Environmental exposure and sensitization patterns in a Swiss alpine pediatric cohort

Background The level of environmental exposure throughout life may contribute to the prevalence of allergic sensitization and allergic disease. The alpine climate has been considered a healthy climate with little allergen exposure and pollution. We conducted a cross-sectional study to investigate local environmental exposure and concomitant prevalence of allergic sensitization among local school children born and raised in an alpine environment. Methods Clinical and demographic data were collected with a questionnaire. Allergen content was assessed in residential settled dust samples, lifetime exposure to pollen and air pollution was calculated using data from national pollen and air pollution monitoring stations, and the allergic sensitization profile was determined with component resolved diagnostics (ISAC®). Univariate and multivariate regression models were used to estimate the relation between exposure and sensitization. Results In a cohort of children born and raised in an alpine environment, sensitization to aeroallergens is quite common (38%), especially to grass (33%) and cat (16%). House dust mite allergen was detected in up to 38% of residential dust samples, but sensitization to HDM was low (2.5%). Pollutant levels were low, but an increasing trend was observed in the amount of ozone and PM10. Living close to a busy road was associated with increased odds OR (95% CI) for being sensitized to any allergen 2.7 (1.0–7.2), to outdoor allergens 2.8 (1.1–7.1) and being sensitized plus reporting symptoms of rhinoconjunctivitis 4.4 (1.3–14.8) and asthma 5.5 (1.4–21). Indoor living conditions, including the presence of visible mold, increased the odds of being sensitized to indoor allergens (1.9 (1.1–3.2) and being sensitized plus reporting symptoms of rhinoconjunctivitis 1.9 (1.0–3.6) and asthma 2.1 (1.0–4.1). Conclusion In a healthy alpine environment, pollution might still be an important factor contributing to allergic sensitization.


INTRODUCTION
2][3] Environmental exposure throughout life, or exposome, likely contributes to this increase. 4The exposome consists of the general external environment (climate, biodiversity, urbanrural, social and economic factors), the specific external environment (diet, physical activity, allergens, microbes, pollutants) and the host dependent internal environment (age, metabolic factors, inflammation, oxidative stress). 5,6The exposome is able to modulate the development of allergic sensitization. 7There is considerable variation in allergic prevalence between countries, within countries and even within cities, indicating the importance of local environmental characteristics. 2iming and duration of environmental exposure are important determinants of the development and progression of allergic diseases. 8veral significant risk factors for the development of allergic disease have been identified, such as indoor allergen exposure and the abundance and diversity of microbial exposure during the first years of life. 4,9,10Indoor allergen exposure levels can be influenced by several environmental, socioeconomic and demographic factors and previous studies have found large differences in allergen exposure between groups. 8The abundance and diversity of microbial exposure during the first years of life have been linked with the risk of allergic sensitization. 92][13] However, there is a complex interplay between the different exposome indoor and outdoor environmental factors, host factors, timing, and location of exposure and subsequent development or exacerbation of allergic disease and no single factor has been identified that dominates disease development in every individual. 8e alpine climate has been considered a healthy climate with little exposure to allergens and little pollution.The Swiss municipality Davos is located at 1560 m altitude in the Alps and its climate is characterized by low presence of aeroallergens such as house dust mites (HDM), pollen and fungal spores and low amounts of TRAP such as particulate matter (PM10), nitrogen oxides (NOx) and ozone (O3). 14he pollen season in Davos can be short and clinically relevant pollen concentrations are sometimes observed only a few days per year.It has been generally accepted that with increasing altitude and decreasing humidity, HDM allergen concentration decreases.So far, few studies have investigated local environmental exposure in the alpine environment together with allergic sensitizations in its local population.
This cross-sectional study aims to investigate the relation between environmental exposure and prevalence of allergic sensitization among school children born and raised in the Swiss alpine municipality Davos.

