Michael T. MUSGROVE
Deana R. JONES
USDA, Agriculture Research Service, Athens, GA, U.S.A.
Introduction
For both food safety and quality reasons, distribution and level of bacteria in a food processing environment are of great concern to plant management, regulatory officials and consumers. Two of the primary methods by which bacteria become distributed in a processing plant are via air and water. In an egg processing facility, airborne bacteria may originate from workers, bird dander, feathers, fecal material, solid waste, wastewater, and dust (Al-Dagal and Fung, 1990; Ellerbroek, 1997; Lutgring et al., 1997; Northcutt et al., 2004). Bacteria may also become aerosolized during cleaning and sanitation of the facility, as high volumes of water hit contaminated surfaces and flood the drains. Processing water may also serve as a vehicle for contamination of eggs, particularly if specific parameters are not monitored during egg washing and egg sanitizer application. Previous research using inoculated simulated egg wash water showed that water temperature, pH and detergent affected the survivability of pure cultures of Escherichia coli, Salmonella, Citrobacter, Enterobacter, Pseudomonas and other bacteria (Kinner and Moats, 1981). The studies described in this paper were conducted in a commercial environment to identify airborne bacteria at various stages of processing and to chemically and microbiologically characterize commercial eggs and egg wash water.
Air sampling methodology
Air was sampled using MicroBio MB2 Air Samplers (F. W. Parrett Limited, London, England), which were attached to standard camera tripods and set to a height of 91.4 cm (center of sampling head). The air samplers were set to draw air for 10 min (1000 L of air) directly on to Rodac plates containing 5 mL of sterile Brain Heart Infusion (BHI) agar. After sample collection, the agar was blended with 10 mL of sterile phosphate buffered saline (PBS) and the blended mixture was plated onto other specialized media. Air was sampled in three facilities that were using different line operations (in-line, off-line, and mixed operations). Sampling sites included areas in or near the following locations: hen house (in-line and mixed operations), farm transition room (in-line and mixed operations), egg washers, egg dryer, packer head, post-processing cooler, nest-run cooler (off-line and mixed operations), loading dock and dry storage area. Air was evaluated for total aerobic bacteria, yeasts/molds, coliforms and pseudomonads.
Egg sampling and wash water analyses
Egg wash water (EWW) was collected and evaluated from three commercial in-line shell egg processing facilities located in the Southeast. Facilities evaluated during this study were designated as Plants X, Y and Z. Water was sampled from each facility during three different visits (replications), and the facilities were visited every 2 weeks on a rotational basis (Plant X, Plant Y, Plant Z, Plant X, Plant Y, Plant Z, etc.). The average processing capacity of each facility was as follows: 373 cases per hour for Plant X; 265 cases per hour for Plant Y; and 292 cases per hour for Plant Z.
During each visit, water was sampled from three different sources: tap water (fresh incoming water), washer 1 (W1) and washer 2 (W2). W1 and W2 samples were collected after the facilities had been operating for 2 hours, and in all cases, this was the first 2 hours of operation on each sampling day. W1 and W2 samples were collected from the waste water stream as the water exited the upper chamber of the washer and before the water passed through the screen and entered the recirculating tank. Water samples were evaluated for temperature at the time of collection, pH, chlorine, soluble iron, total dissolved solids, total suspended solids, total Kjeldahl nitrogen and chemical oxygen demand. For the sake of brevity, this proceeding will focus on EWW temperature, pH, chlorine, iron and total Kjeldahl nitrogen.
Eggs were also collected from the processing line after 2 hours of operation immediately before and during the mid-morning break. Eggs were collected from 12 sampling sites along the processing line; however for the sake of brevity, this proceedings and the accompanying presentation will focus only on eggs collected from W1 and W2. Eggs and water samples were analyzed for total aerobic bacteria, yeasts/molds, Enterobacteriaceae, E. coli and Salmonella.
Results and discussion
Airborne Bacteria
Both the in-line and mixed operations have hen houses directly attached to the processing facilities. The highest counts for airborne total aerobic bacteria were found in the hen house (5.5 to 5.9 log10 cfu/mL air), and these counts decreased as the eggs progressed to the post-processing coolers (2.5 to 3.3 log10 cfu/mL air; Figures 1 and 2). Bacterial counts in the air in the farm transition room, post-processing cooler, loading dock, and dry storage were greater when the samples were collected from a mixed operation (4.7, 3.3, 3.8 and 3.8 log10 cfu/mL air, respectively) as compared to an in-line operation (3.7, 2.7, 2.8, and 3.1 log10 cfu/mL air, respectively). No difference was found due to type of operation for total aerobic bacteria behind the packer heads (3.8 to 4.0 log10 cfu/mL) and in the nest-run coolers (3.5 log10 cfu/mL). The higher microbial counts in the mixed operation may be attributed to contamination from live birds in combination with eggs/carts/flats from different farms and additional product and employee movement through the facility.
