Results
Among the 200 children with AD, 85 cases(42.5%) were positive for egg SPT, 78 cases(39%)egg allergy were diagnosed by OFC . EA group had no lost or withdrew from the study,56 had eaten eggs before diagnosis,29 had simultaneous milk allergies. Follow-up duration was 18 months in 43 children (55.13%) with egg tolerance and 35 with persistent EA. In addition, 48 children in the EA group presented with respiratory symptoms, which revealed that EA was associated with an increased chance of asthma (26.9 vs 10.3%, P = 0.039) when compared to that in the AD group.
EA
In the present study, there was a 39% prevalence of EA in children with AD at 14.71 ± 7.9 months old after an EA diagnosis.According to AD severity classification, 24 cases (30.8%) were mild, 41 (52.6%) were moderate, and 13 (16.7%) were severe. The median score of the itchy skin/sleep VAS self-rating scale (0–10) was 5 (4–7). The SCORAD score (P = 0.012), severity grade (P = 0.003), and itchy skin/sleep VAS score (P = 0.012) in the EA group were higher than that in the AD group. No significant difference was observed in the distribution range of EOS% between the two groups.
In the WEA group, 12 patients (23.5%) had severe AD, and 27 (52.9%) had moderate AD. Meanwhile, in the OnlyEWA group, milder AD (40vs 57%) and moderate AD (60 vs 28%) were observed compared to that in the OnlyEWA group. In addition, we observed that the children in the WEA group had more skin damage (P = 0.045) and sleep impairment (P = 0.041) than those in the other groups.
Avoiding eggs and egg products greatly alleviated allergic symptoms in children with EA with a significant decrease.The changes of SCORAD and VAS scores in EA group were compared by Friedman-M method ,there were significant differences in different time periods(0, 3, 6, 12 and 18 months) .In addition, higher median SCORAD and VAS scores were observed at different follow-up time period in children in the WEA and OnlyEWA groups. Above all, a change in the scores was slower in children in the WEA and OnlyEWA groups than in other groups (P < 0.001).
Egg tolerance and persistent EA
No significant differences were found between the egg tolerance and persistent EA groups in age, sex, premature birth, cesarean section, exclusive breastfeeding, only child, and respiratory diseases. The EA-persistent group comprised more children from a smoking environment than the tolerant group (P < 0.05). There was no significant difference in the EA tolerance rate among children with different AD severity (i.e., mild, moderate, or severe).
Specifically, the average SPT wheal diameter at initial diagnosis (SPTdiag) in EY was 5 mm (3.25–6.875 mm) and in EW was 8.25 mm (6–10 mm) in the tolerance group.Additional follow-up results revealed that SPTdiag in the persistent group was 8 mm (6–9) in the EYA and 10.5 mm (9.25-15) in the EWA. Importantly, children om the persistent group had more persistent milk allergy (25.7 vs 9.3%), nasal symptoms (54.3 vs 30.2%), and asthma (40 vs 16.3%) than those in the tolerant group (Table 1).
Multivariate Logistic analysis was carried out again by using the factors of P < 0.1 in univariate logistic analysis, our results also showed that EW-SPTdiag (P = 0.031) and EY-SPTdiag (P = 0.023) were independent risk factors for persistent EA, with odds ratio (OR) values of 1.197 (95% confidence interval [CI]: 1.016–1.411) and 1.470 (95%CI: 1.055–2.049), respectively (Table 2). In addition, logistic analyses were conducted with persistent EA and persistent milk allergy as independent variables and respiratory disease symptoms (nasal congestion, asthma) as dependent variables. We observed that persistent EA was associated with nasal symptoms (P = 0.034) and asthma (P = 0.022), and that children with persistent EA were more likely to develop nasal congestion and asthma.
Tolerance to different egg ingredients
Egg tolerance was 86% in the OnlyEYA group, 60% in OnlyEWA group, and 49%in WEA group. The cumulative probability of allergic tolerance was calculated,there was a significant difference in the allergic tolerance rate among the three groups (P = 0.03). Specifically, patients in the OnlyEYA group were the earliest and the fastest to develop tolerance, followed by those in the OnlyEWA group and those in the WEA group (Fig. 1A).
