Executive Summary:
First-generation antipsychotics (FGAs) have the longest safety record of any antipsychotic class in pregnancy, with human exposure data spanning over 60 years. The landmark multinational cohort study by Huybrechts et al. (2023) — encompassing over 6,000 FGA-exposed pregnancies across 6 countries — found a malformation prevalence of 3.1% (95% CI 2.7–3.5%) in FGA-exposed infants, compared to 2.7% in unexposed infants, with adjusted relative risks generally close to the null.[1] The 2016 Medicaid cohort (1.3 million pregnancies) similarly found no increased risk for congenital malformations with typical antipsychotics after confounding adjustment (aRR 0.90; 95% CI 0.62–1.31).[2]
FGAs offer a critical metabolic advantage over second-generation antipsychotics (SGAs) in pregnancy. Multiple meta-analyses (2025) have consistently demonstrated that FGAs are not associated with increased gestational diabetes (GDM) risk, whereas SGAs — particularly olanzapine, clozapine, and quetiapine — carry a significantly elevated risk (RR 1.4–1.6).[3][4][5] The Swedish national register study (1.3 million births) confirmed that only the “high-metabolic-risk” SGAs (olanzapine, clozapine, quetiapine) were associated with GDM and large-for-gestational-age infants; FGAs and other SGAs were not.[3]
The primary clinical trade-off — the “Potency-Metabolic Seesaw” — is well-supported by evidence: high-potency FGAs (haloperidol, fluphenazine) carry a higher risk of neonatal extrapyramidal symptoms (EPS) due to potent D₂ blockade, but protect the mother from GDM and excessive weight gain. Low-potency FGAs (chlorpromazine, thioridazine) carry lower EPS risk but higher metabolic, sedative, and autonomic risks. The clinical consensus favors high-potency agents because neonatal EPS is transient and manageable, whereas GDM and macrosomia have lasting implications.[6][7]
Reproductive Safety & Teratogenicity
FGAs provide the most robust long-term safety database of any antipsychotic class in pregnancy.
Large-Scale Human Data:
- Huybrechts et al. (2023) — Multinational Cohort: Among 6,371 FGA-exposed pregnancies across 5 Nordic countries and the US, the prevalence of any major malformation was 3.1% (95% CI 2.7–3.5%), compared to 2.7% in unexposed infants and 4.3% in SGA-exposed infants. After propensity score adjustment, adjusted relative risks for most FGA-malformation combinations were close to the null. The only potential signal was for chlorprothixene (a thioxanthene FGA used primarily in Nordic countries) and cardiac malformations (aRR 1.6; 95% CI 1.0–2.7), though confidence intervals were wide.[1]
- Huybrechts et al. (2016) — US Medicaid Cohort: Among 733 FGA-exposed pregnancies in 1.3 million Medicaid-enrolled women, the adjusted RR for overall malformations was 0.90 (95% CI 0.62–1.31) — no increased risk. The findings for cardiac malformations were similarly null.[2]
- Quinn et al. (2026) — Meta-Analysis: A pooled meta-analysis of 8 studies (>10 million pregnancies) found borderline evidence of association between antipsychotic exposure and congenital malformations overall (OR 1.27; 95% CI 0.996–1.624, p = 0.054), but no association when limited to SGAs (OR 1.16; 95% CI 0.78–1.72). The authors concluded that antipsychotics are “not a major teratogen.”[8]
- Wang et al. (2021) — Hong Kong Sibling-Matched Study: Using sibling-matched analyses to control for shared genetic and environmental factors, this study found no association between prenatal antipsychotic exposure (including FGAs) and ADHD, ASD, preterm birth, or small for gestational age. The authors concluded that “clinicians should not stop administering regular treatment because of a fear of birth outcomes.”[9]
