Although pregnancy has typically been considered a time of emotional well-being, recent studies suggest that up to 20% of women suffer from mood or anxiety disorders during pregnancy. Particularly vulnerable are those women with histories of psychiatric illness who discontinue psychotropic medications during pregnancy. In a recent study which prospectively followed a group of women with histories of major depression across pregnancy, of the 82 women who maintained antidepressant treatment throughout pregnancy, 21 (26%) relapsed compared with 44 (68%) of the 65 women who discontinued medication. This study estimated that women who discontinued medication were 5 times as likely to relapse as compared to women who maintained treatment.
High rates of relapse have also been observed in women with bipolar disorder. One study indicated that during the course of pregnancy, 70.8% of the women experienced at least one mood episode. The risk of recurrence was significantly higher in women who discontinued treatment with mood stabilizers (85.5%) than those who maintained treatment (37.0%).
Although data accumulated over the last 30 years suggest that some medications may be used safely during pregnancy, knowledge regarding the risks of prenatal exposure to psychotropic medications is incomplete. Thus, it is relatively common for patients to discontinue or to avoid pharmacologic treatment during pregnancy.
The New U.S. FDA Pregnancy Labeling and Lactation Rule
In 1975, the U.S. Food and Drug Administration (FDA) provided guidelines to drug companies for labeling medications with regard to their safety during pregnancy. This system of classification used five risk categories (A, B, C, D and X) based on data derived from human and animal studies. While widely used to make decisions regarding the use of medications during pregnancy, many criticized this system of classification, indicating that this type of drug labeling was often not helpful and, even worse, may be misleading.
In an effort to improve the accuracy and usefulness of information regarding the safety of medications used during pregnancy and breastfeeding, the FDA proposed a newly designed system on June 30th 2015. The Pregnancy and Lactation Labeling Rule or PLLR will abolish the letter categories and instead will include more comprehensive information discussing the potential risks and benefits to the mother and the fetus, and how these risks may change during the course of pregnancy.
Companies will be required to remove the pregnancy letter categories from the labeling for all prescription drugs and will have to revise the labeling with updated information. Medications approved before June 30, 2001 are not covered by the PLLR.
Weighing the Risks
Women with histories of psychiatric illness frequently come in for consultations regarding the use of psychotropic medications during pregnancy. Not infrequently, women present with the first onset of psychiatric illness while pregnant. Many pregnancies are unplanned and may occur unexpectedly while women are receiving treatment with medications for psychiatric disorders. Many women may consider stopping medication abruptly after learning they are pregnant, but for many women this may carry substantial risks.
Decisions regarding the initiation or maintenance of treatment during pregnancy must reflect an understanding of the risks associated with fetal exposure to a particular medication but must also take into consideration the risks associated with untreated psychiatric illness in the mother. Psychiatric illness in the mother is not a benign event and may cause significant morbidity for both the mother and her child; thus, discontinuing or withholding medication during pregnancy is not always the safest option.
Depression and anxiety during pregnancy have been associated with a variety of adverse pregnancy outcomes. Women who suffer from psychiatric illness during pregnancy are less likely to receive adequate prenatal care and are more likely to use alcohol, tobacco, and other substances known to adversely affect pregnancy outcomes. Several studies have described low birth weight and fetal growth retardation in children born to depressed mothers. Preterm delivery is another potential pregnancy complication among women experiencing distress during pregnancy. Pregnancy complications related to maternal depression and anxiety in late pregnancy have also been described, including an increased risk for having pre-eclapsia, operative delivery, and infant admission to a special care nursery for a variety of conditions including respiratory distress, hypoglycemia, and prematurity. These data underscore the need to perform a thorough risk/benefit analysis of pregnant women with psychiatric illness, including evaluating the impact of untreated illness on the baby and the mother, as well as the risks of using medication during pregnancy.
What are the Risks of Medication Exposure?
