Translation of Science to Service: Series on Pediatric Bipolar Disorder Scientific Evolution
Mani N. Pavuluri, MD, PhD
Associate Professor, Director of Pediatric Mood Disorders Clinic and Bipolar Research Lab
Institute for Juvenile Research
University of Illinois-Chicago

Programmatic Approach

Pediatric Translational Research in Affective and Cognitive Circuitry and Treatment (P-TRACT)

VISION

• To understand the brain function and behavior in children with pediatric bipolar disorder (PBD) and determine how treatment can reverse the dysfunction.   
• The broad aspects of brain function that are the focus of our investigation are the functioning of circuits which support the feelings or emotion regulation, as well as the circuits which support thinking or cognition,  and the interface between the two.   
• Our work is driven by two objectives: Translation 1: To translate animal and human postmortem and injury findings into building models to probe brain function in PBD; and Translation 2: To translate the hardcore neuroscience findings into the creation of a forum to influence the treatment of those affected by PBD.

PHASE I: A PLATFORM TO LAUNCH OUR PROGRAM FROM THE GROUND-UP

At the beginning of the twenty first century, or of this decade, there was a huge controversy as to what bipolar disorder was, or for some, whether it even existed in children.  I felt the desire and obligation to study these children to see if we could substantiate the diagnosis with biological validation. However, at that time there was no research setting to recruit and study the brain function in children with mood disorders as well as treat them in the state of Illinois. It felt unfair and incomplete to just recruit children for research purposes without medically treating them.  In order to truly understand these children as well as treat them with safe and standardized care, we needed to develop a forum which would facilitate research of the disorder and provide a model of care with evidence-based medication and psychosocial treatment. This led to launching our pediatric mood disorders clinic - the first step in our overall vision.

The studies generated from the work undertaken at the clinic led to the development of:

(1) a parent rating scale to help diagnose a child, called The Child Mania Rating Scale by Parents (CMRS-P)

(2) a formula to help determine which medications to use, called A Pharmacotherapy Algorithm, and

(3) a therapeutic program called Child and Family Focused Cognitive Behavior Therapy (CFF-CBT) or RAINBOW therapy.

The Child Mania Rating Scale by Parents (CMRS-P)

The CMRS-P is the first parent rating scale used to assess for a potential bipolar disorder diagnosis.  It is currently the leading parent-rated measurement and is unique and helpful because parents can quickly check off the answers to 22 questions about their child’s behavior and symptoms.  Each question includes a vivid example to guide the parent or caretaker. A score above 20 suggests that a clinical evaluation is necessary for a potential diagnosis of bipolar disorder. This rating scale separates a healthy youth and an ADHD youth from a possible bipolar patient (Journal of American Academy of Child and Adolescent Psychiatry (JACAAP), 2006).

The Pharmacotherapy Algorithm is uniquely designed as a problem solving model to help determine the most effective drug regimen for the pediatric patient.  Traditionally, clinicians have used a linear approach (one silo at a time in old models vs. systematic problem solving approach where multiple problems can be solved in tandem) when prescribing medications for the child with bipolar disorder.  Our Pharmacotherapy Algorithm  begins by cleaning up potentially harmful medications. This is followed by the first line option to stabilize the child’s mood based on his clinical profile i.e., severity and type of episode. Subsequent steps are based on a multi-tiered approach, offering solutions to treat depression, comorbid conditions (such as ADHD), adverse events, and treatment resistance or tolerance (JACAAP, 2004). Treatment guidelines developed by the group of us (Kowatch et al., 2005) at a later date were very similar to the algorithm model that we had in place. One essential significant development was additional accrued data to support treating mania with second-generation anti-psychotics (SGAs) as first line agents rather than just lithium, and superseding divalproex (Depakote). This clearly shows the need to be constantly updated on developments in pharmacotherapy as research continues to reveal new and better ways to treat these patients.

The CFF-CBT is specifically designed by integrating Cognitive Behavioral Therapy (CBT) techniques with a “self psychology model” (a psychology model based on empathy and building on strengths of the child) to fit the needs of children with bipolar disorder. It involves working with patients’ strengths and fostering interpersonal relationships through empathy and compassion. Parents are taught the same tools as their child with bipolar disorder and are also taught to be their child’s coach. The techniques are highly developmentally sensitive. The catchy mnemonic, RAINBOW, summates the principal ingredients:

R-Routine

A- Affect regulation

I- I can do it

N - No negative thoughts

B- Be a good friend

O- Oh, how can we solve it?