In the WEA group, 38 children were tolerant to EY and 28 children were tolerant to EW. The tolerance rates of EY and EW were also compared. We found that patients with EYA developed a tolerance faster than those with EWA (P = 0.026) (Fig. 1B).
SPTdiag and △SPT6mo were significantly correlated with outcomes of EA
We next analyzed the relationship between the size of SPTdiag and the development of EA tolerance. The children with EA were divided into the EWA group (n = 71) and EYA group (n = 58). The larger SPTdiag for EW or EY, the lower the probability of tolerance in the future. The SPTdiag for EY and EW (P < 0.001) decreased at 0, 3, 6, 12, and 18 months of follow-up (Fig. 2). The median SPTdiag in the persistent allergy group was higher than that in the tolerated group (P < 0.05).
To calibrate the SPTdiag conducted at irregular intervals, the reduction rate of the diameter was calculated at intervals of 6 months (ΔSPT6mo), using the following formula:
(ΔSPT6mo)% = [(V0:SPT-V2:SPT) / V0:SPT] * 100,
where, V0:SPT is the first value of the mean perpendicular diameter of the SPTdiag, and V2:SPT is the value of the diameter of the SPTdiag in the second follow-up (6 months later).
Comparing the predictive power for tolerance acquisition in the EA group, ΔEY-SPT6mo (area under the curve [AUC], 0.811; 95% CI, 0.692–0.929) was better than EY-SPTdiag(AUC, 0.787; 95%CI, 0.655–0.919). The cutoff values were 7.75 mm for EY-SPTdiag and 27% for ΔEY-SPT6mo. The predictive power of using both ΔEY-SPT6mo and EY-SPTdiag (AUC, 0.750; 95%CI, 0.628–0.872) was similar compared to using EY-SPTdiag or ΔEY-SPT6mo alone (Table 3A). When the predictive power was compared in the EWA groups, ΔEW-SPT6mo (AUC, 0.836; 95%CI, 0.746–0.926) was better than that in the EW-SPTdiag (AUC, 0.722; 95%CI, 0.598–0.847). The cutoff values were 10.25 mm for EW-SPTdiag and 39.5% for ΔEW-SPT6mo. The predictive power of using both ΔEW-SPT6mo and EW-SPTdiag (AUC, 0.798; 95%CI, 0.695–0.900) concurrently was similar to using EW-SPTdiag or ΔEW-SPT6mo alone (Table 3B).
Patients were divided based on the cutoff ΔEY-SPT6mo value of 27% and ΔEW-SPT6mo value of 39.5% to perform additional analyses. Results showed that 42 patients had a ΔEY-SPT6mo ≥27%, and 38 patients (90.5%) developed EY tolerance; Of the children with a ΔEW-SPT6mo ≥ 39.5%, 30 (83.3%) developed egg tolerance; The Kaplan–Meier curves showed more developed EYA tolerance in the group with ΔEY-SPT6mo ≥ 27% (P < 0.001) or ΔEW-SPT6mo ≥ 39.5% (P < 0.001) (Fig. 3).
Discussion
Relationship between AD and EA
There is a certain degree of overlap in the allergy march.Results from a population-based cohort14 shown that > 50% of infants with AD have an FA, particularly in those with severe,persistent and early-onset eczema. Infants with eczema were 5.8 times more likely to develop egg allergy by 12 months than infants without eczema.Studies have shown that early introduction is effective in preventing the development of a food allergy in specific groups of high-risk infants 15,16.Mindlessly avoiding food can also increase the risk of malnutrition,it is important to know if you are allergic to eggs. The mechanisms by which a food allergy develops after the onset of atopic dermatitis may involve damaged skin barriers and inflammatory synergism 17. The reason why egg was found to be more closely to related to AD is still unknown. One hypothesis is that the interaction between the microbial proteins and the skin and presence of homologous proteins between microbes and egg allergen may be related to the development of IgE sensitization and egg allergy18.
