Educational draft · awaiting clinical reviewThe full textbook explains uncertainty but does not replace live national or local guidance, specialist advice, or current prescribing information.
Preconception psychotropic review
Psychiatry · UK scope · Sources checked 27 Aug 2026 · Clinical review pending
!
Unexpected pregnancy or acute deterioration
A positive pregnancy test while taking valproate, lithium or another medicine needing specialist review requires prompt coordinated advice, while suicidality, mania, psychosis, severe agitation or inability to maintain safety requires same-day emergency mental-health assessment.
Action: Confirm the medicine, dose, adherence, last menstrual period and immediate mental state; do not advise abrupt self-discontinuation; contact the relevant psychiatric and obstetric or medicines-in-pregnancy specialists urgently, and use emergency services when the person, fetus, baby or another person may be in immediate danger.
Role and principles
Plan conception without destabilising mental health by reviewing reproductive intentions, relapse history, psychotropic exposure, physical health, contraception and the support needed before, during and after pregnancy.
Preconception review is not a search for a medicine that is completely risk free. It is a structured comparison of imperfect options for an individual whose untreated illness may itself harm nutrition, sleep, antenatal engagement, relationships, safety and parenting. Begin with the person's priorities, language, plans for conception and previous experience of pregnancy. Ask whether pregnancy is desired now, later, uncertain or actively avoided; this determines whether the immediate task is treatment optimisation, reliable contraception, fertility advice or urgent pregnancy assessment.
Construct a longitudinal relapse map. Record diagnoses, episode polarity, psychosis, suicide attempts, self-neglect, admissions, compulsory treatment, postpartum episodes, seasonal pattern, time to relapse after stopping, and which medicines and psychological interventions actually helped. Include family history of bipolar disorder or postpartum psychosis. A stable person with repeated severe relapse after withdrawal faces a different balance from someone whose diagnosis is uncertain or whose medicine has never been effective.
Assessment and patient selection
High relapse liability
Previous rapid deterioration after dose reduction, recurrent mania or psychosis, severe depression, admission or postpartum psychosis makes discontinuation particularly consequential.
Medicine-specific reproductive concern
Valproate, lithium, carbamazepine, benzodiazepines and some antipsychotic or antidepressant exposures require different counselling, monitoring and neonatal plans rather than a single class rule.
Fertility and endocrine effect
Hyperprolactinaemia from a prolactin-raising antipsychotic can impair conception; sexual dysfunction, menstrual change, polycystic ovary syndrome and substance use may also affect plans.
Contextual vulnerability
Housing insecurity, domestic abuse, migration barriers, limited support, previous traumatic care or difficulty attending appointments can determine whether a technically sound plan is workable.
Protective continuity
A known effective treatment, insight into early warning signs, supportive relationships and rapid access to a familiar team can reduce avoidable instability.
Red flags requiring action
Current suicidal intent, psychosis, mania, severe self-neglect or rapidly falling function overrides a routine preconception appointment and needs urgent specialist assessment.
Pregnancy while taking valproate requires urgent specialist contact because exposure carries major fetal risks, but abrupt unsupervised cessation can precipitate relapse or seizures.
Baseline assessment
01Pregnancy intention and contraception reviewFirst step
Why
Establish current pregnancy possibility, desired timing and whether contraception is acceptable, effective and safely controlled by the person.
Interpretation and limitations
A late period or uncertain exposure prompts pregnancy testing; an unwanted or coerced reproductive plan requires confidential, non-directive support rather than medication-only advice.
02Longitudinal psychiatric record review
Why
Verify diagnosis, episode severity, postpartum history, treatment response and relapse after changes rather than relying on a single snapshot.
Interpretation and limitations
Past response and consequences of withdrawal are central to the individual risk-benefit balance and may outweigh small differences between medicine datasets.
03Medication and interaction reconciliation
Why
Identify psychotropics, supplements, alcohol, nicotine and non-prescribed substances that could affect conception, pregnancy or mental stability.
Interpretation and limitations
Resolve discrepancies with the prescriber and pharmacy; St John's wort, interacting analgesics or apparently intermittent use can materially alter risk.
04Medicine-specific baseline monitoring
Why
Check weight, blood pressure and indicated renal, thyroid, hepatic, metabolic, prolactin, ECG or plasma-level measures before a planned change.
Interventions
01Shared formulationDefine illness and reproductive prioritiesFirst stepA person taking or likely to need psychotropic treatment may conceive in the foreseeable future.+
1PreferredClarify reproductive intention, contraception, fertility concerns, preferred terminology and who may be involved in decisions.
2Build the relapse and response history, including any postpartum episode and the effect of previous stopping or switching.
3Present continuing, changing and stopping options with both illness-related and medicine-related consequences, then document the person's informed priorities.
