Bipolar I Disorder

Article written and reviewed by Dr Luca Moretti

Bipolar I disorder is a mood disorder defined in the DSM-5-TR by the lifetime presence of at least one manic episode. Major depressive and hypomanic episodes are very common during the course but are not required for diagnosis. In ICD-11, the corresponding category is bipolar type I disorder, code 6A60, characterized by one or more manic or mixed episodes; depressive or hypomanic episodes may also occur.

Mania is not simply euphoria, increased productivity or irritability. It is an episodic syndrome characterized by a clearly abnormal and persistent change in mood associated with increased activity or energy, accompanied by additional cognitive and behavioral changes and severe enough to cause marked functional impairment, require hospitalization or produce psychotic manifestations. The presence of psychosis during a period of elevated mood makes the presentation, by DSM-5-TR definition, manic rather than hypomanic.

Bipolar I disorder is generally a recurrent illness. A patient may spend prolonged periods in remission, but manic, depressive and, in some cases, hypomanic episodes or episodes with mixed features may alternate over a lifetime. The polarity and frequency of episodes vary considerably between individuals and within the same individual over time.

Epidemiological estimates depend on diagnostic criteria and study methods. The World Mental Health Survey Initiative estimated the lifetime prevalence of bipolar I disorder at approximately 0.6%, while broader estimates for the bipolar spectrum are higher. The disorder affects both men and women, with sex differences being more evident in the polarity of the course and in comorbidities than in the simple presence of the diagnosis.

Onset frequently occurs between adolescence and early adulthood. Young age at onset, diagnostic delay and an initial presentation with depressive episodes may make recognition difficult, because many patients first come to clinical attention during depression and do not spontaneously report previous periods of activation. A longitudinal reconstruction of the entire mood history is therefore essential.

The burden of illness does not depend on mania alone. Depressive episodes may occupy a substantial part of the course and are closely associated with disability and suicide risk. Anxiety comorbidity, substance use disorders, obesity, metabolic syndrome, cardiovascular disease and other medical conditions are common and contribute to the reduced quality of life and life expectancy observed in populations with bipolar disorder.

Etiology, Pathogenesis and Pathophysiology

There is no single necessary and sufficient etiologic cause of bipolar I disorder. Evidence converges on a multifactorial architecture in which genetic predisposition, neurobiological development, circadian regulation, environmental exposures and psychosocial factors interact to modulate the probability of developing the illness and its clinical expression.

The genetic component is particularly important. Family and twin studies show strong familial aggregation, and heritability estimates for bipolar disorder as a whole are generally between 60 and 80%. This high heritability does not imply Mendelian transmission: the disorder is highly polygenic, and individual risk derives from the combined effects of numerous common variants, rare variants and nongenetic factors.

Large genome wide association studies have progressively expanded the number of loci associated with the disorder. The multiethnic analysis published in 2025 involving more than 158,000 cases identified hundreds of associated loci and strengthened evidence for genes and cell types involved in synaptic and neuronal function. These findings improve biological understanding of the illness, but polygenic risk scores do not currently have sufficient accuracy for use as diagnostic tests in clinical practice.

Genetic vulnerability is partly shared with other psychiatric disorders, particularly schizophrenia and major depression. This helps explain the overlap of some phenotypes and confirms that clinical diagnostic categories do not correspond to single genetic entities separated by absolute biological boundaries.

Environmental factors associated with onset or course include childhood adversity, stressful life events, sleep deprivation, substance use and disruption of circadian rhythms. These exposures are not deterministic causes and their weight varies between patients. In vulnerable individuals, however, they may contribute to mood destabilization or precede the onset of episodes.

The relationship with sleep is particularly relevant in mania. Reduced need for sleep is a cardinal symptom of the episode, but sleep loss may also precede or facilitate activation in susceptible individuals. Travel across time zones, shift work and marked irregularity of the sleep wake rhythm may therefore have clinical significance, although none is sufficient on its own to cause the disorder.

Pathophysiology involves distributed networks regulating emotion, reward, cognitive control, motivation and the stress response. Large neuroimaging studies have documented group level cortical and subcortical differences and functional alterations in circuits involving the prefrontal cortex, cingulate, amygdala, hippocampus, striatum and thalamocortical systems. None of these alterations is specific or constant in an individual patient.

