- Email: [email protected]
Efficacy of trimming chilled beef during fabrication to control Escherichia coli O157:H7 surrogates on subsequent subprimals
Meat Science 90 (2012) 420–425
Contents lists available at SciVerse ScienceDirect
Meat Science journal homepage: www.elsevier.com/locate/meatsci
Efficacy of trimming chilled beef during fabrication to control Escherichia coli O157: H7 surrogates on subsequent subprimals B.A. Laster, K.B. Harris ⁎, L.M. Lucia, A. Castillo, J.W. Savell Center for Food Safety, Meat Science Section, Department of Animal Science, Texas AgriLife Research, Texas A&M University, 2471 TAMU, College Station, TX 77843-2471, United States
a r t i c l e
i n f o
Article history: Received 15 March 2011 Received in revised form 31 May 2011 Accepted 26 August 2011 Keywords: Beef Escherichia coli O157:H7 Fabrication Subprimals Trimming
a b s t r a c t Effectiveness of trimming external carcass surfaces from subprimals during fabrication to reduce Escherichia coli O157:H7 surrogates was evaluated. Carcass sides (n = 10 sides) were inoculated along the hide pattern opening before entering the blast chill cooler with a gelatin slurry containing a bacterial cocktail of three rifampicin-resistant, nonpathogenic E. coli biotype I strains. Following a 48 h chill, sides were fabricated to produce eight subprimals. Microbiological samples were taken from the original carcass fat surface area, initial lean surface area, trimmed fat surface area (where applicable), and trimmed lean surface area (where applicable). Newly exposed lean surfaces had lower (P b 0.05) counts of rifampicin-resistant E. coli than did the external fat surfaces. However, fat and lean surfaces that were not inoculated became contaminated during the fabrication process. Trimming external surfaces reduced levels of pathogens, but under normal fabrication processes, pathogens were still spread to newly exposed surfaces. © 2011 Elsevier Ltd. All rights reserved.
1. Introduction During 2007 and 2008, the beef industry suffered from an increased number of positive Escherichia coli O157:H7 results and recalls (U.S. Department of Agriculture, 2010a, 2010b). One area of current concern relates to E. coli O157:H7 contamination of beef products intended for the production of non-intact products. Nonintact products include beef that has been enhanced by vacuum tumbling, mechanically tenderized by cubing, needle injected to incorporate a marinade, or subjected to a comminution process such as grinding, chopping, or mincing (U.S. Department of Agriculture, 1999). Pathogens may be introduced below the surface of these products as a result of these processes. Many further processors utilize purchase specifications that require the application of a validated microbial intervention to support the decision that E. coli O157:H7 is not reasonably likely to occur in food safety hazard. In addition, survey data collected in 2004 by Kennedy, Williams, Brown, and Minerich (2006) did not detect E. coli O157:H7 on the surfaces of subprimals. However, the United States Department of Agriculture's Food Safety and Inspection Service (USDA-FSIS) continues to question the ability of further processors to support their decisions that E. coli O157:H7 is not reasonably likely to occur in food safety hazard on the raw materials used to produce non-intact products.
⁎ Corresponding author. Tel.: + 1 979 862 3643; fax: + 1 979 862 3075. E-mail address: [email protected] (K.B. Harris). 0309-1740/$ – see front matter © 2011 Elsevier Ltd. All rights reserved. doi:10.1016/j.meatsci.2011.08.011
The beef industry has continued to search for ways to improve the safety of beef by reducing E. coli O157:H7 contamination. Studies have concluded that the primary sources of contamination during beef slaughter are fecal shedding and hides (Barkocy-Gallagher et al., 2003; Elder et al., 2000). The efficacy of hide washes (Baird, Lucia, Acuff, Harris, & Savell, 2006; Bosilevac, Nou, Osborn, Allen, & Koohmaraie, 2005), water washes (Castillo, Hardin, Acuff, & Dickson, 2002; Márquez-González, Harris, & Castillo, 2010), hot water washes (Castillo, Lucia, Goodson, Savell, & Acuff, 1998b; Graves Delmore, Sofos, Reagan, & Smith, 1997; Márquez-González et al., 2010), steam pasteurization (Gill & Bryant, 1997), and organic acid rinses (Castillo, Lucia, Goodson, Savell, & Acuff, 1998a; Castillo, Lucia, Mercado, & Acuff, 2001; Hardin, Acuff, Lucia, Oman, & Savell, 1995) to reduce pathogens has been reported. Previous studies have evaluated the efficacy of trimming (Castillo et al., 1998a; Hardin et al., 1995) and trimming combined with other interventions (Castillo et al., 1998a; Graves Delmore et al., 1997) in the decontamination of carcasses during harvest. Results showed that the removal of visible fecal contamination by trimming alone reduced bacterial counts (Castillo et al., 1998a; Graves Delmore et al., 1997), and when trimming was combined with treatments such as water wash, hot water wash, and organic acid sprays, reductions also were obtained (Castillo et al., 1998a; Graves Delmore et al., 1997). However, no research has evaluated the trimming of exterior carcass surfaces during normal fabrication processes on bacterial levels of subsequent subprimals. The purpose of this study was to determine if trimming during the fabrication process would reduce or eliminate the number of pathogens present on newly exposed lean and fat surfaces of subprimals.
B.A. Laster et al. / Meat Science 90 (2012) 420–425
To accomplish this purpose, a bacterial cocktail containing three surrogate microorganisms was applied to carcass surfaces during the harvest process to allow for a known quantity of microorganisms to be present before fabrication. Surrogate microorganisms are nonpathogenic microorganisms that grow, survive, and have resistant properties similar to specific pathogens (Marshall, Niebuhr, Acuff, Lucia, & Dickson, 2005). 2. Materials and methods 2.1. Bacterial cultures Three nonpathogenic E. coli biotype I strains previously isolated from cattle hides were obtained from the Food Microbiology Culture Collection (Texas A&M University, College Station, TX) as surrogates based on previous collaborative studies (Marshall et al., 2005). Procedures from Kaspar and Tamplin (1993) were followed to recover rifampicin-resistant (hereafter referred to as rif-resistant) strains from the total population. All strains were stored at − 80 °C, and stock working cultures were prepared on tryptic soy agar (TSA, BD Diagnostics, Sparks, MD). These three rif-resistant strains were designed to be utilized in a “cocktail” to represent possible contamination with enteric pathogens of fecal origin such as Salmonella or E. coli O157:H7. Growth characteristics were compared among the rif-resistant surrogates and their corresponding parent strains. Through previous research, these marker organisms have demonstrated identical thermal and acid resistance to E. coli O157:H7 (Cabrera-Diaz et al., 2009; Marshall et al., 2005; Niebuhr, Laury, Acuff, & Dickson, 2008). 2.2. Preparation of gelatin inoculum Each rif-resistant E. coli was cultured in 250 ml of tryptic soy broth (TSB, BD Diagnostic Systems) the day before inoculation and incubated at 35 °C for 18 ± 2 h. Following incubation, a bacterial cocktail was prepared by mixing equal volumes (200 ml) of each of the three cultures for a total of 600 ml, and the cocktail was used to inoculate the gelatin slurry. These microorganisms were inoculated in an opaque gelatin matrix that mimics fecal slurry. The gelatin mixture was prepared the day before inoculation to allow the mixture to cool to 25 °C. To prepare the gelatin mixture of 5.4 l, 42 g of food-grade unflavored gelatin (Kraft Food North America, Tarrytown, NY) was placed into 1 l of boiling sterile 0.1% peptone water (PW, BD Diagnostic Systems) and was allowed to dissolve for 5 min. This step was repeated to dissolve an additional 42 g of gelatin, and then both of the dissolved gelatins were combined with the remaining PW (3.4 l) and poured into two 4 l plastic beakers that were held for 18 ± 2 h at 25 °C. On each harvest day, the prepared gelatin slurry (5.4 l) was poured into a hand-held, non-corrosive, polyethylene compressed air sprayer (Ortho Heavy Duty Sprayer, The Fountain Group, Inc., New York Mills, NY) and capped. Immediately before spraying onto the carcasses, the cocktail (600 ml) was aseptically added to the sprayer containing the gelatin slurry. The sprayer was recapped and shaken, by hand, for 15 s to distribute the cocktail throughout the gelatin slurry. The average concentration of the gelatin slurry following the addition of the cocktail was ~7 log CFU/ml. 2.3. Harvesting and inoculation A total of five head of cattle were harvested at the Rosenthal Meat Science and Technology Center at Texas A&M University, College Station, TX, on two different dates (two head the first day and three head the second day). The beef slaughter process at this facility is an on-the-rail gravity-flow procedure (Savell & Smith, 2009), and standard slaughter procedures were performed up to carcass splitting.
