Assessing Illness and Medication Treatment in the Perinatal Period

July 1, 2016

Simone Vigod, MD Assistant professor at the University of Toronto and Institute of Health Policy in Toronto, Canada. Dr. Vigod has disclosed that she has no relevant financial or other interests in any commercial companies pertaining to this educational activity.

TCPR: Dr. Vigod, we know that it’s best to avoid medications if we can in pregnant women. But that’s more easily said than done. What’s your approach?

Dr. Vigod: I start by assessing the severity of the symptoms and the impact on function. Women usually fall into two categories: those who are not on medication and have become symptomatic, and those who have been taking medication, who are perhaps in remission, and are unsure about going off medication during pregnancy.

TCPR: For the women who are symptomatic and not on medication, what are the options?

Dr. Vigod: You can leave them untreated; you could treat them with a nonpharmacological option and/or you could treat them with a pharmacological option. For women who have mild symptoms, the hope is that you can treat with psychosocial interventions that might include exercise, stress reduction, increased social support from their social network and/or peer support, or psychological therapies. With women with more moderate symptomatology, the question is, can this be adequately treated with psychological therapies? There are barriers to psychological therapies: time off from work, school, or current parenting responsibilities that can involve time, coverage, and monetary challenges. Another consideration is if the woman either doesn’t respond to psychotherapy or is taking too long to respond. Psychotherapy is usually a 12- or 16-week trial, and that potentially means a woman doesn’t get better for 16 weeks, during which time the fetus is exposed to untreated depression and anxiety.

TCPR: What are the effects of untreated depression and anxiety on the baby?

Dr. Vigod: Untreated anxiety or depression in pregnancy is not benign. It can affect the developing fetus directly. Some research suggests that prenatal stress can increase cortisol levels, which can affect the fetus in various ways (Gover V et al, Neurosci Biobehav Rev 2010 35(1):17–22). We also know that untreated anxiety/depression in pregnancy is associated with smaller babies, babies born more preterm, and infants that are more irritable. There’s also evidence to suggest that untreated anxiety/depression in pregnancy might affect the infant’s developmental outcomes—though it’s very hard to disentangle whether this is due to prenatal stress or anxiety in the mother after the child is born. Untreated anxiety/depression in pregnancy can also affect a fetus indirectly because such women may be more likely to smoke, use alcohol, sleep less well, eat less well, or potentially not have the same level of antenatal care. Untreated anxiety and depression in pregnancy is also the strongest risk factor for postpartum depression and anxiety, which we know clearly have an impact on maternal infant attachment, and can even lead to delays in language and motor development as well as child internalizing and externalizing disorders.

TCPR: And what about the impact of untreated illness on the mother?

Dr. Vigod: For the mother, the longer you leave the depressive or anxiety disorder untreated, the harder it is to treat, and the more likely it is to become chronic. And so the greatest risk factor for postpartum depression and anxiety is untreated antenatal depression and anxiety.

TCPR: So it sounds like there are many really good reasons to treat depression and anxiety in pregnancy. How do you present the risks to new mothers without needlessly alarming them?

Dr. Vigod: In my opinion, we should have a high threshold for medication use in pregnancy, but we should use them when we have to use them. With that said, when talking to a new or potential mother, you want to present an absolute number or absolute risk when possible—because relative risks can be deceiving. For example, if there is a 0.5% risk of an abnormality, I will say that there is a 5 in 1,000 risk of this thing happening—but will add that this also means there is a 995 out of 1,000 chance of it not occurring. If you can present visuals, that’s very helpful. For example, suppose you have an abnormality that occurs in 5 of 1,000 births at baseline, and occurs in 8 of 1,000 births with exposure to a medication. In relative terms, this is a 60% increased risk (3 extra events divided by 5 events at baseline), which can sound pretty terrifying. But if you have a picture of 1,000 faces and you color in 5 for baseline risk and an extra 3 for the increased risk with medication, you can see that’s actually really small. Some women may not want to tolerate even that increased risk, and that’s fine, but you want to make sure they have an accurate view of the amount of risk.

TCPR: Let’s talk about women who may already be taking medication. What do you do if a medication is so new that we don’t have any kind of study data to extrapolate from?

Dr. Vigod: First of all, we would try not to use something so new. But if a woman is coming to me on one of these newer medications, she’s tried three older medications, she has a history of suicidality and it took her five years to really get better, the risk of taking her off that medication and switching her to something else would be awfully high. So I would educate her about the fact that we don’t have data, but that nevertheless we have to weigh that against the risk of coming off a medication that has been so helpful.

