Ketamine in Pregnancy and Lactation: Safety Considerations and Data Gaps

There is no dedicated pregnancy registry, controlled human trial, or long-term neurodevelopmental follow-up study confirming that ketamine is safe during pregnancy or lactation. That gap is why ketamine in pregnancy and lactation safety considerations and data gaps remain a genuine clinical dilemma for psychiatrists and obstetricians treating severe, treatment-resistant depression or acute suicidality in pregnant and postpartum patients. According to Gaynes and colleagues (2005), perinatal depression affects approximately 10-20% of pregnant individuals, and a meaningful share of those cases do not respond to first-line treatment. Untreated severe depression carries its own serious risks, including suicide, a leading cause of maternal mortality, so clinicians sometimes have to weigh the theoretical risks of ketamine exposure against the documented risks of leaving severe depression untreated. Before considering ketamine in this population, clinicians should also review general contraindications and screening guidance and patient selection criteria that apply outside of pregnancy.

What Preclinical Data Show About Ketamine and the Developing Brain

The NMDA receptor, a glutamate receptor in the brain that plays a central role in neuronal signaling and synaptic plasticity, is the primary target of ketamine's mechanism of action. The most consequential preclinical finding involves NMDA receptor antagonism and apoptotic neurodegeneration, a process in which neurons undergo programmed cell death, in the developing brain. Ikonomidou and colleagues (1999) reported in the journal Science that a single dose of the NMDA antagonist MK-801 produced widespread neuronal apoptosis in the neonatal rat brain, affecting the hippocampus, thalamus, and cortex. Later studies found that ketamine produces similar effects: Brambrink and colleagues (2012) showed that a five-hour ketamine infusion at anesthetic-level plasma concentrations significantly increased neuronal apoptosis throughout the cerebral cortex of neonatal rhesus macaques, and follow-up work linked this exposure to long-term learning, memory, and social behavior deficits (Paule et al., 2011).

These findings prompted the FDA's 2016 Drug Safety Communication, which warned that repeated or lengthy use of general anesthetic and sedation drugs, including ketamine, may affect brain development in children under three years old and in the third trimester of pregnancy.

The vulnerable window corresponds to the brain growth spurt, the period of rapid synapse formation during which NMDA receptors govern neuronal survival, migration, and circuit formation. In humans, this period runs from roughly the third trimester through the first two years of life, with peak vulnerability around birth (Bhutta, 2007). First-trimester exposure, before this window opens, appears to carry a different risk profile centered on structural malformation rather than neuroapoptosis, and conventional teratogenicity studies of ketamine have not shown increased major malformation rates at clinically relevant doses, though high-dose mouse studies have reported minor skeletal variations (Brindle et al., 1987).

Dose and duration matter. The studies showing significant neuroapoptosis used anesthetic-level doses sustained for hours, conditions very different from a single 40-minute sub-anesthetic infusion. Whether a brief, low-dose exposure during the vulnerable window carries meaningful risk for a human fetus is unknown but biologically plausible at some threshold. Pharmacokinetic modeling suggests fetal exposure following a maternal 0.5 mg/kg IV ketamine infusion would be substantially lower than the anesthetic concentrations used in animal studies (Ngan Kee et al., 2014).

Human Data Are Limited to Case Reports

Published human data on ketamine in pregnancy consist mostly of case reports and small retrospective series, and most describe anesthetic-dose ketamine used during cesarean delivery rather than the repeated sub-anesthetic dosing used in psychiatric care. Stutsman and colleagues (2010) described a case of chronic recreational ketamine use during pregnancy with normal outcomes at delivery, though the report did not include long-term developmental follow-up. Li and colleagues (2017) found no increase in adverse neonatal outcomes, measured by Apgar scores and NICU admission, among infants exposed to ketamine anesthesia during cesarean delivery compared with other anesthesia methods, but that study also did not assess neurodevelopment.

No prospective pregnancy registry exists specifically for ketamine or esketamine. Esketamine (Spravato), the FDA-approved intranasal S-enantiomer of ketamine used for treatment-resistant depression, is the only ketamine-related compound with company-issued pregnancy guidance: its label recommends pregnancy testing before starting treatment and advises against use during pregnancy based on preclinical neurodevelopmental data (Janssen Pharmaceuticals, 2019). A pregnancy registry for esketamine has been established, but outcome data have not yet been published in enough detail to draw conclusions. Large epidemiological studies of ketamine exposure during pregnancy do not exist. The International Registry of Antiepileptic Drugs and Pregnancy shows what a dedicated surveillance system could look like, but no equivalent registry tracks ketamine psychiatric use in pregnancy.

