Allergy in Young Adults Associates with Elevated Fractional Exhaled Nitric Oxide Levels and IgE-Verified Parental Allergy but Is Confounded by Self-Reported Symptoms

The frequency of IgE-verified allergic diseases is increasing worldwide, both in westernized and developing countries, and may be due to changes in lifestyle and environment [1, 2]. The reasons for developing allergy and asthma are multifactorial, where hereditary and environmental factors cooperate [2‒8]. Environment-related risk and protective factors that influence allergy and asthma development before birth and during early life have been investigated and discussed. Several factors connected to protection against allergy development have been reported, such as a lifestyle with farm animals [9, 10], a high prevalence of breastfeeding [11, 12], having older siblings [5], and pets at home [4, 13]. In opposite, risk factors such as overweight [14, 15], smoking [10, 16], and air pollution have been reported in several studies [17, 18].

Although heredity is frequently considered as an important factor regarding allergy development in young children, less is known about the influence of the hereditary component in allergic diseases later in life. Here, birth and prospective cohort studies provide a unique opportunity to evaluate the incidence and prevalence of, as well as risk and protective factors for allergic disease, at different ages [17]. Indeed, associations between reported allergic diseases in parents and children have been investigated in previous cohort studies [19‒21]. A birth cohort reported that parental allergy was associated with the risk of having any allergy-related disease, at the age of 12 years [19]. In another study concerning asthmatics with allergic multimorbidity (rhinitis and/or eczema), maternal heredity for allergy was higher among the young adults with allergic multimorbidity than in individuals with asthma only [21].

In most large cohort studies, allergic heredity is defined based on the results from questionnaires. As our cohort of young adults, that all were born into the cohort, has well-documented data regarding the parents’ allergy confirmed by IgE sensitization, it is very well suited to explore the role of parental allergy. The aim was to investigate IgE-verified and self-reported allergy and asthma at 20 years of age in association to parental allergy and environmental factors early in life.

Parents expecting a child were recruited at regular visits at maternity care centers in southern and central Stockholm between 1997 and 2000, before the children (n = 281) were born. Parental allergic disease was characterized by medical history and confirmed with skin prick test (SPT) and specific IgE antibodies (sIgE-ab). The children were divided into three groups depending on their parents’ allergic diseases: (1) no parent with allergy, “no heredity” (NH, n = 77); (2) an allergic mother but no allergic father, “single heredity” (SH, n = 84); and (3) both parents allergic, “double heredity” (DH, n = 120). The cohort has been described in detail elsewhere [22‒25].

Several follow-ups have been performed since birth (shown in Fig. 1 and online suppl. eFig. 1; for all online suppl. material, see https://doi.org/10.1159/000539968). In the present study, we have included data from baseline, 2, 5, 10, and 20 years of age. Follow-up adherence was 94%, 86%, 78% at 2, 5, and 10 years, respectively.

For the 20-year follow-up, all individuals from the original cohort were invited in October 2019, and the data collection continued until December 2020. The participants answered a slightly modified web-based questionnaire [26], concerning allergy and asthma as well as demographic questions. All participants were invited for a clinical examination including weight, height, fractional exhaled nitric oxide (FeNO), and blood samples, after completing the web questionnaire.

The participants were characterized for IgE sensitization to common allergens at all follow-ups and classified for asthma, atopic dermatitis, food allergy, and rhinitis [25]. Definitions of allergic disease and asthma are presented in Table 1.

At the age of 2, 5, and 10 years, skin prick test (SPT) was performed for food (egg white, cod, cow’s milk, peanut, and soybean) and inhalant allergens (cat, dog, birch, timothy, and Dermatophagoides farinae) according to the manufacturer’s instructions (ALK, Copenhagen, Denmark). At 10 years of age, mugwort, horse, rabbit, and hazelnut were included. Histamine chloride and allergen diluent were used as positive and negative controls, respectively, and if the wheal was greater than or equal to 3 mm after 15 min, the test was considered positive.

sIgE-ab analyses were performed at 2, 5, 10, and 20 years, using the ImmunoCAP System (Thermo Fisher Scientific, Uppsala, Sweden). At two, five, and 10 years, the allergens selected for SPT were also analyzed for sIgE-ab. At 20 years, the participants were screened for the eleven most common foods (egg white, milk, fish, shrimp, peanut, hazelnut, walnut, pecan, cashew, pistachio, and sesame) and five most common airborne allergens (birch, timothy, horse, dog, and cat), and the test was classified as positive if the sIgE-ab levels were ≥0.35 kU_A/L. Participants were classified as IgE-sensitized if positive for at least one of the SPT and/or sIgE-ab and as allergic when one or several allergic symptom(s), asthma, atopic dermatitis, food allergy, and rhinitis were accompanied by IgE sensitization (Table 1).

