Introduction
Food allergy, a growing public health concern, affects 6–8% of children and 3–4% of adults worldwide, with increasing prevalence in developed countries.1,2
It involves immune-mediated reactions to food proteins, eliciting symptoms ranging from mild cutaneous manifestations to life-threatening anaphylaxis, thereby requiring precise diagnostic and management.3
Oral food challenges (OFCs), the gold standard for diagnosing IgE-mediated food allergies, confirm clinical reactivity to allergens such as cow’s milk, egg, peanuts, tree nuts, and sesame, which are common in young children.4,5 OFCs distinguish true allergies from sensitization, assess tolerance, and guide dietary management but require careful monitoring due to the risk of inducing allergic reactions, including anaphylaxis.6,7
The European Academy of Allergy and Clinical Immunology (EAACI) guidelines recommend conducting OFCs in specialized facilities experienced in managing food challenges, guided by the patient’s history and food-specific IgE (sIgE) cut-off levels, which vary with allergen.8 Recent updates emphasize that high-risk patients, particularly those with a history of anaphylaxis, should undergo OFCs in centers equipped to handle severe reactions.9 However, these recommendations are largely based on expert consensus, with limited studies evaluating their clinical utility.
Clinical reactions during OFCs may include cutaneous (e.g., urticaria, eczema), respiratory (e.g., wheezing), gastrointestinal (e.g., vomiting), cardiovascular (e.g., hypotension), or neurological (e.g., syncope) symptoms, with severity and onset varying widely.9,10 Factors such as food type, patient age, and atopic disease history influence OFC positivity and reaction severity.11
Despite the pivotal role of OFCs in confirming IgE-mediated food allergy, data on clinical outcomes and risk factors remain limited in pediatric populations.4,8 This study investigates reaction profiles, severity, multisystem involvement, and demographic, clinical, and laboratory predictors of OFC positivity to optimize safety and diagnostic precision in children with IgE-mediated food allergies.
Materıals and Methods
Study design
We conducted a retrospective cohort study at a tertiary pediatric allergy center (from January 1 to December 31, 2023. Medical records of children with suspected or confirmed IgE-mediated food allergy who underwent OFC were reviewed, and the clinical outcomes, laboratory results, demographic data, comorbidities, reaction details, and administered treatments were evaluated.
Ethical statement
Written informed consent for OFC procedures and data publication was obtained from parents or guardians. The study was approved by the Institutional Ethics Committee (Approval No: GOA-122) and conducted in accordance with the Declaration of Helsinki.
Participants
Children aged ≥6 months undergoing OFC for diagnostic confirmation or tolerance assessment were included. Patients with incomplete documentation of pre- or post-OFC clinical, laboratory, or demographic data, or absence records of reactions and treatments, were excluded (n = 7). Patient selection and study procedures are illustrated in Figure 1.
Figure 1 Flowchart of patient selection and study procedures.
Skin prick testing and specific IgE
Skin prick tests (SPTs) and prick-to-prick tests were performed with fresh foods (milk, egg, wheat, fish, walnut, sesame, hazelnut, peanut, pistachio, cashew, and lentils). A wheal diameter ≥3 mm and larger than the negative control was considered a positive SPT or prick-to-prick test. Specific IgE levels ≥0.35 kUA/L were considered positive (IMMULITE 2000, Siemens Healthcare Diagnostics, Tarrytown, NY, USA).4
Oral Food Challenge Protocol
Oral food challenges were performed in accordance with the PRACTALL consensus and national guidelines.4 Antihistamines and interfering medications (e.g., tricyclic antidepressants) were discontinued ≥7 days prior; asthma controller therapies (inhaled corticosteroids, leukotriene receptor antagonists) were allowed to continue.8
Increasing doses of dietary protein (3, 10, 30, 100, 300, 1000, and 3000 mg) were administered at 15- to 30-min intervals.
Challenges were terminated upon the onset of objective symptoms (e.g., urticaria, angioedema, vomiting, respiratory distress) or persistent subjective symptoms (e.g., throat tightness, abdominal pain).4
Allergic reactions were managed per symptom severity: cutaneous symptoms with age-appropriate antihistamines,9 and anaphylaxis with immediate epinephrine, with emergency resuscitation equipment available.12
Data collection
Variables included age, sex, age at diagnosis, initial symptoms, allergenic foods, concomitant atopic diseases, family history of atopy, OFC purpose (diagnostic versus Tolerance assessment), reaction type and severity, and treatments administered.
