Gender Differences in Bipolar Disorder Across the Life Span Through an Intersectional Lens
TAMARA LEWIS JOHNSON: Good afternoon and welcome to the 2021 National Institute of Mental Health Women’s Mental Health Research Webinar. The topic and the title for the webinar this afternoon is Gender Differences in Bipolar Disorder Across the Lifespan through An Intersectional Lens. I want to say that this webinar is sponsored by the National Institute of Mental Health’s Office for Disparities research and workforce diversity. My name is Tamara Lewis Johnson and I am the Chief, of the Women’s Mental Health Research Program at the Office for Disparities and Workforce Diversity at the NIMH. I am responsible for providing guidance and advice on matters relating to women's mental health research.
This afternoon we are spotlighting the research of Drs. Hillary Blumberg and Crystal Clark. This research was funded by NIMH’s Division of Translational Research and the Division of Services and Implementation Research.
Two percent of the world's population is living with bipolar disorder, which is twice the rate for schizophrenia. Although the prevalence of this mental health condition is equally prevalent among males and females, there are unique gender differences that affect quality-of-life and emergence and persistence of the condition.
Studies indicate an increased risk of women in Bipolar 2 and hypomania rapid cycling, and mixed episodes, and there are important gender differences found in comorbidity. Pregnant women who have a history of bipolar disorder are 23 times more likely to have a recurrent episode during the postpartum period. In addition, bipolar disorder is likely to worsen with age, if the condition goes untreated.
The speakers this afternoon will highlight NIMH funded research in this area, with a focus on translational research and customizing combination therapeutic interventions for women and men who are at risk for or are living with bipolar disorder.
Next slide. I am going to introduce our speakers, first I am going to introduce Dr. Hilary Blumberg. She is the John and Hope Furth Professor of Psychiatric Neuroscience, Professor of Psychiatry, Radiology, and Biomedical Imaging in the Child Center, and the Director of the Mood Disorders Research Program at Yale School of Medicine. She graduated summa cum laude in neuroscience from Harvard University and completed her medical degree, psychiatry training, and specialty training in brain scanning research at Cornell University Medical College.
Dr. Blumberg’s research is devoted to understanding the brain circuitry differences that underlie mood disorders across the lifespan, with a focus on bipolar disorder. She directs the Mood Disorders Research Program at Yale, which brings together a multi-disciplinary group of scientists to study the genetic, developmental, and environmental factors that cause mood disorders to develop new methods for early detection, more effective interventions, and prevention of the disorders and their associated high risk for suicide. This research includes the use of state-of-the-art brain scanning methods. The program is also known for training young scientists to be new leaders in the field.
Dr. Blumberg has served as principal investigator on awards for the NIMH, the National Institute on Drug Abuse, the Department of Veterans Affairs, the American Foundation for Suicide Prevention, Brain and Behavior Research Foundation, International Bipolar Disorder Foundation, For the Love of Travis Foundation, the MQ Foundation, Stanley Medical Research Institute, and Women’s Health Research at Yale.
She has received numerous awards, including the 2017 Brain and Behavior Foundation Colvin Prize for Research Achievement in Mood Disorders and the 2018 American Psychiatric Association Blanche F. Ittleson Award for outstanding and published research in child and adolescent psychiatry. She is a fellow of the American Psychiatric Association and a member of the American College of Neuropsychopharmacology and Society of Biological Psychiatry.
And now I will introduce Dr. Tennille Clark. Dr. Clark's research interests focus on women and mood and anxiety disorders across the reproductive lifecycle. She is an associate Professor of Psychiatry and Behavioral Sciences and Obstetrics and Gynecology at Northwestern University. Dr. Clark's passion for optimizing treatment in pregnancy for women with bipolar disorder has led her to her current research focused on dosing lamotrigine and lithium in pregnancy.
She is also interested in optimizing dosing of medication for mood and anxiety disorder, and other times of hormonal changes, including those related to infertility treatments and use of oral contraceptives. She maintains an interest in trauma associated and traumatic birth as it relates to the reoccurrence of mood and anxiety symptoms in pregnancy due to past trauma and I am delighted to have both Dr. Clark and Dr. Blumberg to present today.
But before they present, I have a few housekeeping things to share with the audience.
Participants will be muted in silent only mode and cameras will be turned off. Please submit your questions to me via the Q&A box at any time during the presentation. You will find the Q&A box at the bottom of your screen. If you have technical difficulties hearing or viewing the webinar, please note these in the Q&A box and our technicians will work to fix the problem. You can also send an email to the email address online here, as you see on the screen.
Next slide please, I also wanted to make the viewers aware that our Office of Disparities Research and Workforce Diversity has several more webinars throughout the summer. This is the first of our series and so on June 29th, you will see the Strength in the Face of Challenge: Youth Suicide Prevention Research Among the White Mountain Apache and Navajo Nation in The Time of COVID-19. Then in July there will be the NIMH’s Jackson Memorial Award, which will focus on mental health equity research. Then in August, there will be a talk on navigating a neuroscience career for scientists of color, with a focus on research training and in September there will be a webinar focused on sex and gender minority in mental health research focused on youth.
I encourage you to sign up for all of these at the registration site that is on the screen here.
With no further ado, Dr. Blumberg, I am delighted to have you proceed with your presentation.
HILARY PATRICIA BLUMBERG: Thank you for the kind introduction, and many thanks to Tamara Lewis Johnson and the organizers of this webinar for including me. They have chosen an extremely important area for which more research is needed. There are three major points that I hope you will take away from the presentation today. First, we have made major strides in understanding the brain and bipolar disorder that are informing research on new early detection, and intervention strategies so the future is very hopeful that better help is on the way.
Second, there are substantial gender differences in clinical features of bipolar disorder. Although we have only begun to scratch the surface, of underlying brain differences and mechanisms. And third, it is critical that new research initiatives are undertaken, designed, specifically to understand gender differences through an intersectional lens.
With this talk, first, I will outline the clinical features of bipolar disorder, that differ between genders. Second, I will present neuroimaging data supporting the working model of the brain systems involved in bipolar disorder. I provide examples where there is initial evidence of gender differences and of contributing biopsychosocial factors. Finally, I will touch very briefly on the implications of these data for treatment.
