Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
s Microbiologic and Toxicologic Assessment Abstract Microbiologic and toxicologic (chemical) hazards that are vitally important to public health are not addressed in SIS-C. Continuous assessment of such hazards based on scientifically valid sampling and the use of more effective methods must be a part of any meat inspection system. Moreover, research must be conducted on slaughtered cattle to provide statistically valid data to test hypotheses relating to microbiologic and toxicologic hazards in order to enhance and preserve public health and restore public confidence in the current inspection system. Assessment As emphasized in previous Food and Nutrition Board (FNB) reports (NRC, 19SSa, 1987a), microbiologic and toxicologic hazards represent the greatest risk to human health associated with meat. A full assessment of possible microbiologic and toxicologic risks associated with red meat goes well beyond the charge of this committee. These risks have been reviewed in detail in previous Food and Nutrition Board reports (NRC, 19SSa,1987a). Table 5-l is taken from the 19SS report Meat and Poult7y Inspection. Although it includes information on both red meat and poultry, it illustrates the diversity of microbial agents that may be present in or on meat and meat products. Salmonella and Campylobacter jejuni are generally regarded as posing the major microbiologic risks associated with meat and poultry. However, we currently lack the epidemiologic data necessary to calculate the proportion of human infections due to these organisms that are attributable s~ecificaliv to red meat (i.e.. attributable risk data). In unpublished data from the USDA National Surveillance Survey made available to the committee (see Appendix B), Salmonella were isolated from 1.~% of 5,319 frozen red meat samples tested; the significance of these results is unclear, since complete details on study methodology were not available. Recent concern has also focused on contamination of meat by enterohemorrhagic strain of Eschenchia colt, a newly described human pathogen that can cause hemorrhagic colitis and, in severe cases, the hemolytic-uremic syndrome. Risks associated with toxicologic hazards appear to be smaller. However, they have been less well defined and often do not have the immediately obvious health consequences of bacterial or viral infections. . , ~, , a, . . These observations led successive FNB committees to recommend that the Food Safety and Inspection Service (FSIS) intensify its efforts to control and eliminate contamination by microorganisms that cause disease in humans and optimize its 42
Table 5-} Classification of Worldwide Meatborne and Pathogens According to Modes of Transmissiona Poult~yborne Microbial Pathogenic microorganisms transmissible to humans by ingestion of raw or undercooked meat and poultry: Bacillus anthracis Balantidium cold Campylobacter cold Campylobacter fetus subsp. fetus Campylobacter jejuni Escherichia cold Francisella tularensis Salmonella spp. Sarcocystis spp. Taenia saginata Taenia solium Toxoplasma gondii Trichinella spiralis Yersinia enterocolitica Yersinia pseudotuberculosis Pathogenic microorganisms transmissible to humans by ingestion of cooked or otherwise heat-processed meat or poultry that became contaminated after the heat processing or that was improperly stored after initial heat processing: Any of the above Bacillus cereus Clostritlium botulinum Clostridium perfringens Shigella spp. Staphylococcus aureus Streptococcus pyogenes Pathogenic microorganisms transmissible by contact with animal tissue or by inhalation of aerosols or dust from animals: Bacillus anthracis Brucella spp. Chlamydia paittaci Cowpox virus Coxiella burnetii E,ysipelothruc rhusiopathiae Francisella tularensis Leptospira spp. Listeria monocytogenes Newcastle disease virus Pseudomonas mallet Streptococcus pyogenes Toxoplasma gondii Other bacteria sometimes on meat and poultry that have been reported to be pathogens but for which proof is lacking that meat and poultry are vehicles: Aeromonas spp. Bacillus licheniformis Citrobacter spp. Klebsiella spp. Plesimonas shigelloides Proteus spp. Providencia spp. Streptococcus faecalis Streptococcus faecium a Source: NRC, 19SSa. 43
National Residue Program (NRP) for dealing with toxicologic risks (NRC 1985a, 1987a). When the SIS-C program is considered in this context, several questions arise. Some of these issues are dealt with elsewhere in this report, but the committee believes that they are important enough to warrant repeating in this chapter. Have previous FNB recommendations concerning microbiologic and toxicologic hazards been integrated into SIS-C? According to FSTS, the microbiologic or toxicologic issues discussed in previous FNB reports are not directly addressed in SIS-C. Might SIS-C have an indirect effect on microbiologic or toxicologic hazards in red meat? Modifications in inspection procedures could result in reduced microbial contamination of carcasses only if they lead to improved handling of carcasses or plant sanitation. Thus, evaluation of inspection systems should consider possible effects on total microbial counts, and on detection and counts of such pathogens as Salmonella, Campylobacter, and enterohemorrhagic Eschenchia cold It is unlikely that improved inspection procedures could lead to reduced levels of toxicologic or chemical hazards. However, any changes in carcass rinses or equipment sanitizing chemicals should be evaluated for possible effects of these contaminants on human health. Are there data documenting the effects of STS-C on microbiologic or toxicologic hazards? FSIS data on microbiologic and toxicologic hazards are available for some individual plants. Unfortunately, existent data sets are construed inconsistently and appear to be too small to permit meaningful statistical comparisons between SIS-C and traditional systems. Such confounding variables as differences in source and type of cattle also complicate the analysis. The committee received USDA data on only 35 SIS-C samples from five SIS-C plants (see Appendix B). No conclusions about inspection systems could be drawn due to the small sample size, low frequency of contamination, and the freezing of specimens for shipment. Comparable problems arise in trying to draw conclusions from available toxicologic data. Some processing plants (such as the Monfort plant in Greeley, Colorado) have developed data bases on microbial contamination of products. These data are of interest, and their generation should be encouraged. However, industry-derived data can not be substituted for substantive, statistically valid FSIS data that are subject to the scrutiny of scientific peer review processes. 44
