Concentrations and Emissions of Airborne Endotoxins and Microorganisms in Livestock Buildings in Northern Europe

Transcription

1 J. agric. Engng Res. (1998) 70, Concentrations and Emissions of Airborne Endotoxins and Microorganisms in Livestock Buildings in Northern Europe J. Seedorf ; J. Hartung ; M. Schröder ; K. H. Linkert ; V. R. Phillips ; M. R. Holden ; R. W. Sneath ; J. L. Short ; R. P. White ; S. Pedersen ; H. Takai ; J. O. Johnsen ; J. H. M. Metz ; P. W. G. Groot Koerkamp ; G. H. Uenk ; C. M. Wathes Tierärztliche Hochschule Hannover, Institut für Tierhygiene und Tierschutz, Bu nteweg 17p, Hannover, Germany; Silsoe Research Institute, Wrest Park, Silsoe, Bedford MK45 4HS, UK; Danish Institute of Animal Science, Research Centre Bygholm, Dept of Agricultural Engineering, PO Box 536, DK-8700 Horsens, Denmark; Instituut voor Milieuen- Agritechniek, Postbus 43, 6700 AA Wageningen, The Netherlands (Received 27 August 1996; accepted in revised form 24 December 1997) The concentration of airborne endotoxins and microorganisms in livestock buildings (cattle, pig, poultry) was surveyed in four European countries (England, The Netherlands, Denmark and Germany). Measurements were made during the day and night. The endotoxin concentrations were determined from dust samples separated into inhalable and respirable fractions. Airborne microorganisms were classified as total bacteria, Enterobacteriaceae and total fungi. The lowest endotoxin concentrations were found in cattle houses. The highest concentrations of endotoxin were measured in poultry houses, especially percheries, with an overall mean of 692 and 49 ng/m for inhalable and respirable fractions, respectively. Endotoxin concentration was usually higher in the day than at night. These concentrations give cause for concern for the health of stockmen working in such buildings. The corresponding mean emission rates from poultry houses were 678 and 43 μg/h (500 kg) liveweight, respectively, 51 and 6 μg/h (500 kg) liveweight from pig buildings and 9 and 1 μg/h (500 kg) liveweight from cattle houses. A new automated bioaerosol sampler was developed, which allows remote, independent sampling of airborne microorganisms. Its principle of operation is similar to a slit sampler: airborne microbes are collected by impaction on an agar surface. Most measurements of airborne microorganisms were made in Germany. The indoor concentration of total bacteria was 6)43 log colonyforming units (cfu)/m in broiler houses, 5)1 log cfu/m in pig buildings and 4)3 log cfu/m in cattle buildings. During the day, the overall concentration of Enterobacteriaceae ranged between 3 and 4 log cfu/m. The mean fungi concentration for cattle was 3)8, for pigs 3)7 and for poultry 4)0 log cfu/m during the day. Slightly lower concentrations were measured at night. No significant correlation was found between the concentrations of endotoxin and Enterobacteriaceae. The highest emission rate of total bacteria and fungi was measured for broilers, i.e. 9)5 and 7)7 log cfu/h (500 kg) liveweight, respectively. Buildings for laying hens had the highest mean emission rate of 7)1 log cfu/h (500 kg) liveweight for Enterobacteriaceae Silsoe Research Institute 1. Introduction The air in intensive livestock buildings usually contains high concentrations of airborne microorganisms. 1 4 Normally, these microorganisms are associated with dust particles and present a biologically active aerosol (bioaerosol). They can cause or trigger lung-related diseases in animals and man, 5 7 while many common pathogens of pigs and poultry are known to be transmitted aerially. 8 Some non-pathogenic microbes, e.g. the genus Micrococcaceae, can become opportunistic pathogens under certain conditions, such as poor air quality which can interfere with respiratory defence mechanisms. Fungi and their specific mycotoxins are also important, especially the mycotoxin-producing genus Aspergillus, Penicillium or Fusarium, whcih are routinely detected in livestock buildings. 9 A summary of measured concentrations of airborne microorganisms is given in Table 1 and is restricted to livestock buildings in the northern hemisphere. Other summaries have been published by Methling et al. 3 and Zeitler. 10 Table 2 lists some published data on the concentrations of endotoxin. These summaries indicate the /98/050097#13 $25.00/0/ag Silsoe Research Institute

2 98 J. SEEDORF E¹ A. Table 1 Summary of selected published concentrations of airborne microorganisms in livestock buildings Micro Concentration Sampling Animal House type organism (cfu/m 3 air) method Country Authors Cattle* Dairy Total 1)5 10 Filtration, Epi- Sweden Larsson et al. 11 fluorescence Cattle Dairy Fungi Andersen sampler Finland Pasanen et al. 12 Catttle Cow Total 4)4 10 2)8 10 Poland Dutkiewicz et al. 13 Sows Farrowing Total 1)5 10 Andersen sampler Canada Cormier et al. 2 Sows Farrowing Gram -ve 80 Andersen sampler Canada Cormier et al. 2 Sows Farrowing Fungi 150 Andersen sampler Canada Cormier et al. 2 Pigs Total 3 10 Andersen sampler Sweden Clark et al. 1 Pigs Gram -ve 8)8 10 Andersen sampler Sweden Clark et al. 1 Pigs Fungi 300 Andersen sampler Sweden Clark et al. 1 Pigs Fattening Total 4)9 10 Andersen sampler Canada Cormier et al. 2 Pigs Fattening Gram -ve 140 Andersen sampler Canada Cormier et al. 2 Pigs Fattening Fungi 190 Andersen sampler Canada Cormier et al. 