Pregnancy and the Puerperium
Psychotropic adverse effects during pregnancy generally do not differ qualitatively from those that may occur at other times, although pregnancy may be a time of increased susceptibility to certain conditions relevant to adverse drug effects (e.g., gestational diabetes and subsequent metabolic dysregulation with SGAs). Pregnancy also may alter drug metabolism in ways that affect both efficacy and tolerability—for example, the third trimester involves both the greatest increase in volume of distribution as well as estrogen-mediated induction of CYP enzymes. Consequently, dosages of SSRIs and other drugs metabolized by CYP isoenzymes may require upward dosing to maintain consistent efficacy, although rapid fluid shifts at the time of delivery may necessitate subsequent dosage reductions to minimize the risk of toxicity.
Teasing apart potential teratogenic effects from base rates of minor or major malformations in the general population is often difficult if not impossible. Complicating the situation further is the need to consider factors such as the adequacy of prenatal care and the possible deleterious effects of undertreated psychopathology on the developing fetus (e.g., the impact of hypercortisolemia in depression on the developing hippocampus).
Breast-Feeding and Teratogenicity
No absolute contraindications exist to breast-feeding during treatment with any psychotropic drug. Potential benefits of psychotropic drug therapy during breast-feeding must be balanced on a case-by-case basis against the potential for unknown risks. Potential teratogenic risks no longer are described by a letter (A, B, C, D, X) rating system, but rather are delineated by labeling language that summarizes known risks from human and animal data alongside clinical considerations for individual medications, in order to help foster more contextualized decision making about a drug’s relative risks versus benefits for a given patient during pregnancy.
Virtually all psychotropic drugs are detectable in breast milk, although the clinical significance of this, if any, is often unknown. Among mood stabilizers, carbamazepine and divalproex are often considered relatively safe during breast-feeding, although case reports exist of infants developing through breast-feeding hepatic dysfunction from exposure to carbamazepine and thrombocytopenia or anemia from exposure to divalproex (Chaudron and Jefferson 2000). Most practitioners advise against breast-feeding during lithium therapy, although it is no longer an absolute contraindication. Serum lithium levels, as well as renal and thyroid function, require monitoring in the newborn during breast-feeding.
Prior to June 30, 2015, the FDA employed a “letter” rating system (A, B, C, D, X) to categorize teratogenic risks associated with pharmaceuticals. That system, deemed overly simplistic, was replaced with the Pregnancy and Lactation Labeling Rule (PLLR) (see www.fda.gov/Drugs/DevelopmentApprovalProcess/DevelopmentResources/Labeling/ucm093307.htm). Intended goals of the PLLR include providing more complete safety information about pregnancy and lactation when available, placing animal safety data in the context of human exposure, and contextualizing medical/disease factors as influencing pregnancy outcomes. Central to the PLLR is the establishment of pregnancy exposure registries maintained by industry (i.e., drug manufacturers) and academic institutions (e.g., the Center for Women’s Mental Health at the Massachusetts General Hospital) or disease state organizations. While the FDA itself does not develop or maintain such pregnancy registries, it does provide a listing of relevant organizations that do (see www.fda.gov/ScienceResearch/SpecialTopics/WomensHealthResearch/ucm134848.htm).
As of June 30, 2015, manufacturer’s package insert information must contain pregnancy and lactation “Risk Summary” sections that separately identify human and animal data as well as situations for which no data are available. Revised product labeling also includes a “Clinical Considerations” section with subsections that address the following:
Disease-associated maternal and/or embryo/fetal risk
Dose adjustments during pregnancy and the postpartum period
Maternal adverse reactions
Fetal/neonatal adverse reactions
Labor or delivery
A separate product information subsection on lactation also must describe a known risk summary, clinical considerations, and available data to help guide clinical decision making. This includes information on whether a drug is systemically absorbed and its concentration in breast milk, actual or estimated infant daily doses, and effects on breastfed infants, if known.
A final product labeling change relevant to “Females and Males of Reproductive Potential” includes information about the known roles for pregnancy testing, contraception, and infertility.
Some reasonable guidelines in selecting medications for pregnant women and those planning for pregnancy include the following:
Select medications that are likely to benefit the woman following a risk-benefit calculation that includes the risk of mental illness to the woman, because untreated psychiatric disorders are known to increase obstetrical complications and poor neonatal outcomes.
Maintain an awareness that the literature regarding pregnancy and psychotropics is evolving and can be difficult to interpret.
Consider consultation with perinatal psychiatrists, who specialize in helping women and their providers make educated evidence-based decisions that take into account maternal mental health and reproductive safety of medications.
In general, it is best to treat women of reproductive potential with medications that would be reasonable in pregnancy, because many women will require long-term maintenance of psychotropics and will plan pregnancies or become pregnant during their treatment.
A number of psychotropic agents have relatively well-established associations with specific potential adverse effects during pregnancy (either anatomical or behavioral teratogenicity, or gestational complications) that are often small but significant, and that typically represent a topic for risk-benefit discussions to be had between doctor and patient rather than absolute contraindications (Table 21–1).
