Evaluation of Hirst-type spore traps in outdoor Aspergillaceae monitoring during large demolition work in hospital
Evaluation of Hirst-type spore traps in outdoor Aspergillaceae monitoring during large demolition work in hospital
Sophie Tiphaine Loeffert 2 3
Philippe Vanhems 1 2 3
Estelle Tissot 0 3
CeÂ dric DanancheÂ 1 2 3
Pierre Cassier 3
Thomas BeÂ net 1 3
Michel Perraud 3
Michel Thibaudon 3
Marie- Paule Gustin 2 3 4
0 Analyzair , Lyon , France , 4 Laboratoire de Biologie SeÂcuriteÂ Environnement, Groupement hospitalier Edouard Herriot , Lyon , France , 5 ReÂseau National de Surveillance AeÂrobiologique , Brussieu , France
1 UniteÂ d'hygiène, eÂpideÂmiologie et preÂvention, Groupement hospitalier Edouard Herriot , Lyon , France
2 Laboratoire des Pathogènes Emergents-Fondation MeÂrieux, Centre International de Recherche en Infectiologie (CIRI), Inserm U1111, CNRS UMR5308, ENS de Lyon , France, UniversiteÂ de Lyon 1 , France
3 Editor: Miguel Lopez-Ferber, Ecole des Mines d'Ales , FRANCE
4 DeÂpartement de santeÂ publique, Institut des Sciences Pharmaceutiques et Biologiques (ISPB)-FaculteÂ de Pharmacie , UniversiteÂ de Lyon 1 , France
Demolition can generate fungal spore suspensions in association with various adverse health effects, such as high risk of invasive aspergillosis in immunocompromised patients. One block of Edouard Herriot Hospital was entirely demolished. The aim of the present study was to evaluate Hirst-type spore traps utility in monitoring outdoor Aspergillaceae (Aspergillus spp. + Penicillium spp.) spores in part of Edouard Herriot Hospital (Lyon, France) undergoing major demolition. Three periods were scheduled in 2015: (A) Gutting of building and asbestos removal, (B) Demolition of floors, (C) Excavation and earthwork. Outdoor Aspergillaceae fungal load was monitored by cultivable (Air Ideal®, bioMeÂ rieux) and non-cultivable methods (Lanzoni VPPS-2000, Analyzair®, Bologna, Italy). Differences of Aspergillaceae recorded with Hirst-type spore traps were observed between Gerland and Edouard Herriot Hospital. Differences between Aspergillaceae were recorded between day time and night time at Gerland and Edouard Herriot Hospital. Daily paired differences between Aspergillaceae recorded with non-cultivable methodology at Edouard Herriot Hospital and in an area without demolition work were significant in Period A vs Period B (p = 10± 4) and Period A vs Period C (p = 10±4). Weak correlation of daily Aspergillaceae recorded by both methods at Edouard Herriot Hospital was significant only for Period C (r = 0.26, p = 0.048, n = 58). Meteorological parameters and type of demolition works were found to heavily influenced Aspergillaceae dispersion. Non-cultivable methodology is a promising tool for outdoor Aspergillaceae scrutiny during major demolition work in hospital, helping infection control staff to rapidly implement control measures.
Data Availability Statement: All relevant data are
within the paper and its Supporting Information
Funding: Laboratoires ANIOS provided funding in
the form of salaries to first author ST Loeffert.
Laboratoires ANIOS did not have any role in the
study design, data collection and analysis, decision
to publish or preparation of the manuscript.
Additionally, Analyzair provided funding in the form
of salaries to co-author Estelle Tissot. Analyzair did
not have any role in the study design, data
collection and analysis, decision to publish or
preparation of the manuscript. Analyzair is the
society to which we rent the Hirst collector. The lab
technician (Estelle Tissot) from Analyzair was
responsible for the implementation and control of
the Hirst collector at our hospital. The specific roles
of these authors are articulated in the 'author
Competing interests: Laboratoires ANIOS
provided funding in the form of salaries to first
author ST Loeffert. Analyzair provided funding in
the form of salaries to co-author Estelle Tissot. This
does not alter our adherence to PLOS ONE policies
on sharing data and materials.
