is professor of Psychiatry and Vice Chair of Research in the Department of Psychiatry and Neurobehavioral Sciences at the University of Virginia, Charlottesville, VA.

Jennifer L. Payne, MD

My daughter has schizoaffective disorder. She fell ill at age 15 and had to be hospitalized twice. It was a terrible time for all of us, but today, I am happy to say, she is doing well. She graduated from a highly competitive college, got married, and writes amazing poetry and fiction. She still has delusions, but quietly so, and has developed a clear understanding of her illness and the necessity of mood stabilizers, though she still refuses antipsychotics. Why am I revealing the facts of my daughter’s illness in a public forum? In a word, stigma.

As a psychiatrist, I help patients and their families deal with stigma all the time. Having faced stigma in my own life, I know how upsetting it can be. When my daughter was hospitalized (at my institution), several well-meaning individuals told me how brave I was to allow that to happen. Whenever I heard that, I appreciated the sentiment, but also thought that nobody would have called me brave if my daughter had a brain tumor. Apparently it takes courage to reveal to your colleagues, even to your psychiatric colleagues, that your child is mentally ill—as if a serious mental illness in a child or relative is something to be hidden, not discussed. Through the worst of her illness, I lost a few friends who were uncomfortable with the fact that I had a mentally ill child. Some family members hid what was happening from others, I think out of a sense of shame and embarrassment.

Why is mental illness still considered shameful? At least part of the stigma arises because we do not entirely understand what causes mental illness, and because those with mental illness sometimes behave in baffling or even dangerous ways. On the other hand, many members of the public do not understand the gravity of mental illness. They believe that patients with depression, for example, just need to pull their socks up and 

 Finally, part of the stigma is a society level problem that humans tend to marginalize anyone who looks, acts, or thinks differently.

Widespread stigma and misunderstandings can have terrible consequences for patients. As a psychiatrist who specializes in treating psychiatric illness during and after pregnancy, I see them every day in my professional life. Every week I see patients whose doctors advised them to come off their psychiatric medications for pregnancy, not taking into account the severity of their illness, or the impact of untreated psychiatric illness on the developing child. I have patients and families who ask whether they should have children in the first place, as if someone with a mental illness cannot also be a nurturing and loving parent. It took until 2016 for the US Preventative Taskforce to recommend 

 for pregnant and postpartum women, even though peripartum depression is the most common complication of childbirth.

It is time to stop whispering and to start speaking out. The fact is that most, if not all, of us have family members who have struggled with some form of mental illness, and these numbers are only increasing during the pandemic. Across the world, 792 million people have mental health problems, and 

 individuals will experience a mental health or a substance abuse issue at some point in their lifetime.

 How can we continue to stigmatize something that is so common normal?

There are signs of change. A number of organizations are leading conversations about mental illness, educating the public, and speaking out about stigma. These include but are not limited to: 

. These organizations argue that the more we talk about mental illness as a common condition needing treatment, the more other individuals and families with mental illness will speak out too, and stigma will be reduced.

I believe this is the right approach. It is going to take regular doctors and patients speaking out routinely and treating mental illness like any other disease or condition. It is something that needs to be identified and treated, not something to be ashamed of. Discussing mental illness should be commonplace, like talking about diet and exercise and one’s cholesterol level. Only then can we dispel the stigma, support those who need help, and get them the care and treatments they deserve.

I have been open and frank with colleagues, friends, family, and strangers at times. I ask you to join me. Let’s tell the public that mental illness is common, treatable, and we all have been touched by it in one way or another. For the sake of our patients and ourselves, we need to do more to end the stigma of mental illness. Isn’t it past time?

is professor of Psychiatry and Vice Chair of Research in the Department of Psychiatry and Neurobehavioral Sciences at the University of Virginia, Charlottesville, VA. Dr Payne and her daughter have published a 

 about their personal experiences of having or treating mental illness.

1. Do you know the facts? Breaking down the myths about depression. Back to Campus: A Program of Mental Health America. Accessed September 14, 2021. 

https://mhanational.org/sites/default/files/Breaking_Down_the_Myths_About_Depression.pdf

2. Beharry J. 5 myths that prevent men from fighting depression. National Alliance on Mental Illness. March 23, 2018. Accessed September 14, 2021. 

3. Toohey J. Depression during pregnancy and postpartum. 

4. Dattani S, Ritchie H, Roser M. Mental health. Our World in Data. April 2018. Updated August 2021. Accessed September 14, 2021. 

Articles

Mental illnesses are too often shrouded in secrecy and shame. What would happen if we spoke out about the cases we knew best—the ones in our own families?  

