CLINICAL IMPLICATIONS
• Bipolar disorder in children and adolescents is a chronic condition with a high risk of severe depressive episodes.
• It is important for clinicians to distinguish unipolar depression from bipolar depression in youth.
• Lamotrigine and lithium have shown benefit in small open trials but large controlled trials for pediatric bipolar depression are needed.
• Psychosocial therapies show promise in the treatment of bipolar disorder in children and adolescents, but current evidence is not specific to bipolar depression.
Introduction
The prevalence of bipolar disorder was found to be approximately 1% in a community sample of 1,709 adolescents aged 14–17 years.Reference Lewinsohn, Klein and Seeley1 The majority of youths with bipolar disorder in this sample experienced a depressive episode as the onset mood episode, and the mean age of onset of this first episode was 11.75 years. The National Comorbidity Survey Replication—Adolescent Supplement (NCS-A) surveyed 10,123 adolescents aged 13–18 years along with their parents.Reference Merikangas, He and Burstein2 The prevalence of bipolar disorder (bipolar I and bipolar II disorder) in this community sample was 2.9%. The prevalence of bipolar disorder increased with age, with a nearly twofold increase from the 13- to 14-year age group to the 17- to 18-year age group. The lifetime prevalence of mania/hypomania with major depression was 2.5% within the entire sample.Reference Merikangas, Cui and Kattan3 Many prevalence studies do not differentiate between bipolar I and II disorder. A review article by Merikangas and LamersReference Merikangas and Lamers4 reports lifetime prevalence rates of bipolar II disorder in adolescents to be 1.5%.
Diagnosis
Youth with bipolar disorder may have varying mood states at onset of bipolar illness. In a retrospective chart review of 82 youth with bipolar disorder, 52% of youth presented with manic symptoms, 31% with mixed symptoms, and 17% with depressed symptoms.Reference Faedda, Baldessarini and Glovinsky5 Notably, within the group with manic symptoms at onset, the majority of youth experienced a dysphoric mania (65%) versus a classic euphoric mania (35%).
Since depression is a common presentation in child and adolescent clinical settings, it is important to determine whether the depressive symptoms are evidence of unipolar or bipolar depression. For example, in a study of 247 adolescent outpatients presenting to a community mental health clinic with a major depressive episode, 41% met criteria for bipolar disorder I or II.Reference Dilsaver, Benazzi and Akiskal6 The Diagnostic and Statistical Manual of Mental Disorders, 4th ed., text revision (DSM-IV-TR) criteria for a depressive episode do not differ between unipolar depression and bipolar depression, so these are not helpful in the distinction between unipolar and bipolar depression.7
Wozniak etal Reference Wozniak, Spencer and Biederman8 examined a group of 280 children and adolescents with either unipolar or bipolar depression and compared depressive symptoms. Youth with bipolar depression had a more severe depressive episode with suicidality, hopelessness, and anhedonia compared to youth with unipolar depression. Bipolar depression in adolescents has greater clinical severity than unipolar depression, as evidenced by earlier age of onset, more depressive episodes, and more severe depression.Reference Merikangas, Cui and Kattan3 Youth with bipolar depression were also found to have more comorbid disorders, more hospitalizations, and lower functioning than with unipolar depression.Reference Wozniak, Spencer and Biederman8
The clinician should do a thorough assessment for prior or current symptoms of mania or hypomania in children and adolescents who present with symptoms of depression. Diagnostic criteria for bipolar disorder are the same in children and adolescents as for adults. Screening instruments may be helpful in identifying bipolar depression, especially for youth who present with depressive symptoms only. The Mood Disorder Questionnaire—Adolescent version (MDQ-A) aids in identifying prior and current episodes of mania.Reference Wagner, Findling and Emslie9 Other useful screening instruments are the 10-item short form of the Parent General Behavior Inventory (PGBI-SF10) and the Child Mania Rating Scale (CMRS).Reference Youngstrom, Frazier and Findling10, Reference Pavuluri, Henry and Devineni11
Course of Illness
Bipolar disorder in children has a chronic course, with a high percentage of youth experiencing relapse of symptoms. In a 4-year prospective follow-up study of 78 youth with bipolar I disorder, ages 6–17, only 6.4% of youths were considered to be in remission without pharmacotherapy at the end of the 4-year follow up.Reference Wozniak, Petty and Schreck12 Seventy-three percent continued to meet diagnostic criteria for bipolar I disorder. Over 5% of study participants experienced significant depressive symptoms at the end of the 4-year follow up period.
