5
Drug Residues and Microbial Contamination in Food: Monitoring and Enforcement

A principal goal of U.S. food-safety programs is the control of contaminants that might appear in food because of drug use in animals or inadvertent introduction of microorganisms. Drug residue control and microbial-contamination surveillance are accomplished through a rigorous, extensive process of sampling, testing, notification, and enforcement. Tens of thousands of samples are collected and processed annually in routine screening procedures aimed at statistically identifying the occurrence of residues and microorganisms. Three agencies do most of the work to protect the public from residue and microbial hazards: the U.S. Department of Agriculture (USDA), including the Food Safety and Inspection Service (FSIS) and the Agricultural Marketing Service (AMS); the Food and Drug Administration (FDA); and the U.S. Environmental Protection Agency (EPA).

USDA is charged with enforcing the Federal Meat Inspection Act (FMIA), the Poultry Products Inspection Act (PPIA), and the Egg Products Inspection Act (EPIA). Within USDA, FSIS is responsible for the wholesomeness and safety of fresh meat, poultry, and processed meat and poultry products intended for human consumption. It inspects slaughtering and processing establishments and samples and analyzes tissues derived from livestock and poultry at the time of or after slaughter. Inspection and analysis are intended to ensure, among other things, that meat and poultry do not contain residues of drugs, pesticides, or pathogens that cause them to be adulterated as defined in FMIA or PPIA. When residue violations are detected, FSIS notifies FDA, as FDA is authorized to take legal action against violators.

AMS is responsible for the wholesomeness and safety of egg products. It



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 110
The Use of Drugs in Food Animals: Benefits and Risks 5 Drug Residues and Microbial Contamination in Food: Monitoring and Enforcement A principal goal of U.S. food-safety programs is the control of contaminants that might appear in food because of drug use in animals or inadvertent introduction of microorganisms. Drug residue control and microbial-contamination surveillance are accomplished through a rigorous, extensive process of sampling, testing, notification, and enforcement. Tens of thousands of samples are collected and processed annually in routine screening procedures aimed at statistically identifying the occurrence of residues and microorganisms. Three agencies do most of the work to protect the public from residue and microbial hazards: the U.S. Department of Agriculture (USDA), including the Food Safety and Inspection Service (FSIS) and the Agricultural Marketing Service (AMS); the Food and Drug Administration (FDA); and the U.S. Environmental Protection Agency (EPA). USDA is charged with enforcing the Federal Meat Inspection Act (FMIA), the Poultry Products Inspection Act (PPIA), and the Egg Products Inspection Act (EPIA). Within USDA, FSIS is responsible for the wholesomeness and safety of fresh meat, poultry, and processed meat and poultry products intended for human consumption. It inspects slaughtering and processing establishments and samples and analyzes tissues derived from livestock and poultry at the time of or after slaughter. Inspection and analysis are intended to ensure, among other things, that meat and poultry do not contain residues of drugs, pesticides, or pathogens that cause them to be adulterated as defined in FMIA or PPIA. When residue violations are detected, FSIS notifies FDA, as FDA is authorized to take legal action against violators. AMS is responsible for the wholesomeness and safety of egg products. It

