• cow's milk allergy;
  • IgE;
  • patch test;
  • ROC curve;
  • skin prick test

We evaluated the value of the patch test, skin prick test, and milk-specific IgE by CAP RAST in 301 infants with suspected hypersensitivity to cow's milk. The patch test was carried out with milk powder, and the skin prick test with cow's milk-based formula. Hypersensitivity to cow's milk was determined with double-blind, placebo-controlled challenge. An immediate reaction to cow's milk challenge was observed in 100 infants (33%), a delayed reaction in 76 (25%), and a negative result in 125 (42%). Skin prick test wheals were significantly greater in infants with immediate reactions than in infants with delayed or negative reactions. Milk-specific IgE was correlated with the skin prick test (r=0.78, P<0.001, n=268) but did not contribute to further discrimination of immediate reactions from delayed or negative reactions compared to skin prick test alone. In our study population, the skin prick test (diameter ≥3 mm) showed a specificity and sensitivity of 91% and 69%; the results for milk-specific IgE (≥0.7 kU/l) were 88% and 58%, respectively. The patch test did not distinguish subjects with immediate or delayed reactions from those with negative reactions.