Speaking Out on Schizoaffective Disorder

The prevention of maternal psychiatric illness, particularly postpartum depression (PPD), has the potential to usher in far-reaching individual and societal consequences. A 

 suggests that maternal mental health significantly affects children’s physical, intellectual, and emotional development. PPD has been shown to result in lower IQ, slower language development, increased rates of attention-deficit/hyperactivity disorder, as well as an increased risk of behavioral issues and psychiatric illness in the exposed children.

 found a strong and graded relationship between exposure to abuse or household dysfunction during childhood that in turn was linked to multiple risk factors for the leading causes of death in adults. The ACE study screened for psychiatric disorders in the household, but many other ACEs can be linked to adverse maternal mental health, including substance abuse, divorce, and increased rates of abuse in the home.

In sum, targeting maternal mental health could improve both mental and physical outcomes for everyone in the home, and potentially improve outcomes for whole communities. While improving maternal mental health is bigger than preventing PPD, if we start with preventing PPD we can at least improve cognitive and mental health outcomes for both mothers and their children.

So how do we prevent PPD? One possibility is identifying markers or factors that are strongly predictive of PPD on an individual level, preferably prior to the onset of symptoms. These are collectively known as 

: measurable substances whose presence is indicative or predictive of an illness’ onset and outcome.

PPD may be one of the easier psychiatric diseases to study, because you can identify the at-risk population and predict when the disease will occur. If you take 100 pregnant women, about 10 to 15 of them will become depressed in the first few weeks following childbirth. If those pregnant women have a history of a mood disorder and are not taking psychiatric medications the rate is 

 By following women through pregnancy while collecting potential biomarkers and then identifying who becomes ill postpartum, one can identify biomarkers that are predictive of later illness.

There are numerous factors associated with the development of PPD, at least on the group level. For example, having a psychiatric history prior to pregnancy, a family history of psychiatric illness including PPD, and being depressed in pregnancy are all associated with the 

 These increase the odds of developing PPD, but the exact risk for a specific individual is difficult to quantify. The same can be said for various environmental risk factors, such as socioeconomic status, adverse life events, and obstetrical outcomes such as preterm birth.

 As mental health providers, we can screen for these factors in our patients and know that an individual is at an increased risk given her individual or family history, but we cannot predict whether she will develop PPD. More personalized prediction, however, may be possible.

Perhaps our most promising biomarker potential for PPD is epigenetics. Marrying environmental influences and gene function, 

 examines changes to DNA that do not alter the DNA sequence itself, but they do turn the function of the gene on or off. Much of the literature in this area has focused on DNA methylation, which can be altered by such environmental factors as stress, medication use, and hormones. Epigenetic alterations have the potential to explain some psychiatric illnesses, which could be the result of an underlying biological vulnerability that is triggered by environmental stressors. In the case of PPD, one possibility is that hormonal changes during pregnancy lead to epigenetic changes of particular genes that serve either as a marker for or as an underlying cause of PPD.

My colleagues and I have published a series of studies identifying 2 epigenetic biomarkers of PPD. In our first publication, we initially identified genetic loci that were responsive to high dose estrogen in the mouse hippocampus.

 We cross-referenced these loci to DNA methylation differences in samples of blood from pregnant women with mood disorders, who were followed prospectively through pregnancy and postpartum.

The initial sample included pregnant women who were clinically well during pregnancy and either did or did not develop PPD. Two loci were identified at the HP1BP3 and TTC9B genes which, based on methylation differences, were able to correctly identify whether a woman did or did not develop PPD with an 

 We then performed a replication analysis in women who were depressed during pregnancy and either continued to be depressed or recovered postpartum. Our biomarkers were again able to segregate PPD status with 88% accuracy.

 We then went on to replicate these results in an independent sample of 51 pregnant women with preexisting mood disorders, as well as in a sample of 240 pregnant women without a previous psychiatric diagnosis, with an accuracy of at least 80%.

 Most recently, our group again replicated the PPD prediction, but also found that we were also able to predict third trimester depression using blood samples from the first trimester.

The exact functions of the TTC9B and HP1BP3 genes are currently unknown, but bioinformatics analysis suggests that both may be involved in mediating synaptic plasticity in the brain, and both have ties to estrogen signaling.

 Interestingly, mice without functioning HP1BP3 genes exhibit impaired maternal care, leading to a dramatic reduction in pup survival, indicating a role for the gene in maternal behavior.

 Further work on the functions of these 2 genes and their role in PPD may ultimately help elucidate the biological underpinnings of PPD.

What are the next steps? We are validating the biomarkers in a larger sample and examining what factors might be involved when there is a mismatch between what biomarkers predict and what actually happens from a psychiatric perspective postpartum.

The next important scientific step is to attempt the holy grail of psychiatry: a prevention study. Can PPD be prevented when a woman has the biomarkers in pregnancy or does having the biomarkers make the development of PPD inevitable? If PPD can be prevented, are there some interventions that are more effective than others?

is currently Associate Professor of Psychiatry and Director of the Johns Hopkins Women’s Mood Disorders Center at the Johns Hopkins School of Medicine, Baltimore, MD. Beginning September 1, She will become Professor of Psychiatry and Vice Chair of Research at the University of Virginia, Charlottesville, VA.

