Long-Term Prophylaxis in Bipolar Disorder

CNS Drugs 2006; 20 (4): 303-310THERAPY IN PRACTICE 1172-7047/06/0004-0303/$39.95/0

© 2006 Adis Data Information BV. All rights reserved.

Long-Term Prophylaxis inBipolar DisorderMatthew J. Taylor and Guy M. Goodwin

Department of Psychiatry, University of Oxford, Warneford Hospital, Oxford, UK

ContentsAbstract . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3031. When to Consider Prophylaxis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3042. Choice and Use of Medication . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 304

2.1 Lithium . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3042.2 Valproate . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3052.3 Lamotrigine . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3052.4 Olanzapine . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3052.5 Carbamazepine . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3062.6 Other Agents . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3062.7 Drug Combinations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3072.8 Supplementary Medication . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 307

3. Enhanced Clinical Care . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3074. Discontinuation of Prophylaxis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3075. Special Patient Populations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 307

5.1 Elderly . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3075.2 Pregnant Women . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3085.3 Lactating Women . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 308

6. Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 309

Bipolar disorder is a major cause of disability, and the prevention of relapse isAbstracta key management goal. Pharmacological interventions, effectively deliveredthrough enhanced clinical care, are central to long-term management.

This article summarises the available evidence for a range of pharmacologicaloptions, and provides guidance on common issues in clinical management in linewith current practice guidelines. The use of medications for long-term prophylaxisshould be considered in all patients meeting criteria for bipolar I disorder.Increasing high-quality evidence from randomised trials informs managementdecisions relating to both novel agents, such as lamotrigine and olanzapine, andlonger-established therapies, such as lithium and valproate, in monotherapy.Medications taken long-term in bipolar disorder differ in the extent to which theyprotect against manic and depressive relapse. Consequently, the emerging chal-lenge is to understand how combination treatments can enhance efficacy andeffectiveness based on data from controlled trials rather than random polypharma-cy.

Clinical care can be enhanced with effective education about the illness, andthe use of strategies to improve treatment adherence and the recognition and

304 Taylor & Goodwin

management of stressors or prodromal symptoms. Where available, a range ofspecific psychological interventions can be effective as an adjunct to medication.

When discontinuation of prophylaxis is necessary, gradual tapering of doseover weeks or months is recommended.

Bipolar disorder is a major worldwide cause of substantially beyond symptomatic recovery, and ob-disability.[1] The use of preventative strategies to servational data suggest that prevention of earlyreduce rates of relapse is a key component of care, relapse could be associated with a more benignand the mainstay of prophylaxis remains the course of illness.[7] As noted in section 2.1, there iseffective use of pharmacological agents. These evidence that long-term use of lithium, in particular,medicines, taken over the long term to prevent re- is associated with decreased suicide and all-causelapse, are often known as ‘mood stabilisers’, but this mortality in bipolar disorder.term is not without its difficulties, not least that its

2. Choice and Use of Medicationdefinition is variable and therefore it is used incon-sistently.[2]

The choice of which medication to use in theAlthough prophylaxis in bipolar disorder has long-term management of bipolar disorder should

been relatively neglected in the past, there is now a be decided by doctor and patient on the basis ofgrowing evidence base from randomised trials for factors including evidence of efficacy, adverse ef-the efficacy both of novel agents and longer-estab- fect profile and the pattern of illness in the past.lished therapies. One key observation from these Appropriate clinical monitoring for evidence of ad-studies is that agents appear to differ in the extent to verse effects is indicated with all agents. This maywhich they prevent depressive or manic episodes, include simple clinical measures such as monitoringand an understanding of these differences may be of weight, which may be particularly relevant givenimportant in tailoring treatment to individual pa- recent reports of high rates of metabolic syndrometients. in patients with bipolar disorder.[8] Common special

In this article, we review the evidence relating to investigations for a range of medications are sum-a range of medicines in the prevention of recurrent marised in table I; these may need to be supplement-depressive or manic episodes, and summarise issues ed depending on patient characteristics and as clini-of clinical management in line with recent guide- cally indicated.lines.[3,4] Key references to further information areincluded as appropriate. 2.1 Lithium

