If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
The US Food and Drug Administration (FDA) has approved 10 treatments for acute mania, but only 2 for acute bipolar depression. These agents differ from one another with respect to their efficacy and tolerability profiles, such that certain medications may be optimal for some patients but not others.
Objective
Our aim was to compare the therapeutic and adverse effects of treatments for acute mania and acute bipolar depression to inform clinical decision making.
Methods
Using data from large, randomized, double-blind, placebo-controlled acute mania and acute bipolar depression trials, we assessed number needed to treat (NNT) for response, number needed to harm (NNH) for sedation/weight gain, and likelihood to help or harm (LHH = NNH ÷ NNT) compared with placebo.
Results
For acute mania, lithium compared with other FDA-approved agents yielded substantively less sedation (NNH, 27 vs 5−17) yet broadly similar efficacy (NNT, 4 vs 4–8), and thus a more favorable efficacy:sedation likelihood (LHH, 6.8 vs 0.7–3.4). For acute bipolar depression, lamotrigine compared with FDA-approved treatments yielded substantively less sedation (NNH, 42 vs 6–12), weight gain (NNH, −34 vs 6−19), and efficacy (NNT, 12 vs 4–6), but still more favorable efficacy:sedation likelihood (LHH, 3.5) than quetiapine (LHH, 1.0) and efficacy:weight gain likelihood (LHH, −2.8) than the olanzapine plus fluoxetine combination (LHH, 1.5).
Conclusions
For acute mania, lithium compared with other FDA-approved agents yielded less sedation yet similar efficacy, indicating utility for patients sensitive to sedation. For acute bipolar depression, lamotrigine compared with FDA-approved treatments yielded better tolerability but poorer efficacy, suggesting utility in patients sensitive to sedation/weight gain with milder episodes, whereas the FDA-approved treatments might have utility in patients with more severe episodes.
Although pharmacotherapies for acute mania have advanced rapidly over the past decade, with 10 medications now having US Food and Drug Administration (FDA) approval, to date only 2 treatments have FDA approval for acute bipolar depression.
FDA-approved bipolar disorder treatments include 4 mood stabilizers (MSs; lithium, divalproex, carbamazepine, and lamotrigine) and 6 second-generation antipsychotics (SGAs; olanzapine, risperidone, quetiapine, ziprasidone, aripiprazole, and asenapine). Although MSs may be considered foundational treatments for bipolar disorder, recently, utility of SGAs rather than MSs has been emphasized in acute mania treatment reviews,
Comparative efficacy and acceptability of mood stabilizer and second generation antipsychotic monotherapy for acute mania—a systematic review and meta-analysis.
and SGAs (quetiapine monotherapy and the olanzapine plus fluoxetine combination) rather than MSs comprise the only FDA-approved treatments for acute bipolar depression.
However, for certain patients, the potential efficacy advantages of SGAs may be more than offset by tolerability disadvantages, such as the risk of sedation and/or somnolence (referred to hereafter as sedation) and weight gain. Sedation may not be particularly problematic during severe acute manic episodes, but is a common contributor to impaired function and nonadherence during acute bipolar depression, and can be challenging during milder acute mood elevation. Weight gain may be less problematic during acute mania compared with longer term treatment, but during acute bipolar depression, it can constitute a substantive challenge.
Clinicians managing patients with bipolar disorder may choose among varied agents with superior efficacy compared with placebo in controlled trials. However, translating the data from such studies into information that is relevant for clinical practice may prove challenging. Nevertheless, careful concurrent consideration of both efficacy and tolerability may permit such data to crucially inform optimal treatment choices for individual patients.
Tools of evidence-based medicine, such as number needed to treat (NNT), number needed to harm (NNH), and likelihood to help or harm (LHH), may aid clinicians in ascertaining the clinical relevance of efficacy study data.