1. Grace SL, Evindar A, Stewart DE. The effect of postpartum depression on child cognitive development and behavior: a review and critical analysis of the literature. 

2. Felitti VJ, Anda RF, Nordenberg D, et al. Relationship of childhood abuse and household dysfunction to many of the leading causes of death in adults. The Adverse Childhood Experiences (ACE) Study. 

3. Payne JL, Roy PS, Murphy-Eberenz K, et al. Reproductive cycle-associated mood symptoms in women with major depression and bipolar disorder. 

4. Viguera AC, Tondo L, Koukopoulos AE, et al. Episodes of mood disorders in 2,252 pregnancies and postpartum periods. 

5. Kimmel M, Hess E, Roy PS, et al. Family history, not lack of medication use, is associated with the development of postpartum depression in a high-risk sample. 

6. Guintivano J, Manuck T, Meltzer-Brody S. Predictors of postpartum depression: a comprehensive review of the last decade of evidence. 

7. Guintivano J, Arad M, Gould TD, et al. Antenatal prediction of postpartum depression with blood DNA methylation biomarkers. 

8. Osborne L, Clive M, Kimmel M, et al. Replication of epigenetic postpartum depression biomarkers and variation with hormone levels. 

9. Payne JL, Osborne LM, Cox O, et al. DNA methylation biomarkers prospectively predict both antenatal and postpartum depression. 

10. Garfinkel BP, Arad S, Neuner SM, et al. HP1BP3 expression determines maternal behavior and offspring survival. 

Could the emerging field of epigenetics hold the key to catching postpartum depression before it happens?

The Search for Postpartum Depression Biomarkers

Annual Psychiatric Times® World CME Conference™

is Associate Professor of Psychiatry, Associate Professor of Gynecology and Obstetrics, Director, Johns Hopkins Women’s Mood Disorders Center, Johns Hopkins School of Medicine, Baltimore, MD.

Videos

Dr Payne previews her talk at the upcoming Annual Psychiatric Times® World CME Conference™, reminding us why it is important to take care of mothers postpartum.  

Why It Is Important to Treat Postpartum Depression 

Slideshows

Although they may have been the result of good intentions, these mistakes could adversely impact mother and baby.

Top 12 Mistakes in Perinatal Psychiatry

A 37-year-old woman with a history of recurrent major depression is currently stable on vortioxetine, a newer antidepressant. “Lauren” asks for advice regarding psychiatric medication use during pregnancy. She has a history of being nonresponse to classic selective serotonin reuptake inhibitor medications and two psychiatric hospitalizations, one of which was after a serious suicide attempt. How should her psychiatrist advise her?

Managing psychiatric disorders in the setting of pregnancy is confusing and complex; complicating matters is a large amount of conflicting literature that is difficult to absorb, interpret, and use for best practices. Those of us who have specialized in perinatal psychiatry have spent a large amount of time thinking about the best approaches to designing treatment plans for pregnant patients. This article discusses a few common errors in the management of the perinatal psychiatric population (

While medication discontinuation during pregnancy is appropriate for some patients, it is not for many others. Relapse rates for women with mood disorders who discontinue medications in pregnancy are incredibly high: up to 70% in 

 and 85% or higher for women with bipolar disorder.

 Many women suffer symptom relapse when their psychiatric medications are discontinued for pregnancy. In comparison, only one-third of women who do not stop their medications for pregnancy have symptom relapse. Thus, stopping medications practically guarantees that the pregnancy will be exposed to psychiatric illness.

Psychiatric illness (especially depression) during pregnancy has been associated with multiple negative outcomes for the exposed baby, including preterm delivery, low birth weight, decreased motor tone and activity in the baby, higher cortisol levels in the baby, poor reflexes in the baby as well as overall worse health status and increased risk for preeclampsia and gestational diabetes. The negative effects of postpartum depression on exposed babies include lower IQ, slower language development, and higher rates of attention deficit-hyperativity disorder as well as behavior problems and psychiatric illness.

Exposure to maternal psychiatric illness either during pregnancy and/or during the postpartum period can lead to undesirable and negative outcomes for the exposed child. These risks should not be forgotten when considering whether to continue a psychiatric medication during pregnancy.

A second common scenario is that a woman’s medication is continued during pregnancy but at a lower dose than what was needed to get her well. Or, conversely, the medication is continued but not increased or changed in the setting of illness recurrence. In this case, the woman and her baby are being exposed to both the medication and the psychiatric illness. The better approach is to treat to wellness-otherwise, why use the medication in the first place?

In general, perinatal psychiatrists recommend using older medications that we know more about in pregnancy rather than newer ones about which we know little to nothing in pregnancy. This rule of thumb applies to treatment planning before pregnancy. Once a woman is pregnant, the infant is already exposed to the newer medication and switching risks relapse of the illness as well as exposure to the second medication. So, if you are going to make change, do so before pregnancy.

