Diagnostic testing of dogs for food hypersensitivity.

Thirteen food-allergic dogs were studied to evaluate the efficacy of feeding a commercially available egg and rice diet, intradermal skin testing, and serologic testing by ELISA for diagnosing and/or characterizing food hypersensitivity. Feeding of a home-cooked whole lamb meat and rice diet for 3 weeks, followed by challenge with each dog's regular diet, served as the standard for diagnosing food hypersensitivity. Each dog underwent provocative testing with 6 individual ingredients to determine as many of its dietary allergens as possible. Prior to skin testing and serologic testing by ELISA, most dogs had been recently exposed to the offending diet and subsequently manifested clinical signs of allergy. All dogs that tolerated the aforementioned commercial diet were exposed to it for at least 7 weeks; 84.6% of food-hypersensitive dogs ate the commercial diet with impunity. Of the 2 reactors to the commercial diet, only 1 became pruritic in response to provocation testing with chicken eggs. Low sensitivity and high specificity were found for skin testing and the ELISA, indicating a lack of true- and false-positive reactions. Neither the positive nor negative predictive values adequately predicted positive and negative reactions, respectively, for either test. On the basis of these results, the commercial diet, skin testing, and anti-IgE ELISA cannot replace an owner-prepared food elimination diet for food hypersensitivity testing in dogs.