Seth's Blog

A new paper (“Are Prenatal Ultrasound Scans Associated with the Autism Phenotype? Follow-up of a Randomised Controlled Trial” by Yonit K. Stoch, Cori J. Williams, Joanna Granich, Anna M. Hunt, Lou I. Landau, John P. Newnham and Andrew J. O. Whitehouse) takes another look at the results of a randomized trial started in 1989. Half the mothers were given one ultrasound during their pregnancy, the rest got five. This study gave a questionnaire sensitive to autism-like traits to the now-grown-up children. This is potentially more sensitive than binary (yes/no) assessment because the questionnaire has about 50 questions.

Here is the entire abstract:

An existing randomised controlled trial was used to investigate whether multiple ultrasound scans may be associated with the autism phenotype. From 2,834 single pregnancies, 1,415 were selected at random to receive ultrasound imaging and continuous wave Doppler flow studies at five points throughout pregnancy (Intensive) and 1,419 to receive a single imaging scan at 18 weeks (Regular), with further scans only as indicated on clinical grounds. There was no significant difference in the rate of Autism Spectrum Disorder between the Regular (9/1,125, 0.8 %) and Intensive (7/1,167, 0.6 %) groups, nor a difference between groups in the level of autistic-like traits in early adulthood. There is no clear link between the frequency and timing of prenatal ultrasound scans and the autism phenotype.

Parrish Hirasaki pointed out to me that when the study was done the intensity of ultrasounds was eight-fold less than now. Here’s what the paper says about this:

Technological advancements over the past two decades have led to considerable improvements in ultrasonographic capabilities, with corresponding increases in acoustic output.

Did you read that and realize the intensities have increased by a factor of eight? Neither did I. Such a big difference in intensity means the results are not serious evidence — contrary to what the abstract implies — against the idea that sonograms are now causing autism.

More Asked about the failure to make clear the difference in intensity, the corresponding author, Andrew Whitehouse, replied:

The parameters of the ultrasound scans were reported clearly in the Methods section, and we refer to several other articles that include a description of the USS administered in this cohort. We are also very clear in the Discussion when we state that “the current study was not designed to investigate the use of [modern] instruments”.

I disagree. So what if the parameters of the ultrasound scans were reported clearly in the Methods section? That’s not the issue. The issue is failure to make clear the huge intensity difference between the ultrasounds they studied and modern ultrasounds. Saying that “the current study was not designed to investigate the use of [modern] instruments” does not make clear at all (much less “very clear”) how much modern ultrasounds differ from the ultrasounds actually studied. Any study can put caveats like that at the end. Yet few studies are as irrelevant as this one to the question they claim (in the title, abstract and introduction) to help answer.

An article in the New York Times describes research that supposedly linked a rare gene mutation to autism:

Dr. Matthew W. State, a professor of genetics and child psychiatry at Yale, led a team that looked for de novo mutations [= mutations that are not in the parents] in 200 people who had been given an autism diagnosis, as well as in parents and siblings who showed no signs of the disorder. The team found that two unrelated children with autism in the study had de novo mutations in the same gene — and nothing similar in those without a diagnosis.

“That is like throwing a dart at a dart board with 21,000 spots and hitting the same one twice,” Dr. State said. “The chances that this gene is related to autism risk is something like 99.9999 percent.”

It is like throwing 200 darts at a dart board with 21,000 spots (the number of genes) and hitting the same one twice. (Each person has about 1 de novo mutation.) What are the odds of that? If all spots are equally likely to be hit, then the probability is about 0.6. More likely than not. (Dr. State seems to think it is extremely unlikely.) This is a variation on the birthday paradox. If there are 23 people in a room, it is 50/50 that two of them will share a birthday.

When Dr. State says, “The chances that this gene is related to autism risk is something like 99.9999 percent,” he is making an elementary mistake. He has taken a very low p value (maybe 0.000001) from a statistical test to indicate the likelihood that the null hypothesis (no association with autism) is true. P values indicate strength of evidence, not probability of truth.

One way to look at the evidence is that there is a group of 200 people (with an autism diagnosis) among whom two have a certain mutation and another group of about 600 people (their parents and siblings) none of whom have that mutation. If two instances of the mutation were randomly distributed among 800 people what are the odds that both instances would be in any pre-defined group of 200 of the 800 people (defined, say, by the letters in their first name)? The chance of this happening is 1/16. Not strong evidence of an association between the mutation and the actual pre-defined group (autism diagnosis).

Another study published at the same time found an link between autism and a mutation in the same gene identified by Dr. State’s group but again the association was weak. It may be a more subtle example of the birthday paradox: If twenty groups of genetics researchers are looking for a gene linked to autism, what are the odds that two of them will happen upon the same gene by chance?

