Research on early-onset bipolar disorder has often hinged on identifying the key characteristics of the disorder. At a symposium on the course of bipolar disorder in children at the 2013 meeting of the American Academy of Child and Adolescent Psychiatry (AACAP), Jeff Hunt of Brown University discussed findings from COBY, the Collaborative Child Bipolar Network. He described the course of illness in 446 children with bipolar disorder, including 10% who had irritability at baseline, 15% who had elated mood at baseline, and a majority (75%) who had both irritability and elation at baseline.
Most factors such as positive family history of bipolar illness and comorbidities including attention deficit hyperactivity disorder (ADHD) did not differ across the three groups. The three subtypes (irritable, elated, or mixed) did not remain stable, and most of the children eventually converted to the combined irritable and elated subtype. These data contrast with those of Ellen Liebenluft et al., who found that those with severe mood dysregulation or chronic irritability (but not other key characteristics of bipolar disorder) did not go on go on to receive a bipolar diagnosis and tended not to have a family history of bipolar disorder.
The psychiatric community has been preparing for the 2013 release of the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition (DSM-5) by the American Psychiatric Association for years. Each new edition of the manual reflects changing conceptions of illnesses and their diagnosis and treatment.
Tom Insel, the director of the National Institute of Mental Health (NIMH), the largest research organization in the world devoted to the understanding and treatment of mental disorders, caused a stir this past spring when he wrote in a blog post that the DSM-5 lacks validity and that patients deserve better.
The DSM-5 guidelines for diagnosis of mental disorders rely on descriptions of clusters of symptoms, and Insel suggested that new methods of diagnosis that rely on laboratory measure should be developed. The NIMH is launching a project called Research Domain Criteria (RDoC) to incorporate genetics, imaging, and other data in a new classification system for illnesses, and is re-orienting their funding toward projects that “look across current categories,” for example by including all patients in a mood disorder clinic rather than only those who meet DSM-5 criteria for major depressive disorder.
A major concern about this change at the NIMH, which funds much of the research that leads to Federal Drug Administration-approved treatments, is that it will diminish funding for treatment studies in specific diagnostic categories where research is already sparse, such as childhood onset bipolar illness. This may leave many children and adults without a sound evidence base upon which their doctors can base treatment decisions.
Under the NIMH’s new rubric, clinical treatment studies to collect comparative data and evidence-based treatment research would likely lose out to studies focused on the broad collection and identification of biomarkers and the pursuit of new treatment targets. Answering an important clinical question such as whether symptoms of childhood onset bipolar disorder respond to the same medications as oppositional defiant disorder (ODD) or disruptive mood dysregulation disorder (DMDD) might not be a high priority for study.
David Kupfer, chair of the APA’s DSM-5 Task Force, responded to Insel’s statement saying that while it would be great to identify biomarkers and genetic indicators for mental illnesses, “this promise, which we have anticipated since the 1970s, remains disappointingly distant.” Insel acknowledged in his statement that this is only the beginning of development of research domain criteria, and that “for the present” the DSM will continue to be used.
At the 2013 meeting of the International Society for Bipolar Disorders, researcher Barbara Gracious presented evidence that increased levels of high sensitivity c-reactive protein (hsCRP), a marker of inflammation, were associated with an increased risk for developing a full-blown mood episode in 71 youth (average age 13.8) participating in a study called Longitudinal Assessment of Manic Symptoms (LAMS-2). The children were selected for the study because they had manic symptoms that were not severe enough to meet criteria for a diagnosis of bipolar I or II disorder. This research has not yet been published in a peer-reviewed journal, but the abstract can be found in first 2013 supplement of the journal Bipolar Disorders (page 67).
CRP levels are also known to predict cardiovascular disease and Type II diabetes.
Levels of 25-OH vitamin D, TNF?, and IL-6 did not predict a later mood disorder.
Editor’s Note: These data suggest the importance of assessing CRP and other markers in youth who are either prodromal (having early symptoms of a mood disorder) or at high risk because of a family history of a mood disorder.
