03/30/2006 11:21 AM 1

Appendix A Best Practices: Pathogen Control During Tenderizing/Enhancing of Whole Muscle Cuts

Best Practices:

Pathogen Control During Tenderizing/Enhancing of

Whole Muscle Cuts

Supported by:

National Cattlemen’s Beef Association American Meat Institute

National Meat Association Southwest Meat Association

Revised February 2006

February 2006

2

The American Meat Institute (AMI), National Cattlemen’s Beef Association (NCBA), National Meat Association (NMA), and Southwest Meat Association (SMA) are pleased to have developed these industry Best Practices for Pathogen Control for Tenderizing Operations of Whole Muscle Cuts. In September 2003 leading manufacturers of non-intact meat products collaborated under the guidance of the American Meat Institute, National Meat Association, Southwest Meat Association, National Cattlemen’s Beef Association, and developed the Best Practices for review by the Beef Industry Food Safety Council (BIFSCo). The Best Practices for Beef Slaughter (NMA et al., 2003a) and Best Practices for Handling Vacuum Packed Subprimal Beef Cuts (AMI et al., 2003) were used as resources in developing recommendations for non-intact beef products. Substantial updating of this document was completed following the Non-intact Products Processing Workshop (December 2005) based on meeting participants’ comments. A full summary of this meeting is documented in Beef Industry Addresses the Safety of Non-Intact Beef Products (NCBA, 2006). While the operating practices at individual companies may vary, producers of non-intact whole-muscle cuts are urged to consider these Best Practices as guidelines for their own internal practices and documentation. These practices are the best conditions known at the date of publication. The following individuals should be recognized for their contribution to the development of these Best Practices:

Dell Allen, Cargill (retired) Sharon Beals, Tyson Foods Dane Bernard, Keystone Foods Vince DeGrado, Rosen Meat Group Mitch Gilgour, Sysco Corp. Lynn Graves Delmore, California State Polytechnic University, San Luis Obispo Les Glowka, Quantum Foods Randy Huffman, American Meat Institute Foundation Scott Eilert, Cargill Meat Solutions Brian Farnsworth, Hormel Foods Forrest Dryden, Hormel Foods Jerome Lawler, Swift & Company Twila Leierer, Arby’s, LLC Ali Mosheni, American Foods Group Nick Nickelson, Standard Meat Company Jose L. Prego, Cozzini Group Skip Seward, American Meat Institute

February 2006

3

Industry Best Practices for Pathogen Control During

Tenderizing/Enhancing of Whole Muscle Cuts

Purpose This document is designed to discuss Best Practices that can be implemented throughout the tenderizing or enhancing operation, as well as during cleaning and sanitizing operations, to reduce the likelihood that contamination with potential pathogens (specifically E. coli O157:H7) will occur. There are multiple ways to reach the optimal end-result, and each operator must be able to apply the practices and procedures that best fit an individual operation. This document is not designed to mandate the use of any specific system or technology, but rather, to stress the importance of validating that the tenderizing or enhancing system is optimized to reduce the risk of contamination. Introduction FSIS defines non-intact beef products as ground beef; beef injected with solution, beef that has been mechanically tenderized by needling, cubing, frenching, or pounding devices, and beef that has been reconstructed into formed entrees. Whole muscle cuts (e.g., chucks, ribs, tenderloins, strip loins, top sirloin butts, rounds) may be treated to increase tenderness or to add ingredients for quality purposes, a practice that often occurs before subsequent fabrication at the same or external location. Treatments may include solid-needle tenderizing or hollow-needle tenderizing where a solution is pumped into the whole muscle. In the latter case, the solution typically is re-circulated, refrigerated and treated to ensure the quality of the pumping solution. It is important that the management of these operations be such that the equipment, refrigeration, solutions and product are optimized for quality and safety. Producers of raw non-intact beef products recognize that these products may pose a risk if potential pathogens are moved to the interior portions of the meat products (Krizner, 1999; Phebus et al., 2000; Lambert et al., 2001; Hajmeer et al., 2002), and the product is not cooked adequately to destroy the pathogens inside the meat product. As is discussed below, the likelihood of potential pathogens being transferred to the inside from the outside of the product is extremely low because of a very low prevalence of pathogens on meat portions being tenderized or enhanced (Ransom et al., 2002; Warren et al., 2003). If equipment used in the operation is contaminated somehow, and not cleaned and sanitized, the tenderizing or enhancing equipment, and perhaps the solution to be injected, may become the vehicle of the contamination. To reduce the risk, it is extremely important that processors implement Best Practices by focusing on cleaning and sanitation practices for tenderizing and enhancing operations. One of the primary considerations in assessing the likelihood of contamination of products that are tenderized or enhanced is whether or not contamination, especially with E. coli O157:H7, is a hazard reasonably likely to occur on the surface of intact meat portions before the tenderizing or enhancing operation. Several studies indicate that E. coli O157:H7 is not a hazard reasonably likely to occur on the surface of intact meat portions. A study was conducted by Warren et al. (2003) where sponge samples were taken of 1,014 subprimal cuts from six beef processing plants

February 2006

4

over a five-week period. Only two samples (0.2%) tested positive for E. coli O157:H7. Enumeration indicated that each of the two positive samples had <3.0 CFU per 200 cm2 sampled. Two later studies were conducted by ABC Research Corporation (Gainesville, Fla.) throughout 2004 to determine the prevalence of E. coli O157:H7 and indicator organisms on the surface of beef subprimals that would be used as raw materials for tenderizing or enhancing operations. These studies used cuts of meat specifically used for tenderizing or enhancing operations, namely, briskets, rounds, chucks and middle meats. One study (I) focused on raw materials produced during the winter months (January and February); the second study (II) collected data during the late summer and fall (August into November). In Study I, 600 samples comprising six subprimal cut types (100/type) were collected from five plants from the southern Midwest, Midwest, northern Midwest and the Southeast. Each sample was a sponge sample of the entire surface of a subprimal. None of the 600 samples had E. coli O157:H7. In study II, 599 samples (following the same scheme described above for study I) tested negative for E. coli O157:H7. Based on limits of methodologies and the results from Studies I and II, the authors concluded that the overall incidence of E. coli O157:H7 on beef subprimals was < 0.083% (Kennedy and Badnaruk, 2004, 2005). This document provides Best Practices for tenderizing and enhancing operations and can be used by establishments to develop plant specific programs. Although these Best Practices are applicable to both production of raw and fully cooked tenderized and/or enhanced items, this document primarily focuses on the manufacture of raw non-intact products (excluding ground beef). These Best Practices are designed to provide a recommended set of practices and procedures that processors may want to adopt in their entirety, or in part to ensure optimal wholesomeness. Raw Material Control Best Practices begin with optimizing raw material (i.e. whole muscle cuts) quality and safety. Tenderizing and enhancing operations should identify requirements for raw material suppliers and have a system for verification that the requirements are being met and achieving the goals of the quality and safety program. Criteria to select raw material suppliers should include that suppliers have process interventions in place to reduce or eliminate potential enteric pathogens. Raw material suppliers should have validated process interventions and/or validated critical control points (CCPs) in place to prevent, eliminate or reduce E. coli O157:H7 to a non-detectable level. As always, multiple interventions (hurdles) are preferable to single microbial interventions. Validation may include scientific literature and/or plant specific validation using indicator organisms, and it should be specific to the process being applied at the establishment. This validation can be incorporated into the processor’s purchase specifications or other plant programs to ensure that all raw materials are produced using validated CCPs or process interventions. These purchase specifications should have a means to ensure that they are being met. Examples of such verification tools include, but are not limited to third party process reviews, customer audits and microbiological testing. This is true for both domestic and imported suppliers of raw materials to be used in production of non-intact product. Purchase specifications should be updated regularly

February 2006

5

(at least annually). An example letter from a harvest/fabrication facility to meet the processor’s prerequisite program requirements has been provided and is included in Best Practices: Appendix A. Another important criterion for supplier selection is the ability and demonstrated maintenance of cold chain management. This includes rapid chilling of hot carcasses to control microbial growth and proper carcass rotation within the cooler to ensure timely fabrication. Lastly, it is important for non-intact beef processors to have specific data on E. coli O157:H7 incidence to support the position taken during the hazard analysis as “not reasonably likely to occur.” These data must relate to the raw materials and/or finished product(s). Routine microbiological testing may include sampling and testing for E. coli O157:H7. Other microbiological testing includes analyses for Salmonella, Aerobic Plate Count (APC), Total Plate Count (TPC), coliforms, and generic E. coli. For all microbiological testing, it is important that there be a written protocol for sample collection, lab analysis and proficiency testing, as well as the procedures for reporting the results. It is important to establish how the results will be used before the data are collected. Most of these microbiological tests are used for tracking supplier trends over time; however, each establishment must clearly define how they are going to use the information and the consequences of failing to meet internal microbiological guidelines. Supplier Evaluations Raw material suppliers are critical to both food safety and quality aspects of producing tenderized and enhanced products. In addition to well-defined requirements it is important that there are procedures established to evaluate the raw material supply whether from an internal or external vendor source. Guidelines developed for the Raw Ground Products Best Practices can be used to help design a system for evaluating supply sources for other non-intact raw materials. A more detailed discussion of supplier evaluations can be found in the Best Practices for Raw Ground Products document (NMA et al., 2003b; www.bifsco.org/BestPractices.htm). Temperature Control Cold chain management is a continuum from the time a carcass leaves the slaughter process and enters the chilling process through processing, packaging, storage and distribution. The goal is to achieve and maintain the temperature that will inhibit the growth of foodborne pathogens and slow the growth of spoilage microflora. The minimum growth temperatures for the pathogens of most concern are 44.6°F (7°C) for salmonellae and 44.6-46.4°F (7-8°C) for pathogenic E. coli (ICMSF, 1996). If cold chain control is violated at any point in the chain, product safety and quality may be compromised. Cold chain management is especially important at the tenderizing or enhancing operation. Specific points where temperature should be controlled, other control points related to temperature control, and examples of operating limits in tenderizing or enhancing operations include:

• Receiving and storage of raw materials at 40°F or less • Processing raw materials using a “First In First Out” (FIFO) rotation • Monitoring raw materials and finished products using a process room/cooler control

program

February 2006

6

• Verifying the potability of process water • Maintaining process water at 40°F or less • Maintaining finished product temperatures at 40°F or less throughout their shelf life • Controlling brine solutions to 40°F or less • Pre-chilling shipping containers to 40°F or less before loading • Maintaining temperatures at 40°F or less throughout transport

While temperatures are specified at 40°F or less in the above list based on the growth limitations for pathogenic Salmonella and E. coli O157:H7, it is generally recognized that the colder the temperature the better. Process Controls There are three general types of processing that are recognized within tenderizing and enhancing operations. These include needle tenderizing, brine-injecting (enhancing), and suspension injecting. Specific Best Practices will be presented for each of these categories due to unique differences between the processes. Example Standard Operating Procedures (SOP) are provided in the appendix as a reference for cleaning and sanitizing of injector assembly (Best Practices: Appendix B). Every process and enhancement system is unique and appropriate SOP’s should be in place depending on the situation.