Haloperidol — The “Gold Standard” FGA for Pregnancy:
– Haloperidol has the largest individual FGA exposure database. The FDA label for haloperidol decanoate states: “Available data from published epidemiologic studies of pregnant patients exposed to haloperidol have not established a drug-associated risk of major birth defects or miscarriage.”[10]
– The oral haloperidol label states: “No teratogenic effect has been reported in rats, rabbits or dogs.” [Cleft palate](/rare-disease/cleft-palate) was observed in mice at 15× the maximum recommended human dose, but this is considered a “nonspecific response to stress or nutritional imbalance” rather than a drug-specific effect.[11][12]
– There are isolated case reports of limb malformations following maternal haloperidol use, but these occurred with concomitant use of other drugs with suspected teratogenic potential, and causal relationships were not established.[10][13]
– The Schoretsanitis et al. (2020) systematic review found haloperidol’s mean umbilical cord blood penetration ratio was 0.66 (SD 0.40) — moderate placental transfer, similar to olanzapine (0.71) but higher than quetiapine (0.24).[14]
Chlorpromazine:
– The FDA label states: “Safety for the use of chlorpromazine during pregnancy has not been established.” There are reported instances of prolonged jaundice, extrapyramidal signs, hyperreflexia or hyporeflexia in newborn infants whose mothers received phenothiazines.[15][16]
– Reproductive studies in rodents demonstrated potential for embryotoxicity, increased neonatal mortality, and nursing transfer of the drug. Tests in offspring showed decreased performance. The label notes: “The possibility of permanent neurological damage cannot be excluded.”[15]
– This more cautionary FDA language — unique among antipsychotics — reflects chlorpromazine’s older labeling era rather than definitive evidence of harm. However, combined with its unfavorable metabolic and autonomic profile, it supports chlorpromazine’s classification as a second-line FGA in pregnancy.
Fluphenazine:
– The FDA label states: “The safety for the use of this drug during pregnancy has not been established.” The standard class-wide neonatal EPS warning applies.[17]
– Fluphenazine is available as a long-acting injectable (decanoate), which can be advantageous for adherence in patients with schizophrenia.
Thioridazine:
– The FDA label states: “Reproductive studies in animals and clinical experience to date have failed to show a teratogenic effect with thioridazine.”[18]
– However, thioridazine carries a boxed warning for QTc prolongation and risk of Torsades de pointes and sudden death. A dose of 50 mg produced a mean maximum QTc increase of ~23 msec, with greater prolongation expected at clinical doses.[18][19]
– Thioridazine is contraindicated with other QTc-prolonging drugs, CYP2D6 inhibitors, and in patients with congenital long QT syndrome.[18]
Clinical Pearl — FGAs Have a Lower Malformation Rate Than SGAs: A finding that is often overlooked: in the Huybrechts 2023 multinational study, the crude malformation rate for FGA-exposed infants (3.1%) was actually lower than for SGA-exposed infants (4.3%), and very close to the unexposed rate (2.7%).[1] After adjustment, both classes showed risks close to the null. This is an important counseling point: FGAs are not “older and riskier” — they are, if anything, slightly more reassuring than SGAs regarding structural teratogenicity, likely because the FGA class does not include the metabolically active agents (olanzapine, quetiapine) that drive the slightly higher crude SGA rate.