2 Pigs Fattening Total Impingement, Filtration Scotland Crook et al. 14 Pigs Fattening Fungi Impingement, Filtration Scotland Crook et al. 14 Pigs Fattening Total 2)6 10 Reuter-Centr-Sampler Germany Hartmann et al. 7 Pigs Total 1)1 10 Andersen sampler Netherlands Heederik et al. 15 Pigs Gram -ve 7)7 10 Andersen sampler Netherlands Heederik et al. 15 Pigs Total 5)7 10 1)5 10 Poland Dutkiewicz et al. 13 Hens Cages incl. Total 4)2 10 Andersen sampler Sweden Clark et al. 1 egg room Hens Cages incl. Gram -ve 4)1 10 Andersen sampler Sweden Clark et al. 1 egg room Hens Cages incl. Fungi 500 Andersen sampler Sweden Clark et al. 1 egg room Broilers Louisiana Total 7)7 10 Isolated from Germany Hinz et al. 16 volumetrically determined dust samples Broilers Louisiana Gram -ve 810 Isolated from volumetrically determined dust samples Germany Hinz et al. 16 * Counted microorganisms in an epifluorescence microscope (microorganisms/m ) Values of house A, fungi as sum of median values of yeasts and moulds Values of house B, fungi as sum of median values of yeasts and moulds. ubiquitous nature of microbes in the air of livestock buildings. Serological studies have shown that antibodies against various microorganisms can be found in farm workers. 11 In addition to the contribution of viable bacteria to respiratory disease, endotoxins, which are cell-wall components of dead Gram -negative bacteria, also play a major role in inflammatory reactions and the deterioration of lung function. 15,19 21 The emission of these agents into the environment may be a cause for concern. The common methods for sampling airborne microorganisms are based on the principles of impingement, impaction, filtration and sedimentation. At present, there is neither a generally accepted standard sampling method nor a method which can work either continuously or in the absence of the operator. To overcome these disadvantages, an automatic bioaerosol sampler (ABS) was Table 2 Summary of selected published endotoxin concentrations Mass Aerial concentration concentration Animal (μg/mg dust) (ng/m 3 air) Authors Cattle 0.44 Clark et al. 1 Pigs 0) Clark et al. 1 Pigs 0)41 10 Crook et al. 14 7)8 10 Pigs 130 Heederik et al. 15 Pigs 12 Wiegand 17 Pigs Dutkiewicz et al. 13 Hens 0) Clark et al. 1 Broilers 140 Hinz et al. 16 Broilers 0)039 Wiegand et al. 18 Poultry Thelin et al. 19

3 EMISSIONS OF AIRBORNE ENDOTOXINS IN LIVESTOCK BUILDINGS IN NORTHERN EUROPE 99 developed for this project and has been described elsewhere. 22 The object of this study was to determine the concentrations and emissions of airborne endotoxins and microorganisms in different types of livestock buildings in four Northern European countries. This study was part of a larger survey of the emissions of aerial pollutants from livestock buildings, an overview of which is given by Wathes et al Materials and methods 2.1. Endotoxins An overall description of the project has been given by Phillips et al. 24 Dust samples were taken at seven locations within the animal house. A pair of inhalable and respirable dust samplers (SKC) was used in the day ( h) and night ( h) for each sampling site. Flow rates were adjusted to 2)0 l/min for the inhalable dust sampler and 1)9 l/min for the respirable dust sampler. The endotoxin concentrations in the dust samples were analysed after conditioning and weighing of pooled samples of the filters. The basis of the endotoxin analysis was the Limulus Amoebocyte Lysate (LAL) gelation test, modified by kinetic-turbidimetric analysis. All dust samples were analysed for endotoxin content as a batch under identical sampling conditions. All glassware was made from borocilicate glass and was made pyrogen-free by heating at 300 C for 3 h. Other disposable articles (Eppendorf, Germany) were pyrogenfree. Each filter batch was diluted with endotoxin-free water (Acila, Pyroquant Diagnostik GmbH, Germany) over the range from 1 : 10 to 1 : 10 ; the dilution factor depended on the total amount of sampled dust. The optimum ph ranged between 6 and 8 and was adjusted where necessary by the addition of NaOH, HCl or by an endotoxin-free buffer. The water/dust suspension was shaken for at least 3 h at 20 C. The solution was centrifuged at 1000 g for 10 min and an aliquot of 200 μl was diluted with endotoxin-free water to 1 : 10. A volume of 200 μl of this highest dilution was made up with 200 μl endotoxin-free water and 100 μl lysate. For calibration, six standard solutions were made with the endotoxin of Salmonella abortus equi (Pyroquant Diagnostik GmbH, Germany), adjusted to 100 ng/ml. The magnitude of gelation and the recovery of dust-related endotoxins was checked by spiked samples, containing 12)5pg of Salmonella abortus equi endotoxin. A multipoint calibration with 400 μl of solutions with 100, 50, 25, 12)5, 6)25 and 3)125 pg/ml with 100 μl lysate was used and the magnitude of concentration-dependent gelation was measured by photometry. All data were analysed by linear regression and compared with a standard curve obtained from a reference endotoxin of Salmonella abortus equi. A correlation coefficient of better than 0)98 was accepted. A solution of 400 μl endotoxin-free water and 100 μl lysate served as a control. The measurement of gelation was made with a LAL-5000 instrument (Pyroquant Diagnostik GmbH, Germany). The result was expressed as ng of endotoxins per m air. Data were transformed by logarithms and analysed by one-way analysis of variance with country (England, The Netherlands, Denmark, Germany), species (cattle; pig; poultry) and animal types (dairy, beef, calf; sow, weaner, fattening pig; layer and broiler, respectively), species-related housing type (tied/cubicle, litter, slats, mesh, perchery, cages) and season (winter, summer) as separate factors Microorganisms Sampling of airborne microorganisms was carried out with a newly developed automatic bacteria sampler which has been described elsewhere. 