421
Following carcass splitting, samples were obtained from the external surface of the carcass to determine whether rif-resistant E. coli were present before inoculation. Two 10-cm 2 samples were excised randomly using a sterile stainless steel borer, scalpel, and forceps and were composited (20-cm 2 total area) from three locations on the left side of each carcass. Sample locations included the round, rib/plate, and chuck areas. Before entering the blast chill cooler (− 2 to 0 °C), the normal hide pattern opening areas (e.g., brisket, plate, flank, round, and leg extremities) of both sides of each carcass were inoculated using the pump sprayer. The average flow rate of the gelatin inoculum was 23 s/carcass side for an average volume of 565 ml/carcass side. The flow of the slurry exiting the sprayer was calibrated before the first carcass inoculation for each harvest day. Carcass sides remained in the blast chill cooler (− 2 to 0 °C) for approximately 48 h. The average initial count of the carcass following inoculation was ~ 5 log CFU/cm 2.
2.4. Fabrication and microbiological sampling A refrigerated room equipped with two cutting tables with removable polyethylene cutting boards (one table used for forequarters and one table used for hindquarters), a band saw, hand saws, and knives, hooks, and steels were used to fabricate the carcasses. Fabrication followed general laboratory procedures outlined in Savell and Smith (2009) to produce eight subprimals: ribeye, brisket, shoulder clod, chuck roll, short loin, top sirloin, inside round, and bottom round. During the fabrication process, microbiological samples were taken from lean and fat surfaces categorized as either before trimming (BT) or after trimming (AT). The BT fat samples generally were those from exterior subprimal surfaces and the BT lean samples generally were surfaces that were revealed during the initial removal of the subprimal from the forequarters or hindquarters. The AT fat samples generally were those from the remaining fat surfaces that had been exposed to meet the final external fat trim specifications for that subprimal. The AT lean samples were those from the lean surfaces exposed during the final processing of the subprimal. Not all subprimals had the full complement of BT and AT fat and lean samples because of the fabrication method. The beef band (rib and plate section) was separated from the arm chuck section with a cut between the 5th and 6th ribs. The BT rib fat samples were taken from the exterior carcass fat surfaces over the M. longissimus thoracis. The BT lean samples for the rib were taken from the anterior exposed surfaces of the M. longissimus thoracis after the rib was separated from the chuck (6th rib interface). The rib was further processed in compliance with the Institutional Meat Purchase Specifications (IMPS) to make a lip-on ribeye 112A (North American Meat Processors Association, 2010) by removing all bones, lifter meat (M. latissimus dorsi, M. rhomboideus, M. trapezius, and M. subscapularis) with external surface fat, cartilage, and ligamentum nuchae. The AT fat samples for the rib were taken on the intermuscular fat surfaces on the dorsal aspect of the ribeye exposed after the lifter meat was removed, and the AT lean samples were taken on the ventral lean surfaces exposed after the back ribs were removed. The brisket was removed from the arm chuck by making a straight cut 2.5 cm from the end of the M. pectoralis profundus and at the cartilaginous juncture of the 1st rib and the sternum. The BT fat samples for the brisket were taken from the exterior carcass fat surfaces distal to the sternum. The BT lean samples were taken on the M. pectoralis profundus exposed after the sternum was removed because there were no other exposed lean areas that provided the appropriate sample size. The exterior carcass fat surface was trimmed to 0.6 cm, and the AT fat samples for the brisket were taken on the newly exposed fat surfaces. The AT lean samples were taken on the newly exposed M. pectoralis profundus after further trimming was performed to smooth the surface.
422
B.A. Laster et al. / Meat Science 90 (2012) 420–425
The foreshank of the arm chuck was suspended with a j-hook and trolley to allow for on-the-rail removal of the shoulder clod. The shoulder clod included the M. infraspinatus, M. triceps brachii, and M. teres major. The BT fat sample was taken from the exterior carcass fat surface where the M. cutaneus omobrachialis was present. The BT lean samples were taken on the interior lean surfaces of the M. infraspinatus, M. triceps brachii, and M. teres major muscles. Because the lean surface was the same before and after fabrication, an AT lean sample was not taken. Following the trimming of the exterior carcass fat surface to 0.6 cm, the AT fat sample was taken from the newly exposed fat surface. The remaining portion of the arm chuck section was placed back on the boning table for removal of the chuck roll. The BT fat samples for the chuck roll were taken on the external fat surface near the dorsal anterior portion of the chuck primal that was removed during production of the chuck roll. The BT lean sample for the chuck roll was taken on the M. longissimus thoracis face where the saw was used to separate the rib from the chuck (5th rib interface). The chuck roll was fabricated to comply with the specifications for the IMPS 116A (North American Meat Processors Association, 2010) by removing all cartilage, ligamentum nuchae, lymph glands, and bones. On the chuck roll, the external fat was removed during fabrication, so AT fat samples were not taken. The AT lean sample for the chuck roll was taken on the dorsal aspect on the newly exposed lean surface where the external fat was removed. The round was separated from the full loin between 4th and 5th sacral vertebrae and about 2.5 cm anterior to the knob of the aitch bone. The sirloin and short loin were separated between the 5th and 6th lumbar vertebrae and immediately anterior to the hip bone. The BT fat samples for the short loin were taken on the external fat surfaces over the M. longissimus lumborum. The BT lean samples for the short loin were taken from the M. longissimus lumborum faces on the sirloin ends where the top sirloin was separated from the short loin (6th lumbar interface). The external fat surfaces were trimmed to 0.6 cm, and the AT fat samples were taken on this newly exposed surface. Because the short loin remained an IMPS 174 (North American Meat Processors Association, 2010), the lean surface was not changed so an AT lean sample was not taken. The sirloin was separated through the natural seam to obtain a top sirloin and bottom sirloin. The BT fat samples for the top sirloin were taken from the exterior carcass fat surfaces. The remaining portions of the ilium and sacral bones were removed. The BT lean samples for the top sirloin were taken on the ventral lean surfaces exposed after the bones were removed. The external fat surface of the top sirloin was trimmed to 0.6 cm, and the AT fat sample was taken on the newly exposed fat surface. The lean surface did not change after the bones were removed resulting in an AT lean sample not being taken. The round was suspended with a j-hook and trolley for on-the-rail removal of the knuckle, inside round, and gooseneck. The inside round was separated from the bottom round and knuckle through the natural seam and was placed on the cutting table. The BT fat samples for the inside round were taken from the exterior carcass fat surfaces. The BT lean samples were taken on the ventral lean surface where the inside round was separated through the natural seam. Because lean surfaces remained the same before and after fabrication, AT lean samples were not taken from the ventral surface. Instead, the exposed medial portion of the M. semimembranosus that was dried during chilling was removed, and the newly exposed lean surface served as the AT lean sample. The exterior carcass surface of the inside round was trimmed to 0.6 cm, and the AT fat sample for the inside round was taken on the newly exposed fat surface. For the gooseneck (IMPS 170) (North American Meat Processors Association, 2010), the BT fat samples were taken from the exterior carcass fat surface. The BT lean samples were taken on the ventral lean surface of the M. semitendinosus, M. gluteobiceps, and M. gastrocnemius. The gooseneck was processed further by removing the heavy
connective tissue (epimysium), popliteal lymph gland, sarcosciatic ligament, M. semitendinosus, and M. gastrocnemius muscles to produce a bottom round flat (IMPS 171B) (North American Meat Processors Association, 2010). The AT fat sample for the bottom round flat was taken on the newly exposed external fat surface after trimming to 0.6 cm. The AT lean sample for the bottom round flat was taken on the lean surface exposed after the M. semitendinosus, M. gastrocnemius, heavy connective tissue (epimysium), popliteal lymph gland, and sarcosciatic ligament were removed. During fabrication of the carcass sides, worker's knives, hooks, and steels were sanitized in a chemical sanitizer (Biquat, Birko Corporation, Henderson, CO) every 5 min. Following the fabrication of the first side of each carcass, the table tops were flipped before starting fabrication of the second side to minimize contamination from one side to the next. The cutting lab was cleaned and sanitized between carcasses. To ensure that cleaning and sanitizing practices removed any residual surrogate microorganisms, environmental samples were taken using sponges on randomly selected places on the slaughter floor, blast chill cooler, and cutting lab. Before sampling, the sponge was moistened with 25 ml of sterile Butterfield's phosphate buffer (Hardy Diagnostics, Santa Maria, CA), and the sample collection was achieved by firmly rubbing the damp sponge over a selected area of 400 cm 2 with a pre-moistened sterile sponge using the procedure of 10 vertical and 10 horizontal passes. Microbiological samples from subprimals were obtained by excising two 10-cm 2 × 2 mm thick samples using a sterile stainless steel borer, scalpel, and forceps, and compositing them (20-cm 2 total area). Each composite sample was placed into a sterile stomacher bag, placed in an insulated container containing refrigerants, and transported to the Food Microbiology Laboratory. Upon arrival, 99 ml of sterile 0.1% peptone was added to each stomacher bag. The sample then was pummeled for 1 min using a Stomacher 400 (Tekmar Company, Cincinnati, OH). For each sample, counts of rifampicin-resistant E. coli surrogate microorganisms were determined by plating appropriate decimal dilutions on prepoured and dried rifampicin-tryptic soy agar (rif-TSA) plates. The environmental samples were hand massaged for 1 min and appropriate decimal dilutions were plated on rif-TSA plates. Rif-TSA was prepared by adding a solution of 0.1 g of rifampicin (Sigma-Aldrich Inc., St. Louis, MO) dissolved in 5 ml methanol (EM Science, Gibbstown, NJ) to 1 l of autoclaved and cooled (55 °C) TSA. Rif-TSA plates were incubated for 18 ± 2 h at 35 °C before counting and reporting the number of rifresistant E. coli biotype I per cm 2. 2.5. Statistical analysis Plate counts were converted to log CFU per ml or cm 2 before analysis. Data were analyzed using PROC GLM of SAS (SAS Institute, Inc., Table 1 Least squares means for trimming location effect on counts (log10 CFU/cm2) of rifampicinresistant E. coli for the ribeye. Trimming locationa
Log10 CFU/cm2
Before trimming, fat After trimming, fat Before trimming, lean After trimming, lean
2.9a 0.9bc b 0.7cb 1.1b
Means lacking a common letter (a–c) differ (P b 0.05). a The before trimming fat samples were taken on the external fat surface over the M. longissimus thoracis. The after trimming fat samples were taken on the intermuscular fat surface on the dorsal aspect of the M. longissimus thoracis exposed after the lifter meat and all external fat were removed. The before trimming lean samples were taken on the M. longissimus thoracis face where the rib was separated from the chuck (6th rib interface). The after trimming lean sample was taken on the ventral M. longissimus thoracis exposed after the back ribs were removed. b Value denotes samples below the minimum detection level of 0.7 log CFU/cm2.