Clinical Selection: Evidence-Based Risk Stratification
|
Medication |
Potency |
Risk Status |
Primary Indications in Pregnancy |
Key Maternal/Neonatal Considerations |
|
Haloperidol |
High |
Preferred |
Acute mania; psychosis; severe agitation; adjunct for HG nausea |
Gold standard FGA: largest data pool; lowest metabolic/GDM risk; highest neonatal EPS risk; moderate placental transfer (66%) |
|
Fluphenazine |
High |
Preferred |
Chronic psychosis; LAI for adherence |
Low metabolic risk; high neonatal EPS risk; LAI provides stable levels through pregnancy GI changes |
|
Perphenazine |
Mid |
Compatible |
Psychosis with nausea/anxiety |
Moderate sedation; middle-ground safety profile; used in Nordic countries with reassuring registry data |
|
Trifluoperazine |
Mid |
Compatible |
Psychosis; anxiety |
Moderate EPS risk; limited but non-concerning pregnancy data |
|
Chlorpromazine |
Low |
Second-Line |
Refractory nausea (HG); mania; severe agitation |
High metabolic risk (weight gain, sedation); orthostatic hypotension; prolonged neonatal jaundice reported; FDA label uniquely cautionary |
|
Thioridazine |
Low |
Avoid if Possible |
Treatment-resistant cases only |
⚠️ QTc prolongation (boxed warning); risk of Torsades de pointes; CYP2D6-dependent metabolism; multiple contraindications |
The “Potency-Metabolic Seesaw” — Evidence Base
The choice between high-potency and low-potency FGAs involves a deliberate trade-off between neurological and metabolic risks. This concept is well-supported by evidence:
High-Potency FGAs (Haloperidol, Fluphenazine):
– Neurological risk (HIGH): Potent D₂ blockade → higher risk of maternal EPS (dystonia, akathisia) and neonatal EPS (hypertonia, tremors, feeding difficulties).[11][13]
– Metabolic risk (LOW): The Swedish register study confirmed that FGAs were not associated with GDM or LGA infants. Two 2025 meta-analyses confirmed: FGAs showed no significant GDM association (RR 1.31; 95% CI 0.29–2.32 in Liu Zhao; no increased risk in Zhang et al.).[3][4][5]
Low-Potency FGAs (Chlorpromazine, Thioridazine):
– Neurological risk (LOWER): Less D₂ blockade → lower EPS risk, but higher anticholinergic, antihistaminic, and alpha-adrenergic effects.
– Metabolic risk (HIGHER): Potent H₁ blockade → weight gain, sedation, glucose intolerance. Alpha-1 blockade → orthostatic hypotension (risk of falls and reduced uterine perfusion).[15][18]
– Cardiac risk (THIORIDAZINE): QTc prolongation with boxed warning.[19]
The Clinical Consensus: Neonatal EPS is transient (resolving within 2–14 days) and manageable with supportive care in a standard nursery setting.[11][13] In contrast, GDM and macrosomia have lasting implications — increased cesarean delivery risk, shoulder dystocia, neonatal hypoglycemia, and long-term metabolic programming for both mother and child.[3] Therefore, high-potency FGAs are preferred when an FGA is indicated.
Clinical Pearl — The Metabolic Advantage Is Real and Quantifiable: The GDM risk difference between FGAs and high-metabolic SGAs is clinically meaningful. The Swedish register study found an adjusted RR of 1.8 (95% CI 1.3–2.4) for GDM with high-metabolic SGAs (olanzapine, clozapine, quetiapine) vs. untreated women, while FGAs showed no increased risk.[3] The 2025 Liu Zhao meta-analysis confirmed: SGAs had a pooled RR of 1.59 (95% CI 1.24–1.94) for GDM, while FGAs had a non-significant RR of 1.31 (95% CI 0.29–2.32).[5] For patients with pre-existing obesity, metabolic syndrome, or prior GDM, this metabolic advantage makes high-potency FGAs a strong consideration.
Clinical Risks to the Fetus and Neonate
(A) Neonatal Extrapyramidal Symptoms (EPS) and Withdrawal
All FGAs carry the FDA class-wide warning: “Neonates exposed to antipsychotic drugs during the third trimester of pregnancy are at risk for extrapyramidal and/or withdrawal symptoms following delivery.”[11][13][20]
– Symptoms: Agitation, hypertonia, hypotonia, tremor, somnolence, respiratory distress, and feeding disorder.[10][11]
– Severity Spectrum: “These complications have varied in severity; while in some cases symptoms have been self-limited, in other cases neonates have required intensive care unit support and prolonged hospitalization.”[11][20]
– Duration: Typically resolves within 2–14 days postpartum. Most cases are self-limited and manageable with supportive care (swaddling, quiet environment, assisted feeding).[6][7]
– Risk Factors: Higher maternal dose, high-potency agents (haloperidol > chlorpromazine), and use in the third trimester (especially near delivery) increase risk.[11]
The Swedish National Register Study (Heinonen et al., 2022) — the largest population-based study of neonatal morbidity after antipsychotic exposure (n=2,677 exposed, 1.3 million total births) — found:[21]
- – 19.3% of antipsychotic-exposed infants were admitted to neonatal care vs. 7.8% of unexposed infants (aRR 1.7; 95% CI 1.6–1.8).