FeNO, expressed as parts per billion (ppb), was measured at 10 and 20 years of age. We used ≥25 ppb as cutoff for allergic inflammation, according to the American Thorax Society (ATS), evidence-based guidelines, valid for adults [27, 28]. FeNO was performed using Circassia NIOX VERO^® (Uppsala, Sweden) in accordance with published guidelines [29].

Background characteristics were presented for the cohort at baseline and for the participants who were observed in the 20-year study population. Categorical variables were expressed as percentage of the total number that was observed together with the 95% confidence interval (CI) where applicable. To test for differences between groups regarding heredity and allergic phenotypes, the χ² test was used, p values <0.05 were interpreted as statistically significant. Comparisons between proportions of allergic disease and phenotypes over time, from 10 to 20 years old were calculated with McNemar’s test. There was no adjustment for multiplicity as this study is an exploratory study, and type-I error has not to be controlled in a strict sense. All findings are to be interpreted as exploratory findings.

Logistic regression was used to estimate odds ratios (OR) for allergic disease with the corresponding 95% CIs. Associations between allergic disease and early and present potential factors associated with the outcome variable were analyzed. Factors analyzed in this study were collected from the literature: allergic heredity, breastfeeding exclusive and partly, parents smoking at birth, pets at birth, siblings at birth, and parameters from 20 years of age; gender, smoking, body mass index (BMI), pets, and FeNO. The data were analyzed using IBM SPSS Statistics (Version 28) predictive analytics software.

At the 20-year follow-up, 73% (205/281) of the baseline cohort answered the web questionnaire and 51% (145/281) came for the clinical examination. There was a similar adherence across the heredity groups (online suppl. eTable 1). In total, 142 young adults also provided a blood sample and constituted the study population at 20 years which did not differ from the original study population regarding baseline characteristics such as breastfeeding, furred pets, siblings, daycare, sex, and parental smoking (online suppl. eTable 2). Characteristics at 20 years of age (smoking, BMI, furred pets at home, and FeNO) are presented in online supplementary eTable 3.

Among the participants, 76/142 were IgE-sensitized (54%) for one or several allergens. IgE sensitization in the heredity groups NH, SH, and DH were 30%, 50%, and 71%, respectively. Sixty-nine of the 20 years olds had no confirmed allergy up to 10 years; NH (n = 28), SH (n = 17), and DH (n = 24). A newly acquired allergy between 10 and 20 years of age in those 69 individuals was observed among 17 participants (25%); NH 11% (3/28), SH 41% (7/17), and in DH 30% (7/24) (shown in online suppl. eFig. 2).

In total, 64 (45%) 20 years olds were IgE-sensitized and reported allergic symptoms such as asthma, atopic dermatitis, food allergy, and/or rhinitis and were classified as allergic. The prevalence of allergic disease was 23% in the NH group, 40% in the SH group, and 61% in the DH group (online suppl. eTable 4).

Self-reported allergy (71%) in the questionnaires was compared with confirmed allergy and was almost twice as common as allergy where the medical history was supported by IgE sensitization (45%) (online suppl. eTable 4). Self-reported allergy was more common than confirmed allergy in all three heredity groups and highest in the non-heredity group, 60% versus 23% (online suppl. eTable 4). Regarding allergic phenotypes, the self-reported results were in accordance with confirmed allergy for food allergy (18% versus 16%), for atopic dermatitis (13% versus 10%), and for asthma (27% versus 22%). However, allergic rhinitis showed a significant discrepancy in the whole group, where 65% (n = 92) self-reported rhinitis and 42% (n = 60) were classified as having confirmed allergic rhinitis (sIgE-ab and symptoms) (Table 2). Thus, thirty individuals reported symptoms of sneezing, runny, or blocked nose after exposure to allergens and unrelated to infection without being IgE-sensitized.

