Current Treatment Options in Psychiatry

, Volume 1, Issue 1, pp 37–47

Interventions for Youth at Risk of Bipolar Disorder

Child and Adolescent Psychiatry (M DelBello, Section Editor)

DOI: 10.1007/s40501-013-0006-x

Cite this article as:
Duffy, A. Curr Treat Options Psych (2014) 1: 37. doi:10.1007/s40501-013-0006-x

Opinion statement

With the recognition that bipolar disorder (BD) develops in a series of predictable clinical stages, clinical and research focus has shifted increasingly into early intervention and prevention. The heritability of BD is estimated to be around 85 %; therefore, children of affected parents are an identifiable and important high-risk group. Lessons from early psychosis and other areas of medicine suggest that education for high-risk families regarding recognizable clinical stages and modifiable risk factors are a reasonable starting place. Specifically, reinforcing the importance of healthy nutrition, cardiovascular exercise, maintaining normal body mass index, and healthy coping strategies are important topics to cover. Early risk syndromes include sleep and anxiety disorders, which should be addressed with low-risk treatments, including sleep hygiene and individual psychotherapy. Typically, adolescence marks the age of onset of depressive disorders related to the bipolar diathesis. This is also the time when poor coping through substance abuse emerges. It is very important when assessing a depressed adolescent to understand the familial risk of psychiatric disorders. While psychotherapy is typically effective for mild non-psychotic depression, the acute treatment of moderate–severe major depression in adolescents and young adults with a confirmed family history of BD is a topic of considerable debate. The treatment decision should be taken together with the patient and family, with full discussion of the risks and potential benefits. Options include a closely monitored trial of low-dose antidepressant, discontinued immediately upon resolution of the depressive episode, or mood stabilizer that fits the patient profile, or a combination of these two, in addition to psychotherapy and reducing modifiable risk factors. When a high-risk subject manifests a diagnosable manic or hypomanic episode (typically years after the first depressive episode in late adolescence or adulthood) the question arises of whether to initiate prophylactic treatment. Potential candidates are those with a high recurrence risk and/or concern about the recurrence of a severe episode. Research has shown that most patients can be stabilized, with selected monotherapy individualized on the basis of the nature of the clinical course (episodic vs. non-episodic), quality of the spontaneous remission, and family history of psychiatric disorders and treatment response. Novel adjunctive treatments including nutraceuticals, antioxidants, and anti-inflammatory agents are being studied and may be helpful in high-risk individuals during the early stages of illness development.


High-riskBipolar disorderMood disordersTreatmentEarly interventionPreventionClinical stagingNovel treatmentAdolescentNutraceuticalsAntioxidantsAntidepressantsMood stabilizersLithium


Psychiatry is undergoing a paradigm shift in the approach to diagnosis and treatment. The early-intervention movement is rooted in schizophrenia research pioneered by McGorry and colleagues [1••], which highlighted the critical need and benefits of identifying psychiatric disorders earlier in the illness course. In parallel, neurobiological research aimed at understanding the pathophysiology of major psychiatric disorders, including bipolar disorder (BD), has advanced our understanding of biological markers associated with illness activity and helped to identify possible early-intervention targets [2•]. Collectively, the developmental early-intervention approach to diagnosis, coupled with assessment and monitoring of important risk factors and illness activity indicators, will move us toward improved individualized diagnosis, treatment planning, and response monitoring. Happily, based on overwhelming evidence, we appear to be ready to abandon the current approach to diagnosis that emphasizes cross-sectional assessment of non-specific overlapping symptoms while neglecting other important information, such as the individual patient’s clinical course, family history of psychiatric disorder and treatment response, and psychosocial and biological risk factors. The status quo yields broad heterogeneous diagnostic categories, with no utility in predicting outcome or treatment response and completely unsuited toward advancing our understanding of the onset and progress of psychiatric disorders in vulnerable individuals [3•].

This paper aims at providing a brief overview of the clinical course of BD from high-risk status to illness onset and progression. Following this, treatment implications based on the recent advances in our understanding of the natural history of BD and pathophysiology are discussed, with particular emphasis on high-risk individuals.