Bipolar disorder is highly prevalent, but with similar rates in females and males, it is highly heritable, although genetic predisposition may not be the whole story in its expression. It has a peak onset in adolescents, and young adulthood, and multiple factors including environmental, are implicated in the brain trajectories in bipolar disorder. Some early studies suggested higher rates of maternal transmission of interest given a resurgence of study of mitochondrial mechanisms. However, more recent findings on maternal transmission are mixed. With adolescent onset, sex hormones have been implicated that could have gender related effects.
Bipolar disorder has shown important gender differences in its clinical features. Features higher in women, include depressive and mixed episodes, rapid cycling, anxiety and eating disorder comorbidities, medication resistance, and further complicating treatment, greater rates of antidepressant induced mania have been reported. Bipolar disorder is associated with early mortality, from suicide, estimated to be a preventable form of early mortality in about one in five individuals with bipolar disorder, and more women have suicidal ideation and attempts. Multiple medical comorbidities also cause early mortality in bipolar disorder. Including cardiac, and metabolic disturbances worsened by higher rates in women with bipolar disorder of abdominal obesity that can lead to deleterious inflammation.
These factors may also relate to Alzheimer's disease rates that are higher in women and in bipolar disorder. Multiple interacting biopsychosocial factors are implicated in gender differences in clinical presentations of bipolar disorder. These include developmental and aging, hormonal, autoimmune, inflammatory, metabolic, and changes in pharmacokinetics that Dr. Clark will be talking about, as well as stressors and numerous social determinants of health. These interact with life phase, such as during both biological and psychosocial changes that occur from puberty to the peripartum, to menopausal times, which Dr. Clark will be talking more about.
Premenstrual worsening has been reported clinically but it is has been difficult to generate consistent supported evidence, that perhaps can be remedied using new digital technology for longitudinal study, which I will touch upon this at the end of my talk.
So, onto the brain, and how neuroimaging studies have informed the field. Consistent with emotion dysregulation as central to bipolar disorder, brain regions consistently shown to be involved in bipolar disorder are ones that subserve emotion processing and regulation, especially the amygdala and the ventral prefrontal cortex, which have substantial bidirectional inhibitory connections that are also implicated in bipolar disorder pathology.
However, these structures function within the context of more widely distributed brain systems that include regions that have shown preliminary evidence of gender differences in bipolar disorder. Such as the hippocampus, prefrontal cortex, and cerebellum. I will also get back to the hypothalamus, later in the talk, which has important connections in the system.
Throughout the neuroimaging work I will present, for the most part sample sizes are quite modest. This has constrained ability to perform meaningful analyses to investigate gender. Although, even when sample sizes might have been sufficient, gender was often co-varied out and not examined as a variable of interest.
I will be presenting some of the limited gender related findings, that can help us start considering relevant brain regions and connections that necessitate further direct study of gender. For example, amygdala differences in adolescents and young adulthood, are among the most consistent biological findings in the disorder. In youths, amygdala volume is decreased and there are excessive responses to emotional stimuli.
However, in adulthood, results are less consistent and the reasons for the heterogeneity are not clear. Biological and environmental factors such as stress, medication, and substances, are suggested. When volume increases in adults are reported, often they are in males who are psychotic, and have been exposed to antipsychotic medications and lithium. There have been increases in amygdala volume observed with age in young males but not females with bipolar disorder but this finding has not been a well replicated finding, as of yet.
In hippocampus, smaller volume has been seen only in girls with bipolar disorder. Although there are some other variable results, and especially in adults. Given the hippocampus’ sensitivity to stress, as well as its capabilities for plasticity, such as in response to mood stabilizing medications, there are many potential factors that can be influencing the hippocampus in bipolar disorder over the lifespan.
Stress is an environmental factor for which there has been evidence of gender-related brain effects. For example, our team observed preliminary evidence for gender related effects of child maltreatment, physical and emotional, with adolescent girls showing more involvement of hippocampus and amygdala and frontal regions important in internalizing emotional experiences but males, more dorsal reasons subserving impulse control and externalizing behaviors and perhaps this contributes to differences that have been observed in internalizing and externalizing behaviors between girls and boys exposed to trauma.
Gender related effects of child maltreatment and bipolar disorder extend into adulthood. Increasing adverse outcomes, including suicide. Both women and individuals with bipolar disorder are more likely to report having experienced child maltreatment and in our work under review, only woman with bipolar disorder showed decreases in hippocampus related to child maltreatment. This is a possible cause of higher rates in females of comorbidities, such as anxiety.
This finding in women only, highlights the importance of studying women and men and performing analyses specifically through gender. HPA axis mechanisms are implicated including stress hormones and their interactions with estrogens, and later life adversity can also adversely affect the brain and may interact with aging processes. A subset of individuals with bipolar disorder can show progressive cognitive dysfunction in later life and stress and endocrine interactions are implicated in Alzheimer's disease, which has higher rates in women and in individuals with bipolar disorder.
So these are areas for study which could be fruitful to understand and decrease risk for cognitive impairment and dementia development. Stress clearly has an impact on the brain and in bipolar disorder, and there are gender related effects. But study of stressors has been highly limited and will be of critical importance to expand this area of research.
Returning to development, during adolescence and young adulthood, the epoch when bipolar disorder typically develops, frontal cortices and their white matter connections are dynamically maturing. We observed evidence for an exaggeration of grey matter pruning and less increases in the structural integrity of white matter during adolescence and young adulthood in bipolar disorder. Gender related factors can affect the development of frontal cortex and its connections such as hormones, substances and stress, could affect these frontal trajectories altering clinical features in gender related ways.
Though the frontal cortex has been studied a good deal in bipolar disorder, rare gender related frontal findings have been reported. A few reports suggest there might be greater frontal abnormalities in women with bipolar disorder, for example, in responses of right, venture lateral, prefrontal cortex to emotional phases that were lower in adult woman with a polar disorder and in data from colleagues in China, women with bipolar disorder were also observed to have greater magnitude of functional and structural dysconnectivity from the amygdala to the prefrontal cortex.
Interestingly consistent with that, recent preclinical research suggests there may be sex differences in glial gene expression in prefrontal cortex. That might contribute to clinical phenotypes related to the frontal dysconnectivity, which is further related to psychotic symptoms.