A prospective 4–5 year follow-up study examined the course of bipolar I disorder in 25 youth, ages 9–16 years, who presented with a diagnosis of mania at initial assessment.Reference Jairam, Srinath and Girimaji13 All youth experienced recovery from the index episode, with a mean time to recovery of 44 days. Sixteen youth (64%) relapsed during the course of the study, with a mean period of 18 months to relapse. Polarity at time of relapse was manic (58%), depression (23%), mixed (16%), and hypomanic (3%). The youth who experienced relapse were more likely to have had a prior depressive episode. One subject committed suicide during the course of the study while in a depressive episode.
In the Course and Outcome of Bipolar Youth (COBY) study, 413 youths, ages 7–17 years, with bipolar spectrum disorder were followed over 4 years to assess the course of their illness, including recovery and remission rates.Reference Birmaher, Axelson and Goldstein14 After a median time of 12 weeks, 81.4% of the study participants experienced a full recovery. Of these recovered youths, 62.5% experienced a syndromal recurrence a median of 71 weeks after recovery. The most common polarity observed in recurrence was depression (60%), followed by hypomania (21%), mania (15%), and mixed (5%). This large study highlights the significance of bipolar depression in the pediatric population, since the majority of mood symptoms experienced by participants during the 4-year follow-up period were depressive symptoms. Similarly, in a 4-year follow-up of 86 youth ages 7–16 years with bipolar I disorder, manic or mixed, the mean number of weeks that youth were depressed was 99. Switch rates to depression were 1.1 times per year.Reference Geller, Tillman and Craney15
Youth with bipolar disorder may experience depressive episodes into young adulthood. Of 115 children and adolescents with bipolar I disorder, manic or mixed, followed over 8 years, 20% had depressive episodes by young adulthood.Reference Geller, Tillman and Bolhofner16
Treatment
Pharmacotherapy
There are no U.S. Food and Drug Administration (FDA)–approved medications for the treatment of bipolar depression in children and adolescents. Quetiapine and olanzapine/fluoxetine combination have FDA approval for treatment of adults with bipolar depression. Lithium and lamotrigine are also commonly used in the treatment of adults with bipolar disorder. Medications to treat bipolar depression in children and adolescents are frequently those used to treat adults; however there is a very limited evidence base for their use in pediatric bipolar depression.
Quetiapine
There is one double-blind, placebo-controlled, 8-week trial that examined the efficacy of quetiapine for bipolar depression in 32 adolescents, ages 12–18 years.Reference DelBello, Chang and Welge17 The dose range of quetiapine was 100–600 mg/day with a mean of 403 mg/day. The primary outcome measure was change in the Children's Depression Rating Scale–Revised (CDRS-R) scores from baseline to endpoint. Response was defined as a ≥ 50% reduction in CDRS-R score from baseline to endpoint, and remission was defined as a CDRS-R score ≤ 28. There was no statistically significant difference between quetiapine and placebo in change in CDRS-R scores from baseline to endpoint. Similarly, response rates (71% vs 67%) and remission rates (35% vs 40%) did not differ between the quetiapine group and the placebo group. These findings are in contrast to the positive adult bipolar depression studies.