Healthcare establishments are frequently confronted by infrastructure work. Construction
work, particularly demolition, generates 105-fold elevated outdoor levels of thermo-tolerant
fungi compared to non-demolition values and, consequently, augments the risk of invasive
aspergillosis (IA).in immunocompromised patients, a severe infection with 50±90% lethality
caused mostly by A. fumigatus (>80%)[1±3]. Specific guidelines have been published to
increase indoor hospital air control measures during construction [
]. A pilot study,
conducted at EHH in 2013, evaluated the use of Hirst-type spore traps (HTSTs) in environmental
fungal load monitoring of indoor and outdoor hospital units [
]. This work showed that
indoor HTSTs were not useful but confirmed previous results, suggesting that indoor fungal
contamination is highly influenced by outdoor, airborne spores [
]. Large-scale surveillance
systems are needed to detect outdoor fungal spores and alert hospitals to quickly implement
control measures. Surveys with spore traps are required to develop such warning systems [
Some studies have evaluated the reliability of different sampling methods (cultivable and
non-cultivable) in measuring fungal spore concentrations [8±14]. Each sampling method
seems to have advantages and drawbacks. Non-cultivable methods allow the sampling of
numerous spores, useful to carry out surveys. However, fungal spore's identification only
possible at the genus level [
]. Cultivable methods allow spore identification at the species level
but is time-consuming and depends on the substrate plated and culture conditions applied
]. Other factors, such as environment and meteorological parameters, may also affect the
survivability of fungal spores [
]. To the best of the authors' knowledge, no standardized
method of Aspergillus-Penicillium (Aspergillaceae) assessment has been developed by
comparing results with paired samples [
The main objective of the present study was to evaluate the utility of HTSTs for outdoor
Aspergillaceae monitoring during large demolition work in hospital. The secondary objectives
were to correlate the results of HTSTs with an agar impact sampler and to evaluate the impact
of meteorological parameters on Aspergillaceae aero-contamination measured by both
Materials and methods
This study was conducted at University of Lyon-affiliated Edouard Herriot Hospital (EHH,
850 beds) and at Gerland area located a few km south of the hospital and subject to similar
weather conditions (Fig 1A). An entire block, in the center of EHH, was entirely demolished.
The time period was separated into 3 consecutive periods between February and December
2015 (Fig 1B and 1C, video of the demolition work: https://www.youtube.com/watch?v=
Oa7xRufAnhQ). No major demolition work was undertaken in proximity to the Gerland area
which served as control site.
Airborne fungal sampling was undertaken simultaneously, by cultivable and non-cultivable
methods. Outdoor data were recorded continuously by the non-cultivable method, using
HTSTs samplers (Lanzoni VPPS-2000 samplers Analyzair1, Bologna, Italy), in the Gerland
area, where no demolition works happened during the study and EHH. They were placed on
the building roof respectively 35 m above ground level in Gerland and 10 m in front of the
EHH demolition site. Aspergillaceae (Aspergillus spp. + Penicillium spp.) fungal load (AFL),
impacted on adhesive tape, placed on a drum, were counted by microscopy every 2 h and
2 / 10
Fig 1. Study site. (A) Sampler locations in Lyon, France, (B) Edouard Herriot Hospital, and (C) demolition work schedule.
expressed as spores per cubic meters every 2 h (spores/m3/2-h) or per day (spores/m3/day).
The number of Aspergillaceae spores counted was multiplied by a conversion factor of 0.19,
corresponding to a reading of 36.5% of the surface sampled.3 HTSTs drums were changed
weekly over 42 weeks. Inlet mean flow rate was 10 L.min-1 with the non-cultivable method.
With the cultivable method, a daily environmental survey of AFL with agar impact sampler
(Air Ideal1 90 mm, bioMeÂrieux) was undertaken outdoor at EHH all around the demolition
site (100-L samples). Outdoor samplings were realized four days per week at EHH in the
morning (between 10 a.m-12 noon) and afternoon (between 2±4 p.m). Each day, 3 outdoor
sites corresponding to the porch of 3 different units were monitored in the morning and
afternoon. Each sample was gathered by agar impact sample in 90-mm diameter Petri dishes
containing Sabouraud Chloramphenicol agar. Air intake velocity of this agar impact sampler was
100 L/min. Two plates were seeded at each sample site. Each plate was seeded for 1 min,
resulting in air volume of 100 L. One of these plates was incubated for 48 h at 37ÊC to grow
thermotolerant A. fumigatus species [
]. The other plate was incubated for 5 days at 30ÊC to allow
growth of all fungi. Colonies were expressed as spores and colony-forming units per cubic
meter (CFU/m3) [
Meteorological parameters were recorded by Meteociel, a meteorological agency, throughout
the study period at a station located only a few km from EHH (Fig 1A). Relative humidity (%),
temperature (ÊC), wind direction, rain (mm), atmospheric pressure (hPa) and speed (km.h-1)
were monitored every hour throughout 2015.