Lithium has a long history of use in the treatment1. When to Consider Prophylaxisof bipolar disorder, and remains the benchmarkagainst which other agents should be measured.The use of long-term prophylactic medicationPoor medication adherence is a contraindication tocan be considered in patients who have had even athe use of lithium, in view of the increased risk ofsingle episode of mania (i.e. patients meeting crite-relapse on discontinuation.[9] Where used, lithiumria for bipolar I disorder).[3,5,6] Traditionally, Euro-should be given at the highest dose that producespean guidelines have recommended long-term pro-minimal adverse effects, although blood concentra-phylaxis only for patients experiencing multipletions <0.5 mmol/L are probably too low.[3]episodes over a short time interval;[4] changing per-

spectives reflect improved understanding of both the The use of lithium as an active comparator in alikelihood of recurrent episodes and their potential number of recent randomised trials has substantiallyfor harm. In bipolar disorder, functional impairment increased the available data on its efficacy as aassociated with an episode of illness often persists prophylactic agent. A recent meta-analysis was able

© 2006 Adis Data Information BV. All rights reserved. CNS Drugs 2006; 20 (4)

Long-Term Prophylaxis in Bipolar Disorder 305

to include data from five randomised trials involving 2.3 Lamotrigine1237 participants, of whom 770 had received lithi-um or placebo. It found that the use of lithium Results from two 18-month studies have beenreduces the risk of both depressive and manic re- reported in which 638 participants were randomisedlapse (relative risks 0.72 and 0.62, respectively); to lamotrigine, lithium or placebo.[16] Both lamo-thus the effect appeared numerically, if not statisti- trigine and lithium were found to be superior tocally, most robust for avoiding further mania.[10] In placebo in delaying the time to treatment for a newaddition, there is evidence from both observational mood episode (median survival 86 days vs 184 daysstudies and randomised trials that the use of lithium [lithium] and 197 days [lamotrigine]). Interestingly,reduces rates of suicide, deliberate self-harm and the effects of lamotrigine appeared superior in theall-cause mortality.[11,12] prevention of depressive relapse, while lithium was

superior for preventing mania.

2.2 Valproate2.4 Olanzapine

The strongest data on the use of valproate (as Olanzapine has been studied both as a monother-divalproex [valproate semisodium]) to prevent re- apy and as augmentation of other long-term treat-lapse in bipolar disorder comes from two randomis- ment. Results from four randomised trials of 1142ed trials involving 623 participants. One study com- participants have been reported. As noted in sectionpared valproate with lithium or placebo.[13] In this 2.2, in one study comparing olanzapine with val-study, on the primary outcome measure of time to proate continuation after the treatment of mania,any mood episode, there was no statistically signifi- rates of relapse did not differ between the groups.[15]

cant difference between the groups. However, pa- In a further study, olanzapine was as effective astients receiving valproate were less likely than those lithium in preventing symptomatic recurrence,[17]

receiving placebo to withdraw from the study be- and a secondary analysis found olanzapine to because of relapse (relative risk 0.63), and on this superior to lithium in preventing symptomatic ma-secondary measure there was no statistically signifi- nic or mixed recurrence (odds ratio for mania, 1.9).cant difference between the valproate and lithium Another study has found that olanzapine is superiorgroups.[14] A second randomised trial compared val- to placebo in preventing both depressive and manicproate with olanzapine continuation after the treat- relapse, although the median time to relapse in thement of mania;[15] rates of relapse did not differ group abruptly withdrawn from olanzapine to place-significantly between the two treatment groups. bo was only 22 days, which suggests an important

Table I. Commonly indicated investigations for medications used in the long-term management of bipolar disorder

Agent Initial investigations Ongoing monitoring

Lithium Renal and thyroid function Lithium concentrations

Full blood count Renal and thyroid function

Pregnancy test if indicated

Valproate Full blood count Full blood count

Renal and liver function Liver function


Lamotrigine No routine investigations No routine investigations

Olanzapine Glucose and lipid monitoring Glucose and lipid monitoring

Carbamazepine Full blood count Full blood count

Renal and liver function Liver function




Table II. Recent recommendations on the use of combinations of medications in maintenance therapy of bipolar disorder