02Specialist optimisationChange treatment only with a safe bridgeA current medicine has important reproductive concerns, uncertain benefit or monitoring gaps.+
1Obtain perinatal psychiatric, pharmacy, obstetric and condition-specific advice, urgently if pregnancy has already occurred.
Medicines and treatment safety
Valproate may prevent bipolar relapse in selected patients, but its substantial fetal malformation and neurodevelopmental risks mean that alternative effective treatments must be actively evaluated before conception.
Valproate reproductive safety action
Do not initiate valproate in a new patient younger than 55 unless two specialists independently document the MHRA exception; do not prescribe it to a woman or girl able to have children unless Pregnancy Prevention Programme conditions are fulfilled. Any withdrawal is specialist-led and gradual rather than abrupt.
Exposure in pregnancy is associated with physical birth defects in about 11% of babies and neurodevelopmental disorders in up to 30–40% of children. Use the current annual risk materials. Counsel men about the precautionary three-month contraception and sperm-donation interval; the paternal causal association remains unconfirmed.
Lithium may be the most effective relapse-prevention treatment for an individual with bipolar disorder and can be considered when antipsychotic medication has not been effective and the personalised benefit outweighs reproductive risk.
Lithium preconception strategy
Use an individualised specialist regimen titrated to the person's established 12-hour plasma target; there is no universal reproductive dose. If a well pregnant person elects to stop, NICE describes gradual withdrawal over 4 weeks, while high relapse risk may justify supervised continuation or another specialist option.
Discuss uncertain-size fetal cardiac-malformation risk with first-trimester exposure, high transfer into breast milk and infant toxicity concern, renal and thyroid status, interactions, dehydration and the marked postnatal relapse risk. Continuing treatment requires a detailed pregnancy, labour and postnatal monitoring plan.
Targets, monitoring and follow-up
Review reproductive intention and contraception whenever diagnosis, treatment, relationships or life plans change; do not treat one preconception conversation as permanent consent.
Use a shared symptom and function baseline so emerging depression, activation, hypomania, psychosis or withdrawal during a change is identified early.
Track medicine-specific physical measures and plasma levels on the agreed schedule, with additional review after illness, dehydration, interacting treatment or major physiological change.
Confirm that referrals were received and that the person has direct contact details rather than assuming a copied letter created continuity.
At a positive pregnancy test, reconcile actual exposure dates and arrange prompt specialist review; never delay assessment while trying to calculate risk alone.
Revisit the written postpartum plan late in pregnancy, including prophylaxis, sleep, feeding, neonatal observation, crisis thresholds and who will make the first home or clinical contact.
Special situations
No option means no exposure
Untreated relapse, withdrawal, polypharmacy after destabilisation and loss of antenatal care are exposures too, so comparisons must include them explicitly.
Previous response is safety data
The medicine that reliably prevented severe episodes for this person supplies clinically important evidence that population pregnancy tables cannot replace.
Subtherapeutic is not protective
A dose too low to treat illness may expose the pregnancy while providing neither symptom control nor relapse prevention.
Contraception needs agency
A method is not effective in practice if it is unacceptable, inaccessible, contraindicated or controlled by an abusive partner.
Feeding is not a purity test
Breastfeeding benefits, medicine transfer, maternal sleep and relapse prevention belong in one respectful discussion, with support for the eventual choice.
Common pitfalls
01
Giving a generic 'all medicines are dangerous' warning that provokes abrupt discontinuation and avoidable relapse.
02
Changing a uniquely effective medicine solely because another drug has more published pregnancy data, without considering prior non-response.
03
Failing to apply the current MHRA valproate restrictions or to discuss the precautionary paternal-exposure advice.
04
Assuming that a specialist referral removes primary-care responsibility for repeats, contraception, monitoring and prompt communication.
05
Discussing fetal outcomes while omitting suicide, self-neglect, substance use, domestic abuse and the consequences of severe recurrence.
06
Writing 'review after birth' without a date, named clinician, medicine action or out-of-hours escalation route.
Practice
Two practice questions
Your practice draft is saved in this browser.
Question 1 of 2
0 correct
PsychiatryOriginal SBA
Unplanned valproate exposure
A 29-year-old taking valproate for bipolar disorder reports a positive home pregnancy test and asks whether to stop all tablets immediately. What is the safest next step?
Sources and review status4 sources · checked 27 Aug 2026 · clinical review pending+
Sources
Sources and review status
National guidance is shown before implementation-dependent detail. Typical adult dose examples remain subject to patient factors, contraindications and the live BNF or specialist protocol. Source check completed 27 Aug 2026; clinical approval remains outstanding.
NICE CG192 recommendationsPreconception referral, shared decisions, medicine principles and lithium planning.
