Changes in food sensitization with changing allergy practice in Ireland

The CORAL cohort is a cohort of infants born in Ireland during the first SARS-CoV-2 mandated lockdown between March and May 2020. We hypothesized that infants born during this time would have increased rates of allergy possibly mediated through changes in the infant social and physical environment during the lockdown. In this letter, we outline the unexpectedly low incidence of peanut and egg allergy at 2 years despite higher levels of atopic dermatitis (AD) (25.3% CORAL vs. 15.5% BASELINE p < .0001) and egg sensitization at 12 months. Study participants included term, singleton infants recruited postnatally from 2 Dublin maternity hospitals. The details of the CORAL study have been published.1, 2 We have compared allergy-related outcomes with the pre-pandemic Irish BASELINE cohort recruited between 2008 and 2011.3 Food Frequency Questionnaires (FFQs) were completed at 12 and 24 months in CORAL but not in BASELINE. At recruitment, families received standard national weaning advice regarding the early introduction and maintenance in the diet of allergenic foods including egg and peanut, and this was formally reviewed at each study visit. In between research appointments, CORAL infants attended their primary care physician and public health nurse for vaccinations, developmental checks, and weaning advice. In total, 320 infants remained in the study at the 2-year point, 299 attended their appointment in person, and 292 underwent skin prick test (SPT). Each child underwent SPT to cow's milk, egg and peanut irrespective of history. If a child had a clinical history consistent with IgE-mediated food allergy, SPT to that food was completed. A child was considered food-sensitized if SPT was ≥3 mm. Food allergy at 1 year was defined as sensitization plus a history of immediate reaction or positive oral food challenge (OFC). Food allergy at 2 years was defined by positive OFC, using local PRACTALL-derived protocol. Peanut sensitization doubled between 12 and 24 months of age in the BASELINE cohort.3 While AD has been higher in the CORAL pandemic cohort,2 peanut sensitization and peanut and egg allergy are much lower at 24 months. We feel this may be explained through the combination of the widely adopted early introduction and the little studied, regular consumption of peanut, which was encouraged in the CORAL cohort. Egg allergy incidence at 2 years may reflect the use of baked egg ladders in all infants with egg allergy.1, 2 Food Frequency Questionnaire data on egg and peanut consumption are outlined in Table 1. The mean age of completion was 14.3 months for the first FFQ and 24 months for the second FFQ. Peanut introduction increased in the CORAL group from 45/354 (12%) at 6 months to 269/344 (78%) at 12 months, and 299/317 (94%) at 24 months. Children with peanut sensitization and allergy are outlined in Table 2. 3/344 (0.9%) CORAL infants were SPT positive (>3 mm) to peanut at 12-month appointment – 2 of these patients were peanut allergic (Children 1 + 2 in Table 2) and 1 was regularly tolerating peanut having successfully introduced at 7 months at home with no history of reaction (Child 3). Child 4 had a history of severe AD and egg allergy and had screening SPT in clinic, not in the study at 8 months (Peanut SPT: 5 mm) and successful introduction of peanut under medical supervision. Despite this at 14 months (FFQ 1 completed), parents reported peanut avoidance (no history of reactions) and further encouragement was given. At 24-month appointment, Child 4ʼs parents reported only once monthly ingestion of peanut without reaction. Later the same day, parents reported an immediate mild clinical reaction to peanut, and Child 4 failed a formal OFC at 25 months. Of the 2 peanut allergic children at 12 months (Children 1 + 2), one failed and one passed OFC at 24 months, bringing the point prevalence of IgE-mediated challenge proven peanut allergy at 24 months to 2/292 (0.7%). All infants who were peanut-sensitized at 12 months or had a history of reaction to peanut attended at both 12 months and 24 months. Notably, there was no other new peanut sensitization at 24 months. Of the 28/320 (9%) of infants who remained recruited in the study at 24 months but did not have SPT, no child had a history of immediate reaction to peanut. Peanut sensitization was similar between the CORAL and BASELINE cohorts at 12 months (3/344 (0.9%) vs. 15/1540 (1%); p = .9) but was significantly lower in CORAL at 24 m (2/292 (0.7%) vs. BASELINE 33/1260 (2.6%); p = .04). Peanut allergy was 60% lower in the CORAL group than the BASELINE group at 24 months, but this did not reach statistical significance (2/292 (0.7%) vs. 22/1260 (1.8%); p = .2). Despite significantly higher levels of egg sensitization at 12 months (20/344 (5.8%) vs. 49/1540 (3.7%); p = .02), at 24 months egg sensitization was similar between the CORAL and BASELINE (9/292 (3.1%) vs. 49 (3.8%); p = .4), and egg allergy was statistically significantly lower (0 vs. 37/1259 (2.9%); p < .001). All infants (11/344) with egg allergy were started on the baked egg ladder at diagnosis. A further 2/344 with egg sensitization who had never eaten egg were also started on the ladder at 12 months. At 24 months, 8/292 infants remained egg-sensitized and on the ladder. In total, 1/292 other participant developed egg sensitization (SPT: 7 mm) between 12 and 24 months. Of these 9/292 infants invited to attend raw egg OFC, 7/9 attended and all were negative. The BASELINE cohort preceded the Learning Early About Peanut (LEAP) study, which demonstrated the effectiveness of early peanut introduction in high-risk infants in the prevention of peanut allergy.4 The LEAP group ate