Switching to a category B medication is similar to the previous error but it also involves a misunderstanding of the old FDA pregnancy category B. The old pregnancy categories are being phased out and replaced with a summary on the label of all available data on the safety and risks of a medication during pregnancy and lactation. However, the pregnancy categories are still used for some older medications and are readily available on the internet.

Category B combines medications that have reassuring animal data but no human data, with medications that have demonstrated an adverse effect in animals but have no demonstrated risk in humans. Thus, many providers conclude that if a medication is category B, the medication is safer to use during pregnancy compared with a category C or D medication. However, many Category B classified medications simply have no human data at all and therefore are not preferable over medications that we have safety and risk data for in humans.

Switching to a different medication while breastfeeding because it has a lower passage into breast milk is a well-intentioned practice but also increases the number of exposures for the infant: to the medication used during pregnancy (note that the exposure in utero is far larger than the exposure in breastfeeding), to the medication used in breastfeeding, and potentially to psychiatric relapse. Switching to a different medication while breastfeeding should only be done in the case of a history of response to medication that could not be used during pregnancy but can be used during breastfeeding (such as valproic acid) or in the case of taking a medication during pregnancy that should not be used during breastfeeding, such as clozapine and in some cases lithium. A switch can also be made if there is symptom relapse on the current psychiatric medication(s) but not for the purposes of limiting exposure during breastfeeding.

It often makes sense to try to “clean up” a medication regimen and switch to medications with more data prior to pregnancy. However, when a woman is older, her risk of genetic malformations increases with each passing year. In the spirit of treating the whole patient, taking time to change a medication regimen and risk relapse and spend time finding a medication regimen that is helpful may not make sense. In this setting, the provider does best to educate the patient about this risk to make sure she makes an informed decision about her treatment options.

Many clinicians get nervous about benzodiazepine use in pregnancy and, particularly in the setting of an unplanned pregnancy, will refuse to prescribe benzodiazepines when a woman finds herself pregnant. However, sudden discontinuation of benzodiazepines can precipitate withdrawal symptoms including unstable changes in blood pressure and heart rate, and can result in seizures, none of which are particularly good for the patient, her baby, or the pregnancy.

Attempting to taper and minimize the regular use of benzodiazepines during pregnancy is a more reasonable approach that supports the health of both mother and baby. Tapering of benzodiazepines can be accomplished by switching to a longer-acting agent followed by a taper of 20% every 24 hours in the setting of a motivated patient and good supports. Many patients will require a 

I have evaluated a number of patients concerned about their psychiatric medications in pregnancy yet they continue to smoke tobacco, drink alcohol, use marijuana, or take opiates. It is important to address these known risk factors for negative pregnancy outcomes and emphasize that these substances have very clear risks if continued during pregnancy. In comparison, most (but not all) psychiatric medications do not.

I have evaluated a number of patients in which appropriate treatment has been withheld because the patient was pregnant. Providers have the good intention of not exposing the baby to psychiatric medications; however, they neglect the very real risk of exposure to psychiatric illness and unhealthy and unpredictable maternal behaviors. Furthermore, there are very few medical illnesses for which we withhold treatment during pregnancy and psychiatric illness should not be an exception.

Unfortunately, lithium use during pregnancy has a very bad reputation because of early studies that found an increased rate of Ebstein anomaly in in utero exposed infants. However, more recent data reveal that the risk of Ebstein anomaly in exposed infants is less than 1%, while the risk of relapse of bipolar I disorder with discontinuation of 

during pregnancy is approximately 85% to 100%.

The negative effects of postpartum depression on exposed babies include lower IQ, slower language development, and higher rates of attention deficit-hyperativity disorder as well as behavior problems and psychiatric illness.

Lithium should be considered as a possible treatment option during pregnancy for patients who respond well to lithium and have a history of severe bipolar illness. You also may want to consider switching to an atypical antipsychotic although this carries a risk of relapse in patients who respond to lithium.

Lithium use during breast-feeding remains controversial. Although it can be supported in some individual cases, it is not recommended in patients who are disorganized or unlikely to recognize dehydration in an infant.

To be clear, I am not recommending that valproic acid or carbamazepine not be prescribed to women who may become pregnant. I am, however, recommending that contingency plans be in place when these medications are prescribed in women because neither should be used in pregnancy except in exceptional circumstances. If valproic acid or carbamazepine are to be used in a woman who has the potential to become pregnant, it is important that she clearly understands that she should not take these medications during pregnancy and that any pregnancy should be a planned one, and other medications will be used during pregnancy.

Discussion of contraception choices should be carried out and initiation of contraception should be encouraged and if possible undertaken. In addition, the association between valproic acid exposure and the development of polycyctic ovarian syndrome as well as the potential for decreased fertility should be discussed with the patient.