These abnormalities that tend to progress in bipolar disorder during adulthood tend to be deficits in lateral ventral and dorsal prefrontal regions, subserving emotion and cognition. Whereas early abnormalities tend to be of excessive responses. In the amygdala and medial and coddle, orbitofrontal, limbic and paralytic regions. These may relate to one of the still leading models of development in bipolar disorder, the sensitization model proposed by Bob Post and colleagues at the NIH decades ago. They creatively brought together several observations. They noted that similar to pattern to pattern and sensitization processes, untreated episodes of bipolar disorder could lead to an increase in the severity and frequency of subsequent episodes. Individuals with excessive limbic and paralimbic activity because of seizures, had symptoms similar to what is seen in bipolar disorder.
So noting that anticonvulsants could lead sensitization processes, they proposed anticonvulsants to treat bipolar disorder, which are currently a mainstay of treatment. Some anticonvulsants such as valproate, may be especially helpful for rapid cycling, which is more frequently seen in women with bipolar disorder.
Although, as Dr. Clark may discuss further, considerations have to be taken into account such as possible relationships of polycystic ovary disease and reproductive planning. Higher sensitization effects in females have been shown across species, in rodents, nonhuman primates, and humans that are long-lasting and can occur from excessive catecholaminergic inputs.
And in this figure, two amphetamines in women, effects that are estrogen dependent. This raises the possibility that medication and substance exposures that might increase sensitization might also increase risk of clinical features seen more in women, such as rapid cycling and mixed episodes. Moreover, cross sensitization with stress has long been posited as shown in the top figure.
Hypothyroidism, including with the autoimmune cause of anti-thyroid antibodies are another mechanism that could lead to a variant cortical limbic functioning in women with bipolar disorder linked to rapid cycling. Women with bipolar disorder are more likely to have hypothyroidism and antithyroid antibodies which have been linked to rapid cycling and, preliminary PET evidence suggests the antibodies are also related to elevated anterior singulate cortex metabolisms.
These thyroid processes are a potential place, multiple gender related mechanisms converge that could influence bipolar disorder phenotypes. Including endocrine, thyroid hormones and their interactions with estrogen, autoimmune and metabolic and also sleep irregularities which can be potent in precipitating and worsening episodes of bipolar disorder and increasing suicide risk.
The cerebellum has had a relatively little study in bipolar disorder. Though, it has connections to the amygdala and prefrontal cortex, it shows effects of lithium, and is increasingly implicated in emotional behavior. In one study of the cerebellar vermis, we noted higher volume only in males with bipolar disorder. It is possible this could relate to phenotypes in men, including the lithium responsiveness, although it could also relate to exposures such as to lithium or other medications.
Gender related features of suicide thoughts and behaviors have long been known. We observed greater frontal system, structural, functional and connectivity abnormalities in adolescent and young adult suicide attempters with bipolar disorder. In this study of modest sample size, gender related effects were not detected. However, we recently performed a review of 113 neuroimaging studies of suicide thoughts and behaviors from the past two decades. We were struck by how few investigated gender, and either did not attend to it or co-varied out sex or gender. Given the expensive neuroimaging studies, cost can be a culprit, with limiting samplings sizes. However, this is an important issue to solve as study of gender differences in suicide thoughts and behaviors in bipolar disorder can help prevent the devastating outcome of suicide.
In large scale national and international neuroimaging consortia, increased sample sizes can allow for novel comparisons between females and males. Results from the enigma consortia studies of bipolar disorder in thousands of individuals, showed thalamus increases only in females with bipolar disorder, and less pronounced cortical thinning in females with bipolar disorder only if they were under age 26 years. These results differ from some previous smaller scale studies in the field for unclear reasons. While it is tempting to assume that adding analysis of sex and gender to large-scale studies will be the solution, there are limitations, and the field needs to proceed with caution. Technical issues include the use of different scanners, and scanning techniques, and large regions of interest that could obscure findings.
Importantly, most studies were not designed to investigate gender effects. There are many considerations that need to be taken into account, including multiple intersecting factors so studies specifically designed to assess for gender differences and intersectionality are critically needed.
One of the few studies to consider gender in an imaging study of bipolar disorder, and also to consider development as well as comorbidities, showed preliminary perspective evidence that girls who develop substance misuse, here -- alcohol and cannabis, have decreases in grey matter in the area in the red ovals, more in brain regions subserving emotion regulation. Whereas, with boys, showed more decreases in blue ovals, in a dorsal area, involved in impulse control, suggesting different pathways to this substance abuse comorbidity.
Other relatively unexplored areas of study that could be fruitful in uncovering brain gender differences in bipolar disorder, especially with new methods to study structural and functional connectivity are hemispheric lateralization and inter-hemispheric connectivity. Some functional domains have more bilateral brain representation in females, these include language, which could have implication for higher rates of dyslexia in male, and improved ability in females to recover from stroke and language areas via contralateral compensatory abilities.
This highlights the importance throughout to consider not only disadvantages of some factors that we’re discussing today, but also potential advantages, such as in resilience and recovery.
Females have also been suggested to have higher bilateral processing of emotions, might this have adapted as well as maladaptive sides to it? For example, we and others noted decreases in the corpus callosum and the white matter structure measure and bipolar disorder. We also noted increased functional connectivity between the left and right prefrontal cortex. We recently observed it related to mood episodes, especially depression. Might it relate to the vulnerability to cycling and particularly, to depression in females? This needs further study.
The hypothalamus has been relatively elusive to study with traditional neuroimaging techniques but has been highly implicated in bipolar disorder and particularly in multiple gender related factors including endocrine and stress related. The hypothalamus has major projections to both the amygdala and ventral prefrontal cortex. Sleep disruption has been linked to increased inflammatory markers, and to acute mood episodes with affects that may be greater in females. Sleep irregularities are robust risk factors and therefore important targets to prevent and reduce mood symptoms and prevent psychiatric and medical comorbidities and suicide.
During the talk I have just touched briefly on pharmacological treatments, Dr. Clark will be talking more about them. As I near the end of the presentation, I will focus on a psycho-behavioral intervention that our laboratory is studying.
The intervention BE-SMART, is a variation of social rhythm therapy or SRT, a component of IPSRT. Initial findings from an NIMH funded grant to study BE-SMART provide preliminary evidence that by regularizing sleep and other daily rhythms via an intervention delivered largely through video teleconferencing, it can be robust in decreasing both depressive and manic symptoms as well as hopelessness and suicide propensity. We are performing neuroimaging, which is providing preliminary support of beneficial brain changes in bipolar disorder, in which excessive amygdala responses are quieted by increased engagement of the ventral prefrontal cortex through mechanisms which may include interactions with the hypothalamus and In collaboration with the NIMH Intramural Program, we additionally incorporate digital technologies, including smart phones and wearables such as the Actigraphy watch shown here which enables collection of rich longitudinal data at a high temporal resolution and individuals in their usual settings so that we can better understand time patterns and changes and catch changes as soon as they occur to prevent worsening.
