Perinatal EvidenceCompass
Prepopulated workbenchMED-060

Mood stabilizers/antiseizure — core

Valproate sodium

Core mood stabilizer/antiseizure medication; psychiatric and epilepsy indications must be separated

Not a categorical safety rating

This record organizes questions; it does not close them.

The clinical fields and position statements are prepopulated from the public-source corpus. They have not completed individual dual extraction, formal appraisal, exact-label reconciliation, or multidisciplinary sign-off. Do not use this page to justify abrupt discontinuation or reflexive switching.

Inventory identity

What this record represents

Canonical generic name
Valproate sodium
Entity type
single-ingredient drug
Active component(s)
Valproate sodium
Formulation / route / interval
Oral; formulation varies
Attachment entry
Sodium valproate
Separate monograph status
No — cross-reference shared molecule monograph
Identity and cross-reference
Valproate sodium = sodium valproate; one canonical row
Inventory verification confidence
Moderate — molecule-level source; exact product/jurisdiction pending

Decision taxonomy

Five positions kept separate

These are workbench statements for reviewer execution. They are not independently commissioned, patient-specific recommendations.

  1. 01

    Stable continuation

    Workbench statement

    Do not withhold or discontinue solely because of pregnancy or lactation. For this record, assess indication, prior response, relapse severity, dose/formulation, alternatives, exposure timing, and mapped evidence; individual commissioned adjudication remains pending.

  2. 02

    New initiation in pregnancy

    Workbench statement

    For a new start, prefer an effective option with the strongest diagnosis-specific efficacy and reproductive evidence when clinically suitable. This record has not received an individual, independently verified initiation ranking.

  3. 03

    Switch before conception

    Workbench statement

    Before conception, consider a switch only when expected benefit exceeds relapse, withdrawal, failed-switch, and dual-exposure risks. Build and monitor the plan before pregnancy when feasible; no universal switch direction is assigned here.

  4. 04

    Unplanned exposure

    Workbench statement

    After unplanned exposure, confirm product, route, dose, timing, indication, co-exposures, and symptoms; do not advise abrupt cessation. Use the mapped evidence and current label to decide whether any targeted fetal assessment is indicated.

  5. 05

    Delivery, postpartum & lactation

    Workbench statement

    Create separate delivery, postpartum-dose, neonatal-observation, and lactation plans. Do not infer breastfeeding suitability from pregnancy evidence or routinely taper before delivery without drug-specific support.

Linked evidence

9 mapped claim summaries

Mapping can include class-wide, framework, formulation, regulatory, or evidence-gap claims. A linked claim is not necessarily a drug-specific causal estimate.

B-006Provisional public source

Valproate has the highest major-congenital-malformation prevalence among commonly studied antiseizure monotherapies across the cited evidence sets.

Public-source certainty descriptor
high for an association; observational registry evidence
Formal appraisal
Not started — full text and independent appraisal pending
Exposure window
first-trimester monotherapy
Comparator
other studied ASM monotherapies/background prevalence about 2.4–2.9%
Effect or numbers carried forward

AAN/AES/SMFM synthesis: MCM prevalence approximately 9.7%; EURAP 2024: 9.9% (153/1,549; 95% CI 8.5–11.5).

Limitations

Dose, indication, registry methods, and live-birth selection affect absolute prevalence.

Sources, citation controls, and verification state

Linked sources

PMID
Not populated in v0.95
DOI
Pending DOI verification
Exact locator
Pending exact table/figure/page/supplement locator
Access date
2026-07-29
Integrity check
Same-day source surveillance recorded; source-by-source correction/retraction verification remains pending
Internal verification
Source-linked; independent external verification pending
B-007Provisional public source

Valproate is associated with neural-tube defects and other organ-specific malformations.

Public-source certainty descriptor
high for direction; moderate for exact prevalence
Formal appraisal
Not started — full text and independent appraisal pending
Exposure window
first trimester/periconception
Comparator
no valproate or lower-risk ASM
Effect or numbers carried forward

AAN/AES/SMFM synthesis: neural-tube-defect prevalence approximately 1.4%, urogenital approximately 1.2%, and renal approximately 1.4%.

Limitations

Registry heterogeneity and dose/polytherapy differences.

Sources, citation controls, and verification state
PMID
Not populated in v0.95
DOI
Pending DOI verification
Exact locator
Pending exact table/figure/page/supplement locator
Access date
2026-07-29
Integrity check
Same-day source surveillance recorded; source-by-source correction/retraction verification remains pending
Internal verification
Source-linked; independent external verification pending
B-008Provisional public source

Prenatal valproate exposure is associated with lower cognitive scores and increased neurodevelopmental-disorder risk relative to several alternatives.

Public-source certainty descriptor
moderate—consistent observational evidence
Formal appraisal
Not started — full text and independent appraisal pending
Exposure window
throughout pregnancy
Comparator
other ASMs or unexposed children
Effect or numbers carried forward

Directionally lower IQ and higher autism-spectrum/neurodevelopmental risk; no single pooled effect used here

Limitations

Confounding by indication, dose, polytherapy, parental factors, and outcome ascertainment.