In general, I recommend prescribing valproic acid and carbamazepine in women of reproductive age only when other, safer medication choices for pregnancy such as lithium or atypical antipsychotics have failed.

Let’s return to the clinical case described at the beginning of this article. What are some appropriate treatment options for Lauren during pregnancy?

She has a history of severe illness requiring hospitalization, a history of a significant suicide attempt, and a history of non-response to SSRI medications. She is, however, on a newer antidepressant about which we know little during pregnancy.

One option is to attempt to switch her to an older medication about which we know more, as long as it is not an SSRI since she has a history of nonresponse.

However, since Lauren is in her late 30s it might not make sense to take the time to undertake medication trials and risk relapse of her illness, which has been severe in the past. Her options should be discussed in detail, using a risk-risk approach: discussing the risks of psychiatric medication use during pregnancy, the risks of switching to a different medication, and the risks of psychiatric illness during pregnancy to both the patient, her pregnancy, and her infant.

Finally, let’s revisit her case with the added detail that Lauren is, in fact ,currently pregnant. Does this change recommendations? Absolutely. If Lauren is already pregnant, it does not make sense to switch to an older medication because the developing baby is already exposed to the medication and switching increases the risk of relapse and exposing the baby not only to the original antidepressant but also a second medication. In this case, it would be better to continue with a medication known to work for Lauren and try to minimize the number of exposures for the baby.

There are no easy answers in designing a psychiatric medication plan for pregnancy, but a calm, thoughtful approach can help us promote maternal mental health which is not only good for the mother, but also for her baby.

is Associate Professor of Psychiatry, Associate Professor of Gynecology and Obstetrics, Director, Johns Hopkins Women’s Mood Disorders Center, Johns Hopkins School of Medicine, Baltimore, MD. She reports no conflicts of interest concerning the subject matter of this article.

1. Cohen LS, Altshuler LL, Harlow BL, et al. Relapse of major depression during pregnancy in women who maintain or discontinue antidepressant treatment. 

2. Viguera AC, Nonacs R, Cohen LS, et al. Risk of recurrence of bipolar disorder in pregnant and nonpregnant women after discontinuing lithium maintenance. 

3. Viguera AC, Whitfield T, Baldessarini RJ, et al. Risk of recurrence in women with bipolar disorder during pregnancy: prospective study of mood stabilizer discontinuation. 

4. Gopalan P, Glance JB, Azzam PN. Managing benzodiazepine withdrawal during pregnancy: case-based guidelines. 

5. Patorno E, Huybrechts KF, Bateman BT, et al. Lithium use in pregnancy and the risk of cardiac malformations. 

6. Newmark RL, Bogen DL, Wisner KL, et al. Risk-benefit assessment of infant exposure to lithium through breast milk: a systematic review of the literature. 

This article describes common misconceptions about treatment of psychiatric illness during pregnancy and advocates for a logical and thoughtful approach to designing a treatment plan.

Common Errors Psychiatrists Make When Managing Mood Disorders in Pregnant Patients

The controversy surrounding the use of antidepressants during pregnancy is well publicized, widespread, and confusing-which leaves many general psychiatrists unsure about how best to care for pregnant patients. This article will summarize the risks of untreated psychiatric illness during pregnancy as well as the risks and benefits of antidepressant use. The goal is to provide practical guidelines to help clinicians and their patients make informed treatment decisions for pregnancy.

Risks of in utero exposure to maternal depression and anxiety

About 15% of all pregnant women have a psychiatric illness.

 The risk of MDD during pregnancy is roughly comparable to the risk during other times in a woman’s life. However, in women with preexisting MDD, the risk of relapse during pregnancy increases when medication is discontinued. The relapse rate is as high as 68% when drug therapy is stopped during the first trimester.

Anxiety disorders are also extremely common during pregnancy: up to 27% of women are affected.

Beyond the obvious risks of suicide and unhealthy behaviors associated with untreated MDD and anxiety (including higher BMI, decreased compliance with prenatal care, higher rates of substance use, poorer nutrition, and decreased rates of breastfeeding), studies indicate that untreated psychiatric illness is directly correlated with adverse neonatal and child outcomes.

 Untreated MDD has been associated with significantly increased risks of preterm birth and low birth weight.

 Other studies have demonstrated elevated cortisol levels in infants exposed to maternal MDD and increased rates of child behavioral and emotional problems.

 Untreated antepartum depression is also one of the strongest risk factors for postpartum depression,

 which is associated with impaired maternal-infant bonding

; and adverse child outcomes, including lower IQ, slower language development, and behavioral disturbance

; as well as potentially devastating consequences such as suicide and infanticide. Clearly, the decision not to treat a woman during pregnancy should be made with great caution.

