Intestinal permeability and fecal eosinophil-derived neurotoxin are the best diagnosis tools for digestive non-IgE-mediated cow's milk allergy in toddlers

Background: Food allergy is a common problem in France involving 4%-6% of toddlers. As opposed to IgE-mediated cow's milk allergy (CMA), delayed-onset CMA, mostly, non-IgE-mediated, remains difficult to diagnose in toddlers. Our study assessed the diagnostic performances of intestinal permeability and of fecal markers, in comparison with the standard allergic work-up in children referred for CMA diagnosis. Methods: Twenty-five consecutive children, mean age (standard deviation) 6.3 months (4.8) with digestive and/or extra-digestive manifestations suggesting CMA, were prospectively studied based on a standardized allergic work-up (specific cow's protein IgE and IgG, skin prick test, atopy patch test and oral open cow's milk challenge) and digestive work-up including fecal microbiota analysis, intestinal permeability determination (urinary lactitol/mannitol ratio) and fecal markers measurement, i.e., alpha(1)-antitrypsin, tumor necrosis factor-alpha, calprotectin, beta-defensin2, secretory IgA and eosinophil-derived neurotoxin (EDN). Receiver operating characteristic (ROC) curves were calculated for all markers in order to define cut-off levels. Results: The cow's milk challenge was positive in 11 children and negative in 14. The global test performances, i.e., the number of true positive+negative cases/the total number of cases, were 76% for intestinal permeability; 72% for fecal EDN; contrasting with atopy patch test, 68%; IgE, 60%; skin prick test, 55% and IgG, 52%. Conclusions: In this routine diagnosis allergy work-up for CMA in toddlers, the best efficacy was seen for intestinal permeability compared to IgE, IgG, skin prick test and atopy patch test. Moreover, fecal EDN in a single spot sample displayed a similar performance.