Perinatal Psychopharmacology for Prescribers: A Complete Clinical Reference

Quick Reference

| Clinical question | First-line answer | |---|---| | First-line SSRI in pregnancy | Sertraline (most data; no clear teratogenic signal) | | First-line SSRI in breastfeeding | Sertraline (lowest RID; typically undetectable in infant serum) | | New-onset moderate-severe PPD, no prior medication history | Sertraline + therapy (CBT or IPT); consider zuranolone for faster onset | | Severe PPD with suicidality or functional impairment | Inpatient evaluation; brexanolone if accessible; fast-acting SSRI titration | | Perinatal psychosis | Emergency evaluation; olanzapine or haloperidol; hospitalization in most cases | | Patient on SSRI asking about breastfeeding safety | Continue; LactMed data supports most agents; sertraline preferred for new starts | | Patient stopping SSRI to breastfeed | Do not endorse; relapse risk is significant; refer to discussion with prescriber | | Benzodiazepine for acute anxiety in breastfeeding | Short-acting lorazepam at lowest effective dose; limit duration |

---

The Core Decision Framework

Perinatal psychopharmacology is consistently more nuanced than the available evidence supports. Prescribers face three asymmetric risks:

Risk of treating: Fetal or neonatal exposure to psychotropic medication, potential teratogenic effects (most are low or absent for well-studied agents), neonatal adaptation syndrome.

Risk of not treating: Untreated psychiatric illness in the mother, which carries documented risks to fetal development, obstetric outcomes, infant attachment and development, and maternal function and safety.

Risk of untreated illness vs. treated illness: The comparison is not "medication vs. no exposure." It is "fetal exposure to medication vs. fetal exposure to untreated psychiatric illness." Untreated depression elevates cortisol, reduces self-care and prenatal care utilization, increases substance use risk, and is associated with adverse obstetric outcomes. This is not a neutral comparator.

For most PMAD presentations, the risk-benefit calculation favors treatment.

---

Antidepressants in Pregnancy

SSRIs

Sertraline: Best safety profile for pregnancy initiation. Most data available; no consistent teratogenic signal. Neonatal adaptation syndrome (NAS) with third-trimester use: transient jitteriness, irritability, feeding difficulty in approximately 30 percent of exposed neonates; self-resolving in most cases within days. Not a reason to withhold treatment.

Escitalopram: Acceptable second-line. No clear teratogenic signal. NAS profile similar to sertraline.

Fluoxetine: Extensive data; no consistent teratogenic signal. Long half-life complicates management of NAS. Not first-line for new starts in pregnancy but continuation preferred over destabilization in patients who are responding.

Paroxetine: Historical concern about cardiac defect risk (VSDS) based on early registry data; subsequent larger studies have not confirmed a clinically meaningful signal. Most guidelines now consider the risk low but still list it as not-preferred for new starts due to the historical signal and significant discontinuation syndrome.

SNRIs (venlafaxine, duloxetine): Less data than SSRIs. Generally acceptable when SSRIs have not been effective. NAS profile includes hypertension and jitteriness; neonatology should be aware.

Bupropion

Lower seizure threshold is a theoretical concern in pregnancy; not confirmed as a clinical problem at therapeutic doses. Reasonable option when SSRIs produce unacceptable side effects. Limited but generally reassuring teratogenicity data.

Tricyclics

Nortriptyline and imipramine have a long safety record in pregnancy. More side effect burden than SSRIs. Consider when SSRI trials have failed or when the patient has prior response history.

Authoritative references

Teratogen databases: OTIS/MotherToBaby (mothertobaby.org), REPROTOX, TERIS. These are updated with current evidence and should be consulted for individual agents. Package insert data is typically outdated and overestimates risk based on animal data and small early cohorts.

---

Antidepressants in Breastfeeding

See our companion article on medication-safe breastfeeding for complete RID data. Key summary:

• Sertraline: RID 0.5–3%; typically undetectable in infant serum. First-line.
• Escitalopram: RID 3–8%; second-line.
• Fluoxetine: RID 2–15%; active metabolite accumulates; not first-line for new starts in breastfeeding.
• LactMed (drugs.ncbi.nlm.nih.gov/lactmed): Authoritative source for breastfeeding-specific data. Use this, not the package insert.

---

Brexanolone and Zuranolone

Both are neuroactive steroid GABA-A receptor modulators with FDA approval specifically for PPD.

