Management of Bipolar Disorder in Children and Adolescents

The initial step in the successful treatment of children and adolescents with bipolar disorder is making an accurate diagnosis. This remains a significant challenge even for experienced clinicians. Controversies regarding phenomenological presentation and the validity of applying adult-derived criteria to children and adolescents still remain today. Beyond establishing an accurate diagnosis, well-established developmental differences in the presentation between youth and adults further complicate the treatment and management (1). One example is that adults exhibit distinct episodes of depression and mania, whereas the pattern of illness observed in youths is often characterized by mixed or dysphoric mood states accompanied by irritability (2). Moreover, children and adolescents may be more susceptible to rapid cycling and experience more symptomatic periods (3). Most patients with pediatric bipolar disorder also have comorbid or co-occurring conditions, most commonly attention deficit hyperactivity disorder (ADHD) (4). Furthermore, there is a large gap between the data available regarding medication safety and efficacy in adults and children with bipolar disorder. Growing evidence suggests that merely extrapolating from adults is insufficient.

In the midst of the challenges of recognizing and managing bipolar disorder in children and adolescents, it is also clear that children and adolescents with bipolar disorder require prompt treatment to ameliorate symptoms and to reduce the psychosocial morbidity that accompanies the illness. In fact, early detection and treatment seems paramount considering earlier onset and longer duration of illness is associated with poor rates of recovery (5).

Treatment of pediatric bipolar disorder requires a multimodal approach that includes pharmacologic and psychosocial interventions. This article reviews the empirical evidence regarding the efficacy and safety for both pharmacological and psychosocial interventions for bipolar disorder in children and adolescents, summarizes the recent progress of examining these issues, and highlights future challenges.

Empirical Evidence of the Pharmacological Treatment of Bipolar Disorder in Children and Adolescents with Manic Episodes

For the purpose of this article, response and remission rates are reported as indicated in the original publications. Definitions for response and remission vary across different studies, however, and differ based on the scale used and/or the cutoff used (e.g., percentage change and absolute score).

Lithium

The first publication of lithium’s use in adults with bipolar disorder dates back more than 60 years (6). Studies of lithium use in children started to emerge in the medical literature 50 years ago (7). After the first publication of lithium use in children, several case reports and case series that included a mix of patients described as having bipolar disorder followed. A handful of double-blind, randomized, placebo-controlled (DBRPC) crossover trials in children and adolescents tested lithium’s use in patients with a mélange of vaguely defined clinical syndromes that may, or may not, have been pediatric bipolar disorder (i.e., “severely hyperactive children with mood swings” [8], “episodic mood and behavior disturbance” [9], “behaviorally disordered children of parents who are lithium responders” [10]). Although many of these studies yielded positive results, the diagnostic ambiguity limits the conclusions that can be drawn. Despite the limited data, the U.S. Food and Drug Administration (FDA) approved lithium for the treatment of mania in children 12 years and older.

Open-label (OL) studies, with more homogeneously defined diagnostic criteria, have suggested that children treated with lithium achieve a response rate between 38% and 55% as monotherapy and 82% when used in combination with second-generation antipsychotics (SGAs) (e.g., risperidone) (11–13). In one OL lithium study of 100 adolescents (12–18 years of age) with an acute manic episode, 63% achieved response and 26% stayed in remission (14). Furthermore, in a double-blind placebo-controlled discontinuation study, 62% of patients who initially responded to lithium and switched to placebo experienced clinically significant exacerbation of symptoms, while 53% of those who responded and remained on lithium had symptomatic relapse; however, this difference was not statistically significant (15).

Furthermore, there is some evidence suggesting that lithium response may run in families (16) and that lithium may be particularly effective in preventing suicidal behavior (17). Common adverse effects associated with lithium include nausea, headache, acne, weight gain, thyroid dysfunction, diabetes insipidus, and tremor. Baseline complete blood counts, thyroid panels, blood urea nitrogen, creatinine, serum calcium, urinalysis, and pregnancy test are recommended prior to initiation of lithium, as well as every 3–6 months thereafter.

Antiepileptic Medications

Carbamazepine.

Despite its track record in adults, carbamazepine has been studied in only a few OL trials in youth with bipolar disorder. A 6-week OL comparison of carbamazepine with valproate and lithium showed a modest response rate of 38% for carbamazepine, 38% for lithium, and 53% for valproate; these differences were not statistically significant (12). Additionally, an OL study evaluated carbamazepine extended release and showed a similar response rate of 44% (18). To date there are no DBRPC trials of carbamazepine for the treatment of pediatric bipolar disorder.

