Clinical Reflections on the SSRI and Pregnancy FDA Panel

ParamePrizma/AdobeStock

BRAIN TRUST: CONVERSATIONS IN PSYCHOPHARMACOLOGY

-Series Editor Joseph F. Goldberg, MD

Joseph F. Goldberg, MD: Hello everyone. Welcome to our inaugural presentation today in thew new series, “Brain Trust Conversations in Psychopharmacology.” I am Joe Goldberg, a clinical professor of psychiatry at the Icahn School of Medicine at Mount Sinai. It is my pleasure to be your host in this new video series hosted by Psychiatric Times, where we will have a chance to talk with experts in clinical psychopharmacology on topics that are front of mind, controversial, unresolved, or at least should be in your own thinking as a clinical prescriber, psychiatrist, mental health practitioner. Essentially, what is really current and topical in psychopharmacology?

Our guest today is Marlene Freeman, MD. She is professor of psychiatry at Harvard Medical School. She is the associate director of the Center for Women's Mental Health at the Mass General Hospital and the Abra apprentice foundation chair in women's mental health at mg she is the medical director of the Clinical Trials Network and Institute at MGH. Dr Freeman completed medical school at our mutual alma mater, Northwestern Medical School, followed by a residency at Harvard, Longwood and research fellowship in biological psychiatry at the University of Cincinnati, College of Medicine. She previously directed women's mental health programs at University of Arizona, College of Medicine and UT Southwestern Medical Center at Dallas. Dr Freeman is editor in chief of the Journal of Clinical Psychiatry. She is on the board of directors of the American Society of Clinical Psychopharmacology, where she is also a fellow, and a fellow of the American College of Neuropsychopharmacology, and a member and volunteer for Postpartum Support International.

Marlene, I asked you for 3 reasons today on this inaugural program. First and foremost, as a longtime friend and colleague, I hope you will go easy on me if I ask or say anything that is not to your liking. Number 2, the topic of reproductive health and psychopharmacology is something that I think most people think they know more about than they do, and so there is an awful lot of potential misinformation out there, and to have an expert authority such as yourself give us some guidance is really invaluable. And last but not least, the timeliness factor of this meeting today. I am going to start us out with acknowledging that on July 21 this year, the FDA convened a panel on the topic of SSRIs in pregnancy, and there were a lot of issues that came out of that panel discussion about the potential risks vs safety of SSRIs in pregnancy. So Marlene, I would like to start off by asking you, what do you think that panel got right and didn't get right, and what was your reaction to it overall?

Marlene Freeman, MD: So first, thank you so much for having me and and also thank you for inviting me to speak on a topic that I feel is so important. I really hope that health care providers of multiple disciplines tune in, because there are definitely things that I want to make sure, after a panel like that, that people really know.

One issue that any prescriber should keep in mind is that for women of reproductive potential, the rate of unplanned pregnancies in this country is about 50%. Whenever we are writing prescriptions or deciding on treatment plans, we want to keep in mind that even if a woman is not planning on becoming pregnant at the time, she might become pregnant at some time. We always want to keep in mind the reproductive safety of the medications that we are using, and that includes what we select for girls and young women as well. So many of the disorders we treat are chronic or recurrent, and we want to make sure that there is a reasonable plan for medication use across pregnancy, because so many of our patients have moderate to severe illness and will need treatment across pregnancy and postpartum. That is one of the things that tends to get underscored: the risks of untreated safety psychiatric illnesses during pregnancy in the postpartum. What I explain to patients, and what I explain when there is a confusing message from somewhere about use of psychiatric medications during pregnancy, is I always start with the risks of untreated illness. We know with all the major psychiatric disorders that have been studied across pregnancy, that the disorder itself carries risk for the individual, the pregnancy outcomes postpartum, and for the baby, and really for child development. There is an increased risk of obstetrical and neonatal complications. There may be long term neurodevelopmental consequences, and we we know that there is potential risk of the untreated illness itself, which can create a more stressful environment for the fetus, and may also lead to poor self care and engagement in prenatal health with the individual.

But there are also so many other potential variables that are impacted, like use of substances: tobacco, alcohol, smoking. We are seeing so many women using cannabis, assuming it is safe during pregnancy.

Also, more women with psychiatric disorders are obese than the general population, which is a major risk factor for basically every obstetrical complication and also major malformations. Many of our patients do not have adequate social support, good nutrition, or get regular exercise. We have to keep in mind all those factors when we are talking about treatment vs nontreatment.