- – The highest relative risks were for withdrawal symptoms (aRR 17.7; 95% CI 9.6–32.6), neurological disorders (aRR 3.4; 95% CI 2.4–5.7), and persistent pulmonary hypertension (aRR 2.1; 95% CI 1.4–3.1).
- – However, the absolute risks for these outcomes were low among exposed infants: withdrawal symptoms 1.3%, neurological disorders 1.8%, persistent pulmonary hypertension 1.0%.
- – The authors concluded: “These findings do not warrant discontinuation of a necessary treatment but rather increased monitoring of these infants.”[21]
(B) Fetal Growth — The FGA Paradox
Unlike SGAs (which tend to cause macrosomia via maternal hyperglycemia), FGAs are associated with a different growth pattern:
- – Joseph-Delaffon et al. (2024) Meta-Analysis: FGAs were associated with an increased risk of small for gestational age 3rd percentile (OR 1.37; 95% CI 1.02–1.83) and a lower mean birth weight (MD −135g; 95% CI −203 to −66). In contrast, SGAs were associated with large for gestational age >97th percentile (OR 1.56; 95% CI 1.31–1.87).[7]
- – Lin et al. (2022) — Taiwan Cohort: Among women who received antipsychotics, the odds of low birth weight were significantly higher in SGA users compared with FGA users (aOR 1.32; 95% CI 1.04–1.68), suggesting FGAs may be more favorable for birth weight outcomes.[22]
- – Clinical Implication: The slight reduction in birth weight with FGAs is likely related to the absence of metabolic effects (no maternal hyperglycemia → no fetal hyperinsulinism → no overgrowth). This is generally considered a favorable trade-off compared to the macrosomia risk with high-metabolic SGAs.
(C) The “Hypotension Hazard” (Low-Potency Only)
Low-potency FGAs (chlorpromazine, thioridazine) cause significant alpha-1 adrenergic blockade:
- – Maternal Risk: Orthostatic hypotension (fainting), especially in the third trimester when pregnancy-related vasodilation compounds the effect.[15]
- – Fetal Risk: Brief periods of reduced uterine blood flow if maternal blood pressure drops significantly. Chlorpromazine’s FDA label notes that hypotension “rarely resulting in cardiac arrest” has been reported.[15]
- – Prevention: Slow position changes, adequate hydration, compression stockings in T3. Avoid concomitant antihypertensives when possible.
(D) Long-Term Neurodevelopmental Outcomes
- – Kaplan et al. (2026) — Systematic Review: A comprehensive review of 16 studies found that early motor delays may be observed in antipsychotic-exposed infants but did not persist into later childhood. After adjusting for confounders (including maternal illness severity), the risk of neurodevelopmental disorders in school-age children does not appear to be increased.[23]
- – Wang et al. (2021) — Sibling-Matched Analysis: Using sibling-matched analyses to control for shared genetic and environmental factors, no association was found between prenatal antipsychotic exposure and ADHD, ASD, preterm birth, or SGA.[9]
Clinical Pearl — Neonatal EPS Is Manageable; GDM Is Not: The Swedish register data provide the most important counseling framework: while 19.3% of antipsychotic-exposed neonates required neonatal care admission (vs. 7.8% unexposed), the specific serious outcomes (withdrawal, neurological disorders, PPHN) had absolute risks of only 1–2%, and the effects were transient.[21] In contrast, GDM — which is significantly more common with high-metabolic SGAs — has lasting consequences: increased cesarean delivery, shoulder dystocia, neonatal hypoglycemia, and long-term metabolic programming for both mother and child. This is the core argument for preferring high-potency FGAs when metabolic risk is a primary concern.