Guideline Year Recommendation Reference

APA 2002 Patients who continue to experience subthreshold symptoms or breakthrough mood 5episodes may require the addition of another maintenance medication, an atypicalantipsychotic or an antidepressant. There are currently insufficient data to support onecombination over another

BAP 2003 If the patient fails to respond to monotherapy and continues to experience subthreshold 3symptoms or relapses, consider long-term combination treatment. Where the burden ofdisease is mania, it may be logical to combine predominantly anti-manic agents (e.g.lithium, valproate, an antipsychotic). Where the burden is depressive, lamotrigine or anantidepressant may be more appropriate in combination with an anti-manic, long-termagent

CANMAT 2005 There is little evidence to recommend one combination over another. Combinations that 6have demonstrated some efficacy include lithium plus divalproex (valproatesemisodium) or carbamazepine, as well as lithium or divalproex plus olanzapine orrisperidone. No data are available on lithium plus lamotrigine, but this combination isrecommended based on the proven prophylactic efficacy of the two medications asmonotherapies

APA = American Psychiatric Association; BAP = British Association for Psychopharmacology; CANMAT = Canadian Network for Mood andAnxiety Treatments.

withdrawal effect.[18] Like lithium, olanzapine 2.6 Other Agentsshould probably be tapered out over 2–4 weeks.

There are a number of other medicines that couldOlanzapine augmentation of lithium or valproatebe considered for relapse prevention in bipolar dis-was compared with the addition of placebo in oneorder; however, the evidence for their use is not asrandomised trial.[19] There was no statistically sig-clear as for those drugs mentioned in sectionsnificant difference between groups on the primary2.1–2.5.outcome measure of time to syndromic relapse, but

Long-term use of antidepressants in bipolar dis-in a secondary analysis, olanzapine augmentationorder is common in clinical practice; however, aincreased the time to symptomatic relapse (mediansystematic review of the very limited data currentlysurvival 163 days for combination therapy vs 42available has suggested that TCAs are not estab-days for monotherapy).lished as being better than placebo in preventingrelapse.[23] Considering newer antidepressants, such2.5 Carbamazepineas SSRIs, a recent observational study found thatcontinuation of these antidepressants after remissionCarbamazepine was the first antiepileptic drug towas associated with reduced rates of depressivebe used in bipolar disorder. A meta-analysis of earlyrelapse over a 1-year period.[24] However, the vul-randomised trials involving participants with bothnerability of such studies to confounding variablesunipolar and bipolar disorder found no significantand the absence of high-quality randomised trialdifference between lithium and carbamazepine inevidence means that the role of the long-term use ofpreventing relapse.[20] However, the evidence fromantidepressants remains controversial.[25] Currently,two recent randomised trials comparing carba-long-term antidepressant monotherapy for bipolarmazepine with lithium (238 participants) in bipolardisorder would not be generally recommended;disorder has indicated a substantial benefit for lithi-however, antidepressant use in combination withum over carbamazepine in preventing relapse.[21,22]

effective anti-manic agents may be appropriate forIn one study, relapse rates for the two groups weresome individuals (see section 2.7 and table II).27% for lithium and 42% for carbamazepine,[22] and

in the other, also randomised but not of double-blind Another question pertinent to clinical practice isdesign, the rates were 28% and 47%, respective- whether other antipsychotics share the efficacy ofly.[21] olanzapine in relapse prevention for bipolar disor-



der. Preliminary reports indicate that the atypical management. An effective therapeutic alliance be-antipsychotic aripiprazole is more effective than tween the doctor and patient is critical, particularlyplacebo in preventing manic but not depressive re- if management is to be effective in the longer term.lapse.[26] The recognition and management of co-morbid dis-

orders is also important, for example, substance use2.7 Drug Combinations disorders, which are not uncommon in the bipolar

disorder population and associated with an adverseWhere the long-term use of a single agent doesimpact on both the course and management of thenot prove satisfactory, the use of combinations ofdisorder.[30]

drugs is becoming increasingly widespread. ThisA recent systematic review found that a range ofhas previously been guided predominantly by