Study design and participants
We conducted a cross-sectional study in the alpine municipality Davos, Switzerland at 1560 m altitude.School children aged 5-9 years attending school (kindergarten to second grade) in the municipality of Davos were asked to participate in the study.Exclusion criteria were not born in Davos or having been away from Davos for longer than 6 months.Participants were included in the study between October 2017 and May 2019.Written informed consent was obtained from the legal guardians.

Clinical and demographic characteristics
Clinical and demographic data was collected with the use of a questionnaire, filled out by the legal guardians of the child.The questionnaire included questions on sex, age, socioeconomic status and others.Data was collected on the birth of the child, breastfeeding, number of siblings, general health of the child, use of antibiotics, vaccination status, day care attendance, and time spent outdoors.Residential characteristics such as type of residence, total living area, number of persons living in the home, proximity to a busy road, and type of heating were collected.Furthermore, data was collected on exposure to smoke, exposure to animals, presence of pets in the home, washing and changing of bed linens, and family history of atopy.
Presence of atopic disease was assessed with the ISAAC questions for asthma, rhinitis, atopic dermatitis and food allergy.Furthermore doctordiagnosed allergies or atopic diseases were recorded, as well presence of allergies and intolerances by recording use of antihistamines, previous allergy tests, possession of an anaphylaxis emergency kit or previous immunotherapy, subcutaneous immunotherapy/sublingual immunotherapy (SCIT/SLIT).

Allergen content in residential settled dust samples
Individual house dust samples were collected by study participants using a commercially available DUSTREAM Collector (Indoor Biotechnologies, Charlottesville, VA, USA) attached to a household vacuum cleaner.Participants were asked to vacuum the mattress, pillow and floor of the bedroom where the child sleeps, as well as the couch and playroom or other rooms where the child spends much time.Dust extracts were prepared by removing large particles/hairs and dissolving 100 mg of dust in 2 ml of phosphatebuffered saline pH 7.4, 0.05% Tween-20.Samples <10 mg of dust were not considered for further analysis.Proteins were extracted by incubation on a roll mixer for 2 h at room temperature following a centrifugation step at 13,000Âg for 20 min.Supernatants were stored at À80 C until further processing.For allergen content analysis, the dust extracts were thawed and centrifuged again.Supernatants were diluted 1:10, 1:100 and 1:10,000 and examined in a Multiplex Array for Indoor Allergens (MARIA, Indoor Biotechnologies, Charlottesville, VA, USA) using xMAP Technology (Luminex, Austin, TX, USA).The array uses fluorescently labeled beads conjugated to monoclonal antibodies specific for purified allergen molecules.The allergen concentration of Der p 1, Der p 2 and Der f 1 allergens (house dust mite allergens), Mus m 1 (mouse allergen), Asp f 1 and Alt a 1 (Aspergillus and Alternaria mold allergen) as well as Ara h 6 (peanut allergen) was investigated.A 12-point standard curve executed in duplicates was used to quantify the results using the 1:10 dust extract dilution.Additionally, quality controls provided with the test kit were applied.Measurements of fluorescence were performed in the Bio-Plex 200 System (BioRad Laboratories, Hercules, CA, USA).For samples below the lower limit of detection no quantification was calculated, but a present/notpresent was calculated when a sample gave an OD above 3 S.D. of the blank.

Exposure to weather, pollen, and pollution
Davos is located at the bottom of a valley.Predominant wind directions for this city are 0-45 and 180-225 , reaching 1022 mm as annual rate of precipitation and having an average mean temperature of 3.5 C. 14 Meteoswiss operates a national pollen monitoring network and data from the station (PDS) at 1587 m altitude at Davos Wolfgang was collected. 14aily pollen data (pollen grains/m 3 ) for hazel, alder, birch, plane, grass, plantain, mugwort, and ambrosia were collected for each child from the month of birth until blood was collected to calculate cumulative exposure measures.
The federal office for the environment collects data on air pollution through the national air pollution monitoring network. 15Daily pollution measurements (mg/m 3 ) regarding NO, NO 2 , NOx, O 3 , PM 2.5 and PM 10 were collected from the measurement stations DPR 46.79,654 N, 9.82240 E, 1556.8 m (until December 2017) and DBP 46.79,304 N, 9.82088 E, 1556.5 m (from December 2017) in Davos Platz.Welch's t-test showed no differences between the stations and data was combined for the whole time period.

