MRI as potential diagnostic tool for autism

No major structural differences between the brains of people with autism and those without it have been identified, with the exception of brain volume and head circumference in children. So the bulk of neurological research on the disorder focuses on how various regions of the brain communicate with one another.

Now, in a study published in October in the journal Cerebral Cortex, researchers at the University of Utah say they are one step closer to using magnetic resonance imaging (MRI) to diagnose autism.

"This work adds an important piece of information to the autism puzzle," says principal investigator Janet Lainhart, an associate professor of psychiatry and pediatrics. "It adds evidence of functional impairment in brain connectivity in autism and brings us a step closer to a better understanding of this disorder. When you understand it at a biological level, you can envision how the disorder develops, what are the factors that cause it, and how can we change it."

The researchers monitored 53 males with high-functioning autism and 39 typically developing males from late childhood through early adulthood. MRIs of these patients show lower-level communication between regions associated with functions such as facial recognition and motor skills than the levels found in MRIs of people without the disorder.

In the article abstract, the researchers note: "Interhemispheric connectivity differences were better explained by diagnosis of autism than by potentially confounding neuropsychological metrics of language, IQ, or handedness."

While connectivity abnormalities have already been identified in patients with autism in previous studies, this is the first of its kind to characterize abnormalities in the entire brain rather than in specific pathways. Imaging of the entire brain has the potential to help experts identify specific types of the disorder.

"This is a complex disorder that doesn't just fall into one category," Lainhart says. "We hope the information can lead us to characterizing different types of autism that may have different symptoms or prognoses that will allow us to identify the best treatment for each affected individual."

Because the study was limited to high-functioning males, the findings may not translate to females, younger children, or "lower functioning" individuals with autism, the researchers write. One skeptic, Laura Bono of the autism advocacy group "Safe Minds," tells Business Week that using MRI as a primary screening tool is unrealistic and therefore not terribly beneficial:

Knowing that diagnosing autism is about looking for certain behaviors, only children already identified as having issues that might be a function of autism would be screened. So if there's no targeted treatment to follow up on the results of this kind of scan, then it's just another expensive test for parents who are already barraged with so many tests at diagnosis. And in that case, I see this as just a waste of money.

So there's a bit of a Catch-22: the scans must be put to the test to develop targeted treatments, but until there are targeted treatments, using the scans widely is impractical and expensive.

Needless to say, this study is likely not the last to come from Lainhart's autism imaging research group--which includes experts from the University of Utah, Brigham Young University, the University of Wisconsin, and Harvard University.