Lamotrigine
An 8-week, open-label study examined the efficacy of lamotrigine either as monotherapy or adjunctive treatment for bipolar depression in 20 adolescents, ages 12–17 years.Reference Chang, Saxena and Howe18 Seven study participants had a diagnosis of bipolar I disorder, 6 had bipolar II disorder, and 7 had bipolar disorder not otherwise specified (NOS). The primary efficacy measure for response was a score ≤ 2 on the Clinical Global Impressions Improvement scale (CGI-I), and the secondary measure for response was a ≥ 50% decrease in the CDRS-R score from baseline to endpoint. Nineteen youth completed the study, and of those, 84% were considered responders using primary criteria, and 63% were responders based on secondary criteria. There was no difference among youth with bipolar I, II, or NOS in response rate, using secondary criteria. Remission was defined as a CDRS-R score ≤ 28 and a CGI Severity scale (CGI-S) score of ≤ 2; 58% of the youth in the study met remission criteria. The mean dose of lamotrigine at the end of the study was 131.6 mg/day. The most common side effects reported were headache (84%), fatigue (58%), nausea (53%), and sweating (47%). There were no significant rashes reported, and no youth discontinued the study due to adverse effects.
A 12-week, prospective, open-label trial examined the efficacy of lamotrigine in 39 children and adolescents, ages 6–17 years, with bipolar disorder in a manic, hypomanic, or mixed episode.Reference Biederman, Joshi and Mick19 The majority of the study participants had a diagnosis of bipolar I disorder (90%), 1 had bipolar II disorder (2%), and 3 (8%) had bipolar disorder NOS. There was a significant improvement of depressive symptoms from baseline to endpoint, using change in CDRS-R score as a measure for depressive symptom response. A CGI-I score of ≤ 2 for depressive symptoms was found for 46% of youth. Thirteen youth reported skin rashes, and 6 youth discontinued lamotrigine as a result of skin rashes. However, no Steven–Johnson syndrome was reported in any of the youth.
A restrospective chart review examined the response to adjunctive lamotrigine in 5 adolescents with bipolar depression.Reference Soutullo, Díez-Suárez and Figueroa-Quintana20 One youth was diagnosed with bipolar I disorder, 1 with bipolar II disorder, and 3 with bipolar disorder NOS. The mean dose of lamotrigine was 100 ± 87.5 mg/day, and the mean duration of treatment was 28 weeks. All 5 patients were taking a mood stabilizer (lithium, carbamazepine, and/or valproate) and an antipsychotic (olanzapine, risperidone, or quetiapine) before adding lamotrigine. Response was retrospectively rated using the CGI-I scale, and the mean CGI scores significantly improved from 5 (markedly ill) at baseline to 3 (mildly ill) at endpoint. The one youth with bipolar II disorder showed minimal improvement on CGI from baseline to endpoint compared to responses of either much or very much improved in the other two youths (bipolar I or NOS). No significant rashes were reported.
Lithium
A 6-week, open-label study examined the effectiveness of lithium in bipolar I depressed youth, ages 12–18.Reference Patel, DelBello and Bryan21 Lithium was titrated up to a dose yielding a therapeutic serum level of 1.0–1.2 mEq/L. There was a significant reduction from baseline to endpoint in the CDRS-R scores.Reference Poznanski, Freman and Mokros22 Response was defined as a ≥ 50% reduction in the CDRS-R score from baseline to endpoint, and 48% of youth met this criteria. Remission was defined as a CDRS-R endpoint score ≤ 28, and a Clinical Global Impressions Improvement Scale for Bipolar Disorder (CGI-BP) score of 1 or 2; 30% of study participants met criteria for remission.Reference Spearing, Post and Leverich23
Antidepressants
In a chart review of 42 outpatient youth with bipolar disorder in a depressive episode, depressive symptoms were 6.7 times more likely to improve for youth who received selective serotonin reuptake inhibitors (SSRIs) than those who did not.Reference Biederman, Mick and Spencer24 However, youth on SSRIs were 3 times more likely to have emergence or worsening manic symptoms than youth not on SSRIs.
A case report of a 17-year-old girl with bipolar disorder stable on valproate and olanzapine described deterioration of mood with emergence of manic symptoms two days after duloxetine was prescribed for a depressive episode.Reference Desarkar, Bakhla and Sinha25 Her symptoms included decreased sleep, euphoria, grandiosity, and excessive speech, with cycling between euphoria and sadness every four days. Duloxetine was discontinued, the valproate dose was increased, and the manic symptoms subsequently resolved.