My close colleague, Lee Cohen, in assessing risk of a medication always says succinctly, “Compared with what?” It is compared with the untreated illness. Because if we look at women who are on the medication during pregnancy vs those not on a medication during pregnancy, those groups will differ by more than just exposure to the medication. They are probably groups that are quite different for a lot of reasons. The other thing is that women can have the same diagnosis but different severity, so we suspect that women who have more serious illness are likely to take a medication during pregnancy than not.

Goldberg: Just to interrupt on that point before I lose you on this: One of the comments that was made by one of the FDA expert panelists was something along the lines of, “depression is not necessarily a medical illness, and we should just let it pass.” What is wrong with that statement?

Freeman: I think most people realize that there are serious psychiatric disorders that affect individuals and their families, and we have a job as a field to take these illnesses seriously and provide psychoeducation everywhere we turn. To our colleagues, to the public. I think in recent years, there has actually been great campaigns to educate, especially about postpartum depression; less so about depression during the pregnancies. But I think we have actually come a long way. It is unusual to hear people so dismissive of psychiatric disorders and the suffering that they cause.

One of the reasons that drew me to perinatal psychiatry was the opportunity not only to treat the individual woman during such a vulnerable time in her life, but also to affect the children—the baby and other kids in the home, entire families. by treating 1 person who is central to the family, you really improve the mental health of the entire family. One of the things in psychiatry that is more studied than almost anything else are the risks of maternal depression on child development. That includes postpartum depression and maternal depression, and we know that almost every aspect of child development is negatively impacted by having a mother who has depression, and I would say we can extend that to other psychiatric disorders. They are just not as well studied as depression. I think we are so beyond the conversation of whether psychiatric disorders are real or not—we know that they are suffering. We know that they impact really every aspect of the prenatal health of a woman who has a psychiatric disorder and everything that is interrelated. Having an untreated psychiatric disorder increases stress. If a woman is not well during pregnancy in terms of a psychiatric disorder, then the natural course of that is to get much worse abruptly in the postpartum. For most illnesses, if someone is doing what she thinks is the best thing for her baby and not pursuing treatment or taking a medication, if she is really unwell at the time of delivery, the risk of her being much, much worse and experiencing severe postpartum psychiatric illness is very high. The last thing we want is for women to be taking new babies home from the hospital being unwell. As any of you who are parents can understand, the the needs of a new baby, in terms of the emotional demands and the physical demands, are so high that we want people to be at the top of their game. We do not want them suffering from a severe illness, and sometimes medications take a while to work, so they may spend weeks or months recovering. We want to make sure, by the end of a pregnancy, that they are rock solid, stable, and well.

Then after delivery, I want them to know that they will be at risk. I always take the opportunity during pregnancy to meet with a woman, and if she has a partner, bring the partner into the conversation, about what to watch for postpartum in terms of symptoms, how to differentiate between what is normal in the postpartum and what are symptoms that are emerging. It is a really vulnerable time and we need to make sure that first and center in the conversation is the treatment of the illness.

Goldberg: Let me make sure I understand some things: Pregnancy itself is a high risk time for depression in the lives of women?

Freeman: Not really, actually. For some women, it might be, but for most people, it is neutral. The risk of relapse or continuation of a depressive episode is probably about the same in pregnancy as not, and probably the same for bipolar disorder. The issue is that when women discontinue medication that they have been maintained on, there is a very high risk of recurrence. This is the context where we see probably the most individuals trying to come off their medicine, either planning for pregnancy or during a pregnancy, and very often, understandably, patients are terrified if they find out they are pregnant and they are on a psychiatric medication. What we often see is patients abruptly stop the medication, relapse to illness, and then come in for treatment really unwell. Oftentimes, when that happens, patients are really destabilized. It is not as simple as just putting back the medication and they are fine. Many patients end up on more medication than they started on. Oftentimes, people have to be hospitalized. If a woman's planning pregnancy, it may delay her plan to conceive, which may, if she's in her late 30s or early 40s, impact fertility. There are consequences to stopping medication, but the risk of recurrence is probably about the same if treatment is about the same.

Goldberg: The biology of depression has direct, known, negative effects on the fetal development and abstract obstetrical course, right? Just to differentiate what the panel did not so much differentiate about what is attributable to the illness of depression from possible impacts of medicines. That must be really hard to parse out, although we do know things about what untreated depression can do to a developing fetus and the course of pregnancy.

