The reduction of rhinitis symptoms by nasal ﬁlters during naturalexposure to ragweed and grass pollen
Background: Prototype nasal ﬁlters were developed to collect inhaled pollen.
This study evaluated the eﬃcacy of the ﬁlters for prevention of rhinitis symp-
toms during acute outdoor pollen exposure.
Methods: A randomized double-blind design was used. Subjects (n ¼ 46) with a
1Woolcock Institute of Medical Research, Royal
history of autumn exacerbation of rhinitis and positive skin test to ragweed,
Prince Alfred Hospital, University of Sydney, New
Bermuda and/or Bahia grass wore either active or placebo nasal ﬁlters for 2 h in
autumn in a park containing these species. Major and Total Symptoms scores
Department of Rural Health, University of Sydney
were recorded at 0, 30, 60, 90 and 120 min.
and Northern Rivers Area Health Service, LismoreNew South Wales, Australia
Results: Subjects wearing active nasal ﬁlters had signiﬁcantly reduced scores, atall time-points compared with placebo group (all P < 0.05). Of 14 individualsymptoms measured, seven were signiﬁcantly reduced (number of sneezes, runnynose, itchy nose, sniﬄes, itchy throat; itchy eyes and watery eyes) and anotherthree showed a trend towards lower severity. The nasal ﬁlters also enabled the
Key words: allergen avoidance; nose filter; prevention.
resolution of existing symptoms. Maximal diﬀerence in symptoms was seenimmediately after subjects had spent 20 min sitting beside a large patch of
Conclusion: This is the ﬁrst clinical trial of a nasal ﬁlter. The results suggest
it has potential for enhancing rhinitis management during acute allergen
Avoidance of allergens is advocated as the ﬁrst step in
management of allergic rhinitis, and as an adjunct to
medications (1, 2). It is recognized, however, thatavoidance of pollens and fungal spores is diﬃcult to
The nasal ﬁlter is shown in Fig. 1. The airﬂow resistance and cap-
achieve because of their ubiquitous nature (1). Facemasks
ture eﬃciencies for ragweed (Ambrosia artemisiifolia), Bermuda
provide personal protection; however, they are only used
grass (Cynadon dactylon) and Bahia grass (Paspalum notatum)
by 1% of people (3). While the mainstay of rhinitis
pollen (Greer Laboratories Inc., Lenoir, NC) were measured at ﬂowrates of 4.6, 10.3, 21.7 and 32.5 l/min as previously described (4).
management is pharmacotherapy, 74% of people reportthat medications do not adequately control their symp-toms, and 65% report avoiding some medications
The study utilized a double-blind placebo-controlled design. The
We have previously described nasal ﬁlters, worn inside
study was approved by the Human Ethics Committee of Northern
the nose that collect inhaled particles using the principle
Rivers Area Health Service, and subjects gave written informed
of impaction (4). They are easy to breathe through, and
consent. Eligible subjects were over 16 years old, had a history of
have a high capture eﬃciency for particles above 8 lm in
rhinitis exacerbation in the autumn and were skin prick test-positive
diameter (4), which includes all pollens (5). Given the
to mixed ragweed, Bermuda grass and/or Bahia grass (Table 1).
prevalence of allergic rhinitis, the level of dissatisfaction
Exclusion criteria were: complete nasal obstruction, recent sinusitis,history of severe asthma exacerbations, or use of nasal steroids/
with current medications and lack of acceptance of an
antihistamines/systemic decongestants within the last month. The
eﬀective method to prevent exposure outdoors, we tested
study location was a semirural park with abundant ﬂowering Bahia
a prototype nasal ﬁlter to determine if it would reduce
and Bermuda grasses and ragweed. Prior to arriving at the park
symptoms of allergic rhinitis during high natural expo-
subjects wore disposable dust-masks to reduce the development of
baseline symptoms. Subjects were randomly allocated to one of
Self-assessed rhinitis symptoms and peak nasal inspiratory ﬂowwere recorded at baseline and at 30 min intervals during thechallenge. Self-recorded symptoms were combined into twocomposite variables: Major Symptom Complex (MSC) and TotalSymptom Complex (TSC) severity scores (6–8). To ensure that atleast moderate levels of ragweed pollen exposure were experi-enced by all participants, each group sat beside a large patch ofragweed for 20 min, during the period 30–60 min after ﬁlterinsertion.
Ambient pollen levels were measured using a Burkard 7-day volu-metric spore trap, running at 10 l/min, located 3.5 m above theground at the challenge site. Individual pollen exposures weremeasured by the number of pollen grains collected on the adhesivecore of the active nasal ﬁlters. Samples were stained with Calberla’s
Figure 1. The nasal ﬁlter prototype is made up of a soft medical
solution (9) and ragweed and grass pollen were counted under a
grade silicone and has an inner polypropylene core is coated
with an adhesive. The placebo ﬁlters were of identical externalappearance, but did not have an inner core (inner core has been
removed from only one nostril in this picture). To accommodatediﬀerent sized noses, two sizes of nasal ﬁlter were used.
