‘THE PRIMARY CARE COMPANION FOR CNS DISORDERS
`
`Prim Care Companion CNS Disord, 2011; 13(6): PCC, 11101174
`doi: 10,4088/PCC, 11°01174
`
`PMCID: PMC3304688
`PMID: 22454806
`
`Olanzapine-Induced Weight Gain in Patients With Bipolar | Disorder: A Meta-Analysis
`
`Mina G. Nashed, BSc, Maria R. Restivo, BSc, and Valerie H. Taylor, MD, PhD, eRcpcu
`MINDS Neuroscience Graduate Program, McMaster University, Hamilton (Mr Nashed and Ms Restivo); and Women's College Hospital, Department of Psychiatry, University of Toronto (Or
`Taylor), Ontario, Canada
`orresponding author.
`Corresponding author:Valerie H. Taylor, MD, PhD, FRCPC, Department of Psychiatry, University of Toronto, Women's College Hospital, 76 Grenville St, Toronto, Ontario, M5S 7B2 Canada
`(valerietaylor@wchospital.ca).
`Received 2011 Feb 25; Accepted 2011 Jun 13.
`Copyright © 2011, Physicians Postgraduate Press, Inc.
`
`Abstract
`
`Go to:
`
`Objective: The weight impact produced bythe atypical antipsychotic olanzapine has been explored in meta-analyses focusing on patients with
`schizophrenia. However, outcomesidentified for schizophrenia patients cannot always be gencralized to patients with bipolar disorder. This study
`aims to quantitatively estimate the impact of olanzapine on the weight ofpatients with bipolar disorder.
`
`Data Sources: EMBASE, Medline, and PsycINFO were searched using the keywords olanzapine AND (bipolar OR acute mania) in conjunction
`with (weight gain OR weight increase) (last search: October 2010, with no restrictions on dates of publication). English language was used as a
`restriction.
`
`Study Selection: The search identified 110 articles for review. The inclusioncriteria for the chosen studies were a diagnosis of bipolar disorder, the
`presence ofan olanzapine monotherapygroup, a comparator placebo or monotherapygroup, and mean weight gain and/or incidences of weight
`gain data. This process identified 13 studies for inclusion.
`
`Data Extraction: The primary outcome measure was the mean weight change between olanzapine monotherapy and comparator monotherapy,
`reported in kilograms. Standard deviation was extracted directly from studies when possible and imputed for 3 studies. The secondary outcome
`measure wasthe reported incidences of = 7%weightgain.
`
`Data Synthesis: The mean difference in weight gain wascalculated for the continuous data of the primary outcome. Olanzapine monotherapy was
`associated with more weight gain when comparedto placebo (mean difference = 2.10 kg; 95%CI, 1.16—3.05; P < .001) and other bipolar
`monotherapy(mean difference = 1.34 kg; 95%CI, 0.95-1.72: P < .001). Oddsratio analysis of the dichotomous secondary outcome also showed
`more weight gain with olanzapine monotherapy compared to placebo (odds ratio [OR] = 10.12; 95%CI, 1.93-53.14; P = 006) and other bipolar
`monotherapy (OR = 2.09; 95% CI, 1.27-3.44: P = .004).
`
`Conclusions: Currentlyavailable data suggest that olanzapine is associated with significant weight gain in bipolarpatients. Issues related to side
`effect profiles and their impact on treatment compliance and physical health outcomes need to be considered whenselecting pharmacotherapy.
`
`
`Bipolar disorder, a chronic mentalillness that impacts 1% of the population, is defined clinically by a wide range of symptoms: a depressed or
`euphoric mood,lack of activity paralleled at times with energized behavior, and a decreased need for sleep and social interaction that can manifest
`as either the desire for completeisolation or extreme extroversion that can become problematic. To further complicate the picture, individuals in
`either the manic or depressed phases of bipolar disorder can experience psychotic symptomsas well.) Asa consequence, the pharmacologic
`managementof bipolar disorder involves a myriad of options from a variety of drug categories; moodstabilizers, antidepressants, and atypical
`antipsychotics are all recommendedasfirst-line agents, either as monotherapyor in combination." Ofthese options, the most recent class of
`medications to becomefirst line for acute and maintenance treatmentof bipolar disorder is the second-generation atypical antipsychotics (SGAs).
