Exogenous oxytocin administration during labor and autism spectrum disorder.

Background: Oxytocin is a neuropeptide hormone that plays a key role in social behavior, stress regulation, and mental health. Synthetic oxytocin administration is a common obstetrical practice, and importantly, previous research has suggested that intrapartum exposure may increase the risk of neurodevelopmental disorders, such as autism spectrum disorder.

Objective: This study aimed to examine the association between synthetic oxytocin exposure during labor and autism spectrum disorder diagnosis in the child.

Study Design: This population-based retrospective cohort study compared 2 cohorts of children: (1) all children born in British Columbia, Canada between April 1, 2000 and December 31, 2014 (n=414,336 births), and (2) all children delivered at Soroka University Medical Center in Be'er-Sheva, Israel between January 1, 2011 and December 31, 2019 (n=82,892 births). Nine different exposure groups were examined. Cox proportional hazards models were used to estimate crude and adjusted hazard ratios of autism spectrum disorder in both cohorts on the basis of induction and/or augmentation exposure status. To further control for confounding by indication, we conducted sensitivity analyses among a cohort of healthy, uncomplicated deliveries and among a group that was induced only for postdates. In addition, we stratified our analyses by infant sex to assess for potential sex differences.

Results: In the British Columbia cohort, 170,013 of 414,336 deliveries (41.0%) were not induced or augmented, 107,543 (26.0%) were exposed to oxytocin, and 136,780 (33.0%) were induced or augmented but not exposed to oxytocin. In the Israel cohort, 51,790 of 82,892 deliveries (62.5%) were not induced or augmented, 28,852 (34.8%) were exposed to oxytocin, and 2250 (2.7%) were induced or augmented but not exposed to oxytocin. On adjusting for covariates in the main analysis, significant associations were observed in the Israel cohort, including adjusted hazard ratios of 1.51 (95% confidence interval, 1.20-1.90) for oxytocin-augmented births and 2.18 (95% confidence interval, 1.32-3.57) for those induced by means other than oxytocin and not augmented. However, oxytocin induction was not significantly associated with autism spectrum disorder in the Israel cohort. In the Canadian cohort, there were no statistically significant adjusted hazard ratios. Further, no significant sex differences were observed in the fully adjusted models.

Conclusion: This study supports that induction of labor through oxytocin administration does not increase the risk of autism spectrum disorder in the child. Our international comparison of 2 countries with differences in clinical practice regarding oxytocin administration for induction and/or augmentation suggests that previous studies reporting a significant association were likely confounded by the underlying indication for the induction.