The most common adverse events with carbamazepine include nausea and sedation (19). A baseline complete blood count and follow-up measurements should be obtained since carbamazepine has been associated with agranulocytosis and aplastic anemia (20). In addition, carbamazepine increases the risk for developing Stevens-Johnson syndrome in patients of Asian ancestry; therefore, it is recommended that such patients should be screened for the HLA-B* 1502 allele (21).

Topiramate.

Although studies in adults do not support the efficacy of topiramate as monotherapy in acute mania or mixed episodes (22), there are some data suggesting topiramate might have some effect on manic symptoms in youth. Two retrospective studies suggested that adjunctive topiramate reduces the severity of bipolar mania in both inpatient and outpatient children (23, 24). A 4-week, DBRPC trial of topiramate in pediatric mania was prematurely terminated by the sponsor, because of negative results in adults (25). Common adverse effects reported in these studies included decreased appetite, nausea, and weight loss.

Topiramate use has also been explored in preventing weight gain associated with certain SGAs. An 8-week OL study of olanzapine monotherapy compared with olanzapine plus topiramate showed that both treatments were capable of reducing symptom scores and that those patients with topiramate plus olanzapine combination had lower weight gain than the group only treated with olanzapine (26).

Lamotrigine.

There are only two OL studies of lamotrigine monotherapy for manic symptoms in youth (27, 28). In an OL 14-week study, 46 children (6–17 years) received lamotrigine, which was slowly up-titrated for 8 weeks. In this study, lamotrigine appeared to be effective in managing manic symptoms in pediatric bipolar disorder with a response rate of 72% and a remission rate of 56%. In another OL study, lamotrigine was found to produce a response rate of 54%. Common adverse effects reported in this study included gastrointestinal symptoms, headaches, and skin rashes. Although rare, a serious side effect of lamotrigine in youth is Stevens-Johnson syndrome (SJS). The incidence of serious rashes, which includes Stevens-Johnson syndrome, is approximately 0.8% (8 per 1,000) in pediatric patients (2–16 years of age) receiving lamotrigine as adjunctive therapy for epilepsy and 0.3% (3 per 1,000) in adults receiving adjunctive therapy for epilepsy. In clinical trials of bipolar and other mood disorders, the rate of serious rash was 0.08% (0.8 per 1,000) in adult patients receiving lamotrigine as initial monotherapy and 0.13% (1.3 per 1,000) in adults receiving lamotrigine as adjunctive therapy (lamotrigine product label). The rate of rashes in youth with bipolar disorders is unknown.

Valproate.

The response rate for valproate in OL studies of youth with mania has been between 53% and 75% (29–31). However, more recent reports suggest that valproate may not be efficacious for the treatment of acute mania in pediatric bipolar disorder. One study comparing valproate and quetiapine reported larger response (84% versus 56%; p<0.05) and remission rates (60% versus 28%) in the quetiapine group than in the valproate group (32).

A second study suggested that risperidone is superior and better tolerated than valproate (33). This DBRC trial, reported greater response (78.1% for risperidone versus 45.5% for valproate; p<0.01) and remission rates (62.5% versus 33.3%; p<0.05) for risperidone compared with valproate. Subjects taking valproate dropped out because of adverse events, with irritability being the most common.

In a third randomized controlled study, valproate extended release failed to separate from placebo (34) for the treatment of mania in youth. In this well-powered study, there were no significant group differences in Young Mania Rating Scale (YMRS) change from baseline, nor in response (24% versus 23%) or remission rates (16% versus 19%).

Common side effects of valproate include gastrointestinal symptoms, sedation, and weight gain. Pancreatitis, hepatotoxicity, alopecia, thrombocytopenia, and polycystic ovary syndrome have also been associated with this medication. Recommendations for laboratory monitoring include obtaining complete blood counts, liver function tests, and pregnancy screening, at baseline and every 6 months thereafter, as well as, careful monitoring of menstrual cycles in girls who are prescribed valproate sodium.

Second-Generation Antipsychotics (SGAs)

Aripiprazole.

Aripiprazole is an SGA with U.S. FDA approval for the treatment of mixed or manic episodes in patients aged ≥10 years. Empirical evidence of the efficacy and safety of aripiprazole is available in several reports, including retrospective chart reviews (35), OL studies (36, 37), and two DBRPC trials (38, 39). In both DBRPC trials, aripiprazole was significantly superior to placebo in decreasing manic symptoms, response rates, and remission rates. Response rates were 44.8% in the 10-mg group, 63.6% in the 30-mg group, and 88.9% in the flexible dose study. Remission was achieved by 25%, 47.5%, and 72% in the 10-mg, 30-mg, and flexible groups, respectively. Common adverse events reported in these studies included sedation, gastrointestinal complaints, cold symptoms, headache, extrapyramidal symptoms, and akathisia.