Freeman: There have been good studies that look at the risks of depression and anxiety across pregnancy without medication. And there also have been sophisticated studies that look at women with depression on medication and with depression who are not on medication and then normal controls, no medicine, no depression, and in order to make these comparisons so that we can speak more clearly as to risk of antidepressants alone. We have so many studies now on SSRIs and pregnancy, but we still need more high quality research because there are so many confounding variables. Some studies are weak in terms of not controlling for confounding variables, including psychiatric illness, and not having comparison groups.

Our job in perinatal psychiatry is to be able to integrate all those years of studies and to explain to patients what seems to be true risk and what is not. When I am talking to patients and families about this, I really want to target what I think their concerns are. Patients often come in with specific questions like, “Do SSRIs increase the risk of autism?” No, they do not. They actually absolutely do not. We have enough data now that that is clear. But in the past, there were studies that were not well controlled, that showed that individuals on an antidepressant had an association with an increased risk of having a child with autism. So decades later, we now have so much data on this. We have reviews of reviews on this topic right now. You would think that we could reassure people. But if someone does not live in this zone like I do… I have the luxury of being able to go over the literature on this. But if someone is trying to counsel a patient, or a patient finds the information on autism from a very old study out of context, I understand that it would be very alarming, and scary information really sticks with people. That is where our job is to really try and synthesize information for our viewers and listeners.

Goldberg: You make such an important point about this notion of confounding. I am not sure if everybody who is listening or reading understands what that means. When a study gets done, especially what we call an observational study, where you are taking a large group of people, and you are not necessarily controlling for the unique impact of intervention, investigators will talk about confounding effects that would bias the outcome. Could you just say a little more for those who may not be familiar with the notion of confounding?

Freeman: If the question is, does an SSRI in the first trimester lead to major malformations compared with not being exposed in the first trimester, what else do you have to consider? First of all, there is the confounding by illness. The reason that the individual is on the medication because it may cause detrimental effects, either directly or indirectly, by some of those factors we talked about earlier, like substance use, obesity, exercise, nutrition, planned pregnancy, prenatal vitamin use—all those things. Studies are really only as good as the ability to control for all those variables, because otherwise we may mistakenly blame something that is not to blame for some outcome.

For example, at MGH, we have our national pregnancy registry for psychiatric medications, and the main outcome is looking at first trimester exposures to medication and risk of major malformations. But we are collecting information on just about every obstetrical and neonatal variable in child development in order to really enrich that sample. But we collect all that data. We collect everything I was talking about in terms of all those different confounding variables, and along the way, any obstetrical outcomes. Other medications are very important to collect, as well as any substance use over the counter, medications, supplements. A lot of patients assume some things are safe when they may be in question or not safe. What a registry like that does, or a large database where some investigators may look, is to be able to control for all those different variables. You can control for individuals exposed to the medication in question vs those not exposed, and you can look at women and compare, for example, among those who have the same disorders.

Goldberg: If a patient came to me and said, “I want to get pregnant. I have been treated with an SSRI. It has been effective, but I heard this FDA panel say that they are dangerous.” One thing I would want to importantly say to the patient is something that the panel did not really talk about: How do you know cause and effect? Taking an SSRI side by side with having untreated depression, being a smoker, being obese, having poor prenatal care, having substance use disorders, having poor social supports. There must be many, many, many, many factors that could go into the outcome of a pregnancy, untreated depression being not the least of it. But the FDA did not really quite cast that context of how do you understand any risk, if any, from really any intervention, relative to those confounders?

Freeman: I think that there is some background information that we also need to emphasize to patients and health care providers, and that is that pregnancy is inherently risky. The rate of major malformations in this country is something like 3% to 4% in the background population—that is without exposure to psychiatric medications. But I really think that pregnant women cannot think about that every day and get up and function. That is terrifying! Sometimes when people think about risks, or they hear something about risks, in their mind they are comparing it to the risk of not taking something, being nothing. We know that even without medication exposure, that pregnancy complications and complications for the child, such as major malformations even, are common. What we are trying to do is take every variable for that individual patient, then make very patient centered decisions with information about that patient and her history and also her values.

We always want to look at the severity of the illness and how recurrent it has been. We tend to see people who have had very severe depression, very recurrent depression, and are at very high risk of relapse. But we also never prescribe or recommend medication in a vacuum. We always recommend some form of psychotherapy, support, good nutrition, exercise, a lot of nonpharmacologic treatments along with it. We cannot actually give people like a breakdown on exactly how less risky it is with exercise, or those kind of things. In our field, one of the most neglected topics is sort of the groundbreaking value of prenatal vitamins and folic acid, because those are very important to take before someone even conceives. Folic acid periconception, even before conception, has been found to reduce the risk of birth defects, especially central nervous system defects that occur in the very beginning of pregnancy and then through pregnancy. Folic acid is important in terms of neurodevelopment. One of the things for those of us who treat reproductive age women that may become pregnant, is to recommend a simple addition of a multivitamin, which would have a pretty decent amount of folic acid, and for any woman who is pregnant or planning pregnancy, higher doses of folic acid. We always want people to be living their most healthy life, and especially during pregnancy, where that really resonates with people.