Sensitization profile
Study participants provided a capillary or venous blood sample.Blood samples were obtained from the fingertip, incubated at room temperature and centrifugated for 5-10 min at 4350 rpm.Serum was separated and stored at À20 C until further analysis.
Analysis of sera for specific IgE to single purified allergen components was done by means of the allergen multiplex array ImmunoCAP ISAC (Thermo Fisher Scientific, Uppsala, Sweden).According to the manufacturer's protocol, the test was performed with 30 ml of serum (Protocol No. Sensitization to aeroallergens was divided into indoor allergens (dog, cat, mouse, HDM, cockroach, Alternaria) and outdoor (grass, tree, weed, horse) allergens.
Multiple logistic regression analysis was used to estimate the relation between exposure and sensitization, corrected for the confounding factors family history of atopy and sex.Statistical analyses were performed with IBM SPSS statistics for Windows version 23 and GraphPad Prism 7 for Windows (GraphPad Software, Inc, La Jolla, CA, USA).Pollen and pollution graphs and the statistical analysis were carried out with R aeRobiology package (R Core team 2014).Two-sided p values < 0.05 were considered statistically significant.No multiplicity adjustment was used.

Study participants
A total of 131 children participated in the study, of whom 121 also donated blood (Fig. 1).In total 99 dust samples were analyzed, some study participants did not provide a sample, two samples were excluded because not enough dust Fig. 1 Flowchart of study participants and provided blood and dust samples was provided, 25 samples were excluded because their siblings participated in the study and the sample would represent the same household.
Table 1 shows the demographic characteristics of the study population.The mean age of the study participants was 6.8 years and half were girls.Most had at least one sibling, a third were born with c-section.Median breastfeeding time was 8 months.Almost half had pets in the home and a quarter reported living close to a busy road.Almost two thirds had regular exposure to animals.Half of the study population spent more than 2 h outdoors each day.Almost two thirds had atopic family members and one third reported current or previous visible mold in the home.Socio-economic position was calculated using educational level of both parents, number of persons in the household, number of rooms in the house and marital status (Supplementary Table 2).

Indoor allergens
We found detectable levels of the house dust mite allergen Der p 1 in 17%, Der p 2 in 37%, Der f 1 in 38%, Mus m 1 in 50%, Asp f 1 in 6%, Alt a 1 in 3% and Ara h 6 in 99% of households (Supplementary Figure 1).The median of detectable HDM samples was 74 (16-4960) ng/g for Der p 1, the median of detectable Der p 2 was 18 (2.1-1620), and the median for Der f 1 was 29 (7-2460) ng/g.For mouse allergen Mus m 1, the median of positive samples was 2.7 (0.4-36) ng/g.The mold allergens Alt a 1 and Asp f 1 had a median of 3.3 (2.4-3.4)ng/g (3 positive samples) and a median of 6 ng/g (3.3-50) (6 positive samples) (Fig. 2).For peanut allergen Ara h 6, all samples but 1 were positive and the median was 432 (13-11,500) ng/g (Supplementary Figure 2).Exposure to multiple allergens was present in the majority of households (Supplementary Figure 3).In 27% of households the peanut allergen Ara h 6 was the only detected allergen.In 31% of households 2 allergens were detected, in 18% of households 3 allergens were detected, in 14% of households 4 different allergens were detected, in 8% of households 5 different allergens were detected, in 2% of households 6 different allergens were detected.All but 1 household with Der p 1 present also was positive for Der p 2, 78% of samples positive for Der p 2 were also positive for Der f 1.

