It has been suggested that a 30-50 % lithium dose reduction or lithium discontinuation 24-48 h before delivery could minimize neonatal complications. We investigated the maternal lithemia changes around delivery after a brief discontinuation, the placental transfer of lithium at delivery, and the association between neonatal lithemia at delivery and acute neonatal outcomes.A retrospective observational cohort study was conducted in a teaching hospital (November/2006-December/2018). Data was extracted from the medical records. We included psychopathologically stable women, with a singleton pregnancy, treated with lithium in late pregnancy, with at least one maternal and neonatal lithemia at delivery. Lithium was discontinued 12 h before a scheduled caesarean section or induction, or at admission day to hospital birth; and restarted 6-12 h post.Sixty-six mother-infant pairs were included, and 226 maternal and 66 neonatal lithemias were obtained. We found slight maternal lithemia fluctuations close to 0.20 mEq/L, and early postpartum relapse of 6 %. The mean (SD) umbilical cord/mother intrapartum lithemia ratio was 1.10 (0.17). Fifty-six percent of neonates presented transient acute complications. Neonatal hypotonia was the most frequent outcome (N = 15). Mean lithemia were 0.178 mEq/L higher in those with hypotonia than in those without (p = 0.028).It is a retrospective cohort of a moderate sample size of healthy uncomplicated pregnancies and results cannot be generalized to all pregnant treated with lithium.Lithium transfers completely across the placenta. A brief predelivery lithium discontinuation was associated with slight maternal lithemia fluctuations. Neonates exposed intrautero to lithium present frequent but transient acute effects.