3 / 10
Data were expressed as medians and interquartile range (IQR). Continuous non-cultivable
Aspergillaceae data were cumulated per day (over 24 hours). Differences of daily data between
EHH and Gerland were compared between the 3 study periods taken pairwise using Wilcoxon
test. Non-cultivable AFL data recorded every 2 h were averaged for each day over day time (7
a.m.-7 p.m) and night time (9 p.m.-5 a.m.) and compared between day/night and
EHH/Gerland using paired Wilcoxon tests with adjusted p-values for 4 comparisons.
To compare non-cultivable method and cultivable method results for AFL at EHH
Wilcoxon test were used. For the non-cultivable method, only data corresponding to the sampling
times of cultivable method were used (each day between 10 a.m-12 noon and 2±4 p.m). Data
correlations were analyzed with Spearman correlation coefficients: i) between AFL recorded
with the two methods, and ii) between AFL per day (over 24 h) from both HTSTs (EHH and
The impact of meteorological variables on AFL presence estimated in EHH by both
methods for the 3 periods was evaluated by fitting logistic regressions with forward selection of
variables for each period. In case of significant interaction between 2 meteorological variables,
logistic regression was performed in sub-groups, to ease odds ratio (OR) interpretation.
Meteorological findings at 11 a.m. and 3 p.m. were paired in the cultivable method with daily
samplings in the morning and afternoon. P-values <0.05 were considered to be statistically
significant. In case of multiple comparisons, adjusted p-values were computed with the Holm
All data were collected over a period of 42 weeks. Table 1 reports medians and IQR of outdoor
fungal loads sampled by cultivable and non-cultivable methods during the 3 demolition
periods. Differences were observed with each sampler according to sampling site and demolition
Description of airborne fungi loads with cultivable and non-cultivable
With the cultivable method, a total of 511 air samples were collected outdoor at EHH. The
highest median AFL concentrations were observed in Period C (30 CFU/m3, IQR:80) during
Periods of demolition work
excavation and earthwork. AFL concentration median ranged between 0 (IQR:20) and 30
CFU/m3 (IQR:80) in the 3 periods. Penicillium median ranged were of 0 CFU/m3 in the 3
Data were collected according to the non-cultivable method for 296 days (42 weeks) at
EHH and Gerland. The highest median AFL concentration at EHH was recorded in Period B
(median = 23.5 spores/m3/day, IQR:24.75) during demolition of the building floors. In
contrast, the highest AFL median at Gerland occurred in Period A (median = 15 spores/m3/day,
IQR:35). AFL medians at EHH and Gerland ranged from 12 (IQR:10) to 23.5 (IQR:24.75) and
from 2 (IQR:12) to 15 (IQR:35) spores/m3/day, respectively, in the 3 demolition periods.
Comparisons of Aspergillaceae loads in EHH and Gerland
Comparison of demolition time periods. Fig 2 reports apparent time period variations
in AFL contamination of the non-cultivable spore traps located at EHH and Gerland. AFL
medians were observed qualitatively to be much higher at EHH than at Gerland during all
study periods. Daily paired AFL differences between the 2 sites (EHH and Gerland) were
significantly different between Period A vs Period B (median (IQR) of EHH-Gerland differences:
1 (24.3) vs 13.5 (31.8), p<10−4), Period A vs Period C (1 (24.3) vs 8 (13.5), p<10−4) and Period
B vs Period C (13.5 (31.8) vs 8 (13.5), p = 0.032).
Day and night time data per site. At EHH, mean AFL medians were higher during the
day (7 a.m.-7 p.m. with every 2-h data) than the night (9 p.m.-5 a.m.) (median (IQR): 1.37 vs
1.09 (1.51 vs 1.42) spores/m3/2-h, p = 0.007). A similar trend was observed at Gerland (median
Fig 2. Median of Aspergillaceae levels every 2-h by the non-cultivable method at Gerland and EHH during the 3
5 / 10
(IQR): 0.57 vs 0 (2.14 vs 1.48), p = 0.007), attributed to some high AFL (8 peaks>15 spores/m3/
2-h) in the day time. Statistically significant differences in day time data were evident between
EHH and Gerland, with a higher median at EHH (median (IQR): 1.37 (EHH) vs 0.57 (1.51 vs
2.14) (Gerland) spores/m3/2-h, p = 0.010). AFL data were also higher during the night at EHH
than at Gerland (median (IQR): 1.09 (EHH) vs 0 (1.42 vs 1.48) (Gerland) spores/m3/2-h,
Cultivable and non-cultivable sampling methods at EHH
Fig 3 reports the mean daily concentrations of AFL colonies and spores (CFU/m3 and spores/
m3/24-h) recorded by both methods during the 3 demolition periods at EHH. A weak
correlation between daily AFL recorded by the two methods was observed during Period C (r = 0.26,
p = 0.0475, n = 109 days), corresponding to excavation and earthwork. During this period,
higher peaks were recorded by the cultivable than by the non-cultivable method (Fig 3). No
correlation between both methods was observed during Periods A and B (p = 0.6 and p = 0.2,
respectively). In Period A, both methods showed similar orders of peak magnitudes. In Period
B, higher peaks were observed with the non-cultivable method. The cultivable method
produced no significant peaks.