OCR for page 110
The Use of Drugs in Food Animals: Benefits and Risks conducts inspections and analyzes samples for chemical residues to ensure compliance with EPIA at plants that process egg products. FDA enforces the federal Food, Drug, and Cosmetic Act (FDCA). FDA is directly responsible for ensuring the safety of milk and seafood for human consumption and that animal feeds are safe and contain no illegal residues of drugs, pesticides, or other environmental contaminants. FDA also approves drugs used for food-producing animals, establishes tolerance and safe levels for animal drugs and establishes action levels for unavoidable environmental contaminants that might adulterate food. (The section on “Tracking Residues in Food: Regulatory Input” describes tolerance, safe, and action levels.) EPA is responsible for administering and enforcing the Federal Insecticide, Fungicide, and Rodenticide Act, which regulates the manufacture, sale, and use of pesticides. EPA also is responsible under FDCA for establishing tolerance levels and recommending action levels to FDA and FSIS for residues or pesticides in food. Under the Toxic Substances Control Act, EPA also regulates other chemical substances (such as industrial chemicals) that can adulterate food. DRUG RESIDUE STANDARDS AND SCREENING Under the provisions of FDCA, FDA’s Center for Veterinary Medicine (CVM) is responsible for ensuring that drugs are safe and effective for use in animals, and that food derived from animals is safe for human consumption. In line with the requirements for approval of a drug, the company developing and sponsoring the animal drug is responsible for furnishing CVM with the scientific information and experimental data showing that the presence of residues from a compound in edible animal products is safe for consumers. The sponsor often must develop and validate analytical methods to extract, purify, and quantify the residues and metabolites of a drug in tissue. Detection and measurement of drug residues are scrutinized by two approaches, assay level and analytical method status (FSIS 1995b). Methods are classified by level as summarized below: Level I Assay results with highest validation and credibility; considered unequivocal at concentrations of interest; single or combination methods can be used to determine concentration and identity of residue; when used in combination, methods are confirmatory. Level II Assay results are not unequivocal but accurate and capable of detection at the concentration of interest; sufficiently reliable to be used as a reference method. Level III Screening methods developed to detect the presence of residue and needed for the high throughput of samples; samples that are positive by Level III methods are analyzed further by Level I or Level II methods.

OCR for page 110
The Use of Drugs in Food Animals: Benefits and Risks Practical considerations influence the nature of analyses that FSIS will consider and accept as regulatory methods of residue detection. The criteria for acceptance are the following: (1) the method should take no more than 2 to 4 hours to perform; (2) the instrumentation must be common to all analytical chemistry laboratories; (3) the method must have a minimum proficiency level to detect the residue at the concentrations needed; (4) a quality-assurance program must accompany the method; and (5) the method must have been successfully tested and found reliable for detection of residue at 0, 0.5, 1, and 2 times the levels of published (40 CFR 180; 21 CRF 556) tolerance levels. The method status is further classified according to the source and validation of the method. Examples of method classification status are Official Methods of AOAC INTERNATIONAL, interlaboratory-study-validated methods, Federal Register methods, historical official methods, non-validated methods, published methods, and correlated methods. The confirmatory methods are extremely sensitive and validate the presence of exact residue structures and their concentrations as determined by the mass ion of the molecule, using gas or liquid chromatographic separation and isolation procedures followed by mass spectroscopy. Over the years, the action levels of some drug residues have been lowered because of issues related to “sensitivity of the method,” which refers to the accuracy and precision of measuring the lowest concentration of a compound. Over the years, as chemical separation and isolation chemistry methods have advanced, smaller and smaller amounts of compound have been measured. The relative safety of a drug, or its metabolites in edible tissues or milk, is related to the drugs being present in concentrations that have no substantial risk of toxicity or to its being present in such an innocuous form as to be biologically inert. The issues of toxicity become complex and well beyond the scope of this report when the toxic character of the compound arises from the animal’s metabolism of the drug and not from the drug itself. The metabolism of some drugs varies according to species, and the toxic character of a compound in one animal species is not necessarily the same as that in humans. Residue is defined by CVM as any compound or metabolite of a compound that is present in edible tissues from food animals because of the use of a compound in or on animals. Residues can be from the compound itself, its metabolites, or any other substances formed in or on food as a result of the compound’s use. CVM has a rigorous program for establishing the safety of residues present in food-animal tissues. Data are required for toxicity testing, residue and metabolism testing, and development of analytical methods. Toxicity testing is used to establish the maximum safe residue concentration in the edible tissues of the target animal. CVM evaluates toxicity with tests designed to monitor acute, short-term, and chronic toxicity over time. Within the scope of these tests, concentrations of drug residues are determined that affect morbidity and mortality as well as reproductive toxicity, teratology, and carcinogenicity. For monitor-