Sensitization & atopic disease
Of the 121 children who provided blood for ISAC measurements, 51 (42%) were sensitized to at least 1 allergen component, of whom 13 (25%) were mono-sensitized and 38 (75%) were sensitized to multiple allergen components.Sensitization to aeroallergens 46 (38%) was most common, 40 (33%) of children were sensitized to grass (Phl p or Cyn d), 19 (16%) were sensitized to cat (Fel d 1) (Supplementary Table 1).Few children were sensitized to food allergens 8 (7%) and HDM 3 (2.5%).Self-reported rates for existing allergies were 26%, for asthma 24%, for atopic dermatitis 14%, for rhinoconjunctivitis 22% and for food allergy 9%.Reported doctor-diagnosed asthma was 12%, doctor-diagnosed atopic dermatitis 15%, doctordiagnosed rhinoconjunctivitis 18% and doctordiagnosed food allergy 6%.3).Exploratory factor analysis (Principal Component Analysis) of the exposure questions from the questionnaire resulted in three factors explaining 73% of the variance: pollen exposure, parental education and indoor living conditions.Pollen exposure contains the cumulative pollen exposure variables to hazel, alder, birch, plane, grass, plantain, mugwort and ambrosia; the second factor, parental education, contains the education level of mother and father; the third factor, indoor living conditions, contains occupant density, change of bed linen, visible mold in the home, and presence of multiple detectable allergens in residential dust.Factor scores (Anderson Rubin scores) were created and directly used as a variable in the multiple logistic regression analysis.Multiple logistic regression analysis demonstrated that living close to a busy road increased the odds of being sensitized to any allergen aOR(95%CI) 2.7(1.0-7.2) (Table 2).The strength of this finding increased for sensitization together with self-reported symptoms of rhinoconjunctivitis aOR(95%CI) 4.4 (1.3-14.8)or asthma aOR(95%CI) 5.5 (1.4-21).For sensitization to outdoor aeroallergens, living close to a busy road also increased the odds aOR(95%CI) 2.8(1.1-7.1).For sensitization to indoor allergens, the constructed factor indoor living conditions consisting of occupant density, frequency of bedlinen change, visible mold in the home and exposure to multiple allergens in residential dust significantly increased the odds of being sensitized to indoor allergens aOR(95%CI) 1.9 (1.1-3.2) and being sensitized plus reporting symptoms of rhinoconjunctivitis 1.9 (1.0-3.6) and asthma 2.1 (1.0-4.1).

