key: cord-0030170-cz5mp203
authors: Maya‐Manzano, José María; Pusch, Gudrun; Ebner von Eschenbach, Cordula; Bartusel, Elke; Belzner, Thomas; Karg, Erwin; Bardolatzy, Ulrich; Scheja, Michael; Schmidt‐Weber, Carsten; Buters, Jeroen
title: Effect of air filtration on house dust mite, cat and dog allergens and particulate matter in homes
date: 2022-04-21
journal: Clin Transl Allergy
DOI: 10.1002/clt2.12137
sha: 43790e5bd1e191c13d117c4ccf05eaa7dd67566c
doc_id: 30170
cord_uid: cz5mp203

BACKGROUND: Indoor allergens (i.e. from mite, cat and dog) are carried by airborne particulate matter. Thus, removal of particles would reduce allergen exposure. This work aims to assess the performance of air filtration on particulate matter and thus allergen removal in 22 bedrooms. METHODS: Indoor air was sampled (with and without air filtration) with a cascade impactor and allergens were measured using enzyme‐linked immunosorbent assay (ELISA). Particulate matter (including ultrafine particles) was also monitored. RESULTS: The median of allergen reduction was 75.2% for Der f 1 (p < 0.001, n = 20), 65.5% for Der p 1 (p = 0.066, n = 4), 76.6% for Fel d 1 (p < 0.01, n = 21) and 89.3% for Can f 1 (p < 0.01, n = 10). For size fractions, reductions were statistically significant for Der f 1 (all p < 0.001), Can f 1 (PM(>10) and PM(2.5–10), p < 0.01) and Fel d 1 (PM(2.5–10), p < 0.01), but not for Der p 1 (all p > 0.05). PM was reduced in all fractions (p < 0.001). The allergens were found in all particle size fractions, higher mite allergens in the PM(>10) and for pet allergens in the PM(2.5–10). CONCLUSIONS: Air filtration was effective in removing mites, cat and dog allergens and also particulate matter from ambient indoor air, offering a fast and simple solution to mitigate allergen exposome.

with increasing aerodynamic diameter of a particle. Thus both, the smaller particles with capacity of penetrating deeper into the airways 5, 15 and the larger particles containing the majority of the allergen 16 should be removed. 13, [16] [17] [18] Der f 1 has a half-life of 10 years 8, 19 and it is not realistic to rely on natural decay to ensure low allergen levels at homes. The bundle of single measures recommended to mitigate mite allergen exposure (encasings, washing bedding >60°C, humidity <50%, removal of carpets etc.) is quite diverse, 4,20-25 but a combination of them is typically advocated, [26] [27] [28] as the 'one only' approach was often reported as ineffective. 1, 4 Other indoor allergens such as those originating from cats and dogs can also be ubiquitous, even when these animals are absent in the homes. 29, 30 Fel d 1 is the major cat allergen, 31 originating from the sebaceous gland of the skin, 32 whilst Can f one is found mostly in saliva and hair/dander. 33 Removing the pet from the household is considered effective for allergen avoidance, but most pet owners are not willing to do so. 34 Since these allergens are detected in the airborne PM, a good approach to reduce exposure would be filtering the air. 21, 35, 36 In the last few years, significant improvements in allergic rhinitis and asthma symptoms were reported using high-efficiency particulate air (HEPA) air purifiers, 37, 38 which also resulted in a decrease in medication. 38 HEPA air purifiers not only removed the HDM allergens, 39, 40 but also cat allergens. 41 Besides exposure to allergens, there are other concerns about exposure to PM. The health effects of PM, especially PM 2.5 , are well described and include respiratory and cardiovascular diseases, lung cancer and cognitive dysfunctions, 42, 43 especially in vulnerable groups like children, elderly and people with pre-existing cardiovascular diseases. 42, 43 Ultrafine particles (UFP) exposure is a concern too, and evidence suggests adverse effects on cardiovascular and cerebrovascular health. 44, 45 Thus, although the effectiveness of air filtration has been proven for certain particle sizes, 38, 39 the novelty of our study is to simultaneously investigate the airborne levels of three major indoor allergens in different size fractions as well as PM covering a wide range of sizes in homes (bedrooms).

The aim was to determine exposure to airborne allergens from HDM, cats and dogs, PM and UFP in 22 bedrooms in Bavaria, South Germany.