In summary, progress has been made in understanding the brain circuitry of bipolar disorder and its development and researchers are beginning to piece together pathways from multiple genetic variations to differences in clinical presentations that can show substantial gender differences. However, there remains a great deal of work to do and studies designed specifically to elucidate gender related influences are needed. Moreover, there are multiple complex interactions including biopsychosocial factors, including numerous important social determinants of health, that will require careful study and multivariate analytic approaches to move towards more effective personalized treatment approaches.
There is quite a positive and hopeful side. The field is rapidly increasing knowledge on wholistic approaches to bipolar disorder, and other mood disorders and reduction of risk and identifying multiple entry points to have beneficial effects on these modifiable interacting factors and these can include low risk talk and behavioral therapies with potential for wide dissemination.
We need a great deal more research. Importantly, while it is helpful to do secondary analyses on already collected data, these have major limitations and there is critical need for studies designed specifically to investigate gender differences through the lifespan and through an intersectional lens. This includes a great need for longitudinal research that may be facilitated by new technologies. A key aspect of bipolar disorder is that it is characterized by changes in time, over months and years, in thinking about episodes and progression, but also shorter scales such as days and diurnal variations, and even minutes, in acute responses to stress.
Importantly, transitions, are times of greatest vulnerability to symptoms and suicide, but often the most challenging to study. In the next talk, by Dr. Clark, you will hear about the highly vulnerable peripartum period during which rapid changes need to be captured for studies.
Finally, it will be critical to broaden engagement. It is critical to understand the many varying biopsychosocial factors which can lead to vulnerability and which can also lead to resilience. For example, stressors can vary greatly and are major risk factors for symptoms of suicide, and it is critical that we study variations in them. Yet, there is a challenge in that persons with a higher stress level may be least likely to access research and clinical care.
There are many people that I would like to thank in this work. My apologies to the many that I have left out, and we thank the generous funding that we received to perform this work. I am going to go very quickly through references for this presentation, but I would be happy to provide them and answer any questions that you have at the end of this webinar. But first, I will hand it off to my colleague Dr. Clark.
CRYSTAL TENNILLE CLARK: Thank you Dr. Blumberg for that wonderful presentation, I have learned a lot. I will share my screen and begin. Alright, so you have heard a lot from Dr. Blumberg today about the gender differences in bipolar disorder. As she mentioned, I will take us to the next step. I will be thinking about bipolar disorder and treatment particularly through the most vulnerable time for women with bipolar disorder, that is the perinatal period. I will speak specifically to you all today about pharmacokinetics of the most commonly used agents to treat bipolar disorder and how these pharmacokinetics effect treatment and how we might think about further research to optimize management of these medications and maintain wellness.
I have no disclosures.
So I will walk us through intersectionality and bipolar disorder, as it relates to what we have been talking about today, but also, bipolar disorder and specifically the perinatal period. As well as pharmacokinetics and the changes we see in perinatal women with mood stabilizers, and atypical antipsychotics, as well as where we might take future research.
So intersectionality is defined as “Marginalized or disadvantaged social statuses interacting at the micro level of individuals’ lived experience to reflect interlocking systems of privilege and oppression at the macro social structural level, such as racism, sexism, and classism”.
So we know that clinical trials have historically not included women. In preparation for this talk, looking back, it is amazing how recent we have begun to include gender differences in scientific investigations. So we know that sexism has led to the idea that women should not be studied, they are too messy. Messy meaning, they have all these hormones, and things going on with their bodies that make research more complicated so we will just study men, and even male animals so that we can keep it simple then we generalize that to women. Thankfully, in the early 1990s the NIH decided to make a concerted effort to increase the number of women in the NIH funded studies. And then in 2014, the NIH instituted a new policy to study sex differences and experimental design in cell and animal research which was also a very important move. You can appreciate this was just seven years ago, in the 1990s, not even a full three decades ago.
But we know that gender differences are well-recognized today, thankfully, by that messiness, and now we are at a point where we need to better investigate the changes in hormones that lead to menstruation and perimenopause, and accompany postpartum and lactation. We must focus our scientific investigation there because that is a major difference between men and women. These reproductive transitions and life changes that impact treatment course, impact illness course, and needs to be better investigated for us to optimize wellness for women.
Finally, I want to give a nod to PRGLAC, the NIH led task force on research specific to pregnant women and lactating women, which was established in 2016. They established several recommendations through this task force that was submitted to the Health and Human Services. I will highlight three of the 15 here.
Increase need for the quantity, quality and timeliness of research on safety and efficacy of therapeutic products. We need to remove regulatory barriers that prohibit pregnant and nursing mothers from participating in research, and we need to leverage, establish and support new infrastructures and collaborations to perform research on pregnant and lactating women.
Alright, so moving into addressing the gender differences particularly as it relates to this vulnerable perinatal period in the relationship to pharmacokinetics and mood stabilizers and atypical during pregnancy.
We know that women with bipolar disorder are more vulnerable during the perinatal period. Just how vulnerable are they? Fifty percent are more likely to have a postpartum episode than women who have major depressive disorder. They are seven times more likely to be hospitalized for the first onset of a mood episode early postpartum. They have a hundred-fold increase in risk of postpartum psychosis compared to the general population. As you heard from Dr. Blumberg, women in general with bipolar disorder have an increased risk for suicide. That increase risk is further exacerbated in the perinatal period and it’s a leading cause of maternal death.
I won't talk much about all the treatments. You've heard some of the treatments for bipolar disorder in Dr. Blumberg’s presentation, but I will talk today about the mainstay of treatment. particularly pharmacotherapy for bipolar disorder. Of course, psychotherapy, I cannot say enough about the need for IPSRT as an adjunct, as well as chronotherapy and ECT.
