TCPR: Dr. Burt, you’ve spent much of your career in psychiatry treating women and teaching about reproductive psychiatry. How do you do your evaluation of pregnant or postpartum women who have psychiatric disorders? 

Dr. Burt: Typically, when a patient calls to schedule an appointment, she’s been referred to us by an obstetrician, reproductive endocrinologist, internist, or psychiatrist. She may already be pregnant or hope to become pregnant or is post-partum and possibly breastfeeding. I spend time speaking with her on the phone to assess her current state of mental health and functioning, particularly her stability and safety. It’s now pretty widely recognized that pregnancy does not protect against mental illness, and the postpartum period is certainly a time of very high vulnerability for women, especially if they’ve had a history of psychiatric illness. 

TCPR: What if a patient does not seem stable or safe?

Dr. Burt: If she appears to be unstable or unable to function, I will have her seek emergency psychiatric intervention, usually by going to an emergency room, and will not have her wait for an outpatient appointment. 

TCPR: And if she seems stable, how do you proceed?

Dr. Burt: For women who are stable enough to wait for an appointment, I send them a detailed questionnaire so that I have a head start before the initial consultation, which is generally a 90-minute appointment. Some patients find it useful to consolidate their history in a way they’ve actually never done before. I have them describe everything from current symptoms, emotional state, and behavior functioning, to details about past psychiatric history; current health providers; current and past medications, both psychiatric and non-psychiatric, including over-the-counter and alternative agents; substance and alcohol abuse or use now and in the past; medical history; obstetrical history; and of course social and development history. Also, there is the well-validated Edinburgh Postnatal Depression Scale, a 10-item screening instrument widely used to screen for depression both in pregnancy and postpartum which patients can take themselves (Cox JL, Br J Psychiatry 1987;150:782–786). 

TCPR: So if during your screening process you determine that a patient is high risk, what’s your best practice in terms of working with the referring doctor?

Dr. Burt: One of the things I do is provide patients and their health care providers (with the patient’s consent) a written copy of the final consultative report, which summarizes the patient’s clinical status as well as my recommendations. I also include relevant citations from the current literature that support my treatment recommendations. Referring clinicians are often worried about incurring liability for making such recommendations, so I feel that this is the best approach to address any of their medical/legal liability concerns. An important part of that report is a written case formulation. It puts together my understanding of the patient psychodynamically in terms of her psychiatric diagnosis so that my treatment recommendations reflect my best understanding of what will work for that particular patient. 

TCPR: And then do you take over the treatment?

Dr. Burt: It depends on the situation. Some referring clinicians prefer that I follow and treat their patients through pregnancy and postpartum, and then refer them back after six to 12 months postpartum once everything is stable and the patient and her baby are doing well. Other times, we just provide that one-time consultation. Many clinicians feel very comfortable following our recommendations, and we are happy to provide clarification or additional consultations as needed. 

TCPR: What’s your thought process like as you evaluate whether women wanting to become pregnant should be on antidepressants? 

Dr. Burt: When I see patients who want to know the risks of antidepressants in pregnancy, I always start by addressing possible risks and benefits of not using antidepressants in pregnancy. The main risk, of course, is relapse. For example, let’s say I’m seeing a patient who is in her 20s or her early 30s. She’s been on antidepressants for a long time, and she’s stable. It’s reasonable for her to defer pregnancy for a while so that we can taper her off her medication, and we’ll follow her, say, for six months or so. If she continues to be fine off the antidepressant, then she can become pregnant without exposing the fetus to medication, which is the best approach, other things being equal. But if she relapses, I’ll generally recommend restarting the medication to keep her stable. 

TCPR: What about the woman who is older and feels like the “clock is ticking” on her fertility? Do you make different recommendations?

Dr. Burt: Absolutely. If a patient is in her late 30s or maybe her early 40s, we don’t have as much time to wait to see if she’s stable off medication. Continuing the antidepressant she’s been on may be the wiser plan. No two cases are the same, and the best treatment approach is always individualized for a patient’s specific circumstance. 

TCPR: That brings us to the crux of the matter, which is the risks of being on antidepressants during pregnancy. There are plenty of studies, and it can get pretty confusing. Can you summarize where we are with antidepressants right now and what are the things we need to think about?

Dr. Burt: Antidepressants in general may be associated with an increased risk for somewhat shorter gestation—about 3 days or so, certainly less than a week, which is generally not clinically significant. With regard to birth weight, antidepressants may be associated with a reduction in birth weight of about 75 grams, which is about 2–1/2 ounces or so, which, again, is generally not of clinical significance.