Observations in patients with established illness

Prospective and retrospective studies of individuals diagnosed with BD, as well as large epidemiology studies, support typical BD onset starting in adolescence and usually peaking in early to mid-adulthood [46, 7•]. The history leading to the current BD diagnostic construct has been reviewed extensively and is beyond the scope of this paper (see [8, 9•]). However, common to all historical descriptions of classical manic-depressive illness is the recurrent nature of the spontaneous illness course. The frequency of recurrence varies greatly between individual patients, and the patient’s own spontaneous course is the best predictor of future recurrence risk and cycle length (free interval) [8]. The observation of severe depressive episodes alternating with activated hypomanic or manic episodes, often associated with the same indicators of dysfunctions of central feedback mechanisms and brain arousal, have mystified clinicians and researchers alike. Furthermore, patients vary greatly in the proportion of depressed compared to activated episodes over the life course, and it is unclear what factors determine an individual’s course of illness. What we do know is that women tend to have a higher depressive/activated ratio, and untreated, the illness continues with lifelong remittance and recurrence and complications of substance abuse and very high rate of suicide [8, 1012].

More recently, there has been a reconceptualization of BD as a chronic illness, with progressively reduced cycle lengths (remissions) and significant residual symptoms. These observations support the view that BD is a neuroprogressive disorder [13••]. There are several reasons why the understanding of the natural history of the disorder has shifted in this way, including the fact that the diagnostic category has been broadened and includes psychotic and personality disorders not historically included in the diagnosis [14•, 15]. Furthermore, BD patients (typical or otherwise) have a very high risk of developing complicating substance use and medical disorders (cardiovascular, metabolic syndrome, cancer) that likely contribute to poorer outcomes and reduced quality of remission [16]. Generic treatment guidelines do not match the patient profile to the treatment approach, and contribute not only to reduced success in both acute and long-term stabilization, but are also likely to contribute to chronicity of the otherwise spontaneous clinical course [14•, 17, 18••]. In other words, if you treat a spontaneously episodic recurrent lithium-responsive form of BD with chronic antidepressants or antipsychotics, there is a risk of inducing a more chronic refractory illness course with poor quality of remission [14•].

There is already ample evidence to suggest that, as in other areas of medicine (i.e., cancer), there are identifiable subtypes of BD with different treatment response profiles, and adopting this more individualized approach to treatment has the potential to greatly improve outcomes [19]. For example, despite the current wisdom that early-onset BD represents a more severe form of illness that likely will require polypharmacy, evidence suggests that even early-onset disorder can be stabilized with selected monotherapy, chosen on the basis of family history of psychiatric disorders and treatment response in other affected family members, alongside the clinical course of the patient and developmental history [17, 2022].

Observations from high-risk offspring

Similar to other major psychiatric and medical illnesses, BD is a complex trait resulting from the interaction of a number of risk factors in genetically susceptible individuals [23••]. The heritability of BD is the highest of all psychiatric disorders, estimated at about 85 % [24]. Therefore, studying the biological offspring of affected parents is an important strategy to advance understanding of the onset of illness and to map the early clinical course.

There have been reports from several independent international prospective studies of high-risk offspring that have informed our understanding of risk syndromes predicting the onset of and describing the early clinical course of BD [25, 26•, 27]. Specifically, anxiety and sleep disorders (indicative of circadian rhythm disturbance) appear to be early risk syndromes in confirmed high-risk offspring, predicting the later development of major mood episodes [28, 29]. In offspring of parents with a more psychotic spectrum form of BD, childhood risk syndromes also include neurodevelopmental disorders such as learning disabilities and attention deficit hyperactivity disorder [30•, 31].

BD-related mood disorders in offspring with an affected BD parent are most often in the depressive polarity [5, 32]. This, of course, causes a clinical conundrum because, by convention, BD is not diagnosed until there is a documented activated episode. However, given the overwhelming evidence from clinical, high-risk, and epidemiological studies that BD most often debuts as major depressive episodes, major depression in high-risk offspring is increasingly accepted as the onset of a BD-related mood disorder. This is important because these patients are at greater risk for paradoxical responses to high doses of antidepressants and/or stimulants often prescribed on the basis of presenting clinical symptoms without accounting for family history and likely illness trajectory [33, 34]. This concept is supported by the fact that family studies have estimated that recurrent major depression in first-degree relatives of a BD patient has an 80 % likelihood of being related to the BD trait segregating in the family [35, 36].

Observations from high-risk studies are consistent with a clinical staging model as articulated by Duffy and colleagues [37, 38], which maintains that BD in high-risk individuals (with a first-degree relative with confirmed BD) may be preceded by childhood risk syndromes – including sleep anxiety and, in some cases, neurodevelopmental disorders – followed by minor affective disorders in early adolescence and major mood episodes typically starting in mid to late adolescence. Not all high-risk individuals manifest every stage, but once the illness does manifest, clinically psychopathological development follows this predictable sequence. Specifically, diagnosable activated episodes (BDNOS, BDI, or II) tend to follow the initial depressive episode by several years, and rarely prior to puberty. Of concern is the fact that substance abuse is emerging much earlier in the course than in previous generations, with the drug of choice shifting from alcohol to cannabis [39]. The deleterious effect of substance abuse on the clinical course and treatment response is highly significant, and this complication should be aggressively prevented or treated.

Treatment in high-risk individuals

As mentioned above, the movement toward earlier diagnosis based on an evidence-informed appreciation of the natural history of psychiatric disorders has advanced the possibility for early intervention and prevention research and clinical practice. However, most of the research to date has focused on individuals at high risk for developing schizophrenia and other psychotic disorders [1••, 40, 41]. Nonetheless, this research, in conjunction with findings from studies of the pathophysiology of BD, provides us with an evidential and rational basis from which to develop treatment approaches for clinically unwell individuals at known risk of developing BD and related mood disorders (see Table 1).
Table 1

Treatment approach for clinical stages of developing bipolar disorder


Clinical Stage

Intervention Options


Well at genetic risk

○ Optimize nutrition, sleep, exercise and relaxation routines

○ Education*

○ Healthy coping strategies**


Non-specific anxiety & sleep disorders

○ Cognitive and supportive psychotherapy

○ Healthy coping strategies**

○ Consider nutraceuticals (e.g., omega fatty acids)/antioxidants

○ Consider short-term sleep aids (i.e., melatonin)


Subaffective & adjustment disorders

○ Psychotherapy***

○ Optimize nutrition, sleep, exercise and relaxation routines

○ Healthy coping strategies**

○ Nutraceuticals (e.g., omega 3 fatty acids)/antioxidants


Major depression

○ Psychotherapy***

○ Substance use avoidance

○ Healthy coping strategies

○ Nutraceuticals/antioxidants

○ For refractory or moderate to severe depression, consider either short-term closely monitored antidepressant

If psychosis, add short-term low dose atypical


Consider selected mood stabilizer****

○ If prominent suicidal thoughts, consider lithium either as adjunct or monotherapy


Bipolar Disorder

○ Selected mood stabilizer****

○ Psychotherapy***

○ Psychoeducation

○ Family education & support

○ Healthy coping strategies**

○ Long-term follow-up in specialty clinic

*Education about modifiable risk factors (substance abuse) and protective factors (reduce stress), risk syndromes and natural history of disease and treatment options

**Healthy coping strategies include sleep hygiene, cardiovascular exercise, maintaining normal BMI, avoidance and cessation of substance abuse

***Psychotherapy includes cognitive behavioural therapy, mindfulness/rumination, interpersonal therapy, social rhythm and supportive counseling

****Mood stabilizer selection should be individualized and based on the patient profile (developmental history, clinical course, quality of spontaneous remission, psychiatric disorders and response in other family members)

Early-stage treatment – education, healthy lifestyle choices, psychotherapy

It is the logical assumption that intervention earlier in the course of an evolving major psychiatric disorder should have a more favourable benefit/risk ratio [42]. There is accruing evidence that individual and family psychotherapy (CBT, mindfulness-based, supportive) and education is helpful in reducing acute and residual symptoms of anxiety and depression in depressed and bipolar patients [4345]. More recently, there has been preliminary evidence that these interventions may be helpful in symptomatic high-risk individuals, although more systematic study is needed [46, 47]. In regard to this latter point, there is evidence that cardiovascular exercise, stress reduction intervention, sleep hygiene, and substance abuse prevention all have positive early-intervention and prevention effects (see [2•, 16, 48, 49••, 50]). While more systematic research is needed to understand the underlying mechanisms of the positive effects of these interventions and whether they significantly prevent or delay the progression of BD, it seems reasonable to assume that these low-risk, acceptable, and well-tolerated interventions may be helpful in improving quality of life in the short-term [51].

