Objectives: To examine associations between fatty acid supplementation, cytokine level changes, and changes in behavior among children aged 5 to 12 years with ADOS-confirmed autism.
Methods: Plasma fatty acid and cytokine measures (TNF alpha, IL-1 beta, IFN gamma, IL-10) were analyzed pre- and post-intervention of 12 weeks of unsaturated fatty acid supplementation (n=58). These measures were correlated with changes on several behavioral scales: the Aberrant Behavior Checklist (ABC) for parents, the Pervasive Developmental Disorder Behavior Inventory (PDD-BI), and the Clinical Global Impressions (CGI) Scale. Pearson’s correlation coefficients were calculated for changes in fatty acid and cytokine levels, and changes in behaviors based on the ABC, PDD-BI, and CGI scores.
Results: Significant correlations between behavior change scores and cytokine change scores were found. Changes in IFN Gamma levels were associated with a broad range of changes in scores on both the PDD-BI and ABC, including changes in aggressiveness, arousal regulation problems, approach/withdrawal problems, expressive language, expressive social communications abilities, specific fears, learning, memory, and receptive language, receptive/expressive social communications abilities, ritualisms/resistance to change, semantic pragmatic, social approach behaviors problems, social pragmatic problems, hyperactivity/noncompliance, stereotypic behavior, and inappropriate speech. Changes in sensory/perceptual approach scores were also associated with changes in IL-1 beta levels. TNF alpha and IL-10 changes were not correlated with behavioral changes in any domain. Changes in EPA levels also significantly correlated with changes in autism composite scores.
Conclusions: The correlated changes in cytokine levels as biomarkers of inflammation with changes in behavioral scores extends previous findings that autism is associated with various immune system abnormalities. Results may provide a clue as to the mechanism of how these behavioral changes occur, and suggest that more direct methods of reducing inflammation in autism may be appropriate for further study.