TCPR: What about SSRIs specifically?

Dr. Burt: As a group, serotonergic agents—whether we’re talking about SSRIs or SNRIs—are not considered major teratogens, certainly nowhere close to drugs like Accutane or thalidomide. If you do a literature search, individual articles suggest specific malformations with one SSRI or another. The problem is that most of these studies have flaws because ethically you can’t do the gold standard studies that would give us definitive answers. Specifically, we can’t do randomized controlled trials in which we assign pregnant women randomly to antidepressants vs. placebo. The potential risks to the fetus are too great. 

TCPR: So what’s the bottom line on SSRIs and SNRIs?

Dr. Burt: What should be noted, of course, is that any pregnancy carries somewhere between a 3% and 5% risk for major congenital anomalies, and that maternal stress and depression itself during pregnancy have also been associated with adverse reproductive outcomes. The bottom line is that these medications do not appear to increase the risk for major congenital anomalies, with the exception of paroxetine, which does have a number of studies suggesting an increased risk for cardiac malformations, particularly right ventricular defects. But even here, it should be noted that there is some controversy even about whether paroxetine is a problem; because there are enough articles that concern us, we tend to not use it as a first-line agent. The other issue readers may be aware of is persistent pulmonary hypertension (PPHN). This is a serious condition that can be lethal, and some years ago there were some data associating it with SSRI exposure. But a number of more recent studies have called this into question, suggesting that other issues like cesarean sections may be responsible rather than SSRI exposure. In 2011, the FDA issued a public safety announcement advising health care professionals not to alter their current clinical practice of treating depression during pregnancy. They stated, and I’ll quote this, “It is premature to reach any conclusion about a possible link between SSRI use in pregnancy and PPHN.” 

TCPR: Aside from major abnormalities, are there any other concerns about antenatal antidepressant exposure?

Dr. Burt: One issue that does appear to be associated with antenatal antidepressant exposure is that of neonatal adaptive difficulties occurring just after delivery. This has various names, but one that has been gaining traction is PNAS (perinatal neonatal adaptation syndrome). It occurs in about 30% of deliveries, so I’ve encountered this with my patients. The symptoms include jitteriness, decreased muscle tone, grunting (which is suggestive of respiratory distress), and difficulty sucking. These adverse effects tend to be quite mild and generally very transient. In most cases, infants recover quite quickly within hours—and certainly within days—and they do well subsequently. In general, I always suggest that new mothers on antidepressants in pregnancy and their neonates remain in the hospital for 48 hours post-delivery to ensure that the babies are healthy and progressing well.

TCPR: There continues to be some confusion around whether antidepressants should be tapered prior to delivery to minimize these issues. 

Dr. Burt: In my opinion, it makes no sense to taper off a required serotonergic antidepressant prior to delivery. This is when a mother is entering her most vulnerable time psychiatrically—the postpartum period—when she not only has the responsibility for looking after herself, but also a baby that is dependent on her. We always use what I call the minimal effective dose, and those two words “minimal” and “effective” are equal. You want to minimize the dose, yes; but what’s the point of using a dose that’s so low that it’s not effective? Now you have exposure without achieving the desired effect.

TCPR: What are the data about what happens when a woman is on an antidepressant, becomes pregnant, and goes off it? 

Dr. Burt: You may be aware of the three-site naturalistic prospective study from 2006 in which we found that recurrence risk was significantly higher with a 68% relapse rate for women who discontinued antidepressant around the time of conception compared to 26% in those who maintained antidepressant treatment. And recurrences emerged rapidly—50% in the first trimester and 90% by the end of the second trimester. So that’s important information (Cohen LS et al, JAMA 2006;295(5):499–507). The conclusion is that women are at risk of relapse in many cases, and that pregnancy per se is not going to prevent relapse. And especially for the patient who has shown us that, in the past, every time she goes off her medication she relapses, you may want to continue the medication.

TCPR: What are your thoughts on treating anxiety or insomnia during pregnancy in women who’ve never had a psychiatric history?

Dr. Burt: First-line treatments include interventions like cognitive behavioral therapy (CBT), reducing psychosocial stressors, and eliminating stimulants such as caffeine and nicotine. You don’t want to smoke anyway during pregnancy for obvious reasons. Sometimes couples therapy is indicated. For insomnia specifically, you want to ensure sleep hygiene as a first-line treatment. You want to suggest decreasing fluids prior to bedtime, massage, and stretching. CBT for insomnia has been manualized and endorsed by NIH and works well for motivated patients. 