B.A. Laster et al. / Meat Science 90 (2012) 420–425 Table 2 Least squares means for trimming location effect on counts (log10 CFU/cm2) of rifampicinresistant E. coli for the brisket. Trimming location
a
Before trimming, fat After trimming, fat Before trimming, lean After trimming, lean
423
Table 4 Least squares means for trimming location effect on counts (log10 CFU/cm2) of rifampicinresistant E. coli for the chuck roll.
Log10 CFU/cm2
Trimming locationa
Log10 CFU/cm2
3.8a 1.8b 1.2b b 0.7cb
Before trimming, fat Before trimming, lean After trimming, lean
3.3a 0.9b 1.4b
Means lacking a common letter (a–c) differ (P b 0.05). a The before trimming fat samples were taken on the exterior carcass fat surface distal to the sternum. The after trimming fat samples were taken on the newly exposed surface after trimming to an external level of 0.6 cm. The before trimming lean samples were taken on the M. pectoralis profundus exposed after the sternum was removed. The after trimming lean samples were taken on the newly exposed M. pectoralis profundus after further trimming was performed to smooth the surface. b Value denotes samples below the minimum detection level of 0.7 log CFU/cm2.
Cary, NC). Least squares means were generated for main effects and separated using PDIFF option when appropriate with an alpha-level (P b 0.05). 3. Results and discussion This study used rif-resistant surrogate microorganisms that would not typically be found in a commercial beef-processing establishment. No detectable counts of rif-resistant microorganisms were found on the non-inoculated sides (data not shown). Environmental samples taken to ensure that the rif-resistant microorganisms were removed from the facility were below detectable levels (data not shown). There were no differences (P N 0.05) in counts (log10 CFU/cm 2) of initial carcass inoculum levels of rif-resistant microorganisms between harvest Day 1 (5.1) and Day 2 (4.9) and among the brisket (5.1), plate (5.0) and round (5.0). Although we would anticipate inconsistent or sporadic contamination in normal beef harvesting processes, we designed this study to create a consistent level of contamination to test the impact of trimming during the production of subprimals in the fabrication process. The before and after trimming effects for the lean and fat surfaces for each of the subprimals evaluated are presented in Tables 1 through 8. Similar trends were found for each of these subprimals. In six of the seven subprimals where there were direct comparisons between before and after trimming for fat, the after trimming fat surfaces had lower (P b 0.05) counts of rif-resistant E. coli than did the before trimming surfaces (the bottom round did not respond this way). For the most part, trimming external fat surfaces to produce these closely trimmed subprimals resulted in significantly lower pathogens on the resultant surfaces than if no trimming had occurred. For all eight subprimals, the before trimming lean surfaces had lower (P b 0.05) counts of rif-resistant E. coli than did the before trimming fat surfaces, which further implies that significantly fewer pathogens
Table 3 Least squares means for trimming location effect on counts (log10 CFU/cm2) of rifampicinresistant E. coli for the shoulder clod. Trimming locationa
Log10 CFU/cm2
Before trimming, fat After trimming, fat Before trimming, lean
3.7a 0.8b 1.1b
Means lacking a common letter (a and b) differ (P b 0.05). a The before trimming fat samples were taken on the external fat surface where the M. cutaneous omobrachialis was present. The after trimming fat samples were taken on the newly exposed surface after trimming to an external level of 0.6 cm. The before trimming lean samples were taken on the interior lean surface of the M. infraspinatus and M. triceps brachii. There was no after trimming lean sample taken for this subprimal.
Means lacking a common letter (a and b) differ (P b 0.05). a The before trimming fat samples were taken on the exterior carcass fat surface near the dorsal anterior portion of the chuck primal. There were no after trimming fat surfaces for the chuck roll. The before trimming lean samples were taken from the M. longissimus thoracis face where the rib was separated from the chuck (5th rib interface). The after trimming lean samples were taken on the dorsal aspect on the newly exposed surface where the external fat was removed.
will be on the initial lean surfaces when these subprimals are prepared than on the outside fat surfaces of the cut. However, in the limited direct comparisons of rif-resistant E. coli counts between the before and after trimming comparisons for the lean surfaces, in one subprimal they decreased (brisket, Table 2), in one they increased (ribeye, Table 1), and the remainder they did not differ between the two lean surfaces. At this point, the lean surfaces of the subprimals were minimally altered compared to the fat surfaces, which were reduced or removed in the preparation of the subprimals. Although there are a number of studies that report the efficacy of trimming hot carcasses as a decontamination method (Castillo et al., 1998a; Hardin et al., 1995; Prasai et al., 1995; Reagan et al., 1996), there are limited data regarding the impact of trimming chilled carcasses or cuts to reduce pathogens. Lemmons et al. (2011) found that trimming top sirloin butts that had been inoculated with high or low levels of E. coli O157:H7 was an effective method of decontaminating these beef subprimals. Lemmons et al. (2011) support the U.S. Department of Agriculture (2009) recommendations that trimming is a measure that establishments may use to address E. coli O157:H7; however, our study showed that trimming fat surfaces was more efficacious at reducing pathogens than trimming lean surfaces. Although there were variations in the microbiological counts from the impact of trimming exterior carcass surfaces during fabrication, this project showed that there was a general trend that counts from trimmed surfaces were lower than initially exposed surfaces. However, fat and lean surfaces that were not inoculated became contaminated during the fabrication process, which indicates that trimming external surfaces results in reduced, but not completely eliminated, pathogens on the finished products. The beef industry continues to be criticized by USDA-FSIS and consumer groups for not preventing illnesses associated with E. coli O157:H7. To address the concern of E. coli O157:H7, the industry has focused primarily on harvest interventions such as hot water, steam, and organic acid rinses to reduce pathogen contamination because the exterior surfaces may be contaminated with pathogens. Table 5 Least squares means for trimming location effect on counts (log10 CFU/cm2) of rifampicinresistant E. coli for the short loin. Trimming locationa
Log10 CFU/cm2
Before trimming, fat After trimming, fat Before trimming, lean
2.3a b 0.7bb 0.9b
Means lacking a common letter (a–c) differ (P b 0.05). a The before trimming fat samples were taken on the external fat surface over the M. longissimus lumborum. The after trimmed fat samples were taken on the newly exposed surface after trimming to an external level of 0.6 cm. The before lean samples were taken from the M. longissimus lumborum face on the sirloin end where the short loin was separated from the sirloin (6th lumbar interface). There were no after trimming lean surfaces for the short loin. b Value denotes samples below the minimum detection level of 0.7 log CFU/cm2.
424
B.A. Laster et al. / Meat Science 90 (2012) 420–425
Table 6 Least squares means for trimming location effect on counts (log10 CFU/cm2) of rifampicinresistant E. coli for the top sirloin. Trimming locationa
Log10 CFU/cm2
Before trimming, fat After trimming, fat Before trimming, lean
2.9a 1.2b 1.7b
Means lacking a common letter (a and b) differ (P b 0.05). a The before trimming fat samples were taken on the exterior carcass surface of this cut. The after trimming fat samples were taken on the newly exposed surface after trimming to an external level of 0.6 cm. The before trimming lean samples were taken on the ventral lean surface exposed after the ilium and sacral bones were removed. There was no after trimming lean surface for the top sirloin.
Table 7 Least squares means for trimming location effect on counts (log10 CFU/cm2) of rifampicinresistant E. coli for the inside round. Trimming locationa
Log10 CFU/cm2
Before trimming, fat After trimming, fat Before trimming, lean After trimming, lean
3.5a 2.3b 0.9c b 0.7cb
Means lacking a common letter (a–c) differ (P b 0.05). a The before trimming fat samples were taken on the exterior carcass surface of this cut. The after trimming fat samples were taken on the newly exposed surface after trimming to an external level of 0.6 cm. The before trimming lean samples were taken on the ventral lean surface of where the inside round was separated from the bottom round and knuckle. The after trimming lean samples were taken on the exposed medial portion of the M. semimembranosus that was dried during chilling and was trimmed away to expose a new surface for sampling. b Value denotes samples below the minimum detection level of 0.7 log CFU/cm2.