Early-stage treatments – anti-inflammatory, antioxidants, nutraceuticals

Since the 1990s, there has been increased recognition of and interest in inflammation as a central genetically sensitive pathway associated with the onset and progression of depression and BD [49••, 5254]. A pro-inflammatory state has been associated with acute mood episodes, and successful treatment has been shown to re-establish the balance between anti- and pro-inflammatory markers [1••, 13••, 55]. Furthermore, recent studies have reported increased expression of inflammation-related genes in offspring of BD parents [56•].This, along with other findings, suggests that subtle abnormalities in the immune system may be present in high-risk offspring [57]. These observations are extremely compelling when taken together with other abnormalities reported in BD patients and their high-risk offspring in the neuroendocrine [58, 59] and neurotransmitter systems that are moderated by the immune system [49••, 6062]. While levels of peripheral inflammatory mediators are not grossly abnormal (up to two/threefold general population and broadly within normal range), it has been postulated that a mild pro-inflammatory propensity, combined with other risk factors in vulnerable high-risk individuals, may explain the onset and progression of illness [49••]. In addition, over the course of recurrent illness episodes, prolonged inflammatory imbalance may lead to more persistent functional and structural changes via oxidative stress and neurotrophic pathways [13••].

Knowledge regarding the pathways involved in the development and progression of BD raises the possibility of novel early-intervention approaches and serves as clues for mechanisms of known effective treatments. For example, psychotherapy, meditation, and exercise are associated with a reduction in peripheral inflammatory markers (see [49••, 50]). There have been some favourable studies of a number of agents with anti-inflammatory properties (i.e., cyclooxygenase-2 inhibitor, TNF inhibitor, acetylsalicylic acid, minocycline) as adjunctive treatments in otherwise-healthy depressed unipolar and bipolar adults (see [2•, 49••, 63]). As of yet, these interventions have not been systematically studied in high-risk populations.

There is increasing interest in antioxidant and nutritional supplements due to their safety, acceptability, tolerability, and relevance to the risk of psychiatric and related medical illnesses [64]. Evidence from peripheral and central nervous system samples suggest that disruption in antioxidant enzymes, increased lipid peroxidation activity, and decreased levels of antioxidants (including vitamins C, E, and D, zinc, and coenzyme Q10) are related to the propensity to psychiatric disease, including depression and BD [65]. N-acetylcysteine (NAC) has been shown in both basic and clinical studies to reduce markers of oxidative stress and modulate inflammation, with positive effects in bipolar depression when used as an adjunctive treatment [2•, 66, 67]. A recent study provided evidence that NAC may be an effective cessation treatment for adolescents with cannabis dependence [68].

The most-studied nutraceuticals with relevance to the prevention and adjunctive treatment of psychiatric disorders are the long-chain omega fatty acids (OFA). OFAs have been shown to have anti-inflammatory, antioxidant, and neuroprotective effects and associated membrane stabilization [69], and to interact with candidate neurotransmitters (serotonin and dopamine) in both basic and clinical studies [7072]. In a landmark study, a 12-week trial of 1.2 g/day of long-chain omega-3 polyunsaturated fatty acids (PUFAs) prevented conversion to full-blown psychosis for up to one year in individuals at high risk for psychosis, with a number needed to treat of 4 [73••]. In a recent post hoc analysis, therapeutic effects of this 12-week double-blind placebo-controlled trial were evident by four weeks on positive symptoms and 12 weeks on negative symptoms and global functioning [74].