If there is a need for such data, how could they be obtained? There is need for data on microbiologic and toxicologic hazards in meat to address the committee's concerns and those of the public. These data would permit comparisons among various inspection systems and continuing assessment of inspection strategies in any one plant. Relatively simple studies could be designed to provide these basic data and to test appropriate hypotheses. For example: o Hypothesis I: The bacterial contamination (i.e., total plate counts, Salmonella counts, etc.) of carcasses in SIS plants is no different--neither greater nor less- -than that of traditionally inspected carcasses of similar source and type slaughtered under similar conditions. O Hypothesis 2: Bacterial contamination does (or does not) correlate with carcass error rates, as determined by organoleptic techniques. O Hypothesis 3: Bacterial contamination rates are (or are not) more dependent on line speed than on inspection strategy. To evaluate these hypotheses statistically, it would be desirable to determine the sample sizes required to provide acceptable alpha(<xjZ and betas risk levels and also to consider possible confounding effects such as type of cattle being slaughtered. A preliminary study, such as one performed by Agriculture Canada (Appendix F), could provide data for sample size determinations. After methods have been established, a continuing program of data collection should be developed and hypotheses retested to provide ongoing assessment of quality control programs and inspection processes. Sudden occurrence of high aerobic bacterial counts (e.g., standard plate counts, conform counts, or Salmonella counts) in a plant might indicate a breakdown in the control of bacterial contamination that is detectable by quality control procedures and inspection processes. Sudden increases in the incidence of enteric bacterial pathogens might indicate an increase in fecal contamination of carcasses by carrier animals. Similarly, a sudden increase in chemical contamination or residues would indicate the need for traceback to sources of contamination. To accomplish these goals, FSIS should identify the most effective methods and should use rapid diagnostic tests, such as enzyme-linked immunosorbant assays (ELISA) and DNA probes for pathogenic microorganisms, as previously recommended by FNB committees. Programs of this type could also be incorporated A (or type b--refers to the error made if a true hypothesis is rejected. 2§ (or Type IT)--refers to the error made if a false hypothesis is accepted. 45
into ongoing studies of the relationship of antimicrobial use to drug resistance in pathogens isolated from foodborne disease outbreaks in humans (IOM, 1989~. Realistic and appropriate standards to protect public health are complex but very important. Under favorable conditions, pathogenic bacteria multiply rapidly to levels adequate to cause human disease both in the plant and after the carcass enters the market. Therefore, microbiologic standards must be considered carefully. The zero risk concept may be appropriate to consider as an idealistic goal for pathogenic bacteria (which crow and multiolv1. Thus. the detection of low levels of ~ _ . .', , · . . . microorganisms can be important, because such organisms can multiply rapidly to levels adequate to cause illness in humans. On the other hand. the negligible risk standard can be an appropriate goal for levels of toxins or chemicals (NRC, 1987b), because their concentrations do not increase. In fact, levels of chemicals in food are usually substantially reduced by food processing and food preparation. Further research is needed on establishing negligible risk levels for chemicals in food to ensure that sensitive population subgroups, such as infants, children, or immunocompromised people, are not susceptible to risks from chemical levels that would be safe for most people. Specific issues related to the effects of pesticides on infants and children are currently under study by a National Research Council committee in the Board on Agriculture. This report will be released in 1991 (NRC, in press). FSIS is not authorized to conduct basic research. Therefore, the necessary studies must be done by researchers outside FSIS. However, USDA must realize that a modern, scientifically based inspection service must conduct or fund research to evaluate practical problems. Modernization of inspection requires sophisticated computer systems, computer modeling, rapid diagnostics, and the ability to integrate this technology into on-line inspection strategies. In summary, ongoing assessment of microbiologic and toxicologic hazards through well-designed and scientifically valid studies is essential to meat inspection. Results of these assessments must be considered in any comparison of inspection strategies. Recommendations o Microbiologic, toxicologic, and chemical data must be considered by FSIS in designing and evaluating any inspection systems designed to protect public health. o If FSIS is to develop sound, believable, scientifically based inspection strategies, a research group must be established with the appropriate statistical, epidemiologic, microbiologic, and toxicologic expertise to frame and test hypotheses relating food inspection to microbiologic and chemical hazards. This 46
group should be based in USDA. It must have the flexibility to coordinate with other government agencies (e.g., the Food and Drug Administration and the Centers for Disease Control) involved in maintaining the safety of the food supply. It must also have adequate funds to draw on expertise from outside the government (i.e., universities and private industry). 47 1