22,25 Plate-Count-Agar (PC, RPP 013), Violet-Red-Bile-Agar (VRB, RPP 075) and Sabouraud-Agar (SA, RPP 055) were purchased from UNIPATH (Wesel, Germany) for the determination of the total amount of bacteria (mainly Gram -positive and to a lesser extent Gram -negative), Gram -negative bacteria (Enterobacteriaceae) and total fungi, respectively. For six buildings, half-strength nutrient agar for total bacteria and malt extract agar for fungi and yeasts were used. At h and at h the following day, samples of airborne microorganisms were collected, usually in the centre of the animal house. Six agar plates were available for taking samples on each occasion and contained either PC, VRB or SA. The sampling times differed for each plate and were in the range from 15 s to 8 min according to the livestock type. The flow rate during sampling was 2 l/min. The exposed PC and VRB plates were incubated at 37 C under aerobic conditions; SA plates were incubated at room temperature, approximately 20 C. Bacteria and fungi colonies were counted after 24 and 48 h (PC and VRB) and after 72 and 96 h (SA). The defined incubation times minimized the overgrowth and fusion of colonies on the agar surface. The results were expressed as the logarithm of colony-forming units per m (log cfu/m ). The measurements were carried out in 61 animal houses (England, 5; The Netherlands, 9; Denmark, 17; Germany, 30) Emission rate calculation The estimate of emission rate was determined from the product of the ventilation rate, which was based on the

4 100 J. SEEDORF E¹ A. carbon dioxide balance method, 26 and the indoor concentration. Emissions were calculated as an average over 24 h. 3. Results 3.1. Endotoxins The concentrations of airborne endotoxins were measured in a total of 241 animal houses. The results are summarized in Tables 3 and 4. Compared with pigs and poultry, the endotoxin concentration in cattle houses was clearly low. For inhalable endotoxin, mean aerial concentrations ranged between 7)4 and 63)9 ng/m and for respirable endotoxin, concentrations ranged between 0)6 and 6)7 ng/m. Mean endotoxin concentrations were higher for pigs. Inhalable endotoxin concentration ranged between 52)3 and 186)5 ng/m with respirable endotoxin concentrations of between 7)4 and 18)9 ng/m. Concentrations were highest for poultry; mean values ranged between 338)9 and 860)4 ng/m air in inhalable dust fractions and from 29)6 to71)8 ng/m air in respirable dust. The respirable fraction typically comprised about 10% of the inhalable fraction, though there were minor differences in this percentage between day and night samples and between livestock species. The concentration of endotoxin was usually higher in the day than at night. The relative frequency distribution of endotoxin concentration in different concentration classes is shown in Figs. 1 and 2. For inhalable endotoxin concentration in the day and night, 40)2 and 47)5% of all samples, respectively, had endotoxin concentrations between '5 and 50 ng/m. Most, approximately 60%, of the respirable fractions had concentrations in the range 0 5 ng/m The results of the statistical analysis showed that poultry had the highest endotoxin concentrations in each of the four dust samples ( p(0)001), followed by pigs and cattle. Calves had higher endotoxin concentrations in both inhalable samples ( p(0)001) than dairy cows and beef cattle. For the same dust fractions, significant variations between the different housing types were estimated. For inhalable daytime samples, the endotoxin concentration was higher in cattle buildings with litter ( p(0)01), while cattle houses with slats showed higher endotoxin concenrations for the nighttime samples (p(0)04). Pig houses in The Netherlands had the highest endotoxin concentrations in the respirable fraction at night ( p(0)03). The highest endotoxin concentrations for all samples were recorded in weaner houses with mesh or slat flooring. As a consequence, for nearly all dust fractions the endotoxin concentration was higher in the mesh/slats housing type ( p(0)03) than in buildings with litter or slats alone. Housing types with litter showed the highest endotoxin concentrations ( p(0)002) only for inhalable day time samples. Differences between poultry houses were observed. Poultry houses for laying hens in England showed the highest endotoxin concentrations for all samples (p(0)01). Except for endotoxin in the inhalable fraction at night, percheries showed the highest endotoxin concentrations ( p(0)05). Significant seasonal interactions with aerial endotoxin concentrations were not be observed for cattle, pigs and poultry Airborne microorganisms The concentrations of airborne microorganisms were measured mainly in Germany. Total counts of bacteria, Gram -negative bacteria (Enterobacteriaceae) and fungi and yeasts give an overview of the airborne microorganisms in animal houses. The results of 61 daily and 25 Table 3 Mean airborne endotoxin concentrations (ng/m 3 ) for different animal types at day and night Inhalable endotoxin Respirable endotoxin concentration (ng/m 3 ) concentration (ng/m 3 ) Species Number (n) Day Night Day Night Cows 31 15)1 7)4 0)6 1)6 Beef cattle 18 11)8 13)6 1)0 1)3 Calves 18 48)7 63)9 6)3 6)7 Sows )6 52)3 8)3 7)4 Weaners )5 157)4 17)7 18)9 Fattening pigs )1 109)1 13)0 11)4 Layers )4 338)9 58)1 29)6 Broilers )7 784)2 35)1 71)8