Needle Tenderized Products • Documented GMPs (including needle integrity checks) exist for tenderizing operations • If possible, needle the product from the side opposite of the external surface to minimize

any bacterial translocation • Traceability program is in place for all finished products • Food Defense program exists to prevent tampering with operational equipment, and raw

materials

Enhanced/Brine-Injected Products • Letters of guarantee and certificates of analysis exist for ingredients used in pumping

solution (brine or pickle solution) • Documented General Manufacturing Practices (including needle integrity checks) exist

for injecting operations • Chilled water feeding system is preferable to complete chilling of brine following mixing • Maximum age is established for reuse brine (pickle) solutions (e.g., 24 hours), with a

mandatory break in the use cycle (e.g., every 24 hours) • Use of an antimicrobial intervention (e.g., filtration, UV) for recirculating pickle solution

is implemented if deemed necessary by the hazard analysis • Use of bacterostatic ingredients in the brine solution (e.g. lactate, diacetate, sodium

metasilicate) is implemented if deemed necessary by the hazard analysis • If possible, inject the product from the side opposite of the external surface to minimize

any bacterial translocation • Daily needle removal and soaking in sanitation solution is conducted • Established protocol exists for managing rework, including traceability and a time frame

for incorporation into manufacturing

February 2006

7

• Traceability program is in place for all finished products • Food Defense program exists to prevent tampering with operational equipment, raw

materials and pickle solutions

Meat Protein Suspension Injection Products1 • Letters of guarantee and certificates of analysis exist for ingredients used in the

processing of the suspension solution (to include all meat and nonmeat ingredients in the brine or pickle solution, as well as documentation on “supplier evaluation” on the sources the trim raw material used)

• Documented GMPs (including needle integrity checks) exist for injecting operations • Chilled water feeding system is preferable to complete chilling of brine following mixing

and as the suspension is generated from it • Maximum age is established for reuse brine (pickle) solutions (e.g., 24 hours), with a

mandatory break in the use cycle (e.g., every 24 hours) • Maximum age is established for reuse suspension solutions (e.g., 8 hours), with a

mandatory break in the use cycle (e.g., every 16-20 hours) • Use of an antimicrobial intervention (e.g., UV) for re-circulating pickle solution is

implemented if needed as determined by the hazard analysis • Use of bacterostatic ingredients in the brine solution (e.g. lactate, diacetate, sodium

metasilicate) if needed as determined by the hazard analysis • If possible, inject the product from the side opposite of the external surface to minimize

any bacterial translocation • Daily needle removal and soaking in sanitation solution is conducted • Established protocol exists for managing rework, including traceability and a time frame

for incorporation into manufacturing • Traceability program is in place for all finished products • Food Defense program exists to prevent tampering with operational equipment, raw

materials and pickle solutions Lotting All non-intact processors should have a lotting mechanism for coding and recording all products to allow trace back and trace forward of products throughout the manufacturing and distribution system. FSIS recognizes that the establishment will define a lot and expects scientific or other supportive basis for defining the lot. Lotting systems can range from very simplistic, e.g., handwritten numbering, to very elaborate, e.g., computerized, automated bar coding. Lotting is often based on some unit of time (e.g., hour, shift, day); however lotting can be driven by other factors including raw material source, production line or processing room. Some processors may choose to further divide lots of product into sublots. By creating smaller lot units, process control can be demonstrated and documented more frequently; and there is a potential to minimize the

1 Cozzini’s SUSPENTEC TM system is a patented method of reducing meat, poultry or fish trimmings to micron size and incorporating them into traditional brines to create a suspension; the suspensions can then be injected into whole-muscle products. The use of this equipment is governed by FSIS Policy Memo PM041B. At the time this document was put together, Cozzini’s SUSPENTEC TM system was the only such technology available for Beef, Pork and Poultry. These practices may or may not be applicable to other suspension technologies when they become available.

February 2006

8

volume of product implicated in the event a recall is ever required. In tenderizing and enhanced operations, there is some precedence that FSIS will accept a single bag of subprimals as a lot, provided the processing facility can show adequate separation. If lots are intended to be broken at some frequency by needle rotation, accompanying sanitation of the feed-in area (debagging tables, conveyors) is also necessary. Additionally, establishments should maintain records associated with all production lots. Information to be recorded is dependent on the individual system; however the following data typically are recorded:

• Raw material vendor, vendor lot • Process date, time of production • Raw material, brine, room and product temperature • Microbiological data • Equipment evaluations

A more detailed discussion of lotting can be found in the Best Practices for Raw Ground Products document (NMA et al., 2003b; www.bifsco.org/BestPractices.htm ). HACCP System Non-intact products will be produced under FSIS or state inspection, thereby meeting all Federal or State (equal to) requirements pertaining to HACCP systems (9 CFR 417), Sanitation SOPs (9 CFR 416) and pre-requisite programs. All processors should be able to support the decisions that are made in the HACCP program and to use the documentation generated from the program to demonstrate product safety. HACCP is a proactive, systematic approach to food safety designed to prevent, eliminate or reduce food safety hazards to an acceptable level. Processing establishments must consider biological, physical, and chemical food safety hazards. As far as the authors know, there are no data to suggest that through a hazard analysis, E. coli O157:H7 should be considered a hazard reasonably likely to occur in tenderizing or enhancing operations. In fact, as mentioned earlier, data (nearly 1200 data points collected in the winter, fall and summer of 2004) have established that E. coli O157:H7 is not a hazard reasonably likely to occur on whole muscle cuts destined for tenderizing or enhancing operations. Likewise, additional studies have documented the very low incidence of E. coli O157:H7 on the surface of subprimals destined to be enhanced or mechanically tenderized. Data show only three to four percent of surface bacterial populations are translocated to an average interior depth of ¼” of the cuts during processing (Sporing, 1999; Lambert et al., 2001). Thus, mechanically tenderized and enhanced products pose no greater risk than intact cuts when cooked to a rare degree of doneness (140°F) (Marsden et al., 1999). A review of current research results is presented by the NCBA white paper entitled Beef Industry Addresses the Safety of Non-intact Beef Products (NCBA, 2006). However, because these are raw meat processing operations, consideration should be given to E. coli O157:H7 as a potential, sporadic contaminate that could find its way into the processing environment and specific tenderizing or enhancing processing systems. Additionally, FSIS gave notice that all processors must reassess their HACCP systems to consider three foodborne outbreaks of E. coli O157:H7 that may have been linked to enhanced/tenderized beef steaks in their hazard analysis (FSIS-USDA, 2005). Thus, processors must focus on what practical strategies can be applied during the tenderizing or enhancing process to minimize the potential

February 2006

9

for growth of E. coli O157:H7 if present as a process contaminant or as a highly unlikely contaminant of subprimals. These strategies typically involve prevention of harborages and niches through cleaning and sanitation of equipment, maintaining cold temperatures and using antimicrobial interventions on the subprimals prior to processing and during recirculation of enhancement solutions. Occasional verification that E. coli O157:H7 is not being harbored in the plant environment by swabbing equipment is recommended. Sanitation and Facilities Production of tenderized and enhanced products must occur in facilities that meet all Federal regulations (9 CFR 307, 310, 313, 314, 317, 318, 320, and 416) and the equipment used must meet sanitary operating guidelines. Establishments should meet all regulatory requirements of the Sanitation Standard Operating Procedures and should consider the guidelines presented in the Sanitation Performance Standards. For optimal operation, the entire system should be process engineered. The idea of process engineering encompasses facility design, equipment design, product movement, supply movement and employee movement to create an environment that minimizes microbial contamination. The American Meat Institute’s Sanitary Design of Equipment and Facilities (AMI, 2003) serves as a good reference. A checklist and a fact sheet, can be accessed at the following Web sites: (http://www.meatami.com/Content/ContentGroups/Food_Safety_Inspection/Inspection1/Sanitation1/AMIEquipmentdesignChecklist.xls http://www.meatami.com/Content/NavigationMenu/PressCenter/FactSheets_InfoKits/FactSheetSanitaryDesign.pdf). FSIS personnel (Engeljohn, 2005) have suggested that insufficient sanitation of equipment was the biggest issue in the three E. coli O157:H7 outbreaks possibly linked to enhanced/tenderized beef steaks. The agency believes proper sanitation to be the single most important control measure available to processors of mechanically tenderized and enhanced products to prevent foodborne outbreaks. Specifically, enhanced and mechanically tenderized processors should follow sanitation practices much like those adhered to by ready to eat (RTE) operations. A comprehensive review of RTE sanitation and practices are found in the Guidelines for Developing Good Manufacturing Practices (GMPs), Standard Operating Procedures (SOPs) and Environmental Sampling/Testing Recommendations (ESTRs) in Ready to Eat (RTE) Products (NMA, 1999). As the tenderizers/injectors pass through the product they may introduce biological hazards to the interior or the product. Inadequate injection needle sanitation poses the greatest risk to spread any microbial contaminants present on the incoming raw materials, thus needle sanitation is critical. All needles must be removed at least daily and soaked in a sanitation solution prior to inspection and reassembly of the needle injector. Ideally, two sets of needles could be rotated to allow for maximum soaking time and potentially greater sanitation efficacy. Injection systems should be cleaned in place (CIP) using a validated sanitation process of cleaning followed by

February 2006

10

sanitizing. Standard operating procedures should include the chemical concentration, frequency of cleaning, responsible party and how it will be verified. Validation and verification of sanitation practices are always challenging, however the nature of small diameter hollow injection needles further compounds this issue. To validate the efficacy of the sanitation system needles can be sacrificed (broken) to determine if the cleaning and sanitizing procedures are adequate. Likewise, routine verification of sanitation practices for needles can be determined by sacrificing and sampling needles at some frequency. One processor has reported sacrificing one needle per cleaning cycle to verify internal needle cleanliness. Interventions/Inhibitors When called for by the hazard analysis, a validated intervention may be appropriate. The most basic intervention is knife trimming; which can be utilized with primals, subprimals, roasts and steaks prior to penetration. Other current applied technologies include application of antimicrobial solutions to the raw materials before processing, treatment of the brine with an inhibitory process (e.g., ultraviolet and/or filtration), addition of inhibitory ingredient to the brine and the use of an intervention or inhibitor applied to the finished product or packaging materials. New antimicrobial intervention and inhibitors that may be applicable in tenderizing or enhancing operations continue to be developed. A list of potential interventions at the time this document was written is included in Best Practices: Appendix C. For illustrative purposes, an in-plant study on the antimicrobial properties of a tenderizing pickle solution has been provided in Best Practices: Appendix D.

Microbiological Testing Some producers have elected to sample and test for E. coli O157:H7 on subprimals destined for non-intact processing operations. Therefore, their verification testing data would serve as a basis for the hazard analysis. Finished product microbiological testing is a means to verify process control and evaluate that the Best Practices discussed throughout this document are being used effectively to reduce the likelihood of contamination by potential pathogens and the overall microbial load on the finished product. However, finished product sampling cannot be used to ascertain the safety of the product unless enough samples are taken to develop a statistically based rationale for acceptance (e.g., 95 percent confidence that the probability of contamination is no greater than five percent). Generally, the economics of testing finished products and the high numbers of samples required to have a relatively high degree of confidence that a low level of contamination will be detected, make finished product testing impractical. There may be instances where finished product testing has some value, e.g., for periodic verification using indicator organisms, or when a process is out-of-control and an assignable cause is being sought. Processors can achieve verification of the efficacy of a harvest/fabrication facility’s processes to minimize microbial contaminants without microbial testing of incoming raw materials (subprimals). One way is to obtain copies of the harvest/fabrication facility’s latest (at least annually) third-party food safety/HACCP audit. Additionally, processors can request that the harvest/fabrication facilities share their own routine microbiological verification data with the non-intact processor.

February 2006

11

Packaging and Labeling Packaging of non-intact beef cuts must occur in a manner to minimize the likelihood of contamination from packaging equipment, the environment, or food contact surfaces. Routine microbiological audit sampling and testing may be used to verify the efficacy of cleaning and sanitation, both on a routine basis and following equipment maintenance or relocation (AMI et al., 2003). It is the belief of FSIS that consumers do not understand or expect whole muscle steaks and roasts to have been needled. Thus, the agency has suggested that processors consider voluntary labeling of enhanced and mechanically tenderized products to identify them as non-intact and to include cooking instructions. At least one large processor currently includes cooking instructions (145°F for three minutes) on such products. Integrated Approach to Control One way to evaluate the overall safety of a product is by calculating the integrated control measures, which is an evaluation of the baseline incidence and the bateriostatic / bacteriocidal effects of all the variables which contribute to the safety of the end product. The integrated approach to control includes, but is not limited to the following factors:

• Organism incidence rates in live animals • Interventions applied at harvest and fabrication • Raw material incidence rates • Application of industry recognized best practices • Interventions (including knife trimming) applied prior to injection/mechanical

tenderization • Organism translocation rates due to injection/mechanical tenderization • Antimicrobial effects of an enhancement brine • Ingredients affecting the heat liability of the organism • Temperature control to minimize microbial amplification • Cooking practices applied to the products • Integrated time-temperature processing (integrated lethality)—incorporates all heat

treatments, i.e. the increase in temperature as the product heats and the temperature levels as the product cools. Microbial destruction takes place during the entire heating and cooling process, not just at the minimum internal temperature.

• Relationship between depth of possible translocation, cooking time and temperature to effectively destroy microorganisms

By considering all of these variables, the true safety of the product can be determined.

February 2006

12

Best Practices: References American Meat Institute. 2003. Equipment Design Task Force. Sanitary Design Checklist and

Fact Sheet. American Meat Institute, Washington, DC. American Meat Institute, National Meat Association, National Cattlemen’s Beef Association,

and Southwest Meat Association. 2003. Best Practices for Handling Vacuum-Packed Subprimal Cuts. Beef Industry Food Safety Council, Centennial, CO. (www.bifsco.org).

Engeljohn, D. 2005. FSIS perspective on non-intact beef. Non-Intact Product Processing Workshop, Dallas, TX.