Sources, citation controls, and verification state

Linked sources

PMID
Not populated in v0.95
DOI
Pending DOI verification
Exact locator
Pending exact table/figure/page/supplement locator
Access date
2026-07-29
Integrity check
Same-day source surveillance recorded; source-by-source correction/retraction verification remains pending
Internal verification
Source-linked; independent external verification pending
B-009Provisional public source

The current US divalproex label contraindicates migraine prophylaxis in pregnancy and in women of childbearing potential not using effective contraception.

Public-source certainty descriptor
regulatory label
Formal appraisal
Not started — full text and independent appraisal pending
Exposure window
preconception and pregnancy
Comparator
alternative therapy
Effect or numbers carried forward

Contraindication for migraine prevention; epilepsy/bipolar use only when alternatives fail or are unacceptable

Limitations

US labeling; other jurisdictions apply different or stricter controls.

Sources, citation controls, and verification state

Linked sources

PMID
Not populated in v0.95
DOI
Pending DOI verification
Exact locator
Pending exact table/figure/page/supplement locator
Access date
2026-07-29
Integrity check
Same-day source surveillance recorded; source-by-source correction/retraction verification remains pending
Internal verification
Source-linked; independent external verification pending
B-010Provisional public source

UK valproate controls include a pregnancy-prevention programme and two-specialist agreement for initiation under age 55 under current materials.

Public-source certainty descriptor
current regulator requirement
Formal appraisal
Not started — full text and independent appraisal pending
Exposure window
preconception and ongoing prescribing
Comparator
not applicable
Effect or numbers carried forward

Pregnancy-prevention programme for women/girls able to become pregnant; two specialists for new initiation under 55 with defined exceptions

Limitations

Jurisdiction-specific rules and exceptions; requirements can change.

Sources, citation controls, and verification state

Linked sources

PMID
Not populated in v0.95
DOI
Pending DOI verification
Exact locator
Pending exact table/figure/page/supplement locator
Access date
2026-07-29
Integrity check
Same-day source surveillance recorded; source-by-source correction/retraction verification remains pending
Internal verification
Source-linked; independent external verification pending
B-031Provisional public source

Once pregnancy is established, replacing an effective antiseizure medicine requires caution because seizure recurrence can harm the pregnant patient and fetus.

Public-source certainty descriptor
guideline recommendation
Formal appraisal
Not started — full text and independent appraisal pending
Exposure window
after conception
Comparator
continuing effective therapy
Effect or numbers carried forward

No single effect estimate; AAN/AES/SMFM caution

Limitations

Decision varies by seizure syndrome, prior control, dose, and drug-specific fetal risk.

Sources, citation controls, and verification state
PMID
Not populated in v0.95
DOI
Pending DOI verification
Exact locator
Pending exact table/figure/page/supplement locator
Access date
2026-07-29
Integrity check
Same-day source surveillance recorded; source-by-source correction/retraction verification remains pending
Internal verification
Source-linked; independent external verification pending
B-032Provisional public source

ACOG recommends against discontinuing mood stabilizers during pregnancy solely because of pregnancy, except that valproate should be avoided when possible.

Public-source certainty descriptor
guideline recommendation
Formal appraisal
Not started — full text and independent appraisal pending
Exposure window
pregnancy
Comparator
discontinuation
Effect or numbers carried forward

Strong recommendation; moderate-quality evidence in ACOG for continuation except valproate

Limitations

Not a mandate to continue every ineffective or unsafe regimen; indication and relapse history remain decisive.

Sources, citation controls, and verification state
PMID
Not populated in v0.95
DOI
Pending DOI verification
Exact locator
Pending exact table/figure/page/supplement locator
Access date
2026-07-29
Integrity check
Same-day source surveillance recorded; source-by-source correction/retraction verification remains pending
Internal verification
Source-linked; independent external verification pending
B-063Provisional public source

A stable patient should not be switched reflexively after conception solely to use a medicine with more pregnancy data.

Public-source certainty descriptor
guideline recommendation
Formal appraisal
Not started — full text and independent appraisal pending
Exposure window
after conception
Comparator
continuing an effective regimen
Effect or numbers carried forward

No single effect estimate; relapse and fetal polypharmacy tradeoff

Limitations

Important exceptions include valproate and compelling toxicity/ineffectiveness; individual history controls the decision.

Sources, citation controls, and verification state
PMID
Not populated in v0.95
DOI
Pending DOI verification
Exact locator
Pending exact table/figure/page/supplement locator
Access date
2026-07-29
Integrity check
Same-day source surveillance recorded; source-by-source correction/retraction verification remains pending
Internal verification
Source-linked; independent external verification pending
B-065Provisional public source

Breastfed infants in the MONEAD cohort generally had substantially lower antiseizure-medicine concentrations than their mothers.

Public-source certainty descriptor
moderate—prospective mother-infant cohort for studied agents
Formal appraisal
Not started — full text and independent appraisal pending
Exposure window
early lactation
Comparator
maternal serum
Effect or numbers carried forward

Infant concentrations substantially lower than maternal concentrations across studied ASMs

Limitations

Does not cover every ASM equally, rare adverse effects, premature/ill infants, or long-term outcomes.

Sources, citation controls, and verification state

Linked sources

PMID
Not populated in v0.95
DOI
Pending DOI verification
Exact locator
Pending exact table/figure/page/supplement locator
Access date
2026-07-29
Integrity check
Same-day source surveillance recorded; source-by-source correction/retraction verification remains pending
Internal verification
Source-linked; independent external verification pending