psychological interventions reduced relapse ratesclinical experience because of a lack of clear high-when used as an adjunct to medication.[31] The vari-quality evidence[27] (see table II). However, thisous interventions studied (interpersonal socialsituation is now changing as well designed,rhythms therapy, cognitive therapy, family focussedrandomised trials addressing this question havetherapy and psychoeducation) share common fea-started. For example, in addition to the study oftures, addressing at least one of the components ofolanzapine augmentation described in sectiondisease understanding, treatment adherence, recog-2.4,[19] one large study investigating the combinationnition and management of stressors or prodromalof lithium and valproate is in progress, with >230symptoms, and stabilising social rhythms. Even inpatients currently randomised.[28]

settings where the specific interventions are notavailable, these components provide indications of2.8 Supplementary Medicationways in which ‘routine care’ can be optimised with

Continuous, rather than intermittent, use of oral education about the nature of bipolar disorder and,medication to prevent relapse is currently advised. in particular, the seriousness of any recurrence, andThis can be supplemented as necessary in response practical advice on what to do should emergentto acute stressors, early symptoms of relapse or symptoms be noticed.anxiety. One useful strategy is for patients to devel-op a self-medication plan during appropriate discus- 4. Discontinuation of Prophylaxission with their doctor while well. This may involve

Discontinuation of medication that has been usedretaining a small supply of benzodiazepines or an-long term in patients with bipolar disorder shouldtipsychotics for use as required, or a simple agree-not be undertaken lightly. Even for patients whoment to increase the dosage of their long-term medi-have remained illness free for long periods whilecation when indicated. This approach may increaseusing medication, the risk of relapse remains high.adherence through a feeling of greater control overLithium in particular is associated with an increasedtreatment plans, and will also allow early treatmentrisk of manic relapse on rapid discontinuation.[9]

in the acute situation where there might be unneces-When discontinuation is planned, it should be per-sary delay in seeing a psychiatrist.[3]

formed by gradual tapering of the dosage over sev-eral weeks or even months.[32]3. Enhanced Clinical Care

Complex chronic conditions such as bipolar dis- 5. Special Patient Populationsorder require expert management, and doctors needto take responsibility for aspects such as accurate

5.1 Elderlydiagnosis (which can be facilitated through the rou-tine use of standard criteria such as DSM-IV[29]), an In elderly patients who require prophylaxis forappropriate physical examination and investiga- bipolar disorder, the same general principles oftions, and communication of the plan of medical treatment apply as in younger adults. However,



management can be complicated by the increased past, and is around 0.05–0.1%.[39] Women receivinglikelihood of medical co-morbidity and the potential lithium, valproate or carbamazepine in the first tri-for drug interactions. There is also an increased risk mester should be advised of the prenatal diagnosisof adverse effects because of factors such as in- of these abnormalities by α-fetoprotein screeningcreased end-organ sensitivity and impairments of and high-resolution ultrasound at 16–18 weeks’ ges-renal and cardiovascular function. This is perhaps tation.particularly the case for the use of lithium, where Some of the above-mentioned risks may in prac-both lower doses and lower target serum concentra- tice be unavoidable, as rates of unplanned pregnan-tions may be required.[33] cy are high in the general population and are likely

to be even higher in patients with mania. Additional-5.2 Pregnant Women ly, the effects on organ formation are seen within the

early stages of pregnancy before it may have beenThe long-term use of medication for bipolar dis- confirmed. All women who have bipolar disorder

order in pregnant women requires a balancing of and are of childbearing potential should be advisedrisks and benefits. The benefits of remaining re- of the importance of effective contraception, and oflapse-free include enabling normal bonding with the planning any pregnancy in consultation with theirchild and facilitating neonatal development, and psychiatrist. A range of options for treatment may beavoiding the ill-effects of relapse that include an considered on the basis of past history and patientincreased likelihood of co-morbid substance misuse. and clinician preferences. These include continuingAgainst this should be balanced the risks of treat- current medications unchanged throughout pregnan-ment, including teratogenesis, toxicity and with- cy, changing to medications associated with lowerdrawal. fetal risks prior to conception, and withdrawal of