Haloperidol in Hyperemesis Gravidarum (HG) — A Special Indication
While not FDA-approved for this indication, haloperidol and other D₂ antagonists are used as anti-emetics in refractory HG:
– RCOG and ACOG Guidelines: Phenothiazines (promethazine, prochlorperazine) are recommended as first-line pharmacologic treatment alongside doxylamine-pyridoxine. Dopamine antagonists (metoclopramide, domperidone) are recommended as second-line options.[24]
– AGA Clinical Practice Update (2024): A Cochrane review found no significant difference in efficacy among metoclopramide, ondansetron, and promethazine for HG. Phenothiazines and metoclopramide carry a risk of extrapyramidal adverse effects, and the drug should be withdrawn if such symptoms occur.[25]
– Safety Data: A Cochrane review and systematic review exploring the safety of antiemetics (including phenothiazines and dopamine antagonists) reported no safety signals, no increased risk of teratogenesis, or other adverse pregnancy outcomes.[24]
– Clinical Consideration: When haloperidol is used for HG, it is typically at low doses (0.5–2 mg) and for short durations. The metabolic neutrality of haloperidol is an advantage over ondansetron (which carries QTc concerns) in patients already on other QTc-prolonging medications.
Pharmacokinetic Considerations in Pregnancy
Haloperidol:
– Primarily metabolized by CYP3A4 (with contributions from CYP2D6 and CYP1A2). CYP3A4 is significantly induced during pregnancy (activity increases ~50–100% by T3), which may lead to declining haloperidol levels across gestation.[11]
– Clinical Action: If breakthrough psychotic symptoms emerge in T2/T3, consider checking haloperidol trough levels and adjusting the dose upward. Postpartum, CYP3A4 activity returns to baseline — the dose may need to be reduced to avoid toxicity.
– Haloperidol Decanoate (LAI): The FDA label states the recommended dosing interval is every 4 weeks. The LAI formulation provides stable levels that are less affected by pregnancy-related GI changes (nausea, vomiting, altered motility). However, the long duration of action means that if adverse effects occur, they cannot be rapidly reversed.[10]
Chlorpromazine:
– Extensively metabolized by multiple CYP enzymes. The FDA label notes that chlorpromazine is transferred across the placenta and that “nursing transfer of the drug” occurs in animal studies.[15]
Placental Transfer (Schoretsanitis et al., 2020 Systematic Review):[14]
|
FGA |
Umbilical Cord Blood Penetration Ratio (Mean ± SD) |
Clinical Implication |
|
Haloperidol |
0.66 ± 0.40 |
Moderate transfer — significant fetal exposure |
|
Flupentixol (thioxanthene) |
Amniotic fluid ratio: 0.24 |
Relatively low transfer |
|
Quetiapine (SGA comparator) |
0.24 ± 0.07 |
Low transfer — often cited as advantage |
|
Olanzapine (SGA comparator) |
0.71 ± 0.22 |
High transfer — similar to haloperidol |
High-Yield “Clinic Ready” Tools: The Patient Script
Topic: Why We Prefer High-Potency Agents (e.g., Haloperidol)
“When picking a medicine for your pregnancy, we look at a ‘safety seesaw.’ We prefer medicines like haloperidol because they are ‘metabolically clean’ — they don’t cause the weight gain or blood sugar spikes that can lead to pregnancy-related diabetes.
Haloperidol is one of the most well-studied medicines we have for pregnancy. Data from thousands of pregnancies across multiple countries show it does not cause birth defects. Keeping your weight and sugar in a healthy range is the best way to ensure the baby grows at a normal rate.
These medicines can sometimes make the baby a little stiff or shaky for a few days after they are born, but that is a temporary physical symptom that the pediatricians see regularly and know exactly how to handle. A large study from Sweden found that while babies exposed to these medicines are watched a little more closely after birth, the serious problems are very rare — affecting only about 1 in 100 babies — and they resolve on their own. It is much safer for both of you than dealing with high blood sugar during the pregnancy or having an untreated episode of psychosis.”
High-Yield “Clinic Ready” Tools: EMR Documentation Template
Pregnancy Consultation: Typical Antipsychotic Selection Management
- Selected Medication: [Medication Name] ([High / Mid / Low] Potency).
