Cavum Septum Pellucidum and Cavum Vergae in Macrocephaly and Autism Spectrum Condition

Background: Previous studies have suggested that ~20% of individuals with an autism spectrum condition (ASC) exhibit abnormally increased brain volume, termed macrocephaly. Macrocephaly in children without an ASC is associated with an increased rate of errors in fetal development, including midline brain anomalies, and a multiple studies highlight the co-occurrence of macrocephaly with abnormalities of the septum, cavum septum pellucidum (CSP), and cavum vergae (CV) in particular. Despite the well known association between macrocephaly and ASC, septal abnormalities have not yet been systematically investigated in ASC.

Objectives: Compare the rates of CSP and CSP+CV between individuals with an ASC, typically developing (TD) individuals, and individuals with reading disorder (RD), and investigate the link to macrocephaly.

Methods: Magnetic resonance imaging (MRI) was performed in a 1.5-Tesla Marconi MR scanner. Participants in the study comprised 3 groups: ASC (N=38), TD (N=27), and RD (N=22). The RD group was included to test the specificity of CSP/CSP+CV to ASC rather than to language disorders in general. The TD group was enriched with participants who had benign macrocephaly, in order to control for the increased rate of macrocephaly in ASC and study the relationship between septal anomalies and macrocephaly. All participants were male, aged 7 to 32 years, and had a performance IQ>65. Each group was separated into those with and those without macrocephaly. Fisher’s exact test and ANOVA were used to compare CSP between ASC and the comparison groups. Correlations between CSP and the following measures were computed: IQ, autism severity, and total brain volume. The Chi-square statistic was computed for the presence of CSP and CSP+CV. A multiple regression analysis was performed to examine the quantitative relationship between total brain volume and cavum length.

Results: There were no significant differences in cavum rating between the ASC group and the comparison groups. While CSP was present in more than 75% of the ASC and comparison participants, CSP size was generally small. About 15% of the ASC participants and about 18% of the combined comparison group had large CSP. Two ASC individuals and 1 TD participant had CSP+CV. There was no significant difference in type of cavum (none, any, large, combined) in participants with vs. without macrocephaly (X²=3.71, p=0.25) when the groups were combined or analyzed individually. Compared to participants without macrocephaly, both rates of any CSP and enlarged CSP were greater in those with macrocephaly (any CSP: t=2.19, p=0.02; enlarged CSP: Fishers Exact test, p=0.030). There was a significant positive correlation between total brain volume and CSP in the TD group (r=0.441, p=0.035), but this correlation was driven by the macrocephalic participants in that group and became insignificant once they were excluded. Autism severity and FSIQ did not correlate significantly with CSP.

Conclusions: The most distinct finding in this study was that macrocephaly, regardless of ASC or comparison participant status, was associated with CSP. It could be that CSP is a vulnerability marker that is clinically expressed in ASC and previously TD individuals by specific neuropsychiatric and cognitive features when other environmental effects occur.