Olanzapine.

Olanzapine is currently U.S. FDA approved for use for mania in adolescents aged ≥13 years. The empirical evidence of olanzapine’s efficacy in the treatment of children and adolescents with bipolar disorder in acute mania comes from four OL studies (26, 40–42) and a DBRPC trial (43). Response rates for olanzapine were 74%, 61%, 33%, and 47% in the OL studies. In the DBRPC trial, olanzapine was superior to placebo with a response rate of 45% (versus 22% for placebo) and a remission rate of 35% (versus 11.1% for placebo).

Common side effects included sedation, appetite increase, metabolic changes, and weight gain. In the 3-week DBRPC study, adolescents gained an average of 7.4 kg compared with 3.2 kg in adults. Subjects also displayed mean baseline-to-endpoint changes in prolactin, fasting glucose, fasting total cholesterol, uric acid, and the hepatic enzymes aspartate transaminase and alanine transaminase.

Paliperidone.

There are limited data regarding the efficacy and safety of paliperidone for the treatment of acute mania in youth. An 8-week OL study examined paliperidone monotherapy for acute manic, mixed, or hypomanic episodes in youth (N=15; 6–17 years of age) with bipolar spectrum disorders (44); treatment with paliperidone was associated with a 60% response rate and 40% remission rate. Participants showed an increase in body weight (mean ± SD increase 2±2.5 kg) and 40% of patients developed clinically significant weight gain (≥7% weight gain).

Quetiapine.

Quetiapine is approved by the U.S. FDA as monotherapy or adjunctive therapy in bipolar youth older than 10 years for manic episodes. Quetiapine has been shown to be safe and effective in children and adolescents with mania in retrospective chart reviews (45), several OL studies (46, 47), and three DBRC studies (40, 48, 49). Quetiapine is effective in reducing manic symptoms and maintaining stabilization through 48 weeks (47). In randomized controlled trials, quetiapine combined with valproate was superior to valproate combined with placebo (48), and quetiapine was superior to valproate as monotherapy (40) and superior to placebo (49). Response rates for quetiapine in these studies ranged from 58% to 87%. Treatment with quetiapine revealed a statistically significant separation from placebo within the first week of treatment (49). Most common side effects included sedation, gastrointestinal upset, and weight gain. Weight increase for quetiapine was on average 1.7 kg in 3 weeks and 3.4 kg in eight weeks. None of these studies reported statistically significant differences in metabolic parameters.

Youth with bipolar disorder tolerate a rapid titration with quetiapine starting at 100 mg and increasing by 100 mg/day to a target dose of 400–600 mg (50). Of note, half-life for quetiapine may be significantly shorter in pediatric patients compared with adults (3 hours versus 6 hours); an extended-release formulation of quetiapine is available, albeit no studies have established that it yields additional benefits over the immediate release formulation in manic youth (51).

Risperidone.

Risperidone received U.S. FDA approval for the treatment of acute manic or mixed episodes in youth ages 10 and above. Efficacy of risperidone is evident in three OL studies of bipolar disorder children and adolescents (13, 52, 53), one retrospective chart review (54), and a DBRPC trial (55). Subjects in the DPRPC trial were assigned to either low-dosage (0.5–2.5 mg/day) or high-dosage (3.0–6.0 mg/day) risperidone or placebo. In both active medication groups, treatment with risperidone significantly decreased manic symptoms. Response rates in the low-dose (59%) and high-dose (63%) groups were superior to placebo (26%). The results of this study suggest that low-dosage risperidone is just as effective as larger dosages, but better tolerated.

Additionally, risperidone has been compared with valproate in two head-to-head randomized double-blind trials (11, 33) and one with lithium (11). Response and remission rates for risperidone were significantly greater to valproate and lithium. It has also been shown to be as effective as augmentation therapy in lithium nonresponders (13).

The side-effect profile of risperidone in youth is similar to that in adults. Most commonly, risperidone is associated with fatigue, dizziness, dystonia, parkinsonian symptoms, akathisia, abdominal pain, dyspepsia, nausea, vomiting and diarrhea, weight gain, and hyperprolactinemia. Special care should be taken regarding the possibility of hyperprolactinemia in pediatric populations since this adverse effect might affect bone development and bring about undesirable menstrual abnormalities (56).