5 EMISSIONS OF AIRBORNE ENDOTOXINS IN LIVESTOCK BUILDINGS IN NORTHERN EUROPE 101 Table 4 Mean endotoxin concentrations (ng/m 3 ) in relation to country, animal type and housing type Inhalable endotoxin Respirable endotoxin concentration (ng/m 3 ) concentration (ng/m 3 ) Animal and housing type Country* Number (n) Day Night Day Night Dairy, litter/tied NL 8 22)5 4)8 0)7 2)5 DK 7 14)9 9)2 0)5 2)1 D 2 44)8 20)2 1)5 0)5 Dairy, cubicle NL 8 6)2 5)2 0)7 0)5 DK 6 7)3 7)2 0)3 1)6 Beef, litter D 1 41)1 9)9 1)2 0)5 Beef, slats NL 8 6)6 16)0 1)3 2)4 DK 7 14)4 14)4 0)8 0)4 D 2 8)6 3)3 0)1 0)2 Calves, litter DK 8 31)3 11)6 0)4 0)1 D 1 10)4 14)2 0)0 0)1 Calves, slats NL 9 66)4 110)1 12)2 13)3 Sows, litter E 9 52)3 23)6 2)2 2)1 D 4 809)0 324)3 52)6 52)2 Sows, slats E 8 21)3 43)8 1)2 0)5 NL 8 108)7 20)1 2)4 2)2 DK 7 28)8 22)7 3)6 4)8 D 8 4)0 11)5 8)5 4)2 Weaners, mesh E 9 52)4 30)3 3)4 16)1 NL 8 365)1 337)4 37)3 27)8 DK 7 226)9 160)0 20)2 18)5 D 3 18)3 10)5 2)2 3)2 Fatt.pigs, litter E 8 111)5 156)5 14)5 5)3 DK 7 204)7 151)3 20)8 21)2 Fatt.pigs, slats E 6 136)1 75)8 12)1 7)5 NL 8 102)2 100)2 10)8 14)3 DK 6 128)7 71)2 7)6 7)8 D 4 134)5 64)9 10)1 10)7 Layers, perchery E )9 1140)4 171)8 105)3 NL 8 431)3 104)6 19)6 9)5 DK 6 265)3 29)9 26)7 10)4 Layers, cages E 6 549)2 309)8 67)8 28)0 NL 7 20)8 20)0 2)2 2)0 DK 2 116)0 53)3 14)0 12)4 D 4 30)9 11)4 3)1 1)5 Broilers, litter E 4 76)2 64)3 7)5 4)9 NL 9 469)3 293)5 17)7 63)8 DK 6 139)7 116)6 18)5 65)4 D )2 6434)7 218)3 260)4 * E"England, NL"The Netherlands, DK"Denmark, D"Germany. nightly measurements in animal houses are shown in Fig. 3, expressed as the geometric mean concentration for each animal type. Seasonal effects were not considered. The highest total bacteria concentrations were detected in broiler houses with mean concentrations of 6)43 log cfu/m air during the day and night. In contrast to broiler houses, houses for laying hens had lower concentrations of between 4 and 5 log cfu/m. For pigs, mean concentrations of 5)1 log cfu/m and for cattle of 4)3 log cfu/m were detected. In all cases the concentrations were slightly greater in the day than at night. This diurnal distribution was also observed for Enterobacteriaceae with the exception of layers, while the mean concentration was approximately 1 log cfu/m lower than the concentration of total bacteria. The mean daily fungi concentration was 3)8 for cattle, 3)7 for pigs and 4)0

6 102 J. SEEDORF E¹ A. Fig. 1. Frequency distribution of inhalable and respirable endotoxin concentrations for all livestock buildings during the day, n"241 for poultry log cfu/m, respectively: similar values were recorded at night. Figures 4 and 5 show the relative frequency distribution of microbial concentrations as a function of concentration. The modal concentrations of microorganisms during the day were approximately 4)5, 3)5 and 3)5 log cfu/m for total bacteria, Enterobacteriaceae and fungi, respectively. At night, the frequency distribution of Fig. 2. Frequency distribution of inhalable and respirable endotoxin concentrations for all livestock buildings at night; n"241

7 EMISSIONS OF AIRBORNE ENDOTOXINS IN LIVESTOCK BUILDINGS IN NORTHERN EUROPE 103 Fig. 3. Concentrations of airborne microorganisms in livestock buildings. Day: n"61; night: n"25 concentration reflected the overall reduction in concentration with a similar pattern to the daytime values Emission of endotoxins The emission rates for airborne endotoxins were expressed as μg/h (500 kg) liveweight and the average emission rates over 24 h for all livestock buildings (n"235) are summarized in Table 5. Broiler houses had the highest emission rates of inhalable and respirable endotoxins of 817)4 and 46)7 μg/h (500 kg) liveweight, respectively. Emission rates for hens were slower than for broilers. Inhalable endotoxin emissions rates for pigs ranged between 37)4 and 66)6 μg/h (500 kg) liveweight and from 3)7

8 104 J. SEEDORF E¹ A. Fig. 4. Relative frequency distribution of concentrations of microorganisms during the day; n"61: Total; Enterobacteriaceae; Fungi to 8)9 μg/h (500 kg) liveweight for respirable endotoxin emissions. Cattle houses had comparatively low emission rates with mean inhalable and respirable endotoxin emission rates between 2)9 and 21)4 μg/h (500 kg) liveweight and from 0)3 to2)7μg/h (500 kg) liveweight, respectively. There was a strong linear correlation between the emission rates of inhalable and respirable endotoxins (r"0)905, p(0)001). Additionally, maximum values for endotoxin emission rates were determined (Table 5). In broiler houses, maximum emission rates of inhalable endotoxins were 600- fold higher than in cow buildings. This trend was similar for respirable endotoxins, although laying hen houses had the highest maximum emissions. Due to generally higher ventilation rates during summer, the