Research has shown that the prevalence of E. coli O157:H7 is not likely on beef subprimals (Heller et al., 2007; Kennedy et al., 2006). Our results support that trimming of external fat surfaces during the normal fabrication process may reduce contamination of E. coli O157:H7. However, we did show that existing pathogens may spread to newly exposed surfaces. This information will 'help the industry better understand how typical processing practices impact the safety of the products they produce. Acknowledgements This project was funded, in part, by The Beef Checkoff, The E. M. “Manny” Rosenthal Chair in Animal Science, and the Manny and Rosalyn Rosenthal Endowed Fund in the Department of Animal Science.
Table 8 Least squares means for trimming location effect on counts (log10 CFU/cm2) of rifampicinresistant E. coli for the bottom round. Trimming locationa
Log10 CFU/cm2
Before trimming, fat After trimming, fat Before trimming, lean After trimming, lean
2.6a 2.2a 1.1b 0.8b
Means lacking a common letter (a–c) differ (P b 0.05). a The before trimming fat samples were taken on the exterior carcass fat surface. The after trimming fat samples were taken on the newly exposed surface after trimming to an external level of 0.6 cm. The before trimming lean samples were taken on the ventral lean surface of the M. semitendinosus, M. gluteobiceps, and M. gastrocnemius. The after trimming lean samples were taken on the surface exposed after the heavy connective tissue (epimysium), popliteal lymph gland, sarcosciatic ligament, M. gastrocnemius, and M. semitendinosus were removed.
References Baird, B. E., Lucia, L. M., Acuff, G. R., Harris, K. B., & Savell, J. W. (2006). Beef hide antimicrobial interventions as a means of reducing bacterial contamination. Meat Science, 73(2), 245–248. Barkocy-Gallagher, G. A., Arthur, T. M., Rivera-Betancourt, M., Nou, X., Shackelford, S. D., Wheeler, T. L., et al. (2003). Seasonal prevalence of shiga toxinproducing Escherichia coli, including O157:H7 and non-O157 serotypes, and Salmonella in commercial beef processing plants. Journal of Food Protection, 66(11), 1978–1986. Bosilevac, J. M., Nou, X., Osborn, M. S., Allen, D. M., & Koohmaraie, M. (2005). Development and evaluation of an on-line hide decontamination procedure for use in a commercial beef processing plant. Journal of Food Protection, 68(2), 265–272. Cabrera-Diaz, E., Moseley, T. M., Lucia, L. M., Dickson, J. S., Castillo, A., & Acuff, G. R. (2009). Fluorescent protein-marked Escherichia coli biotype I strains as surrogates for enteric pathogens in validation of beef carcass interventions. Journal of Food Protection, 72(2), 295–303. Castillo, A., Hardin, M. D., Acuff, G. R., & Dickson, J. S. (2002). Reduction of microbial contaminants on carcasses. In V. K. Juneja, & J. N. Sofos (Eds.), Control of foodborne microorganisms (pp. 351–381). New York: Marcel Dekker, Inc.. Castillo, A., Lucia, L. M., Goodson, K. J., Savell, J. W., & Acuff, G. R. (1998a). Comparison of water wash, trimming, and combined hot water and lactic acid treatments for reducing bacteria of fecal origin on beef carcasses. Journal of Food Protection, 61(7), 823–828. Castillo, A., Lucia, L. M., Goodson, K. J., Savell, J. W., & Acuff, G. R. (1998b). Use of hot water for beef carcass decontamination. Journal of Food Protection, 61(1), 19–25. Castillo, A., Lucia, L. M., Mercado, I., & Acuff, G. R. (2001). In-plant evaluation of a lactic acid treatment for reduction of bacteria on chilled beef carcasses. Journal of Food Protection, 64(5), 738–740. Elder, R. O., Keen, J. E., Siragusa, G. R., Barkocy-Gallagher, G. A., Koohmaraie, M., & Laegreid, W. W. (2000). Correlation of enterohemorrhagic Escherichia coli O157 prevalence in feces, hides, and carcasses of beef cattle during processing. Proceedings of the National Academy of Sciences of the United States of America, 97(7), 2999–3003. Gill, C. O., & Bryant, J. (1997). Decontamination of carcasses by vacuum-hot water cleaning and steam pasteurizing during routine operations at a beef packing plant. Meat Science, 47(3–4), 267–276. Graves Delmore, L. R., Sofos, J. N., Reagan, J. O., & Smith, G. C. (1997). Hot-water rinsing and trimming/washing of beef carcasses to reduce physical and microbiological contamination. Journal of Food Science, 62(2), 373–376. Hardin, M. D., Acuff, G. R., Lucia, L. M., Oman, J. S., & Savell, J. W. (1995). Comparison of methods for decontamination from beef carcass surfaces. Journal of Food Protection, 58(4), 368–374. Heller, C. E., Scanga, J. A., Sofos, J. N., Belk, K. E., Warren-Serna, W., Bellinger, G. R., et al. (2007). Decontamination of beef subprimal cuts intended for blade tenderization or moisture enhancement. Journal of Food Protection, 70(5), 1174–1180. Kaspar, C. W., & Tamplin, M. L. (1993). Effects of temperature and salinity on the survival of Vibrio vulnificus in seawater and shellfish. Applied and Environmental Microbiology, 59(8), 2425–2429. Kennedy, J. E., Williams, S. K., Brown, T., & Minerich, P. (2006). Prevalence of Escherichia coli O157:H7 and indicator organisms on the surface of intact subprimal beef cuts prior to further processing. Journal of Food Protection, 69(7), 1514–1517. Lemmons, J. L., Lucia, L. M., Hardin, M. D., Savell, J. W., & Harris, K. B. (2011). Evaluation of Escherichia coli O157:H7 translocation and decontamination for beef vacuumpackaged subprimals destined for non-intact use. Journal of Food Protection, 74(7), 1048–1053. Márquez-González, M., Harris, K. B., & Castillo, A. (2010). Interventions for hazard control in foods during harvesting. In V. K. Juneja, & J. N. Sofos (Eds.), Pathogens and toxins in foods: Challenges and interventions (pp. 379–395). Washington, DC: ASM Press. Marshall, K. M., Niebuhr, S. E., Acuff, G. R., Lucia, L. M., & Dickson, J. S. (2005). Identification of Escherichia coli O157:H7 meat processing indicators for fresh meat through comparison of the effects of selected antimicrobial interventions. Journal of Food Protection, 68(12), 2580–2586. Niebuhr, S. E., Laury, A., Acuff, G. R., & Dickson, J. S. (2008). Evaluation of nonpathogenic surrogate bacteria as process validation indicators for Salmonella enterica for selected antimicrobial treatments, cold storage, and fermentation in meat. Journal of Food Protection, 71(4), 714–718. North American Meat Processors Association (2010). The meat buyer's guide™ (6th ed.). Reston, VA: North American Meat Processors Association. Prasai, R. K., Phebus, R. K., Garcia Zepeda, C. M., Kastner, C. L., Boyle, A. E., & Fung, D. Y. C. (1995). Effectiveness of trimming and/or washing on microbiological quality of beef carcasses. Journal of Food Protection, 58, 1114–1117. Reagan, J. O., Acuff, G. R., Buege, D. R., Buyck, M. J., Dickson, J. S., Kastner, C. L., et al. (1996). Trimming and washing of beef carcasses as a method of improving the microbiological quality of meat. Journal of Food Protection, 59(7), 751–756. Savell, J. W., & Smith, G. C. (2009). Meat science laboratory manual (8th ed.). Boston, MA: American Press. U.S. Department of Agriculture (1999). FSIS policy on non-intact raw beef products contaminated with E. coli O157:H7. (Available from). Washington, DC: United States Department of Agriculture, Food Safety and Inspection Service. http://www.fsis.usda.gov/oa/background/O157policy.htm (Accessed 16 July 2010).
B.A. Laster et al. / Meat Science 90 (2012) 420–425 U.S. Department of Agriculture (2009). Measures to address E. coli O157:H7 at establishment that receive, grind, or otherwise process raw beef products. FSIS Notice 05–09. Washington, DC: United States Department of Agriculture, Food Safety and Inspection Service. U.S. Department of Agriculture (2010a). FSIS recalls, recall case archive. (Available from). Washington, DC: United States Department of Agriculture, Food Safety and Inspection Service. http://www.fsis.usda.gov/fsis_recalls/Recall_Case_Archive/index.asp (Accessed 25 September 2010).
425
U.S. Department of Agriculture (2010b). Microbiological results of raw ground beef products analyzed for Escherichia coli O157:H7. (Available from). Washington, DC: United States Department of Agriculture, Food Safety and Inspection Service. http:// www.fsis.usda.gov/Science/ground_beef_e.Coli_Testing_results/index.asp (Accessed 2 October 2010).
Contents lists available at SciVerse ScienceDirect
Meat Science journal homepage: www.elsevier.com/locate/meatsci
Efficacy of trimming chilled beef during fabrication to control Escherichia coli O157: H7 surrogates on subsequent subprimals B.A. Laster, K.B. Harris ⁎, L.M. Lucia, A. Castillo, J.W. Savell Center for Food Safety, Meat Science Section, Department of Animal Science, Texas AgriLife Research, Texas A&M University, 2471 TAMU, College Station, TX 77843-2471, United States
a r t i c l e
i n f o
Article history: Received 15 March 2011 Received in revised form 31 May 2011 Accepted 26 August 2011 Keywords: Beef Escherichia coli O157:H7 Fabrication Subprimals Trimming
a b s t r a c t Effectiveness of trimming external carcass surfaces from subprimals during fabrication to reduce Escherichia coli O157:H7 surrogates was evaluated. Carcass sides (n = 10 sides) were inoculated along the hide pattern opening before entering the blast chill cooler with a gelatin slurry containing a bacterial cocktail of three rifampicin-resistant, nonpathogenic E. coli biotype I strains. Following a 48 h chill, sides were fabricated to produce eight subprimals. Microbiological samples were taken from the original carcass fat surface area, initial lean surface area, trimmed fat surface area (where applicable), and trimmed lean surface area (where applicable). Newly exposed lean surfaces had lower (P b 0.05) counts of rifampicin-resistant E. coli than did the external fat surfaces. However, fat and lean surfaces that were not inoculated became contaminated during the fabrication process. Trimming external surfaces reduced levels of pathogens, but under normal fabrication processes, pathogens were still spread to newly exposed surfaces. © 2011 Elsevier Ltd. All rights reserved.
1. Introduction During 2007 and 2008, the beef industry suffered from an increased number of positive Escherichia coli O157:H7 results and recalls (U.S. Department of Agriculture, 2010a, 2010b). One area of current concern relates to E. coli O157:H7 contamination of beef products intended for the production of non-intact products. Nonintact products include beef that has been enhanced by vacuum tumbling, mechanically tenderized by cubing, needle injected to incorporate a marinade, or subjected to a comminution process such as grinding, chopping, or mincing (U.S. Department of Agriculture, 1999). Pathogens may be introduced below the surface of these products as a result of these processes. Many further processors utilize purchase specifications that require the application of a validated microbial intervention to support the decision that E. coli O157:H7 is not reasonably likely to occur in food safety hazard. In addition, survey data collected in 2004 by Kennedy, Williams, Brown, and Minerich (2006) did not detect E. coli O157:H7 on the surfaces of subprimals. However, the United States Department of Agriculture's Food Safety and Inspection Service (USDA-FSIS) continues to question the ability of further processors to support their decisions that E. coli O157:H7 is not reasonably likely to occur in food safety hazard on the raw materials used to produce non-intact products.
⁎ Corresponding author. Tel.: + 1 979 862 3643; fax: + 1 979 862 3075. E-mail address: [email protected] (K.B. Harris). 0309-1740/$ – see front matter © 2011 Elsevier Ltd. All rights reserved. doi:10.1016/j.meatsci.2011.08.011
The beef industry has continued to search for ways to improve the safety of beef by reducing E. coli O157:H7 contamination. Studies have concluded that the primary sources of contamination during beef slaughter are fecal shedding and hides (Barkocy-Gallagher et al., 2003; Elder et al., 2000). The efficacy of hide washes (Baird, Lucia, Acuff, Harris, & Savell, 2006; Bosilevac, Nou, Osborn, Allen, & Koohmaraie, 2005), water washes (Castillo, Hardin, Acuff, & Dickson, 2002; Márquez-González, Harris, & Castillo, 2010), hot water washes (Castillo, Lucia, Goodson, Savell, & Acuff, 1998b; Graves Delmore, Sofos, Reagan, & Smith, 1997; Márquez-González et al., 2010), steam pasteurization (Gill & Bryant, 1997), and organic acid rinses (Castillo, Lucia, Goodson, Savell, & Acuff, 1998a; Castillo, Lucia, Mercado, & Acuff, 2001; Hardin, Acuff, Lucia, Oman, & Savell, 1995) to reduce pathogens has been reported. Previous studies have evaluated the efficacy of trimming (Castillo et al., 1998a; Hardin et al., 1995) and trimming combined with other interventions (Castillo et al., 1998a; Graves Delmore et al., 1997) in the decontamination of carcasses during harvest. Results showed that the removal of visible fecal contamination by trimming alone reduced bacterial counts (Castillo et al., 1998a; Graves Delmore et al., 1997), and when trimming was combined with treatments such as water wash, hot water wash, and organic acid sprays, reductions also were obtained (Castillo et al., 1998a; Graves Delmore et al., 1997). However, no research has evaluated the trimming of exterior carcass surfaces during normal fabrication processes on bacterial levels of subsequent subprimals. The purpose of this study was to determine if trimming during the fabrication process would reduce or eliminate the number of pathogens present on newly exposed lean and fat surfaces of subprimals.
B.A. Laster et al. / Meat Science 90 (2012) 420–425
To accomplish this purpose, a bacterial cocktail containing three surrogate microorganisms was applied to carcass surfaces during the harvest process to allow for a known quantity of microorganisms to be present before fabrication. Surrogate microorganisms are nonpathogenic microorganisms that grow, survive, and have resistant properties similar to specific pathogens (Marshall, Niebuhr, Acuff, Lucia, & Dickson, 2005). 2. Materials and methods 2.1. Bacterial cultures Three nonpathogenic E. coli biotype I strains previously isolated from cattle hides were obtained from the Food Microbiology Culture Collection (Texas A&M University, College Station, TX) as surrogates based on previous collaborative studies (Marshall et al., 2005). Procedures from Kaspar and Tamplin (1993) were followed to recover rifampicin-resistant (hereafter referred to as rif-resistant) strains from the total population. All strains were stored at − 80 °C, and stock working cultures were prepared on tryptic soy agar (TSA, BD Diagnostics, Sparks, MD). These three rif-resistant strains were designed to be utilized in a “cocktail” to represent possible contamination with enteric pathogens of fecal origin such as Salmonella or E. coli O157:H7. Growth characteristics were compared among the rif-resistant surrogates and their corresponding parent strains. Through previous research, these marker organisms have demonstrated identical thermal and acid resistance to E. coli O157:H7 (Cabrera-Diaz et al., 2009; Marshall et al., 2005; Niebuhr, Laury, Acuff, & Dickson, 2008). 2.2. Preparation of gelatin inoculum Each rif-resistant E. coli was cultured in 250 ml of tryptic soy broth (TSB, BD Diagnostic Systems) the day before inoculation and incubated at 35 °C for 18 ± 2 h. Following incubation, a bacterial cocktail was prepared by mixing equal volumes (200 ml) of each of the three cultures for a total of 600 ml, and the cocktail was used to inoculate the gelatin slurry. These microorganisms were inoculated in an opaque gelatin matrix that mimics fecal slurry. The gelatin mixture was prepared the day before inoculation to allow the mixture to cool to 25 °C. To prepare the gelatin mixture of 5.4 l, 42 g of food-grade unflavored gelatin (Kraft Food North America, Tarrytown, NY) was placed into 1 l of boiling sterile 0.1% peptone water (PW, BD Diagnostic Systems) and was allowed to dissolve for 5 min. This step was repeated to dissolve an additional 42 g of gelatin, and then both of the dissolved gelatins were combined with the remaining PW (3.4 l) and poured into two 4 l plastic beakers that were held for 18 ± 2 h at 25 °C. On each harvest day, the prepared gelatin slurry (5.4 l) was poured into a hand-held, non-corrosive, polyethylene compressed air sprayer (Ortho Heavy Duty Sprayer, The Fountain Group, Inc., New York Mills, NY) and capped. Immediately before spraying onto the carcasses, the cocktail (600 ml) was aseptically added to the sprayer containing the gelatin slurry. The sprayer was recapped and shaken, by hand, for 15 s to distribute the cocktail throughout the gelatin slurry. The average concentration of the gelatin slurry following the addition of the cocktail was ~7 log CFU/ml. 2.3. Harvesting and inoculation A total of five head of cattle were harvested at the Rosenthal Meat Science and Technology Center at Texas A&M University, College Station, TX, on two different dates (two head the first day and three head the second day). The beef slaughter process at this facility is an on-the-rail gravity-flow procedure (Savell & Smith, 2009), and standard slaughter procedures were performed up to carcass splitting.
421
Following carcass splitting, samples were obtained from the external surface of the carcass to determine whether rif-resistant E. coli were present before inoculation. Two 10-cm 2 samples were excised randomly using a sterile stainless steel borer, scalpel, and forceps and were composited (20-cm 2 total area) from three locations on the left side of each carcass. Sample locations included the round, rib/plate, and chuck areas. Before entering the blast chill cooler (− 2 to 0 °C), the normal hide pattern opening areas (e.g., brisket, plate, flank, round, and leg extremities) of both sides of each carcass were inoculated using the pump sprayer. The average flow rate of the gelatin inoculum was 23 s/carcass side for an average volume of 565 ml/carcass side. The flow of the slurry exiting the sprayer was calibrated before the first carcass inoculation for each harvest day. Carcass sides remained in the blast chill cooler (− 2 to 0 °C) for approximately 48 h. The average initial count of the carcass following inoculation was ~ 5 log CFU/cm 2.