Dopaminergic, glutamatergic, GABAergic and serotonergic systems are involved in mood regulation and are direct or indirect targets of effective treatments. Increased mesolimbic dopaminergic transmission has been proposed as one mechanism of manic activation, while dysfunction in glutamatergic and GABAergic transmission may alter network excitability and plasticity. These models are useful for research but do not justify a reductionist theory based on a single neurotransmitter.

Lithium, one of the most effective treatments for bipolar disorder, acts on multiple intracellular pathways, including inositol signaling, glycogen synthase kinase 3, neuroplasticity and regulation of gene expression. The fact that a drug with such broad molecular effects prevents episodes of different polarities is consistent with distributed, multilevel pathophysiology rather than a single neurochemical abnormality.

Mitochondrial dysfunction, oxidative stress, immune inflammatory processes and regulation of the hypothalamic pituitary adrenal axis have also been studied. Alterations in these systems have been found in subgroups of patients and may vary with illness state, medications, obesity, sleep and other comorbidities. They are not diagnostic biomarkers and should not be interpreted as universal abnormalities of the disorder.

Circadian mechanisms represent another pathogenetic level. Mood, sleep, hormone secretion and numerous metabolic functions are temporally organized; patients with bipolar disorder frequently show irregularities in the phase, amplitude and stability of biological rhythms. The close relationship between sleep changes and affective episodes is one reason why stabilization of daily routines is an integral part of clinical management.

Pathophysiology should ultimately be considered dynamically. Mania and depression are not simply two symmetrical extremes of the same process. They may be sustained by partly different combinations of vulnerability, network changes, sleep, stress and reward mechanisms. Mixed features further demonstrate that depressive and manic symptoms can coexist within the same episode.

There is currently no blood, genetic, electrophysiological or neuroimaging biomarker capable of diagnosing bipolar I disorder. The clinical translation of biological knowledge mainly concerns understanding risk and developing new treatments; diagnosis remains based on the longitudinal clinical phenotype.

Clinical Manifestations

Assessment should begin by reconstructing the reason for presentation and the change from usual functioning. The patient may present during mania, depression, partial remission or because of behavioral consequences of an episode. During mania, insight is often reduced, and information from family members or people who know the patient well may be decisive in defining the duration and severity of the change.

The core of mania is a distinct period of abnormally elevated, expansive or irritable mood associated with a persistent increase in activity or energy. Irritability may predominate, especially in severe or mixed presentations, and may result in conflict, verbal aggression or impulsive behavior.

Reduced need for sleep differs from insomnia. The patient may sleep only a few hours and wake feeling rested, without fatigue, continuing activities during the night. The onset of several nights of markedly reduced sleep accompanied by increasing energy is a particularly important historical feature.

Increased self esteem may range from greater self confidence to grandiosity. In severe presentations, the patient may believe that they possess exceptional abilities, powers, wealth or unrealistic roles. When such beliefs become delusional, the episode is severe by definition and requires urgent assessment of risk and the appropriate care setting.

Speech tends to become faster, more abundant and difficult to interrupt. The patient may speak continuously, talk over others and change topics rapidly. Flight of ideas may be reported as racing thoughts or observed as an accelerated succession of associations.

Distractibility causes continuous shifts of attention toward irrelevant external stimuli. Goal directed activity increases in parallel: the patient may start numerous projects, work without rest, intensify social or sexual activity and take on commitments that are incompatible with available time and resources.

An important consequence of activation is involvement in activities with a high potential for harm, including excessive spending, reckless investments, dangerous driving, gambling, risky sexual behavior, substance use or impulsive occupational decisions. Clinical assessment should specifically quantify the financial, legal, relational and health consequences.

Psychomotor agitation may become intense. In severe presentations, the patient may be intrusive, disorganized, aggressive or unable to cooperate. Before taking an extensive history, it is therefore necessary to establish whether there is an immediate risk to the patient or to other people.

Psychotic symptoms may be mood congruent or mood incongruent and include delusions and hallucinations. The presence of psychosis during a period of elevated mood distinguishes mania from hypomania. The temporal relationship between psychosis and mood disturbance is also fundamental in the differential diagnosis with schizoaffective disorder and schizophrenia.

Depressive symptoms such as dysphoria, guilt, hopelessness or suicidal ideation may coexist during mania. The DSM-5-TR uses the specifier with mixed features when sufficient symptoms of the opposite polarity are present without meeting criteria for a separate full episode. These presentations may be particularly unstable and associated with high clinical risk.

