Colonic CD8 and gamma delta T-cell infiltration with epithelial damage in children with autism.

OBJECTIVES: We have reported colitis with ileal lymphoid nodular hyperplasia (LNH) in children with regressive autism. The aims of this study were to characterize this lesion and determine whether LNH is specific for autism. METHODS: Ileo-colonoscopy was performed in 21 consecutively evaluated children with autistic spectrum disorders and bowel symptoms. Blinded comparison was made with 8 children with histologically normal ileum and colon, 10 developmentally normal children with ileal LNH, 15 with Crohn's disease, and 14 with ulcerative colitis. Immunohistochemistry was performed for cell lineage and functional markers, and histochemistry was performed for glycosaminoglycans and basement membrane thickness. RESULTS: Histology demonstrated lymphocytic colitis in the autistic children, less severe than classical inflammatory bowel disease. However, basement membrane thickness and mucosal gamma delta cell density were significantly increased above those of all other groups including patients with inflammatory bowel disease. CD8(+) density and intraepithelial lymphocyte numbers were higher than those in the Crohn's disease, LNH, and normal control groups; and CD3 and plasma cell density and crypt proliferation were higher than those in normal and LNH control groups. Epithelial, but not lamina propria, glycosaminoglycans were disrupted. However, the epithelium was HLA-DR(-), suggesting a predominantly T(H)2 response. INTERPRETATION: Immunohistochemistry confirms a distinct lymphocytic colitis in autistic spectrum disorders in which the epithelium appears particularly affected. This is consistent with increasing evidence for gut epithelial dysfunction in autism.

Tumour necrosis factor/cachectin may mediate growth suppression in children with Trichuris colitis - abstract

As we have argued before (Cooper et al, CCMRC, 1989) the association between chronic Trichuris colitis and linear growth retardation is specific: the variability in the degree of wasting, the pattern of post-treatment catch-up growth in which weight-for-height often remains constant or actually falls and the accelerated height velocity without a change in environment, all point to the existence of some endogenous growth-suppressing factor. A candidate for this factor is the cytokine and mediator of systemic inflammation tumour necrosis factor alpha (TNF), also known as cachectin. Thirteen children with the Trichuris dysentery syndrome had a mean height-for-age of -2.3 (SD 1.34) NCHS Z-scores. Plasma TNF was assayed by ELISA. 9/13 were positive (>20 pg ml-1). Values ranged from 0 to 99 pg/ml,-1, similar to those reported in acute but uncomplicated attacks of malaria. Controls: 10 paediatric surgery follow-up patients (height Z-score + 0.1) had plasma TNF < 20 pg ml-1; in 9 stunted but parasite-free children from the community (study of Grantham-Mcgregor) TNF was undetectable. Culture of colonic mucosal mononuclear cells from four of the Trichuris colitis children yielded higher concentrations of TNF in the supernatants than those of four non-Trichuris mild, non-specific colitis patients. This suggests that the source of the increased plasma TNF may be the lamina propria macrophages, increased in number in trichuriasis as we have already described (MacDonald et al, CCMRC. 1989) (AU)