2.4. Fabrication and microbiological sampling A refrigerated room equipped with two cutting tables with removable polyethylene cutting boards (one table used for forequarters and one table used for hindquarters), a band saw, hand saws, and knives, hooks, and steels were used to fabricate the carcasses. Fabrication followed general laboratory procedures outlined in Savell and Smith (2009) to produce eight subprimals: ribeye, brisket, shoulder clod, chuck roll, short loin, top sirloin, inside round, and bottom round. During the fabrication process, microbiological samples were taken from lean and fat surfaces categorized as either before trimming (BT) or after trimming (AT). The BT fat samples generally were those from exterior subprimal surfaces and the BT lean samples generally were surfaces that were revealed during the initial removal of the subprimal from the forequarters or hindquarters. The AT fat samples generally were those from the remaining fat surfaces that had been exposed to meet the final external fat trim specifications for that subprimal. The AT lean samples were those from the lean surfaces exposed during the final processing of the subprimal. Not all subprimals had the full complement of BT and AT fat and lean samples because of the fabrication method. The beef band (rib and plate section) was separated from the arm chuck section with a cut between the 5th and 6th ribs. The BT rib fat samples were taken from the exterior carcass fat surfaces over the M. longissimus thoracis. The BT lean samples for the rib were taken from the anterior exposed surfaces of the M. longissimus thoracis after the rib was separated from the chuck (6th rib interface). The rib was further processed in compliance with the Institutional Meat Purchase Specifications (IMPS) to make a lip-on ribeye 112A (North American Meat Processors Association, 2010) by removing all bones, lifter meat (M. latissimus dorsi, M. rhomboideus, M. trapezius, and M. subscapularis) with external surface fat, cartilage, and ligamentum nuchae. The AT fat samples for the rib were taken on the intermuscular fat surfaces on the dorsal aspect of the ribeye exposed after the lifter meat was removed, and the AT lean samples were taken on the ventral lean surfaces exposed after the back ribs were removed. The brisket was removed from the arm chuck by making a straight cut 2.5 cm from the end of the M. pectoralis profundus and at the cartilaginous juncture of the 1st rib and the sternum. The BT fat samples for the brisket were taken from the exterior carcass fat surfaces distal to the sternum. The BT lean samples were taken on the M. pectoralis profundus exposed after the sternum was removed because there were no other exposed lean areas that provided the appropriate sample size. The exterior carcass fat surface was trimmed to 0.6 cm, and the AT fat samples for the brisket were taken on the newly exposed fat surfaces. The AT lean samples were taken on the newly exposed M. pectoralis profundus after further trimming was performed to smooth the surface.
422
B.A. Laster et al. / Meat Science 90 (2012) 420–425
The foreshank of the arm chuck was suspended with a j-hook and trolley to allow for on-the-rail removal of the shoulder clod. The shoulder clod included the M. infraspinatus, M. triceps brachii, and M. teres major. The BT fat sample was taken from the exterior carcass fat surface where the M. cutaneus omobrachialis was present. The BT lean samples were taken on the interior lean surfaces of the M. infraspinatus, M. triceps brachii, and M. teres major muscles. Because the lean surface was the same before and after fabrication, an AT lean sample was not taken. Following the trimming of the exterior carcass fat surface to 0.6 cm, the AT fat sample was taken from the newly exposed fat surface. The remaining portion of the arm chuck section was placed back on the boning table for removal of the chuck roll. The BT fat samples for the chuck roll were taken on the external fat surface near the dorsal anterior portion of the chuck primal that was removed during production of the chuck roll. The BT lean sample for the chuck roll was taken on the M. longissimus thoracis face where the saw was used to separate the rib from the chuck (5th rib interface). The chuck roll was fabricated to comply with the specifications for the IMPS 116A (North American Meat Processors Association, 2010) by removing all cartilage, ligamentum nuchae, lymph glands, and bones. On the chuck roll, the external fat was removed during fabrication, so AT fat samples were not taken. The AT lean sample for the chuck roll was taken on the dorsal aspect on the newly exposed lean surface where the external fat was removed. The round was separated from the full loin between 4th and 5th sacral vertebrae and about 2.5 cm anterior to the knob of the aitch bone. The sirloin and short loin were separated between the 5th and 6th lumbar vertebrae and immediately anterior to the hip bone. The BT fat samples for the short loin were taken on the external fat surfaces over the M. longissimus lumborum. The BT lean samples for the short loin were taken from the M. longissimus lumborum faces on the sirloin ends where the top sirloin was separated from the short loin (6th lumbar interface). The external fat surfaces were trimmed to 0.6 cm, and the AT fat samples were taken on this newly exposed surface. Because the short loin remained an IMPS 174 (North American Meat Processors Association, 2010), the lean surface was not changed so an AT lean sample was not taken. The sirloin was separated through the natural seam to obtain a top sirloin and bottom sirloin. The BT fat samples for the top sirloin were taken from the exterior carcass fat surfaces. The remaining portions of the ilium and sacral bones were removed. The BT lean samples for the top sirloin were taken on the ventral lean surfaces exposed after the bones were removed. The external fat surface of the top sirloin was trimmed to 0.6 cm, and the AT fat sample was taken on the newly exposed fat surface. The lean surface did not change after the bones were removed resulting in an AT lean sample not being taken. The round was suspended with a j-hook and trolley for on-the-rail removal of the knuckle, inside round, and gooseneck. The inside round was separated from the bottom round and knuckle through the natural seam and was placed on the cutting table. The BT fat samples for the inside round were taken from the exterior carcass fat surfaces. The BT lean samples were taken on the ventral lean surface where the inside round was separated through the natural seam. Because lean surfaces remained the same before and after fabrication, AT lean samples were not taken from the ventral surface. Instead, the exposed medial portion of the M. semimembranosus that was dried during chilling was removed, and the newly exposed lean surface served as the AT lean sample. The exterior carcass surface of the inside round was trimmed to 0.6 cm, and the AT fat sample for the inside round was taken on the newly exposed fat surface. For the gooseneck (IMPS 170) (North American Meat Processors Association, 2010), the BT fat samples were taken from the exterior carcass fat surface. The BT lean samples were taken on the ventral lean surface of the M. semitendinosus, M. gluteobiceps, and M. gastrocnemius. The gooseneck was processed further by removing the heavy
connective tissue (epimysium), popliteal lymph gland, sarcosciatic ligament, M. semitendinosus, and M. gastrocnemius muscles to produce a bottom round flat (IMPS 171B) (North American Meat Processors Association, 2010). The AT fat sample for the bottom round flat was taken on the newly exposed external fat surface after trimming to 0.6 cm. The AT lean sample for the bottom round flat was taken on the lean surface exposed after the M. semitendinosus, M. gastrocnemius, heavy connective tissue (epimysium), popliteal lymph gland, and sarcosciatic ligament were removed. During fabrication of the carcass sides, worker's knives, hooks, and steels were sanitized in a chemical sanitizer (Biquat, Birko Corporation, Henderson, CO) every 5 min. Following the fabrication of the first side of each carcass, the table tops were flipped before starting fabrication of the second side to minimize contamination from one side to the next. The cutting lab was cleaned and sanitized between carcasses. To ensure that cleaning and sanitizing practices removed any residual surrogate microorganisms, environmental samples were taken using sponges on randomly selected places on the slaughter floor, blast chill cooler, and cutting lab. Before sampling, the sponge was moistened with 25 ml of sterile Butterfield's phosphate buffer (Hardy Diagnostics, Santa Maria, CA), and the sample collection was achieved by firmly rubbing the damp sponge over a selected area of 400 cm 2 with a pre-moistened sterile sponge using the procedure of 10 vertical and 10 horizontal passes. Microbiological samples from subprimals were obtained by excising two 10-cm 2 × 2 mm thick samples using a sterile stainless steel borer, scalpel, and forceps, and compositing them (20-cm 2 total area). Each composite sample was placed into a sterile stomacher bag, placed in an insulated container containing refrigerants, and transported to the Food Microbiology Laboratory. Upon arrival, 99 ml of sterile 0.1% peptone was added to each stomacher bag. The sample then was pummeled for 1 min using a Stomacher 400 (Tekmar Company, Cincinnati, OH). For each sample, counts of rifampicin-resistant E. coli surrogate microorganisms were determined by plating appropriate decimal dilutions on prepoured and dried rifampicin-tryptic soy agar (rif-TSA) plates. The environmental samples were hand massaged for 1 min and appropriate decimal dilutions were plated on rif-TSA plates. Rif-TSA was prepared by adding a solution of 0.1 g of rifampicin (Sigma-Aldrich Inc., St. Louis, MO) dissolved in 5 ml methanol (EM Science, Gibbstown, NJ) to 1 l of autoclaved and cooled (55 °C) TSA. Rif-TSA plates were incubated for 18 ± 2 h at 35 °C before counting and reporting the number of rifresistant E. coli biotype I per cm 2. 2.5. Statistical analysis Plate counts were converted to log CFU per ml or cm 2 before analysis. Data were analyzed using PROC GLM of SAS (SAS Institute, Inc., Table 1 Least squares means for trimming location effect on counts (log10 CFU/cm2) of rifampicinresistant E. coli for the ribeye. Trimming locationa
Log10 CFU/cm2
Before trimming, fat After trimming, fat Before trimming, lean After trimming, lean
2.9a 0.9bc b 0.7cb 1.1b
Means lacking a common letter (a–c) differ (P b 0.05). a The before trimming fat samples were taken on the external fat surface over the M. longissimus thoracis. The after trimming fat samples were taken on the intermuscular fat surface on the dorsal aspect of the M. longissimus thoracis exposed after the lifter meat and all external fat were removed. The before trimming lean samples were taken on the M. longissimus thoracis face where the rib was separated from the chuck (6th rib interface). The after trimming lean sample was taken on the ventral M. longissimus thoracis exposed after the back ribs were removed. b Value denotes samples below the minimum detection level of 0.7 log CFU/cm2.