Depressive episodes in bipolar I disorder may have the same phenomenology as a unipolar major depressive episode, with depressed mood, anhedonia, changes in sleep and appetite, fatigue, psychomotor slowing or agitation, guilt, cognitive symptoms and thoughts of death. Observation of the current depression alone therefore cannot establish its bipolar nature.

The longitudinal history should reconstruct all previous episodes, their polarity, duration and intensity, any hospitalizations, psychotic symptoms, catatonia, suicidal behavior and response to medications. Periods of remission and the presence of residual symptoms should also be documented, because these may cause persistent disability even outside major episodes.

Family history of bipolar disorder, severe depression, psychosis and suicide should be explored. An early onset depressive episode, numerous depressive episodes, activation after antidepressants and periods of high energy with little sleep increase suspicion of bipolarity, but no single feature replaces clinical demonstration of a manic episode.

Use of alcohol, cannabis, cocaine, amphetamines, synthetic substances and medications should be reconstructed with precise attention to timing. Intoxication and withdrawal may mimic or worsen mood activation. Corticosteroids and other medications may also induce manic syndromes in susceptible individuals.

Mental status examination describes appearance, behavior, psychomotor activity, level of cooperation, speech, mood, affect, form and content of thought, perception, attention, orientation, insight and judgment. In classic mania, expansive or irritable affect, racing thoughts, talkativeness, distractibility and reduced insight are common.

Assessment must always include suicide risk. Suicide in bipolar disorder is associated particularly with depressive and mixed phases, but risk may also be present during mania because of impulsivity, agitation, psychosis or dangerous behavior. Ideation, intent, planning, previous attempts, access to means and protective factors should be investigated.

Risk to others, ability to manage finances, fitness to drive, vulnerability to exploitation, ability to care for oneself and capacity to provide informed consent should also be assessed. In more severe presentations, these issues may determine the need for a more protected level of care.

Physical and neurological examination does not diagnose bipolar disorder, but helps identify alternative causes, complications of the episode and conditions that influence treatment choice. Weight, body mass index, blood pressure and metabolic status are also important because many treatments may alter weight, lipids and glucose metabolism.

Assessment and Diagnosis

Diagnosis is clinical and longitudinal. There is no pathognomonic laboratory or instrumental test. The correct process begins by identifying a possible manic episode, continues with urgent safety assessment, verifies diagnostic criteria, excludes substances and medical conditions, reconstructs the course and finally defines the current episode, severity, features and comorbidities.

When the patient is agitated, psychotic, severely disorganized or dangerous, the priority is not to immediately complete every diagnostic detail but to ensure safety and stabilization. Once the emergency is controlled, direct history, collateral information and previous medical records make it possible to reconstruct the episode.

Essential DSM-5-TR criteria for a manic episode

• A distinct period of abnormally and persistently elevated, expansive or irritable mood and abnormally and persistently increased activity or energy, lasting at least one week and present for most of the day nearly every day, or of any duration if hospitalization is necessary.
• During the period of mood disturbance and increased energy, at least three symptoms are present among increased self esteem or grandiosity, reduced need for sleep, increased talkativeness or pressure to keep talking, flight of ideas or racing thoughts, distractibility, increased goal directed activity or psychomotor agitation, and involvement in activities with a high potential for harmful consequences; at least four symptoms are required if the mood is only irritable.
• The mood disturbance is sufficiently severe to cause marked impairment in social or occupational functioning, to necessitate hospitalization to prevent harm to self or others, or psychotic features are present.
• The episode is not attributable to the physiological effects of a substance or another medical condition.

For a DSM-5-TR diagnosis of bipolar I disorder, it is sufficient that criteria for at least one manic episode have been met and that the episode is not better explained by a schizophrenia spectrum disorder or another psychotic disorder. A major depressive episode is not required.

A full manic episode that emerges during antidepressant treatment and persists at a fully syndromal level beyond the physiological effect of treatment may count for diagnostic purposes. The simple appearance of one or two activation symptoms, such as irritability or agitation, is not sufficient to infer bipolarity.

ICD-11 uses diagnostic requirements that do not perfectly overlap with the DSM-5-TR and includes bipolar type I disorder in category 6A60. Diagnosis requires at least one manic or mixed episode, with specification of the current or most recent episode and remission status. Classification systems should be applied according to the system in use rather than by arbitrarily mixing their criteria.

