Download Full Text (280 KB)
Introduction: The accepted clinical practice is to consider oral challenge when peanut IgE levels are 2.0 kUA/L or less in patients with a history of a major reaction or less than 5.0 kUA/L in patients with only sensitization. Certain populations are predisposed to having increased IgE levels, and these patients are often not offered challenges due to elevated numbers.
Methods: We reviewed unidentified patients information from January 2017 to July 2019 on peanut IgE levels and oral challenge outcomes through REDCap database. A ROC using SPSS from IBM was used and the points from the ROC curve were used to make an exponential regression curve. Since the likelihood ratio is the derivative of the ROC, the derivate of the graph was obtained.
Results: There were 106 peanut challenges (56 passed and 47 failed). The ROC points created an exponential regression of 0.417x ˇ 2 + 0.2943x + 0.0083 with an r value of 0.9072. The area under the curve was 0.702 (CI 0.601-0.803). The derivative was 0.834x+0.2943, which represents the likelihood ratio. Thus, solving the equation IgE ¼ (likelihood ratio-0.2943)/0.834.
Conclusion: It is important to remember that every patient is different, and the risk associated with oral challenge varies based on the clinical history. Using the likelihood ratio as opposed to a set threshold IgE level could enable us to better predict the possibility of a reaction with a more dynamic IgE cutoff. By using the patientspecific likelihood ratio, we were able to find an equation to allow us to calculate more meaningful IgE cutoff.
Allergy and Immunology | Pediatrics
U.G., Nalin; Portnoy, Jay; Shroba, Jodi; and Aljubran, Salman, "How to use Peanut IgE Levels to Predict Challenge Outcomes in the Pediatric Population" (2019). Posters. 277.