Antidepressant use during pregnancy: How to avoid clinical and legal pitfalls
Overall risks are low, and fears of lawsuits should not deter appropriate care
Discuss this article at www.facebook.com/CurrentPsychiatry
Recently there has been an increase in advertising soliciting participants for class-action lawsuits involving birth defects and antidepressants, particularly sertraline. Many psychiatrists are unsure why these ads are running in seemingly every medium because there has been no change in the FDA pregnancy classification for most selective serotonin reuptake inhibitors (SSRIs), except for paroxetine going from a C to a D rating in 2005.1 Some studies have found SSRIs increase the risk of adverse birth outcomes and others have not, which makes it difficult for clinicians to know what to discuss with patients regarding the risks and benefits of using antidepressants during pregnancy, as well as the risks of untreated major depressive disorder (MDD).
It can be hard to encourage some patients to take necessary medications in the best of circumstances, let alone suggest that a pregnant woman take a medication that has been labeled “dangerous.” This article seeks to alleviate physicians’ fears about being caught in a no-win situation by:
- explaining factors that may have led to this increase in class-action lawsuits
- clarifying the risks of using certain medications and not treating depression
- suggesting ways physicians can protect themselves and their patients.
The FDA’s position
In July 2006, the FDA issued a public health advisory regarding SSRI use during pregnancy and the possibility of persistent pulmonary hypertension (PPHN).2 This warning was based on a single study that found the risk of developing PPHN (baseline rate: 1 to 2 per 1,000 births) was 6 times greater for fetuses exposed to SSRIs in late pregnancy.3 Many legal websites highlight this 2006 warning as proof of SSRIs’ danger. However, because subsequent studies have had conflicting results, the FDA’s current position is that the risks of using SSRIs during pregnancy are “unknown” (Box).1-4
FDA statements about antidepressant use during pregnancy
2006: In a warning about the risk of persistent pulmonary hypertension (PPHN) with antidepressant use during pregnancy, the FDA acknowledged “decisions about how to treat depression in pregnant women are increasingly complex.”2 The FDA issued the warning based on a study by Chambers et al,3 noting that the study was “too small” to look at individual medications.
This warning also cited a study by Cohen et al4 that found “women who stopped their [antidepressant] medicine were five times more likely to have a relapse of depression during their pregnancy than were the women who continued to take their antidepressant medicine while pregnant.”2 Although the warning identified a potential “rare” danger, the FDA guidance was that “women who are pregnant or thinking about becoming pregnant should not stop any antidepressant without first consulting their physician. The decision to continue medication or not should be made only after there has been careful consideration of the potential benefits and risks of the medication for each individual pregnant patient.”
2011: In this communication,1 the FDA stated “the initial Public Health Advisory in July 2006 on this potential risk was based on a single published study. Since then, there have been conflicting findings from new studies evaluating this potential risk, making it unclear whether use of [selective serotonin reuptake inhibitors (SSRIs)] during pregnancy can cause PPHN.” The FDA also said that the “potential risk with SSRI use during pregnancy remains unknown.”
Risks of depression
Although most physicians know the risks of untreated MDD, they tend to minimize or forget these risks when a woman becomes pregnant. Pregnant women with MDD face not only the expected risks of their psychiatric illness but additionally face risks of pre-eclampsia, suicide (20% of deaths in the postpartum period are due to suicide), and infanticide.5-10 Risks to the fetus include poor prenatal care, increased risk of intrauterine exposure to drugs or alcohol, increased exposure to maternal cortisol with resulting neurodevelopmental changes, preterm delivery, low birth weight, and failure to thrive.6-8 Later difficulties for the child of a mother with untreated depression may include poor stress adaptation, decreased cognitive performance, and behavioral difficulties because of poor mother-child bonding and other factors.6
See Table 1 for key statistics regarding pregnancy and depression.
Statistics on pregnancy and depression
There are approximately 6 million pregnancies each year in the United Statesa
There are approximately 4 million live births each year in the United Statesa
Two percent to 3% of healthy pregnancies result in a birth defect or miscarriageb-d
Sixty percent to 70% of birth complications occur due to an unknown caused
Rates of depression during pregnancy are 7% to 25%b,e,f
Approximately 13% of pregnant women take an antidepressant during pregnancye
Fifteen percent of women with untreated depression in pregnancy attempt suicideb
Twenty percent of deaths in the postpartum period are due to suicidee
Women who discontinue antidepressants are 5 times more likely than women who continue medications in pregnancy to have a relapse of depressiong,h
SSRIs are the antidepressant class most frequently prescribed to pregnant womeni
Sertraline is one of the most frequently prescribed antidepressants perinatally and has low concentration in breast milk and infant serumj,k
SSRIs: selective serotonin reuptake inhibitors
Limitations of research
- the risks associated with depression
- comorbid conditions such as obesity
- maternal age
- poor prenatal care
- how the baby was delivered (eg, Caesarean sections have higher rates of PPHN)13
- illicit substance use
- effects of other medications (80% of pregnant women use medications, including nonsteroidal anti-inflammatory drugs [NSAIDs], which are associated with PPHN).11,16
No medication is 100% safe during pregnancy and all pregnancies have risks. In the United States, 2% to 3% of pregnancies produce infants with malformations,17 which makes it hard to determine if a defect is caused by a medication or occurred spontaneously. This is the baseline risk of all malformations; individual malformations are rare and it is difficult to conduct studies that have adequate power to demonstrate the risk for a specific malformation.
