Signs of disrupted connections between the hemispheres of the brain correlate with a diagnosis of autism spectrum disorder (ASD) in adolescents and adults, researchers say.
The disruption in the white matter of the corpus callosum appears to increase with age, a finding that could improve diagnoses and yield new understandings of the causes of the disease, said Clara Weber, a post-graduate research fellow at Yale University in Newhaven, Connecticut.
"We see that autism has progressive changes also in morphology and not only in behavior," Weber told Medscape Medical News.
She presented an analysis of MRIs at the Radiological Society of North America (RSNA) 2021 annual meeting.
Previous studies have also found differences between white matter microstructure in children with and without ASD. But sample sizes were small, making statistically significant findings difficult.
To address that challenge, Weber and colleagues analyzed data on 583 patients from four cohorts created using National Database of Autism research for prior studies. The four cohorts were:
34 infants with ASD and 121 controls, 65.8% male, with a median age 7 months.
57 toddlers with ASD and 45 controls, 73.5% male, median age 32 months.
106 adolescents with ASD and 124 controls, 50.9% male, median age 158 months.
67 young adults with ASD and 29 controls, 99.0% male, median age 230 months.
They used diffusion tensor imaging (DTI) brain scans, an MRI technique that measures connectivity in the brain by detecting how water moves along white matter tracts.
And they used machine learning to look for correlations among fractional anisotropy, mean diffusivity, radial diffusivity, age, and sex. Fractional anisotropy measures the extent to which water diffusion is restricted to just one direction. A value of zero means that diffusion is unrestricted in all directions. A value of one means that diffusion occurs only in one direction, a sign of good connectivity.
"Mean diffusivity" relates to the overall mobility of water molecules, which reflects how densely cells are packed together. "Radial diffusivity" is the extent water diffuses perpendicular to a white matter tract.
"When cells are very loose, and water can diffuse in many directions, then diffusivity would be rather high," Weber said.
The researchers found statistically significant reduced fractional anisotropy within the anterior/middle tracts of the corpus callosum in adolescents and young adults with autism compared with individuals in the control group. There were also correlations between autism diagnosis, and both mean diffusivity and radial diffusivity in young adults.
The researchers also found that the change in fractional anisotropy was age-related. And they found that fractional anisotropy was lower in female than male subjects, even after controlling for autism diagnosis.
Using computer learning to take all these correlations into account, they found that their model was about 75% accurate (area under the curve, 0.75) in identifying whether patients had autism based on their age, sex and these structural features.
"It would be wrong from a physician to just use a machine-learning algorithm based on MRI to diagnose autism," Weber said. "You obviously need to speak with the patients. You need to assess, then you need to have an image of their capabilities and their behavior to make a diagnosis. So, basically, we could use this as a tool to assist in diagnosis and also to assist in therapy planning."
The finding of a difference in the corpus callosum between patients with and without autism is not new, said Dennis Dimond, PhD, who has researched white matter and autism at the University of Calgary in Alberta, Canada, but was not involved in the study. "But I think having a large sample size like this really consolidates that as a true finding, as opposed to something that could be a fluke or due to other variables that aren't accounted for," he told Medscape Medical News.
The finding that the differences in brain morphology don't show up in infants but only in older individuals is also important, he said. "It really just kinds of drives home the point that this is a neurodevelopmental disorder and that you have ongoing changes over the course of brain development that might be contributing to the characteristics that you see in autism."
More such studies are needed because researchers are still a long way from a full understanding of the underlying mechanism of the condition, he said.
Weber and Dimond have reported no relevant financial relationships.
Radiological Society of North American (RSNA) 2021 annual meeting.
Laird Harrison writes about science, health, and culture. His work has appeared in national magazines, in newspapers, on public radio, and on websites. He is at work on a novel about alternate realities in physics. Harrison teaches writing at the Writers Grotto. Visit him at www.lairdharrison.com or follow him on Twitter: @LairdH
Lead Image: National Institute of Health
Medscape Medical News © 2021 WebMD, LLC
Send comments and news tips to firstname.lastname@example.org.
Cite this: Impaired Brain Connections Linked to Autism - Medscape - Dec 02, 2021.