Assessment of bioaerosol particle characteristics at different hospital wards and operating theaters: A case study in Tehran

Graphical abstract


Method details
Exposure to bioaerosols, as an important class of particulate matter (PM), have become one of the most issues in indoor air quality (IAQ) topics for occupational and public health [1][2][3][4]. In particular, hospitals have a more complex and different environment rather than other occupational or residential ones. This is due to continuous working cycle, and special types of air pollutants, including pathogenic and chemical or biological agents [5]. Bioaerosols can originate from humans (including patients), various indoor hospital characteristics, and outdoor environmental sources [2]. It has been found that about 5-34% of indoor air pollution in different indoor environments such as hospitals, dentist offices, shopping centers, subway systems, public libraries, and other workplaces can be attributable to bioaerosols, particularly bacterial and fungal bioaerosols [3,6]. Bioaerosols are compounds with biological origin, including dead or live bacterial and fungal spores, bacterial peptidoglycans and endotoxins, fungal mycotoxins and β (1, 3)-glucans, viruses, pollen grains, airborne algae, plant debris, and fragments of microbial insects, fur fibres and skin fragments from animals and humans [7][8][9]. Among this wide range of species, bacterial and fungal aerosols have been introduced as the most important bioaerosols [3]. Bioaerosols in general, have been associated to several adverse health effects. Considering the significance of this issue, the World Health Organization (WHO) published a guideline on IAQ, in 2009 [10]. Well-documented environmental health studies demonstrate that exposure to bioaerosols, exceeding maximum acceptable levels can lead to noninfectious diseases, infectious diseases, including nosocomial infections or healthcare associated infections, acute toxic effects, cancer and even death, especially for immunosuppressed persons [1,[11][12][13][14][15][16][17][18]. Nosocomial infections cause significant health and economic problems [19,20]. About 1.4 million people are suffering from such infections at any one time worldwide [21]. The consequences of these infections may include extended hospital stay, long-term disability, increased resistance to antibiotics, a huge financial burden for government and patients, and excess deaths [22]. Several factors can affect the number and types of bioaerosols in hospital environments. The density and diversity of biological contaminants in hospitals depend on various factors such as the type of ward, the number and activity of visitors and patients, hospital room design, disinfection, ventilation, temperature and humidity, etc. [23,24]. Accordingly, assessment of bioaerosols in hospital environments, particularly in high risk operating theatres where patients are more susceptible is highly essential [25]. Despite all these studies, many issues remain debating. First, identifying the most high-risk wards, and the dominant species of microorganisms are always critical. Second, different site-specific factors can affect the type and number of bioaerosols. The effect of number of active beds is a less-documented factor. Third, antimicrobial resistance (AMR) can be detected by these studies [22,26]; since, the presence of settleable bioaerosols can be an indicator for effectiveness of disinfection processes in hospitals. Therefore, it is never repetitive to study bioaerosols in hospitals quantitatively and qualitatively. Regarding these issues, this study was perform to investigate the density and diversity of bacterial and fungal bioaerosols in different wards and operating theaters of a high crowded hospital (Khatam-ol-Anbia) in Tehran, Iran. In addition, the effect of number of active beds on bioaerosols concentrations were investigated.

Sampling sites
This cross-sectional study was conducted at different wards of specialty and subspecialty Khatam-Al-Anbia hospital, located on Tehran from August 2015 to February 2016. All sampling sites were first categorized into 2 groups, which are hospital wards including CCU (coronary care unit), GICU(general intensive care unit), ICU(intensive care unit), and NICU(neonatal intensive care unit) and operating theaters including MS(men surgery), NS(neonatal surgery), and WS(women surgery). Then measurements of bacterial and fungal bioaerosol concentrations were conducted in all sites including CCU, GICU, ICU, NICU, MS, NS and WS.