Statistical analysis
Analyses were performed using IBM SPSS Statistics v22.0 (IBM Corp., Armonk, NY, USA). Continuous variables were reported as median (IQR) for non-normal distributions (Shapiro–Wilk test) or mean ± SD for normal distributions. Categorical variables were expressed as frequencies (%). Between-group comparisons used χ2 or Fisher’s exact test (categorical) and Mann–Whitney U or Kruskal–Wallis test (continuous). Univariate and multivariate logistic regression identified predictors of OFC positivity. Variables with P < 0.10 in univariate analysis were entered into the multivariate model. Results were reported as odds ratios (OR) with 95% confidence intervals (CI), and P < 0.05 was considered statistically significant.
Results
Patient characteristics
A total of 310 OFCs were performed in 211 children with suspected or confirmed IgE-mediated food allergy (63.0% males; median age at OFC: 21 months [IQR: 14–37.3]). Multifood allergy was present in 30.8%. Median age at initial presentation was 9 months (IQR: 5–14.7), with a median follow-up of 11 months (IQR: 1.2–21.8). Initial symptoms included eczema (71.6%), urticaria (20.4%), and anaphylaxis (8.1%). Concomitant atopic conditions comprised asthma (16.6%) and allergic rhinitis (2.4%); family history of atopy was reported in 35.5% (Table 1).
Table 1 Demographic and clinical characteristics of patients and OFC results.
| Value | |
|---|---|
| Gender, n (%) | |
| Female | 78 (37.0) |
| Male | 133 (63.0) |
| Age at first visit (months), median (IQR) | 9 (5–14.7) |
| Age at OFC (months), median (IQR) | 21 (14–37.3) |
| Follow-up duration (months), median (IQR) | 11 (1.2–21.8) |
| Symptoms at first visit, n (%) | |
| Eczema | 151 (71.6) |
| Urticaria | 43 (20.4) |
| Anaphylaxis | 17 (8.1) |
| Concomitant atopic disease, n (%) | |
| Asthma | 35 (16.6) |
| Allergic rhinitis | 5 (2.4) |
| Family history of atopy, n (%) | 75 (35.5) |
| Food allergy, n (%) | |
| Single food allergy | 146 (69.2) |
| Multiple food allergies | 65 (30.8) |
| Eosinophil count (cells/µL), median (IQR) | 270 (190–470) |
| Eosinophil percentage (%), median (IQR) | 3.1 (2.1–5.0) |
| Total IgE (kU/L), median (IQR) | 60.6 (17.6–156.5) |
| Food-specific IgE at first visit (kU/L), median (IQR) | |
| Milk | 4.07 (1.2–9.7) |
| Casein | 0.71 (0.35–4.1) |
| Egg white | 3.08 (0.98–9.03) |
| Egg yolk | 1.72 (0.77–4.6) |
| Wheat | 4.59 (0.49–6.9) |
| Tree nuts (walnut, hazelnut, pistachio, cashew) | 1.07 (0.58–4.35) |
| Foods tested in OFC, n (%) | |
| Egg | 147 (47.5) |
| Milk or dairy products | 113 (36.5) |
| Tree nuts (walnut, hazelnut, pistachio, cashew) | 23 (7.5) |
| Legumes | 21 (6.7) |
| Seafood | 7 (2.3) |
| Aim of the OFC, n (%) | |
| Diagnosis of food allergy | 82 (26.5) |
| Tolerance development | 228 (73.5) |
| OFC results, n (%) | |
| Positive | 38 (12.3) |
| Negative | 272 (87.7) |
IQR: Interquartile Range; OFC: Oral Food Challenge; IgE: Immunoglobulin E. Values are presented as n (%) for categorical variables and median (IQR) for continuous variables.
Of the OFCs, 82 (26.5%) were diagnostic (suspected sensitization based on history and positive SPT/sIgE), and 228 (73.5%) were for tolerance assessment in previously diagnosed patients. The most frequently challenged foods were eggs (47.5%), milk or dairy (36.5%), tree nuts (7.5%), legumes (6.7%), and seafood (2.3%) (Table 1).
