Exposure to Airborne Endotoxins among Sewer Workers: An Exploratory Study

Transcription

1 Ann. Occup. Hyg., 2014, 1 11 doi: /annhyg/met085 Exposure to Airborne Endotoxins among Sewer Workers: An Exploratory Study Philippe Duquenne 1 *, Denis Ambroise 2, Pierre Görner 1, Frédéric Clerc 1 and Guylaine Greff-Mirguet 1 1.Département Métrologie des Polluants, INRS, 1 rue du Morvan CS 60027, Vandœuvre-lès-Nancy Cedex, France 2.Pôle Médico-Social des Côtes de Meuse, 74 rue R. Poincaré, Vigneulles-les-Hattonchâtel, France *Author to whom correspondence should be addressed. Tel: ; fax: ; philippe.duquenne@inrs.fr Submitted 20 June 2013; revised 16 December 2013; revised version accepted 16 December Abstract Exploratory bioaerosol sampling was performed in order to assess exposure to airborne endotoxins during sewer work. Personal samples were collected in underground sewer pipes using 37-mm closed-face cassettes containing fibreglass filters (CFC-FG method) or polycarbonate filters (CFC-PC method). Endotoxins were quantified using the limulus amoebocyte lysate assay. Concentrations of airborne endotoxins at sewer workplaces ( EU m 3 ) were higher than those measured outside the sewer network ( EU m 3 ). Sewer worker exposure to airborne endotoxins depended on the workplace and on the tasks. Exposure levels were the highest for tasks involving agitation of water and matter, especially for chamber cleanup and pipes cleanup with a high-pressure water jet. Airborne endotoxin levels at the workplace tended to be higher when CFC-FG was used as the sampling method rather than CFC-PC. The adjusted mean of the measured concentrations for CFC-PC represents 57% of the mean observed with CFC-FG. The number of samples collected in the descriptive study was too low for drawing definitive conclusions and further exposure investigations are needed. Therefore, our exploratory study provides new exposure data for the insufficiently documented sewer working environment and it would be useful for designing larger exposures studies. Keywords: bioaerosols; endotoxin; occupational exposure; sewer; wastewater Introduction Sewage systems are essential for draining and carrying raw wastewater from municipalities. Sewer systems are complex networks made up of underground pipe galleries that run below city streets (Pfafflin and Ziegler, 2012). They collect rainwater, domestic wastewater, and industrial effluent for treatment and disposal at wastewater treatment facilities. Manholes afford ready access to larger sewer pipes when they intersect or change direction (Pfafflin and Ziegler, 2012). Raw wastewater carries a large amount of suspended matter (sand, sludge, grease, and waste) that may deposit in pipes and disturb the water flow or obstruct the pipes. Sewers also deteriorate over time and thus, the sewer network needs to be periodically cleaned out and repaired. These operations are generally performed by trained employees. Sewer pipes are enclosed spaces where employees can be exposed to chemical pollutants such as hydrogen sulphide (Richardson, 1995). Raw The Author Published by Oxford University Press on behalf of the British Occupational Hygiene Society. 1

2 Page 2 of 11 Exposure to Airborne Endotoxins among Sewer Workers wastewater also contains a large quantity and a wide variety of microorganisms including bacteria, fungi, viruses, protozoa, helminths as well as their metabolites and products (Gerardi and Zimmerman, 2005). Depending on the biological agents and on the route of exposure, these microorganisms can induce biological risks among exposed workers. Indeed, epidemiological investigations suggest a higher prevalence of gastrointestinal tract symptoms, respiratory symptoms, fatigue, and headache among employees at sewage treatment plants (Lundholm and Rylander, 1983; Rylander, 1999; Thorn and Kerekes, 2001; Thorn and Beijer, 2004). Infectious diseases caused by bacteria and viruses (hepatitis A) are also reported but their occurrence seems to be restricted to cluster groups of workers and related to general hygiene behaviours (Thorn and Kerekes, 2001; Jeggli et al., 2004). Airborne endotoxins and Gram-negative bacteria were suggested as possible causative agents of airway and intestinal symptoms among workers in wastewater treatment plants (Rylander, 1999; Thorn and Beijer, 2004). Endotoxins are a major cell wall component of Gram-negative bacteria that are ubiquitous in the environment and may become airborne particles during normal working operations in sewers. Exposure to airborne endotoxins has been associated with workrelated respiratory and general symptoms in different occupational environments (Rylander, 2002; Douwes et al., 2003; Liebers et al., 2006). Thus, the assessment of exposure to endotoxins has received much attention and the concentration of airborne endotoxins has been documented for a wide range of workplaces (Duquenne et al., 2013). For work with sewage, most of the published studies were carried out in wastewater treatment facilities (Laitinen et al., 1994; Laitinen et al., 2001; Thorn et al., 2002; Oppliger et al., 2005; Smit et al., 2005; Visser et al., 2006; Spaan et al., 2008) and data