OCR for page 110
The Use of Drugs in Food Animals: Benefits and Risks ing the hundreds of possible compounds that might create residues, FSIS decides where available resources and testing efforts should be assigned, and assesses “relative concerns for those residues most likely to have the greatest impact on public health” (FSIS 1995b). Those decisions are made on the basis of data related to (1) the nature of the FDA or EPA withdrawal period, (2) the rapidity with which the compound is biodegraded to nontoxic products, (3) the absorption and excretion patterns and temporal profiles, and (4) the physical stability of the drug or metabolite in the environment (FSIS 1995b). If the tolerance levels of a compound are not available through FDA or EPA, the pharmacokinetics of absorption, excretion, and tissue distribution can be obtained from the literature. The chronic toxicity of a compound is often given a higher priority than is its acute toxicity simply because the chances of tissues having acutely toxic concentrations are remote. Finally, concern for the presence of residues also should be based on patterns of exposure. For example, for most chemical residues that occur in meat, USDA considers the likelihood of ill effects of one-time or infrequent eating of the meat to be of negligible consequence and risk to the population. TRACKING DRUG RESIDUES IN FOOD Investigators can use sophisticated chemical detection methods or drugs labeled with radioactive markers to study the pharmacokinetics, tissue distribution, and metabolism of a drug or test compound and establish the total residue content of the drug present in the edible tissues and in specific test site tissues of treated animals. Typically, muscle, liver, kidney, and fat are analyzed because they are the tissues that are typically eaten in large amounts, tissues that function as storage points for fat-soluble residues, or tissues that metabolize the major portion of the drug in the process of bodily elimination. Drug residue levels in milk and eggs are determined when appropriate. The metabolic profile of the test compound is determined in a sample of each representative edible animal tissue and in animal fluids such as urine or milk, when applicable. Urinary and fecal excretion patterns of the drug are useful in determining the biochemical events that regulate elimination of the drug from the body. For example, biochemical events in liver and kidney increase the aqueous solubility of otherwise poorly water soluble compounds by adding glucuronide or sulfate moieties. Most drugs are either metabolized and broken down to inert forms or metabolically conjugated to anions such as sulfate or glucuronide in the liver or kidney. Another use for the metabolizing and elimination data is to assist in establishing withdrawal times for drugs used in food animals. The task of tracking drug residues would be considerably more complicated if all important tissues from animals had to be tested for residues. To facilitate inspection and detection of the carcass or product, regulatory agencies have determined that a single tissue site should be targeted for routine residue monitor-

OCR for page 110
The Use of Drugs in Food Animals: Benefits and Risks ing. On the basis of pharmacokinetic, drug distribution, and accumulation–depletion data, a target tissue is that tissue from which residues deplete at the slowest rate. In the target tissue, either the parent drug or a metabolite is selected as the marker residue. A tolerance level is then determined for the marker residue. The tolerance level is the concentration of the marker residue in the target tissue when all the residues in every edible tissue are at or below what is considered the safe concentration for that drug. This amount or concentration is derived from an acceptable set of toxicology, metabolism, and residue studies conducted by a drug company that has submitted those data as part of a New Animal Drug Application (NADA). A tolerance level for residues of a drug in the meat, milk, and eggs of food-producing animals is also the amount that is formally established and published at the time of CVM’s approval of the drug. The tolerance level is established to facilitate monitoring drug residue entry into the food chain and to further aid in regulating the uses of animal drugs. Some producers and drug developers are concerned that this conservative measurement practice is counterproductive to the use of many animal drugs, because the drug concentrations measured in marker tissues are irrelevant. The residue concentrations actually consumed in tissues frequently are much lower, and they are eliminated much faster than are those in the marker tissue. Some regulatory pharmacologists believe that the conservative approach is justified because of the possibility that another drug or pathophysiological condition could alter drug metabolism enzymes (for example, cytochrome P 450 complex) and slow the clearance of the drugs from the animal’s body. The maximum residue level (MRL) is not used by the CVM in its regulation of animal drugs. It is used by other countries and by the Codex Alimentarius Commission. In general, the MRL approximates the CVM tolerance level. CVM also uses safe level as a conservative estimate of the residue of a drug in food animals that is considered safe by CVM on the basis of the available safety data. However, it might not be sufficient to set a tolerance level. A safe level is intended to serve as a guide for estimating the safety of residues in meat or milk when no official tolerance level exists. Safe levels are not intended to supplant tolerance levels, and they do not have the same legal status. Generally, safe levels are assigned only when residues appear in meat or milk because of an unapproved use of an animal drug and because a formal tolerance level does not exist. The safe levels for an animal drug are the same for every species of food animal for which the drug is approved. An action level is a conservative estimate of a residue level of an unavoidable contaminant in food that will not pose a human health risk. CVM initiates regulatory action if a residue found in food is above the action level. An analytical method for quantifying residues in various tissues and biological fluids and for measuring the concentration of marker compound must be developed for the approval process. This work is done by the drug manufacturers and submitted as part of an NADA. CVM, in conjunction with USDA, submits