`While these agents are heterogeneousin their efficacyand tolerability, studies suggest that SGAs, either alone or in combination with mood
`stabilizers, are currently an efficacious treatment strategyin the management ofboth the depressive and manicstages ofbipolar disorder.2 Added
`benefits in favor of the use of SGAsinclude reduced extrapyramidal side effects and the absence of depressive symptom exacerbation.There are
`concerns associated with the use ofthis medication class, however, and the adverse metabolic profile associated with SGAsneedsto be considered
`when making treatment recommendations.
`In terms ofmarket share, the most commonlyprescribed atypical antipsychotic worldwide is olanzapine. In 2003, olanzapine was approved for
`the treatment ofbipolar depressive episodes in combination with fluoxetine,= and in 2004 it was approved for long-term maintenance treatment of
`bipolar disorder.® Since then, olanzapine has becomethe best-studied SGA in this patient population, but while significant weight gain has been
`consistently reported with the use of olanzapine in the treatment ofbipolar disorder, there has not been a meta-analysis to comprehensively
`investigate the problemin this population. The bulk of work examining the weight gain side effects associated with olanzapine has focused on
`schizophrenia, and a 2009 meta-analysis by Leuchtet alconcludedthat olanzapine was associated with 3.3-kg more weight gain when compared
`with haloperidol monotherapy (95%CI, 2.2-4.4, P < .001) in 9 studies on patients with schizophrenia. In 2 other meta-analyses, olanzapine was
`shown to cause more weight gain than any other SGA, with the exception ofclozapine, in patients in schizophrenia.®2
`
`1
`
`Exhibit 2047
`Slayback v. Sumitomo
`IPR2020-01053
`
`Exhibit 2047
`Slayback v. Sumitomo
`IPR2020-01053
`
`

`

`Results from schizophrenia studies are not always generalizable to other patient populations; therefore, a meta-analysis on the weight gain effects
`of olanzapine on patients with bipolar disorder is warranted. The aim of this study was to compare weight gain outcomesofolanzapine
`monotherapyto placebo and other monotherapiesin patients with bipolar disorder.
`
`Clinical Points
`
`«Olanzapine monotherapyis associated with significantly more weight gain than placebo and other bipolar disorder medications that are
`knownto cause moderate weight gain.
`«Weight gain mayexacerbate other health risks associated with bipolar disorder, such as compromised neurocognitive function.
`*Clinician awareness regarding the adverse metabolic side effects of antipsychotics. such as olanzapine, will ensure that patients are able
`to safely choose the best medication to manage their complicatedillness and improve medication compliance.
`
`
`
`METHOD Go to:
`
`Database Search
`
`The OVIDsearch engine was used to perform a combined search of 3 databases: EMBASE, MEDLINE,and PsycINFO.Thelast search was
`conducted in October of 2010, and there were no restrictions on date of publication. English language was used as a restriction. Abstracts,titles,
`and indexed termsofstudies were searched using the keywords olanzapine AND (bipolar OR acute mania) in conjunction with (weight gain OR
`weight increase). After duplicates and articles with no abstracts were filtered out, 784 results remained (Figure|). Studies that did not investigate
`weight gain with olanzapine monotherapy were excluded. This process identified 110 articles for full-text investigation. Of these, studies that did
`not include at least 1 comparator to olanzapine monotherapy were excluded, along with studies that did not investigate bipolar patients. Studies on
`adolescent bipolar patients were not included in the analysis due to limitations in comparing weight gain between adults and adolescents. Open-
`label and naturalistic studies were included since physical measures, such as weight gain, are not susceptible to placebo effect.