B.A. Laster et al. / Meat Science 90 (2012) 420–425 Table 2 Least squares means for trimming location effect on counts (log10 CFU/cm2) of rifampicinresistant E. coli for the brisket. Trimming location
a
Before trimming, fat After trimming, fat Before trimming, lean After trimming, lean
423
Table 4 Least squares means for trimming location effect on counts (log10 CFU/cm2) of rifampicinresistant E. coli for the chuck roll.
Log10 CFU/cm2
Trimming locationa
Log10 CFU/cm2
3.8a 1.8b 1.2b b 0.7cb
Before trimming, fat Before trimming, lean After trimming, lean
3.3a 0.9b 1.4b
Means lacking a common letter (a–c) differ (P b 0.05). a The before trimming fat samples were taken on the exterior carcass fat surface distal to the sternum. The after trimming fat samples were taken on the newly exposed surface after trimming to an external level of 0.6 cm. The before trimming lean samples were taken on the M. pectoralis profundus exposed after the sternum was removed. The after trimming lean samples were taken on the newly exposed M. pectoralis profundus after further trimming was performed to smooth the surface. b Value denotes samples below the minimum detection level of 0.7 log CFU/cm2.
Cary, NC). Least squares means were generated for main effects and separated using PDIFF option when appropriate with an alpha-level (P b 0.05). 3. Results and discussion This study used rif-resistant surrogate microorganisms that would not typically be found in a commercial beef-processing establishment. No detectable counts of rif-resistant microorganisms were found on the non-inoculated sides (data not shown). Environmental samples taken to ensure that the rif-resistant microorganisms were removed from the facility were below detectable levels (data not shown). There were no differences (P N 0.05) in counts (log10 CFU/cm 2) of initial carcass inoculum levels of rif-resistant microorganisms between harvest Day 1 (5.1) and Day 2 (4.9) and among the brisket (5.1), plate (5.0) and round (5.0). Although we would anticipate inconsistent or sporadic contamination in normal beef harvesting processes, we designed this study to create a consistent level of contamination to test the impact of trimming during the production of subprimals in the fabrication process. The before and after trimming effects for the lean and fat surfaces for each of the subprimals evaluated are presented in Tables 1 through 8. Similar trends were found for each of these subprimals. In six of the seven subprimals where there were direct comparisons between before and after trimming for fat, the after trimming fat surfaces had lower (P b 0.05) counts of rif-resistant E. coli than did the before trimming surfaces (the bottom round did not respond this way). For the most part, trimming external fat surfaces to produce these closely trimmed subprimals resulted in significantly lower pathogens on the resultant surfaces than if no trimming had occurred. For all eight subprimals, the before trimming lean surfaces had lower (P b 0.05) counts of rif-resistant E. coli than did the before trimming fat surfaces, which further implies that significantly fewer pathogens
Table 3 Least squares means for trimming location effect on counts (log10 CFU/cm2) of rifampicinresistant E. coli for the shoulder clod. Trimming locationa
Log10 CFU/cm2
Before trimming, fat After trimming, fat Before trimming, lean
3.7a 0.8b 1.1b
Means lacking a common letter (a and b) differ (P b 0.05). a The before trimming fat samples were taken on the external fat surface where the M. cutaneous omobrachialis was present. The after trimming fat samples were taken on the newly exposed surface after trimming to an external level of 0.6 cm. The before trimming lean samples were taken on the interior lean surface of the M. infraspinatus and M. triceps brachii. There was no after trimming lean sample taken for this subprimal.
Means lacking a common letter (a and b) differ (P b 0.05). a The before trimming fat samples were taken on the exterior carcass fat surface near the dorsal anterior portion of the chuck primal. There were no after trimming fat surfaces for the chuck roll. The before trimming lean samples were taken from the M. longissimus thoracis face where the rib was separated from the chuck (5th rib interface). The after trimming lean samples were taken on the dorsal aspect on the newly exposed surface where the external fat was removed.
will be on the initial lean surfaces when these subprimals are prepared than on the outside fat surfaces of the cut. However, in the limited direct comparisons of rif-resistant E. coli counts between the before and after trimming comparisons for the lean surfaces, in one subprimal they decreased (brisket, Table 2), in one they increased (ribeye, Table 1), and the remainder they did not differ between the two lean surfaces. At this point, the lean surfaces of the subprimals were minimally altered compared to the fat surfaces, which were reduced or removed in the preparation of the subprimals. Although there are a number of studies that report the efficacy of trimming hot carcasses as a decontamination method (Castillo et al., 1998a; Hardin et al., 1995; Prasai et al., 1995; Reagan et al., 1996), there are limited data regarding the impact of trimming chilled carcasses or cuts to reduce pathogens. Lemmons et al. (2011) found that trimming top sirloin butts that had been inoculated with high or low levels of E. coli O157:H7 was an effective method of decontaminating these beef subprimals. Lemmons et al. (2011) support the U.S. Department of Agriculture (2009) recommendations that trimming is a measure that establishments may use to address E. coli O157:H7; however, our study showed that trimming fat surfaces was more efficacious at reducing pathogens than trimming lean surfaces. Although there were variations in the microbiological counts from the impact of trimming exterior carcass surfaces during fabrication, this project showed that there was a general trend that counts from trimmed surfaces were lower than initially exposed surfaces. However, fat and lean surfaces that were not inoculated became contaminated during the fabrication process, which indicates that trimming external surfaces results in reduced, but not completely eliminated, pathogens on the finished products. The beef industry continues to be criticized by USDA-FSIS and consumer groups for not preventing illnesses associated with E. coli O157:H7. To address the concern of E. coli O157:H7, the industry has focused primarily on harvest interventions such as hot water, steam, and organic acid rinses to reduce pathogen contamination because the exterior surfaces may be contaminated with pathogens. Table 5 Least squares means for trimming location effect on counts (log10 CFU/cm2) of rifampicinresistant E. coli for the short loin. Trimming locationa
Log10 CFU/cm2
Before trimming, fat After trimming, fat Before trimming, lean
2.3a b 0.7bb 0.9b
Means lacking a common letter (a–c) differ (P b 0.05). a The before trimming fat samples were taken on the external fat surface over the M. longissimus lumborum. The after trimmed fat samples were taken on the newly exposed surface after trimming to an external level of 0.6 cm. The before lean samples were taken from the M. longissimus lumborum face on the sirloin end where the short loin was separated from the sirloin (6th lumbar interface). There were no after trimming lean surfaces for the short loin. b Value denotes samples below the minimum detection level of 0.7 log CFU/cm2.
424
B.A. Laster et al. / Meat Science 90 (2012) 420–425
Table 6 Least squares means for trimming location effect on counts (log10 CFU/cm2) of rifampicinresistant E. coli for the top sirloin. Trimming locationa
Log10 CFU/cm2
Before trimming, fat After trimming, fat Before trimming, lean
2.9a 1.2b 1.7b
Means lacking a common letter (a and b) differ (P b 0.05). a The before trimming fat samples were taken on the exterior carcass surface of this cut. The after trimming fat samples were taken on the newly exposed surface after trimming to an external level of 0.6 cm. The before trimming lean samples were taken on the ventral lean surface exposed after the ilium and sacral bones were removed. There was no after trimming lean surface for the top sirloin.
Table 7 Least squares means for trimming location effect on counts (log10 CFU/cm2) of rifampicinresistant E. coli for the inside round. Trimming locationa
Log10 CFU/cm2
Before trimming, fat After trimming, fat Before trimming, lean After trimming, lean
3.5a 2.3b 0.9c b 0.7cb
Means lacking a common letter (a–c) differ (P b 0.05). a The before trimming fat samples were taken on the exterior carcass surface of this cut. The after trimming fat samples were taken on the newly exposed surface after trimming to an external level of 0.6 cm. The before trimming lean samples were taken on the ventral lean surface of where the inside round was separated from the bottom round and knuckle. The after trimming lean samples were taken on the exposed medial portion of the M. semimembranosus that was dried during chilling and was trimmed away to expose a new surface for sampling. b Value denotes samples below the minimum detection level of 0.7 log CFU/cm2.
