Bipolar disorder affects approximately 4% of US adults and is associated with several physical and psychiatric comorbidities.1 In comparison, the lifetime prevalence of PTSD is 20%, substance use disorder is 47%, and anxiety disorder is 75% – each of which contribute to the complexity of diagnosis and therapy for bipolar disorder.

When considering pharmacotherapy for bipolar disorder, evidence-supported options are scarce and limited by adverse effects (AEs). Additionally, the consensus in clinical guidance is lacking. This prescribing update highlights the latest recommendations for treating bipolar depression, according to the 2019 Psychopharmacology Algorithm Project at the Harvard South Shore Program (PAPHSS), which factors long-term AEs heavily within their recommendations.2

The following pharmacotherapeutic approaches apply to acute bipolar depression – signifying that the patient has a high/urgent risk of suicide and/or homicidal ideation – and chronic bipolar depression, as part of bipolar disorder 1 and 2.

Bipolar Depression Treatments: Urgent and First-Line Approaches

The following group of pharmacotherapeutic approaches applies to patients presenting with acute bipolar depression – signifying that they have a high/urgent risk of suicide and/or homicidal ideation.

Electroconvulsive Therapy (ECT)

Emerging data points to the recommendation of electroconvulsive therapy (ECT) to manage urgent, acute bipolar depressive episodes. Two trials demonstrated a 74% response rate (≥50% symptomatic improvement) with ECT compared to 35% when using lithium, quetiapine, lamotrigine, olanzapine, or antidepressants.3


For patients who fail or refuse ECT, intravenous (IV) ketamine is recommended for urgent, acute bipolar depression. Trials in treatment-resistant bipolar depression2 found a response in 50% to 80% of participants using a single administration of IV ketamine added to a mood stabilizer which lasted 7 to 10 days; however, eight out of nine participants relapsed within 19 days after the last administration of ketamine.4

In patients who do not have urgently acute bipolar depression or who failed IV ketamine, the PAPHSS algorithm points to the use of either lithium, quetiapine, lurasidone, cariprazine, or lamotrigine. These treatments are considered first-line options as well, however, choice of therapy should be individualized to patient preferences, efficacy, and considerations to AEs.2 Table I provides a summary. Below we take a closer look at the individual medication options.


Lithium is an effective pharmacotherapy that is FDA approved for acute mania and maintenance of bipolar depression; however, efficacy in acute bipolar depression is lacking. When compared to quetiapine and placebo, lithium was no better than placebo (P=0.123) in patients with acute bipolar depression whereas quetiapine performed significantly better when compared to placebo (P<0.001).5

In another study, lithium was compared to quetiapine for 1 year as maintenance therapy and shown to be more effective in reducing mania and depression relapses than quetiapine.6

Additionally, lithium has been suggested to have neuroprotective effects. The use of lithium was associated with higher white matter indicating improved structural integrity.7 This finding may indicate benefits in other neurological disease states as well, such as Alzheimer’s disease and Parkinson’s disease.8 However, when considering the adverse effects of lithium, many clinicians and patients have decided to avoid lithium altogether. Common side effects include cardiovascular, endocrine, metabolic, gastrointestinal, hepatic, renal, immunologic effects, and more. It is, therefore, advised that clinicians consult the literature and provide patient education to address concerns of side effects with lithium before prescribing.2


Quetiapine is FDA-approved for acute bipolar depression based on recommendations from the BOLDEN I and II trials; however, efficacy as a monotherapy maintenance medication is lacking. In these trials, quetiapine was more effective than placebo in treating bipolar depression. The BOLDEN I trial showed 58% of participants treated with quetiapine responded vs 36% with placebo.9 In the BOLDEN II trials, quetiapine was effective in both bipolar I and II. The trials showed that the 300 mg/d and 600 mg/d dose of quetiapine were both equally effective in producing a response and increasing remission rates.10 Secondary outcomes analyzing the AEs of quetiapine were also included. Weight gain, increased triglycerides, reduced insulin sensitivity, and QTc prolongation were all associated side effects of quetiapine.2