In a questionnaire administered 1 week after the challenge, sub-jects were asked to score their global satisfaction with the nasalﬁlters.
Table 1. Baseline subject characteristics
The primary outcome variables were change in MSC and TSC
scores from baseline, using all time-points. Secondary outcome
variables were changes from baseline in the individual components
of MSC and TSC. Overall diﬀerences between groups were exam-
Duration of allergic rhinitis, years (range)
ined by repeated measures anova, and t-tests were used to analyse
diﬀerences between treatment groups at each time-point. Diﬀer-
ences between groups in the frequency of sensitization were exam-
ined by chi-squared test. Statistical signiﬁcance was deﬁned as
In the test rig, pollen capture for Bahia, Bermuda and
MSC, Major Symptom Complex, range 0–1360 (number of nose blows, number of
ragweed averaged 98% for active and 3.5% for placebo
sneezes, runny nose, sniffles, itchy nose and watery eyes); TSC, Total Symptom
ﬁlters, across the range of ﬂow rates. Airﬂow resistance,
Complex, range 0–2448 (MSC symptoms plus itchy eyes, itchy ears, itchy throat,
cough and postnasal drip); SPT-positive ¼ skin prick test weal ‡4 mm2.
* ¼ P < 0.05 between active and placebo groups.
2O/L/s for active and 1.6 cmH2O/L/s for placebo
Pollen exposure on the challenge day was measured at
102 grains/2 h from the spore trap, and 68 pollen grains/
eight groups, and each group received either active (n ¼ 22) or
person/2 h from the active ﬁlters (geometric mean
placebo (n ¼ 24) nasal ﬁlters (Fig. 1). To eliminate visual unblind-ing: neither participants nor group supervisors (medical students)
had previous experience of the nasal ﬁlters; the way the ﬁltersworked was not explained to the participants or supervisors;
supervisors were blinded to the randomization allocation; withineach group all subjects received the same type of ﬁlter; there was no
Baseline MSC scores, prior to ﬁlter insertion, were 339.8
contact between groups; and the external appearance of the active
and 187.3 for the active (n ¼ 22) and placebo (n ¼ 24)
and placebo ﬁlters, once inserted, was identical.
After baseline assessments, subjects removed the dust-masks
ﬁlter groups, respectively (P ¼ 0.02). Over the 2-h chal-
and placed the nasal ﬁlters into their nostrils. Subjects were asked
lenge period MSC decreased in the active group and
to breathe through the nose for 2 h, while engaging in only mild
increased in the placebo group compared with baseline
activity (sitting, walking, eating) in a central location in the park.
(Fig. 2), resulting in highly signiﬁcant diﬀerences between
Nasal ﬁlters for ragweed and grass pollen
Figure 2. Mean absolute change in MSC scores between active
and placebo nasal ﬁlter groups. Overall diﬀerence between act-ive and placebo P ¼ 0.0076, repeated measures anova. Error
bars are 95% conﬁdence intervals (CIs) for each mean.
#P < 0.001, *P < 0.05 (t-test). MSC ¼ Major Symptom
Complex (number of nose blows, number of sneezes, runny
nose, sniﬄes, itchy nose and watery eyes).
–70 –60 –50 –40 –30 –20 –10
Figure 3. Percentage change from baseline in each symptom
the groups (overall, P ¼ 0.0076). At 30, 60, 90 and
score for active and placebo ﬁlter groups at 60 min. *P < 0.05
120 min the net diﬀerence in MSC for active compared
at 60 min (t-test). Individual symptoms that were signiﬁcantly
with placebo ﬁlters were )25, )68, )39 and )50% points
reduced by the ﬁlters during the 2 h study were: number of
respectively. The maximum diﬀerence was seen at 60 min,
sneezes, runny nose, itchy nose, sniﬄes, itchy throat, itchy eyes
immediately after the 20 min period of sitting beside a
large patch of ragweed. Similar diﬀerences between activeand placebo ﬁlters were seen in TSC scores (data notshown, overall P ¼ 0.023).
For seven of the 14 individual symptoms, there was a
signiﬁcant reduction in severity and for a further three
This is the ﬁrst reported clinical trial of nasal ﬁlters for
symptoms there was a consistent trend towards lower
the prevention of symptoms of allergic rhinitis. The nasal
severity in the active ﬁlter group than the placebo group.
ﬁlters collect inhaled particles by impaction (4), resulting
The strongest eﬀect was seen for the symptoms of sniﬄes
in high capture eﬃciency for particles above 8 lm and
(overall, P ¼ 0.004), rhinorrhea (overall, P ¼ 0.035) and
negligible air-ﬂow resistance (4, 10). Most pollen grains
itchy nose (overall, P ¼ 0.034), especially at 60 min
are above 15 lm in diameter and ragweed, Bermuda and
(Fig. 3) where highly signiﬁcant diﬀerences were found.