Interpreting the literature about in utero antidepressant exposure

The literature on the safety of in utero antidepressant exposure is large, robust, and often contradictory. Because many studies are small, retrospective, and observational, it is not uncommon for studies of the same risks and outcomes to have findings that directly contradict one another (for a good example, see the discussion of persistent pulmonary hypertension [PPHN] below). In fact, antidepressants are one of the best-studied classes of medications in pregnancy. To make sense of this vast literature, it is important to bear in mind its substantial limitations, including:

 Unfortunately, many studies are marred by poor control of confounding factors associated with depressive and anxiety disorders. For example, women with MDD differ from the general population of women in several important ways, including health-related behaviors; associated illnesses such as diabetes; and rates of obesity, smoking, and substance use. Studies that have not controlled for these factors may find associations between in utero antidepressant exposure and outcomes that may not be causally linked to exposure to the medication itself, but rather to exposure to these confounding characteristics. The 

 presents results from 4 studies of infant outcomes that have attempted to control for confounding by indication and severity of illness. It demonstrates how results may or may not change with controlling for confounding factors.

 Changes in maternal behavior as well as serum cortisol and neurotransmitter levels associated with active depressive symptoms may have direct effects on infant physiology and development that may be hard to distinguish from the effects of exposure to antidepressants. For example, a recent meta-analysis found an association between antenatal depression and both preterm birth and low birth weight, with a stronger effect for more severe depressive symptoms.

 Few published studies in this area control for active symptomatology. To further complicate the picture, many women who take antidepressants during pregnancy are undertreated and may still experience significant depressive symptoms, thus exposing the infant to both medication and psychiatric illness.

 Many headlines in both the popular press and scientific journals emphasize positive associations with in utero antidepressant exposure without telling the whole story. For example, the popular press (including the 

) recently trumpeted an “87% increased risk of autism” based on a press release by an author’s university about the findings of one study-but the coverage did not reflect that the study found an absolute risk increase from 0.7% to 1.3% and was based on 31 infants.

 Nor was it widely reported that the finding became nonsignificant when rigorous clinical diagnoses of autism were applied. Similarly, many studies find a statistically significant effect of antidepressant exposure on the likelihood of preterm birth-but fail to emphasize that the absolute reduction in gestational age is 3 to 4 days, which is of questionable clinical significance.

Risks studied for in utero antidepressant exposure

 Although findings are conflicting, untreated depression has been associated with preterm birth; thus, it is important to control for depression in antidepressant exposure studies. Several well-designed trials and meta-analyses have found a risk of preterm birth with antidepressant exposure,

 independent of the risk of a diagnosis of depression (see 

). To date, the data indicate that antidepressant use in pregnancy may be associated with an increased risk of preterm birth; however, the clinical significance of this association remains unclear. The risk, if real, may be for 

 untreated illness and antidepressant exposure-and therefore of little use in making treatment decisions.

Cardiac defects and persistent pulmonary hypertension.

 Studies that examined the association between in utero antidepressant exposure and infant cardiac defects exemplify the importance of controlling for confounding factors. Cardiac defects are the most common congenital abnormalities, and early studies seemed to find associations between SSRI use and cardiac defects, especially ventricular septal defects and ventricular outflow tract obstructions.

 Most of these early studies, however, compared psychiatric population outcomes with general population outcomes.

More recent, well-controlled studies have not found an association between antidepressant exposure and heart defects when outcomes are compared in 2 psychiatric populations. Huybrechts and colleagues

 controlled for psychiatric illness comorbidities in a sample of over 900,000 by comparing the outcomes of women with MDD who took antidepressants with those of women with MDD who did not take antidepressants, taking into account both known and unknown differences between these 2 populations (see 

). They found no association. Similarly, a meta-analysis of prospective cohort studies found no association between SSRI use in the first trimester and heart defects when women with MDD who took antidepressants during pregnancy were compared with those who did not.

 Thus, the previously identified association between in utero antidepressant exposure and heart defects appears to be associated with other risk factors.

A similar story has evolved for the possible association between in utero antidepressant exposure and PPHN in the newborn. PPHN occurs when the high pulmonary vascular resistance necessary for fetal circulation in utero fails to diminish appropriately at birth; the result is a right-to-left shunt and decreased neonatal blood flow through the lungs, leading to respiratory failure. Studies to date have been conflicting, but the most recent analyzed close to 3.8 million pregnancies and found no association between SSRIs and PPHN when the analyses were adjusted for potential confounding factors associated with MDD (see 

 There was, however, a significant association when the cases were limited to primary PPHN and term deliveries (odds ratio, 1.28; confidence interval, 1.01-1.64). Notably, a previous study found that PPHN in the newborn was associated with a history of psychiatric hospitalization even when the mother abstained from antidepressant use during pregnancy,

 which indicates that other factors associated with MDD may increase the risk of PPHN rather than in utero antidepressant exposure. The absolute risk remains less than 1%, even if the risk results from SSRI exposure.

 The most recent series of studies examined whether there is an association between in utero antidepressant exposure and autism based on the “serotonin hypothesis of autism,” which posits that fetal exposure to high levels of serotonin in utero leads to the autism phenotype.