If the gene with the de novo mutations is actually linked to autism, then we will have insight into the cause of 1% of the 200 autism cases Dr. Smart’s group studied. When genetics researchers try so hard and come up with so little, it increases my belief that the main causes of autism are environmental.

Thanks to Bryan Castañeda.

A female American engineer named Parrish Hirasaki has started a website devoted to the idea that prenatal ultrasound is a major cause of autism. It includes a long list of supporting research. In spite of all this research, “as of spring 2011, analyzing the ultrasound results was not among the 23 hypotheses [about the cause of autism] being tested by the National Children’s Survey”, says the website.

I first heard this idea from Caroline Rodgers, a science writer, and have blogged about it several times. Nevertheless, the website’s home page taught me several things I didn’t know:

1. “A recent major study of twins supports earlier studies in concluding that environmental exposures during or shortly after gestation cause a majority of autism cases.” This is inconsistent with many other explanations (e.g., vaccine, genetic).

2. “In a recent study, autistic children had reduced connectivity between the two sides of the brain.” Which supports the idea that neural misdevelopment is a cause of autism.

3. “Twenty years ago, the FDA increased the allowable intensity of prenatal ultrasound 8-fold to improve images. Autism rates have risen dramatically since that time.”

4. There is “widespread misunderstanding among ultrasound operators of the safety guidelines.”

5. “A fast-growing commercial business is keepsake ultrasound photos. Franchisers advertise that no medical background or certification is required.”

I asked Hirasaki a few questions.

How did you become interested in this?

1. When my last child was born in 1978, there was one ultrasound machine serving the entire population of Houston.
2. My work experience in the industrial and aerospace industries was extensive in heat and vibration.
3. For several decades, my work included industrial instrumentation in chemical plants and refineries. Most such instruments require routine calibration to stay accurate.
4. In the space program, there are always redundant devices because mechanical and electrical equipment may malfunction.
5. A first cousin died at age 6 of cancer caused by x-ray overdose in the 1940′s. I am old enough to remember getting my feet x-rayed at the shoe store.

Who else, in addition to Caroline Rodgers, has independently had this idea and written or posted about it?

Manuel Casanova, a neurologist. Nancy Evans, a science writer.

When you started looking into possible causes of autism, did you look into other possible explanations?

No. I asked myself what had changed about having a baby. Ultrasound seemed to be the obvious major change. Exposure to caffeine, smoking, alcohol and medications had all decreased.

Was there any particular evidence that you found especially persuasive?

On my website, there are 37 papers pertinent to the topic. My primary theory is that overexposure is the cause. The most persuasive evidence that points to this is:

1. The girls in the Kaiser-Permanente study who have the highest exposure to ultrasound during the second trimester have more than double the expected rate of autism.
2. Interruption of the neural migration patterns in the brains of the overexposed mice in the Yale study.
3. A known cause of autism is the mother having an infection. One theory is that as the mother’s body fights the infection with a fever, the rise in temperature of the fetal brain causes the damage. The temperature rises in tissue exposed to ultrasound are reported in a paper which calls for the FDA to examine the current allowable limits.
4. The much higher rate of autism for children who weighed less than 4.5 pounds at birth. The medical and AIUM protocol for infants this small is an immediate head ultrasound.

1. The researchers have education and experience  [and funding -- SR] in medicine, biology, genetics. This is where they look.
2. In the 1950′s, the FDA turned the regulation of medical ultrasound over to the American Institute for Ultrasound in Medicine (AIUM) which is funded by the government and the manufacturers. In November 2011, I expressed my concerns via email to the AIUM Bioeffects Committee. The chair responded that the AIUM monitors the topic and has conducted a thorough search of existing literature (2008) and found no link except the “weak association” in the Kaiser-Permanente study. The report states that there are potential risks to administering ultrasound tests, which include “postnatal thermal effects, fetal thermal effects, postnatal mechanical effects, fetal mechanical effects, and bioeffects considerations for ultrasound contrast agents.” Another section says “the long-term effects of tissue heating and cavitation have shown decreases in the size of red blood cells in cattle when exposed to intensities higher than diagnostic levels. However, long-term effects due to ultrasound exposure at diagnostic intensity is still unknown.” In 1992, the AIUM and the FDA increased the allowable intensity of prenatal ultrasound 8-fold without knowing whether it was safe. Twenty years later, they still don’t know.
3. Engineers who could do such research receive funding from the manufacturers. Such a researcher would be unlikely to propose research that could decimate the industry.
4. This is the classic search for a needle in a haystack. Less than 1 in 300 ultrasounds is causing noticeable brain damage.

I would add that links between sonograms and left-handedness are more evidence that sonograms can cause important brain changes. Here is a doctor who does sonograms dismissing the connection but what I find important is that the connection has been found repeatedly.

