Antipsychotic Use in Pregnancy for Schizophrenia
Two pregnant women with schizophrenia, one many years into stable remission and one newly diagnosed and still unstable, face the same relapse-versus-exposure tradeoff with very different real stakes behind it.
K.S., a 31-year-old woman, built her nursing career carefully after her diagnosis, choosing a specialty and a schedule that let her manage her illness alongside a demanding job, and describes her marriage, now six years in, as having formed and strengthened entirely during her stable years. She has been in sustained remission from schizophrenia for six years on quetiapine, works full-time as a nurse, and is now nine weeks pregnant with a strongly wanted first pregnancy.
She and her husband come to this visit having already read conflicting information online about antipsychotic use in pregnancy and ask directly whether she should stop the medication that has kept her stable for six years. Her history gives the team real information to work with: six years without a single relapse is a strong predictor that abrupt discontinuation carries genuine risk, and relapse during pregnancy itself carries documented risks to both maternal and fetal wellbeing, including impaired prenatal care engagement and obstetric complications associated with acute psychosis. Quetiapine's reproductive safety data, like that for most antipsychotics, is reassuring on major malformation risk without being large enough to rule out smaller effects with certainty, and neonatal adaptation syndrome — transient symptoms in the newborn following late-pregnancy exposure — is a real, generally self-limited consideration rather than a reason to avoid treatment outright.
I'd continue quetiapine through the pregnancy. Six years without relapse is a real track record, and stopping now carries a documented risk of destabilizing that record at exactly the point in her life when relapse would be most disruptive — to her, to the pregnancy, and to the infant she's carrying.
Her own six years matters, and I'd add the population data behind it so this isn't just one clinician's read of one patient. Kang and colleagues' nationwide Korean registry study, published this year in the British Journal of Psychiatry, followed over three thousand women with schizophrenia through pregnancy and found that continuing antipsychotic treatment cut the risk of postpartum psychiatric hospitalization roughly in half compared with discontinuing. I'd frame the reproductive safety data honestly with her too, in both directions: nothing in the available data raises a major malformation concern above baseline for quetiapine, and between that and the relapse numbers, the better-characterized danger here is discontinuation, not the medication.
Given both of you are pointing the same direction, the one thing left to get right is timing, not the decision itself. Neonatal adaptation syndrome — transient symptoms in the newborn tied to late-pregnancy exposure — is real but generally self-limited, and it's worth flagging to the obstetric and pediatric teams ahead of delivery specifically so it isn't mistaken for something more serious in the newborn, not raised now as a reason to reconsider anything.
Agreed: continue quetiapine unchanged through the pregnancy, with obstetric and pediatric teams informed ahead of delivery to anticipate and monitor for neonatal adaptation syndrome.
A.P., a 24-year-old woman, was in the middle of a graduate program when her first psychotic symptoms appeared eight months ago, and has been on medical leave from her studies since, an interruption she describes as the hardest part of an already difficult year. She was diagnosed with schizophrenia eight months ago and has not yet achieved stable symptom control — she has had two brief hospitalizations since diagnosis and is currently on her second antipsychotic trial, risperidone, started six weeks ago with only partial response so far.
She is eleven weeks pregnant, an unplanned pregnancy discovered shortly after this trial began, and unlike K.S. has no extended period of remission behind her to draw confidence from about what discontinuation, or even continuation at an unstable dose, would mean. The pivot here isn't the pregnancy itself, which both patients share, or the drug class, which is similar — it's that A.P.'s baseline illness control is still actively being established rather than already secured. Stopping or changing her regimen now risks losing ground in an illness that hasn't yet been brought under control at all, while continuing an only-partially-effective regimen through pregnancy means treating with a dose that may itself need adjustment for reasons unrelated to the pregnancy. Her situation doesn't have the six-year track record that made Case A's reasoning comparatively straightforward.
I don't think we can treat this the way we treated Case A. She doesn't have a stable baseline to protect — she has an illness that isn't controlled yet, and I want to actually get her stabilized rather than hold at a partially-effective dose out of caution about the pregnancy alone.
That's consistent with what I'd add from the obstetric side, not separate from it. Zhong and colleagues' study, published in BMC Pregnancy and Childbirth, examined over twenty million US delivery hospitalizations and found real, elevated rates of adverse obstetric and neonatal outcomes specifically among women with active psychosis at delivery — not psychosis in someone's history, active symptoms at the time of birth. For K.S. I cited continuation data because her illness was already controlled; for A.P. the relevant risk is different in kind, because achieving control itself is still the open question. Under-treating her out of pregnancy-related hesitancy doesn't protect the pregnancy, it just changes which risk we're accepting.
Then the actual decision left is speed, not direction. If we're increasing her risperidone dose or considering a switch to get real control, I'd want that decided now rather than drifting for weeks on a partial response — every additional week without stabilization is a week of both the illness risk you just named and fetal exposure to a regimen that isn't even working well yet. Unlike K.S., where the pharmacology question was already settled and the only open item was delivery-day coordination, here the dosing decision itself is the thing we can't afford to leave open-ended.
Agreed: increase risperidone to a standard therapeutic dose now rather than continue at a partial-response level, with close symptom monitoring and obstetric coordination established from this visit forward.
The regimen is held through pregnancy, following the same reasoning ultimately applied to Case A once real stability is achieved.
A switch to a different antipsychotic is considered directly, weighed against her still-early gestational age rather than deferred purely out of pregnancy-related caution.