Ultrasounds and autism: Another media scare!

The Pediatric Insider

© 2013 Roy Benaroch, MD

Emily wrote in about an article about prenatal ultrasounds and autism: “I saw this on The Daily Beast (Newsweek) today. Is the media trying to freak us expecting couples out or what? How big of a question is this in scientific circles or is this just sensational stuff? Sometimes I think there should be studies about how the internet causes anxiety disorders!”

A good question… and another post that I’m going to put under my new category, “Guilt Free Parenting.”

The Daily Beast headline and tag reads: “Are Ultrasounds Causing Autism in Unborn Babies?” and “Scientists are uncovering disturbing evidence that those sneak peaks at baby could damage a developing brain.”

I wonder how all of this accumulated hysterical, sky-is-falling reporting is damaging adult brains. As Emily said, the internet seems to cause anxiety disorders, and it’s articles like this that get everyone worked up.

First: autism is a complex neurodevelopmental disorder that causes problems with communication, social interactions, and repetitive behaviors, starting very early in childhood. The best current evidence shows that whatever’s gone wrong, it’s going wrong very early in life, possibly even before babies are born. There are strong genetic influences, and there is still a lot we don’t know about what causes autism and how to best treat it.

Autism is especially scary because we’re hearing so much about it. It may be as common as 1 in 50 boys, and it seems like the incidence is rising dramatically—but a lot of that apparent increase is because of an ever-widening definition of autism, combined with efforts at early detection and what’s called “diagnostic substitution” (kids who would have once been diagnosed with mental retardation or other disorders are now diagnosed with autism.) But whether the true rates of autism are increasing or not, it’s certainly a huge problem for families and communities, and increased awareness, early detection,  and early treatment are urgently needed.

Because the causes of autism remain unclear, there’s a lot of speculation about what might be going on. If you Google “causes of autism”—and please, please don’t do that—you’ll find all kinds of speculation about toxins and parenting styles and government conspiracies and evil doctors who are eager to profit by harming children. You’ll also find a few tantalizing, genuine leads, things that might actually make sense. Those are what legitimate researchers are trying to study.

So what about ultrasounds? It is true that prenatal ultrasounds are being used more commonly, and the rise in their use generally follows the trend in the observed rise in autism over the last 20 years. But that observation, alone, doesn’t really show that one thing causes the other. After all, over the last 20 years we’ve also seen a dramatic increase in cell phones, cable television, frozen food, Oprah, personal computers, and Starbucks; we’ve also seen a decline in the prevalence of cursive handwriting, licking stamps, and the Sizzler Steakhouse chain. Are any of these connected to autism, or to each other? Maybe. Maybe not.

Is there some basic science about ultrasounds that makes a connection with autism plausible? Again, maybe-sort-of. Research on the effect of ultrasounds on developing mouse brains has shown a difference in the way brain cells move and migrate—but those studies looked at ultrasound exposures for many hours a day, and mouse brains develop much more quickly than ours do. We’ve also got much bigger brains, and much more tissue between our babies and an ultrasound probe than mice. Lab research on the effects of ultrasound on moving cells or bubbles is similarly unconvincing—something to think about, but a huge leap from there to “ultrasounds cause autism.”

Several studies have looked for any direct biological effect of fetal ultrasounds on human children. A 1978 report looked at about 1000 infants of mother who received amniocentesis, ultrasound, or neither—it found no developmental effects of ultrasounds. In 1984, a different group looked at 425 children, finding no biologically significant differences among those who were and were not exposed to diagnostic ultrasounds. There are also many studies looking at potential ill-effects of ultrasound technology for diagnostic use in babies, children, and adults—there are none caused by the ultrasound itself. For those of you who’d like even more detail, an excellent review of these and many other studies about ultrasounds and autism is here (unfortunately the full article is behind a paywall.)

So what’s The Daily Beast talking about—are scientists, as they say, uncovering “disturbing evidence”?

This is what’s actually reported in the article, in order of appearance:

