Our article aimed to investigate one plausible environmental cause of autism spectrum disorders (ASD). Data support that a permanent shift in the balance of the sympathetic-to-parasympathetic tone as result of beta 2 adrenergic receptor overstimulation during critical periods of prenatal development is a biologically plausible mechanism whereby beta 2 adrenergic agonists (B2AA) can induce functional and behavioral teratogenesis. The animal model sited would predict that such a shift in autonomic tone could be demonstrated by examining cardiovascular function. Such a shift is seen in children with ASD and in adolescents who were exposed to B2AA prenatally but delivered at term. It predicts the increase risks for ASD seen with more active beta 2 adrenergic receptor gene variants, which explains one of the genetic predispositions for ASD.

Premature delivery is a significant risk factor for neurobehavioral problems. Pitzer et al, examined B2AA exposure and adverse neuropsychiatric outcome comparing exposed individuals to gestational age–matched controls. Individuals exposed delivering prematurely did not differ from their controls, but exposed individuals delivering at term showed increased adverse neuropsychiatric outcomes. The exposed individuals delivered at term had a median exposure to intravenous B2AA of 10 days and were significantly more likely to be exposed to more oral B2AA therapy. The earlier gestational age at onset of tocolysis may explain why the exposed individuals delivering at term had the worst outcome. Pitzer et al concluded that tocolytic treatment may influence child development adversely. The alternate hypothesis that preterm contraction without preterm delivery adversely influenced child neurodevelopment requires an alternate biologically plausible mechanism, and is inconsistent with the findings that exposed individuals delivered at term have the worst outcome.

Duration of exposure is critical to our hypothesis. If sufficient exposure time does not occur, we would not expect to see an effect. This is why short-term exposure was considered nonexposed in the study by Connors et al. In our discussion of exposure studies that did not support an association, we emphasized that not only does the exposure have to be at the critical period of prenatal development, it must also be of sufficient duration. Short-term exposure for up to 48 hours to allow administration of corticosteroids for fetal lung maturation may be reasonable if no alternative tocolytics are available. However, the longer use of tocolytic agents has not been shown to be effective and may put the fetus at risk for long-term neurophysiologic harm including ASD.