regarding work in sewer pipes are scarce (Neumann et al., 2002). As a consequence, the exposure of sewer workers to endotoxins is as yet insufficiently documented and further research is needed to fill this gap in knowledge. Sampling on filters is the most widely used method in the literature for airborne endotoxins (Duquenne et al., 2013). But the diversity of sampling system (i.e. filter holder + filter) configurations as well as of analytical protocols used in studies may induce variability in exposure estimates from one laboratory to another (Reynolds et al., 2002; Chun et al., 2006; Spaan et al., 2007; Thorne et al., 2010). In particular, numerous published studies use polycarbonate membranes as sampling filters (Duquenne et al., 2013), in spite of Standard EN recommending that fibreglass filters (CEN, 2003) be used. The type of bioaerosol being sampled and the type of sampling filter used have been suggested as important factors in the overall efficiency of the endotoxin measurement process (Gordon et al., 1992; Reynolds et al., 2002). Polycarbonate membranes have been shown to underestimate the endotoxin levels measured in various occupational environments but no data are available for work in sewers. Exploratory measurements of bioaerosols were performed in the underground sewer pipes of a city in France. The aim of the study was to assess the exposure of sewer worker to airborne endotoxins. Additionally, the study presents data from field application of the sampling on filter method with fibreglass filters as well as polycarbonate membranes. Materials and Methods General description of the sewerage system The study was carried out in the sewerage system of a city with a population of several hundred thousand in Southern France. It is the main town (not cited for confidential reasons) of an urban agglomeration that is one of the 10 most populated in the country. The sewer network of the city collects domestic wastewater, rainwater, and wastewater and effluent from workshops, small industries, and commercial buildings. It is made up of underground pipe galleries that channel and convey wastewater to the wastewater treatment facility located a few kilometres away from the city. Manholes allow workers to go down into the largest pipe galleries for inspection and maintenance. Description of the work tasks The sewer workers activities that we focused on were those involved in inspecting and cleaning the sewer. The city employs 179 sewer workers. One group of workers (129 persons) is entirely dedicated to pipes inspection and another one (50 persons) is entirely dedicated to cleaning activities (pipes cleanup, chamber cleanup, tanker lorry unloading). Work is usually done by teams of three to six workers and for a 6-h

3 Exposure to airborne endotoxins among sewer workers Page 3 of 11 work shift. For workers involved in cleaning activities, the pipes cleanup, chamber cleanup, tanker lorry unloading tasks account, respectively, for 38, 50, and 12% of the work shift (estimates from the city). Thus, the occupational exposure of the sewer workers to endotoxins was evaluated for the work tasks during which they were believed to be exposed to bioaerosols. Pipes inspection The aim of the pipes inspection task is to define the places where the sewer network needs to be cleaned or repaired (Fig. 1). To achieve this, the sewer workers walk along the man-accessible pipe galleries for making a visual inspection and recording anomalies. Pipes cleanup The aim of the pipes cleanup task is to clean the pipe galleries in order to clear them of accumulated matter (sand, sludge, grease) that can disturb the wastewater flow. Workers achieve this operation by removing the accumulated matter from the sewer walls and by gathering the resulting sludge under the nearest manhole. The sludge is then removed from the pipes and sent for storage to a tanker lorry by sewer workers using a pumping device. In our study, three different removal and gathering methods were used by the sewer workers. A high-pressure water jet and scraper tools were used for man-accessible galleries. Narrow galleries that could not be visited were washed with a stand-alone water pressure flushing system that flushed out silt and sedimented matter from a manhole to another collection point. The operation was conducted from street level by workers who were not inside the galleries while the water pressure flushing system was in operation. Chamber cleanup Sedimentation chambers specially designed for deposition of suspended matter from water have been built throughout the network. The aim of the chamber cleanup task is to clean these chambers in order to clear them of accumulated matter (sand, sludge, grease). The water flow is bypassed so that sewer workers are able to walk down to the sedimentation chamber to clean it. The sludge is then removed from the chamber and sent for storage into a tanker lorry by workers using a pumping device. Tanker lorry unloading The tanker lorry unloading task is an operation aimed at removing the sludge from the tanker lorry and at 1 A sewer worker using a manhole for accessing underground pipe galleries for pipes inspection.