NNT is a measure of the effect size of a categorical efficacy outcome for a treatment intervention, such as response or remission, and is calculated as the reciprocal of the difference in rates of the outcome between treatment groups. NNH is analogous to NNT, but for adverse events as opposed to efficacy outcomes. NNT and NNH can help clinicians understand how many patients need to be treated with 1 agent compared with placebo to yield 1 additional outcome of interest. A lower NNT indicates greater benefit, whereas a lower NNH suggests greater harm. Thus, a medication with a low (eg, single-digit) NNT coupled with a high (eg, double-digit) NNH could be an appealing treatment option. LHH (= NNH ÷ NNT) serves as an integrative measure of the balance between benefits and harms that can be expected with interventions. Higher LHHs indicate greater likelihoods of benefits compared with harms, with LHH >1 suggesting that a given intervention is more likely to help than to harm patients.
This article reviews placebo-controlled efficacy studies of MSs and SGAs for the management of acute manic/mixed and bipolar depressive episodes in patients with bipolar disorder, with an emphasis on the use of NNT, NNH, and LHH as quantitative tools to enhance the clinical relevance of controlled efficacy study data.
Methods
Data Sources
The literature was searched for large (N >100), randomized, double-blind, placebo-controlled trials of the efficacy of MSs and SGAs in bipolar disorder patients. The PubMed database was searched using the search terms bipolar, bipolar disorder, bipolar I disorder, bipolar II disorder, mania, manic episode, mixed episode, bipolar depression, randomized, controlled, treatment, efficacy, effectiveness, lithium, carbamazepine, divalproex, valproate, lamotrigine, olanzapine, risperidone, quetiapine, ziprasidone, aripiprazole, asenapine, and paliperidone.
Study Selection
Large (N >100), randomized, double-blind, placebo-controlled trials of the efficacy of FDA-approved treatments for acute manic and bipolar depressive episodes as well as key (NNT <20) unapproved pharmacotherapies for acute mania (ie, paliperidone) and acute bipolar depression (ie, lithium, lamotrigine, and olanzapine), were selected in patients with well-defined bipolar I disorder or bipolar II disorder. Acute mania studies of topiramate
Efficacy and safety of adjunctive oral ziprasidone for acute treatment of depression in patients with bipolar I disorder: a randomized, double-blind, placebo-controlled trial.
were excluded due to having NNT >20. Studies whose primary emphasis was not the treatment of bipolar disorder and studies not reporting response rates and sedation rates were excluded. For the purpose of brevity, only acute mania monotherapies were reviewed, so that studies of adjunctive therapies (eg, SGAs added to lithium or valproate) for acute mania were excluded. Also, the first-generation antipsychotic chlorpromazine, which has US FDA approval for acute mania, was omitted from analyses due to the lack of large studies using contemporary randomized, double-blind, placebo-controlled methodology.
Outcome Measures
The efficacy variable was NNT for acute response (percentage of subjects with at least 50% improvement) compared with placebo for manic/mixed and bipolar depressive episodes. The primary tolerability variables were NNH compared with placebo for sedation for acute mania studies and NNH compared with placebo for sedation/weight gain for acute bipolar depression studies. NNH for weight gain for acute mania studies was not considered, as weight gain data were missing from some acute mania studies. The alternative measure of tolerability, NNH for discontinuation due to side effects, was calculated for agents with significantly higher side effect discontinuation rates compared with placebo. The primary integrated efficacy and tolerability variables were LHH for response:sedation likelihood for acute mania studies, and LHH for response:sedation likelihood and for response:weight gain likelihood for acute bipolar depression studies.
NNT and NNH were calculated as the reciprocal of the difference in rates of the outcome of interest (eg, response or sedation/weight gain) between 2 treatment groups, rounded up to the next whole number.
Quantifying risk: the role of absolute and relative measures in interpreting risk of adverse reactions from product labels of antipsychotic medications.
For example, a medication with a 20% higher response rate compared with placebo would yield a NNT of 5 (1/0.2). LHH was calculated as NNH ÷ NNT, with a higher LHH indicative of a greater benefit-to-harm ratio.
Data Extraction
The authors (S.S. and T.K.) extracted data from individual studies. The following data were recorded: authors and years of publication, number of patients using active treatments and control interventions, doses, NNT, NNH, and LHH.
Data Analysis
For medications with FDA approval for acute mania and acute bipolar depression as well as for key unapproved (NNT <20) treatments, analyses of NNT for response, NNH for sedation/weight gain, and LHH for response versus sedation/weight gain were conducted using the retrieved reports. When >1 study was available for a particular agent, overall NNT, NNH, and LHH were obtained by pooling individual study data.