OCR for page 110
The Use of Drugs in Food Animals: Benefits and Risks these analytical methods to independent testing facilities for a systematic battery of tests. The analytical method must be specific, accurate, and repeatable when performed at different laboratories by different personnel. USDA and the FDA review the method and the uniformity of the results developed from the independent testing laboratories to ensure that the assay is based on sound scientific principles and is technically ready for testing. Typically, three USDA and FDA laboratories participate in the trial. If the method passes the trials, it is then submitted to FSIS. DRUG RESIDUES IN MEAT AND POULTRY The National Residue Program (NRP), operated by FSIS, is an essential part of the total inspection effort to prevent adulteration of the meat and poultry supply. Under NRP, FSIS monitors, detects, reduces, and controls violative residues of drugs, pesticides, and other potentially hazardous chemicals and contaminants in meat and poultry products. NRP collects samples of livestock and poultry tissues at slaughtering establishments under its inspection authority and from import shipments at ports of entry. The samples are analyzed for the presence of unacceptable residue concentrations of animal drugs that might contaminate meat and other tissues. Most samples for testing are selected either randomly or based on criteria such as incidence of past violations or questionable practices detected on the farm or processing site. Sometimes, informed sources provide information that leads to testing. To narrow the effort, residue testing in the United States is divided into two major activities: animal population and product sample testing programs (monitoring, exploratory, and surveillance) and violation enforcement. Monitoring provides annual profiles on the occurrence of residue violations in specified animal populations. Compounds are selected on the basis of potential hazard and on the availability of a laboratory method suitable for regulatory monitoring. Information is obtained through a statistically based random selection of samples of normal-appearing tissues from carcasses that have passed visual inspection. Generally, the number of samples provides a 95 percent probability of detecting at least 1 violation when 1 percent of the sampled population is violative. In addition to profile information, the monitoring program can identify producers marketing animals with violative residues. When such producers subsequently offer animals for slaughter, the animals will be subject to surveillance sampling and testing until compliance is demonstrated. The collected data also indicate incidences and levels of residues; enabling evaluation of residue trends and identification of problems within the industry where educational or other corrective efforts might be needed. Exploratory projects are conducted for a variety of reasons, but whatever their objective, they have in common the fact that test results normally are not used to take regulatory action or to trigger follow-up surveillance testing. For

