Women with bipolar disorder are at a higher risk for relapse during pregnancy and that the risk is even higher during the postpartum period.  In addition, women with bipolar disorder are more vulnerable to postpartum psychosis.  It is for these reasons that we typically recommend that women with bipolar disorder maintenance treatment with a mood stabilizer during the postpartum period in order to decrease their risk for relapse during this vulnerable time.

What we must keep in mind is that all drugs commonly used to treat bipolar disorder are excreted into the breast milk, and the risk of toxicity for breastfed infants varies across different medications.  When advising women with bipolar disorder, these risks should be carefully weighed against the increased risk of recurrent psychiatric illness if the women’s medication is changed or discontinued.

While we have some information on the use of more traditional mood stabilizers like lithium and lamotrigine in breastfeeding women and can provide specific information regarding risk, we are seeing an increasing number of women taking the newer or atypical antipsychotics as part of their maintenance regimen.  For these agents, the information regarding breastfeeding is quite limited.  Most of the information we have comes from case reports and small case series.

The table below is adapted from a very thorough review article published last month by Pacchiarotti and colleagues.  At this point, we have the most data on olanzapine (Zyprexa) with a total of 40 exposures.  Looking at the data as a whole, it looks as if the risk of adverse events in olanzapine-exposed infants is low.  One of the 40 infants had icterus and sedation; however, these symptoms persisted even after bottle feeding was initiated so it is not clear that these symptoms were caused by exposure to olanzapine.

While these data regarding the use of olanzapine in breastfeeding women are reassuring, we have much less information on exposures to other atypical agents: aripiprazole (2 infants), quetiapine (9 infants), risperidone (4 exposures).  Most of the reports described no adverse events; however, two of the 9 infants exposed to quetiapine (both from the same study) showed mild developmental delays.  There is no information on lurasidone (Latuda).   Some might conclude that quetiapine may carry more risk than other atypical antipsychotics, but when the numbers are so small, this might be a chance finding.  In addition, antipsychotic agents are often combined with other medications which may affect outcomes.  


Drug Reference Study Design Sample and Maternal Dose Main Outcomes
Olanzapine (OLZ)

Total of 40 exposures

Gilad et al., 2011 Prospective controlled observational study 22 mothers breastfeeding while taking OLZ, 15 mothers taking OLZ but did not breastfeed and 51 breastfeeding mothers with no OLZ Data on lactation and outcome at the age of 1–2 years were obtained.

The rate of adverse outcomes in OLZ-exposed breastfed infants did not differ from those of the control groups.

Whitworth et al., 2010 Case report 1 mother-infant pair

(5–15 mg/day)

Relatively high plasma level in the infant aged four months (Maternal dosage 15 mg, infant?s OLZ concentration 11 ng/ml) probably related to an immature hepatic metabolism.  OLZ plasma levels decreased to very low or even undetectable during the following four months
Lutz et al., 2008 Case report 1 mother-infant pair

(5 mg/day)

No adverse effects were noticed in the infant.  Levels in the infant undetectable.
Ambresin et al., 2004 Case report 1 mother-infant pair

(20 mg/day)

The relative amount of drug transferred to the infant was about 4% of maternal weight-adjusted dose.
Gardiner et al., 2003 Case series 7 mother-infant pairs (5–20 mg/day) OLZ was not detected in the plasma of the 6 infants.  No adverse effects were found in the infants.  
Croke et al., 2002 Case series 5 Mother-infant pairs (2.5–10 mg/day) No adverse effects were found in the infants.
Goldstein et al., 2000 Case series 2 Mother-infant pairs (5–10 mg/day) In one infant, the exposure to OLZ at 5 mg was associated with icterus and sedation; although bottle-fed was initiated, icterus and sedation continued.

No adverse events in the second infant.

Kirchheiner et al., 2000 Case report 1 Mother-infant pair

(10 mg/day)

No adverse events reported.
Aripiprazole (ARP)

Total of 2 exposures

Watanabe et al., 2011 Case report 1 mother-infant pair

(18 mg/day)

No adverse events reported
Lutz et al., 2010 Case report 1 mother-infant pair

(15 mg/day)

Neither ARP nor its metabolite was detected in the infant.  No adverse events reported.
Quetiapine (QTP)

Total of 9 exposures

Rampono et al., 2007 Case report 1 mother-infant pair (400 mg/day) No adverse effects were noted in the infant
Misri et al., 2006 Case series 6 mother-infant pairs (25–400 mg/day) QTP levels in breast-milk ranged from not detected to 32–264 nmol/L depending on QTP doses.  4 out of the 6 babies scored as being within normal limits on the Bayley Scales of Infant Development, whereas 2 showed a mild developmental delay.
Gentile, 2006 Case report 1 mother-infant pair (400 mg/day) No adverse events reported.
Lee et al., 2004 Case report 1 mother-infant pair (200 mg/day) No adverse effects reported.
Risperidone (RIS)

Total of 4 exposures

Aichhorn et al., 2005 Case report 1 mother-infant pair (2–3 mg/day) No adverse effects reported.
Ilett et al., 2004 Case series 3 mothers and 2 infants (1.5–4 mg/day) No adverse effects reported.
Hill et al., 2000 Case report 1 mother

(6 mg/day)

No adverse effects reported.


So what would we typically recommend in a woman who is taking one of these medications (other than olanzapine) and would like to breastfeed?  There is no easy answer here.  With sparse information to guide these decisions, the choices a patient makes may be driven more by personal factors — for example, her commitment to breastfeeding or the severity of her illness.  As clinicians, we must provide the little information we do have to patients while at the same time emphasizing the extreme vulnerability to recurrent illness women with bipolar disorder face during the postpartum period.  

This is also an opportunity to support the decision to forgo breastfeeding.  Most women feel considerable pressure to breastfeed, but for many women with bipolar disorder, not breastfeeding might be the best option.  This allows the woman to maximize her pharmacologic treatment without worrying about exposing the nursing infant to medications.  Bottle feeding also makes it easier to for others to participate in caring for the newborn and may also help to minimize sleep deprivation, which can be a potent trigger for women with bipolar illness.  Of greatest importance here is doing everything we can do to keep the mother healthy.  While breastfeeding is beneficial for the mother and the newborn, all of these benefits are erased if the mother becomes ill during this vulnerable time.


Ruta Nonacs, MD PhD  


Pacchiarotti I, León-Caballero J, Murru A, Verdolini N, et al. Mood stabilizers and antipsychotics during breastfeeding: Focus on bipolar disorder.  Eur Neuropsychopharmacol: 2016 Oct; 26(10):1562-78. Epub 2016 Aug 24. Review.

Related Posts