FSIS-USDA. 2005. Federal Register Notice: HACCP Plan Reassessment for Mechanically Tenderized Beef Products. Fed. Regist. 70(101):30331-30334.

Hajmeer, M.N., E. Ceylan, J.L. Marsden, and R.K. Phebus. 2002. Translocation of natural microflora from muscle surface to interior by blade tenderization. Cattlemen’s Day 2002. Report on Progress 850, pp. 125-126. Kansas State University, Manhattan, KS.

International Commission on Microbiological Specifications for Foods. 1996. Microorganisms in Foods: Microbiological Specifications of Food Pathogens. Blackie Academic & Professional, New York, NY.

Kennedy, J. and P. Bodnaruk. 2005. Survey of the prevalence of Escherichia coli O157:H7 on the surface of subprimal cuts of beef during late summer and fall months. Final report to NCBA, Centennial, CO.

Kennedy, J. and P. Bodnaruk. 2004. Survey of the prevalence of Escherichia coli O157:H7 on the surface of subprimal cuts of beef during winter months. Final report to NCBA, Centennial, CO.

Krizner, K. 1999. Blade-tenderization safe for steaks: Studies suggest process poses little risk of introducing E. coli O157:H7 onto beef steaks. Meat Marketing & Technology (September Issue) p. 120.

Lambert, D.L., R.K. Phebus, H. Thippareddi, J.L. Marsden, and C.L. Kastner. 2001. Salmonella spp. risk assessment for production and cooking of non-intact pork products. IAFP Annual Meeting Program and Abstract Book. (PO54). p. 60 (Abstr.).

Marsden, J.F., R. K. Phebus, H. Thippareddi, and C.L. Kastner. 1999. Escherichia coli O157:H7 risk assessment for the production and cooking of blade tenderized steaks. Final report to NCBA, Englewood, CO.

National Cattlemen’s Beef Association. 2006. Beef Industry Addresses the Safety of Non-Intact Beef Products. Non-Intact Products Processing Workshop, Dallas, TX.

National Meat Association. 1999. Guidelines for Developing Good Manufacturing Practices (GMPs), Standard Operating Procedures (SOPs) and Environmental Sampling/Testing Recommendations (ESTRs) in Ready to Eat (RTE) Products. National Meat Association, Oakland, CA.

National Meat Association, Southwest Meat Association, American Meat Institute and National Cattlemen’s Beef Association. 2003a. Best Practices for Beef Slaughter. Beef Industry Food Safety Council, Centennial, CO. (www.bifsco.org).

National Meat Association, Southwest Meat Association, American Meat Institute and National Cattlemen’s Beef Association. 2003b. Best Practices for Raw Ground Products. Beef Industry Food Safety Council, Centennial, Colo. (www.bifsco.org).

Phebus, R.K., H. Thippareddi, S. Sporing, J.L. Marsden, and C.L. Kastner. 2000. E. coli O157:H7 risk assessment for blade-tenderized beef steaks. Cattlemen’s Day 2000. Report of Progress 850, pp. 117-118. Kansas State University, Manhattan, KS.

February 2006

13

Ransom, J.R., K.E. Belk, R.T. Bacon, J.N. Sofos, J.A. Scanga, and G.C. Smith. 2002. Comparison of sampling methods for microbiological testing of beef animal rectal/colonal feces, hides and carcasses. J. Food Prot. 65:621-626.

Sporing, S.B. 1999. E. coli O157:H7 Risk Assessment for Production and Cooking of Blade Tenderized Beef Steaks. Master Thesis. Kansas State University, Manhattan, Kan.

Warren, W., S. Wood, G. Bellinger, T. Frederick, and G. Smith. 2003. Characterization of E. coli O157:H7 on Subprimal Beef Cuts Prior to Mechanical Tenderization. NCBA Project Report, Centennial, CO.

February 2006

14

Best Practices: Appendix A Example E. coli O157:H7 Purchase Specification Letter for Supplier Evaluations Attention: Customer Name Edible beef products from the plants listed at the end of this letter meet all USDA requirements for the production, sale and distribution of meat products. Such requirements include, but are not restricted to the categories listed below. Updates will be issued annually or as significant changes are made. HACCP/Pathogen Reduction Regulation (Megareg) Compliance Testing of carcasses for E. coli Biotype I (9 CFR Part 310, §310.25), effective June 1997. (all Beef Slaughter plants)

Implementation of SSOP (Sanitation Standard Operating Procedures, 9 CFR, Part 416, §416.11 - §416.17), effective January 26, 1997 for all plants.

Implementation of HACCP Systems (9 CFR, Part 417, §417.1 - §417.8), effective January 27, 1998 for plants with greater than 500 employees.

Implementation of HACCP Systems (9 CFR, Part 417, §417.1 - §417.8), effective June 1, 1998 for smaller plants noted separately by “*”

Testing of carcasses and/or ground beef for Salmonella as conducted by USDA in accordance with §310.25.

Federal Register Docket 00-022N, dated 10/7/02 (E. coli O157:H7 Reassessment) Reassessment of HACCP plans for E. coli O157:H7 in accordance with the Notice 22-04, dated 10/7/02 conducted in all Company Name beef plants effective 12/6/02.

Completion of annual reassessment of HACCP plans in accordance with 9CFR 417.4 (a) (3) effective January each calendar year. This reassessment included review and verification of adequacy of the HACCP plans in addressing E. coli O157:H7.

Directive 6420.2 – Issued 3/31/04 CCP’s in place and effect for zero tolerance requirements for head meat, cheek meat and weasand meat for all plants effective 5/17/04. Note: Zero Tolerance on carcasses has been in place as a CCP since the implementation of HACCP in 1998.

Directive 10,010.1 – revised 3/31/04 Labeling USDA approval for the following label disclaimer/instructional statements are available on site at the producing est.:

o For Cooking Only o Lot Tested and Found Negative for ECH7”

Disposition CCP’s All materials that are tested for E. coli O157:H7 that are not negative are addressed within the HACCP plans under a product disposition CCP.

These materials are controlled, relabeled (when applicable) with the statement, “For Cooking Only” and are cooked or otherwise disposed of to inedible or rendering.

Records reflect appropriate disposition of affected material.

February 2006

15

Testing for E. coli O157:H7 Carcasses – Daily validation testing for E. coli O157:H7 is conducted at each beef slaughter plant. This has been in place and effect since 2000. Carcasses are sampled at the same sites as listed in 9CFR 310.25 for E. coli Biotype I and are retained pending results. Beef Materials Destined For Non-Company Name Grinding In accordance with the intended use described in the plants’ Raw Not Ground HACCP plans (including trim and some variety meats harvested in slaughter), all materials destined for raw ground use are subjected to a statistically based sampling plan1 for E. coli O157:H7. All boxed materials that are “Lot tested and found to be negative for E. coli O157:H7” are labeled with that statement. Combo’d trim does not carry this on the label as combo’d trim materials are tested per customer order and a Certificate of Analysis, (COA), specific to those combos is provided to the contracted end user. Since boxes may be broken down into smaller ship units by a primary (or secondary or tertiary, etc.) distributor, we deemed it necessary to label the individual box so the ultimate end user is aware that the materials were part of sampling lot that tested negative for E. coli O157:H7. These labeling components are addressed in our HACCP plan as they are an integral part of the intended use. Ground Beef All raw materials destined for grinding in the plants listed in this document are pre-tested1 and negative for E. coli O157:H7 prior to grinding.

External sources of trim raw material must have a validated carcass intervention for E. coli O157:H7 in place and a copy of that compliance is maintained on file at the receiving establishment.

External sources of raw material must meet Company Name requirements for outside vendors including but not limited to: validated HACCP systems, 3rd party food safety/GMP audits, E. coli O157:H7 testing programs that meet or exceed 95% confidence for detection capability.

Certificate of Analysis (COA’s) received for all outside materials sent to grind. Laboratory Verification Testing Verification of E. coli O157:H7 lab methods is routinely performed at each Company Name Laboratory in conjunction with the American Proficiency Institute Microbiological Performance Evaluation Program.

February 2006

16

HACCP Critical Control Points in place and in effect at present include: HACCP Category

Critical Control Points

Steam Cabinet operational and functional with regard to ambient temperature and transit time to deliver a minimum of 160°F to the carcass surface to address E. coli O157:H7. Zero Tolerance for feces, ingesta and milk on carcasses. Carcass Chilling to reduce the surface down to 45°F or less within 24 hours to control microbial growth.

Slaughter

Disposition CCP to assure proper disposition of any carcasses that do not test negative for E. coli O157:H7. Pre-cut Carcass Surface Temperature below 45°F to control microbial growth.

Raw Not Ground – Trim

Disposition CCP to assure proper disposition of any products that do not test negative for E. coli O157:H7. Zero Tolerance for feces, ingesta and milk on head, cheek and weasand meat. Chilling to reduce the surface down to 45°F or less within 24 hours to preclude microbial growth.

Raw Not Ground – Variety Meats

Disposition CCP to assure proper disposition of any products that do not test negative for E. coli O157:H7 Inbound Raw Material Temperature < 45°F to preclude microbial growth Functioning metal detector, verified for timing and sensitivity at the start of operations.

Raw Ground

Disposition CCP to assure proper disposition of any products that do not test negative for E. coli O157:H7

A CCP is “A point, step, or procedure in a food process at which control can be applied and, as a result, a food safety hazard can be prevented, eliminated, or reduced to acceptable levels”2 It should be clearly understood that these CCP’s are in place to accomplish just that for E. coli O157:H7; control, eliminate or reduce to an acceptable level. The acceptable level for E. coli O157:H7 is undetectable. Best Practices/Good Manufacturing Practices In addition to the CCP’s, the following practices are utilized in our beef slaughter operations.

Steam Vacuums – are located strategically throughout the slaughter floor and are used on pattern mark areas.

Pre-Evisceration Cabinet System (PECS) – eligible beef carcasses are treated with up to 2.5% organic acid pre-evisceration.

Anti microbial spray – carcasses are treated with an anti microbial spray of organic acid or acidified sodium chlorite after the Steam Cabinet. Heads are treated with an organic acid application immediately after the head wash, prior to USDA Inspection.

February 2006

17

Verification In accordance with the facilities’ HACCP plans, all CCP’s have been validated and are verified at the specified frequencies in the HACCP plan in accordance with 9CFR 417.4.

Company Name is audited on an annual basis by an independent third party auditor. That audit encompasses both regulatory compliance (HACCP, SSOP, 10,010.1, etc.) and good manufacturing practices. A summary matrix of audit scores is available upon request.

Customer Notification Company Name plants have a recall plan on file that includes notification to affected customers of any product that may be adulterated or misbranded.

Last, the Company Name plants listed below are federal establishments and operate under the regulatory requirements promulgated in Title 9 of the Code of Federal Regulations. By dint of the Mark of Inspection, we are obligated to adhere to all applicable requirements contained therein. COMPANY NAME BEEF PLANTS

EST. Location Comments Est. ### City, ST Est. ### City, ST Est. ### City, ST Est. ### City, ST Est. ### City, ST

February 2006

18

Best Practices: Appendix B Standard Operating Procedures for Cleaning and Sanitizing Injector Assembly: Example I Purpose: To effectively clean and sanitize the injector assembly Program: At the end of each production day, production personnel will perform the following tasks: Injector Needles

1. Open the needle assembly and inspect for cleanliness. If any residual brine residue remains, rinse the housing and needles completely.

2. Remove all needles and carefully place the needles in a clean meat lug that has not been used during that day’s production.

3. Rinse housing after needles are removed to ensure that all areas of the head are free of visible residue.

4. Add clean & soak chemicals to the meat lug to a level that completely submerges all needles in the container. Needles must soak for a minimum of 6 hours or as recommended by the sanitation chemical manufacture. If necessary, use a second set of cleaned and sanitized needles to ensure adequate cleaning while meeting production requirements.

5. After the needles have soaked for a minimum of 6 hours, each needle must be “blown out” with clean air before being replaced in the injector assembly.

6. Once clean needles have been placed in the injector assembly, they must be sanitized and rinsed before being used in production.

Cleaning and Sanitizing Solutions

1. The composition of the cleaning solution used for nightly cleaning can be used for cleaning the needles and assembly parts unless other solutions have been validated for efficacy.

2. The cleaning and sanitizing chemicals should be rotated periodically. 3. The amount of chemical solution used and the soak time for cleaning should be

documented, and verified periodically, e.g., quarterly. Monitoring & Verification: QA and Production Management will monitor the cleaning and sanitizing process during cleanup hours to ensure proper compliance. QA will verify sanitation daily during pre-operational inspections. An authorized person verifies solution composition and chemical strength nightly. Microbial sampling of cleaned and sanitized surfaces will be conducted as per the documented microbiological sampling schedule.