So it is a complex picture. We are not treating rats. There was an overemphasis on animal data in that panel, and we are treating people who have lives and families and so many variables and environmental concerns, and we have to take that all together. There are always going to be things we do not know, and pregnancy is a time where there are a lot of outcomes we do not know. If a woman is not deriving benefit from the medication she is on, that really makes a case to find something else or take take her off it. We really want to find out what works for someone, ideally, before they consider getting pregnant.

Goldberg: I imagine it is important to not just think depression, depression, depression. A woman who has had 3 or more prior depressive episodes, perhaps with a history of a suicide attempt or hospitalization or comorbidities, is going to pose a higher risk pregnancy when compared to somebody well with less severe illness or less chronic or less highly recurrent illness. It is a little hard to say one size fits all, I would think.

Freeman: Right. There is absolutely a role for psychotherapy in terms of perhaps some patients needing less medication or being able to come off medication. There have been studies, for example, looking at very intensive evidence based psychotherapies, like mindfulness-based cognitive therapy, to see about preventing relapse of depression during pregnancy and the postpartum part of the problem too. We want to be able to offer everyone really scalable options for evidence-based psychotherapy, really the gold standard for any woman who is in that situation of recurrent severe depression would be psychotherapy plus medication.

Goldberg: Any importance about trimesters? So organogenesis is the first trimester, but then there is obstetrical complications later. Do you ever say, “let's just get to the first trimester,” or let's try to not stop the medicine of the third trimester. How do you think about the timing of trimesters and exposures?

Freeman: I am in a good position in that I round with probably the largest group of perinatal psychiatrists every week at our Center for Women's Mental Health. Collectively, over the years, we have changed our practice. So for example, with some medications, we might have tried to stop a medication for the first trimester and then restart it later. However, we have seen so many patients relapse so severely that we do not do that anymore. As long as a medication is reasonable during pregnancy, we would continue it throughout the first trimester, especially if someone had a severe illness.

But the one thing I want to say about the third trimester: The thing that is most misunderstood, is that it has been reported now for about 20 years that about 25% to 30% of babies may have some symptoms after delivery if the mom is on an antidepressant in the third trimester. In our field, we call this poor neonatal adaptation, and when it was reported initially, without any study—and I am talking zero human study, no clinical study whatsoever—the FDA put into all the labels that women should seriously consider stopping their antidepressants in the third trimester. As a field, collectively, perinatal psychiatrists said time out, wait, we are not doing that. Because women who are at the greatest risk of severe postpartum illness are the individuals who are on medication in the first place, and we talked about before how we do not want women to be ill when they are taking a new baby home from the hospital. We just talk about this with patients. We explain that about up to about a third of women may have babies with some symptoms. We would never know for sure if it is due to the antidepressant or not, but the typical symptoms are mild and short lived, and tend to include SSRI some jitteriness, fussiness, crying that goes away usually pretty quickly and do not require medical management, and sometimes they do not occur at all. But I talk about this with every single patient, and I want their partners to hear it also, because this is the topic that most women get the most misinformation about. Many health care providers will tell a patient to stop their antidepressant in third trimester, and I have had patients have the terrible experience of being in the hospital for labor and delivery, which is such a vulnerable time, and have been told in a judgmental way, “Don't you know you were supposed to stop taking that?” And then the woman’s soul is crushed. She is in a major life transition of becoming a mother, and was just told she has done something majorly wrong, and she is just in the hospital to deliver the baby. To me, that is heartbreaking. I want patients to be like the world's expert about this, because they are so likely to encounter misinformation.

Our best strategy is to provide the education. One of the missions of our group is to have our website, (www.womensmentalhealth.org) which provides new information as it comes out. We have an editor in chief to provide information for patients and families. Postpartum Support International also has great resources for patients and families. We realize not everyone is going to be able to see a perinatal psychiatrist, and so what we want to make sure is people have all the educational resources possible.

Goldberg: One of the comments that came out of this panel was the question of whether or not a boxed warning should be applied to SSRIs. What kinds of consequences do you think might occur if such a thing happens?