4 Page 4 of 11 Exposure to Airborne Endotoxins among Sewer Workers cleaning out the lorry. In our study, this very short operation (a few minutes) was done outside at the wastewater treatment facility. Firstly, the tanker lorry was unloaded automatically. Then, the sewer workers entered the tank to clean it and to remove residual sludge from the inner walls using a high-pressure water jet. Methods for bioaerosol sampling Airborne inhalable endotoxins were sampled by the filtration method using 37-mm polystyrene three-piece closed-face cassettes (CFCs) provided by Millipore (Molsheim, France). Each CFC was mounted with a sterile collection filter and a backing cellulose pad thick cellulose absorbent pad, Millipore, Molsheim, France. Two different filters were used in our study; the CFC-PC contained a sterile polycarbonate filter (Nuclepore, 0.8-mm pore size, Millipore, Molsheim, France) and the CFC-FG contained a fibreglass filter (Whatman, FG/B glass microfibre filter, grade GF/B: 1 μm, thickness: 0.68 mm, Sigma-Aldrich, Lyon, France). Initial endotoxin contamination of filters was below 0.05 EU per filter. To achieve sampling, each CFC was connected to a portable pump (GilAir, Gilian AD Air Solutions, Metz, France) operating at the nominal flow rate of 2 l min 1. The flow rate was calibrated before and after sampling using a bubble flow meter (Gilibrator, Gilan, AD Air Solutions, Metz, France). The duration of sampling depended on the duration of the investigated work tasks and ranged from 9 to 98 min (median: 60 min). The duration of sampling for the tanker lorry unloading task was the shortest (9 min). Sampling strategy and experimental design The measurements of airborne endotoxins were taken at 13 different worksites that were located all across the city and selected on the basis of the scheduled working program. Bioaerosol samples were collected from March to June following a non-balanced experiment plan (Table 1). Unexpected changes in the program of work on the sewers and in the availability of operators Table 1. Experimental design used to assess the exposure of sewer workers to endotoxins Working tasks Working sites Number of collected samples CFC-FG exposure measurements CFC-PC exposure measurements Background measurements Pipes inspection A B C D Chamber cleanup E Pipes cleanup Water pressure flushing system F G H High-pressure jet I High-pressure jet J Scrapper tools K Scrapper tools L Tanker lorry unloading M Duration of sampling (min)

5 Exposure to airborne endotoxins among sewer workers Page 5 of 11 wearing samplers prevented us from following the balanced sampling plan we initially designed. The number of worksites per task was 4, 5, 3, and 1 for the pipes inspection, the pipes cleanup, the chamber cleanup, and the tanker lorry unloading tasks, respectively. Simulated individual samples were collected in triplicate with CFC-FG and CFC-PC at the 13 different worksites and inside the sewer network. Sampling was performed by placing three identical sampling devices (i.e. three CFC-FG or three CFC-PC devices) in the breathing zone of an operator. The sampling devices were not worn by sewer workers since their work activity was hindered by the sampling equipment. Therefore, it was decided to place the sampling devices on an independent person (laboratory technician) who was located as close as possible to the sewer workers to mimic personal exposure (Fig. 2). When a specific working task was investigated, the CFC-FG and CFC-PC sampling devices operated simultaneously and were worn by two different operators who stood very close to each other. The background level of airborne endotoxins was determined for each worksite by sampling bioaerosols outside the sewer network. This was done in parallel with the endotoxin exposure measurements performed inside the sewer network and achieved by placing one CFC-PC filter sampling device on the breathing zone of a laboratory technician. The technician was located at street level, usually some distance from the manhole used by sewer workers to access the sewer areas to clean or to inspect and was not involved in the work tasks (Fig. 2). Field negative controls were made with FG filters and PC filters that were not used for sampling. Bioaerosol samples were sent to the lab using iceboxes within 12 h after sampling and were stored at 4 C pending analysis. Sample analysis The samples were analysed within 24 h after sampling. All the operations were performed in a Microbiological Safety Cabinet and with sterile and pyrogen-free disposable material. Filter samples from the CFCs were pre-treated to extract the endotoxins from the filter material before analysis. Each filter was carefully removed from the sampling cassette and aseptically transferred to a 50-ml sterile and pyrogen-free propylene tube (Cellstar tubes, Greiner Bio-One, Courtaboeuf, France) containing 5 ml of pyrogen-free water (Aqua B. Braun, B. Braun, Chasseneuil, France). Each tube was shaken for 1 min using a vortex agitator operating at 2500 r.p.m. (Reax 2 Schematic representation of pipe galleries and of the positions of samplers for individual measurement of workers exposure to endotoxins. Samplers were operating simultaneously when they were compared; they were carried by two different operators who stood up very close to each other and as close as possible to the sewer workers.

