Objectives: The primary goal of this investigation was to examine 1H MRS metabolite levels in white matter regions in children with autism. The relationship between metabolites levels and age was also investigated.
Methods: Multi-voxel 1H MRS scans were obtained from a group of 23 children with autism and 23 age-, and gender-matched healthy controls. Using a STEAM chemical shift imaging sequence, measurements from white matter structures in the left and right hemisphere were obtained to assess the levels of several metabolites including: Choline, NAA, Glutamate+glutamine, Myo-inositol, and Creatine+ Phosphocreatine. Statistical analyses included a series of mixed-effects regression models with Group (Autism, Healthy Controls), Lobe (Frontal, Temporal, Parietal, Occipital), and Hemisphere (Left, Right) as the fixed effects factors; Age as a covariate, and metabolite concentrations as the repeated measures dependent variable. Regression models were computed separately for each metabolite.
Results: High quality 1H MRS scans were available on 17 boys with ASD (Mean Age = 12.5, SD=1.9, range=8-15) and 17 healthy control boys (Mean Age = 11.6, SD=1.2, range=9-15). The same pattern of results emerged for each metabolite with levels being lower in youth with autism when compared to controls (Group main effect - smallest F(1, 196)=7.44, p=.007). Metabolite levels tended to remain stable or decrease with age in healthy controls but increased with age in youth with autism (Group by Age interaction – smallest F(1,196)=6.45, p=.012).
Conclusions: Lower 1H MRS metabolite levels in the deep white matter observed in the present study is consistent with previous studies identifying structural white matter abnormalities in children and adolescents with autism. These results may reflect an alteration in maturation of deep white matter structure, supporting the existence of abnormalities in long-distance fiber tracts. Developmental increases in metabolite levels in autism may reflect an ongoing compensatory process or simply result from developmental consequences of early life white matter pathology. Findings from this study should also be examined in light of grey matter abnormalities since white matter alterations are closely related to those of grey matter and vice versa. Future longitudinal 1H MRS studies are warranted to elucidate the age-related changes in white matter structures in autism and their relationship with grey matter imaging measures and clinical features.