Lateralization of Brain Networks and Clinical Severity in Toddlers with Autism Spectrum Disorder: A Diffusion MRI Study

Background: Recent diffusion imaging and tractography studies in adolescents and young adults with ASD have reported a loss or an inversion of the left–right asymmetry in regions crucial for socio-communicative skills, including the Cingulate, Arcuate Fasciculus, Uncinate Fasciculus, Superior Temporal Gyrus and neural circuits involving the fusiform. Few studies explored atypical lateralization in younger children and its correlation with clinical severity of ASD.

Objectives: We studied a group of children younger than 36 months with a first clinical diagnosis of ASD. Our aim was to measure the laterality indexes of white matter pathways on brain diffusion MRI, in order to determine their correlation with clinical severity of ASD, as assessed by ADOS-G.

Methods: As part of an ongoing prospective project, started on January 2012, we selected all children referred to our tertiary care centre for social-communication impairment. Subjects were included in the present study if they i) received a clinical diagnosis of ASD based on DSM IV, ii) were aged 36 months or less and iii) had no neurometabolic or genetic disorders. Twenty-four subjects fulfilled our inclusion criteria. Four patients were excluded due to incomplete MRI assessment. The final cohort consisted of 20 subjects (15 males; mean age 28 +/- 4.7 months; age range 20-36 months).

Brain MRI was performed on a GE 1.5 T scanner. High angular resolution diffusion imaging (HARDI) scans were acquired using either 31 or 65 diffusion weighted images (b value = 1000 s x mm^-2). Image post-processing consisted of cortical parcellation (FreeSurfer software) on high resolution T1 image and whole brain fibre tracking by constrained spherical deconvolution (MRtrix software). Data from cortical parcellation were combined with tractography to obtain a connection matrix, and within each pair of connected areas, diffusion indexes (D_i) were obtained including mean fractional anisotropy (FA), number of streamlines (NUM) and tract volumes (VOL). A laterality index was generated for each connected pairs according to the following formula: (right D_i – left D_i / right D_i + left D_i).

We correlated laterality indexes from each pair of connected areas with the ADOS-G total score using Pearson's linear correlation coefficient. Correlation values were corrected for multiple comparisons based on a false discovery rate of 10%.

Results: Laterality indexes of FA were significantly correlated with ADOS total scores only in two intra-frontal connected areas (correlation was positive in 1 case and negative in the other). Laterality indexes of VOL and NUM showed no significant positive correlations while they both showed significant negative correlations (p<0.05) in six connected areas, mainly fronto-temporal.  

Conclusions: This study provides first evidence of a significant correlation between brain lateralization of diffusion indexes and clinical severity in toddlers with a clinical diagnosis of ASD. Significant correlations mainly involved regions within the fronto-temporal circuits, known to be crucial for socio-communicative skills. It is of interest that most correlations were negative, suggesting an inversion of the typical left–right asymmetry in subjects with most severe clinical impairment.