The recent editorial by Vintzileos and Ananth discussed the findings of Gregory et al, who found an association between the induction of labor and autism and postulated that exposure to exogenous oxytocin might be a risk factor for autism.

We would like to propose an alternative interpretation of the data by Gregory et al. Gregory et al found an association with “augmentation or induction of labor” (ie, failure to progress) and autism but did not demonstrate a causative link between the administration of oxytocin and the diagnosis of autism. Autism is a spectrum disorder with multiple etiologies and that appears to be the result of a complex mix of both genetic and environmental components. In fact, the data of Gregory et al suggest that multiple factors of the pregnancy, delivery, and maternal condition influence the risk of autism in the offspring.

Recent mechanistic research on animal models in our laboratory and genomic studies in humans has demonstrated that cellular calcium homeostasis plays a fundamental role in the susceptibility to or development of some forms of autism. Because some of the genes involved in calcium homeostasis are also expressed in the uterus as well as brain, abnormal cellular calcium homeostasis in the uterus can result in poorly developed excitation-contraction coupling. In such a scenario, dysfunctional labor, requiring the use of oxytocin or other forms of labor augmentation, may be an indicator of a genetic variant, which results in altered cellular calcium homeostasis, predisposing the mother to have poor excitation-contraction coupling in the uterus as well as predisposing the infant to autism. Most autism susceptibility genes have very low penetrance, so a mother with the susceptibility gene is most likely not to be diagnosed with autism.

In this interpretation, children have a predisposition to autism prior to the administration of oxytocin or the use of any other stimulant. Providers should not be deterred from using oxytocin as needed for the induction or augmentation of labor based on this interpretation. We therefore concur with the editorial that there is no indication that oxytocin is causative for autism. The raised risk in the study of Gregory et al was modest, but in the absence of more precise biomarkers, further data are clearly needed in this area. However, if a male infant is born in a pregnancy which requires augmentation, this could be used as another potential indicator (along with early milestones) to seek early intervention/behavioral therapy, which can significantly improve autism outcomes.