Lamotrigine has had mixed results in its efficacy as monotherapy maintenance for bipolar depression; it is not currently FDA approved for acute bipolar depression. In multiple trials, lamotrigine demonstrated no efficacy in preventing [hypo]mania; however, the drug had less AEs when compared to lithium. A single, double-blind, placebo-controlled study compared the use of lamotrigine (50 or 200 mg/d) in bipolar I. The trial that found 41% of patients improved when given 50 mg/d, 51% when given 200 mg/d, and 26% with placebo.11

Four large, placebo-controlled studies found conflicting results when studying lamotrigine in acutely depressed bipolar I and II patients. The trials found no difference between lamotrigine and placebo.12 Before prescribing lamotrigine, clinicians must assess the manic profiles of the patients and consider the addition of an agent to address the mania. With a high tolerance profile and concerns for the patient’s manic profiles, lamotrigine may be considered first-line for patients – especially those with bipolar II.2


Lurasidone is FDA approved for bipolar depression in bipolar 1 patients; however, it is not recommended for acute mania nor to prevent recurrences of [hypo]mania. Trials assessing lurasidone monotherapy vs placebo revealed a number needed to treat (NNT) of six when administered 20-60 mg/d and seven when administered 80-120 mg/d after 6 weeks of treatment. The higher dose of lurasidone was associated with greater AEs without higher rates of discontinuation.13 Lurasidone appears to lead to less weight gain than lithium and less QTc prolongation than quetiapine; however, akathisia and nausea are most commonly reported with lurasidone use.2 Additionally, case reports have suggested the possibility for lurasidone to precipitate [hypo]mania, especially in low doses (40 mg).14


Cariprazine was FDA approved in 2019 for bipolar mania with and without mixed features as well as bipolar I depression. In a meta-analysis of all four current studies using cariprazine, the NNT with cariprazine was 10 compared to lurasidone of 5 and quetiapine of 6; however, cariprazine produced less metabolic side effects compared to quetiapine. The number needed to harm (NNH) for weight gain was 16 for quetiapine, 58 for lurasidone, and 50 for cariprazine; however, the cariprazine studies were all 6-week-long studies.15

Long-term effects of cariprazine have not been fully assessed. Major side effects may include extrapyramidal side effects (EPS), nausea, akathisia, and restlessness. When considering the use of cariprazine in [hypo]manic patients, dosing should be assessed. Optimal dosing for bipolar depression is 1.5 mg; however, dosing for mania is higher at 3 to 6 mg.2

Table I: Urgent and First-Line Treatments for Bipolar Depression (BP-DEP)2


Therapy Indication(s) Adverse Effects Clinical Considerations
Urgent Therapy*
Electroconvulsive therapy (ECT) Acute, urgent BP-DEP Headaches, nausea, confusion Patients experiencing longer depressive episodes are more likely to respond
Ketamine Acute, urgent BP-DEP Arrhythmia, nausea, and vomiting (N/V), tachycardia, euphoria, confusion, slurred speech, sedation, hypertension, dissociation of mind from body, change in perception of time, double vision May produce rapid improvement in patients to then seek other treatments for maintenance therapy
First-Line Therapy
Lithium Acute mania and prevention of bipolar mood episodes Hand tremor, sexual dysfunction, weight gain, diarrhea, blurred vision, N/V, renal/hepatic impairment, hair loss, blunted high mood Does not have efficacy in acute BP-DEP

Acute BP-DEP


Orthostasis, xerostomia, weight gain, QTc prolongation, lethargy, dizziness, general weakness Not approved for maintenance monotherapy for bipolar disorder
Lamotrigine Prevention of recurrences of BP-DEP Blurred vision, drowsiness, dizziness, insomnia, tremor, rash (SJS), N/V, fatigue Does NOT prevent [hypo]manic episodes. May decrease weight
Lurasidone Prevention of recurrences of BP-DEP in bipolar 1 Increased triglycerides, psychomotor impairment, dystonia, insomnia, dyspepsia, N/V, dizziness, akathisia, QTc prolongation Does NOT treat acute mania nor prevent recurrences of [hypo]mania. Case reports suggest the precipitation of [hypo]mania at relatively low doses
Cariprazine Bipolar mania +/- mixed features and acute bipolar 1 depression EPS, akathisia, dyspepsia, drowsiness, weight gain, nausea, syncope Not approved in patients with dementia-related psychosis

 *Urgent signifies a patient with a high/urgent risk of suicide and/or homicidal ideation.