The next step for clinical research would be to determine what treatment might decrease CRP and whether it would also prevent the development of mood episodes.
Irritability is common in unipolar depression. Emslie suggested that if a child’s irritability is severe and the child destroys objects and denies being irritable, bipolar disorder might be likely. Irritable unipolar depressed children will generally acknowledge being irritable.
Emslie reported that 96% of youth in his randomized placebo-controlled studies of selective serotonin reuptake inhibitor antidepressants (SSRIs) recovered from their unipolar depression, but 46.6% relapsed. Those children with residual depressive symptoms were at double the risk for relapse into a depression compared to those who remitted completely. In those without residual depressive symptoms, there were no relapses if the children stayed on their medications.
Children were excluded from Emslie’s study if they had a positive family history of bipolar disorder, and perhaps because of this, very few participants switched into mania with antidepressants.
MORAL: Treat to remission and stay on the antidepressants associated with the remission. This has previously been found to be important for adults as well. (Emslie added that he would advise that a child stay on an antidepressant for at least a year after a remission was achieved, and longer if the child had difficulties in academic performance or relationships at school.)
Children with unipolar major depression who had a few manic symptoms at a subsyndromal level had poorer outcomes in Emslie’s study. The presence of subsyndromal manic symptoms in bipolar depressed adults is a risk factor for increased switching into mania when antidepressants are added to a mood stabilizer.
Comorbid substance abuse is another risk factor for poor outcome in childhood depression.
It is hoped that measuring biochemical substances in blood will help in the diagnosis of mood disorders and help direct patients to the most effective therapeutic regimens. J.L. Billbielo reported at the 51st Annual Meeting of the National Institute of Mental Health’s New Clinical Drug Evaluation Unit (NCDEU) in 2011 that investigators from Ridge Diagnostics in North Carolina were using this type of biomarker panel in an attempt to provide a predictive algorithm for a diagnosis of major depressive disorder. The panel of 10 assays was derived from a larger screening set and included: alpha 1 antitrypsin (Alpha 1AT), brain-derived neurotrophic factor (BDNF), cortisol, epidermal growth factor (EGF), resistin, and soluble tumor necrosis factor receptor II (sTNFR2). The research group found that this optimized algorithm distinguished depressed subjects from normal controls with a sensitivity of approximately 90% and a sensitivity of 84%.
Editor’s note: This assay is available commercially and appears to represent an interesting panel of potential neurobiological markers of depression, including neurotrophic factors, endocrine stress hormones, and inflammatory markers. While its diagnostic utility is somewhat doubtful and must be further demonstrated, this editor hopes that similar panels could ultimately predict individual clinical response to a given treatment. For example, patients with high levels of inflammatory markers might respond better to treatments aimed at suppressing inflammation. Read more
Kathleen Merikangas of the National Institute of Mental Health (NIMH) gave a plenary presentation on developmental manifestations of the bipolar spectrum at the 2011 Pediatric Bipolar Disorder Conference in Cambridge, Massachusetts this past March, which was sponsored by Massachusetts General Hospital and the Ryan Licht Sang Foundation. There were several striking take-away messages from her epidemiological research. She found that:
- Rates of bipolar disorder in childhood were relatively similar to rates among adults
- Only 22% of youth with bipolar spectrum diagnoses actually obtained mental health treatment for their conditions
- There was no evidence that these children were being over-medicated, as some non-epidemiological reports had suggested
She also reported that those with subthreshold bipolar spectrum disorders, i.e. those not meeting strict criteria for BP-I (including full-blown mania) or BP-II (including hypomania for four or more days) still were very ill and had considerable disability and dysfunction.
Merikangas reported on interviews of 10,123 youth aging from 13 to 18 found in the National Comorbidity Survey Replication Adolescent Supplement (NCS-A), and found rates of illness among the youth similar to those seen in adults. However, these children with bipolar spectrum disorders were more than ten times more likely to be on antidepressants than mood stabilizers, and more than four times more likely to be on antidepressants than atypical antipsychotics, again suggesting these children were not receiving the treatments recommended by consensus guidelines.