Reconcile every prescribed, over-the-counter and non-prescribed substance. For each psychotropic document indication, current dose, adherence, benefit, adverse effects, prior alternatives, interactions, and monitoring. Explain absolute risks where reliable data exist and the uncertainty where they do not. Distinguish a class signal from product-specific evidence. Avoid presenting background congenital-malformation and pregnancy-loss risks as if every adverse outcome is caused by medication.
NICE advises specialist perinatal preconception counselling for current or past severe mental illness. When medicine is started in pregnancy or the postnatal period, choose the lowest-risk suitable option taking previous response into account, use the lowest effective dose without creating ineffective exposure, and use one drug where possible. These principles do not justify destabilising a well person through repeated switching. A planned gradual change needs observation long enough to establish whether the replacement works before conception whenever feasible.
Reproductive safety rules must be operational rather than a warning in the notes. Current MHRA measures state that valproate must not be initiated in a new patient younger than 55 unless two specialists independently document that no other effective or tolerated treatment exists or that reproductive risks do not apply. A woman or girl able to have children must meet Pregnancy Prevention Programme conditions. Paternal exposure also requires counselling: a possible neurodevelopmental signal has not established causation, but the MHRA advises condoms plus contraception used by the female partner during treatment and for three months afterwards, with specialist review when planning a family and no sperm donation during that interval.
Finish with a plan that can survive handovers. It should cover the intended medicine strategy, what not to stop without advice, contraception until the plan is ready, physical-health optimisation, specialist contacts, actions after a positive test, antenatal symptom monitoring, feeding preferences without pressure, sleep protection and postpartum prophylaxis. With consent, involve a partner or trusted supporter, but see the person alone where coercion or safety may be hidden.
Key points
Ask about pregnancy intentions at routine reviews; do not wait until a prescription is due or pregnancy has occurred.
Map previous episodes, admissions, suicide risk, postpartum illness, treatment response and the consequences of past withdrawal before recommending any change.
Refer a person with current or past severe mental illness who is planning pregnancy for specialist perinatal preconception counselling.
Compare four exposures explicitly: untreated or relapsing illness, continuing treatment, switching treatment, and stopping treatment.
Prefer an effective medicine with the lowest appropriate risk profile, the lowest effective rather than subtherapeutic dose, and monotherapy where clinically possible.
Never frame a positive pregnancy test as a reason to stop psychotropics suddenly; establish an urgent, named review route.
Valproate must not be prescribed to a woman or girl able to have children unless Pregnancy Prevention Programme conditions are met; initiation under age 55 also has current two-specialist restrictions.
If lithium is continued in pregnancy, plan level monitoring, hydration, hospital birth, labour management and postnatal relapse prevention before conception.
Record contraception, folate advice from the relevant obstetric or epilepsy pathway, smoking, alcohol, non-prescribed drugs, BMI, blood pressure and medicine-specific physical monitoring.
A history of postpartum psychosis, bipolar disorder or another severe mental illness predicts a need for specialist perinatal planning before conception and enhanced surveillance after birth.
Lithium toxicity symptoms, dehydration, vomiting, interacting medicines or renal deterioration need an urgent level and clinical review rather than a scheduled reproductive discussion.
Coercive control, reproductive coercion, domestic abuse or inability to use agreed contraception safely requires private enquiry and safeguarding support.
Interpretation and limitations
Choose tests from the actual medicine and comorbidity; normal results do not remove reproductive risk but provide a baseline for pregnancy adjustments.
05Physical and obstetric optimisation
Why
Review diabetes, hypertension, epilepsy, nutrition, immunisation, smoking, substance use and previous obstetric complications with the appropriate team.
Interpretation and limitations
Modifiable health factors may influence absolute pregnancy risk more than small medicine-associated increments and should be addressed without blame.
2When change is agreed, taper and cross-titrate according to the medicine, previous withdrawal sensitivity and relapse risk; avoid a medication-free gap imposed solely by fear.
3Allow enough observation before conception to confirm stability where possible, while maintaining acceptable contraception and a rapid route back to care.
03Continuity planPrepare for pregnancy and birthThe treatment strategy is agreed or conception may occur before optimisation is complete.+
1Record the medicine, dose, monitoring schedule, pregnancy-test action and named prescribers in primary, maternity and mental-health records.
2Set antenatal review frequency, early warning signs, crisis contacts, birth setting, neonatal observation and a feeding discussion that preserves choice.
3Specify postpartum prophylaxis or restart timing, sleep-protection arrangements and who will review mental state promptly after birth.
JurisdictionUnited Kingdom
Give the person a written plan naming who will prescribe, who to contact after a positive test, early relapse signs and the intended first postnatal review.