TCPR: So medications for anxiety and insomnia are further down the list of options?

Dr. Burt: Yes. Serotonergic antidepressants can sometimes be reasonable options for symptoms that are so severe that they suggest generalized anxiety disorder or panic disorder which might be causing the insomnia. And sometimes even occasional, intermittent, low doses of benzodiazepines, especially after the first trimester, do not appear to be harmful to the fetus. But the use of benzodiazepines is still somewhat controversial, with older research suggesting a risk of oral cleft, although more recent data do not appear to support that conclusion. But with regard to other pharmacologic approaches, although antihistamines (like most other sleep aid medications) can have a rebound effect, sometimes obstetricians prescribe Unisom with doxylamine. In fact, something called Diclegis, which is a delayed-release tablet containing doxylamine and pyridoxine, has been FDA approved to treat morning sickness. Obstetricians also frequently prescribe zolpidem for intractable insomnia in pregnancy. Although human data are limited for that, animal studies as well as reports from the FDA adverse event risk monitoring system have found no increased risk for congenital malformations with this agent. 

TCPR: In terms of adjusting medications during pregnancy, are there effects that the pregnancy itself may have on medications?

Dr. Burt: There are significant changes in both estrogen, progesterone, and other hormones, all of which are meant to sustain a pregnancy and also ultimately prepare for the delivery and eventually for breastfeeding. For example, estrogen levels rise quickly and steadily throughout pregnancy, and estrogen enhances the clearance of lamotrigine by inducing p450 liver enzymes involved in this metabolism. So as a result, maternal serum lamotrigine concentration frequently declines as the pregnancy progresses. In fact, over the course of pregnancy, lamotrigine levels may dip by as much as 50%, and therefore some women may experience clinical worsening of their bipolar symptoms. Lamotrigine serum concentrations tend to return to pre-pregnancy values within 3–4 weeks postpartum. If this does occur, serum levels may cause postpartum toxicity, and that can be manifested by symptoms including ataxia, nausea, dizziness, blurred vision as early as 3 days after delivery if the dose isn’t decreased. In other words, pregnant women taking lamotrigine for bipolar disorder may relapse because of declining serum levels and may become toxic after delivery.

TCPR: Speaking of bipolar disorder, what is the latest on the use of lithium in pregnancy? 

Dr. Burt: Lithium is still the gold standard for bipolar disorder in pregnancy, in my opinion. The absolute risk for cardiac defects in exposed fetuses is far less than the risk of bipolar decompensation with inadequate treatment. Having said this, there are a number of caveats to think about when using lithium in pregnancy. Obviously, fetal development really has to be monitored carefully and in general is a part of routine obstetrical management—things like the nuchal translucency test measured at 12 weeks’ gestation, and a high-level structural ultrasound at weeks 18–20, are very good ways to assess and screen for normal fetal development. A fetal echocardiogram is also sometimes done when indicated. 

TCPR: What should we know about how pregnancy might affect lithium levels?

Dr. Burt: Over the course of pregnancy, extracellular fluid levels rise and renal clearance of lithium increases, so it’s important to ensure that lithium levels remain therapeutic—about 0.8 to 1.2 mEq/L generally. Levels should be checked regularly once a month during the first half of pregnancy and more often during the latter half of pregnancy. Situations that tend to increase lithium levels should be avoided, and these would include dehydration, the use of NSAIDs or diuretics, calcium channel blockers, and ACE inhibitors. At times, an obstetrician may encourage a sodium-restricted diet to manage preeclampsia and which can inadvertently increase a patient’s lithium levels. 

TCPR: And how might lithium levels be affected during delivery and shortly thereafter?

Dr. Burt: Maternal lithium toxicity is possible in cases of acute loss of fluids at delivery, or hyperemesis gravidarum or preeclampsia. Because of these potential issues, some have suggested that lithium should be withheld during the first one or two days before a planned delivery or at the start of labor, but in our clinic we tend not to do this and rely on careful surveillance throughout labor and delivery to guard against maternal lithium toxicity. Preconception doses of lithium should be reinstated once the mother is medically stabilized following delivery. Because bipolar women are at high risk for serious bipolar illness during the postpartum, clinical status should be carefully monitored, and lithium dosing adjusted accordingly.

TCPR: Thank you for your time, Dr. Burt.