`
`Further screening of the 110 retrieved articles and their references identified 12 double-blind, randomized, controlled studies and 1 naturalistic
`observational study to be included in this analysis. Fourof the total 13 studies were placebo controlled, and the remaining tested olanzapine
`monotherapyagainst an alternative monotherapy. The coauthors of the present analysis (M.G.N. and M.R.R.) performedthe search and extracted
`data from the studies independently, and disagreements were discussed until a consensus wasreached. Theinclusioncriteria for the final studies
`were a diagnosis of bipolar disorder, the presence of an olanzapine monotherapy group, a comparator placebo or monotherapy group, and mean
`weight gain and/or incidences of weight gain data.
`
`It is worth noting that 2 studies that initially passed all phases of screening were later excluded. Thefirst is a maintenance study by Tohenet all
`in which subjects were randomizedto receive either olanzapineorlithiumafter 6 to 12 weeks of open-label cotherapy, Consequently, the subjects
`were not olanzapine naive at randomization, having gained a mean of2.7 kg during cotherapy. The second study was also a maintenance study by
`Tohenet al! and, similarly, this study randomized subjects to receive either olanzapine or placebo after 6 to 12 weeks ofopen-label olanzapine
`treatment. Thus, the placebo group had gained weight due to olanzapine treatment prior to randomization. Both studies showsignificantly more
`weight gain with olanzapine monotherapyat endpoint but were excluded from the present analysis due to their experimental design.
`
`Theclinical trials registry clinicaltrials,20v was searched for unpublished results using the keywords olanzapine AND bipolar AND weight. This
`search yielded 30 trials, of which 2 were completed with results. Of these, 1 compared placebo to olanzapine combination therapy with divalproex
`and wastherefore excluded onthe basis ofnot having an olanzapine monotherapy group. The other study included schizophrenia and
`schizoaffective subjects, along with bipolar subjects, and reported the weight gainresults for all subjects combined. The investigators ofthistrial
`were contacted via e-mail, requesting the separate weight gain results of the bipolar subject, but no response wasobtained.
`
`Outcome Parameters and Data Extraction
`
`The primary outcomeofinterest for this analysis was the mean weight gain in the olanzapine monotherapy and comparator monotherapy groups.
`For the primary outcome, we needed to extract mean weight gain, standard deviation (SD), and sample size from each study. When reported, the
`last observation carried forward (LOCF)wasused as the samplesize in the analysis. For studies that did not report a LOCF, the randomized
`sample size was used. For 10 of the 13 articles, the SDs were extracted directly from the article, and for the remaining 3 studies that did not
`disclose SDs. the SDs were imputed using the pooled SDs fromall the other articles matched appropriatelyfor intervention group. This method
`was used because borrowing SDs fromother studies to impute data has been empirically shown to be an appropriate remedyfor missing SDsin
`meta-analyses22
`
`The secondary outcome ofinterest was the incidence of weight gain, that is, the numberofpatients in each group whogained weight during the
`intervention. Clinically significant weight gain was defined as 7%or moreofinitial body weight. While it would have been preferable to use body
`mass index or waist circumference as a surrogate marker of weight gain, only 4 of the 13 studies included in this meta-analysis had information on
`body massindex and of these 4 studies, 3 reported baseline body mass index but not change over time. No data on changein waist circumference
`were available.
`
`Analytic Methods
`
`The meta-analysis was conducted using the software Review Manager Version 5.0 (Clicktime.com, Inc, San Francisco, California), with statistical
`significance set at P < 05. As the primary outcome examined continuousdata, the meandifference in weight gain (change in kilograms) was
`calculated with a 95%confidence interval (CI). For the dichotomousdata of the secondary measure incidence of weight gain, an oddsratio (OR)
`analysis was performed. This method allowsfor the inclusion of more types ofstudies andis less proneto outliers than relative risk analysis.2:4
`
`

`

`A random-effects model was used in both primary and secondary outcome analyses, with heterogeneity amongstudiesinvestigated using both C
`(P <.01) and F*tests.
`
`
`
`RESULTS Goto:
`
`Study Characteristics
`Table 1 summarizesthe characteristics of the 13 studies included in the meta-analysis.4222 The duration for most ofthe studies was between 3
`and 12 weeks, with 3 studies each investigating 3-week and 4-week intervals,15-1618.22,24.25 9 studies investigating a 6-week period, 23.26 1 study
`investigating an 8-week follow-up and 4 studies investigating weight gain over 12 weeks.122022 The remaining study looked atside effects
`.