overall emission rate was 253)4 μg/h (500 kg) liveweight for inhalable endotoxins and 18)3 μg/h (500 kg) liveweight for respirable endotoxins. In winter, the rates were 130)4 and 9)5 μg/h (500 kg) liveweight, respectively Emission of airborne bacteria and fungi The emission rates for airborne microorganisms were calculated and expressed as log cfu/h (500 kg) liveweight. The mean emission rates over 24 h are shown for all livestock buildings and microbial types in Fig. 6. The highest emission rates of total bacteria were measured in broiler houses, namely 9)5 log cfu/h (500 kg) liveweight but the range of emission rates amongst the other species and housing types was much less; the average rate was approximately 7 log cfu/h (500 kg) liveweight. The emission rates of Enterobacteriaceae were much lower. Layers had the highest emission rate of 7)1 log cfu/h (500 kg) liveweight, sows had the lowest emission rate of 6)1 cfu/h (500 kg) liveweight. For fungi the range of emission rates was from 7)7 log cfu/h (500 kg) liveweight for broilers to 5)8 log cfu/h (500 kg) liveweight for weaners. 4. Discussion 4.1. Concentration of endotoxins Endotoxins are the cell-wall components of Gram -negative bacteria and their release is linked with bacterial death. In principle, quantification of the load of Gram -negative bacteria via endotoxin detection and vice versa should be possible. Laitinen and co-workers have shown that endotoxin levels in the air of wastewater treatment plants can be estimated from bacterial counts

9 EMISSIONS OF AIRBORNE ENDOTOXINS IN LIVESTOCK BUILDINGS IN NORTHERN EUROPE 105 Fig. 5. Relative frequency distribution of concentrations of microorganisms at night; n"25: Total; Enterobacteriaceae; Fungi by use of a selective medium for Gram -negative bacteria 27 and concluded that the LAL assay can be used to estimate the concentration of viable Gram -negative bacteria in the air. Such findings were not confirmed in this study: there was no significant correlation between the concentrations of Gram -negative bacteria and aerial endotoxin. Differences between sampling sites and times, the use of spot readings of airborne microorganisms compared with continuous sampling for dust and endotoxin over 24 h, and stresses due to the sampling method may have accounted for this failure. Poultry houses, in particular percheries, showed the highest airborne endotoxin concentrations in contrast with pig and cattle houses. Amongst cattle and pigs, calves and weaners had the highest airborne endotoxin concentrations (Table 3). These concentrations are comparable with those recorded elsewhere (Table 2), except for those of Dutkiewicz et al. 13 which were much higher. Differences in concentration may be due to the generation rate of endotoxin at its source and/or clearance by various routes, principally ventilation and settlement. The faeces are the major source of dust and Table 5 Mean emission rates of inhalable and respirable endotoxins over 24 h for different animal types of considering maximum values Mean respirable Maximum inhalable Mean inhalable endotoxin emission endotoxin emission Maximum respirable Species Number (n) endotoxin emission [μg/h (500 kg) liveweight] endotoxin emission Cows 31 2)9 0)3 11)4 1)9 Beef 18 3)7 0)6 22)8 9)3 Calves 17 21)4 2)7 90)1 44)8 Sows 43 37)4 3)7 961)6 68)7 Weaners 25 66)6 8)9 347)8 39)8 Fatt.pig 39 49)8 5)2 299)7 56)1 Layers )3 38)7 5247)1 342)5 Broilers )4 46)7 6836)3 294)6

10 106 J. SEEDORF E¹ A. Fig. 6. Emission rates of microorganisms. n"61: Total; Enterobacteriaceae; Fungi Gram -negative bacteria and their microbial content will be affected by the animal s health, especially the occurrence of intestinal disease. The median concentration of airborne endotoxins in inhalable dust samples was about 50 ng/m (Figs 1 and 2). The corresponding concentration in the respirable fraction was usually less than 5 ng/m. The concentration of endotoxin was normally lower at night and this was probably due to the diurnal rhythm in animal activities. 28 Critical concentrations of endotoxin have been suggested on grounds of human exposure. The proposed guidelines for the development of toxic pneumonitis is 200 ng/m and for inflammation of airways 29 is 10 ng/m. However, similar values have not been established for livestock. Our results and those of others given in Table 1 therefore indicate that endotoxin concentrations in poultry houses and perhaps piggeries give cause for concern for the health of stockmen Airborne microorganisms Airborne microbes arise from several sources in livestock buildings and are subject to multiple stresses, such as dehydration and radiation, which influence their viability, and whose effects vary between microorganisms. 8 As a consequence, fully viable, sublethally injured, as well as dead, microorganisms (and their biologically active components like endotoxins) are present in the atmosphere. Sublethally injured microbes may also be able to recover if appropriate nutrient and environmental conditions are available and self-repair mechanisms become active. Sampling and culture of bacteria and fungi on agar media is a standard technique for assay of airborne microbes and therefore the sampling method ideally should allow for full recovery of sublethally damaged microbes. The effects of impact stress on microbial recovery are now recognized. 