Research has shown that the prevalence of E. coli O157:H7 is not likely on beef subprimals (Heller et al., 2007; Kennedy et al., 2006). Our results support that trimming of external fat surfaces during the normal fabrication process may reduce contamination of E. coli O157:H7. However, we did show that existing pathogens may spread to newly exposed surfaces. This information will 'help the industry better understand how typical processing practices impact the safety of the products they produce. Acknowledgements This project was funded, in part, by The Beef Checkoff, The E. M. “Manny” Rosenthal Chair in Animal Science, and the Manny and Rosalyn Rosenthal Endowed Fund in the Department of Animal Science.
Table 8 Least squares means for trimming location effect on counts (log10 CFU/cm2) of rifampicinresistant E. coli for the bottom round. Trimming locationa
Log10 CFU/cm2
Before trimming, fat After trimming, fat Before trimming, lean After trimming, lean
2.6a 2.2a 1.1b 0.8b
Means lacking a common letter (a–c) differ (P b 0.05). a The before trimming fat samples were taken on the exterior carcass fat surface. The after trimming fat samples were taken on the newly exposed surface after trimming to an external level of 0.6 cm. The before trimming lean samples were taken on the ventral lean surface of the M. semitendinosus, M. gluteobiceps, and M. gastrocnemius. The after trimming lean samples were taken on the surface exposed after the heavy connective tissue (epimysium), popliteal lymph gland, sarcosciatic ligament, M. gastrocnemius, and M. semitendinosus were removed.
References Baird, B. E., Lucia, L. M., Acuff, G. R., Harris, K. B., & Savell, J. W. (2006). Beef hide antimicrobial interventions as a means of reducing bacterial contamination. Meat Science, 73(2), 245–248. Barkocy-Gallagher, G. A., Arthur, T. M., Rivera-Betancourt, M., Nou, X., Shackelford, S. D., Wheeler, T. L., et al. (2003). Seasonal prevalence of shiga toxinproducing Escherichia coli, including O157:H7 and non-O157 serotypes, and Salmonella in commercial beef processing plants. Journal of Food Protection, 66(11), 1978–1986. Bosilevac, J. M., Nou, X., Osborn, M. S., Allen, D. M., & Koohmaraie, M. (2005). Development and evaluation of an on-line hide decontamination procedure for use in a commercial beef processing plant. Journal of Food Protection, 68(2), 265–272. Cabrera-Diaz, E., Moseley, T. M., Lucia, L. M., Dickson, J. S., Castillo, A., & Acuff, G. R. (2009). Fluorescent protein-marked Escherichia coli biotype I strains as surrogates for enteric pathogens in validation of beef carcass interventions. Journal of Food Protection, 72(2), 295–303. Castillo, A., Hardin, M. D., Acuff, G. R., & Dickson, J. S. (2002). Reduction of microbial contaminants on carcasses. In V. K. Juneja, & J. N. Sofos (Eds.), Control of foodborne microorganisms (pp. 351–381). New York: Marcel Dekker, Inc.. Castillo, A., Lucia, L. M., Goodson, K. J., Savell, J. W., & Acuff, G. R. (1998a). Comparison of water wash, trimming, and combined hot water and lactic acid treatments for reducing bacteria of fecal origin on beef carcasses. Journal of Food Protection, 61(7), 823–828. Castillo, A., Lucia, L. M., Goodson, K. J., Savell, J. W., & Acuff, G. R. (1998b). Use of hot water for beef carcass decontamination. Journal of Food Protection, 61(1), 19–25. Castillo, A., Lucia, L. M., Mercado, I., & Acuff, G. R. (2001). In-plant evaluation of a lactic acid treatment for reduction of bacteria on chilled beef carcasses. Journal of Food Protection, 64(5), 738–740. Elder, R. O., Keen, J. E., Siragusa, G. R., Barkocy-Gallagher, G. A., Koohmaraie, M., & Laegreid, W. W. (2000). Correlation of enterohemorrhagic Escherichia coli O157 prevalence in feces, hides, and carcasses of beef cattle during processing. Proceedings of the National Academy of Sciences of the United States of America, 97(7), 2999–3003. Gill, C. O., & Bryant, J. (1997). Decontamination of carcasses by vacuum-hot water cleaning and steam pasteurizing during routine operations at a beef packing plant. Meat Science, 47(3–4), 267–276. Graves Delmore, L. R., Sofos, J. N., Reagan, J. O., & Smith, G. C. (1997). Hot-water rinsing and trimming/washing of beef carcasses to reduce physical and microbiological contamination. Journal of Food Science, 62(2), 373–376. Hardin, M. D., Acuff, G. R., Lucia, L. M., Oman, J. S., & Savell, J. W. (1995). Comparison of methods for decontamination from beef carcass surfaces. Journal of Food Protection, 58(4), 368–374. Heller, C. E., Scanga, J. A., Sofos, J. N., Belk, K. E., Warren-Serna, W., Bellinger, G. R., et al. (2007). Decontamination of beef subprimal cuts intended for blade tenderization or moisture enhancement. Journal of Food Protection, 70(5), 1174–1180. Kaspar, C. W., & Tamplin, M. L. (1993). Effects of temperature and salinity on the survival of Vibrio vulnificus in seawater and shellfish. Applied and Environmental Microbiology, 59(8), 2425–2429. Kennedy, J. E., Williams, S. K., Brown, T., & Minerich, P. (2006). Prevalence of Escherichia coli O157:H7 and indicator organisms on the surface of intact subprimal beef cuts prior to further processing. Journal of Food Protection, 69(7), 1514–1517. Lemmons, J. L., Lucia, L. M., Hardin, M. D., Savell, J. W., & Harris, K. B. (2011). Evaluation of Escherichia coli O157:H7 translocation and decontamination for beef vacuumpackaged subprimals destined for non-intact use. Journal of Food Protection, 74(7), 1048–1053. Márquez-González, M., Harris, K. B., & Castillo, A. (2010). Interventions for hazard control in foods during harvesting. In V. K. Juneja, & J. N. Sofos (Eds.), Pathogens and toxins in foods: Challenges and interventions (pp. 379–395). Washington, DC: ASM Press. Marshall, K. M., Niebuhr, S. E., Acuff, G. R., Lucia, L. M., & Dickson, J. S. (2005). Identification of Escherichia coli O157:H7 meat processing indicators for fresh meat through comparison of the effects of selected antimicrobial interventions. Journal of Food Protection, 68(12), 2580–2586. Niebuhr, S. E., Laury, A., Acuff, G. R., & Dickson, J. S. (2008). Evaluation of nonpathogenic surrogate bacteria as process validation indicators for Salmonella enterica for selected antimicrobial treatments, cold storage, and fermentation in meat. Journal of Food Protection, 71(4), 714–718. North American Meat Processors Association (2010). The meat buyer's guide™ (6th ed.). Reston, VA: North American Meat Processors Association. Prasai, R. K., Phebus, R. K., Garcia Zepeda, C. M., Kastner, C. L., Boyle, A. E., & Fung, D. Y. C. (1995). Effectiveness of trimming and/or washing on microbiological quality of beef carcasses. Journal of Food Protection, 58, 1114–1117. Reagan, J. O., Acuff, G. R., Buege, D. R., Buyck, M. J., Dickson, J. S., Kastner, C. L., et al. (1996). Trimming and washing of beef carcasses as a method of improving the microbiological quality of meat. Journal of Food Protection, 59(7), 751–756. Savell, J. W., & Smith, G. C. (2009). Meat science laboratory manual (8th ed.). Boston, MA: American Press. U.S. Department of Agriculture (1999). FSIS policy on non-intact raw beef products contaminated with E. coli O157:H7. (Available from). Washington, DC: United States Department of Agriculture, Food Safety and Inspection Service. http://www.fsis.usda.gov/oa/background/O157policy.htm (Accessed 16 July 2010).
B.A. Laster et al. / Meat Science 90 (2012) 420–425 U.S. Department of Agriculture (2009). Measures to address E. coli O157:H7 at establishment that receive, grind, or otherwise process raw beef products. FSIS Notice 05–09. Washington, DC: United States Department of Agriculture, Food Safety and Inspection Service. U.S. Department of Agriculture (2010a). FSIS recalls, recall case archive. (Available from). Washington, DC: United States Department of Agriculture, Food Safety and Inspection Service. http://www.fsis.usda.gov/fsis_recalls/Recall_Case_Archive/index.asp (Accessed 25 September 2010).
425
U.S. Department of Agriculture (2010b). Microbiological results of raw ground beef products analyzed for Escherichia coli O157:H7. (Available from). Washington, DC: United States Department of Agriculture, Food Safety and Inspection Service. http:// www.fsis.usda.gov/Science/ground_beef_e.Coli_Testing_results/index.asp (Accessed 2 October 2010).