|
KMID : 1225720240160020179
|
|
Allergy, Asthma & Immunology Research : AAIR
2024 Volume.16 No. 2 p.179 ~ p.190
|
|
|
Predicting the Outcome of Pediatric Oral Food Challenges for Determining Tolerance Development
|
|
Kim Ha-Min
Jeong Kyung-Uk
Park Mi-Reu
Roh Yun-Young
Jung Jae-Hwa
Kim Soo-Yeon
Kim Jong-Deok
Kim Min-Jung
Kim Yoon-Hee
Sohn Myung-Hyun
Lee Soo-Young
Kim Kyung-Won
|
|
|
Abstract
|
|
|
Purpose: Despite the risk of anaphylaxis, oral food challenges (OFCs) are performed clinically for various indications, particularly to confirm tolerance development. This study aimed to assess OFCs by relevant indications and build an outcome prediction model to help determine when to perform OFCs in children who are likely to have developed immune tolerance.
Methods: In total, 432 pediatric OFCs were retrospectively analyzed according to indications. Clinical characteristics, serum total immunoglobulin (Ig) E, blood eosinophils, and specific IgE and IgG4 levels for food allergens were noted and compared. Machine learning was utilized to select the most important variables in determining the passage of the OFCs, and prediction models were constructed using the selected variables.
Results: OFCs were most commonly performed to confirm tolerance development (number, %; 267, 61.8%). The most common food allergens tested were egg (191, 44.2%) and milk (135, 31.3%). Children who passed the egg challenges for confirming tolerance acquisition had significantly lower egg white-specific IgE level (P = 0.008). Similarly, those who passed milk challenges had significantly lower cow’s milk-specific IgE (P = 0.002) and casein-specific IgE levels (P = 0.005). We developed a nomogram to predict the outcome of OFCs to determine the tolerance acquisition with the selected variables; lower food-specific IgE, higher total IgE, and younger age indicated a higher probability of passage. The area under the curve (95% confidence interval) was 0.623 (0.503?0.743) for egg and 0.734 (0.628?0.840) for milk.
Conclusions: Serum total IgE and food-specific IgE combined with age showed trends toward passing OFCs for confirming tolerance development. The constructed model may be used by clinicians as a practical guide for minimizing the risks of OFCs and a timely reintroduction for children with food allergies.
|
|
|
KEYWORD
|
|
|
Food hypersensitivity, milk hypersensitivity, egg hypersensitivity, machine learning
|
|
|
FullTexts / Linksout information
|
|
|
|
|
Listed journal information
|
|
|
|
|