- Formulation: [Oral / IM PRN / Long-Acting Injectable (Decanoate)].
- Current Dose: [X] mg [daily / BID / TID / q4 weeks (LAI)].
- Risk Status: [Preferred / Compatible / Second-Line / Avoid if Possible].
- Indication: [Schizophrenia / Bipolar Mania / Acute Agitation / Schizoaffective Disorder / Adjunct for Hyperemesis Gravidarum / Other: ___].**
- Gestational Age at Consultation: [___] weeks.**
- EDD: [___].**
Teratogenicity Counseling:
- Informed patient that FGAs are not associated with an increased risk of major congenital malformations based on data from over 6,000 FGA-exposed pregnancies across 6 countries (Huybrechts et al. 2023: malformation prevalence 3.1%, consistent with the 2–4% population baseline).
- [If haloperidol]: Discussed that the haloperidol decanoate FDA label states: “Available data from published epidemiologic studies of pregnant patients exposed to haloperidol have not established a drug-associated risk of major birth defects or miscarriage.”
- [If chlorpromazine]: Discussed that the FDA label carries more cautionary language than other FGAs, including reports of prolonged neonatal jaundice and extrapyramidal signs. However, large-scale cohort data have not confirmed an increased malformation risk.
- [If thioridazine]: ⚠️ Discussed boxed warning for QTc prolongation and risk of Torsades de pointes. Documented rationale for use over safer alternatives.
- Background risk of major birth defects: 2–4%.
Potency-Metabolic Seesaw Discussion:
- Documented the rationale for selecting a [high / mid / low]-potency agent.
- [If high-potency]: Selected to minimize GDM and metabolic risk. Discussed that FGAs are not associated with increased GDM risk (Swedish register: aRR null; 2025 meta-analyses: non-significant). Accepted higher risk of transient neonatal EPS as a manageable trade-off.
- [If low-potency]: Selected because [patient cannot tolerate high-potency agents due to: / specific indication requires sedation / other: ]. Discussed higher metabolic and autonomic risks.
Metabolic Monitoring:
- – [If high-potency agent]: Low metabolic risk. Standard prenatal weight and glucose monitoring. Standard GDM screening at 24–28 weeks. No early screen needed unless other risk factors present.
- – [If low-potency agent]: Moderate-to-high metabolic risk. Monthly weight tracking. Consider early GDM screen if BMI >30 or other risk factors.
Cardiac Monitoring (if applicable):
- – [If thioridazine]: ⚠️ MANDATORY. Baseline ECG performed: Date []. QTc: [] ms. Repeat ECG with dose changes. Contraindicated with other QTc-prolonging drugs, CYP2D6 inhibitors, and congenital long QT syndrome.
- – [If haloperidol]: QTc prolongation is less common but can occur at high doses or with IV administration. Baseline ECG recommended if dose >10 mg/day or if concomitant QTc-prolonging drugs are used.
- – [If chlorpromazine]: Monitor for orthostatic hypotension. Fall risk assessment performed.
Neonatal Risk Discussion:
- – ⚠️ Discussed FDA class-wide warning: neonates exposed to antipsychotic drugs during the third trimester are at risk for EPS and/or withdrawal symptoms (agitation, hypertonia, hypotonia, tremor, somnolence, respiratory distress, feeding disorder).
- – Discussed that the Swedish national register study (n=2,677 exposed) found neonatal care admission rate of 19.3% (vs. 7.8% unexposed), but absolute risks for serious outcomes were low (withdrawal 1.3%, neurological disorders 1.8%, PPHN 1.0%) and effects were transient.
- – [If high-potency agent]: Higher risk of neonatal hypertonia (stiffness), tremors, and feeding difficulties due to jaw rigidity.
- – [If low-potency agent]: Higher risk of neonatal sedation and hypotonia (“floppy baby”) due to antihistaminic/anticholinergic effects.
- – [If chlorpromazine]: Additional risk of prolonged neonatal jaundice per FDA label.
- – Duration: Typically resolves within 2–14 days with supportive care.