W- Ways to get support

This program, which was written up in the 2004 issue of JACAAP, immediately gives families a format to use to find meaningful solutions to the real life stress impacting their own family. This is also described in “What works for bipolar kids,” my book written for parents with a PBD child. Dr. Amy West is spearheading the randomized trial and will describe her work in a separate article for you.  

It seemed that whatever we touched turned into pure GOLD! We are so encouraged and energized by the solid foundation to our research program. By now, we feel that we know our kids (yes, our kids that we treat!) very well and our determination to learn about their difficulties and solve the problems has intensified.

PHASE II: UNDERSTANDING BRAIN FUNCTION: ZOOMING-IN

There was an acute ramp-up of studies in cognitive and affective operations over the past several years (cognition can be interpreted as thinking and the processing ability of the brain and affect can be equated with emotion). We are beginning to understand that the brain is comprised of multiple functionally organized systems, rather than isolated regions performing discrete functions. Therefore, in a diagnostic session, I allow myself to say “Your child’s problems may not be explained by  just one disorder such as attention deficit disorder, bipolar disorder or autism; instead, he may have a little bit of each type of dysfunction, i.e., problems with attention, emotion and social skills. Often, multiple “wires” are affected in a complex fashion due to an early onset in these psychiatric illnesses in children. It’s helpful to think of your child having a brain dysfunction, rather than specific diagnostic labels which may not describe your child in exact terms.”

The following are the key contributions from our research:

I. Cognitive Difficulties are Persistent: We initially compared three groups of children: 1) Those with acute manic illness, 2) Bipolar patients who have been treated and are stable, and 3) Healthy peers.  There was no difference in intelligence levels among the three groups. We found that illness or medication status did not make a difference among patient group and all patients suffered from inattention, poor executive function (overall organization, initiation, registering information, planning and flexibility), verbal memory (ability to remember what is heard) and working memory (ability to hold information such as shopping lists for short period), whether they are ill or treated and stable (American Journal of Psychiatry, 2006). We also showed that math and reading are both affected by these cognitive difficulties (Biological Psychiatry, 2006).

This is what I say to parents when they begin to compare their bipolar child with their well child, “We are looking at apples and oranges. You cannot expect an apple to drip as an orange does as you eat it. They are both fruits, but are very different”. As a parent, you need to understand that it is not willful spite that causes your bipolar child to be disorganized with school or homework or to struggle with transitions – he simply can’t as it is not in his nature! We need to understand your child’s strengths and weaknesses based on how he/she came into the world.”

It is high time we consider cognitive function as an “extended phenotype” (Here, phenotype means the clinical manifestation of bipolar disorder such as mania or depression and the cognitive difficulties must be factored as an “extension” of those traditionally understood mood symptoms.) We have DSM criteria to identify the clinical symptoms of mania and depression. This DSM definition of the “disorder” leads to medication and psychological treatment to modulate affect. However, there is no forum for recognizing the frequently associated neurocognitive dysfunction, let alone its treatment.

We, therefore, wanted to see if optimal medication management and psychotherapy over a three year period would yield some progress. That led to our award-winning study “Neurocognitive Function in Pediatric Bipolar Disorder: Three-Year Follow-up Shows Cognitive Development Lagging Behind Healthy Youth,” This study was published in JAACAP in March 2009,  and was awarded by the Klingenstein Third Generation Award for the best paper in the Journal of American Academy of Child and Adolescent Psychiatry in the depression and suicide category, 2008-2009. 

We showed that executive functions and verbal memory did not change over time in ill patients, relative to their healthy peers. An equally alarming fact was, despite a similar degree of improvement in patients with bipolar disorder and healthy peers on attention, working memory, visual memory and visuo-spatial perception tasks, patients continued to do poorly relative to healthy peers with their baseline cognitive functioning being much lower than their healthy peers despite treatment. I want our findings to help scientists and clinicians alike to consider the potential for cognitive impairment while educating families and teachers. These hard-wired biological problems are stubborn to reverse and need to be understood in planning for IEPs in schools. I always repeat myself as I say, “Recognizing cognitive problems and reaching out to these children with compassion is like recognizing women’s right to vote!”

