Normal Rates of Clinical Neuroradiological Abnormalities in Children with High Functioning Autism

Background: Magnetic resonance imaging (MRI) brain scans have been used to analyze highly specific volumetric and morphological features of the brains of people with autism, however there have been no comprehensive studies on the rates of basic clinical neuroradiological abnormalities identifiable in MRI scans.

Objectives: To examine the rates of clinical neuroradiological abnormalities in children with HFA, comparing them with the rates for typically developing (TD) children and a clinical control group comprised of children with Attention-Deficit/Hyperactivity Disorder (ADHD).

Methods: Clinical radiological reports based on MRI brain scans were collected for 87 children with HFA (12 girls), 168 TD children (79 girls) and 125 children with ADHD (47 girls), 7.21-13.85 years old.  These subjects were recruited as part of an ongoing neuroimaging and behavioral study; children with a history of epilepsy or known genetic diagnosis were excluded from participation.  Findings were categorized as normal or abnormal, with 13 abnormal sub-categories defined.  Three-way chi-square analysis was used to compare the within-group rates of normal versus abnormal reports and the incidence of each abnormal sub-category with a usable number of subjects.  The abnormal sub-categories analyzed were asymetrical ventricles, Chiari I malformation/low-lying tonsils, cystic lesions, focal subcortical signal abnormalities, focal white matter signal abnormalities, heterotopia, mega cisterna magna/posterior fossa cyst and mild ventricular prominence/volume loss.

Results: Three-way chi-square analysis revealed no significant effect of group on the overall rate of clinical neuroradiological abnormalities (X ² = 2.06, p = 0.36); percentages of children in each diagnostic group with reported neuroradiological abnormalities of any type were HFA: 10.3%, TD: 12.5% and ADHD: 16.8%.  Analyses of individual sub-categories of abnormalities revealed a significant group difference for mega cisterna magna/posterior fossa cyst (X ² = 7.51, p = 0.02); reported percentages in each group were HFA: 1.2%, TD: 0.0%, ADHD: 4.0%.  Post-hoc two-group comparisons revealed that this result was not related to autism-associated differences, but rather to a higher rate of mega cisterna magna/posterior fossa cyst in children with ADHD compared to TD children (X ² = 6.84, p = 0.009).

Conclusions: The lack of significant differences in rates of clinical neuroradiological abnormalities in children with HFA, TD children and children with ADHD suggests that anatomical abnormalities attributable to diagnosis are more subtle than what is seen at the level of clinical MRI examination.  Implications for routine use of clinical MRI scanning in children with HFA will be discussed.