Bahia pollens which are 18, 28 and 34 lm, respectively
During the 2-h challenge period, signiﬁcant reductions
(5) were captured with high eﬃciency in the nasal ﬁlter
were also observed in number of sneezes, itchy throat,
itchy eyes and watery eyes (P < 0.05, t-test). For number
In the clinical trial of natural outdoor pollen exposure,
of nose blows, nasal blockage and peak nasal inspiratory
the net diﬀerence in MSC scores were the result of
ﬂow, there was a trend to improvement in the active
decreases in symptoms in the active ﬁlter group ()18 to
group but the diﬀerences were not signiﬁcant. The active
)33%) and increases in symptoms in the placebo ﬁlter
ﬁlters did not appear to inﬂuence postnasal drip or cough,
group (+7 to +35%), with a maximum net diﬀerence in
while a signiﬁcant (P < 0.05) increase in itchy ears was
MSC of 68% at 60 min. While this study did not compare
the eﬃcacy of the ﬁlters to rhinitis medications, a surveyof the literature indicates that both the magnitude andonset of symptom reduction with active ﬁlters compares
very favourably with that from medications (6–8, 11). The
Ninety-three percent of subjects said they would be
rapid reduction of pre-existing symptoms in the active
prepared to wear the ﬁlters again, with most people
ﬁlter group may expand the utility of the ﬁlters. In studies
prepared to use them in private situations such as around
of rhinitis medications, which have used similar acute
the house (88%) or in the garden (81%) compared with
challenge experiments, there has been a well-recognized
visiting friends (46%) or playing golf (33%).
placebo eﬀect (6–8), which was not observed for the
placebo ﬁlter group, although the low baseline symptoms
(12). Conventional allergen avoidance strategies such as
may account for this. Of note is the beneﬁcial eﬀect of the
mattress encasing, when practiced with high allergen load
nasal ﬁlters on ocular symptoms (Fig. 3), which is
in a normal domestic setting, often appear to fail to
consistent with the converse observations of ocular
reduce allergens to a level where signiﬁcant improvement
symptoms occurring following direct nasal challenge (1).
in symptoms occur [see meta-analysis (13)]. Additional
Although the active nasal ﬁlter group had signiﬁcantly
studies are required to examine the feasibility of using the
higher baseline symptoms than the placebo group, this
nasal ﬁlters for longer periods, as would be needed for
did not explain the signiﬁcant reduction in symptoms
in the active group. A post hoc analysis of data forindividuals with similar mid-range baseline symptoms inboth groups (n ¼ 10/group) showed a signiﬁcant reduc-
tion in MSC and TSC for the active ﬁlter group ()35%),while symptoms in the placebo group increased by
Authors thank Prof. John Beard, Northern Rivers University,
approximately 28% (P < 0.05). The active and placebo
Department of Rural Health, University of Sydney and Mr Mark
groups also diﬀered in the prevalence of sensitization to
Barlettt, Northern Rivers Area Health Service, for their assistance
Bermuda grass (Table 1); however, this was not related
in setting up this project. Drs Janet Rimmer, Connie Katelaris andNorbert Berend reviewed the trial protocol. Authors appreciate the
to baseline MSC scores (P ¼ 0.89, t-test).
support provided by the NSW Department of State and Regional
The nasal ﬁlters may also have an application for the
Development and Inhalix Pty Ltd. The authors also thank Kath
prevention of exposure to perennial allergens, especially
O’Driscoll for assistance with recruiting and running the trial and
those from house dust mites where the majority of
Dr Diana Bass for help with location of a suitable site. Thank you
allergen is carried on particles above 10 lm in diameter
to all the subjects who volunteered for this study.
port: allergic rhinitis and its impact on
airway resistance loads in older adults.
and safety of cetirizine, loratadine, and
placebo for seasonal allergic rhinitis.
7. Day J, Briscoe M, Wilditz M. Cetirizine,
Selner J. Allergic rhinitis: the patient’s
seasonal allergic rhinitis: effects after
13. Gotzsche P, Hammarquist C, Burr M.
management of asthma: meta-analysis.
MK. Recognition of pollen and otherparticulate aeroantigens by immunoblotmicroscopy. J Allergy Clin Immunol1988;82:608–616.
Publications Abelin, T., D. Pfluger, et al. (1993). "Vorstudie zur Frage gesundheitlicher Folgen des Kurzwellensenders Schwarzenburg." Bericht zuhanden des Bundesamt für Abelin, T., T. Schick, et al. (1991). "Ergebnisse der Untersuchung über die Benützer von Atemhilfsgeräten der Zürcher Liga für Lungenkrankheiten." Bericht zur Ackermann, D. K., R. Fuhrimann, et al. (1
DENOMINATION DU MEDICAMENT Myozyme® 50 mg poudre pour solution à diluer pour perfusion. 2. COMPOSITION QUALITATIVE ET QUANTITATIVE Un flacon contient 50 mg d’alpha alglucosidase. Après reconstitution, la solution contient 5 mg d’alpha alglucosidase* par ml et après dilution, la concentration varie de 0,5 mg à 4 mg/ml. *L’α-glucosidase acide humaine est produite par