AIUM Bioeffects Committee’s Comments on the JAMA Pediatrics Article on Prenatal Ultrasonography and Autism Spectrum Disorder
February 25, 2018
AIUM Bioeffects Committee’s Comments on the JAMA Pediatrics Article on Prenatal Ultrasonography and Autism Spectrum Disorder

A recent article published in JAMA Pediatrics reports on a retrospective study of prenatal ultrasound scans and the later appearance of autism spectrum disorder (ASD). The study was based on a comparison of data taken from past medical records of a group of patients with ASD compared to 2 control groups: one comprising individuals with developmental delays and a second comprising individuals with typical development. The study analyzed patient data and data from images recorded during each ultrasound examination for possible associations of ASD with many different parameters.

Importantly, the study found no statistically significant differences in the total number of prenatal ultrasound scans or the total duration of ultrasound scans between patients with ASD and either of the control groups. The study also found no differences between the ASD group and control groups in the 2 acoustic output metrics related to the ultrasound energy of exposure (ie, the mechanical index and the thermal index). The authors describe a parameter termed the “time-weighted mean depth of ultrasound penetration,” which is not a parameter used in evaluations of ultrasound exposures or bioeffects, nor has it been validated in previous studies. The study reports that the mean depth of the ultrasound penetration parameter in the ASD group was slightly greater than in the typical-development group during the first and second trimesters and also than in the control developmental-delay group in the first trimester.
However, the authors’ mean depth of ultrasound penetration parameter is not closely related to the intensity of ultrasound exposure of the fetus. Furthermore, the differences in their mean depth of ultrasound penetration parameter were too small to be clinically significant.

It should also be noted that the article incorrectly states that ultrasound is “minimally regulated” when, in fact, diagnostic ultrasound devices in the United States are strictly regulated by the Food and Drug Administration to ensure that acoustic outputs are within accepted levels. Overall, the reported lack of an association of ASD with the total number of ultrasound scans, duration of scans, and either acoustic output metric (mechanical or thermal index) is consistent with the long history of safety of ultrasound for medical diagnosis. The AIUM continues to promote the use of diagnostic ultrasound as an important clinical tool to be performed with a clear medical indication, by appropriately trained professionals with respect to acoustic output and the time of the examination (as low as reasonably achievable [ALARA] principle).

© American Institute of Ultrasound in Medicine | 14750 Sweitzer Lane, Suite 100 · Laurel, MD 20707 | Phone: 301-498-4100