Lithium, valproate and carbamazepine are all some or all medications around conception and rein-associated with an increased rate of major congeni- troducing them either after the first trimester or aftertal malformations, although the vast majority of delivery.patients using these medications still deliver a Prescribing may need to be altered during preg-healthy child. There is a relative absence of data for nancy. The use of slow-release preparations in di-newer agents such as the atypical antipsychotics, vided doses may reduce the peak drug concentra-rather than clear evidence as yet of greater safety. tions to which the foetus is exposed. Increased glo-

General population rates of major congenital merular filtration may necessitate dosage increasesmalformations vary from 2% to 4%, increasing with in pregnant women to maintain stable serum con-maternal age.[34] Valproate and carbamazepine are centrations, while dosages will need to return toassociated with increased rates of such malforma- normal after delivery.tions, 11% and 6%, respectively,[35,36] and are spe- Women with bipolar disorder are at a particularlycifically associated with neural tube defects (1–2%) high risk of puerperal psychosis in the post-partumand foetal hydantoin syndrome.[37] These effects are period, which requires prompt recognition and treat-dose related for valproate, and are more likely ment, but there is a lack of high-quality evidence towith coadministration of multiple antiepileptic guide specific prophylactic medication choices.[40,41]

agents.[38] While peri-conceptual folate supplemen- Further discussion of issues related to the manage-tation is recommended for all pregnant women, it ment of bipolar disorder in pregnancy and the puer-remains unclear whether it specifically reduces the perium can be found elsewhere.[42]

rates of neural tube defects associated with theseagents. Lithium use is associated with rates of major 5.3 Lactating Womencongenital malformation of 4–12%, but the risk ofcardiovascular abnormalities, including Ebstein’s All maternally ingested medications enter breastanomaly, appears to have been overestimated in the milk, although the extent to which this happens



5. American Psychiatric Association. Practice guideline for thevaries greatly between individuals and medications,treatment of patients with bipolar disorder. 2nd ed. Washing-

and technical issues often limit the extent to which ton, DC: American Psychiatric Association, 20026. Yatham LN, Kennedy SH, Donovan C, et al. Canadian Networkdata can be directly compared between studies.[43]

for Mood and Anxiety Treatments (CANMAT) guidelines forThe adverse effects this may cause remain uncer-the management of patients with bipolar disorder: consensus

tain, with evidence largely limited to sporadic case and controversies. Bipolar Disord 2005; 7 Suppl. 3: 5-697. Kessing LV, Hansen MG, Andersen PK, et al. The predictivereports. The benefits of breastfeeding need to be

effect of episodes on the risk of recurrence in depressive andweighed against the risks, and vigilance for ill ef-bipolar disorders: a life-long perspective. Acta Psychiatr Scand

fects in the infant is necessary. The narrow therapeu- 2004; 109 (5): 339-448. Fagiolini A, Frank E, Scott JA, et al. Metabolic syndrome intic index of lithium and high infant serum concentra-

bipolar disorder: findings from the Bipolar Disorder Center fortions reported in some cases mean the use of the Pennsylvanians. Bipolar Disord 2005; 7 (5): 424-30drug is a relative contraindication to breastfeed- 9. Goodwin GM. Recurrence of mania after lithium withdrawal:

implications for the use of lithium in the treatment of bipolaring.[44]

affective disorder. Br J Psychiatry 1994; 164 (2): 149-5210. Geddes JR, Burgess S, Hawton K, et al. Long-term lithium

6. Conclusions therapy for bipolar disorder: systematic review and meta-analysis of randomized controlled trials. Am J Psychiatry2004; 161 (2): 217-22Bipolar disorder requires expert management

11. Cipriani A, Wilder H, Hawton K, et al. Lithium in the preven-that can increasingly be guided by high-quality evi- tion of suicidal behaviour and all-cause mortality in patientsdence from randomised trials. The use of medica- with mood disorders: a systematic review of randomised trials.