February 2006

19

Standard Operating Procedure Clean In Place System Cleaning: Example II PURPOSE: To minimize bacterial growth. PROGRAM: A CIP cleaning solution will be ran through the injection process to ensure proper cleaning of the injection process. PROCEDURE:

1. Drain all brine material from lines, pumps, and tanks. During the draining process production personnel will continue to rinse all six tanks with potable water until all visible brine residue has disappeared.

2. Fill the two mixing tanks (# 3 & # 6) with 200 Gal. of cold potable water each. 3. Flush 100 Gal. from the line 1 mixing tank (#3) to each of the rear holding tanks (#2 & #1). 4. Flush 100 Gal. from the line 2 mixing tank (#6) to each of the rear holding tanks (#5 & #4). 5. Flush all water from all holding tanks through the CIP system and a minimum of 50 Gal.

through each of the injectors (line 1 and line 2). 6. Fill mixing tanks( #3) and (#6) again with 200 Gal. of cold potable water and add appropriate

amount of the approved CIP cleaning solution. 7. Mix thoroughly. 8. Flush 100 Gal. of the mixed cleaning solution from the line 1 mixing tank (#3) to each of the

rear holding tanks (#2 & #1). 9. Flush 100 Gal. of the mixed cleaning solution from the line 2 mixing tank (#6) to each of the

rear holding tanks (#5 & #4). 10. Flush all cleaning solution from all holding tanks through the CIP system pumping from each

tank a minimum of 5 minutes. 11. A minimum of 50 Gal. will be pumped from one of the holding tanks of each line through its

designated injector (line 1 and line 2). 12. Fill the two mixing tanks (# 3 & # 6) with 200 Gal. of cold potable water each. 13. Flush 100 Gal. from the line 1 mixing tank (#3) to each of the rear holding tanks (#1 & #2). 14. Flush 100 Gal. from the line 2 mixing tank (#6) to each of the rear holding tanks (#5 & #4). 15. Flush all water from all holding tanks through the CIP system and a minimum of 50 Gal.

through each of the injectors (line 1 and line 2). The currently used cleaning solution is STERIS brand Process Klenz alkaline cleaner used at 2.5% by volume. (5 gallons Process Klenz mixed with 200 gallons potable water.) CORRECTIVE ACTION: Production will not be allowed to start until CIP cleaning has taken place. RELATED FORMS: CIP System Cleaning Verification Process Check MATERIALS NEEDED: Steris brand process klenz alkaline cleaner. FREQUENCY: Daily MONITORED BY: QA and Production Management will routinely monitor to ensure proper

compliance.

General Manager Date

QA Manager

Date

February 2006

20

Standard Operating Procedure Clean In Place System Sanitizing: Example III PURPOSE: To minimize bacterial growth. PROGRAM: A CIP Sanitizing solution will be ran through the injection process to ensure proper

cleaning of the injection process.

PROCEDURE:

1. Fill the two mixing tanks (# 3 & # 6) with 200 Gal. of cold potable water each. 2. Flush 100 Gal. from the line 1 mixing tank (#3) to each of the rear holding tanks (#2 & #1). 3. Flush 100 Gal. from the line 2 mixing tank (#6) to each of the rear holding tanks (#6 & #4). 4. Flush all water from all holding tanks through the CIP system and a minimum of 50 Gal.

through each of the injectors (line 1 and line 2). 5. Fill mixing tanks #3 and #6 again with 200 Gal. of cold potable water and add appropriate

amount of the approved CIP sanitizing solution. 6. Mix thoroughly. 7. Flush 100 Gal. of the mixed sanitizing solution from the line 1 mixing tank (#3) to each of the

rear holding tanks (#2 & #1). 8. Flush 100 Gal. of the mixed sanitizing solution from the line 2 mixing tank (#6) to each of the

rear holding tanks (#5 & #4). 9. Flush all sanitizing solution from all holding tanks through the CIP system pumping from

each tank a minimum of 5 minutes. 10. A minimum of 50 Gal. will be pumped from one of the holding tanks of each line through its

designated injector (line 1 and line 2). 11. Fill the two mixing tanks (# 3& # 6) with 200 Gal. of cold potable water each. 12. Flush 100 Gal. from the line 1 mixing tank (#3) to each of the rear holding tanks (#2 & #1). 13. Flush 100 Gal. from the line 2 mixing tank (#6) to each of the rear holding tanks (#5 & #4). 14. Flush all water from all holding tanks through the CIP system and a minimum of 50 Gal.

through each of the injectors (line 1 and line 2).

The currently used cleaning solution is STERIS brand Process LCS liquid chlorinating sanitizer used at .25 ounce per gallon. (50 ounces mixed with 200 gallons potable water.) Chlorine Days Monday, Wednesday, Friday, Saturday, Sunday. Quat Days: Tuesday, Thursday.

CORRECTIVE ACTION: Production will not be allowed to start until sanitizing has taken place.

RELATED FORMS: NA

MATERIALS NEEDED: Quat or Chlorine FREQUENCY: Daily MONITORED BY: QA and Production Management will routinely monitor to ensure proper compliance.

General Manager

Date

QA Manager

Date

February 2006

21

Standard Operating Procedure Operational Cleaning of Injector Reservoir In-Line Filters: Example IV PURPOSE: To minimize bacterial growth. PROGRAM: Injection filters will be cleaned on a regular basis to ensure the injectors operate at an

optimal level. PROCEDURE:

1. Remove the machine side in-line final filter by rotating its holding cylinder to the vertical position where it will latch against the wall of the reservoir.

2. From this position the end cap can be threaded back and spun out of the way so the filter may be removed for cleaning.

3. Remove filter and clean with tempered water of sufficient pressure to remove any built up residue.

4. Replace filter into its holding cylinder and thread back its end cap to secure filter in the cylinder.

5. Return filter assembly to the horizontal position inside the reservoir tank. 6. Remove the off side in-line final filter by rotating its holding cylinder to the vertical position

where it will latch against the wall of the reservoir. 7. From this position the end cap can be threaded back and spun out of the way so the filter may

be removed for cleaning. 8. Remove filter and clean with tempered water of sufficient pressure to remove any built up

residue. 9. Replace filter into its holding cylinder and thread back its end cap to secure filter in the

cylinder. 10. Return filter assembly to the horizontal position inside the reservoir tank.

CORRECTIVE ACTON: NA RELATED FORMS: NA MATERIALS NEEDED: Tempered Water FREQUENCY: Operational cleaning of injector reservoir filters should be conducted on the hourly

basis in order to maintain consistent pump settings.

NOTE: Each employee who handles injector equipment must change gloves before and after as well as clean any additional utensils needed for the tasks. This ten-step process will be used for the reservoir tanks of both line one and line two injectors. If filters are cleaned one at a time than the injector does not need to be shut down for this SOP. MONITORED BY: QA and Production Management will routinely monitor to ensure proper compliance.

General Manager: Date:

QA Manager:

Date:

February 2006

22

Best Practices: Appendix C Decontamination Interventions for Primals, Subprimals, Trim and Ground Meat

Decontamination Interventions

Intervention Effectiveness in Lab setting Effectiveness in Field / Plant Regulatory Status

MECHANICAL TREATMENT

Irradiation Widely studied. Effective in reducing pathogens at varying levels depending on dose.

Effective, but control of dose is critical to minimize effects on organoleptic factors.

Approved, labeling required

Trimming CSU study indicates surface trimming is as effective as certain chemical treatments. 1.1 log CFU/cm2 reduction (inoculated with 3.7 log CFU/cm2).

Effective and implemented widely

Not a limitation

Steam Initial results are limited, but may have an effect.

Unknown Unknown

Hot water wash CSU study indicates a significant log reduction. 1.0 log CFU/cm2 reduction (inoculated with 3.6 log CFU/cm2).

Unknown Unknown

February 2006

23

Intervention Effectiveness in Lab setting Effectiveness in Field

/ Plant Regulatory Status

CHEMICAL TREATMENT

Acidified Sodium Chlorite

Company data 2.9 log reduction of E. coli O157. 2.0 log reduction of E. coli (generic). KSU 2-3 log CFU/cm2 reduction of APC. ABC Research found up to a 0.63 log reduction of E. coli O157 on inoculated subprimals

Initial trials show approximately a 2 log reduction of APC.

Approved, however weight gain over 0.5% must be labeled.

Lactic Acid CSU data supports 2.5% LA @ 55°C resulted in 1.0 log CFU/cm2, while 5.0% LA @ 55°C resulted in a 1.1 log CFU/cm2 (inoculated at 3.6 and 3.5 log CFU/cm2, respectively).

Unknown. 0.4% by weight, of a 2.5% solution was not effective.

Pending approval at 2.5% and 5.0% levels.

Acidified Calcium Sulfate

Company trials are encouraging.

Unknown Not approved in Beef trim

CPC Company trials show significant log reductions.

Unknown Not approved in Beef trim, residual levels cited as concern.

Peroxyacetic acid ABC Research data found .63 - .71 log reduction of E. coli O157:H7 on inoculated subprimals.

Unknown Approved

Citric Acid Laboratory trials show promise.

Unknown Approved

February 2006

24

Intervention

Effectiveness in Lab setting Effectiveness in Field

/ Plant Regulatory Status

BIOLOGICAL

Lactoferrin CSU study indicates that Lactoferrin applied to inoculated subprimals allowed 4.6 log less growth of E. coli O157:H7. Additionlly 5.0% lactic acid used in combination with activated Lactoferrin at 55°C resulted in 0.9 log CFU/cm2 reduction (inoculated at 3.5 log CFU/cm2).

Unknown Approved for Carcasses and parts Directive 7120.1

Lactobacillus acidophilus

TTU study demonstrated a 90% reduction in E. coli O157:H7 and a 99.9% reduction in Salmonella

Unknown Working on petition

- 25 -

Best Practices: Appendix D Studies on the Antimicrobial Properties of Tenderizing Pickle Solution Preliminary Report

September 10, 2003

Study I

Objective: To determine antimicrobial properties of a pickle solution used in tenderizing whole muscle cuts

Composition of pickle solution: A typical pickle solution will contain phosphate, salt and flavorings. The solution used in this study contained a proprietary formula based on in finished products, e.g., 0.5%.

Measurement of the antimicrobial effect: The antimicrobial effect of the pickle solution was measured using a micro-titer assay (i.e., providing minimum inhibitory concentrations) and traditional laboratory plating procedures. Results: Using micro-titer assays, initial experiments determined that the pickle solution reduced the concentrations of E. coli O157:H7 and Salmonella by at least 2 logs (100-fold). In follow-up experiments, direct inoculation of pickle solution with a cocktail of 3 E. coli O157:H7 strains and 3 Salmonella strains at levels near 106 per mL resulted in complete lethality for all pathogens after 30 minutes of exposure (the first measurement time interval after the zero time measurement).

In a laboratory setting using traditional microbiological techniques, the antimicrobial properties of the pickle solution were determined. Pickle solution was inoculated to 1.73 logs per mL with E. coli O157:H7 and stored at room temperature (~73°F) or under refrigeration (37°F). No E. coli O157:H7 were recovered from the pickle solution after 2 hours at room temperature and after 24 hours under refrigerated conditions.

Storage temp Time Room Refrigerator0 min Positive Positive 30 min Positive Positive 1 hour Positive Positive 2 hour Negative Positive 4 hour Negative Positive 24 hour Negative Negative

These data represent the results of a single study using inoculated organisms, and should not be extrapolated to all situations. The storage temperature and times, while different for room temperature versus refrigerated, simply indicate that the brine solution may exhibit inhibitory properties against E. coli O157:H7. However, further research would be needed to confirm that this is the case, and multiple variables may be contributing to this effect.

- 26 -

Next steps: Additional validation work will be repeated with meat extract added to evaluate effects of meat components on bactericidal activity and with inoculated meat exposed to the pickle solution. Study II Objective: To determine the prevalence of E. coli O157:H7 in injection solutions used to enhance various beef products. Sampling Procedures: One-quart samples of injection solutions were taken from the brine return, before the brine entered the reservoir for recycling with fresh solution, before filtration. Samples were collected at least 20 minutes into production, with each sample set of three samples spaced throughout the scheduled production run. Samples were then sealed and sent to the laboratory for testing. Results: In total, 19 sample sets (57 samples) were collected through July and August 2003. All samples (Table 1) tested negative for the presence of E. coli O157:H7. Preliminary investigation into the recovery of E. coli O157:H7 that were inoculated into brine samples indicated that the organism could be recovered from the brine solution, if present.