30 The determination of the true concentration of airborne microbes is fraught with difficulties. Many instruments, such as the Andersen sampler, Casella slit sampler, Reuter centrifugal sampler or the AGI-30 impinger, have been used to measure the concentration of airborne microbes in livestock buildings but each has its own limitations. 31 A new developed automated bioaerosol sampler was used in this study. 22 It allowed remote, independent sampling, which was especially useful at night. The distinction between day and night sampling is important, because of the diurnal variation in animal activity. Tests of a prototype showed that its sampling efficiency was similar to that of the Andersen sampler. 22 Under practical conditions, the measurements of microbial concentration was similar to those observed by other authors (Table 1). The qualitative composition of airborne microorganisms in livestock buildings was characterized mainly by Gram -positive bacteria such as Staphylococcus spp. and Streptococcus spp with a relative abundance of 90% or more. This high proportion is related to the resistance of these species to environmental stresses due to their tough cell wall. In contrast to Gram -positive bacteria, coliform bacteria, such as Enterobacteriaceae spp., did not constitute more than 1 or 2% of the total bacterial count. The main sources of airborne microorganisms are skin, feed, faeces and occasionally sputum. Pathogens such as Staphylococcus aureus, Streptococcus suis, Pasteurella spp., Bordetella spp., Actinobacillus spp., Salmonella spp.

11 EMISSIONS OF AIRBORNE ENDOTOXINS IN LIVESTOCK BUILDINGS IN NORTHERN EUROPE 107 or Mycoplasma spp. can be released from such sources and are important for human and animal health. Dust can be also a carrier and a source of nutrients for fungi, especially for the dominant species of Aspergillus spp., Penicillium spp. and Mucor spp. Some of these fungi are potent producers of mycotoxins such as aflatoxin, which is a carcinogenic agent. The interaction between airborne bacteria and animal-associated microorganisms may play a part in the development of antibiotic resistance. Prevention and treatment of livestock infectious diseases often involves heavy use of antibiotics. Repeated use of an antibiotic can cause resistance as has been found in some airborne bacteria. 32 Furthermore, exchange mechanisms such as plasmid transmission, can also occur within different species. Resistant microorganisms released into the environment via the aerial route may contribute to the development of antibiotic resistance in wild-type strains of other microorganisms. This study showed that the burden of total airborne bacteria is highest in pig and poultry houses. Within this group, broiler and fattening pig units had the highest concentrations. The burden of airborne microorganisms in livestock buildings can be reduced by frequent disinfection, electroprecipitation or artificial UV radiation, while measures which reduce dust generation such as the addition of fat in feed or spraying plant oil will also decrease microbial contamination. airborne bacteria and fungi in the livestock building. In order to predict the subsequent dispersion of viable microbes outdoors, their biological half-life period under varying environmental conditions must be taken into account. 33 Such calculations can provide the theoretical basis for estimating the risk of airborne transmission of disease. 8 The local topography, weather and the design and position of air outlets determine the potential transmission of such contaminants. A pig unit with roof outlets, surrounded by meadows and relatively low emission rates may be more hazardous than a similar building with wall outlets surrounded by trees. Practical investigations on pig farms have recorded concentrations of airborne bacteria of 3)23 log cfu/m in winter and 2)97 log cfu/m in spring at a distance of 100 m from the source. 34 These outdoor concentrations correlated with indoor concentrations of 6)04 and 5)76 log cfu/m, respectively, which are similar to those measured in this study. Recent studies have shown that bioaerosol emissions can be reduced by installing biofilters 36,37 or bioscrubbers. 36,38 These devices have been developed to reduce ammonia emissions but they can also control emissions of bioaerosols. High-energy costs and frequent maintenance to guarantee cleaning efficiency are the main reasons why such devices have not been adopted commercially Emission of airborne endotoxins While emissions of airborne dust and microorganisms from livestock buildings have been studied for many years, there has been less interest in endotoxins and their environmental impact, even though they have been identified as aetiological agents in the pathogenesis of certain occupational pulmonary diseases. 15,19 21 Dwellings downwind of livestock buildings will also be exposed to the plume of emitted endotoxins. The resultant endotoxin concentration are low as shown by preliminary field measurements at 50 and 115 m downwind of a piggery where endotoxin concentrations were 60 and 15 ng/m, respectively. 