Ziprasidone.

Studies of ziprasidone support its use in pediatric bipolar disorder (57–59). In an eight-week OL study, ziprasidone was evaluated as monotherapy in 21 patients (6–17 years) with significant manic symptoms, resulting in a response rate of 33% and a dropout rate of 33% (58). A larger (N=237), 4-week DBRPC study of ziprasidone in pediatric mania revealed that subjects treated with ziprasidone improved significantly compared with those treated with placebo (59). Separation of ziprasidone from placebo reached statistical significance as early as week one, and the overall response rate was 53% (22% in placebo arm). Interestingly, subjects weighing less than 45 kg did not show a significant improvement compared with placebo.

Ziprasidone is well tolerated in children and adolescents, with no significant weight increases or metabolic changes. The most commonly occurring adverse events (i.e., sedation, dizziness, and somnolence) were similar to what has been reported in adults. Ziprasidone has recently been approved for use in Europe for the treatment of moderately manic or mixed episodes associated with bipolar disorder in children and adolescents aged 10—17 years.

Summary.

Data regarding efficacy of pharmacologic agents for the treatment of acute mania in youth are scarce, compared with what findings are available for adults with mania. Selection of a pharmacological agent to treat an acute manic episode should take into consideration the available empirical evidence suggesting its efficacy and also balance the harm induced by the potential side effects. Overall, SGAs seem to be more effective than lithium and mood stabilizers for acute mania in children and adolescents with bipolar disorder. Two recent meta-analyses determined that SGAs are more effective than lithium and mood stabilizers when considering effect sizes, likelihood of achieving remission, and number needed to treat (60, 61). In the near future, results from ongoing pediatric mania studies of medications that have been effective for mania in adults may become available (e.g., asenepine). Still, research and data collection on those pharmacological agents already available should not be abandoned so that definite recommendations can ultimately be made.

Empirical Evidence of the Pharmacological Treatment of Bipolar Depression in Children and Adolescents

Youth with bipolar disorder may experience full depressive episodes or subsyndromal depressive symptomatology. Youth with bipolar disorder spend nearly 40% of the time with impairing depressive symptoms (3). Furthermore, prospective studies have demonstrated that after remitting from a manic episode, children and adolescents may be more likely to relapse into a major depressive episode than into hypomanic, manic, or mixed episodes (5). Unfortunately, there is a paucity of controlled data regarding pharmacotherapy for depression among youth with bipolar disorder.

Lithium

Lithium has an established efficacy in adults with bipolar disorder and in treatment-resistant depression. However, in youth the only published study exploring lithium is an OL study conducted with 27 adolescents with bipolar depression (62). This study showed a large effect size (d=1.7), a response rate of 48%, and a remission of 30%. Lithium was well tolerated and did not lead to an abnormally high attrition rate. Additional studies exploring lithium’s efficacy are needed.

Olanzapine/Fluoxetine Combination

An 8-week DBRPC trial involving 255 children ages 10–17 years (170 olanzapine/fluoxetine combination [OFC], and 85 placebo), evaluated the efficacy of this combination strategy in patients with bipolar depression. On average, patients in active treatment showed a significantly larger decrease in symptomatology (−28 versus −23; p=0.003), and a larger proportion of them achieved remission (59% versus 43%; p=0.03) and response (78% versus 59%; p=0.003). There was no difference in treatment-emergent suicidal ideation/behavior and in worsening of manic symptoms between placebo and OFC. To date, OFC is the only FDA-approved treatment for bipolar depression in children and adolescents (from clinicaltrials.gov identifier NCT00844857).

Quetiapine

Quetiapine is a well-established treatment for adults with bipolar depression; however, in a DPRPC trial in adolescents with bipolar depression, response rates did not differ from placebo (63). A small study of adolescents (N=32) found quetiapine to be well tolerated, but the difference in response rate for quetiapine and placebo were not statistically different (71% versus 67%). Similar results were found in a larger DPRPC trial of 193 youths with bipolar depression (response rates: quetiapine 63% versus placebo 55%) (64). High placebo responses are not uncommon in pediatric depression studies, which may imply the need for different designs such as placebo or psychosocial lead-in designs (65).

Lamotrigine

In pediatric bipolar depression, several case reports describe the successful use of lamotrigine for periods of up to 8 months (66, 67), as lamotrigine has now emerged as a first-line agent for adult bipolar depression. One study of 46 adolescents with manic or mixed episodes demonstrated reduced depressive symptoms following lamotrigine treatment (28). In an OL study of depressed adolescents with a pediatric bipolar spectrum disorder diagnosis, lamotrigine as monotherapy or adjunctive therapy was well tolerated and resulted in a response rate of 63% and a remission rate of 58% (68).