Am J Psychiatry 2005; 162 (10): 1805-19tions for long-term prophylaxis should be consid-12. Tondo L, Hennen J, Baldessarini RJ. Lower suicide risk withered in all patients meeting criteria for bipolar I long-term lithium treatment in major affective illness: a meta-

analysis. Acta Psychiatr Scand 2001; 104 (3): 163-72disorder. Clinical care can be enhanced by effective13. Bowden CL, Calabrese JR, McElroy SL, et al. A randomized,education about the illness and the use of cognitive

placebo-controlled 12-month trial of divalproex and lithium intherapies to improve treatment adherence, and the treatment of outpatients with bipolar I disorder: Divalproex

Maintenance Study Group. Arch Gen Psychiatry 2000; 57 (5):recognition and management of stressors or prodro-481-9mal symptoms.

14. Macritchie KA, Geddes JR, Scott J, et al. Valproic acid, val-proate and divalproex in the maintenance treatment of bipolardisorder. Cochrane Database Syst Rev 2001; (3): CD003196Acknowledgements

15. Tohen M, Ketter TA, Zarate CA, et al. Olanzapine versusdivalproex sodium for the treatment of acute mania and main-Prof. G.M. Goodwin is in receipt of grant support fromtenance of remission: a 47-week study. Am J Psychiatry 2003;Sanofi-Aventis for the randomised trial BALANCE, has fre-160 (7): 1263-71

quently spoken in industry-supported symposia and has ad-16. Goodwin GM, Bowden CL, Calabrese JR, et al. A pooled

vised most major pharmaceutical companies with an interest analysis of 2 placebo-controlled 18-month trials of lamotriginein this therapeutic area in the last 5 years. Dr M.J. Taylor has and lithium maintenance in bipolar I disorder. J Clin Psychia-received an honorarium from AstraZeneca for workshop fa- try 2004; 65 (3): 432-41cilitation. Dr Taylor is funded by a Wellcome Trust Research 17. Tohen M, Greil W, Calabrese JR, et al. Olanzapine versus

lithium in the maintenance treatment of bipolar disorder: aTraining Fellowship. No other external funding was used to12-month, randomized, double-blind, controlled clinical trial.assist in the preparation of this manuscript.Am J Psychiatry 2005; 162 (7): 1281-90

18. Tohen M, Calabrese JR, Sachs GS, et al. Randomized, placebo-controlled trial of olanzapine as maintenance therapy in pa-Referencestients with bipolar I disorder responding to acute treatment1. Murray CJ, Lopez AD. Global mortality, disability, and thewith olanzapine. Am J Psychiatry 2006; 163 (2): 247-56contribution of risk factors: Global Burden of Disease Study.

19. Tohen M, Chengappa KN, Suppes T, et al. Relapse preventionLancet 1997; 349 (9063): 1436-42in bipolar I disorder: 18-month comparison of olanzapine plus2. Malhi GS, Mitchell PB, Berk M, et al. Mood stabilizers: a labilemood stabiliser vs mood stabiliser alone. Br J Psychiatry 2004;label. Acta Psychiatr Scand Suppl 2005; (426): 5-6184: 337-453. Goodwin GM, Young AH. The British Association for

20. Davis JM, Janicak PG, Hogan DM. Mood stabilizers in thePsychopharmacology guidelines for treatment of bipolar disor-prevention of recurrent affective disorders: a meta-analysis.der: a summary. J Psychopharmacol 2003; 17 (4 Suppl.): 3-6Acta Psychiatr Scand 1999; 100 (6): 406-174. Grunze H, Kasper S, Goodwin G, et al. The World Federation of

Societies of Biological Psychiatry (WFSBP) guidelines for the 21. Greil W, Kleindienst N. The comparative prophylactic efficacybiological treatment of bipolar disorders, part III: maintenance of lithium and carbamazepine in patients with bipolar I disor-treatment. World J Biol Psychiatry 2004; 5 (3): 120-35 der. Int Clin Psychopharmacol 1999; 14 (5): 277-81



22. Hartong EG, Moleman P, Hoogduin CA, et al. Prophylactic 34. Brockington IF. Motherhood and mental health. Oxford: Oxfordefficacy of lithium versus carbamazepine in treatment-naive University Press, 1996bipolar patients. J Clin Psychiatry 2003; 64 (2): 144-51

