ubmslatePT-logo-ubm

PT Mobile Logo

Search form

Topics:

Antidepressants in Pregnancy: Balancing Needs and Risks in Clinical Practice

Antidepressants in Pregnancy: Balancing Needs and Risks in Clinical Practice

SIGNIFICANCE FOR THE PRACTICING PSYCHIATRISTSIGNIFICANCE FOR THE PRACTICING PSYCHIATRIST
Risks of untreated psychiatric illness during pregnancyRisks of untreated psychiatric illness during pregnancy
Tips for reading the psychiatric literature on in utero antidepressant exposure Tips for reading the psychiatric literature on in utero antidepressant...
Studies of infant outcomes adjusted for psychiatric illness and severityTABLE. Studies of infant outcomes adjusted for psychiatric illness and...
Practical advice for designing a psychiatric medication plan for pregnancyPractical advice for designing a psychiatric medication plan for pregnancy

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.1 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.2 Anxiety disorders are also extremely common during pregnancy: up to 27% of women are affected.3

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.4 Untreated MDD has been associated with significantly increased risks of preterm birth and low birth weight.5 Other studies have demonstrated elevated cortisol levels in infants exposed to maternal MDD and increased rates of child behavioral and emotional problems.6 Untreated antepartum depression is also one of the strongest risk factors for postpartum depression,7 which is associated with impaired maternal-infant bonding8; impaired parenting behavior9; and adverse child outcomes, including lower IQ, slower language development, and behavioral disturbance10; 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:

1) Confounding by indication. 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 Table 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.

2) Confounding by active illness. 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.11 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.12

3) Statistical meaningfulness. 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 Washington Post13 and the Huffington Post14) 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.15 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.16,17

Pages

 
Loading comments...

By clicking Accept, you agree to become a member of the UBM Medica Community.