Bipolar disorder presents unique pharmacologic challenges. Depending on the phase of illness a patient is in, the focus may be on acute manic or mixed episodes, acute depressive episodes, maintenance treatment (ie, prevention of recurrent episodes), and/or treatment of comorbidities, such as anxiety, ADHD, and substance abuse. Those treating children or adolescents with bipolar disorder face the additional challenge of having far less information to inform their treatment decisions.

Our field has learned from the failure of tricyclic antidepressants for youth depression that it is not as simple as just extrapolating from adult studies. We need studies specifically focusing on youth to best understand the efficacy and tolerability of these medications.

As recently as a decade ago, there was almost no information available to guide pharmacologic treatment decisions for youth with bipolar disorder. Fortunately, a lot has happened in the past 10 years. In this article we’ll summarize some important findings. (For a more detailed overview, see Pfeifer et al, CNS Drugs 2010;24(7):575–593, and Goldstein et al, Child Adolesc Psychiatric Clin North Amer 2012;21:911–939, which contains a full listing of citations for studies referred to here.)

What Have We Learned?
Treatment of Acute Manic and Mixed Episodes

Most studies about the pharmacologic treatment of bipolar disorder among youth have focused on acute manic or mixed episodes. This makes good sense, as mania is the most distinctive aspect of bipolar disorder, manic/mixed episodes often present significant risks, and using psychosocial treatment alone is not advisable. When faced with a youth who is manic, the main question is which medicine to use rather than whether to use a medication. This scenario differs from unipolar depression or anxiety disorders, for which psychotherapy alone is often a reasonable option.

Multiple large-scale studies of second-generation antipsychotics (SGAs) show clearly that this class of medications works very well for youth with manic/mixed episodes. In each of these large studies, SGAs have clearly and convincingly outperformed placebo. Although there have not been large head-to-head studies of one SGA against another, it appears that they are relatively similar in terms of anti-manic benefits. Where they differ is in side effect profiles, particularly weight gain and metabolic abnormalities. Correll and colleagues examined these changes among treatment-naïve youth receiving SGAs for a variety of conditions (Correll et al, JAMA 2009;302:1765–1773). Similar patterns of findings have been observed in bipolar disorder specifically.

Despite the fact that lithium is arguably the leading medication for the treatment of bipolar disorder in adults, so far there are no published placebo-controlled trials of lithium for acute manic/mixed episodes among youth. An unpublished study found no significant advantage for lithium (41%) versus placebo (30%).

Regarding divalproex (Depakote), although some early open-label studies suggested possible benefits, with response rates in the 50% to 75% range, a large randomized controlled trial for youth found a low response rate (24%) that did not beat placebo (23%) (Wagner et al, J Am Acad Child Adolesc Psych 2009;48:519–532). There was some thought that this study may have been negative because of suboptimal divalproex levels. There are some potentially promising signals from studies of carbamazepine (Tegretol), oxcarbazepine (Trileptal), lamotrigine (Lamictal), and topiramate (Topamax); however there are no positive randomized controlled trials for any of these medications in the treatment of acute episodes of mania among youth.

Recently, the large, multi-site Treatment of Early Age Mania (TEAM) study compared lithium, divalproex, and risperidone (Risperdal) (Geller et al, Arch Gen Psychiatry 2012;59:515–528). The response rate for risperidone (68%) was significantly higher than for lithium (36%) or divalproex (24%), which did not differ significantly from one another. However, in some sites, risperidone was vastly superior to lithium, whereas in other sites there was a minimal difference between the two medications. Could it be that the efficacy of lithium depends in part on clinician characteristics (eg, familiarity with lithium) or patient characteristics (eg, family history)? Clearly, lithium is far too important an option, and the evidence base is far too limited, to prematurely conclude at this point that it does not work for mania among youth.

One recent study examined the efficacy and tolerability of mood-stabilizing medications for manic/mixed episodes among adults and youth. Among youth, but not among adults, SGAs were more efficacious than other mood-stabilizers. Similarly, among youth, but not among adults, SGAs were associated with substantially greater weight gain than other mood stabilizers.