All medications diffuse readily across the placenta, and no psychotropic drug has yet been approved by the Food and Drug Administration (FDA) for use during pregnancy. When prescribing medications during pregnancy, one must consider the following risks associated with prenatal exposure: risk of teratogenesis, risk of neonatal toxicity, and risk of long-term neurobehavioral sequelae.
Risk of Teratogenesis
The baseline incidence of major congenital malformations in newborns born in the United States is estimated to be between 2 and 4%. During the earliest stages of pregnancy, formation of major organ systems takes place and is complete within the first 12 weeks after conception. Therefore, discussion around risks of exposures during pregnancy may be broken down by the timing of exposure or trimester, with particular vigilance around first trimester exposures.
A teratogen is defined as an agent that interferes the in utero development process and produces some type of organ malformation or dysfunction. For each organ or organ system, there exists a critical period during which development takes place and is susceptible to the effects of a teratogen. For example, neural tube folding and closure, forming the brain and spinal cord, occur within the first four weeks of gestation. Most of the formation of the heart and great vessels takes place from four to nine weeks after conception, although the entire first trimester is often considered pertinent.
Risk of Neonatal Symptoms
Neonatal toxicity or perinatal syndromes (sometimes referred to as neonatal “withdrawal”) refer to a spectrum of physical and behavioral symptoms observed in the acute neonatal period that can be attributed to drug exposure at or near the time of delivery. Anecdotal reports that attribute these syndromes to drug exposure must be cautiously interpreted, and larger samples must be studied in order to establish a causal link between exposure to a particular medication and a perinatal syndrome.
Risk of Long-Term Effects
Although the data suggest that some medications may be used safely during pregnancy if clinically warranted, our knowledge regarding the long-term effects of prenatal exposure to psychotropic medications is incomplete. Because neuronal migration and differentiation occur throughout pregnancy and into the early years of life, the central nervous system (CNS) remains particularly vulnerable to toxic agents throughout pregnancy. While exposures to teratogens early in pregnancy may result in clear abnormalities, exposures that occur after neural tube closure (at 32 days of gestation) may produce more subtle changes in behavior and functioning.
Behavioral teratogenesis refers to the potential of a psychotropic drug administered during pregnancy to have long-term neurobehavioral effects. For example, are children who have been exposed to an antidepressant in utero at risk for cognitive or behavioral problems at a later point during their development? To date, few studies have systematically investigated the impact of exposure to psychotropic medications in utero on development and behavior in humans.
Antidepressants and Pregnancy
Of all the antidepressants, fluoxetine (Prozac) is the best characterized antidepressant. Data collected from over 2500 cases indicate no increase in risk of major congenital malformation in fluoxetine-exposed infants. One prospective study of 531 infants with first trimester exposure to SSRIs (mostly citalopram, n=375) did not demonstrate an increased risk of organ malformation.
Several meta-analyses combining studies with exposures to SSRIs do not demonstrate an increase in risk of congenital malformation in children exposed to these antidepressants, with the exception of paroxetine (Paxil). There has been particular controversy around paroxetine use in pregnancy, as past reports have suggested that first trimester exposure to paroxetine was associated with an increased risk of cardiac defects including atrial and ventricular septal defects. Other published studies have not demonstrated increased teratogenicity of paroxetine. Importantly, independently conducted meta-analyses of available data sets have consistently found a lack of association between paroxetine exposure and cardiovascular malformations. Even so, these findings prompted the FDA to change the category label of paroxetine from C to D.
Three prospective and more than ten retrospective studies have examined the risk of organ malformation in over 400 cases of first trimester exposure to tricyclic antidepressants of TCAs. When evaluated on an individual basis and when pooled, these studies do not indicate a significant association between fetal exposure to TCAs and risk for any major congenital anomaly. Among the TCAs, desipramine and nortriptyline are often preferred since they are less anti-cholinergic and the least likely to exacerbate orthostatic hypotension that occurs during pregnancy.
