Transfer rates of pathogenic bacteria during pork processing

To study cross-contamination, Escherichia coli (E. coli KCCM 11591), Staphylococcus aureus (S. aureus KCCM 11335), Listeria monocytogenes (L. monocytogenes KCCM 40307), and Salmonella enterica subsp. enterica (Sal. enteritidis, KCCM 12021) were used as representative model pathogens, as the levels of these strains are regulated by the law and are related to meat products [18]. All bacteria were harvested by centrifugation (2,000×g for 10 min), washed twice with sterile buffered peptone water (Difco, Detroit, MI, USA), and suspended in buffered peptone water (10⁶ colony-forming units [CFU]/mL), which was also used as the diluent in all experiments. E. coli isolates were prepared by overnight incubation (24 ± 2 h) in E. coli (EC) broth at 37°C. S. aureus and Sal. enteritidis were detected by inoculation in Baird Parker broth and XLD (Xylose Lysine Desoxycholate, Difco), respectively, followed by incubation overnight (18‒24 h) at 37°C. L. monocytogenes isolates were prepared by inoculation into PALCAM broth (Difco), followed by incubation overnight (24‒48 h) at 37°C.

Prior to the microbial experiments, sterile gloves (cotton and polyethylene) were placed in a sterile plastic bag until analysis. Meat samples (approximately 25 g) were immersed in 70% ethanol in a 10°C incubator for 5 min to remove the surface microorganisms, followed by sterilization under UV light for 1 h on a clean bench. The cutting board and knife were washed with 2 mL of antibacterial soap for 30 s, rinsed with distilled water for 15 s, dried with a paper towel, sprayed with 70% ethanol, and UV-sterilized for 1 h. These procedures served to standardize the initial levels of each pathogen and helped to avoid contaminated final bacterial counts [19].

To determine the initial concentration of bacteria on gloved hands, bacteria suspended (6 Log₁₀ CFU/g) in Erlenmeyer flasks were dropped onto one hand (1 mL) and rubbed with both hands. Each inoculated glove was left in a biological flow hood for 15 min to facilitate the attachment of bacteria onto the glove surface. The gloves were sampled by squeezing (Seward, London, UK) in 225 g buffered peptone water for 1 min and assayed as described in the culture conditions section. Both procedures were repeated three times; bacteria were counted and converted to a mean, which was used as the initial count on each surface. To determine the initial bacterial concentration in the meat samples, the bacterial suspension was inoculated onto the meat surface and allowed to dry (10⁶ CFU/mL). Each inoculated piece of meat was left in a biological flow hood for 15 min to facilitate the attachment of bacteria onto the meat surface. The meat was sampled by crushing in 225 g buffered peptone water for 1 min followed by analysis as described in the culture conditions section. Both procedures were repeated three times; the bacteria were counted and converted to a mean, which was used as the initial count on each surface.

The transfer rate was analyzed at meat processing plants. To evaluate the transfer of pathogens from contaminated meats to either polyethylene gloves or cotton gloved hands and utensils, the bacterial suspension (6 Log₁₀ CFU/g) was inoculated onto the meats and allowed to dry. The workers picked up the meats with both hands and cut the sample for 10 s using a knife and cutting board. Next, the workers’ polyethylene gloves or cotton gloves were sampled by crushing in 225 g buffered peptone water for 1 min, and the knife and cutting board surfaces were sampled with a PBS swab. The swabbed area included the area in contact with the meat samples (10 × 10 cm). Each swab was incubated in buffered peptone water for 5 min and mixed by vortexing for 1 min, followed by analysis as described in the culture conditions section. All experiments were repeated three times.

At the meat processing plants, the transfer rate of bacteria was analyzed between contaminated gloves, meat, and utensils. The bacterial suspension (6 Log₁₀ CFU/g) was inoculated onto the gloved hands of each worker and allowed to dry. The worker picked up the meat with both hands and cut the meat for 10 s using a knife and cutting board. At the end of the procedure, the meat was sampled by crushing in 225 g buffered peptone water for 1 min, and the knife and cutting board surface were sampled by a PBS swab (3M pipette swab, St. Paul, MN, USA). The swabbed area included the area in contact with the meat samples (10 × 10 cm). Each swab was incubated in buffered peptone water for 5 min and mixed by vortexing for 1 min. The assay was performed as described in the culture conditions section. All experiments were repeated three times.

Three samples for each surface and each treatment were evaluated, and a mean value was calculated. The transfer rate was determined as the ratio of the number of adherent bacteria to the initial bacterial count. The transfer rate (%) was calculated as (CFU on destination / CFU on source) × 100 [20,21]. Analysis of one factor with repeated measures and three replicates for each set of experimental parameters was conducted (SPSS 10.0, Chicago, IL, USA). Analysis of variance was used to determine the differences between sample means. Multiple comparisons among means were performed using the Duncan’s test, with p < 0.05 used as the significance level.