Oral food challenge outcomes
The positivity rate of OFCs was 12.3% (38/310). Diagnostic OFCs had a higher positivity rate (18.3% [15/82]) than tolerance-assessment OFCs (10.1%, P = 0.048). The most common elicitors were milk or dairy (44.7%), tree nuts (23.6%), and eggs (18.4%), followed by legumes (5.3%), sesame (5.3%), and fish (2.6%) (Figure 2).
Figure 2 Distribution of foods associated with positive oral food challenge results.
Reactions occurred at a median cumulative protein dose of 43 mg (IQR: 13–43) and median step 3 (IQR: 2–4). Lower thresholds were observed for peanut or sesame (8 mg, Step 2) and fish (3 mg, Step 1). Symptom onset was within 1 h in 60.5% and 1–6 h in 39.5%. Cutaneous symptoms predominated (94.7%), followed by gastrointestinal (10.6%), respiratory (7.9%), cardiovascular (7.9%), and neurological (5.3%) manifestations (Table 2).
Table 2 Characteristics of patients with positive OFC results.
| Value | |
|---|---|
| Time to OFC positivity (minutes), median (IQR) | 42.5 (15–70) |
| Step of OFC positivity, median (IQR) | 3 (2–5) |
| Foods associated with positive OFC, n (%),Cumulative protein (mg) median (IQR), Positive Stage, median (IQR) | |
| Milk or dairy products | 17 (44.7) |
| Tree nuts (walnut, hazelnut, pistachio, cashew) | 9 (23.6) |
| Egg | 7 (18.4) |
| Legumes | 2 (5.3) |
| Sesame | 2 (5.3) |
| Fish | 1 (2.6) |
| Foods Cumulative protein (mg) median (IQR)/ Step of OFC median (IQR) | |
| Milk or dairy products | 43 mg (13–143)/3 (2–4) |
| Walnut | 43 mg (13–143)/3 (2–4) |
| Hazelnut | 43 mg (3–43)/3 (1–3) |
| Egg | 43 mg (13–43)/3 (2–3) |
| Legumes | 8 mg (3–13)/2 (1–2) |
| Sesame | 8 mg (3–13)/2 (1–2) |
| Fish | 3 mg (3–3)/1 (1–1) |
| Anaphylaxis, n (%) | 5 (13.2) |
| Symptoms during positive OFC, n (%) | |
| Cutaneous | 36 (94.7) |
| Pruritus | 29 (76.3) |
| Urticaria | 16 (44.4) |
| Angioedema | 3 (8.3) |
| Respiratory | 3 (7.9) |
| Cough | 3 (50.0) |
| Tachypnea | 3 (50.0) |
| Gastrointestinal | 4 (10.6) |
| Nausea/vomiting | 3 (75.0) |
| Abdominal pain | 1 (25.0) |
| Neurological | 2 (5.3) |
| Syncope (with normal blood pressure) | 2 (100.0) |
| Cardiovascular | 3 (7.9) |
| Tachycardia | 1 (33.3) |
| Bradycardia | 1 (33.3) |
| Hypotension | 1 (33.3) |
IQR: Interquartile Range; OFC: Oral Food Challenge. Values are presented as n (%) for categorical variables and median (IQR) for continuous variables. Time to OFC positivity is measured in minutes.
Serious reactions
Anaphylaxis occurred in 13.2% (5/38) of positive OFCs, triggered by milk or dairy (n = 4) and walnut (n = 1). Notably, 60% (3/5) had comorbid eczema, and all had sIgE <15 kUA/L and SPT wheal ≤9 mm, which highlighted the limited predictive value of sensitization markers for severe reactivity.
Predictors of OFC positivity
Oral food challenge–positive patients were older (median: 25.5 vs 20 months; P = 0.009) and had longer follow-up durations (median: 12 vs 11 months; P = 0.046). Initial presentation with urticaria (55.3% vs 8.1%; P < 0.001) or anaphylaxis (13.2% vs 4.4%; P < 0.001) was associated with OFC positivity, whereas eczema was more common in OFC-negative cases (31.6% vs 87.5%; P = 0.043). No significant differences were observed in sex, concomitant atopic diseases, total IgE levels, or eosinophil counts (all P > 0.05) (Table 4). Multivariate logistic regression identified milk or dairy (OR: 3.326, 95% CI: 1.437–7.702, P = 0.005), sesame (OR: 7.022, 95% CI: 1.194–41.311, P = 0.031), and hazelnut (OR: 11.286, 95% CI: 3.858–33.017, P < 0.001) as independent predictors of OFC positivity.