A review article examined the potential of antidepressants to induce manic episodes in bipolar youth.Reference Goldsmith, Singh and Chang26 The studies reviewed showed evidence of an increased risk of mania in bipolar youth treated with antidepressants. Placebo-controlled studies are needed to clarify the efficacy and safety of adding antidepressants to existing mood stabilizers in bipolar depressed youth.
Alternative Treatments
A 16-week, double blind, randomized-controlled trial examined the efficacy of adjunctive or monotherapy flax oil in youth, ages 6–17 years, with bipolar I or bipolar II disorder, manic, hypomanic, mixed, or depressed.Reference Gracious, Chirieac and Costescu27 The youth were randomized to either an olive oil placebo or flax oil [550 mg of omega-3 fatty acid α-linolenic acid (α-LNA)]. The dose was titrated to 12 capsules per day as tolerated over 16 weeks. There were no differences between the two groups when comparing scores on all the efficacy measures, including the CDRS-R, which was the primary depression efficacy measure used in this study.
Psychotherapy
The efficacy of adjunctive cognitive-behavioral therapy (CBT) for 16 adolescents with bipolar I, II, NOS, or cyclothymia was assessed in a 12 week randomized study.Reference Feeny, Danielson and Schwartz28 Twelve study participants had a diagnosis of bipolar I disorder, 3 had bipolar II disorder, and 1 had cyclothymia. Eight adolescents received individual CBT along with pharmacological treatment, and 8 adolescents received no adjunctive psychosocial interventions. The group receiving CBT showed a significant decrease in depressive symptoms on the parent-rated General Behavior Inventory (GBI). No significant change in symptoms was reported on the child-rated GBI or on interview-administered measures of depression using the Inventory of Depressive Symptoms (IDS).
Ten adolescents, with bipolar I, II, or NOS, manic, mixed, or depressed received adjunctive dialectical behavioral therapy in an open study.Reference Goldstein, Axelson and Birmaher29 Seven study participants had a diagnosis of bipolar I disorder, 2 had bipolar II disorder, and 1 had bipolar disorder NOS. The course of treatment included 24 weekly sessions over 6 months, then 12 sessions tapering in frequency through 1 year. Sessions included family skills training and individual therapy. Significant improvements in suicidality, emotional dysregulation, and depressive symptoms were found.
One randomized controlled trial examined the efficacy of multifamily psychoeducational psychotherapy for mood symptoms in 165 children.Reference Fristad, Verducci and Walters30 Seventy percent of the subjects had some form of bipolar disorder, and 30% had a depressive disorder. The primary outcome measure was the Mood Severity Index, calculated from a combination of CDRS-R and the Mania Rating Scale ratings at each assessment.Reference Young, Biggs and Ziegler31 Subjects randomized to the multifamily psychoeducational psychotherapy plus treatment as usual showed significant improvement in mood severity scores from baseline to endpoint when compared to subjects in the wait list control plus treatment as usual group, with a decrease in their mean MSI score of 6.48 more than the control group.
Conclusion
Bipolar disorder in children and adolescents has a chronic course, marked by relapse of symptoms in a high percentage of patients. Youth with bipolar disorder spend a significant amount of time in depressive episodes during the course of their illness. They also experience more frequent depressive episodes and lower functioning than those with unipolar depression. Bipolar depression in this population is more likely to present with more severe depressive symptoms than unipolar depression, including a higher risk of suicidality. However, evidence-based pharmacological treatment for bipolar depression in children and adolescents is limited to mostly small open studies, and evidence for psychosocial interventions is not specific to bipolar depression. Large controlled studies are needed to examine the efficacy of pharmacological and psychotherapy treatments for bipolar depression in the pediatric population.
Disclosures
Dr. DeFilippis has nothing to disclose. Dr. Wagner for the period 3/2012 to 3/2013 has received honoraria from UBM Medica, American Psychiatric Association, Slack Inc., Las Vegas Psychiatric Society, Partners Healthcare, Brain and Behavior Research Foundation, NAC CME, and the University of Wisconsin.
