Cognitive function in metformin exposed children, born to mothers with PCOS – follow-up of an RCT
Journal article, Peer reviewed
MetadataShow full item record
Original versionBMC Pediatrics. 2020, 20 (60), . https://doi.org/10.1186/s12887-020-1960-2
Background Metformin is widely used in pregnancy to treat gestational diabetes mellitus and polycystic ovary syndrome (PCOS). Association between PCOS and developmental delay in offspring, and larger head circumference of metformin-exposed newborns has been reported. The objective of this study was to explore whether metformin exposure in utero had any effect on offspring cognitive function. Method The current study is a follow-up of two randomized, placebo-controlled studies which were conducted at 11 public hospitals in Norway In the baseline studies (conducted in 2000–2003, and 2005–2009), participants were randomized to metformin 1700 and 2000 mg/d or placebo from first trimester to delivery. There was no intervention in the current study. We invited parents of 292 children to give permission for their children to participate; 93 children were included (mean age 7.7 years). The follow-up study was conducted in 2014–2016. The Wechsler Preschool and Primary Scale of Intelligence version III and the Wechsler Intelligence Scale for Children version IV were applied for cognitive assessment. Androstenedione and testosterone were measured in maternal blood samples at four time-points in pregnancy. Results We found no difference in mean, full scale IQ in metformin (100.0 (SD 13.2)) vs. placebo-exposed (100.9 (SD 10.1)) children. There was an association between metformin exposure in utero and borderline intellectual function of children (full scale IQ between 70 and 85). Free testosterone index in gestational week 19, and androstenedione in gestational week 36 correlated positively to full scale IQ. Conclusions We found no evidence of long-term effect of metformin on average child cognitive function. The increase of borderline intellectual functioning in metformin-exposed children must be interpreted with caution due to small sample size.