Results
Twenty-four studies meeting the inclusion and exclusion criteria were identified: 17 for acute mania
A multicenter, randomized, double-blind, placebo-controlled trial of extended-release carbamazepine capsules as monotherapy for bipolar disorder patients with manic or mixed episodes.
Extended-release carbamazepine capsules as monotherapy for acute mania in bipolar disorder: a multicenter, randomized, double-blind, placebo-controlled trial.
A randomized, placebo- and active-controlled study of paliperidone extended release for the treatment of acute manic and mixed episodes of bipolar I disorder.
). Thus, 9 FDA-approved monotherapies for acute mania with registration studies using contemporary (randomized, double-blind, placebo-controlled) methodology were reviewed, including 3 MSs (lithium, divalproex, and carbamazepine) and 6 SGAs (olanzapine, risperidone, quetiapine, ziprasidone, aripiprazole, and asenapine). Also, one key unapproved monotherapy for acute mania (ie, paliperidone) was reviewed. In addition, 2 FDA-approved treatments for acute bipolar depression (quetiapine monotherapy and the olanzapine plus fluoxetine combination) and 3 key unapproved pharmacotherapies for acute bipolar depression (ie, lithium, lamotrigine, and olanzapine) were reviewed.
Acute Mania
For acute mania, placebo-controlled studies of the agents listed in the Table and depicted in Figure 1 yielded broadly similar single-digit (ie, <10) NNTs, that ranged 2-fold, from 4 for lithium, carbamazepine, and risperidone to 8 for asenapine. In contrast, NNHs for sedation were more variable, ranging >5-fold, from 5 for olanzapine, risperidone, and ziprasidone to 27 for lithium. Response:sedation likelihoods (LHHs), were even more variable still, ranging almost 10-fold, from 0.7 for ziprasidone to 6.8 for lithium. Thus, lithium, compared with other treatments yielded substantively less sedation (NNH, 27 vs 5–17) yet broadly similar efficacy (NNT, 4 vs 4–8) and, thus, a more favorable efficacy:sedation likelihood (LHH, 6.8 vs 0.7–3.4).
TableTreatments for bipolar disorder: number (N), mean dose, likelihood to help or harm (LHH), number needed to harm (NNH), and number needed to treat (NNT).
Treatment Drug
Acute Mania
Acute Bipolar Depression
N
Dose (mg/d)
LHH (r:s)
NNH (s)
NNT (r)
N
Dose (mg/d)
LHH (r:s)
NNH (s)
LHH (r:w)
NNH (w)
NNT (r)
Mood stabilizers
Lithium
134
1950
6.8
27
4
136
981
1.3
20
−7.5
−112
15
Divalproex, divalproex ER
261
2778
1.0
7
7
—
—
—
—
—
—
—
Carbamazepine ER
223
694
1.5
6
4
—
—
—
—
—
—
—
Lamotrigine
—
—
—
—
—
541
100–400
3.5
42
−2.8
−34
12
Second generation antipsychotics
Olanzapine
124
16
1.0
5
5
351
9.7
0.6
7
0.5
6
12
Risperidone
273
4.9
1.3
5
4
—
—
—
—
—
—
—
Quetiapine, quetiapine XR
209
400–800
1.5
9
6
1166
300–600
1.0
6
3.2
19
6
Ziprasidone
268
121
0.7
5
7
—
—
—
—
—
—
—
Aripiprazole
260
28
1.8
9
5
—
—
—
—
—
—
—
Asenapine
372
18
1.4
11
8
—
—
—
—
—
—
—
Paliperidone ER
195
9
3.4
17
5
—
—
—
—
—
—
—
Combination
Olanzapine + fluoxetine
—
—
—
—
—
82
7.4 + 39.3
3.0
12
1.5
6
4
ER = extended-release formulation; r = response; s = sedation; w = weight gain; XR = extended-release formulation.
LHHs, NNHs, and NNTs are compared with placebo.
See text for references to studies.
*Potential LHH and NNH advantages of lithium for acute mania and of lamotrigine for acute bipolar depression.