David Axelson from Western Psychiatric Institute and Clinic gave a plenary talk on temper dysregulation disorder (TDD) with dysphoria at the 2011 Pediatric Bipolar Disorder Conference, held in Cambridge, Massachusetts in March. Researchers in the field have been discussing whether a diagnosis of TDD or severe mood dysregulation (SMD), a name Ellen Liebenluft of the National Institute of Mental Health has used to describe a similar behavior pattern, is necessary and should be included in the upcoming fifth edition of the Diagnostic and Statistical Manual of Mental Disorders (DSM-5).
The rationale for including a TDD or SMD diagnosis was the upsurge in the diagnosis of bipolar disorder among children. Researchers like Liebenluft believed that bipolar disorder was being over-diagnosed in children, and that some children could instead be classified as having a disorder that was limited to chronic irritability. Temper dysregulation disorder is what researchers eventually settled on. Post-hoc analysis of longitudinal epidemiological studies suggested that some chronic irritability experienced by children and adolescents developed into depressive and anxiety disorders rather than bipolar disorder.
However, as described in the epidemiological data of Merikangas et al. (which we will post later this week) and others, the frequency of youth diagnoses of bipolar disorder are not out of proportion with the number of diagnoses in adults. Now that it does not seem likely that bipolar disorder is being over-diagnosed among children, there is less rationale for the new diagnosis categories. In addition, it seems that TDD may not even capture a specific set of behaviors or symptoms. Read more
At the 2011 Pediatric Bipolar Disorder Conference in March, Steven V. Faraone from the State University of New York Upstate Medical University presented a plenary paper in which he described how to distinguish deficient emotional self-regulation from traditional mood disorders.
Faraone defined deficient emotion regulation as a lack of four regulating behaviors:
- Inhibition of inappropriate behavior related to strong negative or positive emotion
- Self-soothing of physiologic arousal that the strong emotions induced
- Refocusing of attention from strong emotions
- Organization of subsequent behavior in the service of an external goal
He found that deficient emotion regulation was closely related to ADHD, and progressively less similar to oppositional defiant disorder, anxiety disorders, major depression, conduct disorder and, lastly, bipolar disorder.
Deficient emotional self-regulation is associated with considerable functional impairment and is also characterized by these traits:
- Quick to anger
- Easily frustrated
- Emotionally over-reactive
- Easily excited by activities going on nearby
- Loses temper
- Argues with others
- Touchy or easily annoyed by others
- Angry or resentful
The behaviors associated with deficient emotion regulation are also prevalent among the siblings of children who have received the diagnosis. Deficient emotional self-regulation often occurs in families where ADHD is also present, but does not have a familial association with bipolar disorder or other comorbidities.
At the 57th Annual Meeting of the American Academy of Child and Adolescent Psychiatry (AACAP) in October 2010, Gianni Faedda of the Lucio Bini Mood Disorders Center reported that monitoring of activity, sleep, and circadian-rhythm disturbances can be used to distinguish children with a diagnosis of bipolar illness (who showed more hyperactivity and less sleep), from children with ADHD and control children without illness.
At the 65th Annual Scientific Convention of the Society of Biological Psychiatry this year, Giacomo Salvadore reported that significantly higher levels of uric acid are found in patients with mania compared with normal controls.
Editor’s note: This study was particularly interesting because there was a highly significant difference between patients and controls, with very few values overlapping. The data suggest the possibility that uric acid may be a useful biological marker for mania, and is one that should be studied in childhood-onset bipolar illness to determine whether uric acid is a marker for mania in children as well.
Allopurinol, a widely used treatment for gout that reduces levels of uric acid in the blood, is an effective antimanic agent (based on data from two placebo-controlled studies, one by Machado-Vieira et al. and one by Akhondzadeh et al.). The new data on uric acid raise the possibility that high levels of uric acid may be a specific predictor of responsiveness to Allopurinol, although this hypothesis has not yet been explored.
In an article by Chung et al. just published in Psychiatry Research, it was reported that in a very large epidemiological study in Taiwan, patients with bipolar disorder have increased risk of gout.