Residency · Residency · Psychiatry
Psychopharmacology in Pregnancy and Lactation
General Principles of Risk-Benefit Analysis
Framework for Decision-Making
Untreated maternal psychiatric illness carries significant risks: poor prenatal care, substance use, self-harm, impaired bonding, preeclampsia, preterm birth. Every medication decision involves balancing teratogenicity and perinatal risks against risks of untreated illness. No psychotropic medication is FDA category A (adequate studies showing no risk); most are category C (animal risk, no adequate human studies) The FDA letter category system (A/B/C/D/X) was replaced in 2015 by the Pregnancy and Lactation Labeling Rule (PLLR) requiring narrative risk summaries. Shared decision-making with the patient (and partner/family when appropriate) is essential -- document thoroughly.
Timing of Exposure
First trimester (weeks 1-12): period of organogenesis; highest risk for structural malformations. Second/third trimester: risk of neurobehavioral effects, growth restriction, perinatal complications. Peripartum: risk of neonatal adaptation syndrome, neonatal toxicity, withdrawal. Postpartum: medication decisions also affect breastfeeding.
Antidepressants in Pregnancy
SSRIs
Most extensively studied psychotropic class in pregnancy. Sertraline and citalopram/escitalopram: generally considered first-line due to most reassuring safety data. Paroxetine: early reports of cardiac malformations (atrial/ventricular septal defects) led to FDA category D designation; subsequent meta-analyses show small absolute risk increase (~0.1-0.2% above baseline); avoid if possible, but do not abruptly discontinue in a patient who is stable on it. Fluoxetine: extensive data showing no major teratogenic risk; long half-life may complicate neonatal adaptation. All SSRIs: possible small increase in persistent pulmonary hypertension of the newborn (PPHN) with late-pregnancy exposure (absolute risk remains low, ~0.2-0.3%) Neonatal adaptation syndrome: jitteriness, irritability, feeding difficulties, respiratory distress in ~30% of neonates exposed to SSRIs in late pregnancy; usually self-limited (1-7 days)
SNRIs
Less data than SSRIs; venlafaxine and duloxetine not associated with major malformations in available studies. Venlafaxine associated with neonatal withdrawal symptoms similar to SSRIs. If patient is stable on an SNRI, continuation is generally preferred over switching.
Other Antidepressants
Bupropion: limited data; no clear teratogenic signal; may be less effective for anxiety-predominant presentations. Mirtazapine: limited human data; animal studies reassuring; may be useful for hyperemesis gravidarum due to antiemetic properties. TCAs: decades of use; nortriptyline and desipramine preferred (less anticholinergic); therapeutic drug monitoring recommended. MAOIs: generally contraindicated in pregnancy due to hypertensive crisis risk and limited data.
Mood Stabilizers in Pregnancy
Lithium
Classic teratogen: Ebstein anomaly (tricuspid valve malformation) historically reported at 400x baseline rate from the Lithium Baby Register (ascertainment bias) Current data: relative risk approximately 1.2-1.7 for cardiac malformations overall; absolute risk of Ebstein anomaly is 0.05-0.1% (vs. 0.005% baseline) Fetal echocardiography recommended at 16-20 weeks if first-trimester exposure. Lithium levels fluctuate in pregnancy (increased renal clearance, expanded volume of distribution) -- monitor levels monthly, then weekly near term. Reduce dose by 30-50% at onset of labor or planned delivery to prevent neonatal toxicity (acute reduction in renal clearance postpartum) Neonatal risks: floppy baby syndrome, nephrogenic diabetes insipidus, thyroid dysfunction.
Valproate
Known major teratogen -- avoid in women of childbearing potential whenever possible. Neural tube defects (spina bifida) in 1-2% of exposed pregnancies (10-20x baseline) Fetal valproate syndrome: craniofacial anomalies, limb defects, cardiac malformations. Dose-dependent neurodevelopmental effects: reduced IQ (8-9 points on average), increased risk of autism spectrum disorder. FDA pregnancy category X for migraine prevention; strong warnings for bipolar disorder.