Table 3 Clinical features of patients with anaphylaxis during OFC.
| Case 1 | Case 2 | Case 3 | Case 4 | Case 5 | |
|---|---|---|---|---|---|
| Age at OFC (months) | 38 | 24 | 9 | 18 | 44 |
| Symptom at first visit | Eczema | Urticaria | Eczema | Eczema | Urticaria |
| Follow-up duration (months) | 30 | 18 | 5 | 14 | 33 |
| Age at first visit (months) | 8 | 6 | 4 | 4 | 11 |
| Aim of the OFC | Tolerance development | Tolerance development | Diagnosis | Tolerance development | Tolerance development |
| Single/Multiple food allergy | Multiple | Multiple | Single | Multiple | Multiple |
| Food allergens tested | Cow’s milk, Egg | Cow’s milk, Egg, Sesame | Cow’s milk | Cow’s milk, Egg, Legumes | Cow’s milk, Egg, Legumes, Tree nuts |
| OFC food tested | Cow’s milk | Cow’s milk | Yoghurt | Yoghurt | Walnut |
| sIgE (kIU/L) | Cow’s milk: 4.4, Casein: 0.8 |
Cow’s milk: 1.8, Casein: 1.2 | Cow’s milk: 4.8, Casein: 0.3 | Cow’s milk: 11.8, Casein: 1.9 | Walnut: 1.6 |
| SPT (mm) | Cow’s milk: 4 | Cow’s milk: 6 | Cow’s milk: 9 | Cow’s milk: 6 | Walnut: 4 |
| Step of OFC | 2 | 1 | 3 | 3 | 1 |
| Cumulative dose protein (mg) | 13 | 3 | 43 | 43 | 3 |
| Symptoms | Urticaria, Cough, Tachypnea, Vomiting | Urticaria, Cough, Tachypnea, Tachycardia | Urticaria, Vomiting, Syncope | Urticaria, Bradycardia, Hypotension, Syncope | Urticaria, Cough, Tachypnea |
| Grade of anaphylaxis | Grade 2 | Grade 2 | Grade 3 | Grade 3 | Grade 2 |
| Medications administered | Adrenaline, Salbutamol, Antihistamine, IV hydration, Ondansetron | Adrenaline, Salbutamol, Antihistamine | Adrenaline, IV hydration |
Adrenaline, IV hydration |
Adrenaline, Salbutamol, Antihistamin |
OFC: Oral Food Challenge; sIgE: Specific Immunoglobulin E; SPT: Skin Prick Test; IV: Intravenous. Values are presented as n (%) for categorical variables. Anaphylaxis grades are based on Sampson’s clinical severity criteria. Anaphylaxis diagnosis was based on EAACI guidelines.9
Table 4 Comparison of patients with positive and negative OFC results.
| Characteristic | OFC-Positive Patients (n=38) | OFC-Negative Patients (n=272) | p-value |
|---|---|---|---|
| Male, n (%) | 23 (60.5) | 170 (62.5) | 0.814 |
| Age at OFC (months), median (IQR) | 25.5 (19–47.5) | 20 (14–34) | 0.009* |
| Age at first visit (months), median (IQR) | 12 (6–20) | 9 (5–14) | 0.213 |
| Follow-up duration (months), median (IQR) | 12 (8–29) | 11 (1–19) | 0.046* |
| Food allergy, n (%) | |||
| Single food allergy | 22 (57.9) | 174 (64.0) | 0.467 |
| Multiple food allergies | 16 (42.1) | 98 (36.0) | |
| Symptoms at first visit, n (%) | |||
| Eczema | 12 (31.6) | 238 (87.5) | 0.043* |
| Urticaria | 21 (55.3) | 22 (8.1) | <0.001* |
| Anaphylaxis | 5 (13.2) | 12 (4.4) | <0.001* |
| Asthma, n (%) | 5 (13.2) | 43 (15.8) | 0.666 |
| Allergic rhinitis, n (%) | 1 (2.6) | 7 (2.6) | 1.000 |
| Family history of atopy, n (%) | 17 (44.7) | 89 (32.7) | 0.144 |
| Total IgE (kU/L), median (IQR) | 70 (34–158.3) | 58 (14.9–157.5) | 0.189 |
| Eosinophil count (cells/µL), median (IQR) | 270 (195–445) | 270 (187.5–490) | 0.835 |
| Eosinophil percentage (%), median (IQR) | 3.7 (2.5–5.7) | 3 (2–4.9) | 0.244 |
*IQR: Interquartile Range; OFC: Oral Food Challenge; IgE: Immunoglobulin E. Values are presented as n (%) for categorical variables and median (IQR) for continuous variables. p<0.05 indicates statistical significance.
