- Email: [email protected]
Airborne Fungi and Endotoxin Concentrations in Different Areas within Textile Plants in Taiwan: A 3-Year Study
Environmental Research Section A 89, 58}65 (2002) doi:10.1006/enrs.2002.4345, available online at http://www.idealibrary.com on
Airborne Fungi and Endotoxin Concentrations in Different Areas within Textile Plants in Taiwan: A 3-Year Study Huey-Jen Jenny Su,1 Hsiu-Ling Chen,2 Chin-Feng Huang, Chia-Ying Lin, Fang-Chun Li, and Donald K. Milton Graduate Institute of Environmental and Occupational Health, Medical College, National Cheng Kung University, Tainan, Taiwan, Republic of China Received May 15, 2001
INTRODUCTION Bioaerosols have been found in many occupational environments, including animal feeding houses, poultry slaughter houses, and cotton textile plants. This study was undertaken to examine a group of bioaerosols, the endotoxins, fungi, and bacteria, inside two textile factories over 3 years in Taiwan, where temperature and humidity are usually high year-round. Airborne dust was collected with Alter cassettes attached to personal pumps and analyzed by the Kinetic Limulus Assay with Resistant-parallel-line Estimation. For fungi and bacteria determination, samples were collected using duplicated single-stage impactors, and organisms were counted after incubation. Endotoxin was the ajor contamination inside textile plants. Indoor levels were substantially higher than outdoor concentrations by 63- to 278-fold. The average values of fungi inside and outside the plants were not signiAcantly different. Airborne bacteria levels were higher inside the plants as compared to outside. The carding sites, using only cotton, had extremely high endotoxin levels, greater than those at sites using synthetic Abers. Cotton, may be a major source of endotoxin contamination. In conclusion, the early stage of textile processing seems to generate high endotoxin and bacteria contamination. Priorities should be given to occupational hygiene programs for workers at various sites in textile plants. 2002
Bioaerosols are known to be present in many occupational environments and exposure is associated with a range of adverse health effects. Airborne endotoxin has been identi7ed in animal feeding environments, poultry slaughter houses, cotton textile production, and 7berglass manufacturing factories (Smid et al., 1992; Clapp et al., 1993; Hagmar et al., 1990; Pickrell et al., 1993; Olenchock et al., 1990a,b; Walter et al., 1993). Endotoxin is a component of the outer membrane of gram-negative bacteria, known as lipopolysaccharide (LPS) in its pure form. The lipid portion of LPS, lipid A, containing 3-hydroxy fatty acids, is chemically distinct from all other lipids in biological membranes and responsible for the molecule’s characteristic toxicity (Sandstrom et al., 1992). Health effects related to exposure to gram-negative bacteria and related endotoxins have included respiratory distress, fever and malaise, changes in white blood cell counts, shock, and even death (Gordon et al., 1991; Milton et al., 1990; Sandstrom et al., 1992). Textile plants are among the most studied occupational environments for the presence of endotoxin (Christiani et al., 1993; Olenchock et al., 1990a,b) mainly due to the large quantity of biological materials used. Cotton dust has been shown to be a prime source of endotoxin in textile factories, and endotoxin is suggested to be responsible for byssinosis (Rylander and Bergstorm, 1993; Sigsgaard et al., 1992). Studies have found a strong association between altered lung function and exposure to cotton dust in exposed workers (Buck et al., 1986; Sigsgaard et al., 1992). A case-control study showed a dose}response relationship between endotoxin levels in cotton dust and chronic lung impairment among cotton-handling workers, although there was no relationship between acute FEV1 drop and
Elsevier Science (USA)
Key Words: textile plants; bioaerosol; endotoxin; fungi; bacteria.
1 To whom correspondence should be addressed: E-mail: dmilton@ hohp.harvard.edu. 2 Present address: Graduate Institute of Basic Medical Sciences, Medical College, National Cheng Kung University, Tainan, Taiwan, ROC.
58 0013-9351/02 $35.00 2002 Elsevier Science (USA) All rights reserved.
BIOAEROSOL CONCENTRATIONS IN TEXTILE PLANTS
endotoxin level (Kennedy et al., 1987). Other investigations have demonstrated a statistically signi7cant decrease in FEV1 among cotton workers at different work sites after administration of methacholine (Rylander and Bergstorm, 1993). Textile plants usually consist of many operational sites for blowing, cording and combing, drawing, roving and spinning, and winding and packing. Endotoxin concentrations were found to vary widely both within and between two textile mills in southeast and northwest People’s Republic of China. Endotoxin levels ranged from 15.3 to 1462 ng/m3 in mill 1 and from 1 to 1697 ng/m3 in mill 2. The highest concentrations (above 300 ng/m3) were found in opening, cleaning, carding, and drawing areas, while in roving and spinning areas concentrations were below 100 ng/m3 (Christiani et al., 1993a,b). In Shanghai, People’s Republic of China, concentrations of endotoxin close to 500 ng/m3 were found in cleaning and carding areas (Olenchock et al., 1983a,b). A preliminary study of bacterial and fungal exposures found an average bacterial count of 337.8 cells/L in the blowing section and 247.4 cells/L in the carding section as compared to 20.2 cells/L outside the factory, levels of fungi were 144 cells/L in the blowing section and 13.0 cells/L outside (Rochanachin and Ardsmiti, 1987). An investigation of airborne micro8ora in 8ax spinning and 8ax weaving rooms found that bacterial cell counts in spinning rooms ranged from 10,500 to 125,000 and from 2400 to 10,000 per m3 in 8ax weaving rooms. Studies also identi7ed that levels of Mucor sp., Circinella sp., Hormodendrum sp., Vetricillum sp., Aspergillus sp., and Penicillum sp. were between 16.8 and 54% in weaving rooms (Goscicki et al., 1980). Thus, it seems apparent that different operations are likely to result in different levels of microbial exposure. The textile industry has been an important component of economic growth in Taiwan. A number of large-scale textile plants are operating at full scale, and a large number of workers are employed. Yet, there are limited data on either exposure or health assessments. In addition, concentrations of endotoxin and fungal colonies mass vary due to changes in operating processes and the introduction of synthetic 7bers as new manufacturing materials (Christiani et al., 1993a,b; Rochanachin et al., 1987). This study was undertaken to assess endotoxin, bacterial, and fungal exposure in different areas of a textile plant in which all sites are housed within one giant hall in southern Taiwan. In addition to establishing background data, the information was collected in correction with setting priorities for oc-
59
cupational hygiene programs in the textile industry in Taiwan. MATERIALS AND METHODS
Locations A presampling walk-through was conducted in both plants (plant A and B) to evaluate the layout of each work site and their relationships with each other. In principle, the area zones in the plants were based on different operational functions. Areas of blowing, carding and combing, drawing, roving and spinning, and winding and packing were included for study. As shown in Fig. 1, the combing section was housed inside the giant hall, where an open walkway in the middle served to separate areas using different materials, such as cottons or synthetic 7bers, in the same process. The outdoor sample was taken closest to the entrance of the plant in the 7rst year. Because the two plants were adjacent to each other, only one outdoor site was measured in the center of the courtyard to represent the outdoor levels beginning the second year. Sampling The study was conducted in two textile plants and the samples were taken once a year, between April and May, for 3 sequential years. In this way the in8uence of seasonal and climatic factors was minimized. The sampling time was always at the same time of the day, in the morning hours after the shift began. Airborne endotoxin was collected with polycarbonate membrane 7lters of 0.4-m pore size in 37-mm-diameter polystyrene cassettes, connected to personal air pumps sampling at 1.5 or 2 L/min (Su et al., 2001). For airborne fungi and bacteria deter-
FIG. 1.
Combing section in the textile plant.
60
SU ET AL.
mination, samples were taken by a modi7ed duplicated N6 single-stage Andersen sampler (Heida et al., 1995; Anderson, 1958) attached to malt extract agar plates (MEA) and tryptic soy agar plates (Macher et al., 1995; ACGIH, 1999), operating at 28.3 L/min for 30 s. The sampling height was 1.2}1.5 m near the workers’ breathing zones. Endotoxin Assay Desiccation units were used to preserve each 7lter cassette after collection, and they were stored at 4@C until extraction. Samples were analyzed after extraction by 0.05 M potassium phosphate, 0.01% triethylamine, pH 7.5, in a buffer bath for 60 min at 20@C. The Kinetic Limulus Assay with Resistantparallel-line Estimation (Walters et al., 1994; Milton et al., 1990) was used for analyzing all 7lters by estimating endotoxin potency (EU/m3) based on Reference Standard Endotoxin (EC5, 1 ng"10EU). Fungi and Bacteria Count After the 7eld sampling, the MEA plates were incubated at 25@C for 3}5 days. Fungi were identi7ed morphologically according to mycology references (Larone et al., 1987; Barnett and Hunter, 1987). Bacteria concentrations were calculated by counting colonies under a dissecting microscope. The identi7ed genera included Aspergillus, Penicillium, Cladosporium, and Fusarium. The yields of colonies were then corrected by a positive hole conversion table, which was constructed based on a statistical distribution model, before the calculation of 7nal concentration (CFU/m3) (ACGIH, 1999). RESULTS
TABLE 1 Comparison of Bioaerosol Concentrations in the Indoor and Outdoor Areas for 3 Years Fungi (CFU/m3) Endotoxin Bacteria (ng/m3) (CFU/m3) Indoor-1* Outdoor-1 Indoor-2( Outdoor-2 Indoor-3 Outdoor-3
156.7 2.5 111.1 0.4 153.6 41.0
SN SN 4170 2191 8502 1835
Total
PEN ASP CLA FUS
2357 1273 3785 4523 4807 4019
176 20 63 0 118 70
738 284 189 106 642 35
1113 7 991 0 1798 196 3180 0 612 364 2398 35
Note. PEN, Penicillium; ASP, Aspergillus; CLA, Cladosporium; SN, sample not collected. *First year. (Second year. RThird year.
sampling, inside endotoxin concentration, all-site averaged, was 63 to 278 times the outdoor value. In other studies in Taiwan, an average concentration of 5.20 CFU/m3 total fungi was measured in a hospital (Wu et al., 2000), and a geometric mean of 1212 CFU/m3 total fungi was measured in a day-care center. One outdoor measure reported 1032 CFU/m3 (Li et al., 1997). The levels of fungi appear to be higher inside textile plants than in some other environments. In particular, Fusaium concentrations (average range for 3 years: 7}364 CFU/m3 ) are higher in the textile plant than in other settings, such as a hospital (average: 0.06 CFU/m3) and private homes (extremely low) (Su et al., 2001; Wu et al., 2000). The average levels of Penicillium and Aspergillus inside the plant were all higher than outside, except for a much higher level of Cladosporium outside in the second and third years of sampling.
The Bioaerosol Concentration Indoors and Outdoors Table 1 shows the bioaerosol concentrations present inside and outside of the textile plants. The average endotoxin level inside and outside the textile plant, respectively, is 156.7 and 2.5 ng/m3. Fungal counts were 2357 and 1273 CFU/m3 in the 7rst year. In the second year, the endotoxin level inside and outside was 111.1 and 0.4 ng/m3, respectively; bacterial counts were 4170 and 2191 CFU/m3. Fungal counts were 3785 and 4523 CFU/m3. In the third year, endotoxin levels inside and outside the textile plant were 153.6 and 41.0 ng/m3, bacterial counts were 8502 and 1835 CFU/m3, and fungal counts were 4807 and 4019 CFU/m3. Airborne endotoxin levels inside the textile plants are overwhelmingly higher than those outside. In the 7rst and second
Endotoxin Concentration of Different Working Areas Table 2 shows the microbial concentrations found at different work areas inside two textile plants. Basically, while the two plants have similar operations, plant A is an older plant than B. Throughout the plants, indoor endotoxin levels were higher than outside, except for the packing site. Endotoxin concentrations varied among different work sites within the same plant. At the carding site, using entirely cotton materials, notably higher endotoxin concentrations were observed as compared to the site using synthetic 7bers. This was true for most of the 3-year observations. When work sites are grouped into early (blowing to carding), medium (drawing
61
BIOAEROSOL CONCENTRATIONS IN TEXTILE PLANTS
TABLE 2 Comparison of Endotoxin Concentrations at Different Sites of Two Textile Plants Endotoxin (ng/m3) Endo-1*
Endo-2(
Endo-3
Plant A Blowing Carding (all cotton) Carding (OE) Drawing (TC) Drawing (OE) Packing Outdoor?
79.9 283.1 307.8 156.2 45.6 3.3 1.6
104.0 840 12.2 15.3 66.0 ; 0.4
162.7 217.3 189.1 63.7 244.7 2.9 41.0
Plant B Blowing Carding (all cotton) Carding (TC) Combing Drawing (TC) Roving Spinning Winding Packing Outdoor@
104.5 616.2 75.7 285.6 156.4 185.9 9.3 37.5 4.1 3.4
234.0 24.2 54.5 66.0 38.1 81.5 2.7 16.6 0.5 0.4
241.3 648.6 110.5 186.3 130.0 67.4 10.4 18.9 9.6 41.0
Areas
Note. Endo, endotoxin; OE, open-ended; TC, using synthetic 7bers. ?Plant A outdoors. @Plant B outdoors. *First year. (Second year. Third year.
ing and packing sites. Within the textile plants, Cladosporium levels were higher than concentrations of Penicillium and Aspergillus indoors, especially at the packing site. Levels of Penicillium and Aspergillus families were higher inside than outside, while Cladosporium sp. levels outside were higher than those inside. DISCUSSION
Comparison of Bioaerosol Levels Indoors and Outdoors The outdoor levels were used as a reference value to permit a crude examination as to whether indoor contamination was likely to be present for the bioaerosol of interest. In the current investigation, outdoor levels of ambient endotoxin varied signi7cantly over the 3-year study. Such a variation is not of great concern as long as concentrations are relatively low. In addition, the analytical method for environmental endotoxin adopted in this study is the measurement of extracted polyliposacchride as a surrogate entity. Therefore, the degree of variation observed is acceptable and does not interfere with TABLE 3 Comparison of Bacteria Concentrations at Different Sites of Two Textile Plants Bacteria (CFU/m3)
through spinning), and late (winding through packing) stages of process, endotoxin levels decrease from early to late stages. The blowing and carding sites were the most contaminated environments over the 3 years of observation. Bacteria Concentration of Different Working Areas The blowing, carding and combing, drawing, roving, and packing sites were found to have high bacteria concentrations within the textile plants, and the values appear to the less variable within the same plant based on 2-year data (Table 3). The Spearman correlation between bacteria and endotoxin is statistically signi7cant (r"0.463, P"0.014) except when endotoxin concentrations were less than 100 EU/m3 (r"0.258, P"0.303). Fungi Concentration of Different Working Areas Both Penicillium and Aspergillus showed high levels in the blowing and carding sites (Table 4). High Cladosporium values were found in the blow-
Areas
Bacteria-2(
Bacteria-3R
Plant A Blowing Carding (all cotton) Carding (OE) Drawing (TC) Drawing (OE) Packing Outdoor
3887 6855 6148 5442 2686 2332 2191
7410 5857 14750 11503 2611 6563 1835
Plant B Blowing Carding (all cotton) Carding (TC) Combing Drawing (TC) Roving Spinning Winding Packing Outdoor
1484 NAs NAs 7279 3746 4594 2050 2120 5583 2191
12491 15244 7410 8398 19548 9668 1482 1059 3529 1835
Note. NA, not available; OE, open-ended; TC, using synthetic 7bers; NAs, counts not available because sites were not available for sampling at sampling day. (Second year. RThird year.
62
SU ET AL.
TABLE 4 Comparison of Fungi Concentrations in Different Sites of Two Textile Plants Fungi (CFU/m3) Area
Total-1* Total-2( Total-3 PEN-1* PEN-2( PEN-3
ASP-1*
ASP-2(
ASP-3
CLA-1*
CLA-2(
CLA-3
Plant A Blowing Carding (all cotton) Carding (OE) Drawing (TC) Drawing (OE) Packing Outdoor
2583 2127 2087 818 OAa NAs 1273
NAs NAs 4452 4452 NAs 5442 4523
4973 1238 5082 3454 3386 1868 4019
486 514 511 106 NAa NAs 40
NAs NAs 71 0 NAs 0 0
106 71 141 71 0 106 70
1195 952 953 354 NAa NAs 463
NAs NAs 636 565 NAs 0 212
1340 353 2329 1551 1093 35 35
864 588 549 248 NAa NAs 991
NAs NAs 848 1272 NAs 3604 3180
1130 248 671 565 1341 1269 1398
Plant B Blowing Carding (all cotton) Carding (TC) Combing Drawing (TC) Roving Spinning Winding Packing Outdoor
6208 OAa 1256 1986 2733 2094 906 421 5070 1273
5442 NAs NAs NAs 2615 2544 2686 919 5512 4524
20543 7763 3105 5816 6171 3105 1093 1058 3457 4019
40 NAa 70 143 35 108 34 70 0.0 0.0
0 NAs NAs NAs 0 283 71 71 71 0
106 459 318 71 71 177 0 0 70 70
463 NAa 561 1370 1775 1156 0 0 75 106
212 NAs NAs NAs 212 0 71 0 0 0
637 953 106 457 460 283 35 0 0 35
5321 NAa 556 400 814 757 698 282 2393 991
1060 NAs NAs NAs 1201 1272 1908 354 4664 3180
779 530 176 0 353 282 599 459 2393 1398
Note. OAa, over-loaded by Aspergillus niger; PEN, Penicillium; ASP, Aspergillus; OE, open-ended; CLA, Cladosporium; TC, using synthetic 7ber; NAs, counts not available because sites were not available for sampling at sampling day; NAa, speci7c counts not available because culture plates were over loaded with Aspergillus. *First year. (Second year. Third year.
data interpretation for indoor concentrations. The culturable colonies of outdoor airborne bacteria and fungi were quite stable over the study period. We view this as indirect validation of our sampling strategy because we had planned to sample these sites only on the same month of each spring during the study years. Such a test is fundamental for comparing outdoor bacteria and fungi whose levels can be affected by seasonal factors. In this study, airborne endotoxin concentrations inside the textile plant are higher than those outside. This 7nding is consistent with conclusions from previous studies where endotoxin is a major contaminant inside textile plants (Christiani et al., 1993a,b; Olenchock et al., 1990a,b). In our investigation of fungi, the predominant genera in the textile plant are Cladosporium, Penicillium, Aspergillus, and Fusarium. With an extremely high level of Cladosporium observed outside, Cladosporium was not the predominant airborne genus inside the textile plants. We found high levels of Fusarium inside textile plants over the 3-year period. Fusarium is usually only present at low concentrations in most
environments, such as a day-care center (indoor, 3.4 CFU/m3; outdoor, 2.9 CFU/m3) and houses of asthmatic and nonasthmatic children (Su et al., 2001). It is likely that the raw cottons stored and used in the textile plants are sources for the growth of Fusarium. The exact mechanism requires more detailed investigation. Similar fungal genera have been found at higher inside concentrations than outside at other industrial sites. For instance, a study in a Thai textile factory found that the major types of fungi present included Penicillium sp., Rhizopus sp., Aspergillus sp., and Mucor sp. The average number of microorganisms found inside and outside that factory was 143 and 20 cells/L (Rochanachin and Ardsmiti, 1987). While the indoor concentration of Penicillium and Aspergillus were less than that of Cladosporium in the current study, their concentrations inside were higher than those outside. Therefore, Penicillium and Aspergillus seem to be indicative fungi for textile plant studies. Nevertheless, further assessment remains necessary as to whether indoor contamination is truly present.
BIOAEROSOL CONCENTRATIONS IN TEXTILE PLANTS
63
For airborne bacteria, the average level inside the plant is higher than that of outside. Comparing these results with those from day-care centers where bacterial concentration was 735 CFU/m3 indoors and 384 CFU/m3 outdoors (Li et al., 1997), the indoor to outdoor ratio of bacteria in the current study is similar. In year 3, the concentrations indoors were much higher as compared to outdoor levels. It is possible that the cottons are an active source of indoor bacteria as compared to human activities occurring in a day-care center. In addition, the extremely high indoor concentrations of bacteria seen in the third year may be attributed to the different origins of imported cottons in that period, which resulted, in part, in increased bacterial contamination.
was identi7ed, close to outside levels. The results suggest that endotoxin may be the dominant contaminant in textile plants. Sites using only cottons have notably higher endotoxin concentrations than those using synthetic 7bers. Arti7cial 7bers may generate lower endotoxin quanti7ed in dust, which may explain why other studies have reported that chronic bronchitis is more prevalent in cotton workers than in those working with man-made 7bers (Niven et al., 1997). Moreover, endotoxin concentrations appear to vary depending upon sources of cotton (Olenchock et al., 1983a,b). Thus, origin of cottons used in a plant can affect the amounts of endotoxin found. More in-depth study needs to be performed to reach any further conclusions.
Comparison of Endotoxin Concentrations in Different Operating Areas
Comparison of Fungal and Bacterial Concentration in Different Operating Areas
Our investigation showed the lowest endotoxin concentration (5.8 EU/m3) at the packing site and the highest value (10,836 EU/m3) at the carding site. Endotoxin concentrations of vertically elutriated dusts from all areas of textile mills were above the reported threshold of 90 EU/m3 in a study in the People’s Republic of China (Olenchock et al., 1990a,b) and in central America, airborne endotoxin ranged from a low of 18 EU/m3 in the weaving area to a high of 3138 EU/m3 in the opening area (Christiani et al., 1993a,b). The variation in endotoxin concentration appears to be greater in Taiwan’s textile plants than in these other countries. The high year-round temperature and relative humidity of Taiwan may support the high level of bioaerosols identi7ed in our environments. However, the underlying mechanism of this larger variation remains to be further studied. Our results also indicate that endotoxin levels very widely in different operational areas of the two textile plants. Airborne endotoxin at the carding site of plant A is 535.8$82.3 ng/m3, at the drawing site it is 103.2$48.9 ng/m3, at the roving site it is 288.2$87.3 ng/m3, at the combing site of plant B it is 71.1$8.4 ng/m3, and at the spinning site it is 2.1$0.3 ng/m3. In addition to the drawing site in plant A, the endotoxin levels at the different operational areas is consistent with a previous evaluation in the People’s Republic of China (Olenchock et al., 1983). Endotoxin concentrations decrease from the early to late stages of operation, as cottons are processed from raw materials to yarns through spinning. At the packing site, already approaching the outside, the lowest concentration inside the plant
The bacterial and fungal concentrations show less variation inside the textile plant as compared to endotoxin. The highest contamination of bacteria was found in the carding site of plant A and in the carding and drawing sites of plant B. Fungi show the same pattern as bacteria. However, the textile processes vary in plants. In this study, the processing areas were not separated from each other. Therefore, the bioaerosol level in any site can be in8uenced by contaminants from other operations. Our 7nding is similar to that of another study where the highest level of bacteria was found in the blowing and carding sections; the highest fungi concentration was found in the blowing section (Rochanachin and Ardsmiti, 1987). Correlation of Bacteria and Endotoxin Since endotoxin is a component of the outer membrane of gram-negative bacteria (Sandstrom et al., 1992), a high correlation between levels of endotoxin and bacteria is expected, in theory. Our data show a statistically signi7cant correlation between bacteria levels and endotoxin values at several locations in the textile plants (r"0.463). This is consistent with a previous study in which highly signi7cant associations were found between mean endotoxin exposures and gram-negative bacteria levels (using personal samples) (P0.0001) (Alwis et al., 1999). Thus, total bacterial counts may be an acceptable surrogate for the estimation of endotoxin exposures, at least for this type of environment. Further investigation is warranted to identify the major genera of bacteria contributing to endotoxin.
64
SU ET AL.
CONCLUSIONS
In this study in a subtropical environment, high levels of endotoxin, fungi, and bacteria were found inside two textile plants. The dominant genera were Pencillium, Aspergillus, and Fusarium. The current investigation demonstrates a clear variation in indoor microbial concentrations among different operation areas. Processes with high microbial concentrations included blowing, carding, and drawing areas, and the lowest level was found in the packing site near the outside. Therefore, workers in different operational areas may need to receive different priorities in occupational hygiene programs as based upon levels of exposure to different concentrations of airborne microbial aerosols. ACKNOWLEDGMENTS We are most grateful to the textile plants participating in our study with kindness and patience. We are also in great debt to our colleagues at the Environmental Microbiology Laboratory, National Cheng Kung University Medical College, for sampling and analytical assistance.
REFERENCES ACGIH. (1999). ‘‘Bioaersols: Assessment and Control,’’ pp. 2310}23-11. American Conference of Governmental Industrial Hygienists, Cincinnati, OH. Alwis, K. U., Mandryk, J., and Hocking, A. D. (1999). Exposure to biohazards in wood dust: Bacteria, fungi, endotoxins, and (1;3)-beta-D-glucans. Appl. Occup. Environ. Hyg. 14, 598}608. Andersen, A. A. (1958). New sampler for the collection, sizing, and enumeration of viable airborne particles. J. Bacteriol. 76, 471}484. Barnett, H. L., and Hunter, B. B. (1987). ‘‘Illustrated Genera of Imperfect Fungi,’’ 4th ed. Macmillan Co., New York. Buck, M. G., Schachter, E. N., Fick, R. B., et al. (1986). Biologic activity of puri7ed cotton bract extracts in man and Guinea pig. Environ. Health Perspect. 66, 37}44. Christiani, D. C., Velazquez, A., Wilcox, M., et al. (1993a). Airborne endotoxin concentration in various work area within a cotton mill in Central America. Am. Environ. Res. 60, 187}192. Christiani, D. C., Wegman, D. H., Eisen, E. A., et al. (1993b). Cotton dust and gram-negative bacterial endotoxin correlation in two cotton textile mills. Am. J. Ind. Med. 23, 333}342. Clapp, W. D., Thorne, P. S., Frees, K. L., et al. (1993). The effects of inhalation of grain dust extract and endotoxin on upper and lower airways. Chest 104, 825}830. Duchaine, C., Meriaux, A., Brochu, G., et al. (1999). Airborne micro8ora in Quebec dairy farms: Lack of effect of bacterial hay preservatives. Am. Ind. Hyg. Assoc. J. 60, 89}95. Gordon, T. Balmes,. J., Fine, J., et al. (1991). Airway edema and obstruction in guinea pigs exposed to inhaled endotoxin. Br. J. Ind. Med. 48, 629}635.
Goscicki, J., Woldaeczyk, L., and Bielichowska, G. (1980). Evaluation of work environment in the 8ax textile industry. V. Air micro8ora in the 8ax spinning and 8ax weaving rooms. Med. Pracy. 31, 91}97. [Polish] Hagmar, L., Schutz, A., Hallberg, T., et al. (1990). Health effects of exposure to endotoxins and organic dust in poultry slaughter-house workers. Int. Arch. Occup. Environ. Health. 62, 159}164. Heida, H., Bartman, F., and van der Zee, S. C. (1995). Occupational exposure and indoor air quality monitoring in a composting facility. Am. Ind. Hyg. Assoc. J. 56, 39}43. Kennedy, S. M., Christiani, E. A., Eisen, D. H., et al. (1987). Cotton dust and endotoxin exposure-response relationships in cotton textile workers. Am. R. Res. Dis. 135, 194}200. Larone, D. H. (1987). ‘‘Medically Important Fungi: A Guide to Identi7cation.’’ Elsevier, New York. Li, C. S., Hsiu, C. W., and Tai, M. L. (1997). Indoor pollution and sick building syndrome symptoms among workers in day-care centers. Arch. Environ. Health. 52, 200}207. Macher, J. M., Chatigny, M. A., and Bunge, H. A. (1995). ‘‘Air sampling instruments for evaluation of atmospheric contaminants: sampling airborne microorganisms and aeroallergens,’’ pp. 589}617. ACGIH, Cincinnati, OH. Milton, D. K., Godleski, J. J., Feldman, H. A., et al. (1990). Toxicity of intratracheally instilled cotton dust, cellulose, and endotoxin. Am. R. Res. Dis. 142, 184}192. Milton, D. K., Walters, M. D., Hammond, K., et al. (1996). Worker exposure to endotoxin, phenolic compounds, and formaldehyde in a 7berglass insulation manufacturing plant. Am. Ind. Hyg. Assoc. J. 57, 889}896. Niven, R. M., Fletcher, A. M., Pickering, C. A., et al. (1997). Chronic bronchitis in textile workers. Thorax 52, 22}27. Olenchock, S. A., Christiani, D. C., Mull, J. C., et al. (1983a). Endotoxins in baled cottons and airborne dusts in textile mills in the People’s Republic of China. Appl. Environ. Microb. 46, 817}820. Olenchock, S. A., Mull, J. C., and Jonse, W. G. (1983b). Endotoxins in cotton: Washing effects and size distribution. Am. J. Ind. Med. 4, 515}521. Olenchock, S. A., Christiani, D. C., Mull, J. C., et al. (1990a). Airborne endotoxin concentrations in various work areas within two cotton textile mills in the people’s Republic of China. Biomed. Environ. Sci. 3, 443}451. Olenchock, S. A., May, J. J., Pratt, D. S., et al. (1990b). Presence of endotoxins in different agricultural environments. Am. J. Ind. Med. 18, 279}284. Pal, T. M., De Monchy, J. G., Groothoff, J. W., and Post, D. (1997). The clinical spectrum of humidi7er disease in synthetic 7ber plants. Am. J. Ind. Med. 31, 682}692. Pickrell, J. A., Heber, A. J., Murphy, J. P., et al. (1993). Characterization of particles, ammonia and endotoxin in swine con7nement operations. Vet. Hum. Toxicol. 35, 421}428. Rochanachin, M., and Ardsmiti, N. (1987). A primary study of the number and type of microorganisms in a Thai textile factory. Am. J. Ind. Med. 12, 765}766. Rylander, R., and Bergstorm, R. (1993). Bronchial reactivity among cotton workers in relation to dust and endotoxin exposure. Ann. Occup. Hyg. 37, 57}63. Sandstrom, T., Bjermer, L., and Rylander, R. (1992). Lipopolysaccharide (LPS) inhalation in healthy subjects increases neutrophils, lymphocytes and 7bronectin levels in bronchoalveolar lavage 8uid. Eur. Respir. J. 5, 992}996.
BIOAEROSOL CONCENTRATIONS IN TEXTILE PLANTS Sigsgaard, T., Pedersen, O. F., Juul, S., et al. (1992). Respiratory disorders and atopy in cotton, wool, and othertextile mill workers in Denmark. Am. J. Ind. Med. 22, 163}184. Smid, T., Heederik, D., Mensink, G., et al. (1992). Exposure to dust, endotoxins, and fungi in the animal feed industry. Am. Ind. Hyg. Assoc. J. 53, 362}368. Su, H. J., Wu, P. C., Chen, H. L., et al. (2001). Exposure assessment of indoor allergens, endotoxin and airborne
65
fungi for homes in southern Taiwan. Environ. Res. 85, 135}144. Walter, M., Milton, D. K., Larsson, L., and Kreq, T. (1994). Airborne environmental endotoxin: a cross-validation of sampling and analysis techniques. Appl. Environ. Microbiol. 60, 996}1005. Wu, P. C., Su, H. J. J., and Ho, H. M. (2000). A comparison of sampling media for environmental vible fungi collected in a hospital environment. Environ. Res. 82, 153}157.
Airborne Fungi and Endotoxin Concentrations in Different Areas within Textile Plants in Taiwan: A 3-Year Study Huey-Jen Jenny Su,1 Hsiu-Ling Chen,2 Chin-Feng Huang, Chia-Ying Lin, Fang-Chun Li, and Donald K. Milton Graduate Institute of Environmental and Occupational Health, Medical College, National Cheng Kung University, Tainan, Taiwan, Republic of China Received May 15, 2001
INTRODUCTION Bioaerosols have been found in many occupational environments, including animal feeding houses, poultry slaughter houses, and cotton textile plants. This study was undertaken to examine a group of bioaerosols, the endotoxins, fungi, and bacteria, inside two textile factories over 3 years in Taiwan, where temperature and humidity are usually high year-round. Airborne dust was collected with Alter cassettes attached to personal pumps and analyzed by the Kinetic Limulus Assay with Resistant-parallel-line Estimation. For fungi and bacteria determination, samples were collected using duplicated single-stage impactors, and organisms were counted after incubation. Endotoxin was the ajor contamination inside textile plants. Indoor levels were substantially higher than outdoor concentrations by 63- to 278-fold. The average values of fungi inside and outside the plants were not signiAcantly different. Airborne bacteria levels were higher inside the plants as compared to outside. The carding sites, using only cotton, had extremely high endotoxin levels, greater than those at sites using synthetic Abers. Cotton, may be a major source of endotoxin contamination. In conclusion, the early stage of textile processing seems to generate high endotoxin and bacteria contamination. Priorities should be given to occupational hygiene programs for workers at various sites in textile plants. 2002
Bioaerosols are known to be present in many occupational environments and exposure is associated with a range of adverse health effects. Airborne endotoxin has been identi7ed in animal feeding environments, poultry slaughter houses, cotton textile production, and 7berglass manufacturing factories (Smid et al., 1992; Clapp et al., 1993; Hagmar et al., 1990; Pickrell et al., 1993; Olenchock et al., 1990a,b; Walter et al., 1993). Endotoxin is a component of the outer membrane of gram-negative bacteria, known as lipopolysaccharide (LPS) in its pure form. The lipid portion of LPS, lipid A, containing 3-hydroxy fatty acids, is chemically distinct from all other lipids in biological membranes and responsible for the molecule’s characteristic toxicity (Sandstrom et al., 1992). Health effects related to exposure to gram-negative bacteria and related endotoxins have included respiratory distress, fever and malaise, changes in white blood cell counts, shock, and even death (Gordon et al., 1991; Milton et al., 1990; Sandstrom et al., 1992). Textile plants are among the most studied occupational environments for the presence of endotoxin (Christiani et al., 1993; Olenchock et al., 1990a,b) mainly due to the large quantity of biological materials used. Cotton dust has been shown to be a prime source of endotoxin in textile factories, and endotoxin is suggested to be responsible for byssinosis (Rylander and Bergstorm, 1993; Sigsgaard et al., 1992). Studies have found a strong association between altered lung function and exposure to cotton dust in exposed workers (Buck et al., 1986; Sigsgaard et al., 1992). A case-control study showed a dose}response relationship between endotoxin levels in cotton dust and chronic lung impairment among cotton-handling workers, although there was no relationship between acute FEV1 drop and
Elsevier Science (USA)
Key Words: textile plants; bioaerosol; endotoxin; fungi; bacteria.
1 To whom correspondence should be addressed: E-mail: dmilton@ hohp.harvard.edu. 2 Present address: Graduate Institute of Basic Medical Sciences, Medical College, National Cheng Kung University, Tainan, Taiwan, ROC.
58 0013-9351/02 $35.00 2002 Elsevier Science (USA) All rights reserved.
BIOAEROSOL CONCENTRATIONS IN TEXTILE PLANTS
endotoxin level (Kennedy et al., 1987). Other investigations have demonstrated a statistically signi7cant decrease in FEV1 among cotton workers at different work sites after administration of methacholine (Rylander and Bergstorm, 1993). Textile plants usually consist of many operational sites for blowing, cording and combing, drawing, roving and spinning, and winding and packing. Endotoxin concentrations were found to vary widely both within and between two textile mills in southeast and northwest People’s Republic of China. Endotoxin levels ranged from 15.3 to 1462 ng/m3 in mill 1 and from 1 to 1697 ng/m3 in mill 2. The highest concentrations (above 300 ng/m3) were found in opening, cleaning, carding, and drawing areas, while in roving and spinning areas concentrations were below 100 ng/m3 (Christiani et al., 1993a,b). In Shanghai, People’s Republic of China, concentrations of endotoxin close to 500 ng/m3 were found in cleaning and carding areas (Olenchock et al., 1983a,b). A preliminary study of bacterial and fungal exposures found an average bacterial count of 337.8 cells/L in the blowing section and 247.4 cells/L in the carding section as compared to 20.2 cells/L outside the factory, levels of fungi were 144 cells/L in the blowing section and 13.0 cells/L outside (Rochanachin and Ardsmiti, 1987). An investigation of airborne micro8ora in 8ax spinning and 8ax weaving rooms found that bacterial cell counts in spinning rooms ranged from 10,500 to 125,000 and from 2400 to 10,000 per m3 in 8ax weaving rooms. Studies also identi7ed that levels of Mucor sp., Circinella sp., Hormodendrum sp., Vetricillum sp., Aspergillus sp., and Penicillum sp. were between 16.8 and 54% in weaving rooms (Goscicki et al., 1980). Thus, it seems apparent that different operations are likely to result in different levels of microbial exposure. The textile industry has been an important component of economic growth in Taiwan. A number of large-scale textile plants are operating at full scale, and a large number of workers are employed. Yet, there are limited data on either exposure or health assessments. In addition, concentrations of endotoxin and fungal colonies mass vary due to changes in operating processes and the introduction of synthetic 7bers as new manufacturing materials (Christiani et al., 1993a,b; Rochanachin et al., 1987). This study was undertaken to assess endotoxin, bacterial, and fungal exposure in different areas of a textile plant in which all sites are housed within one giant hall in southern Taiwan. In addition to establishing background data, the information was collected in correction with setting priorities for oc-
59
cupational hygiene programs in the textile industry in Taiwan. MATERIALS AND METHODS
Locations A presampling walk-through was conducted in both plants (plant A and B) to evaluate the layout of each work site and their relationships with each other. In principle, the area zones in the plants were based on different operational functions. Areas of blowing, carding and combing, drawing, roving and spinning, and winding and packing were included for study. As shown in Fig. 1, the combing section was housed inside the giant hall, where an open walkway in the middle served to separate areas using different materials, such as cottons or synthetic 7bers, in the same process. The outdoor sample was taken closest to the entrance of the plant in the 7rst year. Because the two plants were adjacent to each other, only one outdoor site was measured in the center of the courtyard to represent the outdoor levels beginning the second year. Sampling The study was conducted in two textile plants and the samples were taken once a year, between April and May, for 3 sequential years. In this way the in8uence of seasonal and climatic factors was minimized. The sampling time was always at the same time of the day, in the morning hours after the shift began. Airborne endotoxin was collected with polycarbonate membrane 7lters of 0.4-m pore size in 37-mm-diameter polystyrene cassettes, connected to personal air pumps sampling at 1.5 or 2 L/min (Su et al., 2001). For airborne fungi and bacteria deter-
FIG. 1.
Combing section in the textile plant.
60
SU ET AL.
mination, samples were taken by a modi7ed duplicated N6 single-stage Andersen sampler (Heida et al., 1995; Anderson, 1958) attached to malt extract agar plates (MEA) and tryptic soy agar plates (Macher et al., 1995; ACGIH, 1999), operating at 28.3 L/min for 30 s. The sampling height was 1.2}1.5 m near the workers’ breathing zones. Endotoxin Assay Desiccation units were used to preserve each 7lter cassette after collection, and they were stored at 4@C until extraction. Samples were analyzed after extraction by 0.05 M potassium phosphate, 0.01% triethylamine, pH 7.5, in a buffer bath for 60 min at 20@C. The Kinetic Limulus Assay with Resistantparallel-line Estimation (Walters et al., 1994; Milton et al., 1990) was used for analyzing all 7lters by estimating endotoxin potency (EU/m3) based on Reference Standard Endotoxin (EC5, 1 ng"10EU). Fungi and Bacteria Count After the 7eld sampling, the MEA plates were incubated at 25@C for 3}5 days. Fungi were identi7ed morphologically according to mycology references (Larone et al., 1987; Barnett and Hunter, 1987). Bacteria concentrations were calculated by counting colonies under a dissecting microscope. The identi7ed genera included Aspergillus, Penicillium, Cladosporium, and Fusarium. The yields of colonies were then corrected by a positive hole conversion table, which was constructed based on a statistical distribution model, before the calculation of 7nal concentration (CFU/m3) (ACGIH, 1999). RESULTS
TABLE 1 Comparison of Bioaerosol Concentrations in the Indoor and Outdoor Areas for 3 Years Fungi (CFU/m3) Endotoxin Bacteria (ng/m3) (CFU/m3) Indoor-1* Outdoor-1 Indoor-2( Outdoor-2 Indoor-3 Outdoor-3
156.7 2.5 111.1 0.4 153.6 41.0
SN SN 4170 2191 8502 1835
Total
PEN ASP CLA FUS
2357 1273 3785 4523 4807 4019
176 20 63 0 118 70
738 284 189 106 642 35
1113 7 991 0 1798 196 3180 0 612 364 2398 35
Note. PEN, Penicillium; ASP, Aspergillus; CLA, Cladosporium; SN, sample not collected. *First year. (Second year. RThird year.
sampling, inside endotoxin concentration, all-site averaged, was 63 to 278 times the outdoor value. In other studies in Taiwan, an average concentration of 5.20 CFU/m3 total fungi was measured in a hospital (Wu et al., 2000), and a geometric mean of 1212 CFU/m3 total fungi was measured in a day-care center. One outdoor measure reported 1032 CFU/m3 (Li et al., 1997). The levels of fungi appear to be higher inside textile plants than in some other environments. In particular, Fusaium concentrations (average range for 3 years: 7}364 CFU/m3 ) are higher in the textile plant than in other settings, such as a hospital (average: 0.06 CFU/m3) and private homes (extremely low) (Su et al., 2001; Wu et al., 2000). The average levels of Penicillium and Aspergillus inside the plant were all higher than outside, except for a much higher level of Cladosporium outside in the second and third years of sampling.
The Bioaerosol Concentration Indoors and Outdoors Table 1 shows the bioaerosol concentrations present inside and outside of the textile plants. The average endotoxin level inside and outside the textile plant, respectively, is 156.7 and 2.5 ng/m3. Fungal counts were 2357 and 1273 CFU/m3 in the 7rst year. In the second year, the endotoxin level inside and outside was 111.1 and 0.4 ng/m3, respectively; bacterial counts were 4170 and 2191 CFU/m3. Fungal counts were 3785 and 4523 CFU/m3. In the third year, endotoxin levels inside and outside the textile plant were 153.6 and 41.0 ng/m3, bacterial counts were 8502 and 1835 CFU/m3, and fungal counts were 4807 and 4019 CFU/m3. Airborne endotoxin levels inside the textile plants are overwhelmingly higher than those outside. In the 7rst and second
Endotoxin Concentration of Different Working Areas Table 2 shows the microbial concentrations found at different work areas inside two textile plants. Basically, while the two plants have similar operations, plant A is an older plant than B. Throughout the plants, indoor endotoxin levels were higher than outside, except for the packing site. Endotoxin concentrations varied among different work sites within the same plant. At the carding site, using entirely cotton materials, notably higher endotoxin concentrations were observed as compared to the site using synthetic 7bers. This was true for most of the 3-year observations. When work sites are grouped into early (blowing to carding), medium (drawing
61
BIOAEROSOL CONCENTRATIONS IN TEXTILE PLANTS
TABLE 2 Comparison of Endotoxin Concentrations at Different Sites of Two Textile Plants Endotoxin (ng/m3) Endo-1*
Endo-2(
Endo-3
Plant A Blowing Carding (all cotton) Carding (OE) Drawing (TC) Drawing (OE) Packing Outdoor?
79.9 283.1 307.8 156.2 45.6 3.3 1.6
104.0 840 12.2 15.3 66.0 ; 0.4
162.7 217.3 189.1 63.7 244.7 2.9 41.0
Plant B Blowing Carding (all cotton) Carding (TC) Combing Drawing (TC) Roving Spinning Winding Packing Outdoor@
104.5 616.2 75.7 285.6 156.4 185.9 9.3 37.5 4.1 3.4
234.0 24.2 54.5 66.0 38.1 81.5 2.7 16.6 0.5 0.4
241.3 648.6 110.5 186.3 130.0 67.4 10.4 18.9 9.6 41.0
Areas
Note. Endo, endotoxin; OE, open-ended; TC, using synthetic 7bers. ?Plant A outdoors. @Plant B outdoors. *First year. (Second year. Third year.
ing and packing sites. Within the textile plants, Cladosporium levels were higher than concentrations of Penicillium and Aspergillus indoors, especially at the packing site. Levels of Penicillium and Aspergillus families were higher inside than outside, while Cladosporium sp. levels outside were higher than those inside. DISCUSSION
Comparison of Bioaerosol Levels Indoors and Outdoors The outdoor levels were used as a reference value to permit a crude examination as to whether indoor contamination was likely to be present for the bioaerosol of interest. In the current investigation, outdoor levels of ambient endotoxin varied signi7cantly over the 3-year study. Such a variation is not of great concern as long as concentrations are relatively low. In addition, the analytical method for environmental endotoxin adopted in this study is the measurement of extracted polyliposacchride as a surrogate entity. Therefore, the degree of variation observed is acceptable and does not interfere with TABLE 3 Comparison of Bacteria Concentrations at Different Sites of Two Textile Plants Bacteria (CFU/m3)
through spinning), and late (winding through packing) stages of process, endotoxin levels decrease from early to late stages. The blowing and carding sites were the most contaminated environments over the 3 years of observation. Bacteria Concentration of Different Working Areas The blowing, carding and combing, drawing, roving, and packing sites were found to have high bacteria concentrations within the textile plants, and the values appear to the less variable within the same plant based on 2-year data (Table 3). The Spearman correlation between bacteria and endotoxin is statistically signi7cant (r"0.463, P"0.014) except when endotoxin concentrations were less than 100 EU/m3 (r"0.258, P"0.303). Fungi Concentration of Different Working Areas Both Penicillium and Aspergillus showed high levels in the blowing and carding sites (Table 4). High Cladosporium values were found in the blow-
Areas
Bacteria-2(
Bacteria-3R
Plant A Blowing Carding (all cotton) Carding (OE) Drawing (TC) Drawing (OE) Packing Outdoor
3887 6855 6148 5442 2686 2332 2191
7410 5857 14750 11503 2611 6563 1835
Plant B Blowing Carding (all cotton) Carding (TC) Combing Drawing (TC) Roving Spinning Winding Packing Outdoor
1484 NAs NAs 7279 3746 4594 2050 2120 5583 2191
12491 15244 7410 8398 19548 9668 1482 1059 3529 1835
Note. NA, not available; OE, open-ended; TC, using synthetic 7bers; NAs, counts not available because sites were not available for sampling at sampling day. (Second year. RThird year.
62
SU ET AL.
TABLE 4 Comparison of Fungi Concentrations in Different Sites of Two Textile Plants Fungi (CFU/m3) Area
Total-1* Total-2( Total-3 PEN-1* PEN-2( PEN-3
ASP-1*
ASP-2(
ASP-3
CLA-1*
CLA-2(
CLA-3
Plant A Blowing Carding (all cotton) Carding (OE) Drawing (TC) Drawing (OE) Packing Outdoor
2583 2127 2087 818 OAa NAs 1273
NAs NAs 4452 4452 NAs 5442 4523
4973 1238 5082 3454 3386 1868 4019
486 514 511 106 NAa NAs 40
NAs NAs 71 0 NAs 0 0
106 71 141 71 0 106 70
1195 952 953 354 NAa NAs 463
NAs NAs 636 565 NAs 0 212
1340 353 2329 1551 1093 35 35
864 588 549 248 NAa NAs 991
NAs NAs 848 1272 NAs 3604 3180
1130 248 671 565 1341 1269 1398
Plant B Blowing Carding (all cotton) Carding (TC) Combing Drawing (TC) Roving Spinning Winding Packing Outdoor
6208 OAa 1256 1986 2733 2094 906 421 5070 1273
5442 NAs NAs NAs 2615 2544 2686 919 5512 4524
20543 7763 3105 5816 6171 3105 1093 1058 3457 4019
40 NAa 70 143 35 108 34 70 0.0 0.0
0 NAs NAs NAs 0 283 71 71 71 0
106 459 318 71 71 177 0 0 70 70
463 NAa 561 1370 1775 1156 0 0 75 106
212 NAs NAs NAs 212 0 71 0 0 0
637 953 106 457 460 283 35 0 0 35
5321 NAa 556 400 814 757 698 282 2393 991
1060 NAs NAs NAs 1201 1272 1908 354 4664 3180
779 530 176 0 353 282 599 459 2393 1398
Note. OAa, over-loaded by Aspergillus niger; PEN, Penicillium; ASP, Aspergillus; OE, open-ended; CLA, Cladosporium; TC, using synthetic 7ber; NAs, counts not available because sites were not available for sampling at sampling day; NAa, speci7c counts not available because culture plates were over loaded with Aspergillus. *First year. (Second year. Third year.
data interpretation for indoor concentrations. The culturable colonies of outdoor airborne bacteria and fungi were quite stable over the study period. We view this as indirect validation of our sampling strategy because we had planned to sample these sites only on the same month of each spring during the study years. Such a test is fundamental for comparing outdoor bacteria and fungi whose levels can be affected by seasonal factors. In this study, airborne endotoxin concentrations inside the textile plant are higher than those outside. This 7nding is consistent with conclusions from previous studies where endotoxin is a major contaminant inside textile plants (Christiani et al., 1993a,b; Olenchock et al., 1990a,b). In our investigation of fungi, the predominant genera in the textile plant are Cladosporium, Penicillium, Aspergillus, and Fusarium. With an extremely high level of Cladosporium observed outside, Cladosporium was not the predominant airborne genus inside the textile plants. We found high levels of Fusarium inside textile plants over the 3-year period. Fusarium is usually only present at low concentrations in most
environments, such as a day-care center (indoor, 3.4 CFU/m3; outdoor, 2.9 CFU/m3) and houses of asthmatic and nonasthmatic children (Su et al., 2001). It is likely that the raw cottons stored and used in the textile plants are sources for the growth of Fusarium. The exact mechanism requires more detailed investigation. Similar fungal genera have been found at higher inside concentrations than outside at other industrial sites. For instance, a study in a Thai textile factory found that the major types of fungi present included Penicillium sp., Rhizopus sp., Aspergillus sp., and Mucor sp. The average number of microorganisms found inside and outside that factory was 143 and 20 cells/L (Rochanachin and Ardsmiti, 1987). While the indoor concentration of Penicillium and Aspergillus were less than that of Cladosporium in the current study, their concentrations inside were higher than those outside. Therefore, Penicillium and Aspergillus seem to be indicative fungi for textile plant studies. Nevertheless, further assessment remains necessary as to whether indoor contamination is truly present.
BIOAEROSOL CONCENTRATIONS IN TEXTILE PLANTS
63
For airborne bacteria, the average level inside the plant is higher than that of outside. Comparing these results with those from day-care centers where bacterial concentration was 735 CFU/m3 indoors and 384 CFU/m3 outdoors (Li et al., 1997), the indoor to outdoor ratio of bacteria in the current study is similar. In year 3, the concentrations indoors were much higher as compared to outdoor levels. It is possible that the cottons are an active source of indoor bacteria as compared to human activities occurring in a day-care center. In addition, the extremely high indoor concentrations of bacteria seen in the third year may be attributed to the different origins of imported cottons in that period, which resulted, in part, in increased bacterial contamination.
was identi7ed, close to outside levels. The results suggest that endotoxin may be the dominant contaminant in textile plants. Sites using only cottons have notably higher endotoxin concentrations than those using synthetic 7bers. Arti7cial 7bers may generate lower endotoxin quanti7ed in dust, which may explain why other studies have reported that chronic bronchitis is more prevalent in cotton workers than in those working with man-made 7bers (Niven et al., 1997). Moreover, endotoxin concentrations appear to vary depending upon sources of cotton (Olenchock et al., 1983a,b). Thus, origin of cottons used in a plant can affect the amounts of endotoxin found. More in-depth study needs to be performed to reach any further conclusions.
Comparison of Endotoxin Concentrations in Different Operating Areas
Comparison of Fungal and Bacterial Concentration in Different Operating Areas
Our investigation showed the lowest endotoxin concentration (5.8 EU/m3) at the packing site and the highest value (10,836 EU/m3) at the carding site. Endotoxin concentrations of vertically elutriated dusts from all areas of textile mills were above the reported threshold of 90 EU/m3 in a study in the People’s Republic of China (Olenchock et al., 1990a,b) and in central America, airborne endotoxin ranged from a low of 18 EU/m3 in the weaving area to a high of 3138 EU/m3 in the opening area (Christiani et al., 1993a,b). The variation in endotoxin concentration appears to be greater in Taiwan’s textile plants than in these other countries. The high year-round temperature and relative humidity of Taiwan may support the high level of bioaerosols identi7ed in our environments. However, the underlying mechanism of this larger variation remains to be further studied. Our results also indicate that endotoxin levels very widely in different operational areas of the two textile plants. Airborne endotoxin at the carding site of plant A is 535.8$82.3 ng/m3, at the drawing site it is 103.2$48.9 ng/m3, at the roving site it is 288.2$87.3 ng/m3, at the combing site of plant B it is 71.1$8.4 ng/m3, and at the spinning site it is 2.1$0.3 ng/m3. In addition to the drawing site in plant A, the endotoxin levels at the different operational areas is consistent with a previous evaluation in the People’s Republic of China (Olenchock et al., 1983). Endotoxin concentrations decrease from the early to late stages of operation, as cottons are processed from raw materials to yarns through spinning. At the packing site, already approaching the outside, the lowest concentration inside the plant
The bacterial and fungal concentrations show less variation inside the textile plant as compared to endotoxin. The highest contamination of bacteria was found in the carding site of plant A and in the carding and drawing sites of plant B. Fungi show the same pattern as bacteria. However, the textile processes vary in plants. In this study, the processing areas were not separated from each other. Therefore, the bioaerosol level in any site can be in8uenced by contaminants from other operations. Our 7nding is similar to that of another study where the highest level of bacteria was found in the blowing and carding sections; the highest fungi concentration was found in the blowing section (Rochanachin and Ardsmiti, 1987). Correlation of Bacteria and Endotoxin Since endotoxin is a component of the outer membrane of gram-negative bacteria (Sandstrom et al., 1992), a high correlation between levels of endotoxin and bacteria is expected, in theory. Our data show a statistically signi7cant correlation between bacteria levels and endotoxin values at several locations in the textile plants (r"0.463). This is consistent with a previous study in which highly signi7cant associations were found between mean endotoxin exposures and gram-negative bacteria levels (using personal samples) (P0.0001) (Alwis et al., 1999). Thus, total bacterial counts may be an acceptable surrogate for the estimation of endotoxin exposures, at least for this type of environment. Further investigation is warranted to identify the major genera of bacteria contributing to endotoxin.
64
SU ET AL.
CONCLUSIONS
In this study in a subtropical environment, high levels of endotoxin, fungi, and bacteria were found inside two textile plants. The dominant genera were Pencillium, Aspergillus, and Fusarium. The current investigation demonstrates a clear variation in indoor microbial concentrations among different operation areas. Processes with high microbial concentrations included blowing, carding, and drawing areas, and the lowest level was found in the packing site near the outside. Therefore, workers in different operational areas may need to receive different priorities in occupational hygiene programs as based upon levels of exposure to different concentrations of airborne microbial aerosols. ACKNOWLEDGMENTS We are most grateful to the textile plants participating in our study with kindness and patience. We are also in great debt to our colleagues at the Environmental Microbiology Laboratory, National Cheng Kung University Medical College, for sampling and analytical assistance.
REFERENCES ACGIH. (1999). ‘‘Bioaersols: Assessment and Control,’’ pp. 2310}23-11. American Conference of Governmental Industrial Hygienists, Cincinnati, OH. Alwis, K. U., Mandryk, J., and Hocking, A. D. (1999). Exposure to biohazards in wood dust: Bacteria, fungi, endotoxins, and (1;3)-beta-D-glucans. Appl. Occup. Environ. Hyg. 14, 598}608. Andersen, A. A. (1958). New sampler for the collection, sizing, and enumeration of viable airborne particles. J. Bacteriol. 76, 471}484. Barnett, H. L., and Hunter, B. B. (1987). ‘‘Illustrated Genera of Imperfect Fungi,’’ 4th ed. Macmillan Co., New York. Buck, M. G., Schachter, E. N., Fick, R. B., et al. (1986). Biologic activity of puri7ed cotton bract extracts in man and Guinea pig. Environ. Health Perspect. 66, 37}44. Christiani, D. C., Velazquez, A., Wilcox, M., et al. (1993a). Airborne endotoxin concentration in various work area within a cotton mill in Central America. Am. Environ. Res. 60, 187}192. Christiani, D. C., Wegman, D. H., Eisen, E. A., et al. (1993b). Cotton dust and gram-negative bacterial endotoxin correlation in two cotton textile mills. Am. J. Ind. Med. 23, 333}342. Clapp, W. D., Thorne, P. S., Frees, K. L., et al. (1993). The effects of inhalation of grain dust extract and endotoxin on upper and lower airways. Chest 104, 825}830. Duchaine, C., Meriaux, A., Brochu, G., et al. (1999). Airborne micro8ora in Quebec dairy farms: Lack of effect of bacterial hay preservatives. Am. Ind. Hyg. Assoc. J. 60, 89}95. Gordon, T. Balmes,. J., Fine, J., et al. (1991). Airway edema and obstruction in guinea pigs exposed to inhaled endotoxin. Br. J. Ind. Med. 48, 629}635.
Goscicki, J., Woldaeczyk, L., and Bielichowska, G. (1980). Evaluation of work environment in the 8ax textile industry. V. Air micro8ora in the 8ax spinning and 8ax weaving rooms. Med. Pracy. 31, 91}97. [Polish] Hagmar, L., Schutz, A., Hallberg, T., et al. (1990). Health effects of exposure to endotoxins and organic dust in poultry slaughter-house workers. Int. Arch. Occup. Environ. Health. 62, 159}164. Heida, H., Bartman, F., and van der Zee, S. C. (1995). Occupational exposure and indoor air quality monitoring in a composting facility. Am. Ind. Hyg. Assoc. J. 56, 39}43. Kennedy, S. M., Christiani, E. A., Eisen, D. H., et al. (1987). Cotton dust and endotoxin exposure-response relationships in cotton textile workers. Am. R. Res. Dis. 135, 194}200. Larone, D. H. (1987). ‘‘Medically Important Fungi: A Guide to Identi7cation.’’ Elsevier, New York. Li, C. S., Hsiu, C. W., and Tai, M. L. (1997). Indoor pollution and sick building syndrome symptoms among workers in day-care centers. Arch. Environ. Health. 52, 200}207. Macher, J. M., Chatigny, M. A., and Bunge, H. A. (1995). ‘‘Air sampling instruments for evaluation of atmospheric contaminants: sampling airborne microorganisms and aeroallergens,’’ pp. 589}617. ACGIH, Cincinnati, OH. Milton, D. K., Godleski, J. J., Feldman, H. A., et al. (1990). Toxicity of intratracheally instilled cotton dust, cellulose, and endotoxin. Am. R. Res. Dis. 142, 184}192. Milton, D. K., Walters, M. D., Hammond, K., et al. (1996). Worker exposure to endotoxin, phenolic compounds, and formaldehyde in a 7berglass insulation manufacturing plant. Am. Ind. Hyg. Assoc. J. 57, 889}896. Niven, R. M., Fletcher, A. M., Pickering, C. A., et al. (1997). Chronic bronchitis in textile workers. Thorax 52, 22}27. Olenchock, S. A., Christiani, D. C., Mull, J. C., et al. (1983a). Endotoxins in baled cottons and airborne dusts in textile mills in the People’s Republic of China. Appl. Environ. Microb. 46, 817}820. Olenchock, S. A., Mull, J. C., and Jonse, W. G. (1983b). Endotoxins in cotton: Washing effects and size distribution. Am. J. Ind. Med. 4, 515}521. Olenchock, S. A., Christiani, D. C., Mull, J. C., et al. (1990a). Airborne endotoxin concentrations in various work areas within two cotton textile mills in the people’s Republic of China. Biomed. Environ. Sci. 3, 443}451. Olenchock, S. A., May, J. J., Pratt, D. S., et al. (1990b). Presence of endotoxins in different agricultural environments. Am. J. Ind. Med. 18, 279}284. Pal, T. M., De Monchy, J. G., Groothoff, J. W., and Post, D. (1997). The clinical spectrum of humidi7er disease in synthetic 7ber plants. Am. J. Ind. Med. 31, 682}692. Pickrell, J. A., Heber, A. J., Murphy, J. P., et al. (1993). Characterization of particles, ammonia and endotoxin in swine con7nement operations. Vet. Hum. Toxicol. 35, 421}428. Rochanachin, M., and Ardsmiti, N. (1987). A primary study of the number and type of microorganisms in a Thai textile factory. Am. J. Ind. Med. 12, 765}766. Rylander, R., and Bergstorm, R. (1993). Bronchial reactivity among cotton workers in relation to dust and endotoxin exposure. Ann. Occup. Hyg. 37, 57}63. Sandstrom, T., Bjermer, L., and Rylander, R. (1992). Lipopolysaccharide (LPS) inhalation in healthy subjects increases neutrophils, lymphocytes and 7bronectin levels in bronchoalveolar lavage 8uid. Eur. Respir. J. 5, 992}996.
BIOAEROSOL CONCENTRATIONS IN TEXTILE PLANTS Sigsgaard, T., Pedersen, O. F., Juul, S., et al. (1992). Respiratory disorders and atopy in cotton, wool, and othertextile mill workers in Denmark. Am. J. Ind. Med. 22, 163}184. Smid, T., Heederik, D., Mensink, G., et al. (1992). Exposure to dust, endotoxins, and fungi in the animal feed industry. Am. Ind. Hyg. Assoc. J. 53, 362}368. Su, H. J., Wu, P. C., Chen, H. L., et al. (2001). Exposure assessment of indoor allergens, endotoxin and airborne
65
fungi for homes in southern Taiwan. Environ. Res. 85, 135}144. Walter, M., Milton, D. K., Larsson, L., and Kreq, T. (1994). Airborne environmental endotoxin: a cross-validation of sampling and analysis techniques. Appl. Environ. Microbiol. 60, 996}1005. Wu, P. C., Su, H. J. J., and Ho, H. M. (2000). A comparison of sampling media for environmental vible fungi collected in a hospital environment. Environ. Res. 82, 153}157.