35. Rosa FW. Spina bifida in infants of women treated with23. Ghaemi SN, Lenox MS, Baldessarini RJ. Effectiveness and carbamazepine during pregnancy. N Engl J Med 1991; 324

safety of long-term antidepressant treatment in bipolar disor- (10): 674-7der. J Clin Psychiatry 2001; 62 (7): 565-9

36. Kaneko S, Battino D, Andermann E, et al. Congenital malfor-24. Altshuler L, Suppes T, Black D, et al. Impact of antidepressant mations due to antiepileptic drugs. Epilepsy Res 1999; 33

discontinuation after acute bipolar depression remission on (2–3): 145-58rates of depressive relapse at 1-year follow-up. Am J Psychia-

37. Hanson JW, Smith DW. The fetal hydantoin syndrome. Jtry 2003; 160 (7): 1252-62Pediatr 1975; 87 (2): 285-90

25. Soldani F, Ghaemi SN, Tondo L, et al. Relapse after antidepres-sant discontinuation. Am J Psychiatry 2004; 161 (7): 1312-3 38. Samren EB, van Duijn CM, Christiaens GC, et al. Antiepileptic[author reply 1313] drug regimens and major congenital abnormalities in the off-

spring. Ann Neurol 1999; 46 (5): 739-4626. McQuade RD, Sanchez R, Marcus R. Aripiprazole for relapseprevention in bipolar disorder: a 26-week placebo controlled 39. Cohen LS, Friedman JM, Jefferson JW, et al. A reevaluation ofstudy [abstract] Int J Neuropsychopharmacol 2004; 7 (Suppl risk of in utero exposure to lithium [published erratum appears1): 161 in JAMA 1994 May 18; 271 (19): 1485]. JAMA 1994; 271 (2):

146-5027. Denicoff KD, Smith-Jackson EE, Disney ER, et al. Comparativeprophylactic efficacy of lithium, carbamazepine, and the com- 40. Brockington IF. Puerperal psychosis. In: Brockington IF, editor.bination in bipolar disorder. J Clin Psychiatry 1997; 58 (11): Motherhood and mental health. Oxford: Oxford University470-8 Press, 1996: 200-84

28. Geddes J, Goodwin G. Bipolar disorder: clinical uncertainty,41. Jones I, Craddock N. Bipolar disorder and childbirth: the impor-

evidence-based medicine and large-scale randomised trials. Brtance of recognising risk. Br J Psychiatry 2005; 186: 453-4J Psychiatry 2001; 178 Suppl. 41: S191-4

42. Yonkers KA, Wisner KL, Stowe Z, et al. Management of29. American Psychiatric Association. Diagnostic and statisticalbipolar disorder during pregnancy and the postpartum period.manual of mental disorders. 4th ed. Washington, DC: Ameri-Am J Psychiatry 2004; 161 (4): 608-20can Psychiatric Association, 1994

43. Yoshida K, Smith B, Kumar R. Psychotropic drugs in mothers’30. Salloum IM, Thase ME. Impact of substance abuse on themilk: a comprehensive review of assay methods, pharmacoki-course and treatment of bipolar disorder. Bipolar Disord 2000;netics and of safety of breast-feeding. J Psychopharmacol2 (3 Pt 2): 269-801999; 13 (1): 64-80

31. Scott J, Gutierrez MJ. The current status of psychological treat-44. Chaudron LH, Jefferson JW. Mood stabilizers duringments in bipolar disorders: a systematic review of relapse

breastfeeding: a review. J Clin Psychiatry 2000; 61 (2): 79-90prevention. Bipolar Disorders 2004; 6 (6): 498-503

32. Suppes T, Baldessarini RJ, Faedda GL, et al. Discontinuation ofmaintenance treatment in bipolar disorder: risks and implica-

Correspondence and offprints: Prof. Guy M. Goodwin, De-tions. Harv Rev Psychiatry 1993; 1 (3): 131-44partment of Psychiatry, University of Oxford, Warneford33. Sproule BA, Hardy BG, Shulman KI. Differential pharmacoki-Hospital, Warneford Lane, Oxford OX3 7JX, UK.netics of lithium in elderly patients. Drugs Aging 2000; 16 (3):

165-77 E-mail: [email protected]


Documents

Long-Term Prophylaxis in Bipolar Disorder