The transfer rate was determined as follows:

This experiment was conducted to determine the duration of use of gloves during general butchering. Pieces of pork tenderloin were purchased from a butcher, and the transported pork was stored in a refrigerator. The cutting board and knife were cleaned with antibacterial soap for 30 s, rinsed with distilled water for 15 s, and then wiped with a paper towel and sterilized. The gloves (cotton, polyethylene) were sprayed with 70% ethanol, and the gloves, knife, and cutting board were examined under UV light on a clean bench for 1 h. Experiments were conducted at 15°C to recreate the environment in the butcher’s shop, and seven experiments were conducted for up to 10 h. The “pre-experiment” included measuring the rate of contamination of the meat and utensils before cutting the pork, and “right after” includes measuring the rate of contamination in a sample of pork 5 min working after obtaining the pre-experiment microorganisms. Similarly, samples were analyzed after 1, 2, 4, 8, and 10 h after 5 min of working. Before each experiment, the gloves were sprayed twice with 70% ethanol and dried for 5 s. The meat was cut with a knife on the cutting board for approximately 1 min; the gloves, cutting board, and knife were sampled using the SWAB method (3M pipette swab). Next, 25 g of meat in 225 mL of sterilized physiological saline was crushed at 300 rpm for 15 min using a stomacher (Seward). The plate counts of total bacteria, E. coli, and Salmonella spp. were measured using PCA medium, EMB medium, and XLD medium, respectively, according to the guidelines provided in the Korean Food Standards Codex [22]. All experiments were repeated three times.

All workers at meat processing shops use cotton gloves, polyethylene gloves, or bare hands when handling meat. Particularly, cotton gloves help prevent the hands from slipping on raw meat. However, the consequences of wearing cotton gloves on hygiene are uncertain, as with few exceptions, published reports describe only the effects of wearing latex or polyethylene gloves during medical procedures or when serving food [23,24]. Meat processing shops have not established detailed regulations regarding the use of utensils and gloves. Therefore, we analyzed the cross-contamination of meat under various conditions. Table 1 summarizes the transfer rates from contaminated meat to cutting boards, knives, polyethylene gloves, and cotton gloves. The initial populations of E. coli, S. aureus, Sal. enterica subsp. enterica, and L. monocytogenes on raw meat were approximately 5–6 Log₁₀ CFU/g. After working with the contaminated meat, the population of pathogens on the cotton gloves (transfer rate 8.98%–15.79%) was monitored for E. coli, S. aureus, Sal. enterica, and L. monocytogenes, which showed values of 4.75, 4.48, 4.86, and 4.79 Log₁₀ CFU/g, respectively; these values on polyethylene gloves (transfer rate 0.24%–1.28%) were 3.07, 3.30, 3.97, and 3.55 Log₁₀ CFU/g. These observations indicate that bacteria are easily transferred from contaminated meats to gloves and utensils, particularly when cotton gloves are used. The relatively-high water absorption of cotton gloves may influence the transfer rate. Cotton is a hydrophilic surface and thus can promote the transfer of pathogens to high-moisture surfaces, such as pork meat.

On cotton gloves, after the workers cut the contaminated meat, the populations of E. coli, S. aureus, Sal. enterica, and L. monocytogenes associated with the cutting board (transfer rate 0.63%–0.97%) and knife (transfer rate 0.30%–1.91%) were 3.50, 3.07, 3.37, and 3.49 and 3.37, 3.29, 3.36, and 3.92 Log₁₀ CFU/g, respectively. When polyethylene gloves were used, after cutting the contaminated meat, the E. coli, S. aureus, Sal. enterica, and L. monocytogene populations retrieved from the cutting board (transfer rate 0.41%–0.76%) and knife (transfer rate 0.24%–0.83%) were 3.46, 3.12, 3.75, and 3.44 and 3.29, 3.33, 3.23, and 3.20 Log₁₀ CFU/g, respectively. For knives and cutting boards, the transfer rate was highest when processing involved cotton gloves rather than polyethylene gloves (p < 0.05). Polyethylene gloves do not absorb bacteria quickly, and thus do not continuously contaminate meat unlike cotton gloves. Our results are similar to those of Kim et al. [21] who reported that the transfer rates of L. monocytogenes from contaminated pork meat were greater when cotton gloves were worn than when bare hands and polyethylene gloves were used. Additionally, Montville and Schaffner [25] observed low cross-contamination rates when using polyethylene gloves, and, Gill and Jones [23] reported that thick rubber gloves prevented the transfer of E. coli.