The severity and features of the episode are defined clinically. Psychotic symptoms, mixed features, associated anxiety, catatonia, seasonal pattern and other specifiers provided by the diagnostic system should be documented when applicable.

The DSM-5-TR specifier with rapid cycling applies when at least four mood episodes meeting criteria for manic, hypomanic or major depressive episodes have occurred in the past year and are separated according to diagnostic rules. It does not simply mean mood fluctuations within the same day.

Questionnaires such as the Mood Disorder Questionnaire, Hypomania Checklist or Bipolar Spectrum Diagnostic Scale may help identify people who warrant further assessment, particularly when bipolarity is suspected in a patient presenting with depression. They should not be used as standalone diagnostic tests because of false positive and false negative results.

Differential diagnosis with bipolar II disorder is straightforward formally but may be difficult from the history: the presence of even one manic episode excludes type II and defines type I. It is therefore necessary to establish whether previous periods of activation reached the severity of mania or remained hypomanic.

Schizoaffective disorder and schizophrenia are considered particularly when psychotic symptoms are present. Psychosis confined exclusively to mood episodes is compatible with bipolar disorder; prolonged periods of psychosis in the absence of affective episodes instead require a different diagnostic formulation according to the specific criteria.

Attention deficit hyperactivity disorder may share distractibility, impulsivity and hyperactivity, but is characterized by a persistent neurodevelopmental course rather than distinct episodes of change from baseline. Borderline personality disorder may also produce affective instability and impulsivity; in bipolar disorder, however, diagnosis depends on the presence of syndromic episodes that can be clearly delimited in time.

Medical differential diagnoses include neurological and endocrine conditions and, especially, presentations secondary to substances or medications. Hyperthyroidism, some neurological disorders, corticosteroids and stimulant substances may produce activation symptoms. The selection of investigations depends on the history and physical examination.

There is no mandatory panel to confirm the diagnosis, but baseline safety testing is often needed before pharmacotherapy is started. Depending on the medication, this may include a complete blood count, electrolytes, renal function, liver function, TSH, blood glucose or glycated hemoglobin, lipid profile, weight, body mass index, blood pressure and a pregnancy test when relevant.

Before and during lithium treatment, renal and thyroid function, calcium and conditions affecting fluid and electrolyte balance should be assessed; the drug requires monitoring of serum concentrations. Valproate requires particular caution because of teratogenicity and reproductive toxicity, and contemporary recommendations strongly restrict its use in people with reproductive potential.

Neuroimaging and EEG are not routinely prescribed. They are indicated when atypical onset, age, neurological signs, seizures, cognitive decline, trauma or other findings suggest a neurological cause. Normal neuroimaging does not confirm bipolar disorder, and an abnormal finding must be interpreted in the clinical context.

The final diagnostic formulation should specify not only the name of the disorder but also the current or most recent episode, severity, any psychotic or mixed features, remission, recurrence pattern, suicide risk, comorbidities and functioning. These elements determine treatment strategy far more than the diagnostic label alone.

Treatment and Prognosis

Treatment should be organized according to the phase of illness: management of acute mania, treatment of bipolar depression and relapse prevention. Strategy varies according to severity, psychotic or mixed features, risk, comorbidities, previous responses, adverse effects, patient preferences and reproductive potential.

In severe mania, particularly when agitation, psychosis or risk are present, hospitalization may be necessary. Substances and medications with potential activating effects should be stopped when clinically appropriate, and intoxication, withdrawal or concurrent medical conditions should be treated.

The 2023 VA/DoD guideline suggests lithium or quetiapine as monotherapy options for acute mania and includes olanzapine, paliperidone, risperidone and additional antipsychotics or mood stabilizers among alternatives according to individual characteristics. CANMAT/ISBD considers several atypical antipsychotics, lithium and divalproex first line, either as monotherapy or, in more severe presentations, in combination.

When response to monotherapy is insufficient or mania is particularly severe, combining lithium or valproate with an antipsychotic may provide faster control and greater efficacy at the cost of more adverse effects. Antipsychotic selection should consider metabolic risk, extrapyramidal symptoms, sedation, prolactin, QT interval and previous response.

Lamotrigine is not a treatment for acute mania and should not be used for this purpose. Similarly, antidepressants are not a treatment for mania and are generally discontinued during a manic episode, taking the clinical context and risk of withdrawal into account.