Evaluation of meteorological impact at EHH
Meteorological variables differed, on average, in the 3 study periods (S1 Table) Table 2
summarizes the results of final, simplified logistic modeling. With the cultivable method, the effects
of relative humidity and wind speed on AFL presence were additive in Period A. When relative
humidity increased by 10%, the OR of Aspergillaceae presence was multiplied by 1.25 (p =
0.035). When wind speed decreased by 10 km/h, the OR of Aspergillaceae presence increased
by a factor of 2.04 (1/0.49) (p<0.001). In Period B, meteorological variables had no significant
effect on AFL. In Period C, when temperature increased by 10ÊC, the OR of Aspergillaceae
presence rose 1.93-fold (p = 0.050). With the non-cultivable method, in Period A, in the
presence of south or southwestern winds, only temperature had a significant effect on AFL. OR
decreased by a factor of 1.41 (1/0.71, p = 0.041) when temperature rose by 10ÊC. With other
Fig 3. Daily average concentrations of Aspergillaceae spores and colonies (spores/m3/day and CFU/m3) recorded by
cultivable and non-cultivable methods during demolition at EHH, in 2015.
6 / 10
1Adjusted OR in case of at least 2 explicative variables
Only univariate regression in the subgroup with south or southwestern winds
OR is given for 10-unit augmentation in case of continuous meteorological variables
wind directions, both temperature and atmospheric pressure had significant, additive effects
on Aspergillaceae presence whose OR increased 1.79-fold when atmospheric pressure rose by
10 hPa (p<10−4). The effect of temperature was similar (OR = 0.71, p<10−3). In Periods B and
C, only relative humidity had a significant impact on Aspergillaceae whose OR increased by a
factor of 1.21 (p<10−4) and decreased by a factor of 1.08 (1/0.92, p = 0.03), respectively, when
relative humidity rose by 10%.
This study revealed: i) the capacity of HTSTs to detect Aspergillaceae aero-contamination of
the surrounding environment, ii) weak correlation (trend) between AFL recorded by
cultivable and non-cultivable methods, and iii) consistent influence of environmental factors, such as
demolition work and meteorological variables, on AFL.
HTSTs seem to be able to detect variations of outdoor AFL, as described in previous studies
]. Higher AFL was detected at EHH which underwent major demolition compared
to Gerland, the control site. These data highlighted the importance to be careful in protecting
immunocompromised patients when they are outside units during demolition at hospital.
Similarly, higher AFLs were recorded at EHH during days than during nights. Even if AFL at
EHH were lower during the night than the day, aerocontamination still stayed higher than at
Gerland area. So period of opening windows during nights have to be realised with caution,
during limited times.
These samplers were highly influenced by environmental conditions and factors, as proved
by differences observed between the Gerland area (control) and EHH [
]. The present
study identified demolition work as the potential source of AFL. At EHH site, some peaks
were observed in the morning and/or the afternoon. A possible explanation to those peaks
could be the presence of high activity of the demolition site at these times. Distinct AFL peaks
according to each demolition period were seen with both sampling methods.
To the best of the authors' knowledge, this is the first report of weak but significant
correlation between both sampling methods of AFL monitoring during earthwork (Period C). As
AFL recorded by cultivable sampling was found to be higher in Period C, the results confirmed
previous studies which suggested that high concentrations of spores and numerous sampling
measures are needed to find statistical relationships between both methods [
]. The kind of
7 / 10
demolition works could explain these high AFL levels. Excavation and earthwork were done at
ground level or just below, at the height where the cultivable sampler was recording. In
contrast, weak AFL contamination was monitored by the non-cultivable method during this
period because of the height separating the collector and the demolition works. The absence of
correlation in other periods could be explained by the location/height of the samplers
according to type of demolition work undertaken. Some experts suggest that particles concentrations
decrease with height, possibly owing to dispersion phenomena [
]. During Period A, only
inside demolition work was done. The absence of correlation may be due to low spore
concentrations observed with the cultivable method, possibly because of its position between
buildings  In Period B, the non-cultivable method showed higher AFL than the cultivable
method, explained by type of demolition work occurring during this period, i.e., demolition of
building floors. Demolition work was done approximately 15 m above ground level,
corresponding to the height at which the HTSTs were located. The cultivable sampler located at
ground level was unable to detect variations due to these demolition works. It may explain the
lack of correlation between both methods. In the present study, the type of demolition work
highly impacted AFL monitored by both methods because of differences in height.
Meteorological effects on AFL data were found consistent with literature [
Aspergillus species are known to gain optimal growth between 30 and 35ÊC with average humidity of
70%. Temperature and relative humidity were closely related to AFL growth and reproduction
capacity. Each demolition period highly corresponded to a season, heavily influencing data
analysis and could help future studies in predicting and modeling AFL peaks.
The main strength of this study resides in the numerous data obtained with both methods,
allowing to answer important questions in environmental monitoring during construction
works and helping infection control staff in planning precautions for immunocompromised
patients. It does, however, have some limitations: the demolition periods were concomitant
with the seasons which could have induced seasonality bias in analysis. No baseline period
without demolition work at EHH could be studied.
It is known that air flow from outdoors can be found indoors, increasing the risk for the
patient to develop IA [
]. To limit exposure, rapid detection of contamination peaks is needed
to implement adequate prevention measures and policies. This study permit to confirm the
careful attention needed to be taken in protecting immunocompromised patients when they
are outside units during demolition. Analysis of day and night AFL variations showed that
although lower AFL were found during the night at EHH, loads still staid higher than control
area. So, to ensure safety of patients, limited period of opening windows during night have to
be recommended. Furthermore, this study give some new insights to improve environmental
monitoring guidelines by showing importance of samplers height position in regards to king
of demolition work applied.
Further studies, including both methods at different ground levels, are needed to better
evaluate sensitivity and reliability as warning systems. Finally, our study determined that
outdoor HTSTs are promising tool for outdoor AFL monitoring during major demolition works
in hospital, helping clinical infection control practitioners to rapidly implement control
S1 Table. Relationship between each meteorological variable and the 3 study periods.
p<0.001; p< 0.01. Multiple comparisons were made between study periods for each
8 / 10
The authors thank Laboratories ANIOS, France and Analyzair1 for invaluable advice.
Conceptualization: Sophie Tiphaine Loeffert, Philippe Vanhems, Thomas BeÂnet, Marie-Paule
Formal analysis: Estelle Tissot, Marie-Paule Gustin.
Investigation: Sophie Tiphaine Loeffert.
Methodology: Sophie Tiphaine Loeffert, Estelle Tissot, Michel Perraud, Michel Thibaudon,
Project administration: Sophie Tiphaine Loeffert.
Resources: Pierre Cassier, Michel Perraud.
Software: Marie-Paule Gustin.
Supervision: Philippe Vanhems, Marie-Paule Gustin.
Validation: Philippe Vanhems, Marie-Paule Gustin.
Writing ± original draft: Sophie Tiphaine Loeffert.
Writing ± review & editing: Philippe Vanhems, CeÂdric DanancheÂ, Thomas BeÂnet, Michel
Thibaudon, Marie-Paule Gustin.
9 / 10
1. Vonberg RP , Gastmeier P . Nosocomial aspergillosis in outbreak settings . J Hosp Infect 2006 ; 63 : 246e254 .
2. Streifel AJ , Lauer JL , Vesley D , Juni B , Rhame FS . Aspergillus fumigatus and other thermotolerant fungi generated by hospital building demolition . Appl Environ Microbiol 1983 ; 46 : 375 ± 8 . PMID: 6354086
3. Marr KA , Patterson T , Denning D. Aspergillosis . Pathogenesis, clinical manifestations and therapy . Infect Dis Clin North Am 2002 ; 16 : 875 ± 94 . PMID: 12512185
4. Gangneux JP , AdjideÂ CC , Bernard L , Botterel F , Carel A , Castel O et al. Quantitative assessment of fungal risk in the case of construction works in healthcare establishments: proposed indicators for the determination of the impact of management precautions on the risk of fungal infection . J Mycol Med 2012 ; 22 : 64e71 .
