Background: This retrospective study was designed in order to compare the usefulness of skin prick tests, serological specific IgE tests and solid phase immunoassays in the diagnosis of infantile allergic diseases.
Methods: Two hundred infants with allergic diseases (120 infants with eczema and 80 infants with asthma), diagnosed in the Pediatric Clinic of our hospital between June of 2011 and June of 2016, were selected to participate in the study. 100 healthy infants were included in a control group. All infants received a skin prick test, a serological specific IgE test or a solid phase immunoassay. A total of 38 allergens were used for the infants receiving skin prick test. The IgE in the serum samples of infants was detected by immunoblotting. And, allergens were detected by solid phase immunoassay in the infants receiving solid phase immunoassay. The positive predictive value, negative predictive value, specificity and sensitivity for these 3 diagnostic methods were analyzed after generating a ROC curve.
Results: In the eczema group, the AUCs (the area under the ROC curve) for skin prick test, serological specific IgE test and solid phase immunoassay were 0.8685±0.0187, 0.8137±0.024 and 0.9682±0.0358, respectively. The sensitivities for skin prick test, serological specific IgE test and solid phase immunoassay were 85.70%, 83.62% and 88.72%, respectively. And, the specificities were 74%, 68% and 83%, respectively. The positive predictive values were 87.27%, 92.37% and 90.00%; and the negative predictive values were 57.04%, 51.11% and 78.15%, respectively. In the asthma group, the AUCs for skin prick test, serological specific IgE test and solid phase immunoassay were 0.8186±0.0161, 0.8037±0.0241 and 0.952±0.0421, respectively. The sensitivities for skin prick test, serological specific IgE test and solid phase immunoassay were 84.70%, 88.62% and 96.72%, respectively. The specificities were 76%, 64% and 81%, respectively; the positive predictive values were 81.34%, 85.59% and 84.54%, respectively. The negative predictive values were 69.29%, 67.14% and 70.08%, respectively.
Conclusions: While all three methods were effective in the diagnosis of infantile allergic diseases, the solid phase immunoassay was superior to the other two methods, as evidenced by the obtained values and we recommend it for clinical application as a goal standard.