6 Page 6 of 11 Exposure to Airborne Endotoxins among Sewer Workers Control, Heildolph-Serlabo-technologies, Entraigues sur la Sorgue, France) and was plunged for 20 min in an ultrasonic bath (Bronson 1210, Bronsonic, Bronson Ultrasonic Corporation, Geneve, Switzerland) operating at 47 KHz and 60 C. After sonication, each tube was shaken for 1 min at 2500 r.p.m. on the vortex agitator. The extracts were filtered through a 0.22-μm pyrogen-free polyethersulfone membrane (Millex -GP, Millipore, Molsheim, France). The amount of endotoxins in the solutions was determined by the kinetic chromogenic limulus amoebocyte lysate (LAL) method as fully described previously (Duquenne et al., 2012). The LAL reagent kit (Kinetic-QCL, Lonza, Lyon, France) was used for the assay according to the recommendations of the supplier. The reagent kit included the lyophilized endotoxin from Escherichia coli 055:B5 (E. coli 055:B5 endotoxin, batch 2L6560) as control standard endotoxin (CSE), the lyophilized LAL reagent (Kinetic- QCL reagent, batch 2L151M), and the LAL reagent water (batch 3L1720). The LAL reagent water and the CSE were used as negative and positive controls, respectively. Samples were spiked with a known concentration of endotoxins to assess the interferences. Analysis results were expressed as endotoxin units per millilitre (EU ml 1 ) and the assay limit of detection was EU ml 1, and values from field negative controls were subtracted from concentration of endotoxins in the extract. Expression of results and statistical analysis The concentration of endotoxins in the air was expressed as EU m 3. As the number of samples collected was too low for a robust statistical analysis, data were presented and discussed on the basis of raw values, means (arithmetic and/or geometric ones), standard deviation, and confidence interval at 95% (CI95). Results Endotoxin levels in field blanks were from <0.025 to 0.03 EU per filter for CFC-PC and <0.025 to 0.3 EU per filter for CFC-FG. Background level of endotoxins Airborne endotoxins were measured outside the sewer network for 11 out of 13 sampling sites. Samples from two worksites were lost during analysis. The recorded endotoxin concentrations were lower than 5 EU m 3 for six samples among all the collected ones, and the endotoxin values ranged from 5 to 10 EU m 3 for three samples (Table 2). Two other samples exhibited higher values: 122 EU m 3 and 51 EU m 3. Simulated personal exposure Sampling with CFC-FG and CFC-PC devices enabled 78 bioaerosol samples to be collected during the work inside the sewer. Exposure values ranged from 13 to 440 EU m 3 when samples were collected using the CFC-FG method with a geometric mean of 85.3 EU m 3 (Table 2). They ranged from 6.0 to 420 EU m 3 when they were collected using the CFC-PC method with a geometric mean of 48.9 EU m 3. The levels of exposure to endotoxins inside the sewerage system were higher than those observed outside (i.e. the background levels). Endotoxin concentrations measured using CFC-FG during the chamber cleanup work task ranged from 79 to 260 EU m 3 with a geometric mean concentration of 163 EU m 3 (Table 2). They ranged from 37 to 239 EU m 3 when measured using the same method during the pipes inspection task with a geometric mean concentration of 85 EU m 3. Data from CFC-FG samples also indicate that exposure levels measured during the pipes cleanup task varied over a wide range as a function of the method used to clean the pipes. Particularly, the use of a high-pressure water jet to clean the pipes led to higher values for endotoxin exposure (from 141 to 440 EU m 3 ) than the use of scraper tools (from 15 to 92 EU m 3 ) and of the water pressure flushing system (from 13 to 16 EU m 3 ). CFC-FG samples collected during the tanker lorry unloading tasks led to concentrations from 33 to 53.4 EU m 3. Similar tendencies were observed when exposure was assessed using CFC-PC (Table 2). However, the endotoxin concentrations measured with the CFC-FG method were higher than those observed with the CFC-PC method for 27 samples among the 39 collected ones (data not shown). The geometric mean of CFC-PC measurements was 57% of the mean observed with CFC-FG measurements. The geometric mean concentration of airborne endotoxins measured with CFC-PC was 123.9, 45.2, 33.9, and 19 EU m 3 for the chamber cleanup, the pipes cleanup, the pipes inspection, and the tanker lorry unloading

