Pediatric bipolar disorder (PBD) is a serious psychiatric illness that impairs children’s emotional, cognitive, and social development. PBD causes severe mood instability that manifests in chronic irritability, episodes of rage, tearfulness, distractibility, grandiosity or inflated self-esteem, hypersexual behavior, a decreased need for sleep, and behavioral activation coupled with poor judgment. While research in this area has accelerated during the past 15 years, there are still significant gaps in knowledge concerning the prevalence, etiology, phenomenology, assessment, and treatment for PBD.

This article briefly summarizes the scientific evidence that has contributed to our understanding of this disorder. The research literature in the areas of prevalence, etiology, pathophysiology, assessment, diagnosis, and treatment is reviewed.

Prevalence of PBD

Unfortunately, there are still no authoritative data on the prevalence of PBD: estimates depend on whether the disorder is defined as a narrow or broad phenotype.¹ Children with the narrow phenotype fit symptom criteria for a bipolar disorder diagnosis as defined by DSM-IV-TR, whereas children with the broad phenotype experience serious mood dysregulation and associated symptoms but may not meet symptom number or duration criteria defined by the narrow phenotype.

One community study showed a lifetime prevalence of bipolar disorder of 1% in youths aged 14 to 18 years, using the Schedule for Affective Disorders and Schizophrenia for School-Age Children (K-SADS).^2,3 However, Brotman and colleagues4 found the lifetime prevalence of severe mood dysregulation to be 3.3% in children aged 9 to 19 years from an epidemiological study sample. These findings indicate that a high percentage of the population may experience symptoms consistent with the broad phenotype of PBD.

Retrospective studies of adults with bipolar disorder have reported that as many as 60% experienced symptoms before age 20 years and 10% to 20% reported symptoms before age 10 years.^5-7 It remains unclear, however, how subthreshold symptoms in childhood relate to adult-onset bipolar disorder, as well as whether there is continuity between the childhood-onset presentation and the more classic presentation of adult bipolar disorder.

Geller and colleagues8 showed continuity in both disorder presence and nature of symptoms in children with bipolar I disorder who were observed for 8 years into adulthood. The findings from their study indicate that the unique symptom presentation of early-onset bipolar disorder continues into adulthood for these children.

Despite a lack of knowledge on exact prevalence and the continuity or overlap between childhood-onset and adult-onset presentations, the psychosocial and interpersonal effects of PBD (whether broad or narrow phenotype) on patients and their families is devastating. A community study showed that PBD is associated with substantial impairment in social (66%), family (56%), and school (83%) functioning.² PBD has been associated with behavioral problems in school, low grades, having few or no friends, frequent teasing, poor social skills, poor sibling relationships, and parent-child relationships characterized by frequent hostility and conflict.^9,10

Etiology and pathophysiology

Heritability for PBD is substantially high with a rate of 15% to 42% risk in first-degree relatives of children with PBD.11 There is some evidence to indicate gene linkages with the brain-derived neurotrophic factor (BDNF) gene (Vall66), the glutamate decarboxylase 1 (GAD1) gene (4s2241165), and the dopamine transporter gene (rs41084). However, none of these associations has been replicated and the sample sizes in these studies significantly limit the detection of small effects.11 Furthermore BDNF in platelets has been shown to be decreased in PBD and increased with treatment.¹²

Although the pathophysiology of PBD remains unknown, MRI studies during the past 5 to 10 years have significantly advanced scientists' understanding of the brain mechanisms related to PBD symptoms and associated impairments. Structural and functional studies have all shown frontostriatal and frontolimbic abnormalities relative to healthy controls. A consistent structural finding is a small amygdala.¹³ There are extensive abnormalities in other related brain regions; overall findings implicate abnormalities in regions of the cognitive, emotional, and reward systems.^14,15 White matter tract integrity is affected in the prefrontal regions of the corticobulbar tracts in patients with PBD relative to healthy peers.¹⁶

A 3-year longitudinal study demonstrates that neurocognitive deficits persist in PBD and show no change during the 3 years in the areas of executive function and verbal memory. While there is improvement in working memory and attention, children with PBD do not catch up with their healthy peers.¹⁴ It is important to evaluate whether there are academic difficulties related to poor neurocognitive abilities.

