Inflammation in pregnancy and child neurodevelopment: A trio polygenic score and Mendelian randomization study in the Norwegian Mother, Father and Child Cohort Study

Background Inflammation in pregnancy induced by chronic disease, infection, or environmental exposures has been associated with neurodevelopmental conditions in observational study designs. Objective To investigate causal effects of pregnancy inflammation on offspring neurodevelopment in the Norwegian Mother, Father and Child Cohort Study (MoBa) using a genetically informed design. Methods In pre-registered analyses, we tested potential causal effects of increased C-reactive protein (CRP), interleukin-6 (IL-6), and glycoprotein acetyls (GlycA) in pregnancy on neurodevelopmental outcomes using trio polygenic score (PGS) and intergenerational Mendelian randomization (MR) analyses. PGS and MR genetic instruments were used to predict neurodevelopmental outcomes in the MoBa cohort, which included mother-reported neurodevelopmental traits registered from age 3 to 8 years and diagnoses of ADHD and autism. In both the trio PGS and intergenerational MR analyses, simultaneous inclusion of mothers’, fathers’, and children’s genotypes allowed for effects consistent with the hypothesized causal pathway (maternal inflammation) to be estimated independent of familial confounding factors. Results Validation analyses of genetic instruments only showed strong predictive value for CRP. Trio PGS in 41,531 complete trios did not reveal maternal indirect effects of genetic liability to higher CRP on offspring neurodevelopmental outcomes. The results of the trio MR also failed to support a causal link between maternal inflammation, proxied by higher CRP, during pregnancy and offspring neurodevelopment. Conclusion This study provided no evidence consistent with causal effects of higher CRP in pregnancy on offspring neurodevelopment. This may imply that observational links are driven by genetic confounding, though other factors – such as selective participation in MoBa – cannot be ruled out as alternative explanations for our null findings.