When we think about pharmacotherapy, the choice to continue a medication in pregnancy is really about thinking about the risk of illness exposure that untreated illness exposure to the fetus versus medication exposure and the risk there. When I am thinking about a patient, which is common as a reproductive psychiatrist, I am commonly thinking about how many times has this woman been hospitalized or had suicide attempts when she has not been taking her medication? What about the severity of her episodes? How long does it take her to recover after having an episode? How quick does her symptoms worsened? What medications have worked for her in the past? As well as looking at the medication and thinking about, what types of adverse effects might the woman experience or might the fetus be exposed to? Such as valproic acid, which I do not use in pregnancy or in childbearing women because of the major congenital malformations that are associated with that drug.
So, just continuing, we are thinking again about the risk and benefits and that balance, discontinuation of mood stabilizers and pregnancy increases BD recurrence. We know that. In fact, it increases recurrence by 85 percent, which is pretty profound. These are modest studies. Rather, modest sample sizes. We just have a snapshot, but I can definitely say this is a consistent finding. We know this because prior to pregnancy and in nonpregnant women and men, discontinuation of medication for bipolar disorder is likely to have a recurrence of a current illness. So we see this in pregnancy is not surprising.
We also know that even for women who continue their treatment, there is some recurrence. Thirty-seven percent will have recurrence despite continuation of treatment. I think what is happening there is the fact that there is something about the management of that medication. Maybe we are under dosing. Maybe the pharmacokinetics changed too much, and we don't have effective concentrations and therefore, lack of drug therapy effectiveness.
But what about untreated bipolar disorder during pregnancy? There is risk there, too. Intrauterine growth retardation, low birth weight, preterm birth, preeclampsia, adverse neurodevelopmental outcomes, and then there is also the behavior associated with bipolar disorder. Impulsivity, for instance, that can lead to substance use or lack of prenatal care and, of course, the risk for suicide.
These are risks that are independent of pharmacotherapy exposure. We are having to think about the two. When I’m thinking about untreated bipolar disorder and that exposure to the woman and fetus, I returned to pharmacotherapy, which is the mainstay of treatment.
With that said, we will transition here and talk about some of the work that has been done to look at mood stabilizer management in pregnancy and postpartum. I will start with lithium because it is the gold standard and if any of my Hopkins colleagues, that is where I trained, are on the webinar today, I have to acknowledge them for their use and training of lithium because today my residents are like I’m the lithium queen. That is because I turn to this drug often because it is still the most effective tried and true for bipolar disorder.
Pharmacokinetics of lithium in pregnancy. We know that, in pregnancy, there are lots of physiological changes. Estradiol rises rapidly to significant numbers during pregnancy, at least to increase cardiac output, 50 percent in fact, increased cardiac output. That leads to increased glomerular infiltration rate. If you think about the pharmacology and the pharmacokinetics related to lithium, we know that it is exclusively renally eliminated. That is related to this glomerular infiltration increase. So that increase leads to increased renal elimination. What does that leave us with? That leaves us with decreased serum lithium concentration.
Why this is important is because when we think, again, of a nonpregnant woman, or the man who is taking lithium, we think of achieving a therapeutic range or a number within that therapeutic range that is personalized to the patient. The range being 0.6 to 1.0, for those who use a more conservative range, or 0.8 to 1.2, and we are usually expecting our patients to fit somewhere in that range.
If they have a therapeutic concentration where they are feeling their best at 0.8, and then they get pregnant and that level drops to 0.5, we would anticipate that they would have a recurrence of symptoms. Why? Because when they are not pregnant, we would anticipate if we dropped the dose or if they were doing something that decreased the concentration from the therapeutic baseline, studies have shown that that would cause a recurrence of illness. Again, we would not expect anything different in pregnancy.
Here I highlight some research that has been done by colleagues in Europe. Looking at the changes in concentration across pregnancy compared to baseline and I note this is retrospective data. As you heard from Dr. Blumberg, we really need more studies that are designed to look prospectively and longitudinally at research around these areas of intersectionality, one being gender differences, particularly in the concentration and changes in lithium.
When we lump the data, we see similar results as far as we can see that the changes from baseline across three trimesters of pregnancy are quite different. Starting off with the baseline of .53, leading up to a trimester change by third trimester, the concentration change to .35. Again, usually a 0.2 drop for lithium can put a patient at risk for occurrence, according to clinical research trials and past data on lithium.
Now we look again at another retrospective study from some of my colleagues in the Netherlands and here, what you see is as you track from baseline across pregnancy trimesters 1, 2 and 3, getting closer to delivery, you can see a trajectory of the concentrations and how they are changing so Baseline, they are elevated, they are declining across the first and second trimesters of pregnancy, and slightly starting to increase as we head into the postpartum where they increase and return to baseline. Again, excellent research by my colleagues, and also I will add that this is lumped data so what we don't get to see here is the personalized trajectory of lithium concentration changes and how, although there is a general trend, how it might affect that individual patient, how changes in concentrations affect the individual patient and how might the trajectory might differ with individuals.
Finally, I bring you to the only study of clearance back in 1973 by Schou et al, where he looked at clearance in healthy women, not with bipolar disorder, clearance of lithium in healthy women where they took lithium for a while, 600 milligram dose, and then he looked at the clearance across 4 to 12 weeks prior to delivery and after delivery. What he found was the clearance significantly went up prior to delivery and after delivery, it went down significantly.
I bring this back to intersectionality because it is so important to have longitudinal studies that are systematically designed to study and analyze how hypotheses that impact the specific population of investigation. When we are thinking about women, we want to have studies that specifically study women, not as a secondary analysis or retrospective chart review or database review, but that are really looking at these individual differences in a particular population. I would say that that is the same when we think about gender differences in pharmacokinetics across reproductive changes in life.
I offer here some pilot data, actually, where we did just that, looking at the longitudinal perspective differences across pregnancy and postpartum in women who are taking lithium. What you see here is, when you look at the third column, we see that elimination clearance compared in trimester one compared to trimester three, is about the same. But it is increased when you compare it to postpartum one and two time points - significantly increased. As a result, what you can see, which I will highlight here with this individual participants data, that you have a 12-hour serum concentration of a woman who has taken 1200 milligrams of lithium once a day. Her doctor, her is a little perinatal savvy, thankfully, split her dose in pregnancy and the reason why that matters is because in pregnancy, because of the physiological changes, having a more consistent concentration across 24 hours is very important. Splitting the dose and going back to older practices, today many people take it once daily, but going back to older practices of splitting the dose can be very helpful to a pregnant woman.