Electroconvulsive therapy may be used in severe mania when a rapid response is required, when pharmacological treatment is ineffective or not tolerated, or when particular clinical conditions are present. It is an effective specialist intervention also in presentations with catatonia or severe impairment.

Bipolar depression requires a different strategy from unipolar depression. The VA/DoD guideline recommends quetiapine as monotherapy and includes cariprazine, lurasidone, lumateperone and olanzapine among evidence supported options. CANMAT/ISBD also includes specific strategies with lithium and lamotrigine, as monotherapy or in combination depending on the phase and previous responses.

The use of antidepressants in bipolar depression is controversial and should be much more cautious than in unipolar depression. They are not recommended as monotherapy in bipolar I disorder; when used in selected cases, they are generally combined with a mood stabilizing treatment and require monitoring for switching, mixed features and acceleration of the course.

After resolution of the episode, maintenance treatment is central because bipolar I disorder has a high tendency to recur. Lithium and quetiapine have strong evidence for preventing mania; lamotrigine is particularly supported for preventing depressive relapses. Other antipsychotics and valproate may be appropriate according to predominant polarity and previous response.

Lithium retains a fundamental role in long term treatment. In addition to its efficacy in relapse prevention, numerous observational and meta analytic data suggest a favorable effect on suicide risk. Treatment nevertheless requires serum monitoring and attention to renal and thyroid function, drug interactions, dehydration and toxicity.

Valproate is effective in several manic and maintenance presentations but has major reproductive safety concerns. Guidelines and regulatory authorities have progressively strengthened restrictions during pregnancy and in people with reproductive potential. Choice should be individualized and comply with current local regulatory requirements.

Second generation antipsychotics may cause weight gain, dyslipidemia, glycemic abnormalities and other metabolic effects. Follow up should therefore include physical health monitoring, including weight, blood pressure, blood glucose and glycated hemoglobin, lipids, physical activity, smoking and cardiovascular risk.

Psychoeducation is an integral part of treatment. Patients and, when appropriate, family members are helped to recognize prodromal signs, sleep changes, signs of relapse, medication effects and destabilizing factors. Good knowledge of the illness facilitates adherence and allows earlier intervention.

Structured psychological interventions, including cognitive behavioral therapy, family focused therapy and interpersonal and social rhythm therapy, may be added to pharmacotherapy especially during maintenance. They do not replace antimanic treatment in severe acute mania.

Regularization of the sleep wake rhythm, reduction of alcohol and substance use, and management of shift work or activities that cause marked sleep deprivation have an important clinical rationale. Exercise and lifestyle interventions also help reduce the cardiometabolic risk associated with the illness and its treatments.

Reproductive planning should be addressed as early as possible. Pregnancy and the puerperium are periods of high relapse risk, and some medications have important teratogenic profiles. Abrupt discontinuation of effective therapy may itself increase episode risk; every change should therefore follow an individualized assessment of the benefit risk balance.

Prognosis is heterogeneous. Many patients achieve substantial remission and good functioning with adequate treatment, whereas others experience frequent episodes, residual symptoms or interepisode impairment. The number of previous episodes, rapid cycling, comorbidities, substance misuse and poor adherence are associated with a more complex course.

Symptomatic remission does not always coincide with full functional recovery. Cognitive symptoms, occupational difficulties and relational problems may persist even between episodes. Long term treatment goals therefore include relapse prevention, reduction of suicide risk, social and occupational recovery and protection of physical health.

Complications

Suicide is one of the main causes of premature mortality in bipolar disorder. Risk is particularly elevated during depressive phases and episodes with mixed features, and in the presence of previous attempts, hopelessness, substance use comorbidity and poor support. No tool can predict suicidal behavior in an individual patient with certainty; assessment must be repeated over time.

Mania may cause acute consequences even without suicidal intent. Impulsivity, grandiosity and impaired judgment increase the risk of accidents, trauma, dangerous driving, risky sexual behavior, financial losses and legal problems. Such consequences may persist long after the episode has remitted.

Severe agitation and psychotic symptoms may create a risk of aggression, exposure to conflict and need for more intensive care. Management should prioritize de escalation, a safe environment and treatment of the underlying syndrome, without automatically equating the diagnosis with dangerousness.

Relapses are a central feature of the disorder. Repeated episodes may interrupt education, careers, relationships and autonomy and increase the burden on families. Failure to continue maintenance treatment after an effective response is one modifiable factor that may increase relapse risk.