Long-Term Neurodevelopmental Counseling:
- Discussed that the most recent systematic review (Kaplan et al. 2026) found that early motor delays may occur but do not persist into later childhood, and after adjusting for confounders, the risk of neurodevelopmental disorders in school-age children does not appear to be increased.
Pharmacokinetic Considerations:
- [If haloperidol]: CYP3A4-dependent metabolism. CYP3A4 is induced during pregnancy → levels may decline in T2/T3. Monitor for breakthrough symptoms. Consider TDM if clinically indicated.
- [If LAI formulation]: Documented that LAI provides stable levels unaffected by pregnancy GI changes. Injection schedule: [q4 weeks]. Next injection due: [Date].
- Postpartum: CYP3A4 activity returns to baseline → may need dose reduction to avoid toxicity.
Smoking Status:
- Current smoking status: [smoker — cigarettes/day / nonsmoker / recently quit: date ].
- [If haloperidol]: CYP1A2 plays a minor role in haloperidol metabolism — smoking cessation effect is less dramatic than with clozapine or olanzapine, but levels should still be monitored.
Monitoring Plan:
- Standard 20-week anatomy scan.
- Monthly maternal weight and BP tracking.
- [If low-potency agent]: Orthostatic BP assessment at each visit.
- [If thioridazine]: Serial ECGs per cardiac monitoring plan above.
- Standard GDM screening at 24–28 weeks [early screen if: ___].**
- Pediatrician/neonatologist notified of:
– Maternal FGA use, specific agent, dose, and potency class
– Risk of neonatal EPS (high-potency) or sedation/hypotonia (low-potency)
– [If chlorpromazine]: Risk of prolonged neonatal jaundice
– Monitoring for withdrawal symptoms, feeding difficulties, respiratory distress
– Symptoms typically resolve within 2–14 days
- Postpartum plan: [maintain current dose / adjust based on PK changes / breastfeeding decision: ___].**
Pregnancy Exposure Registry:
- Patient [enrolled / offered enrollment / declined enrollment] in the National Pregnancy Registry for Atypical Antipsychotics (Massachusetts General Hospital): 1-866-961-2388 or womensmentalhealth.org. (Note: This registry also accepts FGA exposures.)
Clinical Rationale:
[Medication Name] was selected for this pregnancy because [document specific rationale: e.g., patient has been stable on this agent for _* years / metabolic neutrality is prioritized due to pre-existing obesity/GDM history / acute mania requires rapid stabilization / patient has failed SGAs due to: / LAI formulation is necessary for adherence / other: __].** The risk of untreated maternal psychiatric illness****
References
- Association of In Utero Antipsychotic Medication Exposure With Risk of Congenital Malformations in Nordic Countries and the US. Huybrechts KF, Straub L, Karlsson P, et al. JAMA Psychiatry. 2023;80(2):156-166. doi:10.1001/jamapsychiatry.2022.4109.
- Antipsychotic Use in Pregnancy and the Risk for Congenital Malformations. Huybrechts KF, Hernández-Díaz S, Patorno E, et al. JAMA Psychiatry. 2016;73(9):938-46. doi:10.1001/jamapsychiatry.2016.1520.
- Antipsychotic Use During Pregnancy and Risk for Gestational Diabetes: A National Register-Based Cohort Study in Sweden. Heinonen E, Forsberg L, Nörby U, Wide K, Källén K. CNS Drugs. 2022;36(5):529-539. doi:10.1007/s40263-022-00908-2.
- Antipsychotic Use During Pregnancy and Risk of Gestational Diabetes: An Updated Meta-Analysis. Zhang H, Zhang X, Wei L, et al. Journal of Affective Disorders. 2025;:120633. doi:10.1016/j.jad.2025.120633.
- Effect of Antipsychotic Consumption During Pregnancy on Risk of Gestational Diabetes Development: A Systematic Review and Meta-Analysis. Liu J, Zhao Y. Frontiers in Psychiatry. 2025;16:1710101. doi:10.3389/fpsyt.2025.1710101.