The 12-hour level at the first trimester of pregnancy, she has declined significantly. This is marked as trimester one, but she is about 14 weeks, almost 15 weeks at the time of this blood draw. You see she has dropped a complete almost 0.7 in concentration. I can share with you that at that point, she became very symptomatic and complained of worsening depressive symptoms such that her doctor increased the dose. Just increasing it by 300 and still keeping the dose split, she gets to 0.67, not quite back to her baseline, but improves in her symptoms. By postpartum she is back at 600 milligrams twice a day, so she does not get toxic, but it took her a while to get back to her baseline. Although, just at the 300-milligram dose increase was significantly helpful to remit her symptoms and Why this is important about knowing, about monitoring these concentrations is number one, for maintaining wellness and keeping symptoms in remission But also because so often, as I see in my practice, women will come in and, if they have a doctor who is willing to prescribe in pregnancy, too often that is not the case and is not considered necessarily standard of care, the doctor will not increase the medication that has already been working for them. They will add on medications, which increases unnecessary exposure so It is important for us to understand how we can optimize monitoring and dosing in pregnancy to prevent additional unnecessary exposures and also maintain wellness and not overshoot as far as dosing.
So, next I will move on to where my research began with Lamotrigine. I continue to think this is a fascinating drug in pregnancy, and particularly for bipolar disorder and we know that many may be aware that Lamotrigine has been studied well in neurology because it started out as an antiepileptic for epilepsy. It has been well studied in women with epilepsy who are pregnant because there was some understanding a while ago, that actually as the woman progressed in pregnancy, her seizures were returned. So there was a study to see what was happening. Was this because something was happening with the Lamotrigine? Did she need more medication? And What they found that the concentrations of Lamotrigine were decreasing. As a result, there were some association with an increase in seizure frequency.
So, seizures and bipolar disorder, to our knowledge, are not the same thing. So, I decided to take a look at what is happening in bipolar disorder under that same understanding. Taking a closer look at how does lamotrigine perform in patients with bipolar disorder who are pregnant. So, when we look at pregnancy, for the woman who has bipolar disorder, as with anybody, we know the estradiol increases, as UGT1A4, is regulated directly in relation to the estradiol increasing, as well as another minor enzyme, UGT2B7. That leads to increased lamotrigine metabolism, which decreases the major inactive metabolite and leads to increased lamotrigine elimination.
What you are looking at here is a preconception compared to late postpartum. Particularly third trimester, because that is where we expect the clearance changes, pharmacokinetic changes to be most robust. So when we look at third trimester, it is highlighted here, looking at the second column, you can see that the clearance is 8.4, compared to late postpartum which is 3.3 liters per hour. We are looking at late postpartum as a way to appreciate the baseline, but even if you go to back to preconception, the baseline is 3.8. Again, significantly different and significantly lower than what we see in third trimester. And also as you look at the third column, and you look at the half-life of Lamotrigine, it is 50% less than the third trimester compared to late postpartum, which is a pretty significant difference. Finally, if you look at the volume of distribution, which is this column here, you can appreciate that actually the volume of distribution isn’t changing that much.
It is not so much that the woman is diluting the medication because of her change in volume, weight gain, fluid status, it is actually directly tied to changes in clearance, half-life, and the resultant changes in the serum concentration.
Here we just highlight the participants in the study, who was taking 200 mg of Lamotrigine throughout her pregnancy, never changed the dose because her psychiatrist didn't think to do so, and to my knowledge, the patient was not interested doing so either, despite symptom recurrence as she got further along in pregnancy. You are seeing here how the concentration over a 24-hour period looked throughout the first three trimesters. Just to put this in a better context, these patients will come in for a study once each trimester, and they will stay overnight for a 24-hour period, to have a serial sampling done where we are checking the lamotrigine concentration up to 15 times during the 24-hour period and comparing that clearance and calculating that timepoint to the next trimester, and then two times postpartum.
So, you can see here that compared to pregnancy postpartum, her concentrations were much higher. I can tell you that as with most patients, in the study, they felt much better postpartum especially if the doses were never changed by their psychiatrist. This is just a representation of how correlated the Lamotrigine concentration or clearance, I should say, is with the serum estradiol concentration. So, it is almost a one-to-one ratio.
What we can see with the study, as clearance increases, symptoms worsen. As for some people this meant that they would have increases in doses, thankfully. As the PI watching this, it would always make me concerned, and the psychiatrist who was monitoring the patient, everyone would have to have a psychiatrist in the study, would get information from us, and let them know where the concentration was for that particular time point. They can make a decision as to whether or not they want to do something with that – add a medication, increase the medication or not, as well as based on how that patient was feeling.
Most psychiatrists that had patients in the study, would not change doses unless the patient was experiencing some symptoms and we noticed that most people at least have one dose change, and some had as much as four dose changes related to not only concentration dropping, but particularly symptom presentation and the psychiatrist noting that.
Most often, I would add that the symptom recurrence was depressive symptoms or anxiety symptoms, we did not see any mania. So patients would start to say that they are feeling more anxious, I feel more down for some reason, and often times it would not be a full-blown episode. It would just be symptom worsening that they might identify as "I am not feeling like myself and I don't know if it is pregnancy or not".
We know clearance is not the same for everyone. Seventy seven percent of women had an increase in clearance during pregnancy, 219 percent increase in fact, whereas 23 percent of women will only increase 21.3 percent. And there is some thought that that is related to genetic differences, because it has not been related to age, race, weight, but it has not been specifically studied as it relates to age, weight, or race.
In our study we found that UGT1A4 variants *3 or L48V, which is associated with increased emotional clearance, was present in five of our participants. UGT2B7*2a, the 161TT genotype which has a frequency of 48 percent of the population, was associated with decreased lamotrigine clearance and found that some participants had genotypes typical of this variance, where most were homozygous.
With you UGT1A4*2, which is associated with decreased LTG Cle, we found that all participants had that. We are still trying to work through that data to see what does that mean. In a small sample size, it is really difficult to determine anything with pharmacogenetics but this is just the beginning of highlighting and underscoring that there might be differences in biomarkers that we can address early on and anticipate what might happen to drug therapies for pregnant and postpartum women.
I would also highlight the fact that most of the participants had a variance of decreased lamotrigine clearance, as well as the variant that increased lamotrigine clearance, and it is very possible that they cancelled each other out, but we are still investigating that.
We will move on finally to atypical antipsychotics which are increasingly used in women of childbearing age, and must really again, be thoughtfully considered as to how best dose these medications across pregnancy as well as thinking about the complications they can add such as excessive weight gain, which can lead to gestational diabetes and other complications for the fetus.