Airborne concentrations of Der f 1 (n = 20). Left: Total airborne allergen (sum of all size fractions). Right: Concentrations for each size fraction. Dashed red lines represent the LOQ, which for Der f 1 is close to zero. Empty homes mean that they did not meet the criteria to be analyzed, according to section Allergen Sampling. (B) Airborne concentrations of Der p 1 (n = 4). Left: Total airborne allergen (sum of all size fractions). Right: Concentrations for each size fraction. Dashed red lines represent the LOQ. Empty homes mean that they did not meet the criteria to be analyzed, according to section Allergen Sampling Then we tested whether these parameters can be significantly reduced by using air filtration, for which we used a portable air purifier (Philips Air Purifier AC4236, 4000i-series) with a HEPA filter and a clean air delivery rate (CADR) of 500 m 3 /h.

Approvals from the ethical committee of Klinikum rechts der Isar, number 377/19-S-SR, amendment from 23 July 2020 (SARS-CoV2 measures) and from Philips Internal Committee Biomedical Experiments were obtained. Twenty-two homes were selected according to the following inclusion/exclusion criteria and informed consent was obtained in writing.

� All occupants being between 18 and 65 years of age. � Willing not to change their bedding for 2 weeks before each home visits.

� Not traveling for more than 1 week during the study duration (i.e.

� Preferred people sleeping in winter 'always' with closed windows over 'sometimes', over 'never'.

� Fluent German and/or English speaking 'willing and able to provide informed consent'.

� Unwilling or unable to provide informed consent.

� Being absent from their homes for more than 1 week during the duration of the study.

� Households where at least one occupant had asthma.

� Bedroom height more than 3.2 m.

� Bunk beds and water beds in bed room.

� Mattress cleaned by vacuuming within previous 6 months.

� Air brick or ventilation aperture/mechanical ventilation in the bedroom.

� Households using air dehumidifiers in their homes. � Households using house dust mite kill sprays/products. � Households using mattress encasings in bedroom.

The study period was from 5 February 2020 to 22 April 2020 and (after a break due to a SARS-CoV-2 lockdown in Germany) from 1

July 2020 to 28 September 2020 (Table S1 ). Every home completed a control-and intervention visit following a crossover randomized experimental design, having both visits within 4 weeks to avoid any influence or bias provoked by the possible seasonality of mites or pet allergens. Thus, half of cases were first control (without air filtration) and half first intervention (when air filtration was working during sampling).

To collect airborne particles a GMU (Gerhard Mercator Universität, Duisburg) Johnas II cascade impactor (thereafter Johnas 2) was used, connected to a pump with an aspiration rate of 53 l/min.

Flow was permanently monitored in-line with a Bellows BG series gas flow meter. The flow was additionally calibrated before each visit by using a heat-wire anemometer-flowmeter EasySPT200. This Johnas 2 cascade impactor separates PM into three size fractions, PM >10 , PM 2.5-10 and PM 2.5 . Particles were impacted on electrostatic cloth, 46 which was tested to release all allergens best (data not shown). PM was measured using a spectrometer GRIMM model 1.108 version 8.60 in operational mode mass (normal dust mode, expressed in μg/m 3 ), which divides PM into 16 fraction sizes every 6 s. We report PM as the same fractions as were collected with the Johnas 2, PM 2.5 , PM 2.5-10 and Power consumption of air filtration (correlated with air flow) was monitored using a Basetech EM-3000.

A 2-min interval where pillows (30 s), covers (30 s) and sheets (60 s) were shaken represented one dust disturbance event. Each home visit consisted of four dust disturbance events. Air filtration at maximal performance (500 m 3 /h, lower flows can be easily set) was turned on directly after that event, and then the Johnas 2 was also turned on together with the spectrometer and the Nano tracer. After 1 h air filtration the devices were switched off and the procedure was repeated (each home was sampled about 4 h in total). The particle sensors ran continuously during the whole experiment, with doors and windows closed during the experiment. The home owners were told to not change the bedding within 2 weeks before each of the home visits and asked to not clean or vacuum the mattress in the bedroom until the end of (and 6 months before) the study. The same experimenter performed the dust disturbance events for control and intervention at a specific home. Nobody (including pets) was allowed to enter the experiment room or to take showers (humidity) during the experiment to avoid owners' interference in air quality. No home had multiple cats. All sampling inlets were located at 1.2 m' height and air filtration was located at least 2.5 m away from the measurement instruments. The only variation in experimental procedures between both visits was the in-or exclusion of the Philips air purifier. A drawing of the principal set-up guaranteed that the equipment was placed identically for both visits ( Figure S1 ).

ELISA optimization for recovery of Der f 1 and Der p 1 was performed according the guidelines of EAACI. 47 Here HDM allergens D. Homes with at least two values in any fraction >LOQ in control visits were included into analysis. Amongst the included homes, values below LOQ for any size fraction were set to half of LOQ. 50, 51 This represents a conservative approach for statistical analysis since correctness of values below this LOQ can't be ensured. As a consequence, 20 homes qualified for analysis for Der f 1, 4 homes for Der p 1, 10 homes for Can f 1 and 21 homes for Fel d 1.