There are several potential adverse outcomes to consider when prescribing psychotropics to a pregnant woman, including miscarriage, malformation, preterm delivery, perinatal toxicity, and behavioral teratogenesis (Table 2).6,7 SSRIs have been implicated in adverse outcomes, but there is no strong evidence that they increase the miscarriage rate, and several studies found no increase in birth defects.6,13,18-20 Regarding teratogenesis, the FDA switched paroxetine from class C to class D because of a potential 1.5% to 2% risk of fetal cardiac malformation, compared with a 1% baseline rate in the general population.21 Drug toxicity or withdrawal in a neonate also is a risk; however, this condition is self-limited and managed supportively by neonatology.22 Behavioral teratogenesis—neurobehavioral problems that develop later in a child’s life—remains a hypothetical concern; research has been conflicting, and studies often used flawed methodology.
Evidence linking SSRIs to an increased risk of adverse birth outcomes often has been based on large, retrospective health system database cohort studies looking at SSRI exposure and associations with conditions such as PPHN, cardiac anomalies, attention-deficit/hyperactivity disorder, and autism.10,23,24 However, correlation is not the same as causation. It is difficult to prove or disprove the causative factor of adverse outcomes in these studies because:
- these databases were not designed to answer these types of exposure questions (eg, limitations in data collected, such as other potential causes not recorded)
- they have many confounding biases (undocumented illicit substance use, possible minimization of smoking history, publication basis for positive findings, etc.)
- individuals who provided the data did not follow a standardized method (eg, variability among individual clinicians).
Many of these limitations are evident in a 2009 study by Pedersen et al,23 who reported the prevalence of septal heart defects was 0.5% (2,315/493,113) among unexposed children and 0.9% (12/1,370) among children exposed to 1 SSRI during early pregnancy (odds ratio [OR]=1.99 [1.13 to 3.53]). Based on this study’s data, the number needed to harm—the number of patients you would need to treat to encounter 1 adverse outcome—was 246, which suggests a relatively low risk. When data for the entire study is reviewed, the ORs for either minor birth defects (control: 7,373/493,113 vs SSRI exposed: 39/1,370; OR=0.88 [0.54 to 1.41]) or major birth defects (control: 15,518/493,113 vs SSRI exposed: 55/1,370; OR 1.21 [0.91 to 1.62]) were not statistically significant (major and minor malformations were defined using European Surveillance of Congenital Anomalies coding).
Not to case aspersions on this group’s work, it should be noted that this study had limitations, including that the researchers:
- did not take into account SSRI dosage
- did not measure depression severity or remittance
- were not able to fully account for potential exposures (eg, over-the-counter NSAIDs)
- were unable to confirm that patients took their medications because the variable measured was prescriptions filled
- did not interview participants about their medication use or symptoms.
In addition, researchers noted that mothers who filled their antidepressant prescription at least twice also were likely to have other factors that put them at higher risk for having a child with birth defects—such as older age or smoking. The biggest problem with the study was a lack of a control group, such as depressed women who did not receive medication (eg, the risk of depression itself could explain the rise, or those with more severe depression could be prescribed antidepressants).15
In a more recent study,24 33 of 11,014 infants exposed to SSRIs after gestational week 20 developed PPHN (absolute risk: 3 per 1,000 births, compared with an incidence of 1.2 per 1,000 births in the general population), with an adjusted OR of 2.1 (95% CI 1.5 to 3.0). Although the authors warned that the results suggest a “class effect,” the rate of PPHN also was higher for mothers with a history of a psychiatric hospitalization within the last 10 years who were not taking medication (OR=1.3, 95% CI 1.0 to 1.6) and the OR for escitalopram (1.5, CI 0.2 to 10.5) was not statistically significant. This study did include a control group, but the 10-year window may have been too wide to represent a group with similar comorbid risks. Similar to the previously discussed study, mothers prescribed SSRIs were older, 1.7 times more likely to be smokers, and twice as likely to be prescribed NSAIDs. The study did not analyze the risk factors of smoking and body mass index because of an initial subset analysis (which was not reported) finding that these known risk factors for PPHN “did not confound the results.”24