But when we think about the risk/benefit analysis, we are always thinking about what has worked best for the patient, and what may work best moving forward through pregnancy and balancing those risks and benefits. So when we think about atypical antipsychotics, again, we are thinking about the ways in which these drugs are metabolized. Some of them are metabolized with UGT1A4 or CYP2D6, which has been well characterized in terms of its genetic variance for rapid metabolism or slower metabolism, so that it definitely must be considered for the pregnant patient, and also, CYP3A4.
We know physiological changes in pregnancy, this may lead to increased eliminated atypical antipsychotics therefore lowering the concentrations of these drug therapies and leading to recurrence of symptoms or associated with recurrence of symptoms. So here I present some pilot data that my team looked at, in terms of quetiapine, the area under the curve and oral clearance. What you see here is data looking through preconception onto postpartum. You will see this trajectory of declining area under the curve, directly associated with the increases in oral clearance of the quetiapine.
When we take a look at how the oral clearance increases across pregnancy, and returns to baseline postpartum, we can also appreciate that depression trajectory symptoms are also higher in pregnancy compared to postpartum as well as anxiety symptoms. We saw this also with lurasidone, and where you can also see in pregnancy that at least in the latter half of pregnancy, that doses of 80 mg or higher, the concentration was much lower despite getting up to doses of 120 mg. Concentrations are much lower compared to postpartum, where back to 80 mg dose, the concentration is much higher and in conjunction with that, seeing symptoms plummet, anxiety and depression, in particular, and they start to normalize postpartum.
So, in summary, physiological changes during pregnancy and postpartum result in pharmacokinetic changes that impact psychotropic concentration and effectiveness.
Symptoms recur with decreases in the psychotropic serum concentrations compared to a woman’s pre-pregnancy concentration.
Genetic differences may influence the magnitude of pharmacokinetic changes in pregnancy.
There is a need for frequent symptom monitoring and dose adjustments are needed to maintain wellness and prevent poor outcomes. This seems to be consistent and different magnitudes across drug therapies that we commonly used for bipolar disorder.
So, when we think about research, next steps, and what is critical to better understand, I will go back to what the PRGLAC task force has already recommended, as we think about different infrastructures to support this type of research. It is extremely hard to recruit patients with bipolar disorder. Those who are in this field and working in the specialty, already know that getting a very large sample of patients with bipolar disorder is hard. But when you add on having a woman who is pregnant and lactating, who chooses to take medication during pregnancy and postpartum, it makes the samples even smaller. So there is such a need for multi-centered trials to look at this more carefully.
This is particularly in the systemic perspective of pharmacokinetic studies that characterize the course of BD illness over time, to recurrence, the symptom presentation, the risk of postpartum episodes, and suicidality, if they do not have a maintained therapeutic concentration throughout pregnancy. Also, how often do we need to monitor these symptoms? How much do we need to increase dose? What is the magnitude? How might optimizing pharmacotherapy and pharmacokinetics in pregnancy help us to reduce risk and improve mother and baby outcomes.
Dr. Blumberg also highlighted so much awesome work as it relates to imaging too, and it would be great to coordinate and collaborate with her on how we could show the changes in pharmacokinetics and how that pairs with maybe changes in the brain as well.
All of this work must be done through a lens of intersectionality of course, and I highlight today how sexism has affected scientific investigation as it relates to women, but we have so much more to do as it relates to social determinants of health in other populations. Therefore, I highlight intersectionality as it relates to African Americans, Hispanics, and other minorities underrepresented minorities. I did not speak a lot about African Americans or other minorities, in this presentation, because the research is so sparse as it relates to our understanding of prevalence, and the best understanding that we have an ECA study in 1984, which suggested that it was the same prevalence as in whites. But studies since then, have been unable to speak to prevalence, particularly in African American and Hispanic populations. We know that too often, this has been the illness that has been misdiagnosed as schizophrenia in minority populations, especially African Americans.
When we think about perinatal mood disorders, which are thought to be more prevalent in black women in the risk of suicide in black women, we have not yet studied enough how again, how pharmacokinetics’ for instance, in looking at pharmacology, might help to reduce that risk for this patient population. We really need more studies that are directly designed to address these disparities and differences and trajectories in the course of illness.
With that said, acknowledgements to those who helped with the work and especially to the BIRCWH and my K23 that has helped me collect this data. Special thanks to Tamara Lewis Johnson and Dr. Rudorfer, in the NIMH, for having me here today. With that I will turn it over to questions.
TAMARA LEWIS JOHNSON: Thank you, Dr. Blumberg and Dr. Clark for outstanding presentations. Giving us a lot of food for thought and really pointing us in the right direction of research priorities as it relates to bipolar disorder across the lifespan. We have a good number of questions from the audience so I'm just going to pull directly from those. So let's see, this is the first question that has come in. Is there any data, this is for Dr. Blumberg, is there any data to show the impact of transgender hormonal therapy on bipolar disorder? If so, does the data reflect the same results such as biological males to transgender females having the same issues as biological females?
HILARY PATRICIA BLUMBERG: Thank you very much for the important question. I think the main point here is that this is an extremely important area for further research. I wish I had more to tell you and maybe Dr. Clark is aware of some more study and I will give you one example.
In the review that we did of the 113 studies for suicide thoughts and behaviors, in terms of neuroimaging to try to see the brain associations, of them, we only found one study that even mentioned having someone who was transgender. That was one person. So, the studies have not been having this focus. It is extremely important, it is important in the study of bipolar disorder and with high risk, very important in the study of suicide.
TAMARA LEWIS JOHNSON: Thank you so much for responding to that question. Dr. Clark, there are a number of questions here, I am just going to group them. People are really wanting to know more about how often to check lithium levels during pregnancy and postpartum. And is there an ideal monitoring schedule for lamotrigine or lithium levels in pregnancy, or are you really monitoring by symptom change during the pregnancy?
CRYSTAL TENNILLE CLARK: Thank you all so much for those questions because that leads me back to where we need to go for further research. I have definitely studied lamotrigine for a long time now and have plenty of data that has convinced me that because of the rapid change, as quick as 10 weeks in pregnancy that we need to check this drug monthly, for most women. Of course, as I mentioned, there are some women who may not metabolize it as quickly for reasons we don’t understand yet. But that is exactly where I would love to take the research next, with the support of NIH, and really map out, what is the best approach, for number one, checking concentrations and being proactive about those changes when we know how significantly this drug bottoms out. I can quickly have a patient be on 300 milligrams and have a concentration of .4, and by second trimester have less than 0.5 in her system. Which to me would suggest much cannot be at the receptors if we cannot quantify it in the blood.