Bupropion may be an option for women who have not responded to fluoxetine or a tricyclic antidepressant, as data thus far have not indicated an increased risk of malformations associated with bupropion use during pregnancy. The most recent information from the Bupropion Pregnancy Registry maintained by the manufacturer GlaxoSmithKline includes data from 517 pregnancies involving first trimester exposure to bupropion. In this sample, there were 20 infants with major malformations. This represents a 3.9% risk of congenital malformation that is consistent with what is observed in women with no known teratogen exposure. While this information regarding the overall risk of malformation is reassuring, earlier reports had revealed an unexpectedly high number of malformations of the heart and great vessels in bupropion-exposed infants. A retrospective cohort study including over 1200 infants exposed to bupropion during the first trimester did not reveal an increased risk of malformations in the bupropion-exposed group of infants nor did it demonstrate an increased risk for cardiovascular malformations.
Scant information is available regarding the reproductive safety of monoamine oxidase inhibitors (MAOIs), and these agents are generally not used in pregnancy as they may produce a hypertensive crisis when combined with tocolytic medications, such as terbutaline.
With regard to the newer antidepressants, prospective data on 150 women exposed to venlafaxine (Effexor) during the first trimester of pregnancy suggest no increase in risk of major malformation as compared to non-exposed controls. To date, the literature does not include prospective data on the use of duloxetine (Cymbalta).
Another prospective study assessed outcomes in 147 women taking either nefazodone (n=89) or trazodone (n=58) during their first trimester of pregnancy and compared them to two control groups of women exposed to either non-teratogenic drugs (n = 147) or to other antidepressants (n=147). There were no significant differences among exposed and non-exposed groups with regard to rates of congenital malformations. In another report, there were no differences in malformation rates among women who took mirtazapine (Remeron) (n=104) during pregnancy as compared to women who took other antidepressants or controls exposed to known nonteratogens.
While these initial reports are reassuring, larger samples are required to establish the reproductive safety of these newer antidepressants. It is estimated that at least 500 to 600 exposures must be collected to demonstrate a two-fold increase in risk for a particular malformation over what is observed in the general population. In general, the SSRIs, specifically fluoxetine, citalopram, and sertraline, are the antidepressants most commonly used during pregnancy.
Several recent studies have suggested that exposure to SSRIs near the time of delivery may be associated with poor perinatal outcomes. Attention has focused on a range of transient neonatal distress syndromes associated with exposure to (or withdrawal from) antidepressants in utero. These syndromes appear to affect about 25% of babies exposed to antidepressants late in pregnancy. The most commonly reported symptoms in the newborns include tremor, restlessness, increased muscle tone, and increased crying. Reassuringly, these syndromes appear to be relatively benign and short-lived, resolving within 1 to 4 days after birth without any specific medical intervention.
These studies deserve careful consideration, yet one of the major shortcomings is that most have failed to use raters blinded to the mother’s treatment status. The decision to admit a newborn to a special care nursery may represent a reasonable precaution for an infant exposed to medication in utero and may not be an indication of a serious problem. Another limitation is that few studies have attempted to assess maternal mood during pregnancy or at the time of delivery. There is ample evidence to suggest that depression or anxiety in the mother may contribute to poor neonatal outcomes, including premature delivery and low birth weight, and it is important to evaluate the contribution of maternal mood to neonatal outcomes.
Based on these findings, many women are advised to taper or discontinue treatment with SSRIs prior to delivery; however, this strategy has not been shown to change neonatal outcomes. Importantly, neonatal effects have been reported with both untreated mood and anxiety disorders, as well as with medication, and limited studies have adequately teased out these variables. One important consideration is that discontinuation of or reductions in the dosage of mediation in the latter part of pregnancy may increase the risk of postpartum depression, as the postpartum period is a time of increased vulnerability to psychiatric illness and depression or anxiety during pregnancy has been associated with postpartum depression.
