Intestinal cytokines in children with pervasive developmental disorders

OBJECTIVES: A relationship between autism and gastrointestinal (GI) immune dysregulation has been postulated based on incidence of GI complaints as well as macroscopically observed lymphonodular hyperplasia and microscopically determined enterocolitis in pediatric patients with autism. To evaluate GI immunity, we quantitatively assessed levels of proinflammatory cytokines, interleukin (IL)-6, IL-8, and IL-1beta, produced by intestinal biopsies of children with pervasive developmental disorders. METHODS: Fifteen patients, six with pervasive developmental disorders and nine age-matched controls, presenting for diagnostic colonoscopy were enrolled. Endoscopic biopsies were organ cultured, supernatants were harvested, and IL-6, IL-8, and IL-1beta levels were quantified by ELISA. Tissue histology was evaluated by blinded pathologists. RESULTS: Concentrations of IL-6 from intestinal organ culture supernatants of patients with pervasive developmental disorders (median 318.5 pg/ml, interquartile range 282.0-393.0 pg/ml) when compared with controls (median 436.9 pg/ml, interquartile range 312.6-602.5 pg/ml) were not significantly different (p = 0.0987). Concentrations of IL-8 (median 84,000 pg/ml, interquartile range 16,000-143,000 pg/ml) when compared with controls (median 177,000 pg/ml, interquartile range 114,000-244,000 pg/ml) were not significantly different (p = 0.0707). Concentrations of IL-1beta (median 0.0 pg/ml, interquartile range 0.0-94.7 pg/ml) when compared with controls (median 0.0 pg/ml, interquartile range 0.0-60.2 pg/ml) were not significantly different (p = 0.8826). Tissue histology was nonpathological for all patients. CONCLUSIONS: We have demonstrated no significant difference in production of IL-6, IL-8, and IL-1beta between patients with pervasive developmental disorders and age-matched controls. In general, intestinal levels of IL-6 and IL-8 were lower in patients with pervasive developmental disorders than in age-matched controls. These data fail to support an association between autism and GI inflammation.