Enhancing early detection of autism

See allHide authors and affiliations

Science Translational Medicine  25 Sep 2019:
Vol. 11, Issue 511, eaaz3708
DOI: 10.1126/scitranslmed.aaz3708


Electroencephalography might help to predict autism development in infants.

Early detection of autism provides the best opportunity for early intervention when the brain is most plastic. However, behavioral signs of autism spectrum disorders (ASD) typically do not appear until the second year of life or later. Developing biomarkers of ASD in early postnatal life is critical for identifying infants in need of early intervention and optimizing outcomes.

Using electroencephalography (EEG) measures of brain oscillations, Gabard-Durnam and colleagues tested whether, how, and when EEG power might help to predict later ASD outcomes. Participants were infants at low familial risk of ASD or high risk due to having a sibling with ASD. The authors collected EEG measurements when the infants were 3, 6, 9, 12, 18, 24, and 36 months of age. Findings showed that EEG power over the frontal lobes within the first year of life was the strongest predictor of subsequent ASD. Importantly, information from single timepoints as well as developmental change over time both contributed to prediction. By analyzing specific frequency bands, the authors found that changes in delta and gamma power particularly distinguished between infants with and without future ASD diagnoses.

These findings suggest that measures of early brain development could enhance early risk identification, even before behavioral symptoms of autism emerge. A major strength of this work is the examination of the same infants across development. This longitudinal design reveals that developmental trajectories themselves contribute meaningfully to early detection and highlights the importance of considering neurodevelopmental timing. Translational efforts into clinical settings could also benefit from the authors’ finding that the spatial localization of electrode placement mattered for prediction. Lastly, examining the best predictors of ASD outcomes could provide novel insight into the pathophysiology of autism. Delta and gamma frequencies have been associated with excitation/inhibition balance across species and may indicate that disruptions in neuroplasticity are core features of ASD.

Building upon this study, future research will benefit from testing the generalizability of these EEG predictors in larger cohorts and the extent to which the findings are specific to ASD versus other neurodevelopmental disorders.

Highlighted Article

View Abstract

Navigate This Article