With persistent and carefully crafted treatment with antipsychotics in the acute phase followed by lamotrigine (Lamictal), we subsequently showed that working memory caught up with healthy peers in some; however, the group still lagged behind in executive function. We are also currently conducting non pharmacological cognitive intervention studies through computerized tasks and ecologically sound ‘real-life applicable’ strategies to see if they will help the children with bipolar disorder.

II. Social Cognition is Deficient:

The area of social cognition (or the mental processes associated with social interaction) is gaining increased attention among researchers. Although not formally listed as part of the diagnostic criteria for bipolar disorder, a number youth with bipolar disorder show significant difficulties in the ability to identify, process, and respond to the emotional states of others. Additionally, research from our clinic and that of others has documented significant peer and parent-child relationship difficulties among bipolar patients. Impaired social cognition is likely an important contributing factor to poor social interaction and emotion regulation difficulties in these youth, and therefore, an important area of study in the quest to better understand and treat bipolar patients.

We have conducted a number of studies examining emotion processing and interpersonal functioning in youth with bipolar disorder. For example, our work and that of others has demonstrated that these youth have difficulty rating the intensity of emotional expressions; they often misinterpret neutral facial expressions as being more hostile, and they require more intense emotional cues to correctly identify facial expressions (Rich et al., 2006, 2008; Schenkel et al., 2007). Interestingly, these difficulties are most pronounced for negative stimuli, and are present in both stable and manic patients, suggesting a trait-related impairment (i.e., an impairment not related to mood state).

In addition to emotion processing deficits, youth with bipolar disorder also show difficulties with the ability to recognize subtle social cues, as well as the ability to take the perspective of another person (often these problems are termed under “Theory of Mind”). As with facial emotion processing deficits, these difficulties are most pronounced in the context of negative social interactions. When involved in emotionally challenging situations, these youth lose the ability to identify what another individual might be thinking or feeling, and this often leads to their inability to meet the  expectations of others around them (often termed as poor interpersonal functioning).

We are beginning to write up our findings on recall and production of language. Negative emotions and themes in a story heard by the PBD patients directly impact the patients’ ability to recall details relative to that of positive or neutral stories. This reduced recall of stories with negative content is not observed in healthy peers. Therefore, we are beginning to note that emotions impact language production and recall in our patients.   

III. Brain Function Circuits i.e., Mapping Hard-Wire Problems

Often parents ask me if brain imaging can generate clinical biomarkers for diagnosis. Unfortunately, at present there is no way to diagnose a disorder by imaging brains. But we are in an era of research where we are able to find patterns of brain function in groups of individuals with bipolar disorder. We probe the entire brain to see what regions of the brain are activated when we ask the children to perform a specific task. This allows us to measure the increased blood flow (referred to as increased activation) to the engaged regions of the brain that participate in performing the task. It is extremely important to work out the task that elicits the brain function of a specific cognitive domain. In our first iteration of imaging studies, we designed tasks to probe emotion control, recognition of facial emotions, ability to concentrate when exposed to emotions, ability to inhibit response, and process incidental emotions.

The following illustration shows the five brain circuits affected in pediatric bipolar disorder (JACAAP., 2008): 

(1) Fronto-Limbic Circuit: We showed angry, happy and neutral faces to children with bipolar disorder and healthy control subjects while they were lying in the scanner. The amygdala (part of the limbic system, at the lower part of the brain) which receives the initial message on the type of emotion was over active when patients viewed angry faces, relative to healthy children. The prefrontal cortex (the ventrolateral part of  the frontal cortex, which is way at the top) was underactive and perhaps was unable to modulate the excessively active amygdala (Biological Psychiatry, 2007). This loss of top-down control and fronto-limbic dysfunction, specifically in relation to viewing negative faces is a key finding in bipolar disorder. These findings (strengthened by similar findings by Drs. Brandon Rich, Kiki Chang and their colleagues) lead me to say to parents, “Bipolar children can be very sensitive to negative emotions. Our research showed that these emotions can trigger a fountain of emotional reactivity demonstrated by amygdala hyper activation coupled with under activation of the top portion of the brain, or prefrontal cortex. Our findings inform us that these children struggle with modulating emotions with poor top-down control of emotion regulation. This is a biological problem that we need to understand.”