Alternative Treatment for Treatment-Resistant Bipolar Depression

Olanzapine and Fluoxetine Combination

As an alternative treatment, the use of olanzapine and fluoxetine combination (OFC) is also approved for acute bipolar depression. However, due to significant metabolic effects resulting in long-term risk for morbidity and mortality, OFC is not recommended early on in treatment. The use of OFC should be reserved until after all the aforementioned first-line medications and valproate have been trialed (more on valproate below).

In studies comparing OFC to lamotrigine, similar remission rates were found in both arms; however, lamotrigine was better tolerated.16 Recent meta-analysis show combinations of second-generation antipsychotics and antidepressants do not provide the same effectiveness as OFC.2

Additional Medications for Bipolar Depression: Second- and Third-Line Agents


Other options for use in bipolar depression – whether acute or chronic – are limited. Valproate was studied in four small trials (total n=142) as a maintenance treatment to prevent bipolar depression and concluded that valproate may be effective.17 On the contrary, a 2014 meta-analysis found valproate to not be effective.18 Valproate is a teratogen and associated with weight gain, which contributes to the recommendation to use valproate after all other first-line agents have failed in patients with a mixed state and/or have a history of rapid cycling, which is more common in women with bipolar disorder.

Carbamazepine, Oxcarbazepine, and Antidepressants

Other agents such as carbamazepine, oxcarbazepine, and antidepressants (aside from fluoxetine) lack sufficient evidence to treat bipolar depression. The use of antidepressants in bipolar I disorder is widely discouraged. This recommendation follows even when added to a mood stabilizer. This consensus was based on studies showing worse maintenance outcomes when rapid cyclers were given antidepressants.19 On the contrary, the efficacy of antidepressants in bipolar 2 is unclear. More research is required before further recommendations regarding antidepressants in bipolar 2 can be made. Clinicians must properly diagnose individuals according to the DSM-5 criteria before they consider prescribing antidepressants.2


A new drug containing olanzapine and samidorphan (Lybalvi) was recently FDA approved for bipolar 1 disorder (manic or mixed episodes) or as an adjunct to lithium or valproate. It carries the same metabolic limitations as olanzapine, and it cannot be used in patients taking opioids or it may precipitate withdrawal. Cost may also limits is use.20

Prescribing for Bipolar Disorder

Discrepancies remain with recommendations made by the PAPHSS and other guidelines for bipolar depression. The differences are largely due to the PAPHSS emphasis on long-term adverse effects associated with potential treatments. When providing recommendations, other guidelines do not weigh adverse effects heavily in making their recommendations.

For example, here is a brief look at other clinical practice guidelines for treating bipolar disorder:

  • The Canadian Network of Mood and Anxiety Treatments (CANMAT) and International Society for Bipolar Disorders (ISBP) 2018 guideline update recommends quetiapine, lurasidone plus lithium or divalproex, lithium, lamotrigine, lurasidone, or adjunctive lamotrigine as first-line therapies for bipolar I.21
  • The International College of Neuro-Psychopharmacology (CINP) 2016 treatment Guidelines for Bipolar Disorder in Adults recommends lurasidone and quetiapine as first-line options. They note fluoxetine, escitalopram, olanzapine, and OFC as second-line recommendations with lithium recommended only after other options have failed.22

Prescribing for bipolar depression, therefore, remains a challenge for both patients and clinicians. The selection of a first-, second-, or third-line agent is difficult, as therapy should be individualized with recognition to the efficacy and side effect profile of the therapies. Additionally, available guidelines conflict regarding their recommendations for treatment. Clinicians must be alert to emerging evidence in the psychiatric field in order to provide the safest, most effective care.

Last Updated: Jul 14, 2021