Another concern has been that maternal SSRI use may be associated with a higher than expected number of cases of persistent pulmonary hypertension of the newborn (PPHN). In one report, the use of an SSRI antidepressant after the 20th week of gestation was significantly associated with a six-fold greater risk of PPHN. If we assume that these findings are correct, the risk is still relatively small; the authors estimate the risk of PPHN to be less than 1% in infants exposed to SSRIs in utero. Since the initial report on this topic, three studies have found no association between antidepressant use during pregnancy and PPHN, and one study showed a much lower risk than the 1% originally reported. These findings taken together bring into question whether there is an association at all and suggest that, if there is a risk, it is much lower than that reported in the original 2006 report.
To date only two studies have systematically investigated the impact of exposure to antidepressants in utero on development and behavior in humans. The first of these studies followed a cohort of 135 children who had been exposed to either tricyclic antidepressants or fluoxetine (Prozac) during pregnancy (most commonly during the first trimester) and compared these subjects to a cohort of non-exposed controls. Results indicated no significant differences in IQ, temperament, behavior, reactivity, mood, distractibility, or activity level between exposed and non-exposed children followed up to 7 years of age. A more recent report from the same group that followed a cohort of children exposed to fluoxetine or tricyclic antidepressants for the entire duration of the pregnancy yielded similar results. The authors concluded that their findings support the hypothesis that fluoxetine and tricyclic antidepressants are not behavioral teratogens and do not have a significant effect on cognitive development, language or behavior.
For women with bipolar disorder, maintenance treatment with a mood stabilizer during pregnancy can significantly reduce the risk of relapse. However, many of the medications commonly used to treat bipolar disorder carry some teratogenic risk when used during pregnancy.
Concerns regarding fetal exposure to lithium, have typically been based on early reports of higher rates of cardiovascular malformations (e.g., Ebstein’s anomaly) following prenatal exposure to this drug. More recent data suggest the risk of cardiovascular malformations following first trimester exposure to lithium is smaller than previous assessments and is estimated to be between 1 in 2000 (0.05%) and 1 in 1000 (0.1%). Compared to lithium, prenatal exposure to some anticonvulsants is associated with a far greater risk for organ malformation. First trimester use of carbamazepine (Tegretol) has been associated with a 1% risk of neural tube defect. Of all of the medications used for psychiatric disorders, the one with the greatest potential of serious birth defects is valproate (valproic acid, Depakote). Factors which appear to increase the risk for teratogenesis include higher maternal serum anticonvulsant levels and exposure to more than one anticonvulsant. With a risk of neural tube defect ranging from 1 to 6%, this drug is often considered one of last resort in reproductive aged women, since the risk for teratogenicity is high in very early pregnancy, before many women realize they are pregnant.
Prenatal exposure to valproic acid has also been associated with characteristic craniofacial abnormalities, cardiovascular malformation, limb defects and genital anomalies, as well as other central nervous system structural abnormalities. Also, valproate exposure during pregnancy has been associated with poorer neurocognitive development in children followed to three years of age. In the same study, lamotrigine use (discussed below) did not affect neurocognitive development.
While other anticonvulsants are being used more frequently in the treatment of bipolar disorder, there is limited information on the reproductive safety of these newer anticonvulsants, specifically gabapentin (Neurontin), oxcarbazepine (Trileptal), tigabine (Gabitril), levetiracetam (Keppra), zonisamide (Zonegran). One report has raised concerns regarding potential teratogenicity of topiramate (Topamax).
However, there is a growing body of information the reproductive safety of lamotrigine (Lamictal), and this may be a useful alternative for some women. The International Lamotrigine Pregnancy Registry was created by GlaxoSmithKline (GSK) in 1992 to monitor pregnancies exposed to lamotrigine for the occurrence of major birth defects. Data from the Registry did not show an elevated risk of malformations associated with lamotrigine exposure.
Other data from the North-American Anti-Epileptic Drug Registry indicates the prevalence of major malformations in a total of 564 children exposed to lamotrigine monotherapy was 2.7%; however, five infants had oral clefts, indicating a prevalence rate of 8.9 per 1000 births. In a comparison group of 221,746 unexposed births, the prevalence rate for oral clefts was 0.37/1000, indicating a 24-fold increase in risk of oral cleft in infants exposed to lamotrigine. However, other registries have not demonstrated such a significant increase in risk for oral clefts. It is important to put this risk into perspective. If we assume that the findings from the North American registry are true, the absolute risk of having a child with cleft lip or palate is about 0.9%.