`~
`2
`over a longer maintenance phase of 47 weeks.2+
`
`With respect to study type, 4 of the 13 studiesin the present analysis were placebo controlled,t= 18 While the remaining studies compared
`olanzapine monotherapy to various comparators, namely haloperidol12 divalproex/valproate,1®:22-23 lithium224 risperidone,2>2° and
`asenapine.2_ Tohen et al4® conducted a 3-branch examination on olanzapine monotherapyversus placebo versus divalproex monotherapy, while
`Kim et al? conducted a 3-branch monotherapy examination of olanzapine versuslithium versus valproate, so data from both studies were
`included in 2 comparisons. The study comparing olanzapine monotherapy with haloperidol monotherapy! was grouped with the placebo-
`controlled studies, as haloperidol has not been associated with significant weight gain and for the purpose of the present analysis behaves as a
`placebo.2822 Thus, “olanzapine monotherapyversus placebo or haloperidol” constituted the first comparison. Studies with other comparators
`were grouped together under the second comparison, “olanzapine versusother bipolar disorder medication known to cause moderate weight gain.”
`Asthis comparison suggests, studies with nonplacebo, nonhaloperidol comparators were grouped together due to their documented comparable
`effects on weight gain. On the basis of the literature, divalproex, lithium, risperidone, and asenapine haveall been associated with weight gain that
`is significantly greater than placebo, but less than olanzapine.2®32.31
`
`Primary Outcome: Mean Weight Gain
`
`Olanzapine versus placebo/haloperidol. The 5 studiesinthe first analysis compared olanzapine monotherapyversus placebo or haloperidol. In
`
`this comparison, olanzapine wasassociated with significantly more weight gain than placebo or haloperidol (Figure 2). The pooled mean
`difference of this comparison was 2.10 kg (95%CI, 1.16-3.05: P < .001). The results showed significant heterogeneity among the 5 studies (P =
`90%, C = 38.32, P < .001). Sequential removalof single studies from the analysis was performedto test for a possible outlier, but no single
`removal was found to render heterogeneity nonsigniticant.
`
`Olanzapine versus other bipolar disorder medication. The 10 studies in the second comparisoninvestigated olanzapine monotherapy versus
`other bipolar disorder medications that are known to cause moderate weight gain. The outcomeof this comparison showedgreater weight gain
`associated with olanzapine versus other bipolar disorder medication. As expected, the effect size was smaller than that observed in thefirst
`comparison. The pooled mean difference of this comparison was 1.34 kg (95% CI, 0.95—-1.72: P < .001). The heterogeneity of the 10 studies was
`not significant (/* = 27%, 7° = 12.30, P = .20).
`
`Separated analyses. To test the beliefthat the divalproex, lithium, risperidone, and asenapine trials can justifiably be combinedinto 1
`comparison andthat the haloperidol trial can be combined with the placebo-controlled comparison, a second analysis was performed withall
`comparators separated (Figure3), Results from this analysis showed little changein the placebo group's pooled effect size and heterogeneity when
`the haloperidol trial? was separated, and, so, this combination maybe justifiable.
`
`Effect sizes from the lithium and risperidone groups were similar (mean differences of 0.88 to 0.64 kg, respectively), suggesting justifiable
`combination, However, the divalproex and asenapine groups showedvaried effect sizes (mean differences of1.42 and 2.20 kg, respectively). Only
`the divalproex meandifference of 1.42 kg was similar to the combined mean difference of 1.34 kg. Furthermore, all groups in the separated
`analysis showed less heterogeneity, when applicable, than the combined analysis. On the basis of these results, it is unclear whether the combined
`analysesare justifiable; therefore, both the combined and separated analyses are presented.