Figures 3 and 4 shows the airborne counts for yeasts/molds at the different sampling sites in commercial shell egg processing facilities. Differences in yeasts/molds counts were minimal (< 0.5 log10 cfu/mL air) among the types of operations when samples were collected in or near the hen houses, farm transition rooms, egg washers, egg dryers, packer heads, post-processing coolers and nest-run coolers. Yeasts/molds counts were 0.8 and 0.9 log10 cfu/mL higher in the area near the loading dock for the mixed operation as compared to the off-line and in-line operations, respectively. In addition, there were higher yeasts/molds counts in the dry storage area of the mixed operation (3.8 log10 cfu/mL) as compared to the in-line operation (3.1 log10 cfu/mL). In the mixed operation, the loading dock and dry storage areas were in close proximity to one another, but were separated by a physical barrier. This was not the case for the in-line or off-line operations where the dry storage was in a different location in the facility.
Air in the shell egg processing facilities was also evaluated for coliforms and pseudomonads. The off-line operation had the highest prevalence (number of samples testing positive) of coliform, but the counts were not necessarily higher. Coliform counts ranged from 1.0 to 1.8 log10 cfu/mL air in the off-line operation as compared to 1.6 to 2.5 log10 cfu/mL air in the in-line operation and 1.2 to 2.0 log10 cfu/mL air in the mixed operation. Airborne coliforms were not detected in the post-processing cooler or the dry storage areas. Additionally, airborne coliforms were not detected in the nest-run cooler in the mixed operation or the loading docks of the in-line or mixed operations. The nest-run cooler and loading dock area of the off-line operation tested positive for airborne coliforms, but this was a low level of prevalence (2/12 and 1/12).
Counts of airborne pseudomonads were highest in the hen house (3.1 log10 cfu/mL), farm transition room (2.7 log10 cfu/mL) and behind the egg washers (2.7 to 3.5 log10 cfu/mL). Prevalence of pseudomonades was low at all other sampling sites in the mixed operation. Counts for airborne pseudomonades measured behind the egg dryer, packer head, loading dock and dry storage were not significantly different among the different types of operations. Pseudomonad counts were lowest in the post-processing and nest-run coolers.
Egg Wash Water
Table 1 shows the temperature, pH, chlorine, and iron for W1 and W2 samples collected from plants X, Y and Z. EWW temperature ranged from 103.5 to 111.4°F, and was typically more consistent in Plants Y and Z. When the pH of the EWW was measured, both W1 and W2 in Plant Z were found to have the highest pH values (pH 11.4 and 11.2, respectively). The pH of the EWW from Plants X and Y ranged from 10.0 to 10.6. The pH of the EWW in W2 in plant X was found to be below 10.0 during two of the three sample collections (readings pH 9.9, 9.1 and 11.0). These lower pH values were of interest because pH plays an important role in reducing microbiological contamination.
Plant Y reported using city water to process eggs, while Plants X and Z reported using well water to process eggs. Highest chlorine levels were found in EWW from Plant Z (4.5 and 2.3 mg/L), followed by the EWW from Plants X (2.7 and 2.6 mg/L) and Y (0.9 and 0.9 mg/L). The antimicrobial properties of chlorine are pH dependent with a higher efficacy at pH 6.5 to 7.5. With the pH of EWW at 10.0 and above, more of the chlorine exists as hypochlorite ion which is not as destructive to bacteria as the hypochlorous acid (predominate species at pH 6.5 to 7.5).
Iron levels are also important to bacterial growth, and the average values in the EWW ranged from 0.3 to 1.6 mg/L. Iron was found to be above the 2.0 mg/L recommended limit in W1 in Plant X during one of the sample collections (triplicate readings of 2.3, 2.7 and 2.6 mg/L). During another collection time, iron levels were just below the recommended limit in plant x (~1.7 mg/L). This may be related to the age of the plant, which is over 20 years old and may have some corrosion/debris in the water line.
Total Kjeldahl nitrogen reflects the amount of protein that is lost to the waste stream. Highest values of TKN were found in the EWW from W1 in Plant Z (302 mg/L), followed by W1 and W2 in Plant X (205 and 204 mg/L, respectively). Plant Y had the lowest TKN values (128 and 81 mg/L). Variation in TNK can be related to numerous production and processing variables that affect egg shell quality and the number of eggs broken during processing. In the present study, Plants X and Z used rewash belts where eggs were continuously cycled through the washers. In addition, Plant X and Y oiled their eggs and oiling has been reported to improve shell strength by protecting the cuticle (Ball et al., 1976). The combination of these factors may have contributed to the wide range in TKN values among the plants.
Figure 5 shows the level of total aerobic bacteria found in the tap water and EWW from W1 and W2 from Plants X, Y and Z. Counts were lowest in W1 and W2 from Plant Z where the EWW pH was found to be the highest (pH 11.4 and 11.2). The highest counts were found in W1 and W2 samples from Plants X and Y where the EWW pH was well below 11.0. No difference was found in the yeasts/molds counts among the different plants so these data were pooled and reported for all three plants (Figure 6). The highest yeasts/molds counts were found in EWW from W1 (1.4 log), and these counts were 0.6 log units higher than the counts observed in EWW from W2.