Antidepressants

Antidepressants are effective and well tolerated for the treatment of unipolar depression in children and adolescents (69); however, their efficacy in bipolar depression has not been thoroughly examined. Two meta-analyses suggest divergent findings regarding their efficacy and potential to induce manic episodes (70, 71), while there are fewer empirical data available in children and adolescents. Retrospective chart reviews suggest the use of antidepressants may be effective; however, patients treated with antidepressants are three times more likely to develop treatment-emergent mania (72).

The risk of inducing mania as a result of treatment with SSRIs or other antidepressants must be weighed against the potential therapeutic benefits as well as the suicide risk that often accompanies severe depression. Given the scarce data available, caution is indicated when considering the use of antidepressants in depressed patients with bipolar disorder.

Summary

The olanzapine/fluoxetine combination to date is the only U.S. FDA-approved treatment for bipolar depression in children and adolescents. Both lithium and lamotrigine have been shown to be feasible options in the treatment of pediatric bipolar disorder depression. However, double-blind placebo-controlled trials for both these medications are needed to give a definite recommendation. Antidepressants can be considered an option only after carefully considering the risks of treatment-emergent mania and after other options have been explored.

Treatment-Resistant Bipolar Disorder

Despite documented efficacy of several pharmacological agents in the treatment of bipolar disorder in children and adolescents, empirical evidence shows that a substantial proportion of patients fail to achieve an adequate response or remission of symptoms. When evaluating patients with a poor treatment response and persistent mood symptoms, clinicians might consider following a set of basic steps:

Combination Treatments

There have been eight studies evaluating combination therapy: six OL and two double-blind studies. These studies have evaluated mainly the combination of SGA with mood stabilizers.

A study comparing quetiapine with valproate to valproate alone (48) found that response rates were higher in the combination group compared with monotherapy (87% versus 53%). However, combination therapy was associated with more weight gain (4.2±3.2 kg versus 2.5±2.1 kg).

The combination of risperidone and lithium has been tested in two studies. One study included subjects who received a combination of risperidone plus lithium or valproate (53), while the other study was a 1-year OL study for subjects who did not respond adequately to 8 weeks of lithium monotherapy (13). Patients on risperidone plus lithium combination achieved a combined average response rate of 84% and a remission rate of 60%; however, weight gain was significant with this combination strategy, and dropout rates were high (approximately 35%).

An 8-week OL study compared olanzapine monotherapy with the combination of olanzapine and topiramate in youths with pediatric bipolar disorders in mixed or manic episodes (26). The combination of topiramate and olanzapine was associated with less weight gain, but did not show greater symptom improvement compared with olanzapine monotherapy.

Empirical evidence supports the use of combination therapy when monotherapy is unsuccessful; however, a paucity of data precludes a definite recommendation. Although use of polypharmacy is common, it should always be addressed with caution. The use of multiple pharmacological agents increases the likelihood of adverse events that may affect treatment adherence. However, in patients who are treatment resistant, a combination of two antimanic agents may be warranted.

Psychosocial Interventions in the Treatment of Bipolar Disorder

The development of bipolar disorder during childhood or adolescence disrupts ongoing developmental processes. Therefore, a comprehensive, multimodal treatment approach that combines psychopharmacology with adjunctive psychosocial therapies is usually indicated for children and adolescents with bipolar disorder (85). However, there is little research on how psychosocial treatment used in conjunction with medication might enhance treatment, prevent poor outcomes, and improve the quality of life for patients and their families.

Many therapeutic elements are common across treatments, such as the central role of psychoeducation for the family, and affect regulation for the child (86). Although the existing treatments are in differing stages of empirical validation, they nonetheless assist with delineating the features that are critical to the effective multimodal treatment of pediatric bipolar disorder (87).

Cognitive-Behavioral Therapy for Pediatric Bipolar Disorder

Cognitive behavioral therapy (CBT) has been studied in single-family setting (53), multiple-family setting (88), and individual (with limited familial involvement) (89). Child- and family-focused CBT (CFF-CBT) (90) was developed as an adjunctive psychosocial intervention for children 8–12 years old with bipolar spectrum disorders and their families. CFF-BT consists of 12 sessions that integrate psychoeducational, cognitive-behavioral, and interpersonal techniques; that focus on psychosocial factors influencing course of illness (e.g., expressed emotion, stressful life events); and that teach coping, CBT, communication, and problem-solving skills. An open trial using CFF-CBT plus pharmacotherapy in 34 youths (ages 5–17) with bipolar spectrum disorders found the intervention feasible and showed significant improvements in ADHD, aggression, mania, psychosis, depression, sleep disturbance, and global functioning posttreatment. Subjects in the CFF-CBT maintenance phase exhibited preservation of improvements in symptoms and functioning over a 3-year follow-up (91).