Assessment and diagnosis

Assessment and diagnosis of PBD have also advanced rapidly during the past 15 years. Whereas the notion of diagnosing bipolar disorder in childhood has been plagued by controversy in the past, the literature base to date has led to a consensus that PBD represents a discrete cluster of symptoms that can be validated by reliable assessment.¹⁷ The question remains as to how to best classify and describe the heterogeneity in clinical presentation seen in children with early-onset bipolar disorder.

The diagnosis of PBD is complex and requires an in-depth evaluation of the current presenting problem, as well as a history of mood episodes during the child's lifetime to assess chronicity and episodicity. A clinical interview for making a diagnosis is still the gold standard. In research, however, a diagnostic clinical interview (eg, Washington University in St Louis Kiddie Schedule for Affective Disorders and Schizophrenia [WASH-U-KSADS] or Kiddie­SADS­Present and Lifetime Version [K­SADS­PL]) is often supplemented by an in-depth developmental history and multi-informant clinical rating scales to determine global functioning, social and academic impairment, and quality­of­life issues.

Screening measures can be used preliminarily to assess the presence of different types of manic symptoms. The Child Mania Rating Scale­Parent version (CMRS­P) is the first rating scale that was designed and tested specifically to screen for PBD.18 This measure has 21 developmentally specific items corresponding to DSM­IV­TR­based symptoms. A score of 15 or more (of a potential total score of 63) indicates a 92% chance of having PBD. The CMRS-P is a screening instrument with excellent psychometric properties and is available in 12 languages. It is intended to facilitate early identification of PBD with the hope of aiding parents who seek help for their child, but it is not a diagnostic tool.

Once a formal diagnosis of PBD is made, additional evaluation is needed to address systemic problems through psychosocial treatment that complements pharmacotherapy. Issues include:

• The level of distress of the patient and the family
• Knowledge, skills, and attitude of parents
• The child's school functioning and neurocognitive problems

Bipolar I disorder is diagnosed if a child manifests abnormally elevated mood, grandiosity or inflated self-esteem, extreme irritability, hypersexuality (not explained by a history of sexual abuse), decreased need for sleep, and behavioral activation (characterized by agitation and pressured speech) coupled with poor judgment. Elevated mood, grandiosity, and irritability are the core symptoms of mania. While there is some debate about which of these is most central to a bipolar I disorder diagnosis, irritability interspersed with rage episodes tends to be the most readily identified and problematic for parents.

The 2009 Roundtable on Bipolar Disorder in Children and Adolescents recommends that a diagnosis be based on DSM-IV criteria, which require that children demonstrate 1 core symptom and 3 additional symptoms (4 if core symptom is irritability).¹⁷ The symptoms must represent a deviation from the child's baseline and from age-appropriate behavior. The episode must meet the duration criteria of 7 days for mania and 4 days for hypomania, and symptoms must persist for at least 4 hours each day.

The differentiation of elation and grandiosity from developmentally normative silliness and self-aggrandizement can be challenging. Other mania symptoms, such as hypersexuality and impulsivity, may also present in unique ways, depending on the child's developmental stage. Symptom expressions such as inflated self-esteem and increased goal-directed activity are best judged in the context of the child's history because behaviors may be misleading in isolation and must be considered in the context of the child's cognitive, biological, and social development. If several of the symptoms exist together, however, and are episodic with impairment in functioning, then PBD should be considered.

Bipolar II disorder presents with major depressive episodes that alternate with hypomanic episodes; however, the depressive episodes are central to the clinical picture and cause significant dysfunction. It is often hard to characterize hypomania in youth, given developmentally appropriate excitability.