How Ketamine Crosses the Placenta

Ketamine crosses the placenta readily because of its lipophilicity, low molecular weight (237.7 Da), and moderate protein binding of about 47%. During continuous maternal infusion, the fetal-to-maternal plasma concentration ratio reaches approximately 0.7-1.0 at steady state, meaning the fetus is substantially exposed (Ngan Kee et al., 2014). After a single maternal infusion, fetal plasma levels track the maternal kinetic curve with a brief lag, peaking within minutes of the mother's peak concentration.

Fetal ketamine metabolism is limited by immature hepatic cytochrome P450 enzymes, which can extend the fetal half-life and increase cumulative exposure relative to the mother. Norketamine, the active metabolite of ketamine, also crosses the placenta and accumulates in fetal circulation, adding to total NMDA receptor-antagonist exposure. Ketamine and its metabolites have also been detected in amniotic fluid after maternal administration, indicating fetal exposure through swallowing of amniotic fluid in addition to direct transplacental transfer. The clinical significance of this additional exposure route after a single dose or intermittent dosing has not been established.

Ketamine and Breastfeeding: What the Data Show

Data on ketamine transfer into human breast milk come from a single small study. Marchetti and colleagues (2022), published in Breastfeeding Medicine, measured ketamine and norketamine in breast milk from four postpartum women after a single 0.5 mg/kg IV ketamine infusion. Ketamine peaked in milk at roughly 100-200 ng/mL, with a milk-to-plasma ratio of about 0.5-0.8, and norketamine levels in milk matched or exceeded the parent drug.

Relative infant dose (RID), the fraction of the maternal weight-adjusted dose an infant receives through breast milk, is the standard measure used to judge medication safety during lactation; a RID below 10% is generally considered acceptable. Based on the Marchetti data, the estimated infant dose via breast milk after a single maternal infusion is roughly 0.5-3%, below that threshold. That estimate applies to single-dose exposure only. It does not account for how repeated maternal infusions might affect cumulative infant exposure, which has not been studied.

A Severity-Based Framework for Clinical Decisions

Decisions about ketamine during pregnancy should weigh the risks of exposure against the risks of untreated or undertreated maternal psychiatric illness. According to Gaynes and colleagues (2005)perinatal depression affects approximately 10-20% of pregnant individuals, and a meaningful share of those cases is treatment-resistant. Untreated severe depression carries its own well-documented risks, including suicidal behavior, a leading cause of maternal mortality, poor prenatal care adherence, preterm delivery, low birth weight, and impaired maternal-infant bonding.

For mild to moderate depression, ketamine exposure is not justified; first-line treatments such as SSRIs and psychotherapy have established pregnancy safety profiles and should be optimized first, and any concurrent medications should be reviewed for drug interactions before adding ketamine. For severe treatment-resistant depression without imminent risk, ketamine should be considered only after exhausting safer alternatives, including multiple antidepressant classes, psychotherapy, and electroconvulsive therapy, and if used, exposure should be limited to the minimum effective number of infusions while avoiding first- and third-trimester exposure when possible. In an acute suicidal crisis with imminent risk, the immediate threat to maternal life can outweigh the theoretical developmental risk of a single infusion, a calculation similar to using other potentially risky medications, such as anticonvulsants or antipsychotics, when the alternative is a life-threatening outcome.

Electroconvulsive therapy (ECT), a treatment that induces a brief, controlled seizure under anesthesia to treat severe psychiatric illness, has a much larger pregnancy safety record than ketamine, spanning several decades of case series and retrospective studies. It is not risk-free; uterine contractions and transient fetal heart rate changes have been reported, but ECT is generally considered safe during pregnancy with appropriate obstetric monitoring (Anderson and Reti, 2009). For severe treatment-resistant depression in pregnancy, ECT is the better-established alternative and should generally be considered before ketamine.

Informed Consent and Shared Decision-Making

If ketamine is considered during pregnancy, informed consent should explicitly cover the absence of established human safety data, the preclinical evidence of NMDA receptor blockade-related neurodevelopmental toxicity at high doses, the uncertainty about whether brief sub-anesthetic exposure carries meaningful human risk, the available alternatives and their risk-benefit profiles, the plan for maternal and fetal monitoring during and after treatment, and the patient's right to decline treatment without penalty.

Given the complexity, this decision should involve the patient, psychiatrist, obstetrician, and, when relevant, a neonatologist. Documenting the clinical reasoning, the alternatives considered, and the patient's informed decision matters for both clinical care and medicolegal protection.

Research Gaps and What Still Needs to Be Studied

The most pressing needs are a prospective registry for ketamine and esketamine exposure during pregnancy, long-term neurodevelopmental follow-up of exposed children, pharmacokinetic modeling of fetal exposure at sub-anesthetic doses, breast milk kinetics during repeated dosing, and evidence-based consensus guidelines. Animal studies examining outcomes after brief, sub-anesthetic exposure during the equivalent of human third-trimester development would help close the gap between the anesthetic-dose animal data and the sub-anesthetic doses used in psychiatric practice.