Out of 60 individuals with allergic rhinitis at 20 years of age, 48 (80%) were sensitized to timothy pollen. The second allergen associated with allergic rhinitis was birch pollen to which 63% in the allergic rhinitis group had sIgE-ab. IgE sensitization to furred pets and rhinitis symptoms was seen for dog in 50% and for cat in 53%. In total, 92% of the individuals classified with allergic rhinitis were IgE-sensitized to pollen and/or furred pets of which 22 (37%) had sensitization to all four; cat, dog, birch, and timothy pollen (shown in Fig. 2). Divided into the heredity groups, there was almost no difference regarding IgE sensitization to pollen and/or furred pets between the groups; 88% in the NH group and 92% in the SH and DH groups (online suppl. eTable 5).

Since the participants in this cohort have been followed from birth to 20 years of age, the development of allergic diseases and the phenotypes, asthma, atopic dermatitis, food allergy, and rhinitis in the three different heredity groups, was investigated over time (shown in Fig. 3a–e). IgE sensitization increased from 24% to 54% and allergic diseases increased from 19% to 45% between 2 and 20 years of age. Between 5 and 10 years, the increase slowed down and then continued to increase between 10 and 20 years. The same pattern was seen for all the heredity groups (shown in Fig. 3a and in online suppl. eTable 6).

The development of different phenotypes over time, in the three heredity groups, is shown in Figure 3b–e and in online supplementary eTable 6. The prevalence of asthma increased in all heredity groups between 2 and 5 years, and thereafter decreased until 10 years of age. Between 10 and 20 years, the asthma prevalence increased only in participants with DH (shown in Fig. 3b). The prevalence for atopic dermatitis decreased in all heredity groups over time (shown in Fig. 3c) and the food allergy prevalence increased slightly over time in NH and SH groups, while it was rather stable in the DH group over time (shown in Fig. 3d). Allergic rhinitis increased the most over time, from age 2 to 20 years, in all heredity groups (shown in Fig. 3e).

The results for baseline variables during the first 2 years of life in association with allergic diseases at 20 years, such as breastfeeding exclusively at least 3 months, furred pets at home at birth, older siblings, daycare attendance at 2 years, and smoking among the parents at birth, are presented in online supplementary eTable 2. Results for variables at 20 years of age in association with allergic disease, BMI ≥25, FeNO ≥25, furred pets at home, and smoking, are presented in online supplementary eTable 3. Furred animals were present in 16% of the homes at the time of the children’s birth, compared to in 33% of the young adults’ homes. Pets in the home at baseline were most frequently reported in the NH group (26%), while the highest number of pet owners at 20 years was reported in the SH group (51%). Smoking among parents was 4% in all heredity groups at the time of the child's birth, while 13% of the 20 years olds reported current smoking, and females more frequently reported smoking than males (15% vs. 10%). The highest number of smokers at age 20 was reported in the SH group (20%). BMI ≥25 among all young adults was observed in 22%, and after stratification into heredity groups, the DH group had the highest number, 29%, compared to 21% in the NH group, and the lowest number was observed in the SH group, 12%. In the whole study population, 20% had FeNO ≥25 ppb. Differences were seen when comparing the heredity groups; 12% in the NH, 22% in the SH, and 25% in the DH had FeNO ≥25 ppb (online suppl. eTable 3). Children with an increased FeNO level (≥20 ppb) at 10 years of age more often had allergic disease at 20 years, compared to those with a low FeNO at year 10 (p = 0.001) (Table 3).

The results from a logistic regression analysis, both crude estimates (univariate logistic regression) and adjusted estimates (multivariate logistic regression), of factors associated to allergic disease at 20 years, are shown in Table 4. Crude OR for the probability of being allergic at 20 years of age were significantly higher among those with DH compared to the NH group (p < 0.001). There was also an increased risk among participants in the SH group of developing allergy at age 20, compared to the NH group, although not statistically significant. Those with FeNO ≥25 ppb at 20 years had a significantly increased risk for allergic disease, compared to those with a lower FeNO. No other variables showed any statistically significant differences.

