Growing Evidence to Support Early Intervention in Early Onset Bipolar Disorder

Despite evidence that offspring of parents with bipolar disorder (BD) are four times as likely to develop a mood disorder than children without a family history of BD [1], it appears that many clinicians are still reluctant to treat children and adolescents who have both genetic risk factors and presentation with significant affective symptoms suggestive of early expression of a BD diathesis. The reasons for this reticence are likely to be multifactorial, ranging from challenges in accurately diagnosing paediatric bipolar disorder (PBD), to insufficient data demonstrating that genetically high-risk youth with mood instability actually develop bipolar I disorder (BP-I), to challenges in finding safe and effective monotherapy treatments for children diagnosed with BD.

Making an early diagnosis is complicated by the fact that many of the manifestations of BD in children and adolescents differ from the prototypic descriptions noted in adults. At present it is still unclear whether these differences in clinical presentation are a result of developmental differences in symptom expression and/or differences in the aetiology of paediatric versus adult-onset BD [2, 3].

Rationale for early intervention in BP-I

The theoretical importance of early intervention for BD has been reviewed extensively by Post et al. [4]. The authors noted that, based on longitudinal data from adults, the extant evidence suggests that BD appears to be associated with progressive worsening over time.

There is also preliminary information to suggest that BD illness progression (e.g. more severe episodes with increased cycling rate) may occur during childhood. For example, Jairam et al. found that all subjects (n = 25) with juvenile mania recovered from their index episode, but that 64% relapsed after an average follow-up period of 18 months [5]. Significantly, a majority of the relapses (72.4%) occurred while the subjects were on treatment. Retrospective studies examining accounts of age of onset of bipolar symptoms in adult patients with BD have shown that onset before adulthood [6] or before adolescence [7] is associated with a longer duration of illness, more chronic course, and increased comorbidity with other psychiatric disorders, suggestive of a more severe disease process. Concerns about illness progression highlight the theoretical need to both identify young patients with early onset BD as well as a theoretical rationale to begin treatment early in the course of illness.

Comparing treatment response and phenomenology: adults versus youths

One method of examining the rationale for early intervention in bipolar illness is to compare the response rates of similarly designed pharmacotherapy outcome studies in children versus adults. If higher rates of treatment response in young subjects given pharmacotherapeutic interventions are seen when compared to those response rates observed in adults, this observation would provide one form of concrete evidence to support early intervention in BD.

Two multiphase studies in populations with BD provide notable evidence for the aforementioned phenomenon. In a study published in 2005 (a subset of which is described in more detail in a 2003 publication), 139 children and adolescents with syndromal BD (94.2% of whom had BP-I and 5.8% of whom had BP-II) were treated with open-label combination lithium and divalproex sodium (Li + /DVPX) [8, 9]. Sixty youths (43%) responded to treatment and were consequently randomized to Li+ and DVPX monotherapy. In comparison, Calabrese et al. similarly treated 254 adults with rapid cycling BD with open-label Li + /DVPX [10]. However, in that study the investigators observed a response rate of only 24% (n = 60) of subjects who were eligible to be subsequently randomized to Li+ and DVPX monotherapy.

It is also interesting to note that of the 194 non-responders in the Calabrese et al. study [10], 65 did not respond to treatment as a result of treatment-recalcitrant mood episodes, the majority of which (n = 48) were depressive in nature. Conversely Findling et al. found that out of 79 non-responders to Li + /DVPX, treatment-resistant depression was not a reason for treatment failure in any of the young patients with BP-I [9]. The absence of treatment-resistant depression noted in the Findling et al. study [9] supports previous observations by Kraepelin that depressive symptoms are less prominent than manic or hypomanic symptoms in young patients with bipolar illness [11].

When considered together, the studies of Findling et al. and Calabrese et al. provide evidence to support the assertion that there may be differences in therapeutic response to psychopharmacological agents across the life cycle [9, 10]. Therefore, these studies highlight the necessity to specifically conduct psychopharmacological studies in children and to not infer that children will respond to psychoactive medications in the same way as adults [12].

Toward identification of a bipolar prodrome in youths

The finding that children and adolescents with BP-I may respond at higher rates to pharmacological intervention than adults not only highlights the need for accurate identification of BP-I, but also emphasizes the theoretical construct that early intervention might be associated with a more favourable outcome. It has been noted that less malignant expressions of mood disorders in youths may antecede the development of BP-I [13–15]. Thus, if early intervention is associated with positive outcomes, it might be surmised that if treatment is provided early in the course of a bipolar illness, prior to the full expression of BP-I, this might be an important time frame for even more effective intervention.

To facilitate recognition of youth with ‘subsyndromal’ symptoms who might be good candidates for early intervention, the term ‘cyclotaxia’ has been coined by Findling et al. [16, 17]. This construct was developed in order to begin describing youth who are at higher genetic risk (have at least one parent with BD) for developing bipolarity and who also show early symptoms of BD (thereby meet diagnostic criteria for either cyclothymic disorder or BD not otherwise specified, BP-NOS). Although these offspring of parents with a BD may not fulfill criteria for BP-I or BP-II, the extant evidence suggests that these youths may have substantive mood symptoms that lead to psychosocial dysfunction. Although affective symptomatology and impairment is frequently seen, the symptom that most specifically characterizes cyclotaxia and differentiates it from other conditions is distinct episodes of elevated mood with irritability as well as rapid mood fluctuations [17]. Although these patients with cyclotaxia are clearly in need of intervention, at present time, longitudinal data are lacking to either confirm or refute whether these vulnerable youths will indeed eventually develop BP-I.

Treatment research into genetically at-risk youths

As noted, symptomatic youths who are at genetic high risk for BD are frequently impaired. One notable study that described positive results is an open-label trial of divalproex in 24 genetically at-risk youths. Response rate was reported to be high, approximately 78% [18]. However, a key methodological limitation of that work was its open-label design.

Subsequently, Findling et al. published a study of children and adolescents, meeting DSM-IV criteria for BP-NOS or cyclothymia who also had at least one biological parent with bipolar illness [19]. Fifty-six youths were randomized to receive treatment with either DVPX or placebo. Overall, the authors found significant and equivalent benefit for both youths treated with placebo when compared to youths treated with DVPX. However, the authors also noted that youths with the highest family loading of psychiatric illness did in fact respond more favourably to active DVPX treatment when compared to placebo. Because intensive clinical monitoring was incorporated into this pharmacological clinical trial, the results suggest that some youths with cyclotaxia might benefit from psychosocial intervention and may not necessarily require pharmacotherapeutic interventions.

Psychosocial interventions for BD in youth

Because research shows that both DVPX and placebo treatment are both substantively beneficial in treating youth with cyclotaxia, the question of which psychosocial intervention or interventions may be effective in treatment of this population is an important avenue for future research. At the present time there is a growing body of literature to support several psychotherapeutic interventions in children with bipolar illness [20–22]. Currently, the limited numbers of interventions that have been developed have not been specifically studied in youths with cyclotaxia. Although psychosocial interventions have already shown promise as an adjunct to pharmacotherapy in children identified with BP-I, these forms of treatment may also show benefit as a monotherapy intervention for those with less malignant expressions of bipolarity.

Conclusion

Early onset BP-I is difficult to diagnose in children because of symptom overlap with other disorders. Additional diagnostic challenges lie in determining whether or not the clinical presentations of spontaneous dysfunctional mood states are truly prodromal manifestations of BD in these children. Further studies are needed to determine whether genetically at-risk children exhibiting dysfunctional and spontaneous affective fluctuations are in fact expressing prodromal symptoms of BP-I, or if they are suffering from other psychiatric disorders accompanied by mood dysregulation or affective instability. Regardless of whether or not these children will develop BP-I, these children deserve early and effective treatment. Fortunately, these youths appear to be responsive to a variety of interventions.

Research into the identification of children with prodromal symptoms and the advantages of early intervention are evolving and show the promise of improved long term outcomes for these youths. As in any medical illness, early intervention with effective treatment can reduce the risk of disease progression. Hopefully, with increased research regarding the accurate identification and effective treatment with children with cyclotaxia, substantive and avoidable human suffering may be prevented.

Disclosures

Dr Mao receives or has received research support, acted as a consultant and /or served on a speaker's bureau for Astra Zeneca, Bristol-Myers Squibb, Celltech-Medeva, Forest, Janssen, Lilly, McNeil, Novartis, Pfizer, Sepracor and Shire Pharmaceuticals.

Dr Findling receives or has received research support, acted as a consultant and/or served on a speaker's bureau for Abbott, AstraZeneca, Bristol-Myers Squibb, Celltech-Medeva, Cypress Biosciences, Forest, GlaxoSmithKline, Johnson & Johnson, Lilly, New River, Novartis, Otsuka, Pfizer, Sanofi-Aventis, Sepracor, Shire, Solvay, Supernus Pharmaceuticals, and Wyeth.