MRI Scans May Help Test for Autism

Study Shows New Test Using MRIs May Be Useful in Diagnosis of Autism

Medically Reviewed by Laura J. Martin, MD on December 02, 2010
From the WebMD Archives

Dec. 2, 2010 -- Researchers may be getting closer to developing a test to diagnose autism spectrum disorder using magnetic resonance imaging (MRI) scans.

Autism is typically diagnosed through observations, along with educational and psychological testing.

The new test, named the Lange-Lainhart test after the researchers who developed it, uses MRI scans to produce a detailed map of the brain’s wiring in the six regions responsible for language, social, and emotional function.

If validated in larger groups, this test may lead to earlier, more definitive autism diagnoses and help researchers get a better handle on some of the genetic roots of autism.

The new findings appear online in the journal Autism Research.

The CDC estimates that about one in 110 children in the U.S has an autism spectrum disorder, an umbrella term for a group of developmental disorders that can range from mild to severe and that often affect a person’s ability to communicate and relate to others.

The new imaging test was 94% accurate in pinpointing autism among 30 men aged 18 to 26 who had been diagnosed with a high functioning form of autism when compared with 30 men of the same age who did not have any signs of autism. The researchers repeated the test on another smaller set of participants, and it produced similar results.

More Work Needed on Test

The MRI autism test isn’t ready yet, says Nicholas Lange, ScD, an associate professor of psychiatry at Harvard Medical School and director of the neurostatistics laboratory at McLean Hospital in Boston. “Ongoing studies with more subjects in other people’s labs will help us learn how this test holds up in the broader population.”

The Lange-Lainhart test looks at the wiring of the brain in a very detailed way and can identify deviations in brain circuitry among people with autism, he says.

The new test will also be studied in other types of autism, younger children, and people with other brain disorders. Virtually every neurological and psychiatric disorder exhibits signs of faulty brain circuitry, so a test would need to be able to distinguish autism from other disorders.

The new MRI test is not the only medical or biological diagnostic test for autism in development. Blood and urine tests are also being looked at in the U.S. and abroad, as are other imaging tests.

While it’s not yet clear which test -- if any -- will make it to the finish line, autism experts agree that there is a need for a medical test to help diagnose autism.

“We don’t really know what autism is, and all we have at present is a subjective test that is used to diagnose the disorder which involves four-hour interviews with parents and one hour spent observing the child,” Lange says.

This test only measures the child’s behavior and cognitive ability and is subject to a doctor’s call, he says.

Adriana Di Martino, MD, an assistant professor of child and adolescent psychiatry at NYU Langone Medical Center in New York City, is cautiously optimistic about the new findings.

But, Di Martino says, “before we talk about a test that can be used clinically, we need [to study] a large group of subjects with autism and other diseases.”

“I would not go saying there is now a test to diagnose autism with MRI, but we may get there in the future,” she says.

“A really accurate and valid test or biomarker will aid the process, but it is unlikely that this will substitute for the work of a psychologist,” she says. “The work of the psychologist in observing the child is still crucial.”

Such a test could also lead to earlier diagnosis than is currently possible, she says. Autism signs can sometimes be picked up at 18 months or younger, but a reliable diagnosis is usually not made before a child turns 2.

Treatment Potential

Earlier diagnosis and intervention can have dramatic effects on treatment outcomes among some children with autism, she says.

That would be an important use for a test like the one in the new study, saysKevin Pelphrey, PhD, the Harris Associate Professor of Child Psychiatry and Psychology at Yale School of Medicine in New Haven, Conn.

“Autism isn’t reliably diagnosable before age 2 and that is at state of-the-art centers, but if we had an objective diagnostic measure we could do it earlier,” he says.

Behavioral observation of people with autism is “magnificent, but crude compared to what one can do with a quantitative measure like looking at the brain,” Pelphrey says.

There are other implications as well, he says.

Brain imaging could also help uncover the genetic basis of autism by finding patterns among people with autism and their unaffected family members, he says.

A brain scan could also help classify types of autism among people who already have a diagnosis. A better picture and understanding of the nuances in brain’s wiring may also help doctors better target their treatments.

“We could also look at subtle cases where you are not sure and autism is one of the possibilities, but we will never see a situation where we scan every baby who is born to see if they are at risk for autism,” he says.

“In a few years with studies like this continuing to come out, we will be in a place where we will have brain diagnostics, but they will never replace behavioral observations,” he says.

Show Sources


Lange, N. Autism Research, 2010.

Nicholas Lange, ScD, associate professor of psychiatry, Harvard Medical School; director, neurostatistics laboratory, McLean Hospital, Boston.

Adriana Di Martino, MD, assistant professor, child and adolescent psychiatry, NYU Langone Medical Center.

Kevin Pelphrey, PhD, Harris Associate Professor of Child Psychiatry and Psychology, Yale School of Medicine, New Haven, Conn.

© 2010 WebMD, LLC. All rights reserved. View privacy policy and trust info