Two studies in Sweden by the same group have found high rates of defects in ultrasound machines used to scan pregnant women. The line of research began when doctors at the Karolinska Hospital discovered that many of their ultrasound machines were malfunctioning. (more…)

• Intervention bias by Nassim Taleb
• article about self-tracking in the Economic Times (India).
• Autism and PFCs
• Two weeks of FitBit
• Psychology professor Marc Hauser resigns from Harvard
• Stay classy, Larry Summers. Calling people far less powerful than himself “assholes” is entirely consistent with this. I like to think if Spy magazine still existed, it would run a Larry Summers Watch.

Thanks to Dave Lull and Aaron Blaisdell.

I have blogged several times about Caroline Rodger’s idea that sonograms during pregnancy greatly increase the risk of autism in the fetus. Her idea is supported by several lines of evidence, as she explains in this talk.

A new study provides more evidence. It found a high concordance rate among fraternal twins. In the general population from which the new study was drawn (California), about 1 child in 100 has autism. But if you are an identical twin, and your co-twin has autism spectrum disorder, your chance of having the same diagnosis is about 70%. The crucial point of the study is that the concordance was also high for fraternal twins: about 40%.  As one commenter put it, this result “puts a spotlight on pregnancy as a time when environmental factors might exert their effects”.

Another study found more risk of autism if the mom took an anti-depressant during pregnancy. This supports the idea that a bad prenatal environment causes autism.

Thanks to Paul Sas and Gary Wolf.

Robert Delaney is a geologist who does environmental cleanup in Michigan. While cleaning up an abandoned military base, he found remarkably high contamination by a chemical called PFOS. He had been wondering what causes autism. He came across a rodent study that found that the combination of PFOS and ultrasound was much more damaging to the nervous system than either alone. (See also this study.) He remembered that study when he read my posts about ultrasound and autism. He wrote to me: (more…)

Gary Taubes pointed to this PNAS paper about climate change and noted that one of the authors, Stephen Schneider, had a big non-financial conflict of interest: If it turns out the whole argument is wrong, he looks like a fool. The accompanying statement (“The authors declare no conflict of interest”) is, if taken to mean the authors have no conflict of interest, wildly inaccurate. Readers unaware of Schneider’s history wouldn’t know this.

I came across a similar example today. A reader of this blog wrote extensive criticisms (here and here) of the idea that prenatal ultrasound may cause autism. He believed Caroline Rodgers, my source for that idea, misrepresented the evidence. In particular, Rodgers pointed to a study that found ultrasound disturbed neuronal migration in mouse fetuses. She said it supported her idea. The reader disagreed, saying,

The bottom line for me is that Dr. Rakic (from the mouse study) clarified, “Our study in mice does not mean that use of ultrasound on human fetuses for appropriate diagnostic and medical purposes should be abandoned. Instead, our study warns against its non-medical use.” Yes. Okay. No more boutiquey, keepsake ultrasounds. Great. But for Rodgers to skew this data (along with the FDA’s and others’) into claiming that ultrasounds under the care of an Obstetrics professional (and for medical use) are causing autism is disingenuous at best, unethical propaganda for the Midwifery Way at worst.

The reader is a professor who teaches composition. Maybe an English professor.  He or she takes Rakic seriously, where I completely ignore his statement because of a conflict of interest. If Rakic questions “appropriate” ultrasound, he will be attacked in many ways, making his life unpleasant. I have no idea whether this swayed Rakic, but he would be only human if it did.

Of course developing neurons are unable to distinguish appropriate and inappropriate ultrasound. Rakic’s statement is ridiculous as Rakic and all insiders (neuroscientists) know, I believe. All insiders know that there are dozens of examples where findings from mouse brains have turned out to be true for human brains, in spite of the many differences between them, and that there are thousands of grant proposals in which mouse brains are claimed to be a worthwhile model for human brains. All insiders know this, realize the pressure on Rakic to say what he said, and, like me, just ignore it. As far as I can tell, Rakic pays no price for misleading outsiders because the outsiders don’t know they are being misled. (Just as with political lobbying: the public doesn’t understand what’s happening.) The composition professor doesn’t know this, as far as I can tell.

Rodgers is not claiming that ultrasounds “are causing autism”. She is saying they might cause autism, that there are several reasons to think so, and therefore (a) the ultrasound-autism idea deserves further scrutiny and (b) ultrasounds should be avoided as much as possible until more is known.

I previously blogged (also here) about Carolyn Rodgers’s idea that prenatal ultrasound may cause autism. It turns out that she isn’t the only person with this idea; researchers at the University of Louisville recently published the same idea.

I learned about the Louisville study from Anne Weiss, who said the connection has been plausible for a long time.

