in the treatment of depres-sion. Acta Psychiatr Scand 2011;123:4–11.

78. Lewis CM, Ng MY, Butler AW et al. Genome-wideassociation study of major recurrent depression in theU.K. population. Am J Psychiatry 2010;167:949–957.

79. Uher R, Perroud N, Ng MY et al. Genome-wide phar-macogenetics of antidepressant response in the GEN-DEP project. Am J Psychiatry 2010;167:555–564.

80. Garriock HA, Tanowitz M, Kraft JB et al. Associationof mu-opioid receptor variants and response to citalop-ram treatment in major depressive disorder. Am JPsychiatry 2010;167:565–573.

81. Duncan LE, Hutchison KE, Carey G, Craighead WE.Variation in brain-derived neurotrophic factor (BDNF)gene is associated with symptoms of depression. J AffectDisord 2009;115:215–219.

82. Castr�en E, Rantam€aki T. The role of BDNF and itsreceptors in depression and antidepressant drug action:reactivation of developmental plasticity. Dev Neurobiol2010;70:289–297.

83. Chen L, Lawlor DA, Lewis SJ et al. Genetic associationstudy of BDNF in depression: finding from two cohortstudies and a meta-analysis. Am J Med Genet B Neuro-psychiatr Genet 2008;147B:814–821.

84. Uher R, Caspi A, Houts R et al. Serotonin transportergene moderates childhood maltreatment’s effects on per-sistent but not single-episode depression: replicationsand implications for resolving inconsistent results.J Affect Disord 2011;135:56–65.

85. Risch NHRLT et al. Interaction between the serotonintransporter gene (5-httlpr), stressful life events, and riskof depression: a meta-analysis. JAMA 2009;301:2462–2471.

86. Kessler RC, Nelson CB, McGonagle KA, Liu J, SwartzM, Blazer DG. Comorbidity of DSM-III-R majordepressive disorder in the general population: resultsfrom the US National Comorbidity Survey. Br J Psychi-atry 1996;168:17–30.

87. Fava M, Rush AJ, Alpert JE et al. Difference in treat-ment outcome in outpatients with anxious versusnonanxious depression: a STAR*D report. Am J Psychi-atry 2008;165:342–351.

88. Fava M, Alpert JE, Carmin CN et al. Clinical correlatesand symptom patterns of anxious depression amongpatients with major depressive disorder in STAR*D.Psychol Med 2004;34:1299–1308.

89. Newton-Howes G, Tyrer P, Johnson T. Personality disor-der and the outcome of depression: meta-analysis ofpublished studies. Br J Psychiatry 2006;188:13–20.

90. Rief W, Trenkamp S, Auer C, Fichter MM. Cognitivebehavior therapy in panic disorder and comorbid majordepression. Psychother Psychosom 2000;69:70–78.

91. Fiedorowicz JG, Endicott J, Leon AC, Solomon DA,Keller MB, Coryell WH. Subthreshold hypomanicsymptoms in progression from unipolar major depressionto bipolar disorder. Am J Psychiatry 2011;168:40–48.

92. Coryell W, Endicott J, Maser JD, Keller MB, Leon AC,Akiskal HS. Long-term stability of polarity distinctionsin the affective disorders. Am J Psychiatry 1995;152:385–390.

93. Angst J, Cui L, Swendsen J et al. Major depressive disor-der with subthreshold bipolarity in the National Comor-bidity Survey Replication. Am J Psychiatry 2010;167:1194–1201.

94. Akiskal HS, Walker P, Puzantian VR, King D, RosenthalTL, Dranon M. Bipolar outcome in the course of depres-

21

Pharmacological management of depression

Page 17: Clinical overview Pharmacological management of unipolar ...dro.deakin.edu.au/...pharmacologicalmanagement... · Pharmacological management of unipolar depression Malhi GS, Hitching

sive illness. Phenomenologic, familial, and pharmaco-logic predictors J Affect Disord 1983;5:115–128.

95. Angst J, Sellaro R, Stassen HH, Gamma A. Diagnosticconversion from depression to bipolar disorders: resultsof a long-term prospective study of hospital admissions.J Affect Disord 2005;84:149–157.

96. Beesdo K, Hofler M, Leibenluft E, Lieb R, Bauer M,Pfennig A. Mood episodes and mood disorders: patternsof incidence and conversion in the first three decades oflife. Bipolar Disord 2009;11:637–649.

97. Coryell W, Endicott J, Maser JD, Keller MB, Leon A,HS A. Long-term stability of polarity distinctions in theaffective disorders. Am J Psychiatry 1995;152:385–390.

98. Strober M, Carlson G. Bipolar illness in adolescentswith major depression: clinical, genetic, and psychophar-macologic predictors in a three- to four-year prospectivefollow-up investigation. Arch Gen Psychiatry1982;39:549–555.

99. Geller B, Zimerman B, Williams M, Bolhofner K, CraneyJL. Bipolar disorder at prospective follow-up of adultswho had prepubertal major depressive disorder. Am JPsychiatry 2001;158:125–127.

100. Geddes JR, Carney SM, Davies C et al. Relapse preven-tion with antidepressant drug treatment in depressivedisorders: a systematic review. Lancet 2003;361:653–661.

101. Reynolds CF 3rd, Dew MA, Pollock BG et al. Mainte-nance treatment of major depression in old age. N EnglJ Med 2006;354:1130–1138.

102. Hansen DG, Gaynes BN, Thieda P et al. Meta-analysis ofmajor depressive disorder relapse and recurrence withsecond-generation antidepressants. Psychiatr Serv2008;59:1121–1130.

103. Holma IAK, Holma KM, Melartin TK, Ryts€al€a HJ,Isomets€a ET. A 5-year prospective study of predictorsfor disability pension among patients with major depres-sive disorder. Acta Psychiatr Scand 2012;125:325–334.

104. Viguera AC, Baldessarini RJ, Friedberg J. Discontinuingantidepressant treatment in major depression. Harv RevPsychiatry 1998;5:293–306.

105. Papakostas GI, Perlis RH, Seifert C, Fava M. Antide-pressant dose reduction and the risk of relapse in majordepressive disorder. Psychother Psychosom 2007;76:266–270.

106. Belsher G, Costello CG. Relapse after recovery fromunipolar depression: a critical review. Psychol Bull1988;104:84–96.

107. Chong WW, Aslani P, Chen TF. Effectiveness ofinterventions to improve antidepressant medicationadherence: a systematic review. Int J Clin Pract 2011;65:954–975.

108. Kennedy SH, Rizvi SJ. Agomelatine in the treatment ofmajor depressive disorder: potential for clinical effective-ness. CNS drugs 2010;24:479–499.

109. Montejo AL, Prieto N, Terleira A et al. Better sexualacceptability of agomelatine (25 and 50 mg) comparedwith paroxetine (20 mg) in healthy male volunteers.An 8-week, placebo-controlled study using thePRSEXDQ-SALSEX scale. J Psychopharmacol 2010;24:111–120.

110. American Psychiatric Association. Practice guidelinesfor the treatment of patients with major depressive dis-order. Arlington, VA: American Psychiatric Associa-tion, 2010.

111. University of Michigan Health System. Guidelines forclinical care: depression. Ann Arbor, MI: University ofMichigan Health System, 2005.

112. American Psychiatric Association. Practice guidelinesfor the treatment of patients with major depressive dis-order (revision). Am J Psychiatry 2000;157(Suppl. 4):1–45.

113. Wu C-H, Farley JF, Gaynes BN. The associationbetween antidepressant dosage titration and medicationadherence among patients with depression. DepressAnxiety 2012;29:506–514.

114. Berra C, Torta R. Therapeutic rationale of antidepres-sant use in the elderly. Arch Gerontol Geriatr2007;44:83–90.

115. Amodeo L, Castelli L, Leombruni P, CiprianiD, BiancofioreA, Torta R. Slow versus standard up-titration of paroxe-tine for the treatment of depression in cancer patients: apilot study. Support Care Cancer 2012;20:375–384.

116. de la Torre BR, Dreher J, Malevany L et al. Serum lev-els and cardiovascular effects of tricyclic antidepressantsand selective serotonin reuptake inhibitors in depressedpatients. Ther Drug Monit 2001;23:435–440.

117. Whitrayer VG, Dunner DL, Kornstein SG et al. A com-parison of initial duloxetine dosing strategies in patientswith major depressive disorder. J Clin Psychiatry2007;68:1921–1930.

118. Sj€oholm L, Lavebratt C, Forsell Y. A multifactorialdevelopmental model for the etiology of major depres-sion in a population-based sample. J Affect Disord2009;113:66–76.