Freeman: I have to honestly say, I do not know. I do not know what this FDA is going to do. I do not know if there was an agenda for that meeting, because to me, it is puzzling why there was so much emphasis on animal data and nonclinical data. I am not an expert on what a black box warning might do to prescriptions. There are plenty of medications that have black box warnings that we use regularly. We always want to be respectful of the medication adverse effects, but I do not think that this class of medication requires more caution at this time. I would not suggest label changes.

For some background on the labels, just recently in about 2015, the labels were all changed in the pregnancy section, which is Section Eight of the package insert. The FDA used to assign a letter to the medication on pregnancy safety. So A, B, C, D, X, and at the time of letter assignment, there was usually zero human pregnancy data. It was usually based on the small amount of animal data. So after many decades, those letters went away because they really did not make sense. They did not help weigh the medications risk and benefit. Now, if you look in that section, there is a summary of what is known about human exposure to that medication, first and foremost. Also, if there is a human registry for that study that is FDA-approved in terms of its science. Also, if the information is available, what is the use of that medication? Are outcomes different than controls with the same disorder?

Package inserts actually have changed in recent years, so that that was actually a big development. I think health care providers are just wrapping their heads around that.

Goldberg: I would be worried that a label change would serve to be fear mongering and stigmatizing, and feed more misinformation and dissuade women and practitioners from having transparent dialogues about pros and cons and risks and benefits, particularly given the things that you are saying about untreated depression as neurotoxic, both to women and to a developing fetus. It does not really foster the dialogue that I think we are trying to foster.

Freeman: I would like to add one more thing: As a field, we do not lack concern about all medications. I do not want it to seem like we're laid back about everything. I do feel it is important to mention that there is 1 glaring exception in our field: valproic acid, or Depakote, which has the worst risk of teratogenicity or birth defects associated with it. Many women do not know they are pregnant right away, and with valproic acid, the risks for major malformations occur very early. Then, the information just gets keeps getting worse, because valproic acid is associated with poor neurodevelopmental outcomes with exposures throughout pregnancy, sort of like alcohol. The 1 medication we do not want women to be on who have reproductive potential, or girls younger than that, is valproic acid. Since we have so many other mood stabilizers now, I really think that it should not be introduced, because otherwise we have this bind.

We just had a patient presented in our rounds this week who did very well on valproic acid and now wants to become pregnant, and then the issue is that that the individual has to change medications, or even worst case scenario, is she is already pregnant. I do want to mention that, because I think it is so important in our field. There have been studies that have been done that show women are not counseled about the risks of valproic acid in pregnancy, they are not counseled about using contraception. And even if they are, they often are not using it regularly. Again, we have an unplanned pregnancy rate in this country of about 50%, so if there is 1 medication I wish there would have been a panel on, it would have been that.

Goldberg: That was my reaction too! I do not quite understand why the SSRIs are being single out, when there are other medicines, such as valproates or carbamazepine, that actually do have some potential teratogenic risks. It is not really fair, I think, to just sort of broadly speak about a drug class without getting a little more nuanced and granular about what is known where there is risk, where there is not risk, and also how to do the risk benefit analysis that we all must do.

Freeman: As long as you mention that, there are many medications that are prescribed for disorders other than nonpsychiatric disorders, or for other medical conditions, where there is next to nothing known about them during pregnancy. What I wish would have happened would have been a panel that was convened to look at what medicines do we really need to study and have really accurate risk benefit discussions around across all fields of medicine, not just targeting psychiatry and the most commonly used antidepressants, which seems to not be part of a strategic plan.

Goldberg: Well said. Well Marlene, this has been a wonderful opportunity, I think, certainly for me and I am sure for our listeners, to get some top line insights about what is really known in the treatment of depression and pregnancy, and how to think about these topics with clarity and scientific rigor. Any final thoughts you want to add before we sign off?

Freeman: No, I just want to thank you for including me!

Goldberg: Thank you all for joining us today, and I hope you found this helpful. We will look forward to seeing you again real soon!

Dr Goldberg is a clinical professor of psychiatry at The Icahn School of Medicine at Mount Sinai in New York, NY and the immediate-past president of the American Society of Clinical Psychopharmacology.

Dr Freeman is a professor of psychiatry at Harvard Medical School, the Abra Prentice Foundation Chair in Women’s Mental Health at Massachusetts General Hospital, the associate director of the Center for Women’s Mental Health at Massachusetts General Hospital, and the medical director of the MGH Clinical Trials Network and Institute.