35 These concentrations outdoors are similar to those detected within cattle buildings (Table 3). It is not known whether human exposure to such concentrations outdoors is hazardous to health and long-term measurements for various weather conditions and emission sources are needed before an informed assessment of risk can be made Emission of airborne bacteria and fungi Emission rates of microbes were estimated from the product of the ventilation rate and the concentration of 5. Conclusions Samples of dust from livestock buildings were analysed for endotoxin content and an automated bioaerosol sampler was developed to monitor the concentration of airborne bacteria and fungi in cattle, pig and poultry houses. Compared with pig and cattle houses, poultry houses, especially percheries, showed the highest concentrations of aerial endotoxin with average values of 692 ng/m for the inhalable dust fraction. These concentrations give grounds for concern with respect to the health of stockmen though critical values for livestock are not known. In cattle and pig houses, the highest endotoxin concentrations were detected in calf and weaner houses. The highest emission rates of inhalable endotoxin were observed in broiler units [817 μg/h (500 kg) liveweight] in contrast to cow houses with the lowest rates [3 μg/h (500 kg) liveweight]. A similar rank order was observed for the concentration of airborne microorganisms. The highest concentration of total bacteria was measured in broiler houses (6)43 log cfu/m ), which also showed the highest emission rate for total bacteria [9)2 log cfu/h (500 kg) live weight]. Studies of the transmission, distribution and biological effects of airborne microbes and endotoxins are needed

12 108 J. SEEDORF E¹ A. to estimate and evaluate the hazards of bioaerosols issuing from livestock units. Acknowledgements The work was funded mainly by the Commission of the European Union as Project No. PL Supplementary funding was also received in the UK from the Ministry of Agriculture, Fisheries and Food via Commission CC 0204; in Germany from the Hannover School of Veterinary Medicine and the Institut für Biosystemtechnik of the Bundesforschungsanstalt für Landwirtschaft; in the Netherlands from the Ministry of Agriculture, Nature Management and Fisheries and in Denmark from the Ministry of Agriculture and Fisheries. We thank the many technicians in all the partner countries, without whose help the project could not have been completed, and also Professor Th. Blaha, Head of the Unit of Epidemiology of the Hannover School of Veterinary Medicine, at Bakum, Germany, for his organisational and logistic support. We thank Chris Michael and his staff at Meaco Sales and Marketing for their enthusiasm and dedication in developing with us the novel wire-less data logging system. Finally, we thank the many farmers in England, Netherlands, Germany and Denmark who not only allowed access to their buildings for the measurements to be made, but also helped in many other ways. References 1 Clark S; Rylander R; Larsson L Airborne bacteria, endotoxin and fungi in dust in poultry and swine confinement buildings. American Industrial Hygiene Association Journal, 1983, 44(7), Cormier Y; Tremblay G; Meriaux A; Brochu G; Lavoie J Airborne microbial contents in two types of swine confinement buildings in Quebec. American Industrial Hygiene Association Journal, 1990, 51(6), Methling W; Mehlhorn G; Beer K; Erwerth W; Förster K Quantität und Qualtität der mikrobiellen Kontamination der Luft in Schweinezuchtställen (Quantity and quality of microbial air contamination in pig breeding farms). Monatshefte fu r Veterinärmedizin, 1981, 36, Erwerth W; Mehlhorn G; Beer K Die mikrobielle Kontamination der Luft in den Kälbersta llen einer Rindermastanlage (Microbial air contamination in calf livestock buildings of a cattle fattening unit). Monatshefte fu r Veterinärmedizin, 1983, 38, Donham K J; Hagland P; Peterson Y; Rylander R; Belin L Environmental and health studies of farm workers in Swedish swine confinement buildings. British Journal of Industrial Medicine, 1989, 46, Donham K J Health effects from work in swine confinement buildings. American Journal of Industrial Medicine, 1990, 17, Hartmann A L; Zeitler M H; Mauermayer R Luftverunreinigungen in Nutztierställen. Arbeitsmedizinische, arbeitshygienische und tierhygienische Aspekte (Air pollution in animal housing. Medical and hygienic aspects). Zentralblatt fu r Arbeitsmedizin, 1986, 36, Wathes C M Air and surface hygiene. In: Livestock Housing (Wathes C M; Charles D R, eds). Wallingford: CAB International, Thiemann G Toxinbildende Pilze in der Stalluft (Toxin producing fungi in the air of animal houses). Proceedings of the 4th Hohenheimer Meeting about Current Zoonosis, 1992, Zeitler M H Hygienische Bedeutung des Staub- und Keimgehaltes der Stalluft (The hygienic importance of airborne dust and microorganisms in animal houses). Bayerisches Landwirtschaftliches Jahrbuch, 1988, 65, Larsson K; Malmberg P; Eklund A; Belin L; Blaschke E Exposure to microorganisms, airway inflammatory changes and immune reactions in asymptomatic dairy farmers. International Archives of Allergy and Applied Immunology, 1988, 87, Pasanen A L; Kalliokoski P; Pasanen P; Salmi T; Tossavainen A Fungi carried from farmers work into farm homes. American Industrial Hygiene Association Journal, 1989, 50(12), Dutkiewicz J; Pomorski Z J H; Sitkowska J; Krysinska- Traczyk E; Skorska C; Prazmo Z; Cholewa G; Wojtowicz H Airborne microorganisms and endotoxin in animal houses. Grana, 1994, 33(2), Crook B; Robertson J F; Travers Glass S A; Botheroyd E