7 Exposure to airborne endotoxins among sewer workers Page 7 of 11 Table 2. Exposure of sewer workers to endotoxins as a function of working task and working site Working tasks Working sites Endotoxin exposure (EU m 3 ) Background measurements CFC-PC measurements a CFC-FG measurements a AM Min Max SD CI95 AM Min Max SD IC95 Pipes inspection A % % 0.7 B % % 2.6 C % % 4.5 D % % 8.6 Chamber cleanup E % % nd Pipes cleanup Water pressure flushing system F % % 51.4 G % % 2.1 H % % 5.4 High-pressure jet I % % J % % 2.9 Scrapper tools K % % 7.9 L % % 0.4 Tanker lorry unloading M % % nd Bioaerosol samples were collected using 37-mm closed-face cassettes containing fibreglass filters (CFC-FG) or polycarbonate filters (CFC-PC). AM, arithmetic mean; CI95, confidence interval at the confidence level of 95%; nd, not detected; SD, standard deviation. The background level of endotoxins was measured using one CFC-PC. a Calculated from three replicates (i.e. three identical sampling devices placed in the breathing zone of an operator).

8 Page 8 of 11 Exposure to Airborne Endotoxins among Sewer Workers activities, respectively. The highest CFC-PC exposure levels were measured during the chamber cleanup task (from 31 to 421 EU m 3 ) and when a high-pressure water jet was used to clean the pipes during the pipes cleanup task (from 87 to 186 EU m 3 ). Endotoxin concentrations from triplicate samples varied over a wide range (CI from 3 to 53% for CFC-FG) but with no obvious trend emerging when considering worksites, work tasks, or sampling method (Table 2). Discussion Background levels of airborne endotoxins Air samples were collected outside the sewer network to determine the background endotoxin exposure levels of workers when standing in an area not affected by the operational activities. Endotoxin values lower than 5 EU m 3 corroborate those reported in a previous study for the level of endotoxins in the outdoor air of urban areas (Mueller-Anneling et al., 2004; Madsen, 2006). We assumed that values from 5 to 10 EU m 3 were of the same order of magnitude, but values higher than 50 EU m 3 were not expected. For the two values higher than 50 EU m 3, we noted that the sampling devices were worn by operators who were located close to the sewer manhole. In that situation, the sampling devices were located above a sewer area where the pipes cleanup (background level: 122 EU m 3 ) and chamber cleanup (background level: 51 EU m 3 ) activities were in progress. It is supposed that these operators were exposed to bioaerosols generated by the workers inside the sewerage system via the manhole as no other sources were identified around. This suggests that manholes may constitute a source of exposure for people in the surrounding area. Significance of worker exposure to airborne endotoxins Results from CFC-FG and CFC-PC samples suggest that sewer employees were exposed to airborne endotoxins during their work. In our study, higher amounts of airborne endotoxins were observed inside the sewer network compared with background level. Very little data have been published at international level about exposure of workers to bioaerosols in underground sewer networks before any treatment of the wastewater. Prior studies have demonstrated that work activities in a sewer network increase the total number of airborne cultivable microorganisms (Karpinski et al., 2000; Neumann et al., 2002; Haas et al., 2010). Furthermore, endotoxin exposures up to 800 EU m 3 were reported in one German survey carried out in the same occupational environment (Neumann et al., 2002). Thus, our data corroborate those findings and provide a new contribution to the knowledge of occupational exposure to endotoxins in sewers. Sampling devices were not worn by workers in our study. Although operators wearing the samplers stood as close as possible to the target working activities, our measurements might slightly underestimate the real exposure levels for some situations. This is probably the case for the chamber cleanup and pipes cleanup with the high-pressure water jet tasks for which narrow pipe sections did not always allow a rigorous survey of tasks to be conducted throughout their durations. But such situations happened rarely. For the other tasks, monitoring was easier and we think that our measurements