The data are conflicting and plagued by limitations. A recent systematic review and meta-analysis found that controlling for maternal mental illness in a meta-analysis of 4 case-control studies eliminated statistically significant findings except for first-trimester exposure and completely attenuated the relationship in 2 pooled cohort studies (see 

 The authors concluded that further study is required to rule out a positive association with first-trimester exposure and/or residual confounding.

One infant outcome that is clearly associated with in utero antidepressant exposure is poor neonatal adaptation syndrome (PNAS). Symptoms can include respiratory distress, cyanosis, apnea, seizures, temperature instability, feeding difficulty, vomiting, hypoglycemia, hypotonia, hypertonia, hyperreflexia, tremor, jitteriness, irritability and, rarely, seizures. Most cases appear to be mild, self-limited, and without lasting repercussions, although recent data suggest that motor and CNS effects may still be identified at 1 month postpartum, and longer-term studies need to be completed.

Approximately one-third of exposed infants have at least mild symptoms consistent with the syndrome, and this risk increases when multiple agents, particularly benzodiazepines, are used.

A practical approach to antidepressant use in pregnancy

How do we integrate all the information about the risks of both treatment and lack of treatment into a practical clinical approach? The ideal situation is to begin planning for pregnancy before conception. The patient’s psychiatric history, severity of illness, past medication response, and wishes for treatment during pregnancy should be taken into account. Every case should be considered individually. Ultimately, there is no set approach: just the weighing of risks and benefits of the various options for the individual patient.

The primary goal of treatment in pregnancy is to minimize the number of exposures. This means not only minimizing the number of medications but also limiting exposure to psychiatric illness.

When determining which antidepressant to use during pregnancy, remember that in general we have more information about older antidepressants than newer ones. This fact is reflected in the old FDA pregnancy categories. Categories include A, B, C, D, and X, and the hierarchy is based on the 

 for safety in animal and human studies, not on the 

. Medications in category B simply have not been studied adequately in humans to warrant placing them in category A as safe (or in C, D, or X), and most medications new to the market will therefore be in category B. Prescribers may make the mistake of prescribing a category B medication over an older medication in category C or D, thinking it is safer, when in fact less is known about its safety during pregnancy. While these categories are being phased out, they will still apply to drugs approved before 2001 for some time.

Should women switch to better-studied medications before pregnancy? If decisions are being made well in advance, and the woman does not have a history of non-response to the better-studied medication, she can attempt a switch. Switching to another antidepressant during pregnancy rarely makes sense (unless the woman is not responding), since the baby has already been exposed and switching would increase the number of exposures to medications and potentially depressive illness.

What if the patient is already pregnant? The same principles apply, with an important addition: 

don’t stop all psychiatric medications immediately

. Stopping precipitously can cause great distress, precipitate withdrawal, and induce a relapse of mental illness. The best approach is to review the medication list based on the principles outlined above. Keep in mind that the baby has already been exposed, and while stopping some medications may make sense to minimize the impact on the child, doing so in a controlled and logical fashion is ideal.

Despite increasing evidence of safety, many patients decide not to take psychiatric medications during pregnancy. Many feel guilty about taking any medication during pregnancy, and most women overestimate the risks of medication and underestimate the risks of untreated mood disorders during pregnancy. In this situation, it is appropriate to offer close follow-up care so that relapses are identified early and treatment can be offered. Evidence-based nonpharmacological treatments, including psychotherapy, yoga, and acupuncture, should be incorporated whenever possible (whether the patient is receiving medication or not!).

It is also important to keep in mind that the patient and her family 

 feel comfortable with the treatment used during pregnancy, so they do not look back and regret decisions made during this critical period. Educating the family about the risks and benefits of treatment, the risks of untreated psychiatric illness to both mother and child, as well as signs and symptoms of relapse, is essential to providing good clinical care.

Interpretation of the literature on in utero antidepressant exposure is complicated by confounding variables associated with psychiatric illness. Current evidence is generally reassuring and indicates that the absolute risks of negative infant outcomes are small except for PNAS, which largely appears to be self-limited. In contrast, the risks associated with exposure to significant maternal depression are substantial. A plan to minimize the number of exposures both to medication and to illness, along with close psychiatric follow-up and communication with all involved parties, serves to maximize outcomes for both mothers and their children.

Dr. Osborne is Assistant Director, Women’s Mood Disorders Center, Johns Hopkins School of Medicine, Baltimore, MD, and Assistant Professor of Psychiatry and Behavioral Sciences, Johns Hopkins School of Medicine. Dr. McEvoy is a Postdoctoral Fellow in Women’s Mental Health in the Department of Psychiatry, Johns Hopkins School of Medicine. Dr. Payne is Director, Women’s Mood Disorders Center, Johns Hopkins School of Medicine, and Associate Professor of Psychiatry and Behavioral Sciences, Johns Hopkins School of Medicine.

