Bipolar disorder in pregnancy and childbirth: a systematic review of outcomes

BACKGROUND: Bipolar Disorder (BD) is a mental disorder usually diagnosed between 18 and 30 years of age; this coincides with the period when many women experience pregnancy and childbirth. As specific problems have been reported in pregnancy and childbirth when the mother has BD, a systematic review was carried out to summarise the outcomes of pregnancy and childbirth, in mother and child, when the mother has BD diagnosed before pregnancy. METHODS: An a priori protocol was designed and a systematic search conducted in PubMed, CINAHL, Scopus, PsycINFO and Cochrane databases in March 2015. Studies of all designs were included if they involved women with a diagnosis of bipolar disorder prior to pregnancy, who were pregnant and/or followed up to one year postpartum. All stages of inclusion, quality assessment and data extraction were done by two people. All maternal or infant outcomes were examined, and narrative synthesis was used for most outcomes. Meta-analysis was used to achieve a combined prevalence for some outcomes and, where possible, case and control groups were combined and compared. RESULTS: The search identified 2809 papers. After screening and quality assessement (using the EPHPP and AMSTAR tools), nine papers were included. Adverse pregnancy outcomes such as gestational hypertension and antepartum haemorrhage occur more frequently in women with BD. They also have increased rates of induction of labour and caesarean section, and have an increased risk of mood disorders in the postnatal period. Women with BD are more likely to have babies that are severely small for gestational age (<2nd-3rd percentile), and it appears that those women not being treated with mood stabilisers in pregnancy might not have an increased risk of having a baby with congenital abnormalities. DISCUSSION: Due to heterogeneity of data, particularly the use of differing definitions of bipolar disorder, narrative synthesis was used for most outcomes, rather than a meta-analysis. CONCLUSIONS: It is evident that adverse outcomes are more common in women with BD and their babies. Large cohort studies examining fetal abnormality outcomes for women with BD who are not on mood stabilisers in pregnancy are required, as are studies on maternal-infant interaction. ELECTRONIC SUPPLEMENTARY MATERIAL: The online version of this article (doi:10.1186/s12884-016-1127-1) contains supplementary material, which is available to authorized users.