Coinciding with this study, The results of the present study indicated that the prevalence of EA in children with AD was 39%, moderate and severe AD was strongly associated with EA, children with EA had more serious itching/sleep impairment, which resulted in a negative effect on daily life,and skin symptoms improved significantly after avoiding the foods containing the allergens. Although children with AD may have received routine treatment before entering this study.Some studies have mentioned that early onset AD (2 years old) is more easily relieved, and that moderate and severe AD are more likely to develop into persistent disease19. During the follow-up for 18 months the average tolerance age was 32.3 ± 8.7 months, which suggested that the age of 2–3 years was the peak period for development of EA tolerance,however,the severity of AD had no effect on the development of EA tolerance. So far, avoiding eggs is the most important method by which to reduce EA.
Risk factors for the outcome of EA
In their 2014 study, Sicherer et al. 20 have reported a cohort of children with EA aged 3 to 15 months who were followed up for a median of 74 months and have found a significant correlation between low specific-IgE levels; skin symptoms, such as urticaria or angioedema; and the EA outcome. To analyze the factors that influence EA tolerance, in the present study, no statistical differences were found in sex, exclusive breastfeeding, multiple food allergies, premature birth, cesarean section, family history of allergy, and early ingestion of eggs between the EA tolerance and persistent allergy groups.
EOSs are the main immune cells of a delayed type of food allergic reaction 21, and its number in peripheral blood is an easy index by which to determine an allergic status in the clinic. In the present study, we did not find that the using only the range of EOS% in infants and young children was of value in the diagnosis of rapid food allergy. In the clinic, for infants with only a slight increase in the proportion of EOSs, we must combine medical history, serum specific IgE value, and SPT to make a comprehensive judgment and a clear diagnosis before food avoidance.
Some studies have found that there is a strong correlation between EA in infancy and subsequent sensitivity to inhaled allergens, and persistent EA may be associated with the severity of asthma 17. we found that the prevalence of asthma was higher in children with EA, and that 49% of the cases with wheezing and chronic cough appeared after 2 years.Persistent EA is a risk factor for nasal symptoms and asthma. In the allergy march, food allergy is a risk factor for asthma regardless of whether the suspected food is tolerated, and asthma risk at the age of 4 years is two to three times higher than that for those no history of food allergy 22. Our results were consistent with reports from Rhodes et al 23, who have determined in their cohort of infants born to 100 groups of parents with a history of allergic disease that SPT is positive and EA and milk allergy are independent risk factors for adult asthma. In the present study, we found that children with persistent milk allergies were more likely to have persistent EAs, which, in part, coincided with study from Sicherer 20, which suggests that persistent milk allergy may be used as a time predictor of EA tolerance.
Kose et al. 24 have conducted a retrospective study and found that at the end of 6 months of consuming baked hen’s egg (BE), the scrambled egg tolerance was negatively associated with EW-specific IgE levels, EW SPT, and the EW prick-to-prick test (PTP). EW PTP was the most significant parameter 22. The study has found that the size and change rate of egg SPT wind mass is significantly related to the outcome of EA.SPTdiag was observed to be the risk factor for persisting EA, SPTdiag and △SPT6mo were significantly correlated with EA outcomes.
Tolerance of EA with different ingredients
Most of the children in the OnlyEWA group had moderate AD while, those in the OnlyEYA group had the least severe clinical symptoms that were relieved faster than that in the other groups. Children with EWA had more severe clinical symptoms and slower tolerance rates than children with EYA.Pérez-Rodríguez et al. 25 have found that EW protein affects the integrity of the intestinal epithelial cell–barrier function. EY plays an important adjuvant role in the Th2-type response induced by EWA, and EY may promote the sensitization of EW protein by activating innate immune cells 25. EW and EY have different antigenic components, and the tolerance rate for the different egg ingredients is also different.
During the 18-month follow-up, SPT decreased with time, which indicated that the size of the wheal is a predictor of egg tolerance. Specifically, the larger the diameter of the SPTdiagwheal, the less likely it is that egg tolerance will be developed, and the more serious and longer lasting the clinical symptoms. The cutoff value calculated by the ROC curve showed that EW-SPTdiag> 10.25 mm and EY-SPTdiag > 7.75 mm were associated with persistent EA. The decreasing rate of EW-SPT wheal ≥ 40% and EY-SPT wheal ≥ 27% in 6 months suggested the development of egg tolerance. Regardless of whether it was EW or EY, the ability of △SPT6mo to predict the egg tolerance outcome was better than that of SPTdiag.