119. Wade TD, Kendler KS. The relationship between socialsupport and major depression: cross-sectional, longitu-dinal, and genetic perspectives. J Nerv Ment Dis2000;188:251–258.

120. Katzman MA, Tricco AC, McIntosh D et al. Paroxetineversus placebo and other agents for depressive disorders:a systematic review and meta-analysis. J Clin Psychiatry2007;68:1845–1859.

121. Stark P, Hardison CD. A review of multicenter con-trolled studies of fluoxetine vs. imipramine and placeboin outpatients with major depressive disorder. J ClinPsychiatry 1985;46:53–58.

122. Feighner JP, Boyer WF. Paroxetine in the treatment ofdepression: a comparison with imipramine and placebo.Acta Psychiatr Scand 1989;80(Suppl. 350):125–129.

123. Lydiard RB, Laird LK, Morton WA Jr et al. Fluvox-amine, imipramine, and placebo in the treatment ofdepressed outpatients: effects on depression. Psycho-pharmacol Bull 1989;25:68–70.

124. Reimherr FW, Chouinard G, Cohn CK et al. Antidepres-sant efficacy of sertraline: a double-blind, placebo- andamitriptyline-controlled, multicenter comparison studyin outpatients with major depression. J Clin Psychiatry1990;51(Suppl B):18–27.

125. Doogan DP, Langdon CJ. A double-blind, placebo-con-trolled comparison of sertraline and dothiepin in thetreatment of major depression in general practice. IntClin Psychopharmacol 1994;9:95–100.

126. Muijen M, Roy D, Silverstone T, Mehmet A, Christie M.A comparative clinical trial of fluoxetine, mianserin andplacebo in depressed outpatients. Acta Psychiatr Scand1988;78:384–390.

127. Kiev A. A double-blind, placebo-controlled study of par-oxetine in depressed outpatients. J Clin Psychiatry1992;53(Suppl.):27–29.

128. Heiligenstein JH, Tollefson GD, Faries DE. A double-blind trial of fluoxetine, 20 mg, and placebo inout-patients with DSM-III-R major depression andmelancholia. Int Clin Psychopharmacol 1993;8:247–251.

22

Malhi et al.

Page 18: Clinical overview Pharmacological management of unipolar ...dro.deakin.edu.au/...pharmacologicalmanagement... · Pharmacological management of unipolar depression Malhi GS, Hitching

129. Versiani MMD, Amin MMF, Chouinard GMFF.Double-blind, placebo-controlled study with reboxetinein inpatients with severe major depressive disorder. JClin Psychopharmacol 2000;20:28–34.

130. Andreoli V, Caillard V, Deo RS, Rybakowski JK, Versi-ani M. Reboxetine, a new noradrenaline selective antide-pressant, is at least as effective as fluoxetine in thetreatment of depression. J Clin Psychopharmacol2002;22:393–399.

131. Papakostas GI, Nelson JC, Kasper S, Moller HJ.A meta-analysis of clinical trials comparing reboxetine,a norepinephrine reuptake inhibitor, with selective sero-tonin reuptake inhibitors for the treatment of majordepressive disorder. Eur Neuropsychopharmacol2008;18:122–127.

132. Papakostas GI, Thase ME, Fava M, Nelson JC, SheltonRC. Are antidepressant drugs that combine serotonergic

and noradrenergic mechanisms of action more effectivethan the selective serotonin reuptake inhibitors intreating major depressive disorder? A meta-analysis ofstudies of newer agents. Biol Psychiatry 2007;62:1217–1227.

133. RickelsK, Schweizer E, Clary C, Fox I,Weise C. Nefazo-done and imipramine in major depression: a placebo-controlled trial. Br J Psychiatry 1994;164:802–805.

134. Fabre L, Birkhimer LJ, Zaborny BA, Wong LF, Kapik

BM. Fluvoxamine versus imipramine and placebo: adouble-blind comparison in depressed patients. Int ClinPsychopharmacol 1996;11:119–127.

135. Cohn CK, Robinson DS, Roberts DL, Schwiderski UE,O’Brien K, Ieni JR. Responders to antidepressant drugtreatment: a study comparing nefazodone, imipramine,and placebo in patients with major depression. J ClinPsychiatry 1996;57(Suppl. 2):15–18.

23

Pharmacological management of depression