The transfer rates from contaminated cotton gloves and contaminated polyethylene gloves to contact meat, and utensils are presented in Table 2. The initial population of pathogen on the glove was approximately 6 Log₁₀ CFU/g. For workers using contaminated cotton gloves, the populations of E. coli, S. aureus, Sal. enteritidis, and L. monocytogenes associated with the cutting board, knife, and meat were 3.50, 3.96, 3.46, and 3.75 Log₁₀ CFU/g (cutting board); 4.06, 3.81, 3.62, and 3.37 Log₁₀ CFU/g (knife); and 3.73, 4.00, 3.85, and 3.55 Log₁₀ CFU/g, respectively. Thus, the transfer rates were 0.07%–0.17% for cutting boards, 0.07%–0.54% for knives, and 0.10%–0.25% for meat. For workers using contaminated polyethylene gloves, E. coli, S. aureus, Sal. enteritidis, and L. monocytogenes associated with cutting board, knife and meat were 2.70, 2.47, 2.79, and 3.40 Log₁₀ CFU/g (cutting board, transfer rate 0.02%–0.06%); 3.33, 2.35, 2.01, and 3.98 Log₁₀ CFU/g (knife, transfer rate 0.01%–0.18%); and 2.77, 2.40, 2.83, and 2.93 Log₁₀ CFU/g (meat, transfer rate 0.01%–0.02%), respectively. Similar to the results shown in Table 1, these findings indicate that the transfer rate for cotton gloves was significantly greater than that for polyethylene gloves (p < 0.05). This is also related to the moisture-retaining ability of cotton gloves during work, whereas polyethylene gloves protect against cross contamination. Additionally, the rate of cross-contamination differed depending on the cutting board, knife, and bacteria. The highest transfer rate was observed for knives as compared to that for cutting boards and meat. Although the underlying reason is not clear, these results may depend on the workhouse environment [26]. Ravishankar et al. [20] reported that the bacterial transfer rate from contaminated chickens to cutting boards was 1.25% and from chickens to knives was 0.05%, reflecting the larger surface area of the cutting boards causing direct exposure. Understanding the factors that lead to contamination of meat products destined for consumption by humans is important for preventing the spread of pathogenic bacteria and foodborne illnesses [27]. As shown in Tables 1 and 2, the transfer rate of bacteria from contaminated gloves to meat or utensils was lower than that from contaminated meat to glove or utensils (p < 0.05). These results are similar to those of Jimenez et al. [28] who showed that the transfer rate of Salmonella spp. from gloves to green bell peppers was lower than that from green bell peppers to gloves. Thus, the moisture content of pork meat commodities may facilitate detachment of pathogens from the cutting boards and knives by more than from gloves.

Table 3 shows microbial contamination under general butchering conditions when wearing cotton gloves. During 10 h of work, bacteria colonies were detected on the meat (5.44 Log₁₀ CFU/g) and tools (3.37 Log₁₀ CFU/g), although E. coli and Salmonella spp. were not detected. However, the bacterial counts associated with all tools did not exceed the HACCP standard [16]. This result is similar to those of Hilton and Austin [29], who did not detect Salmonella spp. in experiments involving a dishcloth made of a material similar to that used in cotton gloves [21]. Therefore, we predict that safety and compliance with HACCP standards can be maintained if the workers are required to wear cotton gloves for 10 h per day, provided that 70% ethanol is sprayed on the gloves at the end of each round of cutting. The results of microbial contamination analysis while wearing polyethylene gloves are shown in Table 4. Bacteria were not detected after 10 h on the polyethylene gloves, cutting board, or knife. Plate count bacteria were detected in the meat in the range of 4.09‒5.07 Log₁₀ CFU/g; E. coli, S. aureus, Sal. spp., and L. monocytogens were not detected on either the tools or meat. All utensils was could be used for up to 10 h, as the bacterial counts were below the standards prescribed by the HACCP. Comparison of the results shown in Tables 3 and 4 revealed that contamination of utensils with general bacteria was significantly lower when polyethylene gloves were used rather than cotton gloves. The growth rate of microorganisms may be lower on latex gloves, as significant changes were not observed. This is because of differences in the material and the hydrophobic properties of cotton gloves and polyethylene gloves. Therefore, our results show that using polyethylene gloves rather than cotton gloves may effectively reduce cross-contamination during meat processing. Moore et al. [30] showed that the differences in material and the hydrophobic properties of cotton gloves and latex gloves affect microbial transfer associated with cross contamination. In contrast to our results, Robinson et al. [31] suggested that gloves cannot prevent high levels of microbial contamination; thus, it would be desirable to replace gloves frequently considering the condition of the meat and frequency of tool use [32]. In summary, our study provides important information regarding meat-plant hygiene and the transfer efficiency of pathogenic bacteria during meat processing. Further, our results may be helpful for related industries, as improving the processing of meat products will increase consumer confidence and decrease the incidence of food-related illnesses. Further studies are needed to determine the cross-contamination rate while considering more process-specific parameters than actual work processes about removal of microorganisms.