5. DanancheÂ C , Gustin MP , Cassier P , Loeffert ST , Thibaudon M , BeÂnet T et al. Evaluation of hirst-type spore trap to monitor environmental fungal load in hospital . Plos One 2017 ; 12 ( 5 ):e0177263. https://doi. org/10.1371/journal.pone. 0177263 PMID: 28486534
6. Brenier-Pinchart MP , Lebeau B , Quesada JL , Mallaret MR , Borel JL , Mollard A et al. Influence of internal and outdoor factors on filamentous fungal flora in hematology wards . Am J Infect Control 2009 ; 37 ( 8 ): 631 ±7. https://doi.org/10.1016/j.ajic. 2009 . 03 .013 PMID: 19631408
7. Grinn-Gofron A , Strzelczak A . Artificial neural network models of relationships between Alternaria spores and meteorological factors in Szczecin (Poland) . Int J Biometeorol 2008 ; 52 : 859e868 .
8. Grinn-Gofron A . Airborne Aspergillus and Penicillium in the atmosphere of Szczecin, (Poland) ( 2004e2009 ). Aerobiologia 2011 ; 27 : 67e76 .
9. Elvira-Rendueles B , Moreno J , Garcia-Sanchez A , Vergara N , Martinez-Garcia MJ , Moreno-Grau S . Air-spore in Cartagena, Spain: viable and non-viable sampling methods . Ann Agric Environmental Med 2013 ; 20 ( 4 ): 664 ± 71 .
10. FernaÂndez-RodrÂõguez S , Molina RT , Palacios IS , Garijo AG . Two sampling methods for the Petri dish detection of airborne fungi . Grana 2011 ; 50 : 202 ± 7 .
11. Tormo Molina R , Palacios SI , Gonzalo Garijo A , Muñoz RodrÂõguez AF , FernaÂndez-RodrÂõguez S , Recio Aguado D. Use of personal spore traps to complement continuous aerobiological monitoring . Grana 2010 ; 49 : 134e141 .
12. Hirst JM . An automatic volumetric spore trap . Ann Appl Biol 1952 ; 39 : 257 ± 65 .
13. Bellanger AP , Reboux G , Scherer E , Vacheyrou M , Millon L . Contribution of a cyclonic-based liquid air collector for detecting Aspergillus fumigatus by QPCR in air samples . J Occup Environ Hyg . 2012 ; 9(1): D7±D11 . https://doi.org/10.1080/15459624. 2012 .636727 PMID: 22150297
14. Lang-Yona N , Dannemiller K , Yamamoto N , Burshtein N , Peccia J , Yarden O et al. Annual distribution of allergenic fungal spores in atmospheric particulate matter in the Eastern Mediterranean; a comparative study between ergosterol and quantitative PCR analysis . Atmos Chem Phys 2012 ; 12 : 2681 ± 2690 .
15. FernaÂndez-RodrÂõguez S , Tormo-Molina R , Maya-Manzano JM , Silva-Palacios I , Gonzalo-Garijo A . Outdoor airborne fungi captured by viable and nonviable methods . Fungal Ecology 2014 ; 7 : 16 ± 26 .
16. Abdel Hameed AA , Khoder MI , Ibrahim YH , Saeed Y , Osman ME , Ghanem S. Study on some factors affecting survivability of airborne fungi . Sci Total Environ 2012 ; 414 : 696 ± 700 . https://doi.org/10.1016/j. scitotenv. 2011 . 10 .042 PMID: 22137479
17. Boff C , Brun CP , Miron D , Zoppas BC , Pasqualotto AC . Technical note: The effect of different incubation temperatures on the recovery of Aspergillus species from hospital air . Am J Infect Control 2012 ; 40 : 1016 ±7. https://doi.org/10.1016/j.ajic. 2012 . 01 .029 PMID: 22683029
18. Holm SA . Simple sequentially rejective multiple test procedure . Scand J Stat 1979 ; 6 : 65 ± 70 .
19. Tormo Molina R , Maya Manzano JM , FernaÂndez-RodrÂõguez S , Gonzalo Garijo A , Silva Palacios I. Influence of environmental factors on measurements with Hirst spore traps . Grana 2013 ; 52 ( 1 ): 59 ± 70 .
20. Bryant RH , Emberlin JC , Norris-Hill J . Vertical variation in pollen abundance in north-central London . Aerobiologia 1989 ; 5 : 123 ± 37 .