Imaging of the Autism Brain and Surrounding Tissues Using Linear 2D Transcranial Ultrasonography

Background:  Autism spectrum disorders (ASDs) are developmental conditions of uncertain etiology which have now affected more than 1% of the school-age population of children in many developed nations. Cortical abnormalities observed from autopsy specimens, including inflammation and structural aberrations, have been inconsistently reported. Recently, serial brain magnetic resonance imaging studies from near birth to 2 years of age were noted to be highly predictive of autism when increased frontal extra-axial fluid remained persistent. Transcranial ultrasonography (TUS) via the temporal bone acoustical window appeared to be a potential method of investigation to determine the presence of both cortical abnormalities and increased extra-axial fluid (EAF).

Objectives: To evaluate the brain of children with autism and their siblings using 2D TUS.

Methods:  TUS was accomplished using a linear probe typically used for musculoskeletal investigations (10-5 MHz). Multiple images were recorded and real-time measurement of the cortex and extra-axial spaces were accomplished. Specifically, anatomical areas of significance to autism (primary language and auditory cortical locations) are readily imaged with this novel ultrasonographic technique. With this initial cohort, parents volunteered 18 ASD subjects (males =17) for evaluations (all ages < 16, mean = 7.81 years + 2.4 years), and 12 neurotypical siblings were also examined (all ages < 21, mean = 8.89 years + 3.82 years). Childhood Autism Rating Scale, Second Edition, (CARS2®) scores were obtained as a routine in the practice and the ASD score mean was 48.65 + 7.32 (Severe).

Results: Comparisons of the extra-axial spaces indicated increases in the ASD subjects. For EAF we scored based on the gyral summit distances between the arachnoid membrane and the cortical pia layer (subarachnoid space): 1) <0.05 cm, 2) 0.05 – 0.07 cm, 3) 0.07 – 0.10 cm, 4) > 0.10 cm. All of the neurotypical siblings scored 1, whereas the ASD mean score was 3.31 + 0.7. We also defined cortical dysplasia as the following: hypoechoic lesions within the substance of the cortex, or disturbed layering within the grey matter. For cortical dysplasia we scored: 1) none observed, 2) rare hypoechogenic lesions and/or mildly atypical cortical layering patterns, 3) more common, but separated areas of cortical hypoechogenic lesions, or more commonly abnormal cortical layering, 4) very common or confluent areas of cortical hypoechogenicity and/or markedly disturbed layering of the cortex. Again all of the neurotypical siblings scored 1, while the ASD subjects’ mean score was 3.0 + 0.92.

Conclusions:  These initial TUS observations are consistent with and extend the age range of previous MRI observations of increased EAF in ASDs. The new real-time observations of hypoechogenic lesions in the cortical grey matter suggest focal cortical disruption in areas critical to both language and auditory processing (core features of autism). TUS may be a useful screening technique for ASDs which, if confirmed with repeated studies and high resolution MRI, provides rapid, noninvasive qualification of EAF and cortical lesions.