M; Lacey J; Topping M D Airborne dust, ammonia, microorganisms and antigens in pig confinement houses and the respiratory health of exposed farm workers. American Industrial Hygiene Association Journal, 1991, 52(7), Heederik D; Brouwer R; Biersteker K; Boleij J S M Relationship of airborne endotoxins and bacteria levels in pig farms with the lung function and respiratory symptoms of farmers. International Archives of Occupational and Environmental Health, 1991, 62, Hinz T; Hartung J; Wiegand B Air quality in a Louisianatype broiler house. Proceedings of XII World Congress in Agricultural Engineering Milano, Italy, 1994, Report No. 94-C-008, 9pp 17 Wiegand B Occurence and distribution of dust and endotoxins in pig house air. Proceedings of the 8th International Congress of Animal Hygiene, St. Paul, Minnesota, USA, 1994, AH Wiegand B; Hartung J; Hinz T; Wiemann H D Luftqualität in Louisianaställen. Teil 2: Keim- und Endotoxingehalt im luftgetragenen Stallstaub (Air quality in Louisiana buildings. Part 2; concentrations of microorganisms and endotoxins in airborne dust). Landbauforschung Vo lkenrode, 1993, 43, Thelin A; Tegler O, Rylander R Lung reaction during poultry handling related to dust and bacterial endotoxin levels. European Journal of Respiratory Disease, 1984, 65, Burrell R; Shu-Hua Ye Toxic risks from inhalation of bacterial. British Journal of Industrial Medicine, 1990, 47, Bergmann K C; Müsken H Endotoxine als Auslöser von Atemwegsobstruktionen bei Landwirten (Endotoxins as

13 EMISSIONS OF AIRBORNE ENDOTOXINS IN LIVESTOCK BUILDINGS IN NORTHERN EUROPE 109 causative agents for obstructive pulmonary diseases in farmers). Atemwegs- und Lungenkrankheiten, 1994, 20(1), Pahl O J; Phillips V R; Lacey J; Hartung J; Wathes CMValidation of a novel bioaerosol sampler with automatic plate exchange. Journal of Aerosol Science, 1997, 28(3), Wathes C M; Phillips, V R; Holden M R; Sneath R W; Short J L; White R P; Hartung J; Seedorf J; Schröder M; Linkert K H; Pedersen S; Takai H; Johnsen J O; Groot Koerkamp P W G; Uenk G H; Metz J H M; Hinz T; Caspary V; Linke S Emissions of aerial pollutants in livestock buildings in Northern Europe: overview of a multinational project. Journal of Agricultural Engineering Research, 1998, 70, Phillips V R; Holden M R; Sneath R W; Short J L; White R P; Johnsen L O; Hartung J; Seedorf J; Linkert K H; Schröder M; Pedersen S; Takai H; Johnsen J O; Metz J H M; Grook Koerkamp P W G; Uenk G H; Wathes CMThe development of robust methods for measuring concentrations and emission rates of gaseous and particulate air pollutants in livestock buildings. Journal of Agricultural Engineering Research, 1998, 70, Hartung J Luftkeimprobennahme im Stall mit einem neuen automatischen Sammler (Sampling of airborne microorganisms in livestock buildings with a new automated bacteria sampler). Proceedings of the 21st Congress of the German Veterinary Society, Bad Nauheim, Germany, 1995, van Ouwerkerk E N J Computer program STALKL V94. 11, energy balance of animal houses, option air flow calculation after CO measurement. EU project PL , Laitinen S; Nevalainen A; Kotimaa M; Liessivuori J; Martikainen P J Relationship between bacterial counts and endotoxin concentrations in the air of wastewater treatment plants. Applied Environmental Microbiology, 1992, 58(11), Pedersen S Time based variation in airborne dust in respect to animal activity. In: Livestock Environment IV. Proceedings of the 4th International Symposium at the University of Warwick, Coventry, England, 6 9 July 1993, Rylander R; Jacobs R R The endotoxin criteria document. Proceedings of the 21st Cotton and Other Organic Dusts Conference, New Orleans, USA, 9 10 January Stewart S L; Grinshpun S A; Willeke K; Terzieva S; Ulevicius V; Donnelly J Effect of impact stress on microbial recovery on an agar surface. Applied Environmental Microbiology, 1995, 61(4), Wathes C M Bioaerosols in animal houses. In: Bioaerosols Handbook (Cox C S; Wathes C M, eds), Pp Boca Raton, F L: CRC Lewis Publishers, Müller W; Wieser P Dust and microbial emissions from animal production. In: Animal Production and Environmental Health (Strauch, D, ed.) Vol. 6. Elsevier, Amsterdam, Müller W; Wieser P; Kühme H Zur Frage der Ausbreitung von Luftkeimen aus Tierställen (The spread of microorganisms emitted from animal houses). Zentralblatt für Veterinärmedizin, Reihe B, 1978, 25, Platz S; Scherer M; Unshelm J Untersuchungen zur Belastung von Mastschweinen sowie der Umgebung von Mastschweineställen durch atembaren Feinstaub, stallspezifische Bakterien und Ammoniak (Studies on load of immissions of fattening pigs and the environment outside of pig-fattening farms caused by lung passing dust particles, pighouse specific airborne bacteria and ammonia). Zentralblatt fu r Hygiene und Umereltmedizin, 1995, 196, Hartung J; Seedorf J; Trickl Th; Gronauer H Emission and distribution of airborne particulates from a piggery. Congress of the International Society of Animal Hygiene, Helsinki, Finland, August Lais S; Büscher W Biologische Abluftreiniger: Wie sie wirken, was sie kosten (Biological exhaust purification techniques: How do they work, what do they cost). DGS Magazine, 1995, 40, Hu gle T; Mannebeck D Biofilter. Landtechnik, 1993, 49, Lais S; Büscher W; Jungbluth T Leistungen und Kosten biologischer Abluftwäscher (Power and costs of bioscrubbers). Landtechnik, 1995, 50,