- 27 -

Table 1. Injection Solution Results for Study II

Date Meat Cut E. coli O157:H7 Result 1

E. coli O157:H7 Result 2

E. coli O157:H7 Result 3

29-Jul-03 Flat NEG NEG NEG 29-Jul-03 Flat NEG NEG NEG 29-Jul-03 Ribeye NEG NEG NEG 30-Jul-03 Capoff Inside NEG NEG NEG 30-Jul-03 Flat NEG NEG NEG 30-Jul-03 Ribeye NEG NEG NEG 31-Jul-03 Ribeye NEG NEG NEG

05-Aug-03 Capoff Inside NEG NEG NEG 05-Aug-03 Ribeye NEG NEG NEG 05-Aug-03 Capoff Inside NEG NEG NEG 06-Aug-03 Ribeye NEG NEG NEG 06-Aug-03 Capoff Inside NEG NEG NEG 06-Aug-03 Inside NEG NEG NEG 11-Aug-03 Ribeye NEG NEG NEG 13-Aug-03 Ribeye NEG NEG NEG 20-Aug-03 Inside NEG NEG NEG 20-Aug-03 Capoff Inside NEG NEG NEG 20-Aug-03 Capoff Inside NEG NEG NEG 20-Aug-03 Inside NEG NEG NEG

- 28 -

Appendix B Federal Register Notice: HACCP Plan Reassessment for Mechanically Tenderized Beef Products Federal Register: May 26, 2005 (Volume 70, Number 101) Rules and Regulations Page 30331-30334 From the Federal Register Online via GPO Access [wais.access.gpo.gov] DOCID:fr26my05-2 ----------------------------------------------------------------------- DEPARTMENT OF AGRICULTURE Food Safety and Inspection Service 9 CFR Part 417 Docket No. 04-042N HACCP Plan Reassessment for Mechanically Tenderized Beef Products AGENCY: Food Safety and Inspection Service, USDA. ACTION: Compliance with the HACCP system regulations and request for comments. ----------------------------------------------------------------------- SUMMARY: The Food Safety and Inspection Service (FSIS) is publishing this notice to inform establishments that produce mechanically tenderized beef products that their next annual HACCP plan reassessment for these products must take into account the fact that there have been three relatively recent Escherichia coli (E. coli) O157:H7 outbreaks associated with consumption of mechanically tenderized beef. This requirement applies to HACCP plan reassessments for raw and cooked mechanically tenderized beef products, including such products that are injected with marinade (or “enhanced” products). One outbreak that was associated with consumption of mechanically tenderized beef occurred in August 2000, one in June 2003, and one in August 2004. The occurrence of these outbreaks represents a change that would affect the hazard analysis and could alter the HACCP plans of establishments that produce mechanically tenderized beef products. Therefore, establishments that produce such products should consider the significance of the outbreaks and ensure that their HACCP plans adequately address relevant biological hazards, particularly E. coli O157:H7. If an establishment that produces mechanically tenderized beef products has already considered the significance of the three outbreaks as part of a HACCP plan reassessment, it need not repeat this effort. An establishment that has already conducted its 2005 reassessment for mechanically tenderized beef products and has not yet considered the significance of the three outbreaks as part of a HACCP plan reassessment should do so as part of its 2006 annual HACCP plan reassessment. FSIS invites comments on this notice. DATES: The Agency must receive comments by July 25, 2005. ADDRESSES: FSIS invites interested persons to submit comments on this notice. Comments may be submitted by any of the following methods:

- 29 -

Mail, including floppy disks or CD-ROM's, and hand- or courier-delivered items: Send to Docket Clerk, U.S. Department of Agriculture, Food Safety and Inspection Service, 300 12th Street, SW., Room 102, Cotton Annex, Washington, DC 20250. All submissions received must include the Agency name and docket number 04-042N. All comments submitted in response to this notice, as well as research and background information used by FSIS in developing this document, will be available for public inspection in the FSIS Docket Room at the address listed above between 8:30 a.m. and 4:30 p.m., Monday through Friday. The comments also will be posted on the Agency's Web site at http://www.fsis.usda.gov/regulations_&_policies/2005_Notices_Index/index.asp FOR FURTHER INFORMATION CONTACT: Lynn Dickey, Director, Regulations and Petitions Policy Staff, Office of Policy, Program, and Employee Development, FSIS, U.S. Department of Agriculture, 1400 Independence Avenue, SW., Room 405, Cotton Annex, Washington, DC 20250-3700, (202) 720-5627. SUPPLEMENTARY INFORMATION: Background FSIS administers a regulatory program under the Federal Meat Inspection Act (FMIA) (21 U.S.C. 601 et seq.) to protect the health and welfare of consumers by preventing the distribution in commerce of meat products that are adulterated or misbranded. In pursuit of its goal of reducing the risk of foodborne illness from meat products to the maximum extent possible, FSIS issued final regulations on July 25, 1996, that mandated the development and implementation of Pathogen Reduction and Hazard Analysis and Critical Control Point (HACCP) Systems by federally inspected establishments (61 FR 38806). These regulations require that federally inspected establishments take preventive and corrective measures at each stage of the food production process where food safety hazards occur. The HACCP regulations (9 CFR 417.2(a)) require establishments to conduct a hazard analysis to determine what food safety hazards are reasonably likely to occur in the production process of particular products and to identify the preventive measures that the establishment can apply to control those hazards. Section 417.2(a)(1) of the HACCP regulations states that a food safety hazard that is reasonably likely to occur is one for which a prudent establishment would establish control measures because the hazard historically has occurred, or because there is a reasonable possibility that it will occur in the particular type of product being processed, in the absence of those controls. Whenever a hazard analysis reveals that one or more hazards are reasonably likely to occur in the production process, the regulations require that the establishment develop and implement a written HACCP plan that includes specific control measures for each hazard identified (9 CFR 417.2(b)(1) and (c)). Section 417.4(a)(3) of the regulations requires that every establishment reassess the adequacy of its HACCP plan at least annually and whenever any changes occur that could affect the hazard analysis or alter the HACCP plan. Because the outbreaks discussed in this notice are the first known outbreaks associated with consumption of mechanically tenderized beef products, and because there have been three outbreaks, the occurrence of these E. coli O157:H7 outbreaks is a change that could affect the hazard analysis or alter the HACCP plans for such products.

- 30 -

FSIS' Actions To Address E. coli O157:H7 In 1994, FSIS notified the public that raw ground beef products contaminated with E. coli O157:H7 are adulterated within the meaning of the FMIA (21 U.S.C. 601(m)(1)) unless the ground beef is further processed to destroy this pathogen. The public health risk presented by beef products contaminated with E. coli O157:H7 is not limited, however, to raw ground beef products. In the January 19, 1999, Federal Register, FSIS explained that if non-intact beef products, including beef that has been mechanically tenderized by needling or cubing, are found to be contaminated with E. coli O157:H7, they must be processed into ready-to-eat product, or they would be deemed to be adulterated (64 FR 2803). In the October 7, 2002, Federal Register, FSIS informed manufacturers of raw beef products, including manufacturers of mechanically tenderized raw beef products, that they were required to reassess their HACCP plans, in light of certain scientific data on E. coli O157:H7, to determine whether E. coli O157:H7 contamination was a hazard reasonably likely to occur in their production process (67 FR 62325). The data discussed in that Federal Register provided evidence that E. coli O157:H7 was more prevalent than was thought before the data became available, and that this pathogen may be a hazard reasonably likely to occur at all stages of handling raw beef products (67 FR 62328). Although FSIS previously informed establishments producing mechanically tenderized raw beef products that they were required to reassess their HACCP plans based on the availability of specific scientific data related to the prevalence of E. coli O157:H7, only one outbreak (the 2000 outbreak discussed below) associated with such product had occurred at the time these establishments conducted their HACCP plan reassessments. In addition, FSIS has not previously required establishments that produce cooked mechanically tenderized beef products to reassess their HACCP plans to ensure that these HACCP plans adequately address biological hazards, particularly E. coli O157:H7. E. coli O157:H7 Outbreaks Associated With Mechanically Tenderized Beef In August 2004, the Colorado Department of Public Health and Environment (CDPHE) confirmed by culture tests four E. coli O157:H7 cases with matching Pulse-Field Gel Electrophoresis (PFGE) patterns in the Denver, Colorado, metropolitan area. The CDPHE determined that the individuals who became ill in this outbreak ate a tenderized, marinated beef steak product at four separate locations of a national restaurant chain. The CDPHE conducted an age and sex-matched case-control study that showed that consumption of this particular steak product was the only commonality of those who became ill. Although the CDPHE did not test product for E. coli O157:H7, the case-control study provided strong evidence that consumption of this product was associated with the outbreak. The producing establishment voluntarily recalled approximately 406,000 pounds of product. Information on this recall can be found on the FSIS web page (http://www.fsis.usda.gov), through the “FSIS Recalls” link, under recall case number 033-2004. In June 2003, State health departments confirmed by culture tests eleven E. coli O157:H7 cases in five States: Seven cases in Minnesota, one case in Michigan, one case in Kansas, one case in Iowa, and one case in North Dakota. The cases were a two-enzyme PFGE pattern match.

- 31 -

Based on the food intake histories of the persons who became ill, the State health departments epidemiologically linked all cases to a tenderized beef steak product (a boneless beef filet bacon-wrapped steak product injected with marinade). The Michigan Department of Agriculture Laboratory analyzed one sample of product associated with the outbreak and found it positive for E. coli O157:H7. The Minnesota Departments of Agriculture and Health Laboratories analyzed five samples of the product associated with the outbreak and found them positive for E. coli O157:H7. The product samples analyzed matched the two-enzyme PFGE pattern of the outbreak cases. The food histories of the persons who became ill, and the fact that the PFGE patterns in the product samples analyzed matched the outbreak cases, provided strong evidence that consumption of the tenderized steak product was associated with the outbreak. At the time of the outbreak, the establishment that produced the tenderized beef steak product was thoroughly breaking down, cleaning, and sanitizing its injectors only once per week. The establishment subsequently documented a revised plan in its Sanitation Standard Operating Procedures (SOPs) to break down, clean, and sanitize its injection needles, tenderizing needles, and associated processing equipment on a daily basis. Also, after changing its Sanitation SOPs, the establishment incorporated in its production process an antimicrobial treatment of the product prior to the tenderizing and marinating process. The establishment that had produced the product linked to the 2003 outbreak voluntarily recalled approximately 739,000 pounds of product. Information on this recall can be found on the FSIS web page (http://www.fsis.usda.gov), through the ``FSIS Recalls'' link, under recall case number 028-2003. From information obtained from the Centers for Disease Control and Prevention and State health departments, FSIS identified another outbreak that was associated with the consumption of mechanically tenderized steaks. In August 2000, the Michigan Department of Community Health (MDCH) laboratory identified two human isolates of a distinct strain of E. coli O157:H7 with matching PFGE patterns. This strain had not been previously found in Michigan. Local health departments obtained case histories from both of the persons who had become ill. The only similar possible exposure to the pathogen for the two individuals was a steak meal consumed by each on August 12, 2000, at different locations of a local restaurant steakhouse chain. Each individual had eaten a sirloin steak cooked to order with a red or pink center. The sirloin steaks were needle tenderized. The investigation of this matter suggested that the sirloin steak eaten by each person was likely the common source of the distinct strain of E. coli O157:H7 associated with these individuals' illnesses. The fact that both of the ill persons consumed an identical restaurant meal on the same day and had the onset of symptoms on the same date indicated that consumption of the tenderized beef steak product was associated with the illnesses. As a result of this investigation, the supplier of the steaks agreed to procedural changes in its operations, including sanitizing the needle-piercing machine used and testing its beef for E. coli O157:H7. Reassessment in Response to Outbreaks The E. coli O157:H7 outbreaks discussed above that were associated with consumption of mechanically tenderized beef products are events that could alter the hazard analysis, and