Figure 1Likelihood to help or harm (LHH) for response versus sedation from double-blind, placebo-controlled acute mania monotherapy studies for 10 treatments. Data from recent controlled trials of the mood stabilizers lithium (Li),
A multicenter, randomized, double-blind, placebo-controlled trial of extended-release carbamazepine capsules as monotherapy for bipolar disorder patients with manic or mixed episodes.
Extended-release carbamazepine capsules as monotherapy for acute mania in bipolar disorder: a multicenter, randomized, double-blind, placebo-controlled trial.
A randomized, placebo- and active-controlled study of paliperidone extended release for the treatment of acute manic and mixed episodes of bipolar I disorder.
Black bars, treatments approved by the United States Food and Drug Administration for acute bipolar manic/mixed episodes. White bar, treatment not approved by the United States Food and Drug Administration for acute bipolar manic/mixed episodes.
The side effects discontinuation rate was only significantly greater than placebo for valproate (8.8% vs 3.1%), which yielded a NNH for side effects discontinuation compared with placebo of 17.
Acute Bipolar Depression
For acute bipolar depression, placebo-controlled studies of the agents listed in the Table and depicted in Figure 2 yielded similar single-digit (ie, <10) NNTs for the 2 FDA-approved treatments (6 for quetiapine monotherapy and 4 for the olanzapine plus fluoxetine combination) that were at least 50% lower than the NNTs for the key unapproved treatments (12 for lamotrigine, 12 for olanzapine, and 15 for lithium). NNHs for sedation were even more variable, ranging 7-fold, from 6 for quetiapine to 42 for lamotrigine. Response:sedation likelihoods (LHHs) ranged >5-fold, from 0.6 for olanzapine to 3.5 for lamotrigine. NNHs for weight gain ranged even more broadly from 6 for olanzapine and the olanzapine plus fluoxetine combination to −112 for lithium (with negative NNHs and LHHs indicative of greater harm from placebo compared with study drug). Response:weight gain likelihoods (LHHs) also ranged broadly, from 0.5 for olanzapine to −7.5 for lithium. Compared with FDA-approved treatments, lamotrigine yielded substantively less sedation (NNH, 42 vs 6–12), weight gain (NNH, −34 vs 6–19), and efficacy (NNT, 12 vs 4–6), but still more favorable efficacy:sedation likelihood than quetiapine (LHH, 3.5 vs 1.0) and efficacy:weight gain likelihood than the olanzapine plus fluoxetine combination (LHH, −2.8 vs 1.5).
Figure 2Likelihood to help or harm (LHH) for response versus sedation/weight gain from double-blind, placebo-controlled acute bipolar depression studies for 5 treatments. Data from recent controlled trials of the mood stabilizers lithium (Li)
Data graphed are for the less favorable LHH for each treatment. Black bars, treatments approved by the United States Food and Drug Administration for acute bipolar major depressive episodes. White bars, indicate treatments not approved by the United States Food and Drug Administration for acute bipolar major depressive episodes. (r:s) = LHH for response versus sedation (for Li, LTG, and QTP); (r:w) = LHH for response versus weight gain (for OLZ and OFC).
Side effect discontinuation rates were only significantly greater than placebo for quetiapine (12.9% vs 5.5%) and olanzapine (9.2% vs 5.0%), which yielded NNHs for side effect discontinuation compared with placebo of 14 and 24, respectively.
Discussion
We found FDA-approved treatments for acute mania had broadly similar efficacy compared with placebo, but more variable tolerability profiles and thus substantially different response:sedation likelihoods. Thus, for acute mania, lithium compared with other FDA-approved agents yielded less sedation yet similar efficacy, indicating utility for patients sensitive to sedation. Although key unapproved treatments compared with FDA-approved treatments for acute bipolar depression had less efficacy, they had potentially mitigating tolerability advantages. Thus, for acute bipolar depression, lamotrigine compared with FDA-approved treatments yielded better tolerability but poorer efficacy, suggesting utility in patients sensitive to sedation/weight gain with milder episodes, whereas the FDA-approved treatments might have utility in patients with more severe episodes. Side effect discontinuation rates were only significantly greater than placebo for valproate in acute mania and for quetiapine and olanzapine in acute bipolar depression, all of which yielded double-digit NNHs compared with placebo (range, 14–24), suggesting patients were willing to endure side effects for the limited durations (3 weeks for acute mania, 8 weeks for acute bipolar depression) of the trials.