`
`Secondary Outcome:Incidence of Weight Gain
`
`Olanzapine versus placebo/haloperidol. When incidences of reported weight gain were investigated, olanzapine wasagain associated with
`significantly more weight gain than wasplacebo or haloperidol (Figure4+). The pooled OR ofthis comparison was 10.12 (95%CI, 1.93-53.14; P =
`.006). Theresults for these trials were heterogeneous (/? = 78%, 72 = 13.68, P = .003). The heterogeneity was rendered nonsignificant with the
`removal of2 studies by Tohenet al: Tohenet al!2 (J? = 0%, y? = 1.12, P = .57) and Tohenet al!2 (7? = 51%, 77 = 4.10, P = .13), andit is possible
`that at least 1 ofthese studies is an outlier. It is worth noting that exclusion ofthe Tohenet al study!2 increased the pooled OR to 17.35 (95%Cl,
`3.20—94.01). and exclusion of the Tohenetal study!2 decreased the OR to 3.68 (95%CI. 2.39-5.67). However. exclusion ofthese studies did not
`affect the conclusion of the results, as they remained significant in all cases.
`
`Second comparison: olanzapine versus other bipolar disorder medication. Analysis on incidences of weight gain for the second comparison
`showed more people gaining weight with olanzapine versus other bipolar disorder medication, with an effect size smaller than that observedin the
`first comparison, The pooled OR ofthis comparison was 2.09 (95% CI, 1.27-3.44; P = .004). Theresults for these trials were heterogeneous -
`66%, 72 = 17.51, P = .008), but heterogeneity was rendered nonsignificant with the removal ofthe study by Novicket al2° (/? = 38%, 72 = 8.00, P
`= .16), indicating that it maybe an outlier. Removalofthis studyhadlittle effect on the pooled OR, however, and did not affect the conclusion of
`the analysis.
`
`
`Separated analyses. As with the primary outcome, a second analysis was performed with all comparators separated (Figure 5). For this outcome,
`separation of the haloperidol study from the placebo-controlled group caused anincrease in the pooled ORofthe placebo group (from 10.12 to
`17.42). In addition, heterogeneity of the results decreased, indicating the combined analysis maynot be justifiable.
`3
`
`

`

`Effect sizes from the divalproex, lithium, asenapine, and risperidone groups were relatively similar (ORs ranging from 1.68 to 6.45). Individually,
`the divalproex group showedless heterogeneity (P=1 1%) when separated, whereas the risperidone group showed more heterogeneity (= 92%),
`3
`‘
`,
`2
`4
`;
`6
`F
`3
`ee
`possibly owing to the previously discussed outlier effect of the study by Novick et al. As with the primary outcome, it is unclear whether a
`combinedanalysis ts justifiable; therefore, both the combined and separated analyses are presented.
`
`Possible covariates. Linear regression analysis failed to showa significant relationship between study duration and mean weight gain whenall of
`the studies were included (R = 0.39; F'1,12 = 2.10; P = .17). However, when the maintenance study by Tohenet al2! was removed from the
`analysis, a significant relationship was observed betweenstudy duration and mean weight gain (R = 0.78; F| 1) ~ 17.42; P = .0015). The same
`pattern with nearly identicalstatistics was observed whenincidences of weight gain were used in place of mean weight gain, suggesting a possible
`plateau effect on weight gain between 12 and 47 weeks. A second linear regression was performedto assess the effect of mean olanzapine dosage
`on mean weight gain. This analysis found nosignificant effect (R = -0.25; F), 19 = 0.62; P = 45). Additionally, no significant effect of mean
`olanzapine dosage onincidencesof weight gain was found (R = —0.18: F1,7 = 0.24; P = .64).
`
`DISCUSSION
`
`
`Goto:
`
`To our knowledge, this is the first meta-analysis investigating the extent of weight gain associated with olanzapine monotherapyin the treatment
`of patients with bipolar disorder. The results clearly showthat olanzapine monotherapyis associated withsignificantly more weight gain than
`placebo and other bipolar disorder medicationsthat are known to cause moderate weight gain. These medications include other SGAs(risperidone
`and asenapine). a first-generation antipsychotic (haloperidol), a mood stabilizer (lithium), and an anticonvulsant (divalproex/valproate ). These
`results held wheninvestigating mean weight gained during monotherapy, as well as incidences of reported weightgain.