As a general rule, the clinician should strive during patients’ pregnancy to minimize the number of exposures to different drugs, each with their own (often unknown) risks; this principle often favors the retention of an existing psychotropic drug once a pregnancy is confirmed rather than switching to a new drug that introduces additional teratogenic uncertainties. One point of consideration regarding most teratogenic risks is that their relevance often pertains mainly (but not exclusively) to the first trimester, inasmuch as organogenesis is largely completed by the end of the twelfth week of life (cf. the risk for preterm delivery or bleeding diatheses in the newborn due to vitamin K deficiency with third-trimester divalproex exposure). An example of possible teratogenic risk beyond the third trimester is persistent pulmonary hypertension of the newborn (PPHN), a condition typically arising in 1 in 700 live births but found in one study to have an approximate 6-fold increased risk when mothers had been exposed to an SSRI specifically after the twentieth week of gestation (Chambers et al. 2006). Several subsequent studies, however, have failed to support this risk, with most demonstrating no association whatsoever. Therefore, the relationship between PPHN and antidepressants is questionable at this time, but patients must be made aware of this literature and its inconsistent findings. Individual risk-benefit analyses for a given patient must consider the severity of symptoms and the magnitude of response to an SSRI, as well as additional factors that may independently contribute to PPHN (e.g., presence of obesity or diabetes in the mother). Of all the psychotropic medications that have received systematic study in pregnancy, divalproex is the medication most associated with teratogenicity, with a reported rate of 1%–6% of neural tube defects, and long-term neurocognitive deficits demonstrated in exposed 3-year-olds (Meador et al. 2009). In comparison, the risk of cardiovascular anomaly with first-time lithium use is small, less than 1%, and the association between lamotrigine and oral clefting has been only inconsistently observed in registries.
Conflicting and nonsystematic data exist on the potential relationship between antidepressant exposure during pregnancy and low birth weight or intrauterine growth retardation. Although some studies suggest that antidepressant exposure may lead to lower birth weight and size (e.g., length and head circumference), the potential confounding effects of parental size and metabolic parameters, as well as maternal depression influencing fetal growth and development, often are underappreciated.
Neural tube defects, spina bifida, head and facial deformities, cardiac malformations; neonates may have seizures, vomiting, diarrhea, pulmonary problems.
Drug is relatively contraindicated in pregnancy unless benefits are thought to outweigh risks (more often the case in epilepsy than other conditions).
Neural tube defects (■), potential vitamin K deficiency in newborn.
Some clinicians advise supplemental folic acid for divalproex recipients of childbearing potential, although it is not known whether this practice meaningfully counteracts the potential disruption of CNS formation due to first-trimester divalproex exposure.
Approximate 10-fold increased risk of cleft lips or palates in infants with first-trimester in utero exposure to lamotrigine compared with nonlamotrigine-exposed deliveries (Holmes et al. 2008).
Patients should be apprised of current information as part of the informed consent process. Most experts and the FDA attach little importance or generalizability to this never-replicated observation from a single-case registry.
Approximate 1/1,000 to 1/2,000 risk for Ebstein’s anomaly during cardiac development; increased risk for polyhydramnios.
Contemporary perspectives often favor benefits over risk, depending on severity of response to lithium.
Increased risk for cardiac malformations.
Although paroxetine has become relatively contraindicated in pregnancy, some authorities believe that existing data regarding its possible teratogenicity are too preliminary to draw firm conclusions.
As a broad class, may increase risk for preterm labor (specifically, may decrease gestational period by ~7–10 days).
SSRIs, particularly fluoxetine and sertraline, are considered among the safest psychotropic drugs in pregnancy.
Increased incidence of cleft lip or palate (1.4%) relative to other antiepileptic drugs (0.38–0.55%).
Given the minimal psychotropic effects of topiramate, benefits would seldom outweigh risks.
Note. ■=FDA boxed warning; CNS=central nervous system; FDA=U.S. Food and Drug Administration; SSRI=selective serotonin reuptake inhibitor.
Withdrawal in the Newborn
In general, little rationale exists for stopping or lowering a mother’s dosages of antidepressants before delivery during pregnancy. The likelihood of antidepressant discontinuation symptoms in the neonate is low, and if they are suspected, are typically transient and can be managed conservatively with supportive measures. By contrast, the risks for third-trimester or postpartum affective relapse may be considerable and on the whole outweigh the minimal risk for withdrawal phenomena occurring in the newborn.
SSRIs in neonates have been described in case reports to cause possible withdrawal phenomena, comprised of features such as constant crying, shivering, increased muscle tone, and feeding problems. So-called neonatal abstinence syndrome (NAS) has been reported to occur in about 30% of newborns during the first 1–2 weeks of life after maternal exposure to an SSRI; features may include a high-pitched cry, sleep disturbance, difficulty feeding, tachypnea, hyperthermia, frequent yawning, tremors, exaggerated Moro reflex, and hypertonicity or myoclonus (Levinson-Castiel et al. 2006). Phenomena related to NAS are captured and quantified using a measure known as the Finnegan Score (which can be easily determined; see www.thecalculator.co/health/Finnegan-Score-For-Neonatal-Abstinence-Syndrome-(NAS)-Calculator-1025.html).
Such phenomena, if they even occur, are typically self-limited and readily managed by supportive care. Furthermore, naturalistic case registry data have shown no significant differences in neonatal health outcomes in pregnant women for whom SSRIs were discontinued versus continued throughout the third trimester (Warburton et al. 2010).
In December 2010, the FDA Center for Drug Evaluation and Research modified the warnings and precautions for all antipsychotics to include a statement identifying the risk for extrapyramidal side effects and withdrawal in neonates. Such occurrences appear to be self-limited and of little medical consequence.