TPS 771: Diet and lifestyle, Exhibition Hall, Ground floor, August 26, 2019, 3:00 PM - 4:30 PM
Background/Aim: Maternal nutrition during gestation may modify the effect of toxicants on child neurodevelopment. Caffeine, most notable for its stimulation of the central nervous system, may affect offspring neurodevelopment. However, little is known about the relationship between gestational caffeine intake and autistic behaviors. We aimed to assess the relation between daily maternal caffeine intake and children’s behavior traits associated with Autism Spectrum Disorders (ASD), measured by the Social Responsiveness Scale (SRS).
Methods: We harmonized data from two pregnancy cohorts, the Health Outcomes and Measures of the Environment (HOME) (n= 269), a general population cohort, and Early Autism Risk Longitudinal Investigation (EARLI) (n=120), an enriched-risk autism cohort of mothers who previously had child with autism. Caffeine-containing food and beverage intake was assessed twice during pregnancy via maternal report. Parent-reported SRS scores were ascertained when children were ages 3-8 years, with higher scores indicating greater ASD severity. We estimated changes in continuous SRS scores per interquartile range increase in maternal caffeine intake, adjusting for sociodemographic and perinatal factors.
Results: Median maternal caffeine intake during pregnancy was 18 mg/day (25th and 75th quartiles: 44, 56) in the pooled and individual cohorts (HOME 25th, 75th percentile: 44, 56; EARLI 25th, 75th percentile: 43, 55). In the pooled sample there was a positive association between caffeine intake and SRS scores (β: 1.1; 95%CI: 0.5, 2.2). However, the association was significantly different in each cohort (cohort x caffeine interaction p-value<0.05), where increasing caffeine intake was associated with increased SRS scores in EARLI (β: 2.7; 95%CI: 1.2, 4.2), but not in HOME (β: 0.3; 95%CI: -0.8, 1.3).
Conclusion: Higher levels of maternal caffeine intake during pregnancy were associated with increases in ASD-related traits among children in EARLI, but not in HOME. We speculate differences between these two cohorts could be due to underlying genetic susceptibilities.