In two studies of children with suspected food allergy and going through double-blind placebo-controlled food challenge, in total 19/1012(1.9%) in placebo group compared to 113/1744(6.5%) in verum group experienced worsening of atopic dermatitis without other allergy symptoms. This would translate to about 5% of children with food allergy, only reacting with worsening of atopic dermatitis. The study results were inconsistent, which warrants caution in interpretation. In the second study patients with current atopic dermatitis, only 26/272(9.6%) reacted with verum compared to 13/272(4.8%) with placebo by worsening of atopic dermatitis with or without other symptoms.
The studies also reported a risk of worsening atopic dermatitis when having other allergy symptoms. In the placebo group, in total 7/1012(0.7%) and in the verum group 92/1744(5.3%) experienced worsening of atopic dermatitis concurrent with other allergy symptoms. The results suggest that about 5% of patients with food allergy may experience worsening of atopic dermatitis concurrently with other symptoms.
| Reference | Study type | Population | Intervention and comparison | Outcomes | Risk of bias |
|---|---|---|---|---|---|
| RCT=randomized controlled trial; SR=systematic review; MA=meta-analysis; DBPCFC= double-blind, placebo-controlled food challenge; AD=atopic dermatitis | |||||
| «Ahrens B, Niggemann B, Wahn U, ym. Positive reacti...»1 | Cohort | 639 Caucasian children with suspected food allergy. | DBPCFC; 740 placebo challenges and 1472 verum food challenges with cow's milk (n 556),
hen's egg (n 387), soy (n 263), wheat (n 240), peanut (n 17), and hazelnut (n 9).
Food allergy prevalence about 49%. Challenge blocks of placebo or verum on two different days in a randomized order. |
Worsening of atopic eczema within 24 hours after the challenge. | High |
| «Roerdink EM, Flokstra-de Blok BM, Blok JL, ym. Ass...»2 | Cohort | 317 children with current atopic dermatitis, and suspected food allergy. | Clinical reactions in 272 patients with AD. All patients went through both verum and placebo challenges in randomized order. At least 2 weeks between the challenges. Food allergy prevalence 52.1%. | Exacerbation of AD during placebo vs. verum food challenge within 48 hours after the challenge. | High |
| Reference | Comments |
|---|---|
| «Ahrens B, Niggemann B, Wahn U, ym. Positive reacti...»1 | Double-blind placebo controlled-food challenges. Blinding of analysts not reported. Children could have either two of the same challenge (placebo or verum) or both or just one in some cases, in randomized order. Retrospective analysis. |
| «Roerdink EM, Flokstra-de Blok BM, Blok JL, ym. Ass...»2 | Double-blind placebo controlled-food challenges. Missing data from 14% of the participants. Retrospective analysis. Late reactions were ascertained by semistructured telephone interviews. |
| Reference | Number of challerges placebo/verum | Follow-up time | Absolute number of events (%) during placebo | Absolute number of events (%) during verum | Relative risk (95% CI) |
|---|---|---|---|---|---|
| I= intervention; C=comparison; CI=confidence interval; RR=Relative risk; AD=atopic dermatitis | |||||
| «Ahrens B, Niggemann B, Wahn U, ym. Positive reacti...»1 | 740/1472 | 24 h | 12/740 (1.6%) | 105/1472 (7.1%) | 4.40 (2.44-7.94) |
| «Roerdink EM, Flokstra-de Blok BM, Blok JL, ym. Ass...»2 | 272/272 | 48 h | 7/272 (2.6%) | 8/272 (2.9%) | 1.14 (0.42-3.1) |
| Risk ratios were not reported but calculated by the original studies from the absolute
numbers. Level of evidence: low The quality of evidence is downgraded due to risk of bias and inconsistency. |
|||||
| Reference | Number of challenges placebo/verum | Follow-up time | Absolute number of events (%) during placebo | Absolute number of events (%) during verum | Relative risk (95% CI) |
|---|---|---|---|---|---|
| I= intervention; C=comparison; CI=confidence interval; RR=Relative risks; AD=atopic dermatitis | |||||
| «Ahrens B, Niggemann B, Wahn U, ym. Positive reacti...»1 | 21/725 | 24 h | 1/740 (0.1%) | 74/1472 (5.0%) | 37.2 (5.2-267.1) |
| «Ahrens B, Niggemann B, Wahn U, ym. Positive reacti...»1 | 272/272* | 48 h | 6/272 (1.5%) | 18/272 (6.6%) | 3.0 (1.2-7.44) |
| Risk ratios were not reported by the original studies but calculated from the absolute
numbers. Level of evidence: moderate The quality of evidence is downgraded due to risk of bias. |
|||||