are representative of real exposure. Our study and the few previous ones suggest that endotoxin exposure levels do not exceed 1000 EU m 3 during work with raw wastewater, which is lower than those reported in other occupational environments. By way of comparison, ambient endotoxin concentrations have been measured from 0.1 to 3500 EU m 3 (Laitinen et al., 1994; Thorn et al., 2002; Visser et al., 2006) and exposure levels from 0.6 to 4160 EU m 3 (Thorn et al., 2002; Smit et al., 2005; Visser et al., 2006, Spaan et al., 2008) in wastewater treatment plants. Endotoxin concentrations higher than 5000 or even EU m 3 have been recorded in the air of composting facilities, and during grain handling or transformation (Spaan et al., 2006; Halstensen et al., 2007; Duquenne et al., 2012). Effects of work task on exposure to endotoxins The number of samples collected in the present study is too low for a robust statistical analysis and no definitive conclusions can be drawn from our results with regard to the effect work tasks on endotoxin levels. In our study, employees were mainly exposed to endotoxins during work practices for which agitation of wastewater and deposits occurred, such as the chamber cleanup task and the pipes cleanup task with a high-pressure water jet. Indeed, agitation of water generated bioaerosols in the workers breathing zones

9 Exposure to airborne endotoxins among sewer workers Page 9 of 11 and the phenomenon might be amplified by the confinement of work tasks. Such a relationship between wastewater or sludge agitation and occupational exposure to bioaerosols has already been observed in wastewater treatment facilities (Laitinen et al., 1994; Thorn et al., 2002; Sanchez-Monedero et al., 2008). The variability in concentrations during the pipes cleanup task using scraper tools and the pipes inspection task might be due to changes in the amounts of deposits that were shifted by workers, to the variable levels of contamination of these deposits, or to the proximity of agitated wastewater. Interestingly, CFC-FG measurements (up to 240 EU m 3 for worksite B) indicate that workers may be exposed to endotoxins during the pipes inspection task (Table 2); they probably visited areas with agitated water during their inspection. The lower endotoxin levels (<50 EU m 3 ) observed during the pipes cleanup task using the water pressure flushing system were not surprising since the sewer workers were some distance away from the aerosols created by the water pressure jet (see task description). The tanker lorry unloading is a very short task; it was investigated once throughout the study and further measurements are required. These results suggest that exposure levels to endotoxins were dependent on the work tasks, but a larger study is needed to confirm this statement. A possible influence of filter material on endotoxin measurements In our study, the use of FG filters tended to lead to higher endotoxin exposure levels than PC filters. Direct comparison of the CFC-FG and the CFC-PC sampling methods cannot be made rigorously as the number of samples is low and as samplers were worn by different workers. Previous field studies showed that fibreglass filters yielded higher airborne endotoxin estimates than PC filters (Thorne et al., 1997; Laitinen, 1999; Görner et al., 2006). Differences between filters were ascribed to the ability of the filter material to set endotoxins free during the extraction step and particularly to the capacity of endotoxins to be adsorbed in the filter material. Fibreglass filters may be unfavourable to this phenomenon. Other published studies also suggested that the results of airborne endotoxin measurements depended both on the characteristics of the collection filter and on the nature of the sampled aerosol (Gordon et al., 1992; Reynolds et al., 2002; Spaan et al., 2007). Thus, our results suggest, without demonstrating it, the suitability of FG filters for measuring endotoxins in bioaerosols during work in a sewerage system. Expected health effects of the observed exposure levels There is no admitted occupational exposure limit (OEL) for endotoxins at international level. The health-based recommended OEL (HBROEL) proposed in the Netherlands of 50 EU m 3 and then adjusted to 200 EU m 3 (Heederik and Douwes, 1997; DECOS, 1998) are the widely used reference in the