DSM Through the Looking Glass: Corpus Callosum Volume in High-Functioning Autism and Asperger Syndrome

Background:  

Mounting evidence suggests that the corpus callosum (CC) may be central to understanding the pathophysiology of autism spectrum conditions.  As the largest white matter tract in the brain, the CC mediates the interhemispheric communication underpinning higher cognitive functioning.  Subsections of the callosum have been shown to vary dramatically in quantitative characteristics of axon density, size, and degree of myelination (LaMantia and Rakic, 1990).  Discrepancies remain, however, as to which subregions of the CC are affected in autism (Brambilla et al., 2003), and the question as to whether CC volume differs between individuals with High-Functioning Autism (HFA) and Asperger Syndrome (AS) remains unanswered.  This question has particular relevance in the context of the proposed changes within DSM-V that will consolidate HFA and AS into one diagnostic category of Autism Spectrum Disorder.   

Objectives:  

(1) To acquire a robust sample of CC volumetric measurements in male adolescents with HFA and AS; (2) to correlate volumetric differences in the CC with Autism Diagnostic Interview (ADI-R) clinical measures and task performance on an executive function measure; and (3) to test for volumetric differences in the CC in DSM-IV subgroups of HFA and AS. 

Methods:  

High-resolution structural magnetic resonance images of the brain were obtained on a cohort of adolescent males (12-18 years) with HFA (n=19), AS (n=16), and typically developing controls (n=20). The CC was partitioned into the seven functional subregions according to the Witelson method (Witelson, 1989) and traced in midsagittal and parasagittal sections using Analyze software.  100% of the dataset was traced by two independent raters blinded to group identity (intraclass correlation = 0.98), and all values were averaged.  The CANTAB (Cambridge Neuropsychological Test Automated Battery) ‘Stockings of Cambridge’ task was administered to all participants as a measure of planning ability.

Results:  

Structural neuroimaging data include CC subregion and total CC volume measurements relative to global white matter.  Significant differences were observed between HFA and controls that were not present in AS subjects. Total CC volume was increased in HFA compared to controls (p=0.016), as was the volume of the anterior (p=0.008) and posterior midbody (p=0.016).  Volume of the CC rostrum volume was significantly correlated with ADI-R clinical symptomatology in the AS cohort, notably on communication (p=0.005, R=0.684) and restricted, repetitive and stereotyped behavior (p=0.018, R=0.600) subdomains.  The volume of the genu was correlated at trend level (p=0.052) with performance on the Stockings of Cambridge task.

Conclusions:  

Significant differences in CC subregion size were found between HFA and controls, but not between AS and controls. This has implications for our understanding of cognitive deficits associated with autism spectrum conditions and for the proposed DSM-V changes.