`
`Ourresults are consistent with meta-analyses investigating the weight gain effects of olanzapine on schizophrenia subjects that have also focused
`on comparisons between different SGAs. In an analysis of 16 studies by Rummel-Klugeetal,? for example, the mean difference in weight gain
`between olanzapine and risperidone was 2.44 kg (95%CI, 1.61—3.27) in favorofrisperidone, while in a 13-study analysis by Komossa et al, the
`meandifference between olanzapine and risperidone was 2.61 kg (95%CI, 1.48—3.74) in favor of risperidone. The results from the present
`analysis, in comparison, showa mean difference of 0.64 kg (95%CI, -0.12 to 1.40) in favor of risperidone, although the results are based on only
`2 studies. A 9-study meta-analysis by Leuchtet al2 investigated the weight gaineffects of olanzapine versus haloperidol on schizophrenia subjects
`and suggests a meandifference of 3.3 kg (95% CI, 2.24.4) in favor of haloperidol, similar to the present analysis that suggests a meandifference
`of 2.80 kg (95%CI, 1.90-3.70)in favor ofhaloperidol the basis of 1 study only. Due to the low powerofthe olanzapine versusrisperidone and
`olanzapine versus haloperidol comparisonsin this analysis, it is still difficult to draw strong conclusions. Ourfindingsare supported bya study by
`Treueret al2 in which olanzapine monotherapy was examinedin both schizophrenia and bipolar subgroups.In this study, both groups gained
`weight, with the schizophrenia patients gaining a larger proportion of weightearlier than the bipolar subgroup (25% vs 11%, respectively).
`
`A limitation of this analysis is the small numberofstudies available. Without active exclusion ofpossible outliers, the combined analyses for both
`outcomes showed significant heterogeneity, with the exception of the combined meandifferences for bipolar disorder medications known to cause
`moderate weight gain (primary outcome, second comparison). Performing separate analyses (with all comparators separated) helped remedythis
`problem in most cases but also decreased the numberoftrials and total numberof subjects in each comparison, resulting in a decrease in the
`powerofthe individual analyses.
`
`Another point of caution stems from the conclusionsofthe individual studies. While all placebo‘/haloperidoltrials reported more weight gain in
`their olanzapine monotherapygroup, this was not the case for the other comparators. The naturalistic study by Kim et al?2 reported greater mean
`weight gain with lithium monotherapy when compared to olanzapine monotherapy, while a study by Novick et al*® found nodifference in the
`mean weight gain between the olanzapine group andthe risperidone group, but reported more incidences of weight gain in the risperidone group.
`Thistrial, however, was targeted as a possible outlier in our analysis. Anotherlimitation of our results is a confounder that impacts manyreviews
`on medication use: only2 trials in our analysis*22 did not receive funding from Eli Lilly, the maker ofolanzapine. Weight gain was not the
`primary endpointofthe trials included, however, and weight gain was reportedin all studies.
`
`The results of this analysis highlight the significant weight gain associated with olanzapine in the treatment of bipolar disorder and illustrate the
`need to focus onthis side effect. Olanzapine wasthe most effective SGA in the Clinical Antipsychotic Trials of Intervention Effectiveness
`schizophrenia trial®3 and has proveneffectiveness in bipolar disorder as well. Issues related to weight gain havea significant impact ontreatment
`compliance, and a consensus statement byclinical and research experts lists SGA-associated weight gain as 1 of the main risk factors for
`medication adherence in both patients with bipolar disorder and patients with schizophrenia,*+ putting patients at risk for relapse.>=
`
`Weight gain mayalso exacerbate other health risks associated with bipolar disorder, as both obesity and mood disorders are chronic low-grade
`proinflammatorystates, and the 2 conditions existing together may be associated with problems such as compromised neurocognitive function.*®
`Rates of obesity-related medicalillness, such as cardiovascular disease, type 2 diabetes, and metabolic syndromeare also increasedin patients
`with bipolar disorder and mayincrease the risk of premature mortalityin this population2232 Therefore, continued research on the metabolically
`adverse effects of SGAsis warranted in order to counter these effects, increase treatment compliance, and confer better patient care to individuals
`with bipolar disorder.