Brexanolone (Zulresso):

• 60-hour continuous IV infusion administered in a certified healthcare facility (REMS program)
• Rapid onset: measurable improvement within 24 to 36 hours in clinical trials
• Response rates in pivotal trials: approximately 70 to 75 percent response, 50 percent remission
• Breastfeeding: pumping and discarding during infusion; may resume after clearance
• Practical barrier: inpatient infusion requirement limits access; typically reserved for moderate-severe PPD with functional impairment

Zuranolone (Zurzuvae):

• Oral, 14-day course (50 mg nightly with a fatty meal)
• Approved August 2023; first oral agent specifically approved for PPD
• Onset: response within days 3 to 7 in clinical trials
• Not approved for MDD (only PPD); no Schedule IV designation
• CNS depression: do not drive for 12 hours after each dose
• Breastfeeding: avoid during treatment and for one week after last dose
• Clinical positioning: patients who need faster onset than SSRI titration allows, or who have not responded to SSRIs

For a complete clinical guide see our article on brexanolone and zuranolone for prescribers.

---

Benzodiazepines in Perinatal Anxiety

Benzodiazepines in pregnancy require individual risk-benefit discussion. The historical concern about oral cleft association has not been confirmed in larger studies; current evidence does not support a significant teratogenic signal at therapeutic doses. Risks in late pregnancy include neonatal sedation and hypotonia with delivery-proximate use.

Clinical positioning for benzos in pregnancy:

• Short-term use for acute anxiety that cannot wait for SSRI titration is defensible
• Limit duration; avoid late third-trimester use when possible
• Lorazepam is generally preferred (no active metabolites)

In breastfeeding:

• Short-acting agents (lorazepam) at the lowest effective dose for the shortest duration
• Single bedtime dosing minimizes infant exposure
• Avoid continuous scheduled dosing in breastfeeding patients when possible

SSRIs for chronic anxiety: For perinatal patients with anxiety disorders, SSRIs are first-line for maintenance treatment. Benzodiazepines are adjunctive for acute management, not ongoing treatment.

For a complete prescribing reference see our article on benzodiazepine use in perinatal anxiety.

---

Perinatal Psychosis

Perinatal psychosis is a psychiatric emergency. Onset is typically within the first two weeks postpartum; early signs can emerge within 24 to 72 hours of delivery.

Clinical features: Rapid onset (days, not weeks); fluctuating level of consciousness; confusion and disorientation; delusions (often involving the infant); command hallucinations; severe mood disturbance. Distinguishing feature from postpartum depression: the rapid fluctuating course and confusion.

Risk factors: Prior postpartum psychosis (60 to 80 percent recurrence without prophylaxis), personal or family history of bipolar disorder, first postpartum period.

Management:

• Hospitalization in virtually all cases
• Olanzapine or haloperidol for acute stabilization; both are acceptable in breastfeeding at appropriate doses
• Mood stabilizer consideration if bipolar etiology
• Lithium is evidence-supported for postpartum psychosis in bipolar patients but requires careful monitoring; LactMed guidance on breastfeeding with lithium

See our companion article on perinatal psychosis recognition and triage for the full clinical protocol.

---

When to Start vs. Discontinue Psychiatric Medication

Starting in pregnancy:

• New PMAD onset: treat. Risk of untreated illness exceeds medication risk in most presentations.
• Existing patient planning pregnancy: preconception taper discussion should weigh recurrence risk carefully. Many stable patients elect to continue; guideline recommendation is generally to continue when there is significant psychiatric history.
• Stopping "just in case": not evidence-supported. Untreated illness is not a safe default.

Tapering before delivery:

• First trimester exposure: teratogenicity window.
• Third trimester taper to avoid NAS: the evidence does not clearly support this approach for SSRIs. NAS is typically mild and transient; the relapse risk from tapering a stable patient may be greater than the NAS risk.
• Do not taper without a clear clinical rationale and patient-specific risk-benefit analysis.

For the complete decision framework see our article on when to start or discontinue psychiatric medication in pregnancy.

---

Co-Prescribing With Therapy

Medication plus therapy produces better outcomes than either alone for moderate-to-severe PMADs. The evidence base for therapy in perinatal patients is strongest for CBT and IPT; EMDR is evidence-based specifically for birth trauma/PTSD.

For prescribers: Referral to a PMH-C certified therapist (Postpartum Support International certification) is the most specific referral you can make. General therapists may not have perinatal specialization.

Communication with co-treating therapist: A brief shared understanding of the treatment plan improves coordination. "Patient is on sertraline 50mg, titrating to 100mg; we are targeting both anxiety and depressive symptoms; I expect improvement within 4 to 6 weeks" gives the therapist useful context for pacing and interpreting the clinical picture.

For the co-prescribing framework see our article on co-prescribing for PMAD patients in therapy.

---