Dr. Osborne reports no conflicts of interest concerning the subject matter of this article. Dr. Payne receives research support from and is a consultant for Sage Therapeutics; she has provided expert testimony for Eli Lilly, Astra Zeneca, Johnson and Johnson, and Abbott Pharmaceuticals.

 Andersson L, Sundstrom-Poromaa I, Bixo M, et al. Point prevalence of psychiatric disorders during the second trimester of pregnancy: a population-based study. 

 Cohen LS, Altshuler LL, Harlow BL, et al. Relapse of major depression during pregnancy in women who maintain or discontinue antidepressant treatment. 

 Brunton RJ, Dryer R, Saliba A, Kohlhoff J. Pregnancy anxiety: a systematic review of current scales. J Affect Disord . 2015;176:24-34.

 Kim HG, Mandell M, Crandall C, et al. Antenatal psychiatric illness and adequacy of prenatal care in an ethnically diverse inner-city obstetric population. 

 Grigoriadis S, VonderPorten EH, Mamisashvili L, et al. The effect of prenatal antidepressant exposure on neonatal adaptation: a systematic review and meta-analysis. 

 Field T. Prenatal depression effects on early development: a review. Infant Behav Dev . 2011;34:1-14.

 . Robertson E, Grace S, Wallington T, Stewart DE. Antenatal risk factors for postpartum depression: a synthesis of recent literature. 

 Akman I, Kuscu K, Ozdemir N, et al. Mothers’ postpartum psychological adjustment and infantile colic. Arch Dis Child . 2006;91:417-419.

 McLearn KT, Minkovitz CS, Strobino DM, et al. The timing of maternal depressive symptoms and mothers’ parenting practices with young children: implications for pediatric practice. 

 Grace SL, Evindar A, Stewart DE. The effect of postpartum depression on child cognitive development and behavior: a review and critical analysis of the literature. 

 Jarde A, Morais M, Kingston D, et al. Neonatal outcomes in women with untreated antenatal depression compared with women without depression: a systematic review and meta-analysis. 

 Kimmel M, Hess E, Roy PS, et al. Family history, not lack of medication use, is associated with the development of postpartum depression in a high-risk sample. 

 Cha AE. Maternal exposure to anti-depressant SSRIs linked to autism in children. 

 Almendrala A. Major study links autism to antidepressant use during pregnancy. 

 Boukhris T, Sheehy O, Mottron L, Berard A. Antidepressant use during pregnancy and the risk of autism spectrum disorder in children. 

 Huybrechts KF, Sanghani RS, Avorn J, Urato AC. Preterm birth and antidepressant medication use during pregnancy: a systematic review and meta-analysis. 

 Sit D, Perel JM, Wisniewski SR, et al. Mother-infant antidepressant concentrations, maternal depression, and perinatal events. 

 Louik C, Lin AE, Werler MM, et al. First-trimester use of selective serotonin-reuptake inhibitors and the risk of birth defects. 

 Alwan S, Reefhuis J, Rasmussen SA, et al. Use of selective serotonin-reuptake inhibitors in pregnancy and the risk of birth defects. 

 Pedersen LH, Henriksen TB, Vestergaard M, et al. Selective serotonin reuptake inhibitors in pregnancy and congenital malformations: population based cohort study. 

 Huybrechts KF, Palmsten K, Avorn J, et al. Antidepressant use in pregnancy and the risk of cardiac defects. 

 Wang S, Yang L, Wang L, et al. Selective serotonin reuptake inhibitors (SSRIs) and the risk of congenital heart defects: a meta-analysis of prospective cohort studies. 

 Huybrechts KF, Bateman BT, Palmsten K, et al. Antidepressant use late in pregnancy and risk of persistent pulmonary hypertension of the newborn. 

 Kieler H, Artama M, Engeland A, et al. Selective serotonin reuptake inhibitors during pregnancy and risk of persistent pulmonary hypertension in the newborn: population based cohort study from the five Nordic countries. 

 Harrington RA, Lee LC, Crum RM, et al. Serotonin hypothesis of autism: implications for selective serotonin reuptake inhibitor use during pregnancy. 

 Brown HK, Hussain-Shamsy N, Lunsky Y, et al. The association between antenatal exposure to selective serotonin reuptake inhibitors and autism: a systematic review and meta-analysis. 

 Salisbury AL, O’Grady KE, Battle CL, et al. The roles of maternal depression, serotonin reuptake inhibitor treatment, and concomitant benzodiazepine use on infant neurobehavioral functioning over the first postnatal month. 

 Oberlander TF, Misri S, Fitzgerald CE, et al. Pharmacologic factors associated with transient neonatal symptoms following prenatal psychotropic medication exposure. 

This article summarizes the risks of untreated psychiatric illness during pregnancy as well as the risks and benefits of antidepressant use.