`
`Side effect profiles of medications need to be considered when making treatment decisions, as doesthe use of interventions to counteract these
`side effects, Current literature suggests that most ofthe weight gain associated with olanzapine occurs within the first 12 weeks of treatment,>2
`with a plateau usually reached between 36 and 39 weeks.=232 These results mayhavelittle to do with the pharmacologic properties ofthe
`medication, however, and instead mayberelated to interventions andlifestyle changes that are put in place once weight changes are observed. A
`numberoftrials examining the effects of medication,4°-+! cognitive-behavioral therapy,” andlifestyle** changes have beenable to successfully
`impact weight gain and cause weightloss in patients taking olanzapine. As clinicians, we need to be more aware of medication side effects and
`better able to educate patients on howto minimize these outcomes. This awareness will ensure that patients are able to safely choose the best
`medication to manage their complicated illness,
`
`

`

`Drug names: asenapine (Saphris), clozapine (Clozaril, FazaClo, and others), divalproex sodium (Depakote and others), fluoxetine (Prozac and
`others), haloperidol (Haldol and others), lithium (Lithobid and others), olanzapine (Zyprexa), risperidone (Risperdal and others).
`
`Potential conflicts ofinterest: None reported.
`
`Funding/support: None reported.
`
`Acknowledgments
`
`Goto:
`
`Acknowledgments: The authors would like to thank David Streiner, PhD (Department of Psychiatry and Behavioural Neurosciences, McMaster
`University, Hamilton, and Department of Psychiatry, University of Toronto, Ontario, Canada), for his help and direction regarding the meta-
`analytical techniques. Dr Streiner reports no conflicts ofinterest related to the subject ofthis article.
`REFERENCES Goto:
`
`
`
`1. Yatham LN, KennedySH, Schaffer A, et al. Canadian Network for Mood and Anxiety Treatments (CANMAT)andInternational Society for
`Bipolar Disorders (ISBD) collaborative update of CANMATguidelines for the managementof patients with bipolar disorder: update 2009.
`Bipolar Disord, 2009; 11(3):225-255. [PubMed: 19419382]
`
`2. Scherk H, Pajonk FG, Leucht S. Second-generation antipsychotic agentsin the treatment of acute mania: a systematic review and meta-analysis
`of randomized controlled trials. Arch Gen Psychiatry. 2007;64(4):442-455. [PubMed: 17404121]
`
`3. Liauw SS, McIntyre RS. Atypical antipsychotic tolerability and switching strategiesin bipolar disorder. Expert Opin Pharmacother.
`2010; 11(17):2827-2837. [PubMed: 20726821]
`
`4, Cascade EF, Kalali AH, LiebermanJ, et al. Use of antipsychotics pre- and post-dissemination of CATIE data. Psychiatry (Edgmont)
`2007;4(5):21-23. [PMCID: PMC2921309] [PubMed: 20806026]
`
`5. Symbyax Approval Letter. Department of Health and Human Services. Food and Drug Administration, 12-24-2003. NDA 21-520.
`
`6. Zyprexa Approval Letter (NDA 20—592/S-019), Department of Health and Human Services. Food and Drug Administration, 01-14-2004
`www.accessdata.fda.gow/drugsatfda_docs/appletter/2004/20592se1-01 91tr.pdfAccessed October 3, 2011
`
`7. Leucht 8S, Corves C. Arbter D, et al. Second-generation versusfirst-generation antipsychotic drugs for schizophrenia: a meta-analysis. Lancet.