Carbamazepine
Neural tube defects in ~0.5-1% (2-5x baseline), craniofacial anomalies. Lower teratogenic risk than valproate but higher than lamotrigine. Folic acid supplementation (4-5 mg/day) recommended starting preconception.
Lamotrigine
Safest mood stabilizer in pregnancy based on available registry data. No significant increase in major malformations at doses <300 mg/day. Clearance increases by up to 300% during pregnancy (particularly in second/third trimester due to UGT1A4 induction by estrogen) -- dose increases often necessary. Monitor levels and clinical status; reduce dose postpartum to prevent toxicity. Cleft palate signal in initial registry data not confirmed in larger studies.
Summary: Reproductive Safety of Psychotropic Medications
| Medication Class | Teratogenicity Risk | Key Concerns | Breastfeeding Compatibility |
|---|---|---|---|
| SSRIs (sertraline, escitalopram) | Low | Neonatal adaptation syndrome (~30%); PPHN (absolute risk 0.2-0.3%) | Sertraline preferred (RID 0.5-2%) |
| SSRIs (paroxetine) | Low-Moderate | Small increase in cardiac septal defects (0.1-0.2% above baseline) | Compatible (low transfer) |
| SNRIs | Low | Neonatal withdrawal; less data than SSRIs | Limited data; generally acceptable |
| Lithium | Moderate | Ebstein anomaly (absolute risk 0.05-0.1%); neonatal toxicity | Relatively contraindicated; requires infant monitoring |
| Valproate | HIGH | NTDs 1-2%; IQ reduction 8-9 points; autism risk; fetal valproate syndrome | Compatible (low milk transfer) |
| Carbamazepine | Moderate | NTDs 0.5-1%; craniofacial anomalies | Compatible |
| Lamotrigine | Low | No significant malformation increase <300 mg/day; clearance increases 300% in pregnancy | Variable infant levels (25-50% of maternal); monitor |
| Atypical antipsychotics | Low | Gestational diabetes (olanzapine/clozapine); neonatal EPS | Limited data |
| Benzodiazepines | Low | Neonatal sedation, floppy infant, withdrawal with chronic use | Use with caution; prefer short-acting |
Antipsychotics in Pregnancy
General Considerations
Atypical antipsychotics are increasingly prescribed in pregnancy (for bipolar disorder, psychosis, augmentation) Large registry studies (National Pregnancy Registry for Atypical Antipsychotics) show no consistent increase in major malformations. Metabolic effects: gestational diabetes risk may be increased with olanzapine and clozapine. Neonatal EPS and withdrawal symptoms reported with third-trimester exposure (FDA class-wide warning, 2011)
Specific Agents
Quetiapine and olanzapine: most human pregnancy data among atypicals; no confirmed teratogenic signal. Aripiprazole: growing safety database; no clear teratogenic risk. Clozapine: limited data; risks include gestational diabetes, agranulocytosis monitoring challenges, neonatal seizures.
Benzodiazepines in Pregnancy
Historical concern for cleft lip/palate with first-trimester exposure (early case-control studies); more recent meta-analyses do not support a significant association. Late pregnancy exposure: neonatal sedation, hypothermia, feeding difficulties, "floppy infant syndrome". Neonatal withdrawal (if chronic maternal use): irritability, tremor, seizures. Use the lowest effective dose for the shortest duration; prefer agents with no active metabolites (lorazepam)
Breastfeeding Considerations
General Principles
Relative infant dose (RID) <10% of the maternal weight-adjusted dose is generally considered acceptable. Sertraline: very low breast milk transfer (RID ~0.5-2%); first-line antidepressant in breastfeeding. Paroxetine: low breast milk transfer; acceptable for breastfeeding. Fluoxetine: higher breast milk transfer and long-acting metabolite; not first-line but not contraindicated. Lithium: excreted in breast milk; infant monitoring required (lithium levels, renal function, thyroid); generally considered relatively contraindicated but case-by-case decisions. Valproate and carbamazepine: relatively low breast milk transfer; considered compatible with breastfeeding (ironic given pregnancy risks) Lamotrigine: variable breast milk levels; infant serum levels can reach 25-50% of maternal levels; monitor infant for rash and sedation. LactMed database (NIH): free resource for evidence-based breastfeeding safety information.
<image> A risk comparison matrix for psychotropic medications in pregnancy. The x-axis shows medication categories (SSRIs, SNRIs, lithium, valproate, carbamazepine, lamotrigine, atypical antipsychotics, benzodiazepines). The y-axis shows risk categories (major malformations, neurodevelopmental effects, neonatal adaptation syndrome, obstetric complications). Use a color-coded heat map (green = minimal risk, yellow = some concern, red = significant risk) with specific malformation types noted in each cell. Include absolute risk percentages where known. Clean medical reference style. </image>
<image> A clinical decision flowchart for managing psychiatric medication in pregnancy. Start with "Patient on psychotropic medication discovers pregnancy" or "Pregnant patient requires psychiatric treatment initiation." Branch by diagnosis (depression, bipolar disorder, psychosis). For each branch, show first-line and alternatives with risk annotations. Include decision points for each trimester and peripartum period. Show monitoring requirements (lithium levels, fetal echo, lamotrigine levels). Color-coded by risk level with clear annotation boxes. </image>
<image> A pharmacokinetic changes in pregnancy diagram showing how pregnancy physiology affects drug metabolism. Illustrate increased plasma volume, increased GFR, increased hepatic UGT activity, decreased albumin binding, and altered CYP enzyme activity. Show specific examples: lamotrigine clearance increasing 300%, lithium clearance increasing requiring dose adjustment, and the postpartum reversal of these changes. Use a split-panel format comparing non-pregnant vs. third trimester pharmacokinetics. </image>
Clinical Pearls
The most dangerous medication decision in pregnancy is often abrupt discontinuation of a necessary psychotropic -- relapse of severe psychiatric illness carries its own teratogenic risks (cortisol exposure, poor nutrition, substance use, suicide) Sertraline is the best-studied antidepressant in pregnancy and breastfeeding and is the default first-line choice. Valproate is the one psychotropic that should be actively avoided in all women of childbearing potential unless no alternative exists -- document counseling about contraception and teratogenicity. Lamotrigine levels must be monitored closely in pregnancy (clearance increases dramatically) and reduced postpartum to avoid toxicity. Lithium management requires monthly level monitoring, dose reduction at delivery, and awareness of postpartum rebound toxicity. Fetal echocardiography at 16-20 weeks is recommended for lithium-exposed pregnancies. The LactMed database (freely available from NIH) is the gold standard resource for breastfeeding medication safety.
References
- Epstein RA, et al. Psychopharmacology in pregnancy. In: Schatzberg AF, Nemeroff CB, eds. The American Psychiatric Association Publishing Textbook of Psychopharmacology. 5th ed. 2017.
- Patorno E, et al. Lithium use in pregnancy and the risk of cardiac malformations. N Engl J Med. 2017;376(23):2245-2254.
- Meador KJ, et al. Fetal antiepileptic drug exposure and cognitive outcomes at age 6 years (NEAD study). Lancet Neurol. 2013;12(3):244-252.
- Cohen LS, et al. National Pregnancy Registry for Atypical Antipsychotics. Reports available at womensmentalhealth.org.
- LactMed. Drugs and Lactation Database. National Library of Medicine. https://www.ncbi.nlm.nih.gov/books/NBK501922/
- ACOG Committee Opinion No. 757: Screening for Perinatal Depression. 2018.