Youths with bipolar disorder not otherwise specified (NOS) either have fewer symptoms or have symptoms of shorter duration or less intensity than those required to meet full DSM-IV-TR diagnostic criteria. As many as 35% of children with onset at a younger age meet criteria for bipolar disorder NOS.^19,20 To distinguish between bipolar I disorder and bipolar disorder NOS, it is important to clearly establish cycling or episodicity or both and not to confuse episodes with cycles.

The discovery of the unique phenomenology of PBD led to the description of ultradian cycling in children, defined as cycling at least once within a day (or more than 365 cycles a year).

Rapid cycling as defined by DSM-IV-TR should be differentiated from the rapid cycling variants of bipolar disorder as characterized by Geller and colleagues.²¹ In DSM-IV-TR, rapid cycling is defined as having more than 4 episodes of depression or mania or mixed state during a 1-year period. In phenomenological studies conducted by Geller and colleagues,²² parents often described mood dysregulation in their young children as rapid mood swings, sometimes multiple times a day. This discovery of the unique phenomenology of PBD led to the description of ultradian cycling in children, defined as cycling at least once within a day (or more than 365 cycles a year).

The discrete episodes (of mania, mixed symptoms, hypomania, and depression) with respite in between that are seen more often in the adolescent and adult-onset presentation clearly exist in some children, but they are are less common. In children, respite between episodes is less common, and often children experience subsyndromal symptoms or chronic impairment. Symptom respite in between recurrent episodes depends on severity of illness, availability of proper treatment, and the ability of the family to successfully manage the illness.

Mixed episodes and comorbid attention-deficit/hyperactivity disorder (ADHD), oppositional defiant disorder, and anxiety are common in PBD. A good rule of thumb for determining whether a symptom represents true comorbidity or is part of bipolar disorder is to evaluate whether it appears to be rooted in mood disturbance (ie, whether it intensifies with mood fluctuation). Also, bipolar disorder NOS evolves into bipolar I or II disorder in some children as they mature. One study found that bipolar disorder NOS evolved into bipolar I or II disorder in 38% of youths during a 4-year period.²⁰

It is critical to define the "signature of the illness" for each child. In addition, the pattern of symptom experience, onset and offset, and timeline of episodes help establish the diagnosis. A comprehensive developmental history is essential to understand the emergence of symptoms in a developmental context. For example, there is some evidence that PBD diagnosis may be preceded by a difficult temperament style in infancy and toddlerhood compared with children without PBD.²³

Comorbid ADHD is diagnosed if symptoms precede the onset of PBD symptoms or if the child has very early-onset PBD with symptoms of ADHD that do not resolve after successful treatment of the PBD; however, it is possible that these symptoms represent residual symptoms of intrinsic overactivity and inattention observed in PBD. A family history of bipolar disorder, given the familial risk, adds value in considering the diagnosis.²⁴ However, that family history should not be considered as a diagnostic feature. Conversely, the absence of a family history does not rule out the diagnosis of PBD.

A practical approach to therapy

Evidence suggests that a combination of atypical antipsychotics or mood stabilizers or both is effective in the treatment of mania in PBD.²⁵ Some children may need to have medications added to the primary mood stabilizer to treat comorbid symptoms of attention deficit, aggression, anxiety, psychosis, and sleep disturbance; however, efficacy results remain mixed.1

Pavuluri and colleagues²⁶ integrated the current evidence updated from the previously published evidence-based pharmacotherapy algorithm. The basic tenets of this approach are:

• Prescription hygiene (ie, discontinuation of medications that are worsening the clinical state)
• Mood stabilization
• Overcoming obstacles to mood stabilization by addressing breakthrough symptoms
• Problem solving for comorbid conditions or adverse medication effects or both

A pharmacotherapeutic plan for mood stabilization consists of 2 parts. The first is a history of medications that worsened the patient's clinical status, those that were ineffective, and those that were transiently useful or helpful; the second is the discontinuation of all ineffective medications.

There is no evidence to show which patients worsen with stimulants and which do not. Mood stabilization is the primary treatment objective in PBD; controlling symptoms of ADHD is secondary. However, if the patient's parents report that stimulants have been helpful and there is a clear pattern of response independent of affect dysregulation, a long-acting stimulant at the lowest dose may be prescribed.²⁶

The treatment of choice for mood stabilization is either an atypical antipsychotic, such as risperidone or aripiprazole - which has been FDA­approved for children as young as 10 years - or mood stabilizers, such as lithium (FDA­approved for youths older than 12). Data on quetiapine, olanzapine, and ziprasidone have been presented to the FDA; these agents may be used off label for children and adolescents aged 10 through 17 years.²⁵

Because a single drug may not be effective or its onset of action may be slow, atypical antipsychotics are being used as monotherapy when mania stabilization is a priority. These agents may also be used in combination with a mood stabilizer.²⁶ Atypical antipsychotics alone may be effective when irritability is prominent and demands a faster response not possible with first-line mood stabilizers.²⁶

A combination of an atypical antipsychotic plus lithium or divalproex is an effective first-line strategy for severe cases, especially those with psychotic features.^27,28 With this approach, doses of atypical antipsychotics need to be lower than those often required during monotherapy. Thus, if adverse effects occur, they are far less severe. While the clinician excludes medications that failed to be effective, he or she can choose the next best option in the list either as an alternative monotherapeutic agent or as a combination regimen.

PBD presents a multitude of clinical challenges beyond acute mood stabilization that must be factored into both the acute and maintenance phases of treatment. If there are prominent symptoms of depression, lithium and lamotrigine are chosen as primary mood stabilizers - either alone or adjuvant to other partially effective agents.^29,30 Lamotrigine must be titrated very slowly at the rate of 12.5 mg each week to an optimal dose of 200 mg twice daily.³¹ In circumstances of severe depression, an SSRI in small doses of 2.5 to 5 mg in a time­limited manner, under close supervision, and with psychoeducation may be effective in combination with a mood stabilizer.³² It is important to balance the risks versus the benefits given the black box warning associated with SSRI use in children.

In cases of psychosis, atypical antipsychotics must be added after working down the list of agents that worked previously in a particular patient.^26,33 If mild aggression is present, switching to atypical antipsychotic monotherapy may be helpful. For moderate to severe aggression, the combination of a mood stabilizer and an atypical antipsychotic may be beneficial. Clonidine can be used to subdue uncontrolled rage attacks.^34,35 However, in our experience, children can become disinhibited or become more aroused after persistent use of clonidine, although this particular observation needs to be further examined.

If sleeping difficulties arise, the clinician can increase the evening dose of a sedating mood stabilizer. Beyond that, 1 to 3 mg of melatonin or 25 to 50 mg of trazodone can be administered to establish a sleep routine, which is critical in managing PBD.^36-38 While these compounds are not empirically supported by specific research on sleep in PBD, they are known to be sedating and safe in pediatric populations. Also, they interfere minimally with rapid eye movement sleep. Because of abuse potential, benzodiazepines are not recommended.³⁹

Since the start of our algorithm project and the subsequent publication of our feasibility study, we have continued to update our strategies and tactics on the basis of new information.²⁶ Aripiprazole has been added to the list of atypical antipsychotics. Lamotrigine has been elevated as a first-line agent for depression. Atomoxetine was added as a second-line medication after trying stimulants for comorbid ADHD.

In our experience, clonidine has been shown to exacerbate symptoms in a subgroup of patients with PBD despite excellent short-term response for autonomic arousal. We are closely monitoring this phenomenon. Consequently, we are currently choosing guanfacine (given the longer half-life than that of clonidine) or propranolol as an alternative for extreme hyperarousal that does not respond to mood stabilization. Given the recent review presenting equivocal evidence of trazodone's efficacy as sleep medication, this agent was placed lower on the list after medications such as melatonin that had better safety and efficacy data in idiopathic insomnia, even though trazodone was not tested for sleep problems directly in patients with PBD.^36,40 Chronic unremitting symptoms must be treated by using:

• Alternative monotherapy
• At least 2 trials of combination mood stabilizers plus an atypical antipsychotic
• Moving on to triple therapy and addressing comorbidities (eg, an additional stimulant for comorbid ADHD)

Comorbidities

While ADHD is a distinct disorder separate from PBD, it is not understood whether the ADHD-like symptoms in children with PBD warrant additional treatment beyond mood stabilization. In our study, several participants continued to show symptoms of inattention after mood stabilization that warranted stimulant medication.²⁶ Cognitive difficulties, such as shifting attention and executive dysfunction, seen in both ADHD and PBD, potentially can be addressed with stimulants. Stimulants are almost always given in long-acting formulations unless an additional after-school dose is required to sustain the benefits.

Among psychostimulants, long-acting methyl-phenidate and mixed amphetamine salts are equally effective.41 Atomoxetine is an alternative treatment if stimulants have been ineffective or have not been tolerated. No data establish the safety or efficacy of atomoxetine in treating youth with comorbid ADHD and PBD. Atomoxetine is a selective norepinephrine reuptake inhibitor with potential antidepressant effects and could theoretically trigger or exacerbate symptoms of mania in patients with PBD. Atomoxetine should be used with great care in youth with PBD.

Anxiety disorders, including generalized anxiety disorder and separation anxiety disorder, are relatively common, especially in those with bipolar I disorder. Psychotherapeutic interventions, such as cognitive-behavioral therapy (CBT), remain the first choice of treatment in children and adolescents with comorbid PBD and anxiety disorder. Small doses of SSRIs - such as escitalopram XR as adjuvant medication - may be effective if mania is stabilized, although there are no controlled trials for anxiety comorbid with bipolar disorder. SSRIs are the only medications consistently shown to be effective in controlled trials for childhood anxiety disorders.^42,43

SSRI treatment intervention requires educating the family about the risk of a manic switch, and close monitoring of the treatment response is necessary. Guanfacine may be considered if vigilance and autonomic hyperarousal are prominent.⁴⁴ Benzodiazepines and buspirone are alternative choices.²⁶ The risk of developing dependence needs to be considered for long-term use of benzodiazepines in adolescents. Buspirone may not be effective in all cases. Propranolol may be considered in cases of performance anxiety. Medication is often used in small doses to reduce risks of exacerbating bipolar disorder and to enable patients to benefit from psychotherapeutic interventions.

Adverse effects

Lower doses and slow titration are fundamental principles used to minimize the occurrence of treatment-emergent adverse effects. If problems continue, switching to an alternative medication may be necessary. For persistent adverse effects in the context of good treatment response, lower the dose of the medication before switching medications or use pharmacological management strategies. When a risk-benefit analysis indicates that the offending drug needs to be retained at the desirable therapeutic dose, strategies must be implemented to manage adverse effects.

Despite several antidotes for weight loss, the single most important intervention is diet and exercise. If possible, consultation with a dietician is helpful. In our experience, simple weight management programs (eg, Weight Watchers) have been successful but should only be undertaken after consultation with a health professional. Timely meals and wise food choices cut down excessive calories.

Switching from a medication that is causing weight gain to another that is less likely to have this effect is warranted. Among the atypical antipsychotics, ziprasidone causes the least weight gain and olanzapine causes the most.²⁵ It is important to check the metabolic profile every 3 months.

Weight gain from lithium may partially be the result of increased consumption of high calorie soft drinks or juice to compensate for the thirst caused by this medication. Limiting fluid intake to low calorie drinks is an easy way to prevent unnecessary weight gain.

Benztropine 0.5 to 2 mg once a day or once every 2 days is effective in combating extrapyramidal symptoms. Akathisia in patients treated with an atypical antipsychotic is sometimes missed and often responds to low doses of propranolol. Nighttime dosing decreases problems with sedation. In the event of residual morning somnolence, the evening dose may be moved earlier in the evening.

GI upset from divalproex is dose-related. Administering the drug with food or using a long-acting formulation or both may decrease GI upset. Taking an atypical antipsychotic with a small snack 30 minutes to 1 hour before bed may decrease GI effects.

