The aim of this study was to investigate the cumulative incidence and predictive variables of treatment failure with a whey-based extensively hydrolyzed formula (w-eHF) in children with cow's milk allergy (CMA). All children were diagnosed with CMA, using double-blind placebo-controlled food challenge (DBPCFC) with amino acid-based formula as placebo, and receive w-eHF treatment after diagnosis. Forty-nine children with CMA were included. w-eHF treatment failure was defined as incomplete resolution of original CMA symptoms upon w-eHF treatment and disappearance of these symptoms upon replacement of w-eHF with amino acid-based formula. A multiple logistic regression model was used to investigate which variables could predict treatment failure. Twenty-five (51%; 95% confidence interval (CI) 38-64%) of the children with CMA failed on w-eHF. Only "gastrointestinal discomfort" was found to contribute independently to the probability of failing w-eHF, odds ratio (95% CI) 8.994 (1.007-79.457).
Conclusions: In half of the children with proven CMA, there is incomplete resolution of symptoms upon w-eHF treatment. This study needs to be repeated including DBPCFC with w-eHF to provide more definitive diagnosis, especially since gastrointestinal discomfort seems to be the sole predictive variable for treatment failure. In the meantime, a change in formula should be considered in children with incomplete symptom resolution upon w-eHF treatment.