So to be continued. Love to continue the research this to provide guidelines but what I do now is monitor it monthly and monitor the symptoms of the patient monthly as well. As far as lithium I think that one is a little more – I think based on what we know in nonpregnant persons, that we can extrapolate a lot here. We know that going down 0.2, 0.3 in the lithium concentration is just a matter of time, we don't know how much time for the individual other than what they tell us about their history. But we know that the risk of recurrence is there.
So I check that level once a trimester. It is not as robust of a change in metabolism as a drug like lamotrigine, so I don't feel the need to check it as frequently. But definitely once a trimester, and of course, letting them know what might happen so your patient can let you know if they have symptoms recur.
Of course, with both drugs if you do make changes, you need to be very careful to, if you do make increases in the drug, you need to be very careful to decrease doses postpartum because the patient can get toxic in the postpartum period. I hope I answered all of those.
TAMARA LEWIS JOHNSON: Yes, that was very good. Dr. Blumberg, I have a couple of questions for you. Your research findings showed gender related effects on child maltreatment on individuals living with bipolar disorder, how can these findings inform intervention development for youth living with bipolar disorder? And what are the implications on their mental health trajectory over time, with effective treatment and undertreated illness?
HILARY PATRICIA BLUMBERG: Thank you very much for those questions. Many things that we need to research and many things that we need to do. First, of course, it is critical to do everything that we can to prevent childhood maltreatment.
I also want to make a very important point, a couple of important points about some of the neuroimaging research. One is that the differences that we talked about are very subtle differences. I just want it to be clear that we look at a group of individuals and compare them to another group of individuals, and the differences that we see are statistical differences on average. But these are really subtle differences in the brain and we could not hold up two scans and say, this person has bipolar disorder or this person has been exposed to one of the stressors.
I wanted to make sure to mention that and also, a really important point is that, where we do see the brain changes, the brain is very plastic, it is very plastic throughout life, that means that it has the potential for recovery and growth of new connections. And especially in youth. There is every reason to believe that these changes can be reversed. So it is very important to give every kind of help to the youth. Coping skills, all sorts of ways we can to improve their mental health and of note, as just one example, with our BE-SMART intervention, we are seeing improvements in the brain and the way youths feel and behave, where they are feeling more hopeful where you can even see decrease in suicide propensity, these are early findings, but we are finding that by providing them healthy strategies that they can use themselves lifelong, such as getting regular sleep and physical activity, we can really be of help and we need to think of all of the strategies we can to help reduce their symptoms of depression, anxiety, and importantly, of suicide risk, which, as you know, is high in individuals who have suffered from child maltreatment.
Finally, while I presented the data on child maltreatment, I just wanted to make the point, as Dr. Clark did too, that there are many stressors that can have adverse effects over the lifespan, they change over different phases of the lifespan, and it is great that we have some data on child maltreatment, and there we need a lot more information. But other forms of stress, caregiving stress, and in many different kinds of stressors, can have an impact and they have received little study. Those are extremely important to be focused on in studies, and in particular, with intersectionality lenses.
TAMARA LEWIS JOHNSON: Thank you so much, Dr. Blumberg. Dr. Clark, I had a question, one of the viewers wanted to know let’s see about the effects of lithium on the thyroid.
CRYSTAL TENNILLE CLARK: Yes. The effects of lithium on the thyroid, we tend to see this much further down the line, after treatment has been initiated and been consistent for years. That is why we check once a year, the thyroid. TSH and T4 and T3, to see how the thyroid is functioning with the patient who is taking lithium. There has been more recent thought that keeping, making sure that you have concentrations between 0.6 and 0.8 in a patient who is stable, reduces the chances or the risk of the kidney function being affected. I mention kidney function because the kidney and thyroid are the two things that we are concerned about as we think about adverse effects of lithium treatment long-term.
I am not aware, I would have to look, if there has been data to look at what concentration is most likely to prevent long-term effects on the thyroid as well but definitely, this is something we are more concerned about as the patient continues their treatment and hypothyroidism in particular, is the adverse effect we are looking out for.
TAMARA LEWIS JOHNSON: Great. Dr. Blumberg, there is a question that just came in wanting to know, is the BE-SMART intervention currently available for patients?
HILARY PATRICIA BLUMBERG: Thank you for the question, I am pretty fond of the BE-SMART intervention, as you have gathered. We have active studies currently with BE-SMART. So if there is interest, please feel free to contact me. You can email me or if there are ways to contact the people that are involved in this program, we would be happy to speak with you.
TAMARA LEWIS JOHNSON: Great. I think I just wanted to close out with one question, often times people, this is a question, oftentimes people living with bipolar disorder may have a substance abuse addiction and may suffer from other things, like job stress or ability to keep a job. What would you say to individuals or families that are working with young people that may have those co-occurring conditions, Dr. Clark?
CRYSTAL TENNILLE CLARK: I definitely think there needs to be much more research into these different social stressors, as well as compliance and how we can increase compliance of treatment. Not that there are some treatment resistant cases out there, which is a whole other area that needs to be studied. But more often than not, it is the buy-in, and then having appropriate follow-up and compliance but also some finding of ways to manage those stressors in a way that is helpful for the illness. Dr. Blumberg mentioned IPSRT, interpersonal social rhythms therapy, and the structure of that therapy creates is extremely important, of course, some stressors disrupt that structure. We have to find ways to help patients manage that. There needs to be research there too.
TAMARA LEWIS JOHNSON: Great. We are right at 2:30. I want to thank both Drs. Clark and Blumberg for a very informative and insightful and research agenda setting, presentation. I thank all of the viewers who stayed with us through the entire webinar. If you have any questions, whatsoever, feel free to send an email to the email address on the screen here.
This webinar, as I said earlier, is being recorded and there will be a transcript available for all who have registered, and it will be on our NIMH ODWD website in the coming weeks. Thank you everyone who attended, thank you Dr. Clark, thank you everyone who supported the production of this webinar and I hope that many of you tune in for the upcoming webinars that our office is offering. Thank you again.
(Webinar adjourned at 2:30 p.m.)