A multiple-family adaptation of CFF-CBT consisting of 12 concurrent parent and child group sessions was examined in an open trial with 26 children (ages 6–12) with bipolar disorder and their families (88). Multiple-family CFF-CBT was deemed feasible and acceptable to parents; it also revealed significant improvement in children’s manic symptoms and psychosocial functioning and nonsignificant improvement in parents’ knowledge and perceived self-efficacy in coping.

A program of CBT for adolescents with bipolar disorder spectrum diagnoses has also been developed and tested. It consists of 12 weekly sessions of acute-phase treatment followed by 6–10 biweekly sessions and biannual booster sessions (89). Treatment involves psychoeducation about symptoms and course of bipolar disorder, medication compliance, mood monitoring, anticipating stressors and problem solving, sleep regulation and relaxation, family communication and assertiveness, and relapse prevention; there are optional modules for substance abuse, social skills, anger management, and contingency management. A pilot study compared eight youths (ages 10–17) with bipolar spectrum disorders who received acute CBT with eight matched historical controls. CBT showed no significant between-group differences posttreatment or at 8-week follow-up, and the CBT group reported nonsignificant improvement in depression and mania at both time points. Parents of CBT adolescents also reported significant posttreatment improvement in youths’ depression and mania; however, significant improvements were maintained only for depression.

Multifamily Psychoeducation Groups

The multifamily psychoeducation groups (MFPG) consist of eight 90-minute group sessions that occur separately, but simultaneously for parents and children (92). The goals of MFPG include teaching families about the patient’s illness, treatment, symptom management, combined with relaxing activities in four categories (creative, physical, social, and relaxing), and teaching children valuable coping skills. The MFPG intervention has been empirically tested in a randomize trial comparing treatment as usual (TAU) plus MFPG with TAU plus waiting list control (WLC+TAU) (93). This study involved 165 children, 70% of whom had a bipolar spectrum diagnoses. Children receiving MFPG+TAU had a significant improvement in mood symptom severity compared with WLC+TAU over 1-year follow-up.

Dialectical Behavior Therapy for Adolescents

Dialectical behavior therapy (DBT) is an evidence-based psychotherapy designed for adults with borderline personality disorder; however, it has been adapted for pediatric bipolar disorder, which consists of 6 months of weekly 60-minute psychotherapy sessions followed by another 6 months of bimonthly sessions, with the aim of increasing and maintaining individual and family skills (94). Family skills training includes psychoeducation about bipolar disorder, as well as teaching coping skills (i.e., mindfulness, distress tolerance, emotion regulation, and interpersonal effectiveness). Individual therapy sessions address target behaviors outlined in the DBT hierarchy (i.e., life-threatening, treatment interfering and quality-of-life interfering behaviors, skills development) by using problem-solving strategies within a validating environment.

A small open trial of DBT in 10 adolescents with bipolar disorder indicated that the treatment is feasible and acceptable for this patient population (95). Participants exhibited decreased suicidality, nonsuicidal self-injury, emotional dysregulation, and depression symptoms following treatment; however, no significant improvements in mania or interpersonal functioning were noted.

Family-Focused Treatment for Adolescents

Family-focused therapy, a treatment originally developed for adults with bipolar disorder, has been adapted by Miklowitz and colleagues for use with adolescents (FFT-A) (96). The FFT-A intervention lasts 9 months and consists of 21 (12 weekly, 6 biweekly, and 3 monthly) 50-minute sessions; it includes the parents, siblings, and the adolescent with bipolar disorder. The psychoeducation component focuses on the family developing a common understanding of the symptoms, etiology, and course of the disorder. The communication enhancement training component includes role-playing and rehearsal to develop skills for active listening, providing positive feedback, delivering constructive criticism, or requesting changes in another person’s behavior. Finally, the problem-solving component involves teaching participants how to identify problems in daily life, generate solutions, and implement those solutions.