Table 2 shows the microbiological counts on the eggs collected from W1 and W2 in Plants X, Y and Z (Musgrove et al., 2005). The counts for total aerobic bacteria were higher on the eggs collected from Plant X compared to the counts found on eggs collected from Plants Y and Z. Total aerobic bacteria were also higher on the eggs collected from W1 than W2 (mean values of log 2.35 and 1.9, respectively). Although there were some significant differences in yeasts/molds, Enterobacteriaceae and E. coli counts on eggs collected from the different plants, these differences were all less than 0.5 log, and therefore would have little practical significance. Moreover, low levels of yeasts/molds, Enterobacteriaceae and E. coli were found on all of the eggs collected from the washers (Musgrove et al., 2003).
Salmonella was found in one of the tap water samples collected from Plant X; however, it is not clear how this contamination occurred. For recovery of Salmonella from eggs, two different methods were used. Overall, Salmonella was found on 35/396 eggs (8.8% positive) following enrichment. This is slightly higher than the previously reported incidence (4.4%) where recovery was based on egg rinsing. A higher recovery is likely due to methodology which combined egg rinsing with an egg shell "crushing" procedure (Musgrove et al., 2004).
Conclusion
During air sampling, the levels of airborne bacteria were highest in the presence of the live birds. Airborne counts of yeasts/molds were relatively consistent throughout the egg processing facilities, and coliform/E. coli was either not detected or was detected at low levels in the final processing stages (post-processing egg cooler, and egg dry storage area). Air sampling data showed that there is a definite need for physical barriers during processing, control of ventilation and control of product/worker movement. Other factors that have a significant effect on aerosolized bacteria in processing facilities are air flow, air distribution, temperature, relative humidity, design of the facility and facility maintenance (Lutgring et al., 1997). In the processing plants where airborne bacteria were monitored, the facilities had been expanded to accommodate increased production. When the facilities were expanded, product flow/ventilation were not considered, and this is likely the reason for some of the data presented here.
Recovery of bacteria from eggs and EWW verified the need to monitor and control wash water temperature, pH, iron and organic debris. Lowest counts on eggs and in water were associated with the processing plants that maintained EWW pH above 11.0.
References
Al-Dagal, M. and D. Y. C. Fung. 1990. Aeromicrobiology: a review. Crit. Rev. Food Sci. Nutr. 29:333-340.
Ball, R. F., J. F. Hill, V. Logan and J. Lyman. 1976. The effect of washing, oiling, holding and temperature of eggs on shell strength. Poult. Sci. 55:335-340.
Ellerbroek, L. 1997. Airborne microflora in poultry slaughtering establishments. Food Microbiol. 14:527-531.
Kinner, J. A. and W. A. Moats. 1981. Effect of temperature, pH, and detergent on survival of bacteria associated with shell eggs. Poult. Sci. 60:761-767.
Lutgring, K. R., R. H. Linton, N. J. Zimmerman, M. Peugh and A. J. Heber. 1997. Distribution and quantification of bioaerosols in poultry-slaughtering plants. J. Food Prot. 60:804-810.
Musgrove, M. T., D. R. Jones, J. K. Northcutt, M. A. Harrison, and N. A. Cox. 2003 Microbiology of commercial shell egg processing. 32, 118-126. Tsukuba, Ibaraki, UJNR. Proceedings of the 32nd Annual Meeting of the United States-Japan
Cooperative Program in Natural Resources Food and Agriculture Panel. 11-9. (Conference Proceeding).
Musgrove, M. T., D. R. Jones, J. K. Northcutt, N. A. Cox, M. A. Harrison, S. R. Ladely, and P. J. Fedorka-Cray. 2004. Characteristics of Salmonella, Escherichia coli, and other Enterobacteriaceae isolated from U.S. Commercial Shell Eggs. Cherry, J. P. and Pavlath, A. E. 33, 336-341. 2004. Wyndmoor, PA, USDA. Proceedings of the United States-Japan Cooperative Program in Natural Resources (UJNR). 12-12. (Conference Proceeding)
Musgrove, M. T., D. R. Jones, J. K. Northcutt, M. A. Harrison, and N. A. Cox. 2005. Impact of commercial processing on the microbiology of shell eggs. J. Food Prot. 2005. (In Press)
Northcutt, J. K., D. R. Jones, K. D. Ingram, A. Hinton, Jr., and M. T. Musgrove. 2004. Airborne microorganisms in commercial shell egg processing facilities. Int. J. Poult. Sci. 3:195-200.
Northcutt, J. K., M. T. Musgrove and D. R. Jones. 2005. Chemical analyses of commercial shell egg wash water. J. Appl. Poult Res. (accepted 12/8/04).