DISCUSSION
In this cross-sectional study we investigated the relation between environmental exposure and prevalence of allergic sensitization among school children born and resident in the Swiss alpine municipality Davos.
Sensitization to any allergen was 42% in our cohort of children aged 5-9 years and selfreported rate for any allergies was 26%.7][18] In our study cohort, the self-reported rates for asthma was 24%, for rhinoconjunctivitis 22%, for eczema 14%, and for food allergy 9%.Compared to other studies in Western Europe and Switzerland, rates in our cohort were > three-fold higher. 2,19,20owever, ISAAC also demonstrated large variability in prevalence and severity of atopic disease, not only between countries and cities, but also within cities. 2 Local environmental characteristics are thought to play a crucial role in the development and prevalence of atopic disease.A Norwegian study reported an asthma prevalence of 10% in a nonpolluted rural mountain area, with a comparable climate to our cohort. 21,22It is hypothesized that the relatively high prevalence of asthma may be associated with a high incidence of keeping animals at home combined with frequent heavy physical activity in cold weather. 22In Davos, alpine sports and cross-country skiing are also quite popular among school children and 43% of households reported the presence of pets inside the home.In Los Alamos located at 2194 m, a mitefree area with low outdoor pollution, asthma diagnosis was also high (15%) among children born and raised there, and possibly associated with the frequent presence of indoor pets (77%), especially cat. 23,17 our cohort, sensitization to grass was 33% and self-reported rhinoconjunctivitis was 22%.Several studies have reported a higher prevalence of pollen sensitization and rhinoconjunctivitis at high altitude compared to lower altitudes.In a French study comparing school children living at sea level (Marseille) and high altitude (Briancon, 1260 m a.s.l.), positive SPTs for grass pollen were found in 21.7% of children born and living at high altitude, whereas this was 8.5 % in children living at sea level. 24An American study including children living at 2194 m reported a 47 % prevalence of pollen sensitization. 23A Norwegian study reported 33% prevalence of rhinoconjunctivitis and pollen sensitization. 22However, pollen counts of grass and birch at these altitudes were much lower compared to sea level.It could be that the pattern of sensitization reflects the exposure to locally available allergens.Children living at altitude may spend more time outside and therefore may have a higher cumulative pollen exposure.pollen exposure does not directly represent allergen exposure.Daily allergen release per pollen fluctuates and pollen potencies in the alpine climate may also be affected by the presence of pollen coming from long-distance geographical origin. 25,26lergen exposure outside the geographical area of common residence, for example during summer holidays in places with higher pollen concentrations, may also contribute to the high prevalence of rhinoconjunctivitis.Furthermore, it is possible that populations born and raised in high altitude may have been more isolated and thus, House dust contains several allergens but most commonly mites, animal and fungal allergens.Detectable levels of HDM allergens were found in up to 37% of residential dust samples, but generally in low concentrations.The median concentrations of Der p 1, Der p 2 and Der f 1 were 74 ng/g, 18 ng/g and 29 ng/g, respectively.However, there was large variability among samples and some households reached concentrations as high as 4690 ng/g for Derp 1, 1622 ng/g for Der p 2 and 2457 ng/g for Der f 1.These concentrations were considered elevated in other studies as well. 28nother study already demonstrated no significant change of HDM concentration with increasing altitude. 29However, HDM sensitization in our cohort was rare (2.5%), which is line with other studies conducted at altitude. 22,23,27,30art from indoor allergen exposure, other indoor living conditions demonstrated a significant influence on sensitization to indoor allergens.We found that a combination of occupant density, frequency of bedlinen change, visible mold in the home and exposure to multiple allergens in  Table 2. Multiple logistic regression analyses assessing the relation between exposure and any sensitization, exposure and sensitization plus doctor confirmed diagnosis of rhinoconjunctivitis, exposure and sensitization plus self-reported symptoms of rhinoconjunctivitis and exposure and sensitization plus self-reported symptoms of asthma.Sensitization was assessed as sensitized to any allergen, sensitized indoor allergens and sensitized to outdoor allergens.All models were adjusted for sex, having atopic family members and socioeconomic position.aOR ¼ adjusted Odds Ratio, CI ¼ Confidence Interval, p ¼ p-value.Male sex (vs female sex), atopic family members (vs no atopic family members), socio-economic position index (continuous), living close to a busy road (vs not living close to a busy road), indoor living conditions (Anderson Rubin factor score consisting of occupant density, change of bed linen, visible mold in the home, and presence of multiple detectable allergens in residential dust), ever breastfed (vs never breastfed) residential dust significantly increased the odds of being sensitized to indoor allergens.Visible mold in the home has already been identified as a risk factor for developing atopic disease. 31Combined data from several European birth cohorts also demonstrated an association between visible mold in the home and a statistically significant increased risk for asthma and allergic rhinitis in school children. 32,33 our cohort, living close to a busy road was associated with increased odds OR (95%CI) for being sensitized to any allergen (3.6 (1.3-10)) and to outdoor allergens (2.8 (1.1-7.2)).These associations increased when self-reported symptoms were added to sensitization as the outcome.8][39] Recently, allergen specific analyses suggested increased risk for sensitization to birch, grass and cat may be related to exposure to specific pollutants. 181][42][43] To accurately evaluate environmental determinants, both pollen allergen and pollutant contents should be considered. 41ttle is known about regions with low levels of TRAP exposure and risk of allergic respiratory disease.Even though pollution levels in Davos are relatively low, we still found a significant association between living close to a busy road and allergic sensitization.Another study in a low pollution setting also demonstrated associations between living <200 m from a major road and current wheeze and allergic sensitization. 44These findings suggest that even in settings with relatively low pollution levels, TRAP exposure could increase the risk for allergic sensitization.
This study was designed as a cross-sectional observational study.Therefore, it is not possible to determine whether the outcome followed exposure in time and identified associations should be interpreted with this study design in mind.Furthermore, the use of questionnaires may lead to misclassification because of recall bias.Another limitation of the study is possible selection bias.Families with children with current atopic symptoms may have been more interested to participate in the study, leading to a possible overestimation of the prevalence.
Strengths of this study are the detailed assessment of sensitization using ISAAC and the extensive assessment of exposure to indoor environmental allergens as well as cumulative exposure to pollen and air pollution.