(2) Face Response Circuit: The children with bipolar disorder were underactive in the face response circuit in response to positive as well as negative faces. The underactive circuitry involves “visual cortex, superior temporal sulcus and fusiform gyrus,” conveying the initial messages on interpreting the color, shape, and facial expressions (Biological Psychiatry, 2007). This may be the reason why children with bipolar disorder have poor ability to decode others’ facial emotions, and may appear to have poor social skills (despite their not having a pervasive developmental disorder). This circuit is linked with the fronto-limbic system that is known to be further engaged in interpretation and modulation of facial emotions.

(3) Fronto-Striatal Circuit: We designed a task called “Response Inhibition” that involved an instruction to keep on pressing a button when the patients see an airplane (also known as Go Trials), or to stop pressing the button  when a stop sign pops up ( Stop Trials). The ventrolateral prefrontal cortex (VLPFC) and the striatum are overactive while exerting the effort to stop or inhibit responses (relative to the period while responding) in patients compared to healthy subjects. This helped us to understand that children with bipolar disorder need to exert increased effort in recruiting the fronto-striatal circuit (known to be involved in inhibiting responses even in healthy subjects) to do the same job as their healthy peers.   

(4) Interfacing Circuit: The thinking part of the brain or dorsolateral prefrontal cortex (DLPFC) and the feeling part of the brain (VLPFC) in the hierarchical (“CEO like” role within the brain) plane of prefrontal cortex are both closely connected, and they communicate with the amygdala at the sub cortical, plane, in the brain (Psychiatry Neuroimaging, 2008). So, we wanted to see if we could get our bipolar children and their healthy peers to match the colored words with the colored dots in order to determine how the cognitive and affective circuits interact. In other words, we asked them to focus while matching colors (a cognitive or thinking task) while the words are either neutral, negative or positive (the latter two were emotional words, which triggered emotional feelings).

While we probed this ability to perform the cognitive task under emotional challenge, patients with bipolar disorder switched off their DLPFC and VLPFC (the interfacing hierarchical region in the cortex) in response to both the negative and positive word matching (although it was more pronounced in response to negative words). This was not the case with their healthy peers.  Further, similar to the response to negative facial emotions described above, the amygdala was over reactive in response to negative word matching in patients.

These findings help me to instruct parents to consider that negative expressed emotions are likely to impact the youth’s ability to problem solve. Given the clinical reality of high level of rejection sensitivity in these children with bipolar disorder, it is imperative to understand their high degree of sensitivity to criticism when treating them.  The self psychology model is one such ‘fail proof’ model that we incorporated in our RAINBOW therapy. It may be that direct punishment/negative consequences in traditional contingency based behavior therapy may be harmful for children with bipolar disorder.    

(5)Occipito-limbic Associative Circuit: We designed a task that involved estimating the age of a face which exhibited a noticeable degree of emotion  This meant that the child was thinking hard about the age of the face while also being exposed to emotion.  In a separate  part of the task, the child was directly assessing emotion in the emotional face without divided attention.  Relative to healthy children, the amygdalas of bipolar children were over-active when estimating the age of an emotionally charged face.   This was not the case when the bipolar child estimated the emotion of an emotional face (JACAAP, 2009).  What this told us was that if even if a person is communicating something pleasant, but they have an angry face, then it is bound to activate the amygdala and cause a reactive emotional state in children with bipolar disorder. This idea may be a hypothesis applicable in real life based on our findings. However, the principle concept derived from these studies is that incidental emotions lead to amygdala over reactivity in children with bipolar disorder.   

I’d like to add a disclaimer here. We did not correlate the regions of activation to determine that they are all linked in unison in each circuit as we did not take the brain pictures in split second, but more like a repetition every 2 seconds. Nonetheless, we believe that these regions are activated while performing the task and are assumed to be functioning as a system. We are developing better time resolution in our future studies that will be allow a more sophisticated circuitry level understanding.