TCPR: Can we rely on animal data?

Dr. Vigod: Not necessarily. Thalidomide was not teratogenic in rats or mice, and there are examples where drugs were teratogenic in animals but not in humans. Most human teratogens have been discovered in post-marketing drug surveillance—either a physician doing a case report or large epidemiological studies.

TCPR: Just to shift gears now, you recently published a study on the risks of antipsychotics in pregnancy. What did you find?

Dr. Vigod: So there have already been a few studies on antipsychotics in pregnancy, and the data thus far are pretty reassuring. There were some concerns about metabolic issues like gestational diabetes and hypertension. In our study, we used a special matching method that approximates the quality of data you might get in a randomized controlled trial (Vigod SN et al, BMJ 2015;350:h2298. doi:10.1136/bmj.h2298). We found almost no differences between groups in terms of outcomes for mom or for baby after the matching. However, women in both groups (the antipsychotic group and the matched group) had negative outcome rates that were much higher than one would expect in the general population. This means that antipsychotic users are likely to be a high-risk population around the time of pregnancy, but that the risk for adverse outcomes shouldn’t be blamed on the drug.

TCPR: Is your consideration the same when the indication is a mood disorder versus a psychotic disorder?

Dr. Vigod: I think that the answer is yes for severe disorders such as schizophrenia or bipolar disorder, because it seems fairly clear that women with these conditions will relapse when medication is discontinued. Antipsychotics are now being used more and more for major depressive disorder, usually as adjunctive therapy. In depression, I would always say that monotherapy is better. If we can avoid it, we probably should, since you never know how two drugs mix together. But I guess it all comes back to my thesis, which is we should be treating people with what they need.

TCPR: Can you tell us what’s coming down the pike in perinatal psychiatry?

Dr. Vigod: Sure. Our team created an online patient decision aid to help women navigate decisions around antidepressant use in pregnancy. We are now finishing a pilot randomized controlled trial where we compared use of this decision aid to a PDF resource sheet. Our hope would be that if we can show it reduces women’s decision-making difficulty, then this would be an online intervention that we can update centrally as often as we need with the appropriate evidence and can be potentially scaled back for use in clinical practice.

TCPR: Any other research we should be aware of?

Dr. Vigod: We are studying the feasibility of a brain stimulation treatment for depression in pregnant women—transcranial direct current stimulation (tDCS). Women receive 30 minutes of treatment, 5 days a week over 3 weeks (15 sessions). We are doing it at the hospital to allow for continuous fetal monitoring around the time of the stimulation. It’s going to take some time before we finish our pilot study and do a much larger study, but the hope would be to provide another safe and effective treatment option for depression in pregnancy.

TCPR: That is exciting. Thank you for your time, Dr. Vigod.

Developmental Effects of Prenatal Selective Serotonin Reuptake Inhibitor Exposure in Perspective: Are We Comparing Apples to Apples?

(Oberlander TF and Vigod SN, J Am Acad Child Adolesc Psychiatry 2016;55(5):351–352).

Dr. Vigod recently co-authored an editorial in response to a cohort study in Finland that found a 1.8-fold increased incidence of depression, but not autism or ADHD, by early adolescence among offspring exposed to SSRIs in the prenatal period. Here’s her summary of that editorial.

There were two main points that we wanted to get across. Point one is that when we’re looking at outcomes for SSRIs, we want to be comparing apples to apples to isolate the effect of the SSRI. The authors attempted to do this as much as possible, but there were unmeasurable factors that may have been more highly associated with SSRI use. For example, a big issue is that depression is a heritable disorder. Without knowing what the genetic basis is of a given woman who is also taking an SSRI, it is hard to know what her child’s risk of depression is going to be. If a woman is more likely to take an SSRI during pregnancy if she has a bona fide genetic depression, then wouldn’t her child also have an increased risk of having depression compared to a woman who has less severe depression? And then also, when you’re looking at adolescents, you have to consider all the stuff we know about the environment of postnatal depression and anxiety. Someone who is depressed enough in pregnancy to take a medication—and depression is a chronic and recurrent disorder—what does that mean about that child’s early childhood environment putting them at risk for depression?

Point two is around the fact that there are many different possible maternal and child development outcomes, and the idea that we might want to shift away from a research model where SSRIs are either “good” or “bad.” If you look at some of Dr. Oberlander’s work, you can see how SSRIs may be protective for some outcomes in some environmental circumstances but not others (Hanley GE and Oberlander TF, Birth Defects Res A Clin Mol Teratol 2012;94(8):651–659). So the story about SSRIs may be more complex than some have previously considered.