- 32 -

ultimately the HACCP plan, of any establishment that produces mechanically tenderized beef products. Therefore, as part of their next annual HACCP plan reassessment for such products, establishments that produce raw or cooked mechanically tenderized beef products (with or without marinade), hereafter referred to as mechanically tenderized beef products, must take into account the E. coli O157:H7 outbreaks discussed above to determine whether their HACCP plans for these products adequately address biological hazards, particularly E. coli O157:H7. Establishments that produce mechanically tenderized beef products that have already taken these three outbreaks into account in a HACCP plan reassessment for these products are not required to consider these outbreaks in their next annual HACCP plan reassessment, provided the establishments have evidence of their reassessment in their hazard analysis or HACCP plans, or a record of reassessment, and make this evidence available to FSIS inspection program personnel. When conducting a reassessment that takes these outbreaks into account to determine whether HACCP plans for mechanically tenderized beef products adequately address biological hazards, E. coli O157:H7 in particular, establishments may need to evaluate the adequacy of any E. coli O157:H7 interventions applied to the products' source materials. If they have not already done so, establishments producing mechanically tenderized beef products may wish to consider implementing purchase specifications that require that incoming product has been treated to eliminate or reduce E. coli O157:H7 to an undetectable level. If establishments producing mechanically tenderized beef products require their suppliers to meet such purchase specifications, they should also ensure that their suppliers actually meet these purchase specifications. Establishments could incorporate such purchase specifications in their HACCP plans, in their Sanitation SOPs, which FSIS has recognized as prerequisites for HACCP, or in other prerequisite programs. Establishments producing mechanically tenderized beef products might also consider applying an allowed antimicrobial agent to the surface of the product prior to processing or tenderization. FSIS has made available on its web site a document entitled, “Guidance on Ingredients and Sources of Radiation Used to Reduce Microorganisms on Carcasses, Ground Beef, and Beef Trimmings.” This document provides guidance on the use of antimicrobials on beef products. A link to the document is found with the October 7, 2002, Federal Register notice entitled, “E. coli O157:H7 Contamination of Beef Products,'' on the “Interim and Final Rules” page of FSIS' web site http://www.fsis.usda.gov/Regulations_&_Policies/2002_Interim_&_Final_Rules_Index/index.asp. When conducting their reassessment, establishments producing mechanically tenderized beef products should consider the number of times tenderizers pass through the product. In addition, they should evaluate the adequacy of their sanitation procedures for mechanical tenderizers, including needles, and for associated processing equipment, including reservoirs and piping associated with the tenderizing and enhancing operations. Because tenderizers pass through the product, they may introduce biological hazards, including E. coli O157:H7, into the interior of the product. Therefore, sanitation procedures are particularly important in the production of mechanically tenderized beef products. Thus, Sanitation SOPs, other prerequisite programs, or HACCP plans should address procedures that ensure that all mechanical tenderizers and

- 33 -

associated processing equipment are cleaned on a regular basis to minimize the potential for translocating E. coli O157:H7 from the exterior surface of the product to the interior and to minimize the potential for cross contamination within and among lots of production. Establishments producing raw, mechanically tenderized beef products might also consider including cooking instructions, in addition to required safe handling instructions (e.g., cook to at least 140 degrees F), on packages of raw, mechanically tenderized beef products, or other labeling, to ensure that these products are cooked adequately to destroy E. coli O157:H7, should it be present. Such cooking instructions, or other labeling, however, cannot serve as a control or critical control point (CCP) to address E. coli O157:H7 in the production process of raw, mechanically tenderized beef products. FSIS itself is considering requiring that raw, mechanically tenderized products be labeled to indicate that they have undergone mechanical tenderization, that the product is non-intact, and that it should be cooked to an adequate internal temperature to destroy any pathogens that may have been translocated from the surface of the product to the interior. Although the Federal meat and poultry products inspection regulations require that any marinade injected in a product be listed as an ingredient on the product's label, they do not require that product be labeled to indicate that it has been mechanically tenderized, and it is not possible to discern visually whether product has been mechanically tenderized. Finally, establishments producing cooked mechanically tenderized beef products may need to consider whether their cooking procedures are adequate to destroy E. coli O157:H7, should it be present. Information on a study concerning the effects of cooking on E. coli O157:H7 in blade tenderized steaks is included in the following section of this document. This section also includes information on published studies concerning bacteria other than E. coli O157:H7 in the interior of mechanically tenderized beef. In addition, it provides information on guidelines developed by industry associations regarding pathogen control in mechanically tenderized and enhanced beef products. Research and Guidance on the Production of Mechanically Tenderized Beef Products FSIS asked the National Advisory Committee on Microbiological Criteria for Foods (NACMCF) to answer several questions with regard to E. coli O157:H7 in mechanically tenderized beef. NACMCF met on August 3, 2001, and January 23, 2002, to discuss these questions. A report on NACMCF's responses to FSIS questions is available on the Internet at http://www.fsis.usda.gov/OPHS/NACMCF/2002/rep_blade1.htm. The report is entitled, “Escherichia coli O157:H7 in Blade-tenderized, Non-intact Beef'' (updated September 9, 2002). FSIS asked NACMCF whether non-intact, blade tenderized beef steaks present a greater risk to consumers from E. coli O157:H7 compared to intact beef steaks, if prepared similarly to intact beef steaks. Based on information from a Master's thesis (Sporing, 1999), NACMCF concluded that non-intact, blade tenderized beef steaks do not present a greater risk to consumers from E. coli O157:H7 than intact beef steaks if the blade tenderized beef steak is oven broiled and cooked to an internal temperature of 140 degrees F or above. However, NACMF concluded that blade tenderized beef steaks present a greater risk from E. coli O157:H7 than intact beef steaks,

- 34 -

particularly to immuno-compromised individuals, when served very rare with cold spots (less than 120 degrees F internal temperature). FSIS also asked NACMCF whether non-intact, blade tenderized beef roasts present a greater risk to consumers from E. coli O157:H7 compared to intact beef roasts, if prepared similarly to intact beef roasts. NACMCF concluded that there were insufficient data to answer this question adequately. Finally, FSIS asked NACMCF whether available evidence supports the need for a labeling requirement to distinguish between intact and non-intact products in order to enhance public health protection. Again, NACMCF concluded that there were insufficient data to make a response to this question at the time the committee met. The NACMCF report lists research needs at the end of the document. Participants at the 2004 Conference of Food Protection discussed the handling of blade tenderized steaks at retail facilities and restaurants. Participants discussed the fact that blade tenderized products typically are not labeled to indicate that the products have been tenderized. They considered data from the Master's thesis that NACMCF reviewed (Sporing, 1999). These data showed that 3 to 4 percent of the surface bacterial load of blade tenderized beef steaks is transferred to the interior of the product. According to the thesis, among three methods of preparation--oven cooking, commercial grilling, and skillet cooking--skillet cooking provided the least effective and most variable reduction in E. coli O157:H7. Participants in the 2004 Conference for Food Protection recommended that the Food and Drug Administration (FDA) and USDA work together to develop guidance for retail facilities and restaurants on the safe cooking of blade tenderized steaks and other non-intact steaks. The participants recommended that this guidance be included in the Annex of the Food Code, and that FDA and USDA submit this guidance at the 2006 Conference for Food Protection. FDA and USDA intend to prepare this guidance. Several articles in peer-reviewed journals discuss studies on the penetration of bacteria other than E. coli O157:H7 into the interior of mechanically tenderized beef products. For example, one study concerning salmonellae inoculated in beef rounds found that mechanical tenderization increased the level of salmonellae in core samples by about 1 logarithm, that dripping inoculated rounds into a 50 parts per million (ppm) chlorine solution did not prevent the occurrence of salmonellae in core samples of mechanically tenderized units, and that Salmonella survived in the core and on the surface of some, but not all, inoculated rounds cooked to an internal temperature of 130 degrees F (“The Effect of Mechanical Tenderization on Beef Rounds Inoculated with Salmonellae,” Johnson, R.W.; Harris, M.E., and Moran, A.B., Journal of Food Safety. 1978; 1(3): 201-209. In another study, samples of mechanically tenderized beef were subjected to enumeration of aerobes, coliforms, E. coli, and organisms that formed black or grey on Harlequin TM agar (a medium formulated for recovery of Listeria). The study concluded that cooking mechanically tenderized beef to a medium rare condition may be adequate for ensuring the microbiological safety of this product, provided it is devoid of excessive contamination of deep tissues

- 35 -

(“Microbiological Conditions for Mechanically Tenderized Beef Cuts Prepared at Four Retail Stores,” Gill, C.O.; McGinnis, J.C., International Journal of Food Microbiology. 2004; 95(1): 95-102). Another study found that cleaning and sanitizing the tenderizer with an iodine-based sanitizer (25 ppm titratable iodine) decreased the bacterial levels of mechanically tenderized rounds (“Microbial Aspects of Mechanical Tenderization of Beef,” Raccah, M.; Henrickson, R.L., Journal of Food Protection. 1979. 42(12): 971-973. Several industry associations (the American Meat Institute, the National Cattlemen's Beef Association, the National Meat Association, and the Southwest Meat Association) have developed guidelines to address pathogen control in mechanically tenderized beef products and enhanced beef products. These guidelines are currently available on the Internet, on the Beef Industry Food Safety Council Web site at http://www.bifsco.org/BestPractices.aspx. The guidelines present recommended practices throughout tenderizing or enhancing operations and during cleaning and sanitizing operations. FSIS Actions To Enforce and Facilitate Compliance with the Reassessment Requirement The Agency intends to instruct its inspection program personnel to determine whether establishments have considered the significance of the three outbreaks discussed in this notice as part of an annual HACCP plan reassessment for mechanically tenderized beef products. FSIS will also instruct inspection program personnel to ensure that all establishments producing mechanically tenderized beef products, including small and very small establishments that may not belong to a trade association, are aware that the Agency has issued this notice. Finally, FSIS intends to instruct its inspection program personnel to collect data concerning the outcomes of the required reassessment. Paperwork Reduction Act FSIS has reviewed the paperwork and recordkeeping requirements in this notice in accordance with the Paperwork Reduction Act and has determined that the paperwork requirements for the regulations that require establishments that produce mechanically tenderized beef products to reassess their HACCP Plans have already been accounted for in the Pathogen eduction/HACCP Systems information collection approved by the Office of Management and Budget (OMB). The OMB approval number for the Pathogen Reduction/HACCP Systems information collection is 0583-0103. Additional Public Notification Public awareness of all segments of rulemaking and policy development is important. Consequently, in an effort to ensure that the public and in particular minorities, women, and persons with disabilities, are aware of this notice, FSIS will announce it on-line through the FSIS web page located at http://www.fsis.usda.gov/regulations_&_policies/2005__Notices__Index/index.asp. FSIS also will make copies of this Federal Register publication available through the FSIS Constituent Update, which is used to provide information regarding FSIS policies, procedures, regulations, Federal Register notices, FSIS public meetings, recalls, and other types of

- 36 -

information that could affect or would be of interest to our constituents and stakeholders. The update is communicated via Listserv, a free e-mail subscription service consisting of industry, trade, and farm groups, consumer interest groups, allied health professionals, scientific professionals, and other individuals who have requested to be included. The update also is available on the FSIS web page. Through Listserv and the web page, FSIS is able to provide information to a much broader, more diverse audience. In addition, FSIS offers an email subscription service which provides an automatic and customized notification when popular pages are updated, including Federal Register publications and related documents. This service is available at http://www.fsis.usda.gov/news_and_events/email_subscription/ and allows FSIS customers to sign up for subscription options in eight categories. Options range from recalls to export information to regulations, directives and notices. Customers can add or delete subscriptions themselves and have the option to password protect their account. Done at Washington, DC on: May 20, 2005. Barbara J. Masters, Acting Administrator. FR Doc. 05-10471 Filed 5-25-05; 8:45 am

- 37 -

Appendix C Food Safety Risk Associated with Non-intact Tenderized Beef Products

Abstract As the representative of the U.S. beef industry, the National Cattlemen’s Beef Association (NCBA) has been proactive in addressing the potential public health concern of E. coli O157:H7, Salmonella and Listeria monocytogenes in mechanical tenderized, needle injected or restored beef products. This paper is a summary of several research studies funded by the beef checkoff. Three surveys of 2,213 intact beef subprimals showed that the incidence of E. coli O157:H7 was very low. This pathogen was isolated from only two samples and the quantitative count was <3.0 CFU per 200 cm2. After inoculating the surface of subprimals with E. coli O157:H7, it was demonstrated that mechanical tenderization transposed a low level of the pathogen to subsurface muscle tissue. However, there is no public health risk if the consumer cooks steaks and roasts to an endpoint temperature of at least 140oF. The effect of different intervention treatments on the survival of E. coli O157:H7 on subprimal beef products was also studied. Acidified sodium chlorate (ASC) or peroxyacetic acid (PAA) were added to the surface of subprimals inoculated with E. coli O157:H7 as intervention treatments prior to mechanical tenderization. Both chemical treatments caused a significant reduction in the level of E. coli O157:H7 on the surface but not in the subsurface muscle tissue. A combination of a hot (176°F) lactic acid dip for 2 to 4 seconds followed by a 60 to 70 microwave treatment after packaging resulted in a 1.0 to 1.5 log reduction in the number of E. coli O157:H7 on 5.0 lb beef blocks. Checkoff funded studies were conducted to determine if Salmonella or Listeria monocytogenes on beef subprimals presented a public health risk. Neither pathogen grew in inoculated tenderized steaks during storage at 45°F for 21 days. It was concluded that Salmonella and L. monocytogenes did not present a public health risk if steaks and roasts are cooked to at least 140° F. Introduction E. coli O157:H7 was first recognized as a foodborne pathogen in 1982 when it was associated with illness in people that had consumed contaminated undercooked ground beef. It is estimated that 62,000 cases of E.coli O157:H7 infection occur yearly in the US. In 1994 the USDA/ FSIS notified the public that raw ground beef contaminated with E. coli O157:H7 would be considered to be adulterated under the Federal Meat Inspection Act unless the ground beef is further processed to destroy the pathogen. In 1994, the FSIS began testing ground beef for E. coli O157:H7. In January 1999 the USDA/FSIS published a notice in the Federal Register that that expanded their E. coli O157:H7 adulteration policy to include non-intact beef products. Beef products that were affected by this new policy were (a) beef primal or subprimal cuts that are blade or needle tenderized, (b) those that are injected with chemicals, and (c) steaks and roasts made by combining pieces of beef through restructuring. The concern is that the mechanical tenderization or needle injection procedures will contaminate subsurface muscle tissue with E. coli O157: H7 and therefore a higher cook temperature would be required to render the product safe for human consumption. The new policy states that any non-intact beef product found to be contaminated