In the multivariate logistic regression including the early life factors; breastfeeding, parental smoking at birth, pets at birth, siblings at birth, and sex, together with heredity, and factors at 20 years of age; smoking, BMI ≥25, pets, and FeNO ≥25 ppb, results show that only DH and FeNO ≥25 ppb were significantly associated with allergic disease at 20 years of age, OR 5.0 (95% CI; 2.0–12.9) and OR 7.4 (95% CI; 2.4–22.2), respectively (Table 4).

In this study, individuals born into the cohort of parents with well-characterized allergic disease, confirmed with skin prick test and sIgE-ab, have been followed longitudinally from birth to 20 years of age, regarding IgE-verified allergic diseases. At 20 years of age, the prevalence of IgE sensitization and allergic disease, where more than two-thirds had one or two allergic parents, was 54% and 45%, respectively. We observed an almost doubled frequency of self-reported allergic symptoms compared to confirmed allergy, especially for rhinitis. The most influential factor for being allergic at age 20 was having two allergic parents. In addition, we report that the 20 years olds with allergic disease were characterized by an increased FeNO. From a longitudinal perspective, IgE-verified allergic diseases increased in all three heredity groups from two to 20 years of age.

The prevalence of allergic diseases at 20 years in our repeatedly followed cohort was 23–61% depending on heredity. A direct comparison of previously published studies is difficult as these only present either self-reported allergic disease or only IgE sensitization. However, one publication shows that 25–30%, in an unselected adult population mainly from Europe, are IgE-sensitized [18]. This is a lower number than we report. A plausible explanation to this discrepancy might be due to the inclusion criteria where two-thirds of our cohort have allergic parents.

Rhinitis was the most common allergic symptom at age of 20 in all heredity groups in our study, and the most common cause of allergic rhinitis was IgE sensitization to timothy pollen followed by birch pollen, in line with previous findings in adults [30, 31]. Also, a Swedish study examined pollen-related allergic rhinitis from 4 to 24 years of age and showed that 75% had persistent disease up to 24 years of age [32].

We report a notable discrepancy between self-reported versus confirmed allergy, most prominent for rhinitis where 92 participants self-reported rhinitis compared to 60 of these having confirmed allergic rhinitis. The greatest difference was seen in the group without allergic parents. It is possible that young adults mistake a runny nose due to viral infections with allergic rhinitis during the pollen season. Another consideration is whether the pollen grain in itself leads to an allergy-like reaction in the nasal mucosa, as studies show that the pollen triggers a nonallergic inflammation when there is a high amount of pollen in the air [33]. In a cross-sectional study regarding self-experienced adverse reactions to food, there was a poor correlation between self-reported and IgE-confirmed food allergy in general. Approximately a third reported hypersensitivity reactions to foods versus a sixth were IgE-sensitized [34]. One can therefore speculate whether the prevalence of allergic diseases is overestimated if it is based on studies where the outcome is only self-reported symptoms.

In our study, only double allergic heredity was significantly associated with allergy at 20 years of age. The other risk factor for allergy at 5 and 10 years of age, being born in the winter season [25], had no longer impact at 20 years of age. We speculate that environmental factors that were important in childhood may have lost their significance for allergy development in young adults or have a low significance, which may mean that a very large number of individuals is required to see any effect. Our results are also in line with a cohort where the reported family history of allergy was the only factor associated with IgE sensitization onset at 19 years of age [35]. Further, a cohort, followed to 20 years of age, showed that family history of asthma had strong impact on persistent asthma versus transient asthma [36]. Besides maternal allergy, paternal allergy has been considered as important for the offspring. In an Austrian study of 501 adolescents at the age of 12–21 years old, an allergic father, but not an allergic mother, was linked to IgE sensitization [37].

Further, at our 20-year follow-up, we noticed that young adults with allergic diseases had significantly higher FeNO, regardless of asthma, as compared to those without allergic diseases, confirming the results observed at the 10-year follow-up in our cohort [25]. In addition, 89% with high FeNO at 10 years old had allergies at 20 years. The relation between IgE sensitization and increased FeNO has also been seen in IgE-sensitized dairy farm workers who were compared to controls from the near community. Although no differences were seen between the groups concerning asthma, the dairy farm workers who were IgE-sensitized to bovine had higher FeNO [38]. Similarly, in a cohort, IgE sensitization to peanut storage proteins in young adults was associated to elevated FeNO [39]. Also, others have seen FeNO as a robust biomarker for type 2 inflammation in asthma [40]. However, we speculate whether non-asthmatic subclinical type 2 inflammation in the airways also results in elevated FeNO, as a previous study has shown that there is a correlation between a higher expression of IL4 and IL13 and elevated FeNO [40]. We therefore speculate that FeNO above all is a marker of IgE sensitization and allergic inflammation rather than a proxy for asthma. Most of the clinical investigations at 20 years of age were concluded during fall/winter, lowering the risk of tree- or grass-pollen allergy at the time.