IV. Diffusion Tensor Imaging Studies-White Matter Cable Connections in the Brain: We compared children with bipolar disorder, those with ADHD, and healthy children by examining the integrity of white matter tracts that run across their brains.  It’s believed that the white matter connects the brain’s grey matter. Functional anisotropy (FA) is, simply put, the measure of the ability of the water molecule to slide along in a specific direction along the fiber tracts of the brain. The prefrontal regions of the children with bipolar disorder are affected quite similarly to those of ADHD patients (Biological Psychiatry., 2009). However, the white matter tract abnormalities are more extensive across seven other tracts in ADHD, which is distinctly different from those with bipolar disorder. While there are similar outward behavioral abnormalities common to those with bipolar disorder and ADHD, there may be different pathways of pathophysiology that underlie these two disorders. That is, the cause of the common behaviors may be quite different.  This work needs to be replicated in further ongoing efforts. Dr. Alessandra Passarotti, a cognitive neuroscientist in our laboratory, will describe her efforts in teasing  out bipolar disorder from ADHD through our functional studies in a separate article.        

V. Genomic Studies: We are fortunate to collaborate with our colleague Professor Ghanshyam Pandey, PhD. He is a world renowned neuroscientist. Together, we published studies showing cellular changes in patients with bipolar disorder. First, we studied the role of protein kinase C (PKC), a cellular enzyme known to be affected in adult bipolar disorder (Psychiatric Research., in press). We found that the enzymatic activity of PKC was significantly decreased in platelets obtained from medication-free PBD patients compared to healthy subjects. Eight weeks of treatment with mood stabilizing drugs increased the enzymatic activity of the PKC. These results indicate to us that decreased PKC activity may be a sign of  a constitutional predisposition towards bipolar and treatment with mood stabilizing drugs results in an increase and normalization of PKC activity along with improvement in clinical symptoms. Another very interesting finding was the levels of messenger RNA (that helps build DNA i.e. genetic material) for brain derived neurotrophic growth factor (BDNF, important for nerve growth) in lymphocytes (a type of white blood cells) of PBD subjects were significantly decreased compared with those of normal subjects, but after 8 weeks of treatment with mood stabilizing drugs these levels were normalized (JACAAP., 2008). These studies suggest that the messenger RNA for BDNF present in lymphocyte may be a potential biomarker for PBD. Lymphocyte BDNF levels may also serve as a potential treatment predictor and prognostic indicator in PBD.

These types of studies will ultimately lead to identifying intracellular biomarkers and drug discovery to reverse the cellular genomic abnormalities.   

PHASE III: MULTI-FACETED SCIENTIFIC PROGRESS: MODERNIZING MEDICINE 

Once we acquired the clues of how to further advance this work, and encouraged by our growing understanding of the pathophysiology of bipolar disorder, we transitioned into multi-modal approaches for studying brain and behavioral systems in parallel.

This means (1) Newly designed treatment studies looking at the effects of medications and how they reverse dysfunctionas well as longitudinal designs looking at disease evolution; (2) New methods of using software to model the decay or rise and fall of affect; (3) New ways of analyzing the data such as functional circuitry integration; (4) Integrating diffusion tensor imaging findings with fMRI images; and (5) Understanding the genetic effects on brain function.

We have already showed how pharmacotherapy increased prefrontal activity over the duration of treatment. We are entering into the era of demonstrating how medications alter the functioning of the brain systems along with symptomatic change. We are now in position to assess if symptomatic improvement is associated with improvement in cognitive function through the use of multiple tasks before and after treatment in our research studies.

We love to treat bipolar youth in our clinic and regularly see patients from across the globe. Some come every month and some as infrequently as every 3-6 months as they get better! Whenever possible they are recruited into our studies.  

Connection to Where There is Action:

1. Clinic Appointments Contact: Ms. Luz Martinez or Ms. Tawanna Daniels,

    312 996-7723

2. Research Inquiry Contact: Ms. Stephanie Parnes, 312 413 1710

3. My Assistant: Mr. Richard Whitley, 312 413 1722

3. Community Support in the Chicago Region: Headquartered in Chicago, but available everywhere via the internet - bpkids.org , and our own program, www.psych.uic.edu/pmdc/