Evidence of FFT-A’s efficacy comes from a two-site randomized controlled trial with a 2-year follow-up (97). Adolescents with bipolar spectrum disorders (N=58) were randomized to receive FFT-A in combination with regular pharmacotherapy or enhanced care (EC) and regular pharmacotherapy. Blind evaluators assessed outcome measures, with no significant differences between groups, in regard to completion and adherence. There were no differences between FFT-A and EC in rates of recovery from index episode; however, patients in the FFT-A group recovered faster from depressive episodes. Adolescents in families that rated high for expressed emotion showed greater reductions in depressive and manic symptoms in FFT-A than those in EC (98).

Interpersonal and Social Rhythm Therapy for Adolescents

Interpersonal and social rhythm therapy (IPSRT) is a manual-based psychotherapy founded on the theory that at least one aspect of the biological diathesis to bipolar disorder is a vulnerability of the circadian system and the neurotransmitter systems involved in its regulation (99). Psychosocial stressors are hypothesized to precipitate and/or exacerbate bipolar episodes through their ability to disrupt social and sleep routines. Observational studies support this theoretical framework by elucidating the effects that stressful life events, social support, social routine disruption, and sleep disturbances have in the onset and maintenance of mania and depression (a review can be found in reference 100).

IPSRT attempts to address three interrelated pathways to symptom exacerbation in bipolar patients: medication nonadherence, disruptions in social and sleep routines, and psychosocial stressors. IPSRT addresses the interplay between the interpersonal and the biological spheres by helping patients identify and manage psychosocial stressors and social role transitions, illustrating the importance and maintenance of circadian integrity.

Hlastala and colleagues developed an adapted version of interpersonal and social rhythm therapy for adolescents (IPSRT-A) with bipolar disorder (101). IPSRT-A consists of 16–18 sessions delivered over 20 weeks, mostly with the adolescent alone. In an open pilot study of 12 adolescents diagnosed with bipolar spectrum disorders, feasibility and acceptability of IPSRT-A were high; IPSRT-A participants experienced significant decreases in manic, depressive, and general psychiatric symptoms over the 20 weeks of treatment; global functioning increased significantly; and effect sizes ranged from medium-large to large.

Complementary Medicine

Omega-3 Fatty Acids

Omega-3 fatty acids are long-chain polyunsaturated fatty acids (LC-PUFAs) and are so named because of the position of the last double bond in their chemical structures. Omega-3s are essential LC-PUFA, meaning they are obtained only from diet sources, as mammals are unable to synthesize them de novo, and can only elongate and desaturate them—i.e., changing plant-based alpha-linoleic acid (ALA) into marine-based eicosapentanoic acid (EPA) and EPA into docosahexaenoic acid (DHA) (102). Omega-3 LC-PUFAs have gained attention for their role in the management of depressive symptoms and ADHD (103). Their use in bipolar disorder has grown; however, data concerning their overall efficacy are still lacking.

To date, two published studies for omega-3 use exist in the literature. The first one is an 8-week OL trial of combined EPA/DHA as monotherapy (starting at 1125 mg EPA/165 mg DHA, with flexible dosing up to EPA 1290 mg/DHA 4300 mg) in 20 children and adolescents with bipolar I, II, or NOS, and a YMRS of >15 (104). Symptom reduction at study endpoint was noted, along with 50% (n=10) of subjects with a 30% reduction in baseline YMRS scores and 35% (N=7) with a 50% reduction in baseline YMRS scores; depressive symptoms also improved. The other study is a pilot randomized controlled trial of flax oil up to 12,000 mg/day in children and adolescents with bipolar I and II disorder for 16 weeks and found no difference for primary mood measures, but moderate effects for clinician-rated global improvement and severity of overall illness in those who displayed decreases in arachidonic acid and increases in ALA and EPA (105). This study suggests the importance of change in blood biomarkers as measures of compliance and outcome. Ongoing and future research into omega-3 LC-PUFAs should take biologic marker changes into account in analyses as determinants of compliance and downstream metabolism in PUFA pathways and should also determine whether dietary intake data can predict outcome.

Omega-3 supplementation may also positively benefit attention in the more than half of youth with bipolar disorder who also have comorbid ADHD. DHA supplementation in healthy boys has been found to increase activity in attention networks in the prefrontal cortex during sustained attention tasks (106). The side-effect profile for omega-3 LC-PUFA administration is typically benign, with gastrointestinal disturbance including transient nausea or diarrhea the most common concern, especially in younger children. Although typically well tolerated, rare cases of bleeding have been reported for concomitant use with aspirin and anticoagulants, especially when taken in exceptionally large doses (107).

To our knowledge there are no studies describing the efficacy or safety of other complementary treatment options (St. John’s wort, inositol, S-adenosyl-L-methionine, acupuncture) in youth with bipolar disorder.