The findings that demonstrated potential efficacy:sedation advantages for lithium over other treatments for acute mania were in contrast to the findings of recent meta-analyses of antimanic agents, which emphasized that SGAs compared with MSs had superior efficacy for the management of acute mania.
Comparative efficacy and acceptability of mood stabilizer and second generation antipsychotic monotherapy for acute mania—a systematic review and meta-analysis.
Our approach, which emphasized a specific side effect (sedation), might have been more sensitive regarding tolerability challenges with SGAs (and for that matter divalproex and carbamazepine) than the approaches used in recent meta-analyses. A recent analysis of Medicaid claims found that less than two-thirds of a cohort taking clinically recommended doses of SGAs were adherent to treatment, a finding that might in part reflect such tolerability challenges.
Thus, using NNT, NNH, and LHH as tools to interpret clinical trials data could provide a different perspective by comparing agents in terms of benefit-to-harm ratios for specific side effects. This might help inform personalized treatment decision making. Clinicians who anticipate maintaining their patients on the same agent once the acute episode resolves might prefer to initiate treatment with lithium due to its enhanced acute tolerability. In contrast, management of more severe acute manic or mixed episodes with very prominent insomnia and psychomotor agitation might require prioritization of efficacy at the expense of tolerability, with the options of decreasing antimanic medication dose(s) or switching to better-tolerated agent(s) once the acute manic/mixed episode remits.
The management of acute bipolar depression presents a different set of arguably even greater challenges. Despite the predominance of depression compared with mood elevation in the course of bipolar illness,
only quetiapine monotherapy and the olanzapine-fluoxetine combination have FDA approval for acute bipolar depression, and the utility of these agents can be substantially limited by sedation and weight gain, respectively. Pharmacotherapies lacking FDA approval for acute bipolar depression are commonly used, perhaps related to the small number of and substantial tolerability limitations of FDA-approved agents. The need to balance potential benefits with potential harms may be more clinically relevant for acutely depressed patients who, compared with acutely manic patients, are likely to be more chronically ill, more sensitive to medication side effects, and already struggling with sedation and weight gain (related to their depressed mood state). For severe depressive episodes, the compelling need for efficacy may mitigate the substantial tolerability concerns of the FDA-approved agents. For other patients, however, it may be preferable to look beyond the FDA-approved treatments and consider an agent, such as lamotrigine with its superior tolerability, even if it has less robust evidence of efficacy.
Our overall findings were consistent with those of previous studies evaluating pharmacotherapies for bipolar disorder in terms of NNT and NNH.
Quantifying risk: the role of absolute and relative measures in interpreting risk of adverse reactions from product labels of antipsychotic medications.
Number needed to treat to harm for discontinuation due to adverse events in the treatment of bipolar depression, major depressive disorder, and generalized anxiety disorder with atypical antipsychotics.
Thus, our study was consistent with previous studies that noted prominent weight gain associated with both olanzapine and olanzapine-fluoxetine combination, and prominent sedation associated with quetiapine, as well as significantly increased side effect discontinuation rates for olanzapine and quetiapine in acute bipolar depression. Our study extended the findings of previous work by highlighting potential efficacy:tolerability advantages for lithium in acute mania and lamotrigine in acute bipolar depression.
This study has several noteworthy limitations. The large, randomized, double-blind, placebo-controlled studies included in our analyses entailed strict inclusion and exclusion criteria that limited generalizability to patients more typically encountered in clinical practice (eg, with heterogeneous psychiatric and medical comorbidities and taking concurrent psychotropic and/or non-psychotropic medications). Additionally, our analysis only included large, randomized, double-blind, placebo-controlled efficacy studies, while omitting comparative effectiveness studies that could have provided an additional, complementary source of clinically relevant data. Studies included in this report were of only 3 to 8 weeks duration, potentially underemphasizing side effects that could have developed more insidiously over time (such as weight gain), or overemphasizing side effects that might have improved over time (such as sedation/somnolence). This article omitted analyses of longer term maintenance treatments for bipolar disorder due to methodological limitations in interpreting NNHs and LHHs for such studies, insofar as such calculations failed to adequately account for adverse events occurring during the open stabilization phase. Finally, assessment of NNT and NNH relies upon categorical parameters, such as the presence or absence of response or an adverse effect, thus setting arbitrary thresholds that might not be clinically meaningful and fail to capture data just outside of the threshold (eg, a 49% reduction in a depression rating scale score would be considered a nonresponse despite significant clinical improvement, and 6% weight gain would not be considered clinically significant despite its potential for substantial physical and psychological sequelae).
