Bipolar disorder in the postpartum period: the impact of a prenatal mood episode on maternal improvement at postpartum discharge after joint inpatient hospitalization.

Bipolar disorder (BD) is linked to a high risk of relapse in the year postpartum. The aim of this study was to search for an association of a mood episode during pregnancy with a lack of maternal improvement after a post-partum episode requiring joint hospitalization. In an observational, naturalist, and multicentric study, 261 women suffering from a BD and jointly hospitalized with their child in a Mother-Baby Unit (MBU) were assessed for risk factors associated with a lack of maternal improvement at discharge. A directed acyclic graph (DAG)-based approach was used to identify confounders to be included in a multiple regression model. In bivariate analyses, a lack of improvement (16.9%) was associated with pregnancy specificities (decompensation, psychotropic treatment, antipsychotics, and benzodiazepines intake), as well as maternal smoking during pregnancy and baby's neonatal hospitalization. In a multivariate analysis based on DAG, a lack of improvement was linked to psychiatric decompensation during pregnancy (OR = 3.31, 95%CI [1.55-7.35], p = 0.002), independently from maternal age, mother's maltreatment during childhood, low level of education, single status, low familial social support, and diagnosis of personality disorder. This study shows the critical importance of mental health during pregnancy in women with BD. Clinical screening and evaluation of the benefit/risk balance of psychotropics during pregnancy are essential.