Treatment of Acute Bipolar Depression

A handful of small open-label studies have looked at the impact of mood stabilizing medications on the depressed phase of bipolar disorder among youth. Response rates in studies of lithium and lamotrigine have been good, around 50% to 60%. Only one placebo-controlled study (N=32 adolescents) has been published so far, comparing quetiapine (Seroquel) with placebo. Enthusiasm about the high response rate for quetiapine (71%) was tempered by the equally high response rate for placebo (67%). Since quetiapine is the leading option for bipolar depression among adults, future larger studies among youth are likely. Treatment guidelines discourage the use of antidepressants as monotherapy, but acknowledge a potential role when used in combination with a mood-stabilizing medication. Unlike the adult data, there are no rigorous placebo-controlled studies of antidepressants for bipolar depression among youth, so caution is warranted.

Comorbidity and Treatment

ADHD is one of the most common comorbid conditions among youth with bipolar disorder. It seems that youth with comorbid ADHD have less robust response to mood stabilizing medication. In addition, there is the concern that stimulants could potentially destabilize mood among youth with bipolar disorder. Fortunately, studies have shown that once mood is adequately stabilized with mood stabilizing medications, youth with comorbid ADHD can often be safely and effectively treated with adjunctive stimulants. Although anxiety is very common among youth with bipolar disorder, and is associated with greater overall clinical severity, no treatment studies to date have specifically endeavored to improve comorbid anxiety among youth with bipolar disorder. There is support from a small placebo-controlled trial that lithium may help to reduce substance use and improve global functioning among youth with bipolar disorder and comorbid substance use disorders. A recent randomized trial examined topiramate or placebo in combination with quetiapine, for manic youth with comorbid cannabis use disorders. Topiramate was associated with reduced cannabis use, and was also associated with reduced appetite, lessening quetiapine-associated weight gain.

Maintenance Treatment

No large-scale, placebo-controlled studies have been conducted on the topic of maintenance treatment among youth with bipolar disorder. One study examined youth with bipolar disorder who had stabilized on the combination of lithium and divalproex. Those youth were randomized to discontinue one of the two mood-stabilizers, and the findings suggested that in this context lithium and divalproex were equally effective at preventing/delaying mood episode recurrences. There is some evidence from prospective open-label or naturalistic studies that extended (>six months) treatment with lithium and/or divalproex may help to reduce the burden of mood symptoms among youth with bipolar disorder. This includes prevention of recurrences and further reductions in subsyndromal symptoms.

Where to Next?

There is an important need for continued research regarding pharmacologic treatment of youth with bipolar disorder. Areas in obvious need of progress include combination treatment, maintenance treatment, bipolar depression, and the treatment of comorbid anxiety.

Another important topic that has received very little attention is the treatment of bipolar-spectrum conditions including bipolar II disorder, cyclothymia, and bipolar disorder not otherwise specified. There is compelling evidence that the latter progresses to bipolar disorder type I or II in 40% to 60% of cases, depending on the presence of a family history of bipolar disorder. Irrespective of diagnostic conversion, bipolar spectrum conditions are highly impairing and persistent, and these conditions are far more similar to bipolar I disorder than they are different. This invokes unique pharmacologic treatment considerations that have not to date been examined rigorously.

Our field needs some traction in terms of objective markers that can be used in the selection and monitoring of pharmacologic treatment. Examples include blood tests for serum biomarkers, neurocognitive testing, neuroimaging, and genetic testing. Despite recent gains in these areas, so far the existing science has not translated into clinical decision making. That type of progress is urgently needed, and overlaps substantially with the need for these same technologies to help optimize diagnostic accuracy. Bipolar disorder is a systemic illness associated with inflammation and oxidative stress. Incorporating this evidence in our search for novel treatments may lead us to “out of the box” approaches based on existing treatments from other branches of medicine, including medications with anti-inflammatory properties.

Finally, we cannot ignore the deleterious impact of treatments, particularly SGAs, on the cardiovascular system, and the vastly increased risk for premature cardiovascular disease and related conditions that is independently associated with bipolar disorder. Future studies should incorporate primary tolerability outcomes as well as primary efficacy outcomes. The potential role of complementary and alternative treatments, such as fish oils and curcumin, warrants further investigation. These agents may present options for bipolar depression, early intervention in milder bipolar spectrum conditions, and adjunctive treatment that improves residual symptoms with attractive risk-benefit balance.

The past decade has seen progress in the pharmacologic treatment of bipolar disorder among youth. With continued research we will hopefully be able to offer more personalized treatment options that better balance efficacy and tolerability for youth with bipolar disorders across all phases and stages of illness.