Ultrasound exams during pregnancy are common. And researchers have now looked at whether a diagnosis of autism spectrum disorder is more likely among children exposed to this technology in the womb.
There is no association between the number or duration of prenatal ultrasounds and a later diagnosis of autism in the child, according to a new study published Monday in JAMA Pediatrics.
However, the researchers did find a statistical association between deep ultrasound wave penetration during the first and second trimesters and autism.
"Depth of penetration has to do with the distance between the ultrasound transducer (probe) on the skin and the point at what you're looking at on the ultrasound," said Dr. Jodi Abbott, a co-author of the new study and a physician with Boston Medical Center.
The depth -- as indicated on the image and as read by the authors for the study -- is the farthest down the ultrasound beam reaches, explained Dr. Jacques S. Abramowicz, who is chair of the safety committee of the World Federation of Ultrasound in Medicine and Biology. "It has nothing to do with where the fetus and his/her parts are. The depth could indicate 20 centimeters and the fetus be at 12 centimeters," he said. Abramowicz did not participate in the new research.
Absorption of the wave traveling through tissue increases with frequency, so the higher the frequency, the shallower the ultrasound penetrates, according to the Journal of Ultrasound in Medicine. Ultrasound technology gives the technician some leeway in adjusting the frequency to get a better image of the fetus.
Depth of penetration "varies based on the size of the woman and the amount of tissue that she has on her belly between the transducer and the fetus," Abbott said. This is not something a woman can control, she said.
Dr. N. Paul Rosman, first author of the study and a pediatric neurologist at Boston Medical Center, emphasized that an association does not prove cause and effect. The study results, then, do not mean that ultrasounds cause autism.
"We think this study was done well, but there are deficiencies, and that's why we call for additional studies," Rosman said. Other researchers say that the methodology of the study was not rigorous enough to draw firm conclusions.
Autism spectrum disorders are a group of developmental disabilities characterized by challenges with social skills, speech and nonverbal communication. About 30% of people with autism have a defined genetic abnormality, Rosman said, while another 20% of cases are due to recognizable diseases, such as a mother who gets rubella while pregnant. No specific cause has been determined for the remaining 50% of cases, he said.
Over the past few decades, the prevalence of autism has steadily increased. The disorder affects one in every 68 people, according to the US Centers for Disease Control and Prevention. During the same few decades, expectant mothers have undergone an increasing number of ultrasound exams. Some past research has suggested but not proved that autism may be associated with frequent prenatal ultrasound exams.
To investigate, Boston Medical Center researchers looked at the medical records of 420 children; 107 of these children had been diagnosed with autism.
The researchers examined the sonograms of each child with autism and compared these images with those of two typically developing children and those of one child with a developmental delay.
A higher frequency of prenatal ultrasounds was not seen in the autism group.
The children with autism received, on average, about 5.9 scans, estimated the researchers. Typically developing children averaged 6.3 scans, while children in the developmental delay group averaged 6.1.
The autism group also had not been exposed to longer ultrasound sessions. Compared with the typically developing group, children with autism experienced a shorter duration of ultrasound exposure during the first trimester (290.4 seconds vs. 406.4 seconds) and second trimester (1,687.6 seconds vs. 2,011.0 seconds), the researchers found.
"Despite the increased number of ultrasounds over the years, the concerning findings in the past have not been borne out in the study," Abbott said.
Yet one important difference between groups of children was seen: The autism group averaged greater depth of penetration than typically developing children during the first trimester (12.5 centimeters vs. 11.6 centimeters) and second trimester (12.9 centimeters vs. 12.5 centimeters). These differences are statistically significant, the study authors noted.
Similarly, depth of penetration in the first trimester for the autism group was greater than the developmental delay group (12.5 centimeters vs. 11.6 centimeters).
Rosman suggested readers "look at the study critically." He and his colleagues did not examine multiple factors that might affect a child in the womb, including whether a woman became ill or smoked during her pregnancy.
It is necessary, he said, to do more research to verify the new findings. If future research supports these results, he has a theory why ultrasounds and autism may be linked: It has to do with fetal brain development.
"The cells that give rise to the mature brain originate deep within the brain and migrate outward to form the mature cerebral cortex," he said. "They do this in a very orderly fashion -- one generation of cells following another."
If the ultrasound waves penetrate deeply, they may disturb the deep-lying cells within the brain and "potentially interfere with their migration outward to form a normal cerebral cortex," Rosman said.
Persuasive theories aside, other scientists say pregnant women should not fear a medically necessary ultrasound exam.
Differences 'not significant'
Dr. Marvin Ziskin, a professor of radiology and medical physics at Temple University's School of Medicine, said, "Unfortunately, the authors do not appear to know what is meant by the ultrasound penetration depth. It does not relate to the amount of ultrasound entering the body."
Ziskin was not involved in the study.
"The factors that determine the amount of ultrasound entering the body, and what would affect the fetus, are the mechanical index, the thermal index, and ultrasound power and intensity, for which no significant differences were found," Ziskin said. "The amount of ultrasound imparted into pregnant patients has no association with autism spectrum disorder."
Abramowicz, who is also a professor and director of ultrasound services in the Department of Obstetrics and Gynecology at the University of Chicago, also said the study methodology is "completely flawed."
"As the ultrasound penetrates the tissues, it actually loses energy," he said. "The difference between the groups of children is minimal in terms of the depth. And clinically not significant."
The research team also examined the scans retrospectively and was not present when they were taken, and this is another study flaw, Abramowicz said. They could not know, he said, how long an ultrasound exam lasted based on the timestamps on the first and last sonograms of a session. There could have been half an hour of scanning before the technician took the first image and another half-hour of scanning after the last, he explained.
Asked whetherf individual technicians control depth of penetration, Abramowicz said the newest machines, when in obstetric mode, simply cannot penetrate deeper than what is safe for a fetus.
Finally, the authors claim that ultrasound technology is "minimally regulated," a claim to which others also take exception.
"It is extremely regulated," Abramowicz said, "particularly in the United States, with the Food and Drug Administration looking very closely at the machines."
What is true is that ultrasound usage is less regulated, he said. Still, any patient whose ultrasound exam is done by a qualified medical technician is safe.
However, the American Pregnancy Association notes that commercial keepsake ultrasound services, which offer families a sneak preview of their growing baby, have opened across the nation. An expectant mother should never undergo an ultrasound in a shopping mall or other commercial space to obtain a "keepsake" sonogram.
"Absolutely not!" Abramowicz said.
Thomas Frazier, chief science officer of Autism Speaks, said the new study doesn't provide "enough evidence to make a recommendation for or against ultrasound."
Frazier, who was not involved with the research, added that even if depth of ultrasound somehow increased the risk of autism, it would probably be "a small increase in risk and autism would not be due specifically to ultrasound in most cases."
"Please don't cause any undo fear in the minds of pregnant women. They have enough to worry about," Ziskin said. If anything, the study "should give expectant mothers confidence that ultrasound will not cause autism spectrum disorder or delayed development," he said.