A Better Approach to Measuring Brain Volume in Autism Spectrum Disorder

Background: Total brain volume (TBV) and intracranial volume (ICV) have been extensively researched in individuals with Autism Spectrum Disorder (ASD). Though, findings are often inconsistent. Additionally, some past studies did not account for age and gender, and were often based on small to moderate sample sizes.

Objectives: Brain volume data from studies examining autistic and typically developing (TD) participants were gathered to develop the Autism Neurophysiological Parameters Atlas (ANPA). Brain volume trends in light of age were compared between these two groups.

Methods: Following a systematic search, 34 and 30 studies measuring TBV and ICV, respectively, were included. Raw data, or the means and standard deviations of brain volume and age, were obtained from 3,316 autistic participants and 2,744 TD participants. ANCOVA’s, controlling for age, compared raw data of autistic participants to TD individuals for males and females separately across a number of age groups. The frequency of extreme (±1.5 SD) TBV and ICV were also explored.

Results: Compared to TD males, findings revealed autistic males to have larger TBV and ICV during late childhood and adolescence (p<.05), respectively, as well as smaller TBV during adulthood (p<.05). In contrast, brain volume was not different between autistic and TD females. Extremely small (OR=3.11), and extremely large (OR=7.17), TBV was more common in autistic than TD males aged two to four years. Compared to controls, autistic males were more likely to have extremely large TBV and ICV between nine to 12 years (OR=3.73), and between 13 to 17 years (OR=6.02), respectively. During adulthood, autistic males were more likely to have extremely small TBV than TD males between 18 and 24 years (OR=3.41), and above 25 years (OR=4.04).

Conclusions: We conclude that for those diagnosed with autism, brain volume is dynamic over age, and varies considerably between autistic individuals. Although some mean differences were observed, a greater amount of variance in brain volume was present in autistic than TD males. Future research investigating brain volume in ASD is needed to account for such variance rather than comparing means between groups. Unlike meta-analyses and independent studies, the ANPA properly accounts for outliers by capturing the entire variance in the data. Hence, the ANPA more effectively describes brain volume in ASD compared to previous methods.