Psychopharmacological Treatment of Bipolar Disorder with Comorbid Conditions

Anxiety

Despite evidence that patients with comorbid anxiety disorders are at increased risk for poor treatment response and treatment-emergent episodes (4), no clinical trials have explicitly studied the treatment of comorbid anxiety in pediatric bipolar disorder. Lack of data regarding the safety and efficacy of SSRIs for comorbid anxiety in pediatric bipolar patients implies that there is a great need for pharmacologic and psychosocial studies for the treatment of this diagnosis.

Substance use Disorders

As noted previously, substance use and substance use disorders are common among adolescents with bipolar disorder. However, few clinical trials have specifically examined the effective pharmacological treatments when comorbid substance use disorders are present. In a 6-week randomized placebo-controlled study using lithium for pediatric bipolar spectrum disorders with secondary substance dependence, subjects were less likely to test positive for a urine drug assay after 6 weeks and had significantly greater improvement on the Children’s Global Assessment Scale (108). A 16-week, DBRPC study examined the efficacy of topiramate (average dose 175 mg/day) versus placebo, adjunctive to quetiapine, in 75 manic adolescents with co-occurring cannabis use disorders (ClinicalTrials.gov identifier: NCT00393978), illustrating a significantly greater reduction in cannabis use in the topiramate group versus the placebo group.

Psychosocial interventions that target some of the known risk factors for substance use (e.g., parental alcoholism or other substance abuse, low parental support, inconsistent discipline, and poor parent supervision) may be useful while treating adolescents with bipolar disorder and substance abuse rather than treatment with a specific focus on bipolar disorder or substance abuse.

Attention Deficit Hyperactivity Disorder

Attention deficit hyperactivity disorder is the most common comorbid psychiatric disorder in youth with bipolar disorder (4). These patients are less likely to respond to treatment and may be at increased risk of substance use disorders. Furthermore, conventional first-line agents in the treatment of ADHD have been associated with treatment-emergent mania and worsening of symptoms.

Specifically, the American Academy of Child and Adolescent Psychiatry treatment guidelines advise that symptoms of bipolar disorder should be stabilized first, and if impairing symptoms of ADHD persist, they may be judiciously treated with psychostimulants. Few trials for antimanic agents have reported their efficacy in treating ADHD symptoms. In patients aged 6–17 years these studies have found ADHD rates of response of 33% for ziprasidone (57), 30% for risperidone (109), and 60% for aripiprazole (36, 39).

When significant symptoms of ADHD persist despite mood stabilization, adding a medication for ADHD may be warranted. In a study of 30 pediatric patients with bipolar disorder and ADHD who were stabilized on valproate and then randomized to placebo or mixed amphetamine salts, those in the amphetamine group showed significant reduction of ADHD symptoms and no statistically significant side effects or worsening of manic symptoms (110). Methylphenidate has been studied in youth with bipolar disorder in two studies, one with prior stabilization with valproate and/or lithium (111); the other was in bipolar disorder patients previously stabilized with aripiprazole (112). In both studies, patients had significantly improved symptoms of ADHD, without significant worsening of manic symptoms. Additionally, atomoxetine has been studied in a retrospective chart review (113) and an OL study (114) and appears to be effective in lowering ADHD symptoms, without worsening of mood symptoms.

Conclusions and Future Directions

The previously reviewed literature suggests that effective treatments for pediatric bipolar disorder, in manic/mixed or depressed phases, are available. Specifically, SGAs are highly efficacious for the treatment of acute manic/mixed episodes, although tolerability remains a major concern. The evidence regarding the traditional mood stabilizers (e.g., valproate) suggests that they are less efficacious than SGAs among youth; however, this conclusion remains tentative pending additional placebo-controlled trials.

Furthermore, evidence for additional pharmacological and nonpharmacological treatments of pediatric bipolar disorder is still needed. The issue of maintenance treatment and relapse prevention in bipolar youth has scarcely been addressed in the scientific literature; future studies should address the efficacy and effectiveness of pharmacological and psychosocial interventions in preventing relapses. In addition, because it is highly unlikely that any given medication will be universally effective in youth with bipolar disorder, more studies searching for adequate predictors of treatment response are essential to achieve individualized treatment strategies. Consideration of predictors could include demographic characteristics, symptomatic clusters, family history, pharmacogenetic markers, cognitive deficits, and neuroimaging traits. Systematic data collection, similar data designs, and uniform rater reliability in efficacy measurements will be necessary to identify predictors of treatment response. Future efforts to address these and other questions will ensure that the momentum of progress in the treatment of youth with bipolar disorder continues.