GI upset from lithium is also dose-related; management depends on whether the intolerance is specific to the upper or lower GI tract. Upper GI tract effects from lithium tend to be associated with high doses that directly irritate the stomach mucosa and with high peak serum levels. Switching to a sustained-release formulation or dividing doses twice to three times a day and/or administering the drug with food will reduce GI upset.

Lower GI track effects such as diarrhea are correlated with high doses and high drug serum levels. This may also occur secondary to residual lithium left in the large intestine, which may create an osmotic effect, drawing excess water into the lower GI tract, resulting in diarrhea. Lower GI tract effects may be managed by reducing the overall dosage or by switching from sustained-release formulations to immediate-release tablets, capsules, or liquid formulations. Dividing high every day doses of immediate-release formulations to twice or three times a day may also help.

Elevations in thyroid-stimulating hormone (TSH) levels occur in approximately 15% of patients. This usually occurs within the first 4 weeks of therapy; levels may normalize after a transient elevation although they may persist for up to 4 years. In accordance with standard lithium monitoring guidelines, baseline monitoring of free thyroxine and TSH and then follow-up at 1 month, 6 months, then yearly until the fourth year, is recommended. Additional monitoring may be needed after substantial dosage increases. If hyperthyroidism occurs and lithium needs to be continued, this may be effectively treated with levothyroxine titrated on the basis of the levels of TSH.

Psychosocial treatment

Despite the obvious need for psychosocial interventions to address the complex presentation of symptoms in PBD and the effect of the disorder on peer interaction and school and family functioning, there are few evidence-based psychosocial treatment methods. Fristad and colleagues⁴⁵ developed a multiple family group treatment for parents and youths aged 8 through 12 years who have bipolar and depressive spectrum disorders. This modality focuses primarily on psychoeducation. The goals include teaching parents and children about the child's illness and its treatment, symptom management, and improving problem-solving and communication skills. Study results suggest that the multifamily psychoeducation groups approach is effective in improving symptoms and social and family functioning.

Miklowitz and Goldstein46 adapted their manual-based, family-focused therapy for adults with bipolar disorder to adolescents.⁴⁷ Such therapy is based on the premise that patients with bipolar disorder will experience symptom relief as they acquire a greater awareness of how to cope with the disorder; there will also be decreased levels of expressed emotion from caregivers and better family problem-solving and communication skills. A recently published 2-year randomized clinical trial of such family-focused therapy for adolescents with bipolar disorder indicates that those who received the intervention had shorter times to recovery from depression, shorter depressive episodes, and lower depression severity scores for 2 years following treatment.⁴⁸

A few alternative treatments for adolescents with bipolar disorder are being developed or adapted. These include interpersonal and social rhythm therapy and dialectical behavioral therapy.^49,50

CBT is designed to meet the specific developmental needs of children with PBD and their families. Child- and family-focused CBT exists in both individual and group therapy formats and offers a clear theoretical basis of amalgamating CBT principles and interpersonal psychology principles. There is an intense focus on parent training and therapy along with child work to affect both symptom experience and psychosocial and interpersonal functioning. Essential ingredients of child- and family-focused CBT include maintaining a routine, affect regulation, self-efficacy boosting, cognitive restructuring, social skills building, life skills training, problem solving, and social support enhancement. Research suggests that this form of CBT may improve a child's symptom experience and psychosocial functioning as well as parental attitudes and coping and that these changes may be sustained over time through maintenance child- and family-focused CBT.⁵¹

Conclusion

PBD affects cognitive, behavioral, and affective domains of a child's well-being. Affect dysregulation is the central feature of this disorder. It is not clear that PBD is contiguous with the adult variant of bipolar disorder. The primary goal of pharmacotherapy is mood stabilization with a problem-solving approach to deal with complex comorbid, residual, breakthrough, and associated symptoms. It is imperative that any medication management be coupled with a meaningful holistic therapeutic approach that is practical and tailored to the patient. A comprehensive treatment approach includes evidence-based psychosocial intervention and ongoing parental and family support by linking them to community resources, including other families affected by PBD.

Go to posttest

Go to posttest