Psychiatrists who treat women of childbearing age should consider that those women may become pregnant, and that women with psychiatric illness are more likely to have unplanned pregnancies.1 Thus, thoughtful perinatal medication choices should begin before pregnancy. Pregnancy is a time of vulnerability to psychiatric illness for many reasons, including physiologic changes that can affect mental status; changes in medication efficacy; and numerous stressors, such as new responsibilities and limited sleep.1,2 For the treatment of pregnant—or potentially pregnant—patients, we recommend the following.
Do not panic! Knee-jerk medication changes in response to learning a patient is pregnant can lead to an exacerbation of psychiatric symptoms, as well as decrease trust in clinicians.2 Switching to a medication with a purportedly “safer” reproductive profile may worsen psychiatric illness, while also exposing the fetus to a medication of unknown benefit. 2
Recognize the risk of untreated or undertreated psychiatric illness, either of which has the potential to harm both the woman and her fetus. For example, a pregnant woman in a manic state may be more likely to engage in risky behaviors, such as drug use or risky sexual activity, which can lead to adverse fetal outcomes. They may also present with a higher risk of suicide. Compared to nonpregnant women, pregnant women for whom lithium was discontinued were equally likely to experience illness recurrence and significantly more likely to experience postpartum illness recurrence.3 In addition, the risk of recurrence was greater after rapid discontinuation compared with gradual discontinuation.3
Accurately communicate research findings. Pregnancy risk categories are no longer used. A nuanced interpretation of the potential adverse effects of a medication, such as malformations, impaired fetal growth, birth outcomes (such as preterm birth), and neurodevelopmental sequelae is necessary. Physicians must accurately convey information about risks to their patients, including both the absolute risk of an adverse event and the possible range of severity. For example, lithium use during pregnancy confers a higher relative risk of Ebstein’s anomaly (a cardiac defect).4 However, the absolute incidence of this risk remains low: 0.6% of lithium-exposed infants vs 0.18% among unexposed infants.4 Ebstein’s anomaly also varies significantly in severity—serious cases may require surgery, but less serious cases need only monitoring. A reliable database that compiles the latest evidence may help in staying abreast of the latest data.
Treat the psychiatric illness. Consider the optimal treatment for the psychiatric illness. Lithium remains the gold standard treatment for bipolar I disorder, regardless of reproductive status. Olanzapine and quetiapine are also commonly used and effective during pregnancy. This is an opportunity to conduct a detailed review of the patient’s previous medication regimens, including a review of medication trials and efficacy. Keep in mind that untreated bipolar disorder also carries an increased risk of adverse pregnancy outcomes.5
Consider pregnancy timing. Most organs form between weeks 3 to 8 of pregnancy. For example, if a medication potentially affects heart formation, but the patient is in the third trimester, explain to her that the heart has already been formed. Consider that medication may be required long-term and affect future pregnancies. Pregnant women require more frequent monitoring, because blood volume changes in pregnancy and postpartum can affect medication levels and efficacy. In addition, note whether a woman plans to breastfeed and be mindful of a medication’s profile in breastfeeding.
Ensure the patient can provide informed consent. Communicate your diagnostic formulation and treatment options. Consider involving the patient’s partner and/or support system in the discussion, if the patient consents. If a patient cannot provide informed consent, a surrogate decision-maker should be identified.6
Continue to: Collaborate with other clinicians