- 38 -

with E. coli O157:H7 would be considered to be adulterated and therefore must be processed into a ready-to-eat product. It is a common practice to subject beef subprimals to a tenderization procedure to improve tenderness. The subprimals are penetrated with double edge blades or needles and the subprimals are then cut into individual steaks or roasts. In 2001, the National Advisory Committee on Microbiological Criteria for Foods (NACMCF) was asked by USDA/FSIS to address several questions with the regard to the potential public health risk by the presence of E. coli O157:H7 in tenderized, non-intact beef steaks and roasts. In their final report published in 2002, NACMCF concluded that non-intact blade tenderized beef steaks do not present a greater risk to consumers if the meat is oven broiled and cooked to an internal temperature of 140°F or above (NACMCF, 2002). Although the data were more variable at temperatures below 140°F, it was still possible to achieve a 3.2 log reduction of E. coli O157:H7 for tenderized beefsteaks. The NACMCF concluded that additional research was needed to determine whether non-intact tenderized beef roasts present a greater risk to consumers from E. coli O157:H7 when prepared in the same way as intact roasts. Since 1999, the NCBA has managed a series of checkoff funded research projects to determine the potential public health risk associated with E. coli O157:H7, Salmonella and L. monocytogenes in tenderized beef subprimals. The results from these research studies are summarized in this paper. Summary of Beef Checkoff-Funded Research Projects Prevalence Surveys Three independent studies were conducted to determine the prevalence of E. coli O157:H7 on the surface of beef subprimals. The first study, conducted by the ABC Research Corporation, was designed to determine the prevalence of E. coli O157:H7 and indicator organisms on the surface of 200 samples of each of the following six beef subprimals: chuck tenders, trimmed strips, bottom round flats, trimmed brisket, cap-on top rounds and cap-off inside rounds (Kennedy and Bodnaruk, 2004). 1200 subprimal beef products from 5 plants were examined prior to mechanical tenderization for E. coli O157:H7, total aerobic plate count, coliforms and generic E. coli. E. coli O157:H7 was not detected in any of the 1200 beef samples tested. The aerobic plate counts, total coliforms and generic E. coli counts from the 600 samples were variable within each sample set of each subprimal and between the six subprimal types. The levels of generic E. coli were either very low or undetectable for most of the subprimal samples. It was concluded from these results that internal contamination of beef subprimals with E. coli O157:H7 by mechanical tenderization is an improbable phenomenon. Food Safety Net Services, Ltd. and Colorado State University conducted a similar study of subprimal beef cuts prior to mechanical tenderization for the presence of E. coli O157:H7 (Warren et al., 2002). The surface of 1,014 beef samples from 6 processing plants during the months of June and July were examined. Only two of the subprimal beef samples tested positive for E. coli O157:H7 and the quantitative count of the two positive samples was <3.0 CFU per

- 39 -

200 cm 2. The results from this study indicate that E. coli O157:H7 is not a common contaminant on the surface of subprimal beef cuts prior to mechanical tenderization. It was concluded that internal contamination of subprimal beef cuts by this pathogen via mechanical tenderization is unlikely to occur. Translocation Studies Luchansky (2004) reported that Kansas State University conducted a study to determine the degree of translocation of E. coli O157:H7 to the interior muscle tissue of a steak during tenderization. The results from the first study showed that after a single pass through a blade tenderizer, 3-4% of E. coli O157:H7 on the surface was translocated to the interior muscle tissue.

Checkoff dollars were also used to fund a study to determine the potential public health risk associated with the presence of either Salmonella or Listeria monocytogenes present on the surface of subprimals. Silliker Laboratories conducted a study to determine the frequency and distribution of pathogenic and non-pathogenic bacteria on the surface and in the core of non-intact beef products throughout the US (Gangar and Curiale, 1999a). The pathogens were L. monocytogenes, Salmonella, E. coli O157:H7, Campylobacter jejuni/coli, Clostridium perfringenes, and coagulase positive Staphylococcus aureus. The non-pathogenic bacteria included in this survey were aerobic plate count, generic E. coli and coliform. A total of 49 cubed steaks, 30 rolled roasts, 28 corned beef, 12 pumped roasts and 25 needle-tenderized steaks were included in this study. The samples were obtained from retail food stores and food service sources. Individual 25-gram samples were used for quantitative and qualitative analysis. L. monocytogenes was isolated from 100% of the needle-tenderized steaks and from 7% of the other products. The distribution of the L. monocytogenes between the interior and exterior samples was equal. The average number of L. monocytogenes from the exterior and core of the different beef products was 0.68 and 0.45 MPN/g respectively. These values are not significantly different. The single isolation of Salmonella was from the surface of a rolled roast. E. coli O157:H7 and Campylobacter were not present in or on any of the analyzed beef products. The aerobic plate count, coliform count and generic E. coli count were higher on the surface of products than in the core. The average counts of S. aureus and C. perfringenes were <11 CFU/g. Cooking Studies Kansas State University researchers conducted a study to evaluate the effectiveness of cooking procedures on the destruction of E. coli O157:H7 in intact and mechanically tenderized beefsteaks (Marsden et al., 1999). This heating study was designed to determine the thermal destruction of E. coli O157:H7 in tenderized steaks during broiler cooking to different endpoint temperatures. The surface of top sirloin subprimals was inoculated with 107 CFU/cm2 E. coli O157:H7 prior to being tenderized in a blade tenderizer. The steaks, cut from the tenderized inoculated subprimals, were broiled to endpoint temperatures of 120°F, 130°F, 140°F, 150°F, 160°F and 170°F. After they reached the designated endpoint temperature, the steaks were immersed in an ice bath to quickly halt the

- 40 -

heating process. Steaks cut from inoculated subprimals that were not tenderized were used as controls. Initial statistical analysis of all main effects (treatment, weights, and temperature) and all possible combinations of interaction, with bacterial reduction as the dependent variable, revealed a significant (p<0.05) interaction between the treatment (tenderized and non-tenderized) and temperature. Heating tenderized steaks and non-tenderized steaks to 120°F resulted in a significant (p< 0.05) difference in the destruction of E. coli O157:H7. The results showed a 3.2 log reduction and a 5.2 log reduction in tenderized and non-tenderized steaks respectively. While the endpoint temperature of 120°F was sufficient to destroy 5 logs of E. coli O157:H7 on the surface of non-tenderized subprimals, it was not high enough to eliminate the pathogen in the interior of tenderized steaks. It should be noted that the 2001 FDA Food Code does not include heating times and temperatures below 130oF for raw animal products (FDA, 2001). At 130oF the log reduction of E. coli O157:H7 for the tenderized and non-tenderized steaks was 5.6 and 5.0 respectively. This difference was not significant. At endpoint temperatures of 140°F, 150°F, 160°F, and 170°F, there was a 6-log reduction in both the tenderized and non-tenderized steaks. It was concluded that a target temperature of at least 140°F provides the necessary thermal destruction required to eliminate the public health risk from E. coli O157:H7. It was noted in this study that even though the steaks were rapidly chilled in ice water after broiling to a specific endpoint temperature, the internal temperature continued to rise by approximately 11°F. This is an important margin of safety for broiled steaks being served in a food service setting. ARS scientists conducted a study to determine the effectiveness of six grilling temperatures for destroying E. coli O157:H7 in steaks cut from inoculated subprimals (Luchansky, 2004). In this study, subprimals were inoculated on the lean side of subprimals with 106

CFU/cm2 and passed once through a mechanical tenderizer. After tenderizing, the subprimals were cut into steaks that were 0.75, 1.0, and 1.25 inches thick. The steaks were grilled to internal temperatures of 120oF, 130 oF, 140 oF, 150 oF, 160 oF and 170°F. Five core samples from predetermined locations, were aseptically removed from each steak and tested for surviving E. coli O157:H7.

The results from the grilling study showed the following:

1. Greater lethality was observed as the target cooking temperature increased. 2. With the exception of 130°F, greater lethality was observed as the thickness of the

steaks increased. 3. In some instances greater lethality was observed for two core samples than in the

other three core samples. 4. E. coli O157:H7 was recovered with direct plating methods from steaks cooked to

120°F and 140°F, whereas the pathogen was only recovered by enrichment from some of the steaks that were cooked at 150°F and from all the steaks cooked to160oF and 170°F.

- 41 -

Silliker Laboratories personnel conducted a study in which cubed steaks, rolled roasts, corned beef, pumped roasts and needle-tenderized steaks were inoculated with a five strain cocktail of Salmonella, L. monocytogenes and E. coli O157:H7 and then heated to temperatures ranging from 145°F to 160°F (Gangar and Curiale, 1999b). A syringe was used to inoculate these same five beef products in the center to a level of 107 cells per gram. Four cooking procedures were used. The cubed steaks were pan-fried and the corned beef was water cooked. The needle tenderized steaks and pumped steaks were oven broiled. Rolled roasts were cooked in an electric convection oven. All five products were cooked to end point temperatures of 145°F, 150F°, 155°F and 160°F. Non-cooked inoculated beef samples were used as controls to establish the initial inoculum levels. Non-cooked, non-inoculated samples were analyzed to demonstrate the methods to differentiate the pathogenic test organisms from the non-pathogenic background microflora of the samples. In this study, there was no effort to rapidly chill the five beef products after heating and the internal temperature continued to rise after they reached their target internal temperature. This temperature increase ranged from 3°F for pan-fried steaks to 10°F for the broiled tenderized steaks, which added to the cooking period. The rate of inactivation of the three pathogens varied with product type and pathogen type. Pathogen reduction in the five beef products cooked to 145°F varied greatly by product and to a lesser degree with pathogen type. The reductions ranged from 1.1 to 5.3 logs. In fully cooked product (160°F) the log reduction of the pathogenic bacteria ranged from 2.5 to 6.6 logs. The level of pathogen reduction at 150°F and 155°F was between those observed for 145°F and 160°F. In general, the lowest reductions were in pan-fried cubed steak and the highest in water cooked corned beef. The visual cooked appearance of the five beef products cooked to the 4 different temperatures was documented in this study. When cooked to an endpoint temperature of 145°F, all five products had a lightly browned surface color and a rare “bloody” internal appearance. When the products were heated to the 160°F endpoint temperature, the outer surface was described as dark brown and with the exception of the needle tenderized steak, the core was still slightly pink and juicy. The core of the tenderized steak was considered to be well done, but juicy. Food Safety Net Services, Ltd. evaluated the effectiveness of cooking procedures, on ten different commercial beef products (Bellinger et al., 2002). This study included two whole muscle products, loin steak and top sirloin. While all of the beef products were labeled with the “Safe Heating Instructions” including “Cook Thoroughly” as required by USDA, FSIS, only one included recommended cooking instructions. Three vendors produced the loin steak and top sirloin. The loin steak and top sirloin from vendors #1 and #2 were not mechanically tenderized. Vendor #3 mechanically tenderized both products. Vender #1 was the only one to include cooking instructions for grilling and pan-frying. When the grilling and pan-frying instructions were followed, the final internal temperatures for