Treatment of later-stage illness (major mood episodes)

Treatment of major depressive episodes in an individual at confirmed familial risk of BD is a controversial topic, and also one in need of more systematic study. These episodes typically occur in high-risk individuals who are at least in their adolescence and do not occur in pre-pubescent children. Adolescent depressive episodes in those with a bipolar diathesis have been described as having an abrupt onset, melancholic and psychotic features, and agitation associated with antidepressant treatment [75••, 76]. Given the lack of substantial evidence directly comparing the risks and benefits of treating adolescent depression as per guidelines or based on the familial risk of BD, it seems reasonable to suggest a trial of psychotherapy administered with either a low-dose short-term antidepressant (plus antipsychotic if psychosis is present), carefully monitored for paradoxical worsening or activation with psychotherapy, or a trial of the minimum effective dose of the mood stabilizer that was effective in the other family members (i.e., lithium, atypical antipsychotic, anticonvulsant) with psychotherapy [20, 21].

Continuation treatment and prophylaxis for recurrent mood disorders

Critical emphasis in treating high-risk youth should be given to involving the patient and family in the treatment plan, which should include a full discussion of potential risks and benefits of any suggested treatment and the state of evidence or lack thereof. The limited data available suggest that the chronic use of antidepressants is not helpful in this population, and sometimes is harmful, associated with chronic depression, rapid cycling, or activation. Furthermore, the continuation of mood stabilizers after stabilization of the acute episode needs to be weighed against the recurrence risk, tolerability, acceptability, and effectiveness in individual patients. If the recurrence risk is unknown or of low frequency, a reasonable option would be to discontinue the mood stabilizer after the remission of the acute episode and follow the patient clinically, providing education, support, and counseling, along with incorporation of more benign interventions as described above. Further, biomarkers of illness activity and recurrence risk, such as the dexamethasone suppression test (indicating HPA activation) and peripheral immune markers (immune activation), can be incorporated into the prospective follow-up, along with daily ratings of mood, anxiety, sleep, and energy.

The option of long-term prophylactic medication should be discussed with patients who have an established risk of recurrence. Unselected prophylactic treatment has low success and tolerability rates. Selected prophylactic treatment based on individual patient profile, however, is associated with very high success and tolerability rates [77]. Evidence suggests, for example, that BD patients who respond to long-term lithium prophylaxis are characterized by an episodic remitting illness course, a family history of recurrent mood disorders (not schizophrenia) or poorly remitting psychotic disorders, and a family history of lithium response and classical acute episodes with or without psychosis (lithium has antipsychotic effects) [78, 79]. Lithium is also very effective in treating suicidal thoughts and behaviour, even as an adjunct in patients who would otherwise not be good long-term responders to lithium monotherapy [10, 14•]. Compliant lithium-responsive patients can often be taken off treatment and monitored for recurrence clinically and biologically, as described above, and restarted at times of high recurrence risk (sleep disruption, depressive or activated symptoms, abnormal DST) [18••]. This requires good therapeutic alliance and close monitoring. Any medication should be used in conjunction with education, psychotherapy, and exercise, and carefully monitored to prevent weight gain and other medical disorders for which BD patients are at increased risk, including metabolic syndrome, type II diabetes, and cardiovascular and thyroid disease ([80, 81]).


A better understanding of the natural history of mood and psychotic disorders has underscored the dramatic unmet need for more vigilant and informed research in earlier identification of serious psychiatric disorders and providing effective and acceptable treatment to alleviate acute suffering and prevent illness progression. Longitudinal studies of high-risk youth has further highlighted that there are reliably identifiable at-risk populations, based on family history or clinical symptoms or both, that should be provided with education and clinical support, preferably with effective interventions that are safe and well-tolerated. While there is a major gap in our knowledge regarding the efficacy of some interventions with a favourable benefit/risk ratio, some evidence exists in the general and other high-risk populations that can inform rational early-intervention strategies for youth at risk of BD-related mood disorders. Clinical staging lends a conceptual framework that guides these early interventions, as well as future research on associated biomarkers and effectiveness mechanisms. What is astounding to me is that we are still having the same debates about the “risk” of over-labeling and the cost of high-risk surveillance programs, when these would not even be questioned in other areas of medicine, such as cancer care. What is very positive is the increasing recognition in psychiatry of the need for a developmental approach to diagnosis, an essential strategy for advancement of early-stage intervention, and for furthering our understanding of the onset and progression of these disorders.

Copyright information

© Springer International Publishing AG 2014

Authors and Affiliations

  1. 1.Department of Psychiatry, Mathison Centre for Mental Health Research & Education, Hotchkiss Brain InstituteUniversity of CalgaryCalgaryCanada