- Antipsychotic Therapy During Early and Late Pregnancy. A Systematic Review. Gentile S. Schizophrenia Bulletin. 2010;36(3):518-44. doi:10.1093/schbul/sbn107.
- Neonatal Outcomes After in Utero Exposure to Antipsychotics: A Systematic Review and Meta-Analysis. Joseph-Delaffon K, Eletri L, Dechartres A, et al. European Journal of Epidemiology. 2024;39(10):1073-1096. doi:10.1007/s10654-024-01156-y.
- Antipsychotic Drug Use During Pregnancy and Neonatal Outcomes: A Systematic Review and Meta-Analysis. Quinn D, Donnelly M, O’Neill C. Archives of Women’s Mental Health. 2026;29(1):12. doi:10.1007/s00737-025-01651-5.
- Association Between Prenatal Exposure to Antipsychotics and Attention-Deficit/Hyperactivity Disorder, Autism Spectrum Disorder, Preterm Birth, and Small for Gestational Age. Wang Z, Chan AYL, Coghill D, et al. JAMA Internal Medicine. 2021;181(10):1332-1340. doi:10.1001/jamainternmed.2021.4571.
- Haloperidol Decanoate. Food and Drug Administration. Updated date: 2025-11-26.
- Haloperidol. Food and Drug Administration. Updated date: 2026-04-23.
- Haloperidol. Food and Drug Administration. Updated date: 2024-11-13.
- Haloperidol. Food and Drug Administration. Updated date: 2022-01-25.
- Excretion of Antipsychotics Into the Amniotic Fluid, Umbilical Cord Blood, and Breast Milk: A Systematic Critical Review and Combined Analysis. Schoretsanitis G, Westin AA, Deligiannidis KM, Spigset O, Paulzen M. Therapeutic Drug Monitoring. 2020;42(2):245-254. doi:10.1097/FTD.0000000000000692.
- Chlorpromazine Hydrochloride. Food and Drug Administration. Updated date: 2025-06-16.
- Chlorpromazine Hydrochloride. Food and Drug Administration. Updated date: 2026-02-27.
- Fluphenazine Hydrochloride. Food and Drug Administration. Updated date: 2025-11-25.
- Thioridazine Hydrochloride. Food and Drug Administration. Updated date: 2025-01-15.
- QTc-interval Abnormalities and Psychotropic Drug Therapy in Psychiatric Patients. Reilly JG, Ayis SA, Ferrier IN, Jones SJ, Thomas SH. Lancet (London, England). 2000;355(9209):1048-52. doi:10.1016/s0140-6736(00)02035-3.
- Haloperidol. Food and Drug Administration. Updated date: 2021-11-16.
- Neonatal Morbidity After Fetal Exposure to Antipsychotics: A National Register-Based Study. Heinonen E, Forsberg L, Nörby U, Wide K, Källén K. BMJ Open. 2022;12(6):e061328. doi:10.1136/bmjopen-2022-061328.
- Antipsychotic Use in Early Pregnancy and the Risk of Maternal and Neonatal Complications. Lin HY, Lin FJ, Katz AJ, Wang IT, Wu CH. Mayo Clinic Proceedings. 2022;97(11):2086-2096. doi:10.1016/j.mayocp.2022.04.006.
- Systematic Review: Antipsychotic Medication in Pregnancy and Neurodevelopmental Outcomes in Children. Kaplan CA, Poels EMP, van den Heuvel MI, et al. Journal of the American Academy of Child and Adolescent Psychiatry. 2026;65(3):354-368. doi:10.1016/j.jaac.2025.04.008.
- Hyperemesis Gravidarum. Nana M, Painter R, Williamson C, Nelson-Piercy C. Lancet (London, England). 2026;407(10523):78-89. doi:10.1016/S0140-6736(25)01454-0.
- AGA Clinical Practice Update on Pregnancy-Related Gastrointestinal and Liver Disease: Expert Review. Kothari S, Afshar Y, Friedman LS, Ahn J. Gastroenterology. 2024;167(5):1033-1045. doi:10.1053/j.gastro.2024.06.014.
Responses