- 42 -

the loin steak and top sirloin ranged from 148°F to 155°F. In the food service industry, this is considered to be medium rare. When the grilling and pan frying cooking instructions from vendor #1 were followed for preparing the loin steak and top sirloin from vendors # 2 and #3, the final internal temperature ranged from 147°F to 160°F. The NACMCF report concluded that intact or tenderized steaks cooked to an internal temperature of at least 140°F do not present a public health risk in relation to E. coli O157:H7 (NACMCF, 2002). Based on the results from this study, it was suggested that the meat industry should provide adequate cooking instructions on all raw beef products, including intact and tenderized whole muscle products. The cooking instructions should include at least two cooking methods. The ABC Research Corporation conducted a study to determine the fate of Salmonella and L. monocytogenes in mechanically tenderized beef products (Kennedy et al., 2001). After inoculating the surface of top butt subprimals with a low and high level(list the levels) of either pathogen, the subprimals were passed twice through a commercial blade tenderizer. The subprimals were then cut into steaks and vacuum packaged and stored at 0°F, 28°F and 45°F. After 0, 3, 7, 14 and 21 days storage, triplicate samples were tested for microbial growth by direct plating and the more sensitive ELISA assay. For each inoculum level, a total of 45 steak samples were examined for all storage times and temperatures. Steaks from subprimals inoculated with a high level of L. monocytogenes had a mean inoculum level of 7,440 CFU/g. The count in steaks inoculated with the low inoculum level ranged from below the detection level to 10 CFU/g by direct plating. Steaks made from subprimals inoculated with a high and low level of Salmonella had a mean inoculum level of 390 CFU/g and below the detection level to 10 CFU/g, respectively. The inoculated steaks were heated on a flat top grill on both sides to an internal temperature of 125°F, 135°F, 145°F, 155°F and 165°F to represent rare, medium rare, medium, medium well and well-done steaks. The cooked steaks were cooled at ambient temperature to 110°F or less to simulate actual food service preparation before they were tested microbiologically. Neither L. monocytogenes nor Salmonella grew in the tenderized steaks during storage at 28°F or 45°F for 21 days. The numbers of both pathogens gradually decreased. It is speculated that the growth of the background microflora duringstorage at these temperatures may have caused the decrease in the numbers of Salmonella and L. monocytogenes. Regardless of storage time and temperature, increasing the internal endpoint temperature of steaks during cooking resulted in a greater reductions in L. monocytogenes and Salmonella. At the high inoculum level, both pathogens were recovered by direct plating or with the ELISA assay at each endpoint temperature. At the low inoculum level, L. monocytogenes was recovered from 6 of 45 samples, 4 of 45 samples and 2 of 45 samples cooked to an internal temperature of 125oF, 135oF and 145oF, respectively. L. monocytogenes was not recovered from the 45 samples cooked to 155oF or 165oF.

- 43 -

At the low inoculum level, Salmonella was recovered from 3 of 45 samples, 2 of 45 samples and 4 of 45 samples when the steaks were cooked to 125°F, 135°F, or 145°F respectively. None of the 45 samples cooked to 155°F tested positive for Salmonella and one of 45 samples cooked to 165°F tested positive. The North American Meat Processors and Kansas State University conducted a survey to determine the cooking practices and methods for beef steaks and roasts (Rasor et al., 2004). Five hundred individuals were surveyed on their cooking practices and methods for preparing steaks and roasts. The results of the survey showed that most participants used color as an indicator of doneness of steaks and cooking time was used for roasts. While none of the individuals surveyed knew that 145°F is the FDA’s recommended minimum internal temperature for cooking steaks and roasts, 82% of respondents cook their beef products above this temperature to a level of doneness of medium or above. Intervention Studies A pilot plant study was conducted by ABC Research Corporation to determine the effectiveness of two intervention treatments to reduce the level of surface contamination on beef top sirloin butt subprimals prior to mechanical tenderization (Kennedy, 2003). An area on the top surface of individual subprimals was inoculated with a cocktail of four strains of E. coli O157:H7. This resulted in an initial count of 106 per cm2. The inoculated subprimals were then spray treated with (a) acidified sodium chlorate (ASC) or (b) peroxyacetic acid (PAA) prior to tenderization. The subprimals were mechanically tenderized within one minute after the spray treatment. Five untreated controls were included in this study. Uninoculated subprimals were also given the two-intervention treatment before tenderization. These samples were used to determine the aerobic plate count and total Enterobacteriaceae on the surface and in a core sample. A 100 cm2 surface sample, approximately 0.5 cm thick, was excised from the surface of the inoculated product and tested for E. coli O157:H7. A core sample of approximately 75 grams was also aseptically removed for E. coli O157:H7 analysis. The uninoculated tenderized subprimal were also sampled in the same way for testing for aerobic plate count and Enterobacteriaceae. The PAA and ASC intervention treatments applied to the surface of the beef top sirloin butt subprimals were modestly, but significantly, effective in reducing the level of E. coli O157:H7. The PAA caused a 0.63 to 0.71-log reduction in the numbers of E. coli O157:H7 in the two pilot plant trials. The ASC treatment caused a significant 0.63 log reduction in one of two pilot plant trials. There was no significant reduction in the second trial. Neither the PAA or ASC treatment caused a significant reduction in the aerobic plate count or in the level of Enterobacteriaciae on the surface. The examination of the core samples showed that there was no significant reduction in E. coli O157:H7, aerobic plate count or Enteriobacteriociae in the samples treated with PAA and ASC. Kansas State University conducted a study to evaluate the combined effect of a hot (80°F) lactic acid dip, vacuum packaging and subsequent microwave treatment on E. coli O157:H7 on the surface of 5.0 pound blocks of beef (Fung, 2000). A hot lactic acid dip lasting 2 or 4 seconds and

- 44 -

a 60 or 70 second microwave treatment resulted in a 1.0 to 1.5 reduction of E. coli O157:H7 and list what was analyzed. The acid and microwave treatments resulted in minimal color changes in the lean portion. Conclusions The incidence of E. coli O157:H7 on the surface of subprimals is rare. Only two of 1,614 subprimals examined in two surveys were positive for this pathogen and the quantitative count in both samples was < 3.0 CFU per 200 cm2. During mechanical tenderization, the translocation of E. coli O157:H7 from the surface of subprimals into the subsurface muscle tissue does occur at a low level. However, cooking studies showed that non-intact beefsteaks and roasts provide no significant food safety hazard when they are heated to an endpoint temperature of at least 140°F. The NCBA funded studies also showed that when steaks and roasts are cooked properly, Salmonella and L. monocytogenes do not present a public health risk.

- 45 -

References Bellinger, G., W. Warren, and J. Roman. 2002. Evaluation of cooking instructions and methods

for uncooked beef products. Food Safety Net Services, Ltd. FDA. 2001. FDA Food Code. U.S. Department of Commerce. Section 3-401.11 Raw Animal

Foods. Fung, D. Y. C. 2000. Reduction of natural microflora and Escherichia coli O157:H7 by

combination of hot lactic aid, vacuum packaging, and microwave treatment of subprimal beef cuts. Kansas State University.

Gangar, V. and M. Curiale. 1999a. National survey to determine levels and types of pathogens in non-intact beef products. Silliker Laboratories.

Gangar, V. and M. Curiale. 1999b. Determination of minimum cook temperatures for non-intact beef products. Silliker Laboratories.

Kennedy, J. and P. Bodnaruk. 2004. Survey of the prevalence of Escherichia coli O157:H7 on the surface of subprimal cuts of beef during winter months. ABC Research Corporation.

Kennedy, J., W. Birbari, and W. Brown. 2001. Risk assessment of Listeria monocytogenes and Salmonella in the production and cooking of blade-tenderized beefsteaks. ABC Research Corporation.

Kennedy, J. 2003. Evaluation of interventions to reduce microbial contamination in the manufacture of blade tenderized beef. ABC Research Corporation.

Luchansky, J.B. 2004. Microbiological safety of needle-tenderized beef steaks. USDA. Agriculture Research Service. Eastern Regional Research Center.

Mardsen, J., R. Phebus, H. Thippareddi, and C. Kastner. 1999. Escherichia coli O157:H7 risk assessment for the production and cooking of blade tenderized beef steaks: Determination of lethality requirements for blade tenderized steaks. Kansas State University.

National Advisory Committee on Microbiological Criteria for Foods. 2002. Escherichia coli O157:H7 in blade-tenderized non-intact beef. (http://www.fsis.usda.gov/OPHS/NACMCF/2002/repblade1.htm.)

Rasor, A., J. Miller, J. Franklin, E.J. Harvey, R. K. Phebus, C. Pearsall, and J. L. Marsden. 2004. Cooking practices and methods for beef steaks and roasts. North American Meat Processors Association and Kansas State University.

Warren, W., G. Bellinger, S. Wood, T. Frederick, and G.C. Smith. 2002. Characterization of E. coli O157:H7 on subprimal beef cuts prior to mechanical tenderization. Food Safety Net Services, Ltd. and Colorado State University.

- 46 -

Appendix D Additional References Bogen, K.T. and G.A. Keating. 2001. U.S. dietary exposure to heterocyclic amines. J. Expo.

Anal.Enviro. Epidemiol.11(3):155-168. Elder, R.O., J.E. Keen, G.R. Siragusa, G.A. Barkocy-Gallagher, M. Koohmaraie, and

W.W.Laegreid. 2000. Correlation of enterohemorrhagic E. coli O157 prevalence in feces, hides, and carcasses of beef cattle during processing. Proc. Natl. Acad. Sci. 97(7):2999-3003.

Food and Drug Administration. 2000. Audits International/FDA home Cooking Temperature Database. (http:www.foodriskclearinghouse.umd.edu/ColdFusion/ Cooking/).

FSIS-USDA. 1994. Nationwide Beef Microbiological Baseline Data Collection Program: Steers and Heifers—October 1992-September 1993.

FSIS-USDA. 2001. Draft Risk Assessment of the Public Health Impact of Escherichia coli O157:H7 in Ground Beef. Risk Assessment Division, Office of Public Health and Science. Currently under review by the National Academy of Science. (http://www.fsis.usda.gov/OPPDE/rdad/FRPubs/00-023NReport.pdf and http://www.fsis.usda.gov/OPPDE/rdad/FRPubs/00-023N/InterpretiveSummary.pdf).

FSIS-USDA. 2002. Guidance on Ingredients and Sources of Radiation Used to Reduce Microorganisms on Carcasses, Ground Beef, and Beef Trimmings. October 7, 2002, Federal Register, E. coli O157:H7 Contamination of Beef Products. (http://www.fsis.usda.gov/Regulations_&_Policies/2002_Interim_&_Final_Rules_Index/index.asp).

Gill, C.O. and J.C. McGinnis. 2004. Microbiological Conditions for Mechanically Tenderized Beef Cuts Prepared at Four Retail Stores, Int. J. Food Microbiol. 95(1): 95-102.

Johnson, R.W., M.E. Harris, and A.B. Moran. 1978.The Effect of Mechanical Tenderization on Beef Rounds Inoculated with Salmonellae, J.Food Safety. 1(3): 201-209.

Kause, J. 2001. Food Safety and Inspection Service. Risk Assessment Plan for Non-Intact Beef. Hazard Analysis and Regulatory Affairs Staff, Office of Public Health and Science. December 18, 2001.

Kaplan, S., E. Ebel, and W. Schlosser. 2002. Food Safety and Inspection Service. Technical Report: Comparative Risk Assessment for Intact (Non-Tenderized) and Non-Intact (Tenderized) Beef. Modeling and Exposure Assessment Staff, Office of Public Health and Science.

Kansas State University. 2001. Evaluation of Pathogen Risks Associated with Blade Tenderized Beef Cooked to Varying Degrees of Doneness. Presentation to National Advisory Committee for Microbiological Criteria for Foods (NACMCF) meeting, Washington, DC.

Lammberding, A. 2002. Health Canada . Reported cases of illness associated with tenderized beef. Quebec Center of Food Inspection and Animal Health.

Mead, P.S., L. Slutsker, V. Dietz, L.F. McCaig, J.S. Bresee, C. Shapiro, P.M.Griffin, and R.B. Tauxe. 1999. Food-related illness and death in the United States. Emerg. Infect. Dis. 5(5):607-625.

Michigan Department of Community Health/Communicable Disease and Immunization Division August 2000. Report: Foodborne Illness Investigation: E. coli O157.

- 47 -

National Advisory Committee for Microbiological Criteria for Foods. January 25, 2002. Final Report: E. coli O157:H7 in Blade Tenderized Beef. (http://www.fsis.usda.gov/OPHS/NACMCF/2002/rep_blade1.htm).

Raccah, M. and R.L. Henrickson. 1979. Microbial Aspects of Mechanical Tenderization of Beef, J. Food Prot. 42(12): 971-973.