Discussion
This analysis of 310 OFCs in 211 children yielded valuable insights into pediatric IgE-mediated food allergy outcomes. The positivity rate of 12.3% is lower than the 33% reported by Abrams et al.,13 which likely stems from the inclusion of baked or fermented milk and eggs (often tolerated due to protein denaturation14) and the exclusion of patients with recent severe reactions. The most frequently challenged foods, egg (47.5%), milk or dairy (36.5%), and tree nuts (7.5%), align with prevalent pediatric allergens.6
Oral food challenge–positive children were older (P = 0.009) and had longer follow-up (P = 0.046), consistent with delayed resolution of milk and egg allergies. Initial urticaria (P < 0.001) or anaphylaxis (P < 0.001) strongly predicted positivity, corroborating links between prior severe reactions and persistent clinical reactivity.11 Conversely, eczema was associated with OFC negativity (P = 0.043), likely reflecting the predominance of baked milk or egg challenges in this subgroup, where sensitization rarely translates to clinical allergy.15 This contrasts with Jacob et al.,16 who reported increased reactivity risk with eczema (OR 1.99, particularly to peanut [OR 3.2]), underscoring allergen- and context-specific effects that warrant further study.
Anaphylaxis occurred in 13.2% (5/38) of positive OFCs (milk/dairy: n = 4; walnut: n = 1), despite sIgE <15 kUA/Land SPT wheal ≤9 mm in all cases, reinforcing the poor predictive value of sensitization markers for severe reactions.4,10 Notably, 60% (3/5) had comorbid eczema, aligning with Perry et al.’s 11% anaphylaxis rate in low-sIgE patients.17 These observations highlight the indispensable role of OFCs in unmasking latent risks and the imperative for immediate epinephrine access and trained personnel.7,18
Multivariate logistic regression identified milk/dairy (OR 3.326, 95% CI 1.437–7.702, P = 0.005), sesame (OR 7.022, 95% CI 1.194–41.311, P = 0.031), and hazelnut (OR 11.286, 95% CI 3.858–33.017, P < 0.001) as independent predictors of OFC positivity. The pronounced associations with sesame and hazelnut reflect regional dietary exposure in the Eastern Mediterranean, where early and frequent consumption sustains reactivity.19 These findings parallel Brough et al.’s multicenter data on elevated risks for sesame and tree nuts20 and underscore hazelnut’s persistence as a high-risk allergen.19-21
The clinical implications of our findings are significant for allergy management. OFCs not only confirm diagnoses but also prevent unnecessary dietary restrictions, improving the quality of life for children and families.1 By identifying the determinants of OFC positivity, clinicians can customize protocols, prioritize closer monitoring of children with a history of severe reactions, and consider testing at earlier stages to confirm tolerance in those with eczema.
These insights are particularly valuable in regions such as the Middle East, where food allergy management is evolving, and standardized protocols are needed to address the rising prevalence.10
Limitations include the retrospective, single-center design, which potentially introduces selection bias, mitigated by standardized protocols and limiting generalizability beyond Mediterranean diets. Reliance on prick-to-prick testing may reduce reproducibility, though it mirrors real-world practice. The small anaphylaxis cohort (n = 5) precludes robust severe reaction modeling. Multicenter prospective studies are essential in validating these predictors and refining OFC strategies.
Conclusion
Oral food challenges remain the gold standard for diagnosing and managing pediatric IgE-mediated food allergies. This study demonstrates that severe reactions, including anaphylaxis, can occur despite low sensitization markers. Milk, sesame, and hazelnut are identified as the robust predictors of reactivity, particularly in regions with high exposure. Cutaneous symptoms predominated, yet multisystem involvement underscored reaction unpredictability. These findings reinforce the need for risk-stratified, vigilantly monitored OFCs to ensure safety, facilitate tolerance confirmation, and personalize management, ultimately improving outcomes for children and families.