  1. References to a study showing that among low-risk pregnancies, routine ultrasounds don’t improve outcomes. This is true. It’s irrelevant to the title or thesis of the article, but it’s true. Media lesson #1: if you don’t have a study to prove your point, talk about a different study that says something else entirely.
  2. Ultrasounds drive up the cost of care. Again, correct. Again, irrelevant. See point #1.
  3. Women who undergo frequent ultrasounds are more likely to have a pregnancy where the baby is found to have growth restriction. Well, this is true. It’s also true that if you look outside you’re more likely to know if it is raining. Fetal growth restriction is diagnosed by ultrasound. If you don’t look, you don’t know it’s happened. But looking outside doesn’t make it rain; and looking at an unborn baby with an ultrasound doesn’t cause the baby to be small. And, in any case, this is again irrelevant to autism. See point #1.
  4. The author of the article has written a book in part about her assertion that ultrasounds are to blame for what she calls “an astronomic rise in neurological disorders among America’s children.”
  5. The mice studies I referenced before—those come up now, several paragraphs in, the first even remotely relevant material. The lesson here: if you are a mouse, do not get seven hours of ultrasounds a day.
  6. A neurologist named Manuel Casanova shares the author’s concerns, and says he and colleagues have been testing the ultrasound-autism hypothesis for three years. However, and this is important: after several technical paragraphs about his ideas, he’s uncovered zero evidence to support this claim. What he’s saying are generalities about brain development that are true, and he’s juxtaposing this against information about ultrasounds and information about autism, but he doesn’t in any way refer to any of his or anyone else’s actual research establishing a connection. These are ideas. Ideas are not evidence.

That’s it. The whole article.

Now, there is a reasonable point that I will agree with—prenatal ultrasounds do not necessarily improve the health of babies, and they’re often unnecessary. Vanity ultrasounds to take 3-D pictures of unborn babies use far more energy, and it’s not implausible that there could be ill-effects—things like this are not medical uses, and ought to be discouraged until there is definitive proof of their safety.

That being said, there is no evidence for a link between ultrasounds and autism, none whatsoever. It’s not being uncovered. It’s just not there.

Explore posts in the same categories: Guilt Free Parenting, In the news, Medical problems, The Media Blows It Again

Tags: , , ,

You can comment below, or link to this permanent URL from your own site.

13 Comments on “Ultrasounds and autism: Another media scare!”

  1. Stu Says:

    Thanks for your continued info, Dr. Roy. This type of thing reminds me of the (joking) rule I had for my wife: No internet during pregnancy!


  2. Thanks, Dr. Roy. I’m always amazed how the mainstream media can distort these issues. You would think Newsweek would be more dependable. I enjoyed reading your thoughts. It helps me get better at spotting the holes in their analysis.


  3. In 1991, the FDA increased allowable intensity of prenatal ultrasound 8-fold. The safety data cited is for much lower exposure. The probable culprit with ultrasound is overexposure through device or operator error. A Swedish survey of hundreds of hospital echocardiograms found 40% to be transmitting at different intensities than the control panel indicated. The device have the same basic components as prenatel ultrasounds and are made by the same companies. As to operator error – surveys on knowledge show poor understanding, and medical training is not required in most states.


  4. Dr. Roy Says:

    Parrish has a site that’s dedicated to promulgating an ultrasound-autism link. If you click through, there’s well-documented page of references with summary statements. I appreciate all of the work that had to have gone into collecting this information into one place.

    However, most of the references do not support the conclusion that ultrasounds cause autism, not by a long shot. Many document that ultrasounds are overused (agree), many document that the intensity of the transmissions may not be accurately reflected in the labeling (agree), some document that training may be inconsistent (perhaps); one refers to a lack of warrantees on resold ultrasound machines (OK, now that’s just padding the list.) None even come close to showing that there is an important clinical outcome even close to autism associated with prenatal ultrasound exposure.


  5. lilady Says:

    Hmmm, about those “mouse studies” and autism…from Emily Willingham a Forbes science blogger (and the parent of a child diagnosed with an ASD).


    Mouse Behavior From Autism Studies Not Reproducible

    “…Using the mouse condition to examine the human condition serves a purpose–we can’t very well, for example, purposely knock out a gene in a large group of humans, breed them, and then examine what they do in behavior testing. But mice also are not humans and not even that closely related to us. In a previous post of mine, in writing about best practices for authoring articles about autism, I noted:

    If the study in question is about mice, never talk about how the results will lead to a therapy or a cure or write about the mice as though somehow, they are just tiny humans with tails. Mice have misled us before. They are only a way to model what might happen in a mammal sorta kinda related to us. They are not Us, otherwise we’d live in tiny, crowded places, having 10 children at once and ignoring them when they grow fur, and this autism thing wouldn’t be an issue…”


  6. Adrienne Says:

    Thank you for quelling our fears Doc! I did want to give some input from a patient standpoint regarding the 3D ultrasounds.

    For my pregnancy (and every other pregnant woman in Japan going to a Birthing Clinic), they gave me a traditional and 3D ultrasound at every single appointment after 16 weeks. Now, I’m not sure how long this has been the standard of care here in Japan. But, I think that if there were a link in ultrasounds or 3D ultrasounds, it would have been uncovered in this nation. What do you think?