literature. The Dutch Expert Committee on Occupational Standards published a report proposing a new HBREOL of 90 EU m 3 (DECOS, 2010). In our study, the exposure measurements with CFC-FG were not excessively high compared with other occupational environments, but 17% exceeded 200 EU m 3 and 51% were over 90 EU m 3. The exposure levels we measured during sewer work were close to those reported in the wastewater treatment occupational environment where endotoxins are highly suspected to induce airway symptoms, general symptoms, and diarrhoea (Marth et al., 1997; Thorn and Kerekes, 2001; Hansen et al., 2003; Thorn and Beijer, 2004; Rylander, 2002, 2005). Such symptoms might be more prevalent in the population of workers we studied. Conclusions The sewer workers were exposed to endotoxins during their occupational tasks, especially the tasks involved in cleaning the pipes. In our exploratory study, cleaning the sedimentation chambers and the pipe galleries with a high-pressure water jet led to the highest levels of exposure. Thus, the exposure of the sewer workers to endotoxins tended to be the highest for tasks involving agitation of water and matter, and the method used to clean the pipes might be one of the determinants of the concentration of airborne endotoxins. However, the number of samples collected in the present descriptive study is too low for drawing definitive conclusions. The study only provides data regarding the level of exposure to endotoxins in underground sewer pipes and further exposure investigations are needed. As sewer workers are potentially exposed to a variety of biological agents other than endotoxins, future exposure investigations would include bacterial

10 Page 10 of 11 Exposure to Airborne Endotoxins among Sewer Workers and viral pathogens, mycotoxins, and other hazardous agents. The exposure levels measured in our study did not exceed 420 EU m 3 but such concentration levels of airborne endotoxins may cause adverse effects in sensitive workers. These results would be useful for designing larger exposures studies. Acknowledgements The authors would like to acknowledge Dr Christine Da Passano-Amiot from the worker health department of the city as well as Pascal Wild, Regis Colin, and Vincent Baudin from INRS for their valuable technical assistance during the study and their help for data analysis. References CEN (Comité Européen de Normalisation). (2003) Workplace atmosphere - determination of airborne endotoxins (CEN standard EN 14031). Brussels, Belgium: CEN. Chun DT, Bartlett K, Gordon T et al. (2006) History and results of the two inter-laboratory round robin endotoxin assay studies on cotton dust. Am J Ind Med; 49: Douwes J, Thorne P, Pearce N et al. (2003) Bioaerosol health effects and exposure assessment: progress and prospects. Ann Occup Hyg; 47: Duquenne P, Marchand G, Duchaine C. (2013) Measurement of endotoxins in bioaerosols at workplace: a critical review of literature and a standardization issue. Ann Occup Hyg; 57: Duquenne P, Simon X, Koehler V et al. (2012) Documentation of bioaerosol concentrations in an indoor composting facility in France. J Environ Monit; 14: Dutch Expert Committee on Occupational Standards (DECOS). (1998) Endotoxins: health based recommended recommended occupation exposure limit. Rijswijk, The Netherlands: The Health Council of the Netherlands, Publication No. 1988/03WGD. Dutch Expert Committee on Occupational Standards (DECOS). (2010) Endotoxins: health-based recommended occupational limit. The Hague, The Netherlands: The Health Council of the Netherlands, Publication No. 2010/04OSH. Gerardi MH, Zimmerman MC. (2005) Wastewater pathogens - wastewater microbiology series. Hoboken, NJ: John Wiley & Sons. 179 p. Gordon T, Galdanes K, Brosseau L. (1992) Comparaison of sampling media for endotoxin-contaminated aerosol. Appl Occup Environ Hyg; 7: Görner P, Fabries J-F, Duquenne P, et al. (2006) Bioaerosol sampling by a personal rotating cup sampler CIP 10-M. J Environ Monit; 8: Haas D, Unteregger M, Habib J et al. (2010) Exposure to bioaerosol from sewage systems. Water Air Soil Pollut; 207: Halstensen AS, Nordby KC, Wouters IM et al. (2007) Determinants of microbial exposure in grain farming. Ann Occup Hyg; 51: Hansen ES, Hilden J, Klausen H et al. (2003) Wastewater exposure and health--a comparative study of two occupational groups. Occup Environ Med; 60: Heederik D, Douwes J. (1997) Towards an occupational exposure limit for endotoxins? Ann Agric Environ Med; 4: Jeggli S, Steiner D, Joller H et al. (2004) Hepatitis E, Helicobacter pylori, and gastrointestinal symptoms in workers exposed to waste water. Occup Environ Med; 61: Karpinski C, Scheidt-Illig R, Wenzel E et al. (2000) Investigations of possible infection sources during sewer working. Gefahrst Reinhalt Luft; 60: Laitinen SK. (1999) Importance of sampling, extraction and preservation for quantification of biologically active endotoxin. Ann Agric Environ Med; 6: Laitinen SK, Kangas J, Husman K et al. (2001) Evaluation of exposure to airborne bacterial endotoxins and peptidoglycans in selected work environments. Ann Agric Environ Med; 8: Laitinen SK, Kangas J, Kotimaa M et al. (1994) Workers exposure to airborne bacteria and endotoxins at industrial wastewater treatment plants. Am Ind Hyg Assoc J; 55: Liebers V, Brüning T, Raulf-Heimsoth M. (2006) Occupational endotoxin-exposure and possible health effects on humans (review). Am J Ind Med; 49: Lundholm M, Rylander R. (1983) Work related symptoms among sewage workers. Brit J Ind Med; 40: Madsen AM. (2006) Airborne endotoxin in different background environments and seasons. Ann Agric Environ Med; 13: Marth E, Reinthaler FF, Schaffler K et al. (1997) Occupational health risks to employees of waste employees of waste treatment facilities. Ann Agric Environ Med; 4: Mueller-Anneling L, Avol E, Peters JM et al. (2004) Ambient endotoxin concentrations in PM10 from southern California. Environ Health Persp; 112: Neumann HD, Buxtrup M, Balfanz J et al. (2002) Belastungen durch biologische Arbeitsstoffe bei der Kanalreinigung. Gefahrst Reinhalt Luft; 9: Oppliger A, Hilfiker S, Vu Duc T. (2005) Influence of seasons and sampling strategy on assessment of bioaerosols in sewage treatment plants in Switzerland. Ann Occup Hyg; 49: Pfafflin JR, Ziegler EN. (2012) Encyclopedia of environmental science and engineering. 6th ed. New York, NY: CRC Press. Reynolds SJ, Thorne PS, Donham KJ et al. (2002) Comparison of endotoxin assays using agricultural dusts. Am Ind Hyg Ass J; 63: Richardson DB. (1995) Respiratory effects of chronic hydrogen sulfide exposure. Am J Ind Med; 28:

11 Exposure to airborne endotoxins among sewer workers Page 11 of 11 Rylander R. (1999) Health effects among workers in sewage treatment plants. Occup Environ Med; 56: Rylander R. (2002) Endotoxin in the environment--exposure and effects. J Endotox Res; 8: Rylander R. (2005) Health effects among workers in sewage treatment plants. Occup Environ Med; 56: Sanchez-Monedero MA, Aguilarb MI, Fenolla R et al. (2008) Effect of the aeration system on the levels of airborne microorganisms generated at wastewater treatment plants. Water Res; 42: Smit LAM, Spaan S, Heederik D. (2005) Endotoxin exposure and symptoms in wastewater treatment workers. Am J Ind Med; 48: Spaan S, Heederik DJJ, Thorne PS et al. (2007) Optimization of airborne endotoxin exposure assessment: effects of filter type, transport conditions, extraction solutions and storage of samples and extracts. Appl Environ Microbiol; 73: Spaan S, Smit LA, Eduard W et al. (2008) Endotoxin exposure in sewage treatment workers: investigation of exposure variability and comparison of analytical techniques. Ann Agric Environ Med; 15: Spaan S, Wouters IM, Oosting I et al. (2006) Exposure to inhalable dust and endotoxins in agricultural industries. J Environ Monit; 8: Thorn J, Beijer L, Jonsson T et al. (2002) Measurement strategies for the determination of airborne bacterial endotoxin in sewage treatment plants. Ann Occup Hyg; 46: Thorn J, Kerekes E. (2001) Health effects among employees in sewage treatment plants: a literature survey. Am J Ind Med; 40: Thorn T, Beijer M. (2004) Work-related symptoms and inflammation among sewage plant operatives. Int J Occup Environ Health; 10: Thorne PS, Perry SS, Saito R et al. (2010) Evaluation of the limulus amebocyte lysate and recombinant factor C assays or assessment of airborne endotoxin. Appl Environ Microbiol; 76: Thorne PS, Reynolds SJ, Milton DK et al. (1997) Field evaluation of endotoxin air sampling assay methods. Am Ind Hyg Assoc J; 58: Visser MJ, Spaan S, Arts HJJM et al. (2006) Influence of different cleaning practices on endotoxin exposure at sewage treatment plants. Ann Occup Hyg; 50: