Identification of Sources of Endotoxin Exposure as Input for Effective Exposure Control Strategies

Identification of Sources of Endotoxin Exposure as Input for Effective Exposure Control Strategies Abstract Objective Aim of the present study is to investigate the levels of endotoxins on product samples from potatoes, onions, and seeds, representing a relevant part of the agro-food industry in the Netherlands, to gather valuable insights in possibilities for exposure control measures early in the process of industrial processing of these products. Methods Endotoxin levels on 330 products samples from companies representing the potato, onion, and seed (processing) industry (four potato-packaging companies, five potato-processing companies, five onion-packaging companies, and four seed-processing companies) were assessed using the Limulus Amboecyte Lysate (LAL) assay. As variation in growth conditions (type of soil, growth type) and product characteristics (surface roughness, dustiness, size, species) are assumed to influence the level of endotoxin on products, different types, and growth conditions were considered when collecting the samples. Additionally, waste material, rotten products, felt material (used for drying), and process water were collected. Results A large variation in the endotoxin levels was found on samples of potatoes, onions, and seeds (overall geometric standard deviation 17), in the range between 0.7 EU g−1 to 16400000 EU g−1. The highest geometric mean endotoxin levels were found in plant material (319600 EU g−1), followed by soil material (49100 EU g−1) and the outer side of products (9300 EU g−1), indicating that removal of plant and soil material early in the process would be an effective exposure control strategy. The high levels of endotoxins found in the limited number of samples from rotten onions indicate that these rotten onions should also be removed early in the process. Mean endotoxin levels found in waste material (only available for seed processing) is similar to the level found in soil material, although the range is much larger. On uncleaned seeds, higher endotoxin levels were found than on cleaned seeds, indicating that cleaning processes are important control measures and also that the waste material should be handled with care. Conclusions Although endotoxin levels in batches of to–be-processed potatoes, onions, and seeds vary quite dramatically, it could be concluded that rotten products, plant material, and waste material contain particularly high endotoxin levels. This information was used to propose control measures to reduce exposure to endotoxins of workers during the production process. agriculture, agro-food products, endotoxin, exposure control measures, exposure control strategy, worker exposure, workplace Background Many workers in the agricultural industry experience health problems due to exposure to organic dust. Endotoxins are well-known contaminants of organic dusts and many studies have reported relationships between exposure to endotoxins and health effects (Dutkiewicz et al., 2011). Even though the risk of occupational exposure to endotoxin is recognized, no occupational exposure limits are available other than the health-based recommended occupational exposure limit (HBROEL) of 90 EU m−3 (8-h time-weighted average) (Dutch Expert Committee on Occupational Safety (DECOS)/Nordic Expert Group (NEG), 2010). Many studies have investigated personal and/or static exposure levels to endotoxins in different industries, like agriculture, sewage treatment plants, herb-processing plants, and wood-processing plants (Dutkiewicz et al., 2011; Duquenne et al., 2012; Paba et al., 2013). In agricultural industries, the HBROEL of 90 EU m−3 is still exceeded up to 100–1000 fold (Spaan et al., 2006), indicating that many workers are at risk of developing health problems. Additionally, also a large variability in exposure levels between sectors, jobs, and tasks performed has been observed. Furthermore, emission of organic dust and thereby endotoxins is observed during the entire processing process of agro-food products. This makes it hard for companies to apply effective control measures to reduce the exposure at the workplace. Nevertheless, it is known that at an individual level companies sometimes invest large amounts of money in control measures, with not always the expected reduction in exposure. Effectiveness of control measures has rarely been investigated in these types of industry and is not a standard part of the implementation of control measures. Since the agro-food products that are produced and processed in the whole of the agricultural industry can be assumed to be the main source of (personal) endotoxin exposure in this sector, gathering more insights into the endotoxin levels on the different parts of these products (soil, plant materials, outer side products, etc.) as they enter a company would be valuable. If it is known where the endotoxins are (mainly) present, this would give insight in how these endotoxins can be removed from these products and therefore how reduction of exposure may be achieved. The sooner the endotoxins are removed from the process, the lower the endotoxins levels are during the rest of the production process. Information about endotoxin levels on agro-food products is currently lacking. The aim of the present study is to investigate the levels of endotoxins on different product samples from potatoes, onions, and seeds, representing a relevant part of the agro-food industry in the Netherlands, to gather insights for exposure control measures early in the production process. Material and methods Sample collection This study was organized in collaboration with three sector organizations in the Netherlands: Nederlandse Aardappel Organisatie (NAO), Frugi Venta, and Plantum, representing respectively the potato, onion, and seed (processing) industry. Representative companies within these sectors were contacted, with a focus on (industrial) processing of products, leading to the inclusion of 18 companies: four potato-packaging companies, five potato-processing companies, five onion-packaging companies, and four seed-processing companies. An inventory of the processes and process circumstances (time from entrance in the company until processing, storage time, storage conditions) in the companies was made to get insight in the possibility for growth of micro-organisms in the companies. As the temperature and humidity during storage were low, and the throughput time of the products in the process was generally short, additional bacterial growth during the stay of the products in the companies was expected to be minimal. Therefore, the focus was on products as they arrive in the companies and not later on in the production process, since these were considered to be indicative as source of endotoxin exposure during the course of the production process. As variation in growth conditions and product characteristics are assumed to influence the level of endotoxin on products, products with variable growth conditions, and characteristics were included, taking into account special process characteristics as well. For potatoes, type of soil they are cultured on and surface roughness were considered, as well as process water used to wash the potatoes and felt material used to dry the potatoes after washing. For onions, type of soil the products are cultured on and onions with different gradations of looseness of the skin (indicated by their color), were considered, and some rotten onions were collected. For seeds, type of sector (agriculture or horticulture), growth type, surface roughness, size, and dustiness of the seeds were considered. Since perennial rye-grass covers ~80% of the grass seeds produced in the Netherlands, this type was selected representing agriculture. During November and December 2013, the companies collected sets of product samples from batches of product entering the company by means of a predefined sampling protocol. Companies were instructed how to collect the samples, received material for collection and storage of the samples, and forms to report contextual information like date of entry in the company, type of sample, type of soil grown they are cultured on. After sampling, the samples were stored (cool) and collected by a researcher from TNO within 1 week, and stored at TNO for a maximum of 5 days at 2–10°C until extraction. To differentiate between the sources of endotoxin in a batch of products that enters a company, for each of the products different types of samples were identified. In general, a distinction was made between the outer side of the product (to look for contamination on the surface of the product), soil material (originating from where the product was cultured), and plant material (e.g. leaves). In case of seeds, soil material was not collected as this was not part of batches of product that enter seed-processing companies. However, waste material from the seed-cleaning process was collected, and a distinction between uncleaned seeds (before the seed-cleaning process) and cleaned seeds (after) was made. Results of a pilot experiment indicated that endotoxin levels at the inside of the products were negligible (results not shown), and therefore not further considered. Extraction method and analysis of samples A pilot study was performed to develop a protocol for extraction and analysis of the samples (Gröllers-Mulderij and Spaan, 2014) (results not shown). For extraction, each sample (1 g of the collected sample, with the exception of the outer side of onions and potatoes, in which case the whole onion or potato was used) was immersed in pyrogen-free water (PFW) with 0.05% Tween-20 in a Greiner tube and rocked vigorously at room temperature on a horizontal shaker for 1 h. After 10 min of centrifugation at 1000 g, per sample 1 ml supernatant was collected, vortexed, and stored in three cryovials at 20°C until analysis. Analysis for the endotoxin content in the extracts was performed in PFW in different concentrations using a quantitative kinetic chromogenic Limulus amebocyte lysate (LAL) assay (Kinetic-QCL 50-650U kit, Lonza, Walkersville, MD, USA). Samples were assayed at three different dilutions in PWF. These dilutions depended on the product type ranging from 1:10 to 1:100000. Samples were retested at higher or lower dilutions when the measured concentrations were too close to respectively the upper or lower limit of detection (LOD) of the assay (LOD was 0.005 EU ml−1). All samples were analyzed in duplicate. Results are presented in EU g−1 product sample. Statistical analysis Data were analyzed with SAS statistical software (version 9.3; SAS Institute, Cary, NC, USA). Endotoxin levels were log-normally distributed. Therefore, all calculations were performed with natural log-transformed concentrations. Crude descriptive endotoxin levels were calculated as arithmetic mean, geometric mean (GM), and geometric standard deviation (GSD), minimum, 25th percentile, 50th percentile (median), 75th percentile, and maximum of the observed distribution of endotoxin levels. SigmaPlot (version 12.5; Systat Software Inc, CA, USA) was used to prepare the figures. The boxplots represent the lower outliers, 10th percentile, 25th percentile, 50th percentile (median), 75th percentile, 90th percentile, and upper outliers. Results Table 1 gives characteristics of the collected product samples. In total 330 samples were collected, of which 86 samples from potato packaging and processing companies, 94 samples from onion-packaging companies, and 150 samples from seeds-processing companies. Table 1. Characteristics of collected product samples.   Total  Potatoes  Onions  Seeds  # companies  18  9  5  4      Processing companies: 5    Agriculture companies: 1      Packaging companies: 4    Horticulture companies: 3  # samples  330  86  94  150  Outer side product  150  25  28  97          Uncleaned seeds: 49          Cleaned seeds: 48  Soil material  55  27a  28  —  Plant materialb  61  20  28  13  Waste material  40  —  —  40  Process water  7  7  —  —  Felt material  7  7  —  —  Rotten onions  10  —  10  —    Total  Potatoes  Onions  Seeds  # companies  18  9  5  4      Processing companies: 5    Agriculture companies: 1      Packaging companies: 4    Horticulture companies: 3  # samples  330  86  94  150  Outer side product  150  25  28  97          Uncleaned seeds: 49          Cleaned seeds: 48  Soil material  55  27a  28  —  Plant materialb  61  20  28  13  Waste material  40  —  —  40  Process water  7  7  —  —  Felt material  7  7  —  —  Rotten onions  10  —  10  —  a Including two lumps of clay. b Plant material comprises leaves and other (dead) parts of the potato plant is case of potatoes, the tails of the onions, and the dried plant material that holds the seeds in case of seeds. View Large Overall, a large variation in the endotoxin levels was found on samples of potatoes, onions, and seeds (overall GSD 17, see Fig. 1 and Table S1 in supplementary material, available at Annals of Work Exposures and Health online), ranging between 0.7 EU g−1 (process water from potato processing) to 16400000 EU g−1 (rotten onion). The highest mean endotoxin levels were found in samples from the onion-packaging industry, followed by the potato processing and packaging industry and the seed-processing industry (GMs respectively 93000 EU g−1, 30900 EU g−1, and 17800 EU g−1). In general, the highest mean endotoxin levels were found in plant material (GM 319600 EU g−1), followed by soil material (GM 49100 EU g−1) and the outer side of products (GM 9300 EU g−1). The mean endotoxin level found in waste material (only available for seed processing) was similar to the level found in soil material, although the observed variation was much larger. Figure 1. View largeDownload slide Overview of endotoxin levels (in EU g−1, on a log-scale) found on the different samples taken of potatoes, onions, and seeds altogether, for the different type of samples (outer side product, soil material, plant material, and waste material) and per type of product. Figure 1. View largeDownload slide Overview of endotoxin levels (in EU g−1, on a log-scale) found on the different samples taken of potatoes, onions, and seeds altogether, for the different type of samples (outer side product, soil material, plant material, and waste material) and per type of product. A large variation in endotoxin levels was found on samples of potatoes from the processing and packaging industry (overall GSD 14, see Fig. 2), with a GM endotoxin level of 30900 EU g−1 and a range between 0.7 EU g−1 (process water) to 2842700 EU g−1 (plant material). On potatoes grown on sand, 10-fold higher GM levels of endotoxins were found than on potatoes grown on clay. As most samples in the potato-processing industry were grown on sand and most samples from the potato-packaging industry were grown on clay, the difference in growth conditions largely explains the observed difference between the two types of industry. On potatoes with an intermediate surface roughness (rough/smooth), highest mean endotoxin level was found. However, based on the distributions, the differences between surfaces type did not appear to be significant and thus the level of skin roughness of potatoes does not seem to influence the endotoxin level present on the skin. In general, the highest endotoxin levels were found on plant material, followed by soil material and the outer side of products. Same differences between plant material, soil material, and the outer side of products were observed after subdividing the results with regard to type of soil and skin roughness. Plant material contains more endotoxin than soil material and the outer side of the potatoes (over 85% of the total amount of endotoxin per gram product is found on plant material). And although the average to extreme quantity of plant material in batches that enter the companies is reported to be very low, in the end plant, material comprises ½ to ¾ of the source of endotoxin in batches of potatoes (see Fig. S1 in supplementary material, available at Annals of Work Exposures and Health online). Figure 2. View largeDownload slide Overview of endotoxin levels (in EU g−1, on a log-scale) found for potatoes; differentiated for the different product samples, type of industry, grown conditions, and skin roughness. Figure 2. View largeDownload slide Overview of endotoxin levels (in EU g−1, on a log-scale) found for potatoes; differentiated for the different product samples, type of industry, grown conditions, and skin roughness. A limited number of samples were collected for felt material and process water, of which especially the process water showed a wide range in observed levels. The endotoxin levels in felt material were all high. The GM endotoxin level found in lumps of clay was GM 1300 EU g−1 (results not shown). A large variation in the endotoxin levels was found on samples from the onion-processing industry (overall GSD 12, see Fig. 3), with a GM endotoxin level of 93100 EU g−1 and a range between 630 EU g−1 (outer side of product) to 16396800 EU g−1 (rotten onion). On onions grown on sand, 2-fold higher levels of endotoxins were found than on onions grown on clay. However, the number of samples from sand were limited, and based on the distributions the difference did not appear to be significant. No apparent difference in endotoxin levels from samples from the different types of onions (yellow, red, and pink) was observed. A limited number of samples of rotten onions (n = 10) was collected, which showed a GM level of 270300 EU g−1. When looking at the different types of samples collected, in general, the highest levels of endotoxins were found on plant material, followed by soil material and the outer side of products. The same differences between plant material, soil material, and the outer side of products were observed when making a distinction between type of soil and type of onion. As onion plant material contains more endotoxin than soil material, rotten onions, and the outer side of the onions, even though the average or even maximum content of plant material in batches that enter the companies are low, in the end the plant material forms ¾ of the source of endotoxin in batches of onions (see Fig. S2 in supplementary material, available at Annals of Work Exposures and Health online). Figure 3. View largeDownload slide Overview of endotoxin levels (in EU g−1, on a log-scale) found for onions; differentiated for the different product samples, grown conditions, and type of onion. Figure 3. View largeDownload slide Overview of endotoxin levels (in EU g−1, on a log-scale) found for onions; differentiated for the different product samples, grown conditions, and type of onion. A large variation in the endotoxin levels was found on samples from the seeds processing and packaging industry (overall GSD 21, see Fig. 4) with a GM endotoxin level of 17800 EU g−1 and a range between 16 EU g−1 (cleaned seed) and 10219500 EU g−1 (waste material from cleaning process). Although, according to the companies only 5–15% of the content of the batches that enter the companies is waste material (consisting of amongst others soil material and plant material), after calculating the relative contribution of waste material as a source of endotoxin exposure, based on amongst others the measured concentrations, waste material can comprise 2/5 (in case of horticulture seeds) to 3/4 (in case of agriculture seeds) of the source of endotoxins from batches of seeds being processed (see Fig. S3 and S4 in supplementary material, available at Annals of Work Exposures and Health online). Figure 4. View largeDownload slide Overview of endotoxin levels (in EU g−1, on a log-scale) found for seeds; differentiated for the different product samples, type of industry, surface roughness, size, dustiness, type of cleaning, and grown conditions. Figure 4. View largeDownload slide Overview of endotoxin levels (in EU g−1, on a log-scale) found for seeds; differentiated for the different product samples, type of industry, surface roughness, size, dustiness, type of cleaning, and grown conditions. The mean endotoxin level in grass seeds (agriculture) was 40 times higher than the mean level found in horticulture seed samples. For horticulture seeds, the highest levels were found in waste material, followed by plant material, uncleaned seeds, and cleaned seeds. A 10-fold higher GM level of endotoxins was found on seeds from open-ground cultivation compared with fruit vegetable. For both growth types, the highest mean levels were found on waste material from the cleaning process. Comparing uncleaned versus cleaned seeds, for open-ground cultivation, the highest levels were found on uncleaned seeds while this is the other way around for fruit seeds. This is probably due to the fact that inside fruits no endotoxins are present yet, but during the seed harvesting process the conditions are favorable for growth or micro-organisms which in the end results in the presence of endotoxins on the seeds after harvesting/cleaning. Pre-cleaned seeds (seeds that have been cleaned before arriving in a company) showed a higher mean level of endotoxins compared with seeds after an additional second round of cleaning. On medium-to-large seeds 3.5 times higher GM endotoxin levels were found than on small seeds, with for both in general more endotoxins on the uncleaned seeds than on the cleaned seeds. In general, higher endotoxin levels were found on samples from seeds with a rough skin or those classified as dusty compared to those with a smooth skin or classified as non-dusty. For both dusty and non-dusty seeds, higher endotoxin levels were found in uncleaned seeds compared with cleaned seeds. The difference in endotoxin levels found on uncleaned and cleaned seeds is smaller for seeds with a rough surface than for seeds with a smooth surface, indicating that organic dust containing amongst others endotoxins is more difficult to remove from a rough surface. A large variation in endotoxin levels was found for different categories of crops, ranging from 1800 EU g−1 (tuberous plant) to 51500 EU g−1 (root crop) (see Table S1 in supplementary material, available at Annals of Work Exposures and Health online). For the different types of seeds (crops) sampled, a high variation was found in mean endotoxin levels, ranging from 200 EU g−1 (lettuce) to 373400 EU g−1 (spinach) (see Table S1 in supplementary material, available at Annals of Work Exposures and Health online). Observed differences between types of seeds are related to the differences in the characteristics of these seeds. As an example, spinach seed has a rough surface and is dusty, whereas lettuce seed has a smooth surface and is not dusty (see Table S2 in supplementary material, available at Annals of Work Exposures and Health online). Discussion In this study, levels of endotoxins on product samples from the potato-, onion-, and seed-processing companies have been investigated. A very large variation in endotoxin levels was found on samples collected in the three industries, with the highest levels observed on samples from onions. The highest levels of endotoxins on potatoes and onions were found on plant material (in the case of onions in the so-called tails) and soil material. The high levels found in the tails of onions can be explained by the relatively good growing conditions for micro-organisms in these tails, namely a combination of availability of moisture and nutrients and a more or less closed environment within the tails. Also, very high levels of endotoxins were found on the limited number of samples of rotten onions, indicating the high level of micro-organisms present in these onions as one would expect with a rotting process. For both agricultural and horticultural seeds, the highest endotoxins were found in the waste material from the cleaning process. This is in line with the finding that in general lower levels of endotoxins were found on cleaned seeds compared to uncleaned seeds. The mean endotoxin level in grass seeds (agriculture) was 40 times higher than the mean level found in horticulture seed samples. This difference might in part be explained by the fact that batches of horticulture seeds have often been pre-cleaned before they enter the seed-processing companies that participated in this study. Another reason might be that grass seeds are grown and processed in much larger volumes than horticulture seeds. This may result in circumstances more favorable for growth of micro-organisms, since the levels found on grass seeds samples are in general also higher than the levels found on horticulture seed samples from open-ground cultivation, which is comparable to growing of grass seed. Highest endotoxin levels in grass seed samples were found in waste material followed by uncleaned seeds and cleaned seeds. A source of endotoxins in the potato packaging and processing companies seems to be the felt material used for drying the potatoes after washing, which may result in transfer of endotoxins from the felt material to the surface of the potatoes again when these are dried. However, as only a limited number of felt samples are collected, no firm conclusions can be drawn. One should take into account that the endotoxin levels are presented in EU g−1 of product, without correcting for the difference in surface/mass ratio of the different types of samples. As endotoxins are only found on the outside of the product samples that have been investigated, the surface/mass ratio of for instance the outside of an onion or a potato is smaller than that of seeds or waste material. Although it was not possible to correct for the differences in surface/mass ratios for the different sample types, it is good to take this into account when looking at the results. Growth conditions seem to have a large influence on the endotoxins levels found. In the case of the potatoes and onions much higher levels were found for products grown on sandy ground compared to clay. Unfortunately, only a limited number of samples of onions grown on sand were collected. It seems probable that the growing conditions in sand are more beneficial for micro-organisms due to the more loose composition of sand compared to clay. The companies also reported that onions grown on sand contain higher levels of micro-organisms than onions grown on clay, as these usually appear more ‘weathered’. These onions are also usually less suitable for long-term storage due to more rotting of onions per batch. As indicated in the results section, characteristics of the seed processing process, like the way seeds are grown and harvested, dustiness and roughness of the skin, amount handled, and batches of seeds that have been pre-cleaned or not before entering the processing plant seem to influence the level of endotoxins found in these batches of seeds. It is assumed that the endotoxin levels on products handled is related to the level of endotoxin exposure of workers handling these products. A limited number of studies investigated endotoxin exposure during certain tasks in relation to endotoxin levels on the products handled during these tasks. Madsen et al. (2009) found that cucumber and tomato leaves are reservoirs of endotoxins, which is in line with the present finding that plant material contains high endotoxin levels. In another study, Madsen et al. (2012) investigated the exposure to endotoxins when handling a problematic batch and a reference batch of grass seeds (Festuca arundinacea) that caused organic dust toxic syndrome in workers. Very high endotoxin levels were found in this problematic batch of seeds, with also much higher levels of endotoxins in uncleaned seeds (for the reference batch 80000 EU g−1 and for the problematic batch 2630000 EU g−1) compared to cleaned seeds (for the reference batch 14800 EU g−1 and for the problematic batch 1710000 EU g−1). Furthermore, the dustiness of the seeds was related to the level of endotoxin found in the air: higher dustiness resulted in higher endotoxin levels. Smit et al. (2006) studied the concentrations of endotoxins in extracts of grass seeds. They found levels between 59 to 361000 EU g−1, which is somewhat lower than the levels found in the present study. Next to grass seeds, Smit et al. (2006) also measured endotoxin levels on vegetable seeds, showing a large variation (40 EU g−1 in lettuce seed to >1000000 EU g−1 in lamb’s lettuce seed). In the present study, a limited number of samples of lettuce and lamb’s lettuce seed were collected, showing endotoxin levels in the same range. They also suggested that characteristics of seeds (size, skin roughness) and external factors like harvesting methods and conditions might influence the endotoxin levels found. Although not considered in the present study, several studies found that the age of plants influences the levels of biological agents (including endotoxins) found on leaves of plants with older plant showing higher levels of endotoxins (Enya et al., 2007; Hansen et al., 2012; Madsen et al., 2014). This further endorses the influence of growth conditions on endotoxin levels present. The results were looked at from the perspective of exposure control as early in the process of processing agro-food products as possible and based on expert judgement. These include general measures such as the use of monitoring systems for example to monitor rotting of products, good practices regarding waste handling, enclosing processes during which (much) dust is emitted, ventilation and filtering of ventilated air, and good cleaning practices could lower exposure to endotoxins. These measures are considered effective to reduce and/or control organic dust and thus endotoxin exposure, but cannot always be implemented due to the type of production process. Furthermore, they can be very costly. Next to insight in the production process, product characteristics are important with regard to choosing control measures. The products studied here are very different from each other: they all have specific product characteristics that should be taken into account when selecting control measures. Furthermore, control measures should not have a negative effect on the quality of the agro-food products. For instance, onions should not come into contact with moisture to prevent them from rotting, and potatoes and seeds should not fall from a large height to prevent them from bruising and germination, respectively. Since the highest levels of endotoxins were found on plant material, measures to remove plant material seems to be an interesting strategy to control exposure for all industries that deal with this type of products. As an example, for the processing industries in this study, the following measures could be implemented for the removal of plant material: for potatoes a sorting machine could be used to automatically sort potatoes and remove plant material in a closed system, and potatoes could be brushed to remove excess soil material. The processing of onions should be arranged in such a way that their tails are removed as early in the process as possible. With regard to seeds, sorting based on cleaning steps using air streams in closed systems could be applied, to be able to remove the waste material (with high endotoxin levels) without these becoming airborne. The fact that removal of plant material can be an effective way to reduce endotoxin levels was also found by Madsen et al. (2014), who observed that clearing of non-dried cucumber plants significantly reduced the exposure to endotoxins compared to clearing of dried cucumber plants. This effect can be explained by the fact that micro-organisms die during the drying process, resulting in the release of endotoxins. In addition to the high endotoxin levels found on plant material, very high levels of endotoxins were found on rotten onions. Removal of these onions as early in the process as possible using a sorting machine should therefore also be recommended. High levels of endotoxins were also found on the limited number of samples from felt material used for drying potatoes. Measures should be taken to prevent (excess) contamination of the washing water and (thus) the felt material, and both should be regularly cleaned or renewed to prevent contamination of the potatoes after the washing step. This study gives valuable information about endotoxin levels on potatoes, onions, and seeds, which can be used when developing effective exposure control strategies taking into account these sources of endotoxin exposure to control this exposure in for instance the agro-food industry. The next step would be to study the effectiveness of the possible control measures, for instance, by means of intervention studies including personal exposure assessment, to help companies choose the best solution for their specific situation. Supplementary Data Supplementary data are available at Annals of Work Exposures and Health online. Conflict of Interest The authors declare no conflict of interest relating to the material presented in this article. Its contents, including any opinions and/or conclusions expressed, are solely those of the authors. Declaration This research was funded by the Ministry of Social Affairs and Employment and three sector organizations in the Netherlands: The Nederlandse Aardappel Organisatie (NAO), Frugi Venta and Plantum, representing respectively the potato, onion and seed (processing) industry. Acknowledgements We would like to thank the companies participated in this study for their effort, and for their information about their products and production process. References Dutkiewicz J, Cisak E, Sroka Jet al.  ( 2011) Biological agents as occupational hazards - selected issues. Ann Agric Environ Med ; 18: 286– 93. Google Scholar PubMed  Duquenne P, Marchand G, Duchaine C. ( 2012) Measurement of endotoxins in bioaerosols at workplace: a critical review of literature and a standardization issue. Ann Occup Hyg ; 57: 137– 72. Google Scholar PubMed  Enya J, Shinohara H, Yoshida Set al.  ( 2007) Culturable leaf-associated bacteria on tomato plants and their potential as biological control agents. Microb Ecol ; 53: 524– 36. Google Scholar CrossRef Search ADS PubMed  Dutch Expert Committee on Occupational Safety (DECOS)/Nordic Expert Group (NEG). ( 2010) (2010) Endotoxins. Health-based recommended occupational exposure limit . The Netherlands: The Hague: Health Council of the Netherlands; publication no. 2010/04OSH. PubMed PubMed  Gröllers-Mulderij M, Spaan S. ( 2014) Ontwikkeling van een protocol voor het verzamelen van (product)monsters, en de opslag, extractie en analyse van deze monsters op endotoxinen . TNO report R10379 (in Dutch). Hansen VM, Meyling NV, Winding A, Eilenberg J, Madsen AM.( 2012) Factors affecting vegetable growers’ exposure to fungal bioaerosols and airborne dust. Ann Occup Hyg ; 56: 170– 81. Google Scholar PubMed  Madsen AM, Hansen VM, Nielsen SHet al.  ( 2009) Exposure to dust and endotoxin of employees in cucumber and tomato nurseries. Ann Occup Hyg ; 53: 129– 38. Google Scholar PubMed  Madsen AM, Tendal K, Schlünssen Vet al.  ( 2012) Organic dust toxic syndrome at a grass seed plant caused by exposure to high concentrations of bioaerosols. Ann Occup Hyg ; 56: 776– 88. Google Scholar PubMed  Madsen AM, Tendal K, Frederiksen MW. ( 2014) Attempts to reduce exposure to fungi, β-glucan, bacteria, endotoxin and dust in vegetable greenhouses and a packaging unit. Sci Total Environ ; 468–469: 1112– 21. Google Scholar CrossRef Search ADS PubMed  Paba E, Tranfo G, Corsetti Fet al.  ( 2013) Indoor exposure to airborne endotoxin: a review of the literature on sampling and analysis methods. Ind Health ; 51: 237– 55. Google Scholar CrossRef Search ADS PubMed  Smit LA, Wouters IM, Hobo MMet al.  ( 2006) Agricultural seed dust as a potential cause of organic dust toxic syndrome. Occup Environ Med ; 63: 59– 67. Google Scholar CrossRef Search ADS PubMed  Spaan S, Wouters IM, Oosting Iet al.  ( 2006) Exposure to inhalable dust and endotoxins in agricultural industries. J Environ Monit ; 8: 63– 72. Google Scholar CrossRef Search ADS PubMed  © The Author(s) 2017. Published by Oxford University Press on behalf of the British Occupational Hygiene Society. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Annals of Work Exposures and Health (formerly Annals Of Occupational Hygiene) Oxford University Press

Loading next page...
 
/lp/ou_press/identification-of-sources-of-endotoxin-exposure-as-input-for-effective-WphMoBJSMX
Publisher
British Occupational Hygiene Society
Copyright
© The Author(s) 2017. Published by Oxford University Press on behalf of the British Occupational Hygiene Society.
ISSN
2398-7308
eISSN
2398-7316
D.O.I.
10.1093/annweh/wxx102
Publisher site
See Article on Publisher Site

Abstract

Abstract Objective Aim of the present study is to investigate the levels of endotoxins on product samples from potatoes, onions, and seeds, representing a relevant part of the agro-food industry in the Netherlands, to gather valuable insights in possibilities for exposure control measures early in the process of industrial processing of these products. Methods Endotoxin levels on 330 products samples from companies representing the potato, onion, and seed (processing) industry (four potato-packaging companies, five potato-processing companies, five onion-packaging companies, and four seed-processing companies) were assessed using the Limulus Amboecyte Lysate (LAL) assay. As variation in growth conditions (type of soil, growth type) and product characteristics (surface roughness, dustiness, size, species) are assumed to influence the level of endotoxin on products, different types, and growth conditions were considered when collecting the samples. Additionally, waste material, rotten products, felt material (used for drying), and process water were collected. Results A large variation in the endotoxin levels was found on samples of potatoes, onions, and seeds (overall geometric standard deviation 17), in the range between 0.7 EU g−1 to 16400000 EU g−1. The highest geometric mean endotoxin levels were found in plant material (319600 EU g−1), followed by soil material (49100 EU g−1) and the outer side of products (9300 EU g−1), indicating that removal of plant and soil material early in the process would be an effective exposure control strategy. The high levels of endotoxins found in the limited number of samples from rotten onions indicate that these rotten onions should also be removed early in the process. Mean endotoxin levels found in waste material (only available for seed processing) is similar to the level found in soil material, although the range is much larger. On uncleaned seeds, higher endotoxin levels were found than on cleaned seeds, indicating that cleaning processes are important control measures and also that the waste material should be handled with care. Conclusions Although endotoxin levels in batches of to–be-processed potatoes, onions, and seeds vary quite dramatically, it could be concluded that rotten products, plant material, and waste material contain particularly high endotoxin levels. This information was used to propose control measures to reduce exposure to endotoxins of workers during the production process. agriculture, agro-food products, endotoxin, exposure control measures, exposure control strategy, worker exposure, workplace Background Many workers in the agricultural industry experience health problems due to exposure to organic dust. Endotoxins are well-known contaminants of organic dusts and many studies have reported relationships between exposure to endotoxins and health effects (Dutkiewicz et al., 2011). Even though the risk of occupational exposure to endotoxin is recognized, no occupational exposure limits are available other than the health-based recommended occupational exposure limit (HBROEL) of 90 EU m−3 (8-h time-weighted average) (Dutch Expert Committee on Occupational Safety (DECOS)/Nordic Expert Group (NEG), 2010). Many studies have investigated personal and/or static exposure levels to endotoxins in different industries, like agriculture, sewage treatment plants, herb-processing plants, and wood-processing plants (Dutkiewicz et al., 2011; Duquenne et al., 2012; Paba et al., 2013). In agricultural industries, the HBROEL of 90 EU m−3 is still exceeded up to 100–1000 fold (Spaan et al., 2006), indicating that many workers are at risk of developing health problems. Additionally, also a large variability in exposure levels between sectors, jobs, and tasks performed has been observed. Furthermore, emission of organic dust and thereby endotoxins is observed during the entire processing process of agro-food products. This makes it hard for companies to apply effective control measures to reduce the exposure at the workplace. Nevertheless, it is known that at an individual level companies sometimes invest large amounts of money in control measures, with not always the expected reduction in exposure. Effectiveness of control measures has rarely been investigated in these types of industry and is not a standard part of the implementation of control measures. Since the agro-food products that are produced and processed in the whole of the agricultural industry can be assumed to be the main source of (personal) endotoxin exposure in this sector, gathering more insights into the endotoxin levels on the different parts of these products (soil, plant materials, outer side products, etc.) as they enter a company would be valuable. If it is known where the endotoxins are (mainly) present, this would give insight in how these endotoxins can be removed from these products and therefore how reduction of exposure may be achieved. The sooner the endotoxins are removed from the process, the lower the endotoxins levels are during the rest of the production process. Information about endotoxin levels on agro-food products is currently lacking. The aim of the present study is to investigate the levels of endotoxins on different product samples from potatoes, onions, and seeds, representing a relevant part of the agro-food industry in the Netherlands, to gather insights for exposure control measures early in the production process. Material and methods Sample collection This study was organized in collaboration with three sector organizations in the Netherlands: Nederlandse Aardappel Organisatie (NAO), Frugi Venta, and Plantum, representing respectively the potato, onion, and seed (processing) industry. Representative companies within these sectors were contacted, with a focus on (industrial) processing of products, leading to the inclusion of 18 companies: four potato-packaging companies, five potato-processing companies, five onion-packaging companies, and four seed-processing companies. An inventory of the processes and process circumstances (time from entrance in the company until processing, storage time, storage conditions) in the companies was made to get insight in the possibility for growth of micro-organisms in the companies. As the temperature and humidity during storage were low, and the throughput time of the products in the process was generally short, additional bacterial growth during the stay of the products in the companies was expected to be minimal. Therefore, the focus was on products as they arrive in the companies and not later on in the production process, since these were considered to be indicative as source of endotoxin exposure during the course of the production process. As variation in growth conditions and product characteristics are assumed to influence the level of endotoxin on products, products with variable growth conditions, and characteristics were included, taking into account special process characteristics as well. For potatoes, type of soil they are cultured on and surface roughness were considered, as well as process water used to wash the potatoes and felt material used to dry the potatoes after washing. For onions, type of soil the products are cultured on and onions with different gradations of looseness of the skin (indicated by their color), were considered, and some rotten onions were collected. For seeds, type of sector (agriculture or horticulture), growth type, surface roughness, size, and dustiness of the seeds were considered. Since perennial rye-grass covers ~80% of the grass seeds produced in the Netherlands, this type was selected representing agriculture. During November and December 2013, the companies collected sets of product samples from batches of product entering the company by means of a predefined sampling protocol. Companies were instructed how to collect the samples, received material for collection and storage of the samples, and forms to report contextual information like date of entry in the company, type of sample, type of soil grown they are cultured on. After sampling, the samples were stored (cool) and collected by a researcher from TNO within 1 week, and stored at TNO for a maximum of 5 days at 2–10°C until extraction. To differentiate between the sources of endotoxin in a batch of products that enters a company, for each of the products different types of samples were identified. In general, a distinction was made between the outer side of the product (to look for contamination on the surface of the product), soil material (originating from where the product was cultured), and plant material (e.g. leaves). In case of seeds, soil material was not collected as this was not part of batches of product that enter seed-processing companies. However, waste material from the seed-cleaning process was collected, and a distinction between uncleaned seeds (before the seed-cleaning process) and cleaned seeds (after) was made. Results of a pilot experiment indicated that endotoxin levels at the inside of the products were negligible (results not shown), and therefore not further considered. Extraction method and analysis of samples A pilot study was performed to develop a protocol for extraction and analysis of the samples (Gröllers-Mulderij and Spaan, 2014) (results not shown). For extraction, each sample (1 g of the collected sample, with the exception of the outer side of onions and potatoes, in which case the whole onion or potato was used) was immersed in pyrogen-free water (PFW) with 0.05% Tween-20 in a Greiner tube and rocked vigorously at room temperature on a horizontal shaker for 1 h. After 10 min of centrifugation at 1000 g, per sample 1 ml supernatant was collected, vortexed, and stored in three cryovials at 20°C until analysis. Analysis for the endotoxin content in the extracts was performed in PFW in different concentrations using a quantitative kinetic chromogenic Limulus amebocyte lysate (LAL) assay (Kinetic-QCL 50-650U kit, Lonza, Walkersville, MD, USA). Samples were assayed at three different dilutions in PWF. These dilutions depended on the product type ranging from 1:10 to 1:100000. Samples were retested at higher or lower dilutions when the measured concentrations were too close to respectively the upper or lower limit of detection (LOD) of the assay (LOD was 0.005 EU ml−1). All samples were analyzed in duplicate. Results are presented in EU g−1 product sample. Statistical analysis Data were analyzed with SAS statistical software (version 9.3; SAS Institute, Cary, NC, USA). Endotoxin levels were log-normally distributed. Therefore, all calculations were performed with natural log-transformed concentrations. Crude descriptive endotoxin levels were calculated as arithmetic mean, geometric mean (GM), and geometric standard deviation (GSD), minimum, 25th percentile, 50th percentile (median), 75th percentile, and maximum of the observed distribution of endotoxin levels. SigmaPlot (version 12.5; Systat Software Inc, CA, USA) was used to prepare the figures. The boxplots represent the lower outliers, 10th percentile, 25th percentile, 50th percentile (median), 75th percentile, 90th percentile, and upper outliers. Results Table 1 gives characteristics of the collected product samples. In total 330 samples were collected, of which 86 samples from potato packaging and processing companies, 94 samples from onion-packaging companies, and 150 samples from seeds-processing companies. Table 1. Characteristics of collected product samples.   Total  Potatoes  Onions  Seeds  # companies  18  9  5  4      Processing companies: 5    Agriculture companies: 1      Packaging companies: 4    Horticulture companies: 3  # samples  330  86  94  150  Outer side product  150  25  28  97          Uncleaned seeds: 49          Cleaned seeds: 48  Soil material  55  27a  28  —  Plant materialb  61  20  28  13  Waste material  40  —  —  40  Process water  7  7  —  —  Felt material  7  7  —  —  Rotten onions  10  —  10  —    Total  Potatoes  Onions  Seeds  # companies  18  9  5  4      Processing companies: 5    Agriculture companies: 1      Packaging companies: 4    Horticulture companies: 3  # samples  330  86  94  150  Outer side product  150  25  28  97          Uncleaned seeds: 49          Cleaned seeds: 48  Soil material  55  27a  28  —  Plant materialb  61  20  28  13  Waste material  40  —  —  40  Process water  7  7  —  —  Felt material  7  7  —  —  Rotten onions  10  —  10  —  a Including two lumps of clay. b Plant material comprises leaves and other (dead) parts of the potato plant is case of potatoes, the tails of the onions, and the dried plant material that holds the seeds in case of seeds. View Large Overall, a large variation in the endotoxin levels was found on samples of potatoes, onions, and seeds (overall GSD 17, see Fig. 1 and Table S1 in supplementary material, available at Annals of Work Exposures and Health online), ranging between 0.7 EU g−1 (process water from potato processing) to 16400000 EU g−1 (rotten onion). The highest mean endotoxin levels were found in samples from the onion-packaging industry, followed by the potato processing and packaging industry and the seed-processing industry (GMs respectively 93000 EU g−1, 30900 EU g−1, and 17800 EU g−1). In general, the highest mean endotoxin levels were found in plant material (GM 319600 EU g−1), followed by soil material (GM 49100 EU g−1) and the outer side of products (GM 9300 EU g−1). The mean endotoxin level found in waste material (only available for seed processing) was similar to the level found in soil material, although the observed variation was much larger. Figure 1. View largeDownload slide Overview of endotoxin levels (in EU g−1, on a log-scale) found on the different samples taken of potatoes, onions, and seeds altogether, for the different type of samples (outer side product, soil material, plant material, and waste material) and per type of product. Figure 1. View largeDownload slide Overview of endotoxin levels (in EU g−1, on a log-scale) found on the different samples taken of potatoes, onions, and seeds altogether, for the different type of samples (outer side product, soil material, plant material, and waste material) and per type of product. A large variation in endotoxin levels was found on samples of potatoes from the processing and packaging industry (overall GSD 14, see Fig. 2), with a GM endotoxin level of 30900 EU g−1 and a range between 0.7 EU g−1 (process water) to 2842700 EU g−1 (plant material). On potatoes grown on sand, 10-fold higher GM levels of endotoxins were found than on potatoes grown on clay. As most samples in the potato-processing industry were grown on sand and most samples from the potato-packaging industry were grown on clay, the difference in growth conditions largely explains the observed difference between the two types of industry. On potatoes with an intermediate surface roughness (rough/smooth), highest mean endotoxin level was found. However, based on the distributions, the differences between surfaces type did not appear to be significant and thus the level of skin roughness of potatoes does not seem to influence the endotoxin level present on the skin. In general, the highest endotoxin levels were found on plant material, followed by soil material and the outer side of products. Same differences between plant material, soil material, and the outer side of products were observed after subdividing the results with regard to type of soil and skin roughness. Plant material contains more endotoxin than soil material and the outer side of the potatoes (over 85% of the total amount of endotoxin per gram product is found on plant material). And although the average to extreme quantity of plant material in batches that enter the companies is reported to be very low, in the end plant, material comprises ½ to ¾ of the source of endotoxin in batches of potatoes (see Fig. S1 in supplementary material, available at Annals of Work Exposures and Health online). Figure 2. View largeDownload slide Overview of endotoxin levels (in EU g−1, on a log-scale) found for potatoes; differentiated for the different product samples, type of industry, grown conditions, and skin roughness. Figure 2. View largeDownload slide Overview of endotoxin levels (in EU g−1, on a log-scale) found for potatoes; differentiated for the different product samples, type of industry, grown conditions, and skin roughness. A limited number of samples were collected for felt material and process water, of which especially the process water showed a wide range in observed levels. The endotoxin levels in felt material were all high. The GM endotoxin level found in lumps of clay was GM 1300 EU g−1 (results not shown). A large variation in the endotoxin levels was found on samples from the onion-processing industry (overall GSD 12, see Fig. 3), with a GM endotoxin level of 93100 EU g−1 and a range between 630 EU g−1 (outer side of product) to 16396800 EU g−1 (rotten onion). On onions grown on sand, 2-fold higher levels of endotoxins were found than on onions grown on clay. However, the number of samples from sand were limited, and based on the distributions the difference did not appear to be significant. No apparent difference in endotoxin levels from samples from the different types of onions (yellow, red, and pink) was observed. A limited number of samples of rotten onions (n = 10) was collected, which showed a GM level of 270300 EU g−1. When looking at the different types of samples collected, in general, the highest levels of endotoxins were found on plant material, followed by soil material and the outer side of products. The same differences between plant material, soil material, and the outer side of products were observed when making a distinction between type of soil and type of onion. As onion plant material contains more endotoxin than soil material, rotten onions, and the outer side of the onions, even though the average or even maximum content of plant material in batches that enter the companies are low, in the end the plant material forms ¾ of the source of endotoxin in batches of onions (see Fig. S2 in supplementary material, available at Annals of Work Exposures and Health online). Figure 3. View largeDownload slide Overview of endotoxin levels (in EU g−1, on a log-scale) found for onions; differentiated for the different product samples, grown conditions, and type of onion. Figure 3. View largeDownload slide Overview of endotoxin levels (in EU g−1, on a log-scale) found for onions; differentiated for the different product samples, grown conditions, and type of onion. A large variation in the endotoxin levels was found on samples from the seeds processing and packaging industry (overall GSD 21, see Fig. 4) with a GM endotoxin level of 17800 EU g−1 and a range between 16 EU g−1 (cleaned seed) and 10219500 EU g−1 (waste material from cleaning process). Although, according to the companies only 5–15% of the content of the batches that enter the companies is waste material (consisting of amongst others soil material and plant material), after calculating the relative contribution of waste material as a source of endotoxin exposure, based on amongst others the measured concentrations, waste material can comprise 2/5 (in case of horticulture seeds) to 3/4 (in case of agriculture seeds) of the source of endotoxins from batches of seeds being processed (see Fig. S3 and S4 in supplementary material, available at Annals of Work Exposures and Health online). Figure 4. View largeDownload slide Overview of endotoxin levels (in EU g−1, on a log-scale) found for seeds; differentiated for the different product samples, type of industry, surface roughness, size, dustiness, type of cleaning, and grown conditions. Figure 4. View largeDownload slide Overview of endotoxin levels (in EU g−1, on a log-scale) found for seeds; differentiated for the different product samples, type of industry, surface roughness, size, dustiness, type of cleaning, and grown conditions. The mean endotoxin level in grass seeds (agriculture) was 40 times higher than the mean level found in horticulture seed samples. For horticulture seeds, the highest levels were found in waste material, followed by plant material, uncleaned seeds, and cleaned seeds. A 10-fold higher GM level of endotoxins was found on seeds from open-ground cultivation compared with fruit vegetable. For both growth types, the highest mean levels were found on waste material from the cleaning process. Comparing uncleaned versus cleaned seeds, for open-ground cultivation, the highest levels were found on uncleaned seeds while this is the other way around for fruit seeds. This is probably due to the fact that inside fruits no endotoxins are present yet, but during the seed harvesting process the conditions are favorable for growth or micro-organisms which in the end results in the presence of endotoxins on the seeds after harvesting/cleaning. Pre-cleaned seeds (seeds that have been cleaned before arriving in a company) showed a higher mean level of endotoxins compared with seeds after an additional second round of cleaning. On medium-to-large seeds 3.5 times higher GM endotoxin levels were found than on small seeds, with for both in general more endotoxins on the uncleaned seeds than on the cleaned seeds. In general, higher endotoxin levels were found on samples from seeds with a rough skin or those classified as dusty compared to those with a smooth skin or classified as non-dusty. For both dusty and non-dusty seeds, higher endotoxin levels were found in uncleaned seeds compared with cleaned seeds. The difference in endotoxin levels found on uncleaned and cleaned seeds is smaller for seeds with a rough surface than for seeds with a smooth surface, indicating that organic dust containing amongst others endotoxins is more difficult to remove from a rough surface. A large variation in endotoxin levels was found for different categories of crops, ranging from 1800 EU g−1 (tuberous plant) to 51500 EU g−1 (root crop) (see Table S1 in supplementary material, available at Annals of Work Exposures and Health online). For the different types of seeds (crops) sampled, a high variation was found in mean endotoxin levels, ranging from 200 EU g−1 (lettuce) to 373400 EU g−1 (spinach) (see Table S1 in supplementary material, available at Annals of Work Exposures and Health online). Observed differences between types of seeds are related to the differences in the characteristics of these seeds. As an example, spinach seed has a rough surface and is dusty, whereas lettuce seed has a smooth surface and is not dusty (see Table S2 in supplementary material, available at Annals of Work Exposures and Health online). Discussion In this study, levels of endotoxins on product samples from the potato-, onion-, and seed-processing companies have been investigated. A very large variation in endotoxin levels was found on samples collected in the three industries, with the highest levels observed on samples from onions. The highest levels of endotoxins on potatoes and onions were found on plant material (in the case of onions in the so-called tails) and soil material. The high levels found in the tails of onions can be explained by the relatively good growing conditions for micro-organisms in these tails, namely a combination of availability of moisture and nutrients and a more or less closed environment within the tails. Also, very high levels of endotoxins were found on the limited number of samples of rotten onions, indicating the high level of micro-organisms present in these onions as one would expect with a rotting process. For both agricultural and horticultural seeds, the highest endotoxins were found in the waste material from the cleaning process. This is in line with the finding that in general lower levels of endotoxins were found on cleaned seeds compared to uncleaned seeds. The mean endotoxin level in grass seeds (agriculture) was 40 times higher than the mean level found in horticulture seed samples. This difference might in part be explained by the fact that batches of horticulture seeds have often been pre-cleaned before they enter the seed-processing companies that participated in this study. Another reason might be that grass seeds are grown and processed in much larger volumes than horticulture seeds. This may result in circumstances more favorable for growth of micro-organisms, since the levels found on grass seeds samples are in general also higher than the levels found on horticulture seed samples from open-ground cultivation, which is comparable to growing of grass seed. Highest endotoxin levels in grass seed samples were found in waste material followed by uncleaned seeds and cleaned seeds. A source of endotoxins in the potato packaging and processing companies seems to be the felt material used for drying the potatoes after washing, which may result in transfer of endotoxins from the felt material to the surface of the potatoes again when these are dried. However, as only a limited number of felt samples are collected, no firm conclusions can be drawn. One should take into account that the endotoxin levels are presented in EU g−1 of product, without correcting for the difference in surface/mass ratio of the different types of samples. As endotoxins are only found on the outside of the product samples that have been investigated, the surface/mass ratio of for instance the outside of an onion or a potato is smaller than that of seeds or waste material. Although it was not possible to correct for the differences in surface/mass ratios for the different sample types, it is good to take this into account when looking at the results. Growth conditions seem to have a large influence on the endotoxins levels found. In the case of the potatoes and onions much higher levels were found for products grown on sandy ground compared to clay. Unfortunately, only a limited number of samples of onions grown on sand were collected. It seems probable that the growing conditions in sand are more beneficial for micro-organisms due to the more loose composition of sand compared to clay. The companies also reported that onions grown on sand contain higher levels of micro-organisms than onions grown on clay, as these usually appear more ‘weathered’. These onions are also usually less suitable for long-term storage due to more rotting of onions per batch. As indicated in the results section, characteristics of the seed processing process, like the way seeds are grown and harvested, dustiness and roughness of the skin, amount handled, and batches of seeds that have been pre-cleaned or not before entering the processing plant seem to influence the level of endotoxins found in these batches of seeds. It is assumed that the endotoxin levels on products handled is related to the level of endotoxin exposure of workers handling these products. A limited number of studies investigated endotoxin exposure during certain tasks in relation to endotoxin levels on the products handled during these tasks. Madsen et al. (2009) found that cucumber and tomato leaves are reservoirs of endotoxins, which is in line with the present finding that plant material contains high endotoxin levels. In another study, Madsen et al. (2012) investigated the exposure to endotoxins when handling a problematic batch and a reference batch of grass seeds (Festuca arundinacea) that caused organic dust toxic syndrome in workers. Very high endotoxin levels were found in this problematic batch of seeds, with also much higher levels of endotoxins in uncleaned seeds (for the reference batch 80000 EU g−1 and for the problematic batch 2630000 EU g−1) compared to cleaned seeds (for the reference batch 14800 EU g−1 and for the problematic batch 1710000 EU g−1). Furthermore, the dustiness of the seeds was related to the level of endotoxin found in the air: higher dustiness resulted in higher endotoxin levels. Smit et