Atypical antipsychotic agents (discussed in greater detail below) are commonly used often to manage the acute symptoms of bipolar illness, as well as for maintenance treatment. While the data regarding the reproductive safety of these newer agents is limited, no studies thus far have indicated any teratogenic risk associated with this class of medications. For this reason, some women may choose to use an atypical antipsychotic agent during pregnancy (especially during the first trimester) in order to avoid using a known teratogen, such as lithium or valproic acid.
The consequences of prenatal exposure to benzodiazepines have been debated for over twenty years. Three prospective studies support the absence of increased risk of organ malformation following first trimester exposure to benzodiazepines. More controversial has been the issue of whether first trimester exposure to benzodiazepines increases risk for specific malformations. Although initial reports suggested that there may be an increased risk of cleft lip and palate, more recent reports have shown no association between exposure to benzodiazepines and risk for cleft lip or palate. This risk– if it exists — is calculated to be 0.7%, approximately a ten-fold increase in risk for oral cleft over that observed in the general population. Nonetheless, the likelihood that a woman exposed to benzodiazepines during the first trimester will give birth to a child with this congenital anomaly, although significantly increased, remains less than 1%.
Currently, no systematic data are available on the reproductive safety of non-benzodiazepine anxiolytic agents such as buspirone and hypnotic agents zolpidem (Ambien) and zalepion (Sonata). Therefore, these medications are not recommended for use in pregnancy.
In addition to the atypical antipsychotic medications described above, recent studies have not demonstrated teratogenic risk associated with high-or medium-potency neuroleptic medications; however, a recent meta-analysis of the available studies noted a higher risk of congenital malformations after first trimester exposure to low-potency neuroleptic agents. In clinical practice, higher potency neuroleptic agents such as haloperidol (Haldol), perphenazine (Trilafon), and trifluoperazine (Stelazine) are recommended over the lower potency agents in managing pregnant women with psychiatric illness.
Atypical antipsychotic medications are increasingly being used to treat a spectrum of psychiatric disorders, including psychotic disorders and bipolar disorder, as well as treatment refractory depression and anxiety disorders. The first and largest published prospective study on the reproductive safety of the atypical agents provided reassuring data regarding the risk of malformations in the first trimester, although aripiprazole (Abilify) was not among the medications studied. Investigators prospectively followed a group of 151 women taking olanzapine (Zyprexa), risperidone (Risperdal), quetiapine (Seroquel), or clozapine (Clozapine) and compared outcomes to controls without exposure to known teratogens. There were no differences between the groups in terms of risk for major malformations, or rates of obstetrical or neonatal complications.
While this information is reassuring, it is far from definitive, and larger studies are required to provide more information about the reproductive safety of these medications. To this end, the National Pregnancy Registry has been created to prospectively gather information regarding outcomes in infants exposed in utero to these newer atypical antipsychotic medications.
The U.S. Food and Drug Administration (FDA) recently updated labels for the entire class of antipsychotic drugs to include warnings regarding the use of antipsychotic drugs (both the typical and atypical agents) during pregnancy. The new drug labels now contain more details on the potential risk for abnormal muscle movements (extrapyramidal signs or EPS) and withdrawal symptoms in newborns exposed to these drugs during the third trimester of pregnancy. These recommendations were derived from adverse event reporting. While this may signal a potential problem associated with exposure to antipsychotic medications, it does not yield accurate information regarding the prevalence of an adverse event.
For the latest information on psychiatric disorders during pregnancy, please visit our blog.
How do I get an appointment?
Consultations regarding treatment options can be scheduled by calling our intake coordinator at 617-724-7792.
Pregnancy Research at the CWMH
Currently we are conducting several studies related to depression and pregnancy. To learn more about this study or to sign up for our research patient registry, please click here