CONCLUSION
In a cohort of children born and raised in an alpine environment, sensitization to aeroallergens is common (38%), especially to grass (33%) and cat (16%).House dust mite allergen was detected in up to 38% of residential dust samples, but sensitization to HDM was low (2.5%).Living close to a busy road and indoor living conditions increased the risk for allergic sensitization.In a clean and healthy alpine environment, air pollution might still be an important factor contributing to allergic sensitization.
20-01-02-6).The resulting fluorescent signals were measured with a confocal laser scanner (LuxScan-10K, CapitalBio, Beijing, China).Data were analyzed using Phadia Microarray Image Analyzer (MIA) software and transformed into semiquantitative ISAC Standardized Units (ISU).Specific IgE values !0.3 ISU were considered positive.Participants with a positive value to any of the 112 allergens on the ImmunoCAP ISAC were considered sensitized.Statistical analysis Descriptive statistics were used to describe sensitization patterns of the study participants.Univariate regression models were constructed to assess correlations between specific IgE values (Der f 1, Der p 1, Der p 2, Der f 1, Mus m 1, Alt a 1, Asp f 1, and Ara h 6) and corresponding allergen component concentrations in residential dust and cumulative pollen exposure (Amb a 1, Aln g 1, Bet v 1, Bet v 2, Cyn d 1, Phl p 1-12, Cor a 1.0101, Plane a 1-3, Pla l 1 Art v 1 Art v 3).Time series analysis was conducted for air pollutants NO, NO x , NO 2 , O 3 , PM 10 and pollen with an Annual Pollen Integral of APIn >100 pollen * day/m3 for the studied period (Alnus, Corylus, Betula, Poaceae and Plantago) with LOESS smoothing to determine trends.Principal component analysis was used to group exposure variables from the questionnaire.

Fig. 3
Fig. 3 Daily pollen concentrations of the most common pollen.The black line shows the mean along the period (2011-2019) and the colored areas show the ranges for the concentration values during the studied period

Fig. 4
Fig. 4 Monthly average concentrations of gaseous pollutants NO, NO 2 , NOx, O 3 , and fractions of particulate matter PM 10 and PM 2.5 .The colors indicate the measuring stations, red for Davos Promenade DPR and orange for Davos Bubenbrunnenplatz DBP.The move had no statistical significant effect on the values (see text) . average yearly exposure in the period from 2011 to 2019 was for alder 717 AE 425 (pollen * day/m 3 ), hazel 147 AE 114, birch 564 AE 3774, plane 11 AE 51, grass 1578 AE 3418, plantain 136 AE 29, mugwort 7 AE 8 and ambrosia 3 AE 2 (Supplementary Figure

Table 1 .
(Continued)Clinical and demographic data of study participants.SD ¼ standard deviation, IQR ¼ inter quartile range, ISAC ¼