al. (2006) studied the concentrations of endotoxins in extracts of grass seeds. They found levels between 59 to 361000 EU g−1, which is somewhat lower than the levels found in the present study. Next to grass seeds, Smit et al. (2006) also measured endotoxin levels on vegetable seeds, showing a large variation (40 EU g−1 in lettuce seed to >1000000 EU g−1 in lamb’s lettuce seed). In the present study, a limited number of samples of lettuce and lamb’s lettuce seed were collected, showing endotoxin levels in the same range. They also suggested that characteristics of seeds (size, skin roughness) and external factors like harvesting methods and conditions might influence the endotoxin levels found. Although not considered in the present study, several studies found that the age of plants influences the levels of biological agents (including endotoxins) found on leaves of plants with older plant showing higher levels of endotoxins (Enya et al., 2007; Hansen et al., 2012; Madsen et al., 2014). This further endorses the influence of growth conditions on endotoxin levels present. The results were looked at from the perspective of exposure control as early in the process of processing agro-food products as possible and based on expert judgement. These include general measures such as the use of monitoring systems for example to monitor rotting of products, good practices regarding waste handling, enclosing processes during which (much) dust is emitted, ventilation and filtering of ventilated air, and good cleaning practices could lower exposure to endotoxins. These measures are considered effective to reduce and/or control organic dust and thus endotoxin exposure, but cannot always be implemented due to the type of production process. Furthermore, they can be very costly. Next to insight in the production process, product characteristics are important with regard to choosing control measures. The products studied here are very different from each other: they all have specific product characteristics that should be taken into account when selecting control measures. Furthermore, control measures should not have a negative effect on the quality of the agro-food products. For instance, onions should not come into contact with moisture to prevent them from rotting, and potatoes and seeds should not fall from a large height to prevent them from bruising and germination, respectively. Since the highest levels of endotoxins were found on plant material, measures to remove plant material seems to be an interesting strategy to control exposure for all industries that deal with this type of products. As an example, for the processing industries in this study, the following measures could be implemented for the removal of plant material: for potatoes a sorting machine could be used to automatically sort potatoes and remove plant material in a closed system, and potatoes could be brushed to remove excess soil material. The processing of onions should be arranged in such a way that their tails are removed as early in the process as possible. With regard to seeds, sorting based on cleaning steps using air streams in closed systems could be applied, to be able to remove the waste material (with high endotoxin levels) without these becoming airborne. The fact that removal of plant material can be an effective way to reduce endotoxin levels was also found by Madsen et al. (2014), who observed that clearing of non-dried cucumber plants significantly reduced the exposure to endotoxins compared to clearing of dried cucumber plants. This effect can be explained by the fact that micro-organisms die during the drying process, resulting in the release of endotoxins. In addition to the high endotoxin levels found on plant material, very high levels of endotoxins were found on rotten onions. Removal of these onions as early in the process as possible using a sorting machine should therefore also be recommended. High levels of endotoxins were also found on the limited number of samples from felt material used for drying potatoes. Measures should be taken to prevent (excess) contamination of the washing water and (thus) the felt material, and both should be regularly cleaned or renewed to prevent contamination of the potatoes after the washing step. This study gives valuable information about endotoxin levels on potatoes, onions, and seeds, which can be used when developing effective exposure control strategies taking into account these sources of endotoxin exposure to control this exposure in for instance the agro-food industry. The next step would be to study the effectiveness of the possible control measures, for instance, by means of intervention studies including personal exposure assessment, to help companies choose the best solution for their specific situation. Supplementary Data Supplementary data are available at Annals of Work Exposures and Health online. Conflict of Interest The authors declare no conflict of interest relating to the material presented in this article. Its contents, including any opinions and/or conclusions expressed, are solely those of the authors. Declaration This research was funded by the Ministry of Social Affairs and Employment and three sector organizations in the Netherlands: The Nederlandse Aardappel Organisatie (NAO), Frugi Venta and Plantum, representing respectively the potato, onion and seed (processing) industry. Acknowledgements We would like to thank the companies participated in this study for their effort, and for their information about their products and production process. References Dutkiewicz J, Cisak E, Sroka Jet al.  ( 2011) Biological agents as occupational hazards - selected issues. Ann Agric Environ Med ; 18: 286– 93. Google Scholar PubMed  Duquenne P, Marchand G, Duchaine C. ( 2012) Measurement of endotoxins in bioaerosols at workplace: a critical review of literature and a standardization issue. Ann Occup Hyg ; 57: 137– 72. Google Scholar PubMed  Enya J, Shinohara H, Yoshida Set al.  ( 2007) Culturable leaf-associated bacteria on tomato plants and their potential as biological control agents. Microb Ecol ; 53: 524– 36. Google Scholar CrossRef Search ADS PubMed  Dutch Expert Committee on Occupational Safety (DECOS)/Nordic Expert Group (NEG). ( 2010) (2010) Endotoxins. Health-based recommended occupational exposure limit . The Netherlands: The Hague: Health Council of the Netherlands; publication no. 2010/04OSH. PubMed PubMed  Gröllers-Mulderij M, Spaan S. ( 2014) Ontwikkeling van een protocol voor het verzamelen van (product)monsters, en de opslag, extractie en analyse van deze monsters op endotoxinen . TNO report R10379 (in Dutch). Hansen VM, Meyling NV, Winding A, Eilenberg J, Madsen AM.( 2012) Factors affecting vegetable growers’ exposure to fungal bioaerosols and airborne dust. Ann Occup Hyg ; 56: 170– 81. Google Scholar PubMed  Madsen AM, Hansen VM, Nielsen SHet al.  ( 2009) Exposure to dust and endotoxin of employees in cucumber and tomato nurseries. Ann Occup Hyg ; 53: 129– 38. Google Scholar PubMed  Madsen AM, Tendal K, Schlünssen Vet al.  ( 2012) Organic dust toxic syndrome at a grass seed plant caused by exposure to high concentrations of bioaerosols. Ann Occup Hyg ; 56: 776– 88. Google Scholar PubMed  Madsen AM, Tendal K, Frederiksen MW. ( 2014) Attempts to reduce exposure to fungi, β-glucan, bacteria, endotoxin and dust in vegetable greenhouses and a packaging unit. Sci Total Environ ; 468–469: 1112– 21. Google Scholar CrossRef Search ADS PubMed  Paba E, Tranfo G, Corsetti Fet al.  ( 2013) Indoor exposure to airborne endotoxin: a review of the literature on sampling and analysis methods. Ind Health ; 51: 237– 55. Google Scholar CrossRef Search ADS PubMed  Smit LA, Wouters IM, Hobo MMet al.  ( 2006) Agricultural seed dust as a potential cause of organic dust toxic syndrome. Occup Environ Med ; 63: 59– 67. Google Scholar CrossRef Search ADS PubMed  Spaan S, Wouters IM, Oosting Iet al.  ( 2006) Exposure to inhalable dust and endotoxins in agricultural industries. J Environ Monit ; 8: 63– 72. Google Scholar CrossRef Search ADS PubMed  © The Author(s) 2017. Published by Oxford University Press on behalf of the British Occupational Hygiene Society.

Journal

Annals of Work Exposures and Health (formerly Annals Of Occupational Hygiene)Oxford University Press

Published: Mar 1, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 12 million articles from more than
10,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Unlimited reading

Read as many articles as you need. Full articles with original layout, charts and figures. Read online, from anywhere.

Stay up to date

Keep up with your field with Personalized Recommendations and Follow Journals to get automatic updates.

Organize your research

It’s easy to organize your research with our built-in tools.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

Monthly Plan

  • Read unlimited articles
  • Personalized recommendations
  • No expiration
  • Print 20 pages per month
  • 20% off on PDF purchases
  • Organize your research
  • Get updates on your journals and topic searches

$49/month

Start Free Trial

14-day Free Trial

Best Deal — 39% off

Annual Plan

  • All the features of the Professional Plan, but for 39% off!
  • Billed annually
  • No expiration
  • For the normal price of 10 articles elsewhere, you get one full year of unlimited access to articles.

$588

$360/year

billed annually
Start Free Trial

14-day Free Trial