Sampling procedure
Passive sampling provides a valid risk assessment as it measures the harmful part of the airborne population which falls onto critical surfaces [27]. It is the most readily available, economic, and unobtrusive method of bioaerosol sampling, and is based on particles settling by means of gravity, on a collection substrate housed in a settle plate. The concentration is usually expressed as the number of colony-forming units (CFU) within the area of the settling plates for the duration of a specified timeperiod (for example in units of CFU/m 2 /h) [28]. This is also known as the index of microbial air contamination (IMA) [29]. Passive method captures only the fractions of bacteria and fungi that can sediment in a specific period of time. This fraction can be very critical in hospital environments, because bacteria and fungi can contaminate different surfaces and equipment such as surgical cut in operating theatres [29,30]. To determine the Index of Microbial Air Contamination (IMA) in hospital wards and operating theaters, passive sampling was performed [25,[31][32][33], because there was no official permission to use any active sampling equipment. The sampling was performed from 10:00 AM to 16:00 PM, during a six months' period. 90-mm petri dishes containing Tryptic cycloheximide soy agar and Sabouraud chloramphenicol dextrose agar were used for bacterial and fungal samples, respectively [24,[34][35][36] The 1/1/1 scheme as a standardized method was used to collect bioaerosols [28]. The plates containing the sterile culture medium were located for sampling at the height of 1 m above the ground level and with a distance more than 1 m from the walls, doors, windows and barriers for 1 h. At the end of the sampling, the petri dishes were immediately transferred to the laboratory and incubated at 37 AE 1 C for 2 days for bacteria [37,38] and at 25 AE 1 C for 4 days for fungi [39][40][41]. 36 samples were taken from each site, including 18 samples for fungal, and 18 samples for bacterial analyses. In total, 252 samples (126 bacterial samples and 126 fungal samples) were collected for bacterial and fungal bioaerosols. Bacterial and fungal samplings were simultaneously conducted in all sampling sites. In addition, the numbers of the patient beds in all sampling sites was recorded by observation in each sampling day to assess the association between the numbers of the beds and the concentrations of bacterial, fungal and total bioaerosols.

Identification of bacterial and fungal bioaerosols
After adding 500 mg/L of cycloheximide to inhibit the proliferation of fungi, bacterial samples were incubated at 37 C for at 24-48 h. Genera of bacteria was identified according to Bergey's manual. After Gram staining, a biochemical test was carries out to identify genera using an automatic identification system -VITEK (Model VITEK 32 system, bioMerieux Inc., France) [36,42]. After adding 100 mg/L of chloramphenicol to inhibit the proliferation of bacteria, fungal samples were incubated at 25 C for 72 h. An optical microscope with magnification of 400Â was used to assess the shape and color of colonies and spores, vegetative hyphae, and sexual and asexual reproductive organs. The identification of genera was accomplished based on the taxonomic method of Ainsworth and Baron [2,34]. Finally, the results were expressed in colony forming units per square meter per hour (CFU/m 2 /h) [10,25].

Data analysis
Descriptive statistics were calculated to describe the concentration of bacterial and fungal aerosols. To evaluate differences between bacterial, fungal and total bioaerosols (sum of the bacterial and fungal bioaerosols) in different sampling sites, one-way ANOVA test was employed. In addition, a simple linear regression model was used to evaluate the association between the numbers of beds and bioaerosol concentrations. All the statistical elaborations were conducted using Minitab software version 17. P-values less than 0.05 were considered as statistically significant in all analyses.

