Ultrasound, neuronal migration, and autism

Autism is thought to be a disorder of neuronal migration with a genetic and intrauterine environmental etiology. Although it is not clear that the overall prevalence of childhood cognitive disorders has substantially increased, the diagnosis of autism is made far more frequently than in years past. This may represent a change in how we conceptualize cognitive disorders of childhood, or it may reflect a true change in disease pattern and/or frequency.

There is strong evidence for a polygenic predisposition to autism; a collection of genetic traits that favor neuronal migration disorders. The evidence is fairly strong that this is not related to mercury levels, thimerasol preservatives, or immunization practices. What else can one look for in the environment? What has changed in the past 20 years? The list is very long, which is why this study is mildly interesting (emphases and annotation mine):

Ultrasound Affects Embryonic Mouse Brain Development

.... The prolonged and frequent use of ultrasound on pregnant mice causes brain abnormalities in the developing mouse fetus, Yale School of Medicine researchers report August 7 in the Proceedings of the National Academy of Sciences.

'Proper migration of neurons during development is essential for normal development of the cerebral cortex and its function,' said Pasko Rakic, M.D., chair of the Department of Neurobiology and senior author of the study. 'We have observed that a small but significant number of neurons in the mouse embryonic brain do not migrate to their proper positions in the cerebral cortex following prolonged and frequent exposure to ultrasound.'

Neurons in mammals multiply early in fetal development and then migrate to their final destinations following an inside-to-outside sequence. The destination defines the neurons' connectivity and function. It has been reported earlier by others that abnormal cortical function may result when this process is grossly altered by genetic or environmental factors such as alcohol and drugs.

The study reported on August 7 is believed to be the first to look at the possible effect of ultrasound waves (USW) on neuronal migration in mice at a late stage of embryonic brain development, when the migratory pathways are the longest and may be most vulnerable. The Yale team injected more than 335 fetal mice at embryonic day 16 with special markers to track neuronal development. Exposure to USW for 30 minutes or longer [jf: adjusting for mouse lifespans and gestation, this would be hours to days in a human] caused a small but statistically significant number of neurons to remain scattered within inappropriate cortical layers and/or in the adjacent white matter.

'The magnitude of dispersion of labeled neurons was highly variable but increased with duration of exposure to ultrasound waves,' Rakic said. 'These findings suggested the desirability of further work in this area. We do not have any evidence ourselves that USW cause behavioral effects in mice or have any effect on the developing human brain.'

So mouse exposed to proportionately very long periods of ultrasound have some changes to neuronal migration, but it's unclear if there are any behavioral effects. It would be very interesting to repeat these studies in a mouse population that's genetically predisposed to develop "murine autism".

When I was doing obstetrical care, I and the thoughtful colleagues I worked with shared a common suspicion of the complete safety of ultrasound. There have always been small studies in animals suggesting worrisome effects, but it was absolutely clear that for any significant concern in pregancy ultrasound was indicated. I was less convinced of the case for routine dating ultrasounds, but I also recognized that there was no way we were going to avoid those given current practice and litigation risks. I was, and am, very opposed to recreational ultrasound. I knew enough of the history of medicine to worry about "perfectly safe" interventions.

Which is all to say that I'm predisposed to find this study interesting, and I may be giving it undeserved attention. We can expect it to be well investigated.

I suspect if there is an effect in humans, it may turn out to only be true for a genetically vulnerable fetus exposed at a very critical time for unusually long ultrasounds. I would also wonder if rather than creating a new cognitive disorder it biases the nature of neuronal migration dysfunction in the direction of autism rather than some other cognitive disorder pattern.

This study should discourage recreational ultrasound use. I would also be very slighly more reluctant to use ultrasound on a mother with a strong family history of autism or with a personal history of autism.

Update 9/7/06: There's a weird sequelae to this story.