Conclusions
FDA-approved treatments for the management of acute mood episodes in patients with bipolar disorder demonstrated broadly comparable superior efficacy compared with placebo in controlled studies, yet differed substantively from one another, and from key unapproved treatments, with respect to tolerability and hence likelihood of help or harm. For acute mania, lithium compared with other FDA-approved agents yielded a superior response:sedation likelihood, and thus might be a treatment of choice for patients who are especially sensitive to the sedating effects of medications. For acute bipolar depression, lamotrigine compared with FDA-approved treatments yielded less robust efficacy yet superior response:sedation and response:weight gain likelihoods, suggesting it might be favored for patients with milder illness and/or greater sensitivity to sedation or weight gain. Evaluating therapies in terms of NNT, NNH, and LHH could facilitate clinical decision making that optimizes care while recognizing the divergent needs and priorities of individual patients.
Conflicts of Interest
Dr. Srivastava has received honorarium support from Pamlab. Dr. Ketter has received grant/research support from Abbott Labs; AstraZeneca; Bristol-Myers Squibb; Cephalon; Eli Lilly; GlaxoSmithKline; Pfizer; Repligen; Sunovion; and Wyeth; consulting fees from Abbott Labs; AstraZeneca; Bristol-Myers Squibb; Cephalon; Dainippon Sumitomo; Eli Lilly; Forest; GlaxoSmithKline; Janssen; Jazz; Merck; Novartis; Organon; Solvay; Valeant; Vanda; Wyeth; and XenoPort; and lecture honoraria from Abbott Labs; AstraZeneca; Bristol-Myers Squibb; Eli Lilly; GlaxoSmithKline; Noven; Otsuka; and Pfizer. Also, Dr. Ketter's spouse is an employee and owns stock in Johnson & Johnson. None of these companies provided support for or were in any way involved in the development of this manuscript.
Acknowledgments
Dr. Srivastava has received National Institute of Mental Health (NIMH) T-32 training grant support. Dr. Ketter has received research support from the Agency for Healthcare Research and Quality (AHRQ), National Alliance for Research on Schizophrenia and Depression (NARSAD), NIMH, Pritzker Foundation, and Stanley Foundation.
Drs. Srivastava and Ketter jointly designed the study, interpreted the data and prepared the table and figures. Dr. Srivastava performed the literature search and data tabulation, and wrote the draft manuscript. Dr. Ketter revised the draft manuscript.
References
Merikangas K.R.
Akiskal H.S.
Angst J.
et al.
Lifetime and 12-month prevalence of bipolar spectrum disorder in the National Comorbidity Survey replication.
Comparative efficacy and acceptability of mood stabilizer and second generation antipsychotic monotherapy for acute mania—a systematic review and meta-analysis.
Efficacy and safety of adjunctive oral ziprasidone for acute treatment of depression in patients with bipolar I disorder: a randomized, double-blind, placebo-controlled trial.
Quantifying risk: the role of absolute and relative measures in interpreting risk of adverse reactions from product labels of antipsychotic medications.
A multicenter, randomized, double-blind, placebo-controlled trial of extended-release carbamazepine capsules as monotherapy for bipolar disorder patients with manic or mixed episodes.
Extended-release carbamazepine capsules as monotherapy for acute mania in bipolar disorder: a multicenter, randomized, double-blind, placebo-controlled trial.
A randomized, placebo- and active-controlled study of paliperidone extended release for the treatment of acute manic and mixed episodes of bipolar I disorder.
Number needed to treat to harm for discontinuation due to adverse events in the treatment of bipolar depression, major depressive disorder, and generalized anxiety disorder with atypical antipsychotics.
00771-5&id=gr1.jpg){kind=link}
00771-5&id=gr2.jpg){kind=link}