Blood or skin: what is best in predicting cow's milk allergy diagnosis?

Cut-off values for both skin prick tests (SPT) and specific IgE (sIgE) levels for predicting cow´s milk allergy (CMA) diagnosis are not universally defined. This study is a retrospective analysis of consecutive children (0-18 years-old) with suspected CMA tested with SPT and sIgE for cow's milk (CM) and its fractions between 2016-2017. CMA diagnosis was defined by a positive oral food challenge or a highly suggestive clinical history of CMA and SPT and/or sIgE positive to CM and/or its fractions. One hundred and five patients were included, 58% males with a median age of 2.5 (P25-P75:1-6) years and the diagnosis was confirmed in 83 patients (79%). The variables associated with CMA diagnosis were SPT with CM (p minor 0,05) and casein (p minor 0,05) and all sIgE to CM and its fractions (alfa-Lactalbumin, beta-Lactoglobulin and casein; p<0,05 for all). Optimal cut39 off points (Youden's index) for CMA diagnosis were, for the mean wheal diameter, to CM milk of 4.5mm and to casein of 3mm. For sIgE levels the optimal cut-off points were: for CM of 4.36 kUA/L, alfa-lactalbumin of 1.6 kUA/L, beta-lactoglobulin of 1.7 kUA/L and for casein of 2.6 kUA/L. The role of SPT and sIgE levels to cow´s mlik and its fractions is unequivocal in CMA follow-up. Moreover, sIgE levels seem to be more discriminatory than SPT. 

Key words 
IgE; milk allergy; sensitivity; specificity; skin tests