  7. Hi, I would like to mention something about all of the epidemiological data to date supporting the safety of prenatal sonography.

    The WHO admits this in: http://onlinelibrary.wiley.com/doi/10.1002/uog.6328/full

    We are currently unable to effectively quantify dose of ultrasound exposure. The number of scans is a terribly inaccurate representation of dose, which is the only measure of dose that population studies have gone by so far. This complicates things and it seems to invalidate nearly all epidemiology to date supporting the safety of fetal ultrasonography.

    To explain with example:
    Two women, A and B, are scheduled for ultrasounds. Woman A receives one scan while Woman B receives two. At first, it seems like Woman B receives more exposure. However, what if I add that Woman A’s scan lasted an hour while Woman B’s both lasted 5 minutes? That changes the game a lot.

    It also matters where the ultrasound passes through. I am not likely to get brain damage from my elbow getting exposed to excessive ultrasound. Since none of these variables are taken into consideration, the data is very blurry and adverse health outcomes would not show up unless it was so common that a majority of exposures caused the effect in the first place.

    Likely, if there are adverse effects (such as aggravating autism symptoms) they are only going to occur when specific criteria are met. For example, overexposure of a specific part of the brain, or hormone regulatory control center.

    While epidemiology does not support that ultrasound and autism are linked, there are feasible mechanisms that serve as a direct links between ultrasound exposure and symptoms of autism. More research is needed – it should not be something that is just tossed off as a farce with the wellbeings of so many involved.


  8. A Yale Medical School study of prenatal ultrasound exposure using primates is scheduled for publication in 2014. It is the same lab that found the link with mice as the test subjects.


  9. Janice Hedges Says:

    There needs to be a control group (women in low-risk pregnancies who choose to have no ultrasound exposure). When the children in the no-ultrasound group are 4-5 years old, they can be evaluated/screened for autism spectrum disorder, preschool/daycare staff can be interviewed and parents can be interviewed. If the autism rate in that group of children is the same as the autism rate was 40 years ago, you have a smoking gun-indication of what the “something in the environment” is that has increased the rate so profoundly.


  10. Dr. Roy Says:

    “If the autism rate in that group of children is the same as the autism rate was 40 years ago…”

    That would be incorrect. The diagnostic criteria for ASD are so different from what was used 40 years ago that such a conclusion would not be valid. You’d need to compare groups simultaneously, not relying on historical controls.

    Also: women who choose no ultrasound may be have the same health risks and circumstances as women who wish to undergo ultrasounds. That needs to be randomized to get reliable data, not selected.

    Science can be hard, but that’s why it works.


  11. I did my own survey research and found those children who were exposed ultrasound in utero had a 10 times greater odds for autism diagnosis. That is a far cry from coincidental and I challenge others to do quality research with a “control group” that is unexposed. Commonsense tells you when you point ultrasound toward a developing embyro/baby that creates unknown amount of heat (known teratogen) and that creates changes in the cell membranes through mechanical action and sheering force has great potential to mess up the delicate process of cell migration and differentiation. The increase in use perfectly correlates with the increase in autism and other birth defects on the rise. Using studies from the 70’s to conclude ultrasound is safe is like using the model T to today’s rockets to the outer space. The machines today are many times more powerful.


  12. Dr. Roy Says:

    Dr. Gilliland, I couldn’t find your research on pubmed, though I do see you’re saying you’re publishing a book on this topic soon.

    Whether your 10 x claim (or the 9 x claim on your website) is coincidence or not depends on actual statistics. You’d also want to control for the huge confounder: babies in trouble, for any reason, get more ultrasounds (including those with suspected chromosomal anomalies, those with older parents, those not growing well, those experiencing preterm labor, etc, etc), and all of those factors are known to increase the risk of autism (and birth defects, as well.) You’d need to separate out the cause and effect for your claim to be meaningful. It may not be the ultrasounds causing the mischief, but the medical reasons that prompted the ultrasounds.

    “The increase in use perfectly correlates with the increase in autism and other birth defects on the rise.” Yeah. If you’re going to go the correlation = causation route, it also correlates with cell phones, the success of Oprah Winfrey, organic food sales, and all sorts of other irrelevant things. And although the rates of some birth defects are rising, the rates of others are dropping. No fair picking and choosing to report and believe only evidence that supports your theory.


    Best of luck with your publication and future research.


  13. We need to ask why the people in charge of autism research and its funds continue to avoid studying this obvious suspect even after HHS-funded research shows the influence is during gestation? “Follow the money” could be the route to this answer.


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: