This study will compare the efficacy of lithium, divalproex, and placebo in treating the acute phase of symptomatic bipolar I disorder, mixed or manic episode, in children and adolescents.

Primary Aim:

To compare the efficacy of LI, DVP, and PBO in the acute phase treatment of symptomatic bipolar I disorder, mixed or manic episode, in children and adolescents. Our hypothesis is that differential efficacy will be observed with the following predicted order of response: DVP = LI > PBO.

Secondary Aims:

1. To collect systematic safety data on the incidence of weight gain and the development of insulin resistance and hyperandrogenism in bipolar adolescent females treated with LI, DVP, or LI + DVP.
2. To collect data on possible predictors of acute treatment response to the two active treatments.
3. To provide descriptive information on the stability of acute phase response to monotherapy with either LI or DVP over 6 months of continuation phase treatment.
4. To develop safety and efficacy data about the use of stimulant medications in these subject while treated with a mood stabilizer.

Inclusion Criteria:

1. 7.0 – 17 years of age
2. Bipolar I Disorder, mixed or manic episode, psychotic or non-psychotic, according to DSM IV criteria
3. Score of > 14 on the Y-MRS
4. Normal intelligence
5. Ability and willingness to provide assent and informed, written consent from at least one parent or legal guardian
6. No current general medical illnesses requiring medication

Exclusion Criteria:

• 1. A current or lifetime DSM-IV diagnosis of schizophrenia, autistic disorder, schizoaffective disorder, pervasive developmental disorder, or obsessive compulsive disorder 2. IQ < 70 3. Patients with serious suicide risk 4. Concurrent cognitive-behavior therapy that is specifically focused on the child or adolescent’s bipolar symptoms within 6 weeks of enrolling in this trial.

  5. Any use of psychotropic agents within the preceding 2 weeks, including neuroleptics, monoamine oxidase inhibitors, stimulants, antidepressants, or depot neuroleptics or fluoxetine in the past month 6. Current or history in past 3 months of a DSM-IV diagnosis of Substance Abuse/Dependence or use of illicit drugs or alcohol in the past 3 weeks. Patients who have a positive drug screen at intake, who would otherwise be eligible for the study, will be given the opportunity to repeat the drug screen 3 weeks later. They will be excluded if the second drug screen is positive.

  7. Pregnancy or sexually active females not using a reliable form of contraception 8. Previous adequate trial of either LI or DVP defined as; 3 weeks of DVP at serum levels between 75-125 OR dosage of at least 20 mg/kg; Lithium for at least 4 weeks at serum levels of .8 – 1.2 or dosage of at least 30 mg/kg.

  9. Allergies to LI, DVP or chlorpromazine. 10. Bipolar subjects who are currently stable on mood stabilizers or atypical neuroleptics.

  11. Bipolar subjects with Bipolar I disorder and ADHD who are stable on stimulants with or without concurrent mood stabilizers.

  12. Inpatient hospitalization within 6 months prior to screening. Partial hospitalization is acceptable.