Concentration of bacterial and fungal bioaerosols
A summary of statistics for the bacterial and fungal bioaerosol (CFU m -2 h À1 ) concentrations in the sampling sites are given in Table 1. A total of 746 and 875 CFU of fungal and bacterial bioaerosol from 252 plates (126 SDA plates and 126 TSA plates) were collected by passive sedimentation technique in all the sampling sites. As can be seen in Table 1, the bacterial, fungal and total bioaerosol levels varied on a large scale within the sampling sites. According to Table 1 respectively. Based on the average concentrations, the total fungal bioaerosols were higher than the bacterial bioaerosols in the CCU, MS, and WS, whereas total bacterial bioaerosols were higher than the fungal bioaerosols observed in the NICU, GICU, ICU, and NS. There was a considerable statistic difference in the concentration of fungal bioaerosols in the hospital wards and operating theaters (p < 0.05). The same difference was found for bacterial bioaerosols between hospital wards and operating theaters (p < 0.05). In overall, the average of total bacterial concentration (919 CFU m -2 h À1 ) was significantly higher than fungal bioaerosol (796 CFU m -2 h À1 ) in current study (Table 1). Our results are consistent with those of [2,10,34,36,40,[43][44][45], who showed that bacterial bioaerosol concentrations were higher than fungal species. This finding can be attributed to the larger number of bacterial bioaerosol than fungal species in natural resources, such as the soil [46], and the air over vegetated regions [47]. On the other hand, the higher sedimentation of fungal bioaerosols due to their larger aerodynamic diameter (1-30 mm) compared with bacterial bioaerosols (< 2.5 mm) can change the contribution of settled bacterial and fungal species to the total burden of microbial species [8,10]. Fig. 1shows the ratios of bacterial/fungal (B/F) bioaerosol in all the sampling sites. The mean value of bacterial/fungal (B/F) bioaerosol ratio during the study ranged from 0.89 to 2.78, with a value of 1.55, 2.78, 1.08, 0.89, 0.99, 1.04, and 2.51 for NICU, GICU, ICU, CCU, MS, WS, and NS, respectively. Therefore, the B/F ratio in NICU, GICU, ICU, WS, NS were above 1.0. Likewise, the highest B/F bioaerosol concentrations ratio observed in GICU (2.78) and the lowest values recorded in CCU (0.89). Our findings indicated that the mean total bacteria/total fungi concentration ratio during the study was 2 (Fig. 1).

Number of beds
We found the highest fungal bioaerosol concentrations in CCU. This result can be attributed to the larger numbers of the beds in CCU. As shown in Fig. 3, linear regression analysis revealed a significant strong association between fungal bioaerosols and the numbers of beds (R 2 = 0.76, p < 0.05) [44,50]. On the other hand, a poor association was found between the concentrations of bacterial bioaerosols and the numbers of beds (R 2 = 0.02, p < 0.05), indicating that bacterial bioaerosols concentrations in the sampling sites have been affected from other parameters other than the number of beds. These factors may include the number of health-care workers, visitors and patients' coughing and sneezing [2,44,58,59]. Fig. 4 shows the boxplots of the total bioaerosol concentrations in the sampling sites compared to the available standards. As illustrated in Fig. 4, total bioaerosol concentrations indicated a statistically noticeable difference (p < 0.05) in all sampling sites, especially between CCU and NS. Since no national guidelines or standard limits for the index of microbial air contamination are provided by Iranian official documents, we considered the Swiss Hospital Association standards and other standards available as maximum levels of the index of microbial air contamination (MAL of IMA) in operating theatres (red line, 786 CFU m -2 h À1 or 5 CFU/90 mm diameter plate/h), the optimal (black long  (Fig. 4). In addition, as shown in Fig. 4 the mean total bioaerosol concentrations in the CCU, ICU, GICU, and NICU wards were 2.93, 2.43, 2.29, and 2.23 times higher than the AL of IMA, respectively; whereas, 4.84, 4.04, 3.81, and 3.71 times higher than the OL of IMA. In case of operating theaters, the mean total bioaerosol concentrations in the MS, WS, and NS were 2.19, 1.90, and 1.77 times higher than the MAL of IMA, respectively. Thus, we could conclude that the investigated indoor hospital environments were heavily contaminated with airborne bacterial and fungal bioaerosols at a high level. Therefore, patients, staff, and other subjects were heavily exposed to total bioaerosol concentrations exceeded the standards available in the whole sampling sites.

Conclusion
The present study was carried out to investigate the compositions and the concentrations of airborne bacterial and fungal bioaerosols in hospital wards and operating theaters in Tehran. Bacterial and fungal bioaerosols were isolated from all the samples. Our results showed that concentrations of bacterial, fungal and total bioaerosols at these sites were highly variable, especially for fungal bioaerosols. In addition, our findings suggested that the numbers of the beds have a significant positive and strong effect on airborne fungal concentrations. Despite their significant impact on human health, hospital airborne bioaerosols and their affecting factors are not well understood. Comprehensive studies are necessary to determine the role of other factors including the number of health-care workers and visitors, coughing, sneezing, etc. on the concentrations of airborne bioaerosols. The results of this study can be used to set indoor air quality standards for hospital wards and operating theatres, especially in case of maximum allowable number of beds.

Conflict of interest
The authors declare that they have no conflict of interest.