Newly acquired allergic diseases between 10 and 20 years of age were seen in 25% of individuals, mainly due to development of rhinitis. Divided into heredity, onset of allergic disease between 10 and 20 years was significantly more common among individuals with allergic parents compared to those with nonallergic parents. IgE sensitization over time has also been studied by others. In an American study that has analyzed how sIgE-ab against respiratory allergens changes with age, it can be seen that until the age of 13, the IgE sensitization increases and then decreases slightly up to the age of 20, which is in agreement with our data [41].

We acknowledge some limitations of our study. The cohort is relatively small and is not a general population-based cohort, as we intentionally included parents with allergic disease to a greater extent than in society in general. Also, the 20-year follow-up was affected by the COVID pandemic, which contributed to canceled appointments for study participants due to regulations in the hospital. One of the heredity groups has an allergic mother and a nonallergic father, and we were not able to study paternal allergy alone as this group was not included when this study was initiated. At that time, the in utero environment was believed to be of most importance for allergy development in the offspring. A major strength with this cohort is the thorough validation of heredity, and parents were characterized at recruiting, before the birth of the child, with sIgE-ab, and skin prick tests which needed to be consistent with reported allergic symptoms. Another strength is that the children were born into the cohort and were divided into heredity groups. We also have massive and reliable data during a 20-year period, since all data has been collected prospectively.

Findings from our well-characterized cohort strongly suggest that having two allergic parents as well as increased FeNO levels significantly increases the likelihood of being allergic at age 20 years. Specifically, rhinitis emerges as the predominant allergic phenotype that intensifies from childhood to young adulthood. However, our analysis also highlights a notable inconsistency between self-reported rhinitis and confirmed diagnoses, raising questions about the accuracy of prevalence estimates derived solely from self-report data. This underscores the importance of utilizing confirmed diagnostic methods in epidemiological studies to ensure accurate assessments of allergic disease prevalence.

We would like to thank all participants and staff in the BIAS study, Carina Wallén for the analysis of the IgE antibodies, and Jan Kowalski for advice regarding statistical analyses.

The study was approved by the Regional Ethic Committee in Stockholm (Dnr; 0–2 years – 75/97, 117/97, 5 years – 331/02, 10 years – 2007/858-31/2, 20 years – 2019/02034). The participants gave written informed consents at 20 years of age, while for earlier time points, their parents provided informed verbal consent. At the time of the initiation of the study, no written documentation of participant approval was required, which was agreed to by the Ethics Committee.

C. Nilsson reports grants to institution from Aimmune Therapeutics, a Nestlé Company. Material for IgE analyzes from Thermo Fisher used in academic studies. Lecture fees were received from: MEDA, ALK, Thermo Fisher, and GSK. Eva Sverremark-Ekström has received honoraria for lectures and a grant for another research project from BioGaia AB. No conflict of interest reported from the other authors.

This study was financially supported by Hesselmans Foundation, Stiftelsen Samariten, the Asthma and Allergy Association’s Research Foundation, the Golden Jubilee Memorial, Föreningen Mjölkdroppen, the Cancer and Allergy Foundation, the Swedish Research Council (2020-01839 and 2023-02616), the Swedish Heart-Lung Foundation, and the Swedish Order of Freemasons.

Data collection was managed by C.N. and U.H. C.N., E.S.E., A.N., and U.H. participated in the design of the study. Statistical analysis was conducted by U.H. with support from a senior biostatistician. Analysis and drafting of the manuscript were conducted by U.H. and C.N. All authors participated in the critical revision of the manuscript, provided important input, and approved the final manuscript.

Edited by: H.‐U. Simon, Bern.

The data supporting the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.