We checked the hypothesis whether air filtration after dust disturbance leads to a reduction in exposure to allergens and PM. Nonnormal distributions were found for allergens and PM by using a Shapiro-Wilk test. Thus, the Wilcoxon signed-rank test was used to test our hypothesis. The only exceptions were those cases having n < 6 (Der p 1). Here, Wilcoxon cannot be used (n < 6) and a paired t-test was used instead. Reduction (%) was calculated as = 100 − ((Intervention/ Control) � 100).

For PM analysis, the area under the curve (AUC) was calculated by the trapezoid method for the times air filtration was running (2-60 min). In the PM analysis, the starting point of each repetition was defined as the first increase of the baseline 52 and time was put to zero, except for UFP, where clock time was chosen instead since dust disturbances did not result in consistent peaks. In all cases (allergens and PM), statistical differences were reported when p < 0.05. To estimate the speed by which the air purifiers reach their maximum effect (i.e. after this time no more reduction in airborne particle reduction was measured) the average time until no more change in AUC is reached was calculated. All calculations were carried out by using R software 53 and the AUC was calculated with the pracma package. 54 

Experiments in four homes showed lower Der f 1 concentrations (median = 63.2% less) in living rooms compared to bedrooms ( Figure S2, Supplementary material) . Hence, to avoid dilution of the pooled samples by low HDM allergen-containing particles from the living room, we restricted our experimentation to the bedrooms.

The ELISA for Der f 1 was the most sensitive of our assays. We therefore used Der f 1 as a marker for HDM exposure. The Der p 1-ELISA was sensitive enough to be able to detect allergen levels if they would have occurred at the same level as measured by the Der f 1 assay, but Der p 1 was detected only in four homes. The Der p 2 and

Der f 2 assays were tested with pure HDM preparations from commercial suppliers and had a similar LOQ to the Der p 1 ELISA (data not shown), but did not detect sufficient allergen in pre-experiment samples from homes. This indicates that they are infrequent allergens in and around Munich, and were omitted from the further study. Figure 2 ) .

The majority of Der f 1 and Der p 1 was detected in the fractions PM >10 and PM 2.5-10 ( Figure 2 ). The descriptive statistics for HDM allergens can be seen in supplementary material (Table S2 ). and PM >10 as shown in Figure 4 . Can f 1 concentrations in the two homes with the highest levels of this allergen were two and four times higher than the maximum for Fel d 1 ( Figure 3A ,B).

Air filtration resulted in a statistically significant reduction of total Fel d 1 (p < 0.01) and total Can f 1 (p < 0.01), with medians of reduction of 76.6% and 89.3%, respectively (Table 1) . Fel d 1 was reduced significantly for PM 2.5-10 (p < 0.01; Figure 4 ). Can f 1 was reduced in all fractions that had a measurable concentration of allergen, obtaining medians in reductions of 87.5% for PM >10 (p < 0.01) and 93.7% for PM 2.5-10 (p < 0.01).

The (Table S2 ). The percentage of reduction as calculated using the AUCs for each home and fraction size and corresponding medians can be seen in Table 2 . Running air filtration led to statistically significant reductions of PM concentrations in all particle sizes (all p-values <0.001) with strongest effect for the medium size fractions.

Failure to measure any HDM allergen concentrations in homes is frequently reported, 13,53 as their concentrations are close to or below the limit of detection in many homes 54, 55 or they deposit fast on the floor unless they were disturbed. [55] [56] [57] For instance, a large study in U.S detected HDM allergens in only 38% of homes, 58 

Europe, 59 detectable HDM allergens were reported for 49% of the samples. In the current study, the detection of airborne HDM allergen was successful in all homes even after splitting the allergens into three size fractions, due to optimization of the sampling, extraction and measurement protocols. 47 Der f 1 was the dominant HDM allergen in Munich and surroundings (n = 20) and Der p 1 was only occasionally present (n = 4), but then at sometimes high levels (two sums for all fractions exceeded 200 pg/m 3 , Figure 1 ). contrary to what we found in our homes (modern homes having better isolation and heating, which decreases humidity), which may explain the abundance of Der f 1 we observed in this study. Other authors also found a higher concentration of Der f 1 than Der p 1. 61 Even in homes not infested by D. farinae, allergens can be transferred from mite-infested clothing or car seat materials as a source of HDM allergens, as was reported by different authors. 61, 62 Since outflow of the air purifier was vertical at 80 cm above the floor, and the inletflow (0.9-1.1 m/sec) was close to the floor, but was filtered before emission, the air purifier itself could not cause resuspension of deposited allergens.