`2009;373(9657):31-41. [PubMed: 19058842]
`
`8. Komossa K, Rummel-Kluge C, HungerH, et al. Olanzapine versus other atypical antipsychotics for schizophrenia. Cochrane Database Syst
`Rev. 2010:(3) CD006654. [PMCID: PMC4169107]
`
`9. Rummel-Kluge C, Komossa K, Schwarz §,et al. Head-to-head comparisons of metabolic side effects of second generation antipsychotics in the
`treatmentof schizophrenia: a systematic reviewand meta-analysis. Schizophr Res. 2010;123(2—3):225-233. [PMCID: PMC2957510]
`[PubMed: 20692814]
`
`10. Tohen M, Greil W, Calabrese JR, et al. Olanzapine versus lithium in the maintenance treatmentof bipolar disorder: a 12-month, randomized,
`double-blind, controlled clinical trial. Am J Psychiatry. 2005;162(7):1281—1290. [PubMed: 15994710]
`
`11. Tohen M, Calabrese JR. Sachs GS,et al. Randomized. placebo-controlled trial of olanzapine as maintenance therapy in patients with bipolar |
`disorder responding to acute treatment with olanzapine. Am J Psychiatry. 2006; 163(2):247-256. [PubMed: 16449478]
`
`12. Furukawa TA, Barbui C, Cipriani A, et al. Imputing missing standard deviations in meta-analyses can provide accurate results. J Clin
`Epidemiol. 2006; 59(1 ):7—-10. [PubMed: 16360555]
`
`13. Higgins JPT, Green §, Cochrane Handbook for Systematic ReviewsofInterventions: Version 5.1.0. 2011 eds. The Cochrane Collaboration.
`www.cochrane-handbook.org. Accessed October 3, 2011.
`
`14. Streiner DL. PowerPoint presentation presented at the Departmentof Psychiatry, University of Toronto; Toronto, Ontario, Canada: September
`21, 2010. Meta-analysis.
`
`15. Tohen M, Sanger TM, McElroySL, et al. Olanzapine HGEH Study Group. Olanzapine versus placebo in the treatment of acute mania. Am J
`Psychiatry. 1999; 156(5):702—709. [PubMed: 10327902]
`
`16. Tohen M, Jacobs TG, GrundySL, et al. The Olanzipine HGGW Study Group. Efficacy of olanzapine in acute bipolar mania: a double-blind,
`placebo-controlled study. Arch Gen Psychiatry. 2000;57(9):84 1-849. [PubMed: 10986547]
`
`17. Tohen M, Vieta E, Calabrese J, et al. Efficacy of olanzapine and olanzapine-fluoxetine combination in the treatment ofbipolarI depression.
`Arch Gen Psychiatry. 2003;60(11): 1079-1088. [PubMed: 14609883]
`
`18. Tohen M, Vieta E, Goodwin GM,etal. Olanzapine versus divalproex versus placebo in the treatment of mild to moderate mania: a
`randomized, 12-week, double-blind study. J Clin Psychiatry. 2008;69(11):1776—-1789. [PubMed: 19014751]
`
`19. Tohen M, Goldberg JF, Gonzalez-Pinto Arrillaga AM,et al. A 12-week, double-blind comparison ofolanzapine vs haloperidol in the treatment
`of acute mania. Arch Gen Psychiatry. 2003;60(12):1218—1226. [PubMed: 14662554]
`
`20. Zajecka JM, Weisler R, Sachs G,et al. A comparison ofthe efficacy, safety, and tolerability of divalproex sodium and olanzapinein the
`treatmentofbipolar disorder. J Clin Psychiatry. 2002;63(12):1148—1155. [PubMed: 12523875]
`
`

`

`21. Tohen M, Ketter TA, Zarate CA, et al. Olanzapine versus divalproex sodiumfor the treatment of acute mania and maintenanceof remission: a
`47-week study. Am J Psychiatry. 2003c:160(7):1263—1 271. [PubMed: 12832240]
`
`22. Kim B, KimSJ, SonJI, et al. Weight changein the acute treatment ofbipolar | disorder: a naturalistic observational study of psychiatric
`inpatients. J Affect Disord. 2008; 105(1-3):45—S2. [PubMed: 17498812]
`
`23. Novick D, Gonzalez-Pinto A, Haro JM, et al. EMBLEM Advisory Board. Translation of randomised controlled trial findingsinto clinical
`practice: comparison of olanzapine and valproate in the EMBLEMstudy. Pharmacopsychiatry. 2009;42(4):145—152. [PubMed: 19585393]
`
`24. Niufan G, Tohen M, Qiuging A,et al, Olanzapine versus lithium in the acute treatment of bipolar mania: a double-blind, randomized,
`controlled trial. J Affect Disord. 2008; 1