peanut 3 times per week for 5 years. Weaning guidelines globally have changed to include the early introduction of allergenic foods including peanut. Current Irish weaning guidelines for healthcare professionals advise not delaying the introduction of common allergenic foods and notably advise the maintenance of the foods in the diet at least 3 times per week.5 National parent information does not outline the importance of maintaining allergenic foods in the diet. A recent Australian population-based, cross-sectional study with a different methodology to both CORAL and BASELINE also compared two pre- and post-guideline recommended early peanut introduction populations, HealthNuts (2007–2011) and EarlyNuts (2018–2019).6 The implementation of the new Australian national feeding guideline7 resulted in a threefold increase in peanut consumption at 12 months.8 Despite impressive uptake of the guideline, this did not translate as expected to a significant reduction in peanut allergy at 12 months (2.6% EarlyNuts vs. 3.1% HealthNuts). The EarlyNuts study aimed to assess the uptake and impact of guideline recommended allergenic food introduction on the prevalence of infant food allergy measured once at 12 months in Melbourne. The study design was cross-sectional rather than longitudinal to avoid changes in participant feeding behaviour caused by participation in the study.8 EarlyNuts infants attended no additional weaning or allergy-specific appointments before or after 12 m. In contrast, the longitudinal nature of the CORAL birth cohort study meant that CORAL families had planned interactions with an allergy-focused research team at 6 and 12 months, in addition to state-provided community supports. During these appointments and interval contacts, CORAL families were repeatedly encouraged to maintain peanut in the diet, based on both the existing national advice and emerging allergy experience of the relationship between interruption or reduced frequency of peanut consumption and increased risk of developing peanut allergy after initial successful introduction. Peanut consumption at 12 months was statistically significantly higher in the Australian cohort,8 compared with the CORAL cohort (653/737 (89%) vs. 269/344 (78%); Χ2(1, N = 1081) = 20.24, p < .001)2 The mean age of peanut introduction in the Australian cohort was 6 months, and 553/737 (75%) of EarlyNuts infants had eaten peanut more than 4 times (approx. once a month or perhaps for 1 month and then stopped) by 12 months.8 This can be compared to the 243/280 (87%) of infants in the CORAL cohort who ate peanut at least once a month at the time of the first FFQ (mean age 14.5 months) (Χ2(1, N = 1017) = 16.48 = p < .001). The EarlyNuts study did not use the same validated FFQ for peanut consumption as the CORAL study. The maintenance of peanut in CORAL infants' diets is evidenced by the consistency in frequency of peanut consumption between the first and second FFQs (Table 1). Irish national parent information does not outline the importance of maintaining allergenic foods in the child's diet. Therefore, the maintenance of peanut in CORAL infants' diets is likely secondary to regular contact with CORAL researchers. The opposite is true of the Australian EarlyNuts cohort where national parent information does advise maintaining allergenic foods which have been successfully introduced but infants did not attend face-to-face appointments with allergy-focused researchers. CORAL cohort size was limited by the time sensitivity of recruitment during the first SARS-CoV-2 lockdown. This is a limitation of the study, increasing the risk of type 2 error. However, peanut sensitization and egg allergy are both statistically significantly lower. Despite not reaching significance for peanut allergy, these findings are hypothesis generating and may inform larger studies in more generalizable, non-lockdown, populations. The notable change in egg allergy incidence at 24 months between the CORAL and BASELINE cohorts may reflect the shift towards the active management of egg allergy using baked egg ladders. We suggest that the lack of new peanut sensitization and allergy after 12 months in the CORAL cohort is due to the combination of the established and accepted early introduction and the repeated encouragement of the state's advice for persistence of peanut in infants' diets through regular contact with a clinical research team. The potential importance of regular consumption is not yet widely appreciated outside the allergy specialist field; a recently published rostrum of international infant feeding guidelines highlighted the need for further research to clarify the necessary frequency and quantity of peanut to maintain tolerance.9 We want public health strategies to include more emphasis on regular peanut consumption after the now widely adopted early introduction of peanut. SH recruited participants, collected data, analysed data, and wrote the paper. RF recruited participants, collected data, and edited the paper. DM is the Principal Investigator (PI) and curator of the BASELINE study. She provided input on CORAL study design, contributed BASELINE data to the study, and edited the paper. CV provided input into the use of food frequency questionnaires, contributed to data interpretation, and edited the paper. JH is PI of the CORAL study and is a PI of the BASELINE study. He recruited participants, collected data, contributed to data interpretation, and wrote the paper. He is the guarantor of the work. CORAL study was funded by Temple St Foundation and Clemens von Pirquet Foundation. JOBH is a board member of Clemens von Pirquet Foundation. There are no other conflicts of interest to disclose. Open access funding provided by IReL.