Air filtration resulted in a statistically significant reduction of HDM allergen Der f 1 but was not statistically significant for Der p 1 (see Figures 1 and 2) . Hence, we think we missed statistical significance on Der p 1 due to too few homes containing that allergen in and around Munich.

Some authors reported 80% of HDM allergens to be associated with 10-40 μm particles, 11, 12 settling down within 15 min and not remaining airborne due to rapid sedimentation. 63 We also found most of Der f 1 and Der p 1 in the fraction PM >10 , but our results showed that substantial allergen amounts stay airborne longer as half of total HDM allergen was carried by particles smaller than 10 microns. Furthermore, our Der f 1 ( Figure 2 ) and PM ( Figure 5 and Table 2 ) data show that despite faster sedimentation, significant exposure reduction can be achieved by means of air filtration even for large particles and associated allergens.

Some authors discussed that Fel d 1 is more suitable to be removed by air filtration, due to the higher percentage of allergen carried by smaller particles, 36 compared to mite allergens, this difference was modest. Consequently, we were able to show that not only cat and dog allergens, but also those from mites can be significantly reduced by means of air filtration.

Our results are also in good alignment with previous findings from Custovic et al., 64, 65 who found 42% and 49% of total airborne

Can f 1 and Fel d 1 to be associated with particles >9 μm. Luczynska et al. 66 for Fel d 1 also reported a 75% association to particles larger than 5 μm. Tovey et al. 12 reported that the 80% of Der p 1 was found in PM > 10 μm and De Blay et al. 63 reported that 78% of the group I allergens of mite were detected in particles >6 μm. A direct comparison with our results is not possible as these studies used cascade impactors with different size fraction and methods of resuspension.

Still, our study agrees with previous results that a large fraction of HDM was detected in the larger size fractions. [67] [68] [69] [70] We show however that particles <10 µm also carry substantial amounts of HDM allergens.

Because airborne allergens like HDM are PM themselves, and air filtration used in this study was very effective in removing PM across a wide size range, removal of ambient PM also removed allergens. All PM fractions were significantly reduced by air filtration (all p < 0.001). The larger a particle, the faster it deposits by gravity and consequently air filtration has less time to 'catch' these particles.

Consequently, small particles that stayed airborne longer like PM 1 were more efficiently removed by air filtration than the larger particles like PM 10 ( Table 2) .

Raulf et al. 47 discussed that allergens on smaller particles remain airborne longer, and thus have more chance of being inhaled. Our time-resolved PM measurements confirm this by showing a slower natural decay of smaller particles compared to larger ones ( Figure 5 ).

The same data also show, however, that all larger particles (except the largest size fraction PM 10-22.5 ) remain elevated for at least 1 

This study was sponsored by Philips Inc. Philips provided air filtrations type AC4236 and the Philips Aerosense Nano tracer for UFP.

Open access funding enabled and organized by Projekt DEAL.

Michael Scheja is employee in Philips. The investigations were carried out in compliance with good scientific practices. Support provided by

Philips had no effect on the results presented. 

https://orcid.org/0000-0003-3581-5472

The voice of the patients: allergic rhinitis is not a trivial disease

The house-dust mite (Dermatophagoides pteronyssinus) and the allergens it produces. Identity with the house-dust allergen

Allergenic identity between the common floor mite (Dermatophagoides farinae Hughes

and house dust as a causative antigen in bronchial asthma

Dust mite allergens: mitigation and control

The role of dust mites in allergy

Serological, genomic and structural analyses of the major mite allergen Der p 23

Mite (Der p 1, Der f 1) and cat (Fel d 1) allergens in the homes of babies with a family history of allergy

Decay of house-dust mite allergen Der f 1 at indoor climatic conditions

Sensitization rates to common inhaled allergens in Germany -increase and change patterns over the last 20 years

Mite faeces are a major source of house dust allergens

Indoor allergens and allergic respiratory disease

The distribution of dust mite allergen in the houses of patients with asthma

Mite allergen (Der p 1) is not only carried on mite feces

An immunochemical approach to indoor aeroallergen quantitation with a new volumetric air sampler: studies with mite, roach, cat, mouse, and Guinea pig antigens

Dust mite allergens are carried on not only large particles

Allergens, sources, particles, and molecules: why do we make IgE responses?