Ultrasound was introduced into obstetrics in the 1970′s and was generally restricted to high-risk pregnancies.  By the 1980′s policy statements were issued by ACOG, the NIH and equivalent bodies in Europe and Canada stating that its use should remain limited to high-risk cases. Despite these recommendations, ultrasound technology became common in hospitals and doctors’ offices and routinely applied to low-risk populations. Within a short time the majority of pregnant woman were being exposed at prenatal visits, during multiple scans in hospitals, and during continuous monitoring during labour (which could mean 12 to 14 hours during childbirth alone). Skills and techniques used to monitor the fetus prior to the introduction of ultrasound (in utero and during the birth process) were slowly undermined by the technology and often underutilized.  Iatrogenic effects from false positive readings, – unnecessary C- sections, inductions, instrumental deliveries etc. caused harm to moms and babies, especially in the early 1980′s.

Three important names in the 1980s were (1) Robin Mole, who presented a paper “Possible Hazards of Imaging and Doppler Ultrasound in Obstetrics” to the Royal Society of Medicine Forum on Maternity and the Newborn:  Ultrasonagraphy in Obstetrics, April 1985. She was former director of the Medical Research Council Radiobiology Unit, England.  Also the work of (2) M.E. Stratmeyer – Research in ultrasound.  A public health view.  Birth and Family Journal 1980 and (3) Doreen Liebeskind – still at Albert Enstein and a prof of radiology- presented at a symposium at Columbia in 1983.   She was concerned that ultrasound may be producing subtle changes in the fetal brain perhaps affecting behavioral mechanisms, possible changes in reflexes, IQ, attention span or some of the more subtle psychological, psychiatric or neurological phenomena.  Referred to animal and lab studies that showed ultrasound may cause chromosomal damage, breakdown of DNA, etc.  There are others who sounded the warning that this was not a benign technology but these voices were crowded out for varied reasons like threats of litigation, loss of the traditions skills of birthing etc.

There were also Japanese studies that raised concerns about ultrasound. Weiss continued:

Unfortunately the use of ultrasound in obstetrics has not declined, despite safety concerns and the lack of research to rule out serious neurological effects.  It’s so entrenched in modern obstetrical practice.    Doctors use the machines to protect themselves from litigation – in the case of fetal abnormalities, undetected multiples, placenta previa, neurological or physical damage to the fetus during childbirth, stillbirth etc.  It has almost become a form of entertainment – you can get photos and videos of baby’s ultrasound.  It’s disturbing how benign it appears.

Within the context of the work I do, ultrasound is just one of many concerns I have with the over-management and medicalization of childbirth.  My clients come to me to find ways to subvert this within the hospital setting or to prepare for a home birth with a midwife.   I also get referrals from doctors whose patients are dealing with difficult issues while pregnant.

Caroline Rodgers, whose ideas I blogged about yesterday, wrote to me about lack of research on the possibility that prenatal ultrasound causes autism:

I have heard confidentially that applications for funding of prenatal ultrasound studies (not specifically investigating autism) have been repeatedly denied over the years — which helps explain the great paucity of safety studies, especially since the early ’90s, when the FDA approved an allowable eightfold increase in acoustic output. As recently as this year, funding was denied an ambitious, multi-site study that would have investigated if there was a relationship between ultrasound and autism.

In 2006 when Yale neuroscientist Pasko Rakic announced the results of his study that found prenatal ultrasound interrupted neuronal migration in mice in a way that was consistent with the brains of autopsied autistics, I was surprised that several scientists, including Rakic, did their best to downplay the results. At the time, Rakic was one of many of Autism Speaks’s scientific advisors.

I have spoken with various people throughout the NIH about my concerns [about ultrasound]. They all pointed to various large studies they believe are investigating ultrasound as a possible environmental cause of autism — most recently, the National Children’s Study and EARLI, but when I tracked down the study designs, it turned out that ultrasound is not being studied.

In a report at the time Rakic’s study was published, he indeed downplayed the results:

Dr. Pasko Rakic, chairman of the Yale department of neurobiology and leader of the study, was quick to offer parents reassurance about the safety of ultrasound — done for the proper reasons — in human pregnancies.

“If I had a daughter and she was pregnant, I would recommend she had it for medical reasons,” Rakic said.

Another researcher agreed:

“I couldn’t agree with him more,” said Dr. Joshua Copel, a professor of obstetrics, gynecology and reproductive sciences at Yale and spokesman for the American College of Obstetrics and Gynecology (ACOG). He was not involved in the study. . .

The researchers noted that mice are very different from humans, so the results of their study must be interpreted with caution.

“The forms of migration [of brain cells] and the timing of migration differ in primates like humans than in mice,” Copel said. “In humans, there is a much longer period in which neurons [nerve cells] are migrating.”

Does that sound “very different”?