Alcohol and other substance use disorders are common and may worsen adherence, impulsivity, sleep, suicidal behavior and treatment response. Their presence requires integrated treatment rather than separate management of the two problems.

Anxiety comorbidities are also common and may persist during mood remission. Residual anxiety and insomnia may increase disability and complicate treatment; benzodiazepines and other sedatives require caution because of dependence, tolerance and interactions.

Obesity, diabetes, dyslipidemia and cardiovascular disease are overrepresented in populations with bipolar disorder. Contributing factors include illness related factors such as smoking and sedentary behavior, metabolic effects of some treatments and disparities in access to prevention. Physical health should therefore be part of routine psychiatric follow up.

Some patients experience cognitive impairment in attention, executive function and memory even outside acute episodes. These deficits are modest on average and highly variable, but may affect work and autonomy and should not be mistaken for dementia without adequate assessment.

Adverse effects of treatment are an additional source of morbidity. Lithium toxicity, thyroid and renal abnormalities, metabolic effects of antipsychotics, severe skin reactions to lamotrigine and reproductive problems associated with valproate require specific monitoring and patient education.

Abrupt discontinuation of maintenance medications may be followed by relapse and, in the case of lithium, by a particularly rapid increase in recurrence risk in some patients. Any treatment change should therefore be planned, except in medical emergencies requiring immediate discontinuation.

Pregnancy and the postpartum period are phases of particular vulnerability. The risk of relapse, including postpartum mania or psychosis, may be high in people with bipolar disorder and must be balanced against fetal and neonatal medication risks. Specialist multidisciplinary planning is indicated.

The overall impact of the disorder is therefore not limited to acute episodes. The combination of recurrence, suicide risk, psychiatric and physical comorbidity, treatment effects and interepisode disability requires a continuous and preventive model of care focused on the person rather than solely on resolution of the current episode.

Bibliography
1. American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition, Text Revision (DSM-5-TR). American Psychiatric Association Publishing; 2022.
2. World Health Organization. Clinical descriptions and diagnostic requirements for ICD-11 mental, behavioural and neurodevelopmental disorders. World Health Organization; 2024.
3. Yatham LN et al. Canadian Network for Mood and Anxiety Treatments and International Society for Bipolar Disorders 2018 guidelines for the management of patients with bipolar disorder. Bipolar Disorders. 20(2), 2018: 97-170.
4. Keramatian K et al. The CANMAT and ISBD Guidelines for the Treatment of Bipolar Disorder: Summary and a 2023 Update of Evidence. Focus. 21(4), 2023: 344-353.
5. Department of Veterans Affairs et al. VA/DoD Clinical Practice Guideline for the Management of Bipolar Disorder. Version 2.0. Department of Veterans Affairs and Department of Defense; 2023.
6. National Institute for Health and Care Excellence. Bipolar disorder: assessment and management. Clinical guideline CG185. National Institute for Health and Care Excellence; 2014, updated 2025.
7. Singh B et al. Bipolar disorder. The Lancet. 406(10506), 2025: 963-978.
8. McIntyre RS et al. Bipolar disorders. The Lancet. 396(10265), 2020: 1841-1856.
9. McIntyre RS et al. The clinical characterization of the adult patient with bipolar disorder aimed at personalization of management. World Psychiatry. 21(3), 2022: 364-387.
10. Carvalho AF et al. Bipolar disorder. New England Journal of Medicine. 383(1), 2020: 58-66.
11. O’Connell KS et al. Genomics yields biological and phenotypic insights into bipolar disorder. Nature. 639, 2025: 968-975.
12. Mullins N et al. Genome-wide association study of more than 40,000 bipolar disorder cases provides new insights into the underlying biology. Nature Genetics. 53(6), 2021: 817-829.
13. Ching CRK et al. What we learn about bipolar disorder from large-scale neuroimaging: findings and future directions from the ENIGMA Bipolar Disorder Working Group. Human Brain Mapping. 43(1), 2022: 56-82.
14. Cipriani A et al. Comparative efficacy and acceptability of antimanic drugs in acute mania: a multiple-treatments meta-analysis. The Lancet. 378(9799), 2011: 1306-1315.
15. Miura T et al. Comparative efficacy and tolerability of pharmacological treatments in the maintenance treatment of bipolar disorder: a systematic review and network meta-analysis. The Lancet Psychiatry. 1(5), 2014: 351-359.