Airborne allergens associated with small particle fractions

Airborne allergens associated with asthma: particle sizes carrying dust mite and rat allergens measured with a cascade impactor

The decay of house dust mite allergens, Der p I and Der p II, under natural conditions

The biology of dust mites and the remediation of mite allergens in allergic disease

Home environmental interventions for house dust mite

The role of water temperature and laundry procedures in reducing house dust mite populations and allergen content of bedding

Optimal conditions for the removal of house dust mite, dog dander, and pollen allergens using mechanical laundry

Nonwoven in contrast to woven mattress encasings accumulate mite and cat allergen

Reducing relative humidity to control the house dust mite Dermatophagoides farinae

Clinical practice guideline: allergic rhinitis executive summary

The diagnosis and management of rhinitis: an updated practice parameter

Environmental assessment and exposure control of dust mites: a practice parameter

Distribution and removal of cat, dog and mite allergens on smooth surfaces in homes with and without pets

Animal allergens and their presence in the environment

An update on molecular cat allergens: Fel d 1 and what else? Chapter 1: Fel d 1, the major cat allergen

Airborne indoor particles from schools are more toxic than outdoor particles

The major dog allergens, Can f1 and Can f2, are salivary lipocalin proteins: cloning and immunological characterization of the recombinant forms

Environmental assessment and exposure control: a practice parameter--furry animals

Does air filtration work for cat allergen exposure?

A placebo-controlled trial of a HEPA air cleaner in the treatment of cat allergy

Effects of air purifiers on patients with allergic rhinitis: a multicenter, randomized, double-blind, and placebo-controlled study

Effects of indoor air purifiers on children with asthma

Efficacy of air purifier therapy in allergic rhinitis

Efficacy of air purifier therapy for patients with allergic asthma

Efficacy of air cleaners in asthmatics allergic to cat in ALYATEC® environmental exposure chamber

Air Quality -Health Effects of Particulate Matter. Policy Implications for Countries in Eastern Europe

Danger in the air: air pollution and cognitive dysfunction

Health effects of ultrafine particles: a systematic literature review update of epidemiological evidence

Long-term exposure to ultrafine particles and incidence of cardiovascular and cerebrovascular disease in a prospective study of a Dutch cohort

Passive airborne dust sampling with the electrostatic dustfall collector: optimization of storage and extraction procedures for MAYA-MANZANO ET AL. endotoxin and glucan measurement

Monitoring of occupational and environmental aeroallergens -EAACI position

Endotoxin, cat, and house dust mite allergens in electrostatic cloths and bedroom dust

A multi-center ring trial of allergen analysis using fluorescent multiplex array technology

Effectiveness of a portable air filtration device in reducing allergen exposure during household chores

Indoor determinants of Der p 1 and Der f 1 concentrations in house dust are different

Novel method using baseline normalization and area under the curve to evaluate differences in outcome between treatment groups and application to patients with cervical spondylotic myelopathy undergoing anterior versus posterior surgery

R: A Language and Environment for Statistical Computing

Practical Numerical Math Functions

Measurement of allergen concentration in the air as an aid in controlling exposure to aeroallergens

Risk levels for mite allergens. Are they meaningful?

The effect of home characteristics on dust antigen concentrations and loads in homes

Early-life house dust mite allergens, childhood mite sensitization, and respiratory outcomes

Current eczema in children is related to Der f 1 exposure but not to Der p 1 exposure

Correlations between levels of mite and cat allergens in settled and airborne dust

Mite and mite allergen removal during machine washing of laundry

Airborne dust mite allergens: comparison of group II allergens with group I mite allergen and cat-allergen Fel d I

Woodcock A. Distribution, aerodynamic characteristics, and removal of the major cat allergen Fel d 1 in British homes

Aerodynamic properties of the major dog allergen Can f 1: distribution in homes, concentration, and particle size of allergen in the air

Airborne concentrations and particle size distribution of allergen derived from somestic cats (Felis domesticus): measurements using cascade impactor, liquid impinger, and a two-site monoclonal antibody assay for Fel d I

The effect of air filtration on airborne dog allergen

Clinical effects of air cleaners in homes of asthmatic children sensitized to pet allergens

Can the use of HEPA cleaners in homes of asthmatic children and adolescents sensitized to cat and dog allergens decrease bronchial hyperresponsiveness and allergen contents in solid dust?

Air filtration units in homes with cats: can they reduce personal exposure to cat allergen?

Efficacy of a novel air filtration pillow for avoidance of perennial allergens in symptomatic adults

Effect of an air cleaner with electrostatic filter on the removal of airborne house dust mite allergens

SUPPORTING INFORMATION Additional supporting information can be found online in the Supporting Information section at the end of this article. How to cite this article