OCR for page 110
The Use of Drugs in Food Animals: Benefits and Risks example, FSIS might conduct a study to develop information on the incidence and concentrations of a trace metal, industrial chemical, or animal drug for which no safe level has been established. Surveillance is instituted for investigating and controlling the movement of potentially adulterated products. Sampling is biased and is directed at particular carcasses or products in response to information from monitoring programs. In surveillance, the carcasses and organs might be retained until test results are available. In enforcement testing, specimens are obtained from individual animals or lots based on herd history. Testing is performed to detect individual animals with violative levels of residues. It is emphasized in problem (high-prevalence) populations and used as a tool to prevent residues from entering the food supply. Through 1996, NRP made use of the Compound Evaluation System (CES) to provide a systematic approach to categorizing compounds with respect to the likelihood of their occurrence in meat and poultry and their potential consequences for public health. CES evaluated the risk of residues in meat and poultry on the basis of hazard (adverse effects that might result from a given compound), and exposure (residues and factors affecting concentrations, such as drug use patterns, withdrawal times, and frequency of consumption) (FSIS 1995b). CES ranked compounds in 24 categories. Compounds of greatest concern were designated A-1—those with a high health hazard potential and high likelihood of residue occurrence. Compounds of least concern were designated D-4—those with a negligible health hazard potential and negligible likelihood of residue occurrence. The CES coding was an alphanumeric system expressing the two parts of the risk assessment where it is apparent that the detection of a residue would be possible. Compounds are included in the NRP monitoring plan if they leave a detectable residue in meat and poultry and have an established tolerance level, action level, or other referenceable regulatory limit. FSIS must have a suitable regulatory method that has been validated as capable of confirming the identity and quantity of the residue. A compound can be cycled out of the NRP monitoring plan when its residue potential has been evaluated and is no longer of concern. The exposure potential of such compounds is evaluated annually. When information indicates a possible increase in exposure potential, the compound is reconsidered in the plan. In 1997, FSIS performed monitoring analyses of 7,375 samples of meat and poultry for residues of 12 antibiotics and 7,284 samples for residues of 4 sulfonamide drugs. The service also monitored 1,056 samples of food animals for arsenic compounds; 7,409 samples for 27 chlorinated hydrocarbons and organophosphates; 1,196 samples for halofuginone; 3,327 for ivermectin; and 4,101 samples for levamisole. In addition, enforcement-testing analyses were performed: 219,193 samples for antibiotics; 15,638 samples for sulfonamides; 12 samples for arsenic; 296 samples for chlorinated hydrocarbons and organophos-

OCR for page 110
The Use of Drugs in Food Animals: Benefits and Risks phates; 31 samples for ivermectin; 1 sample for levamisole; and 324 samples for clenbuterol (FSIS 1997). Data on the number of violations identified and violation rates are shown in Table 5–1. The data in the table suggest that the risk of violative residues entering the food chain is very low. Most measured residues are obtained from tissues (liver, kidney, fat) that are much slower to clear (deplete) residues than is muscle (measurable amounts do not in themselves constitute violative contamination levels). There is a relatively large safety margin in the use of animal drugs when proper withdrawal times and uses are followed. The introduction of more rapid tests (such as FAST, the fast antimicrobial screen test) for antibiotics, which allow more samples to be tested suggests that the actual violation rate could be lower than that estimated given fewer samples. Greater assurance that the food chain is protected from contamination with drugs, chemicals and other compounds is evolving in the restructuring of new monitoring processes by FSIS (1998) and with a reclassification of health risk ranking of the various monitored drugs. The assessment of risk embraces safety aspects of residues in foods as they might affect the health of populations where the effect is greatest, even though the size of populations could be very small. For example, penicillin is (and was under the old CES system) ranked relatively high (A-1) in the risk assessment. This was not due to any inherent toxicity of penicillins (as might be more readily equated with chlorinated biphenyls, for example) but rather is a result of the fact that, for some persons with hypersensitivity to penicillins, contact with these residues might pose a life-threatening risk (Hoffman, M. 1998. FSIS, personal communication). Certainly, penicillins are used in human medicine, but the risk to the individual is basically minimized by the patient’s knowledge of the intended use and because of physician prescription and oversight. In the case of food residues, the consumer does not usually know what drugs have been used in the food animal from which a product is obtained. Protection is enhanced by assigning a relatively high risk score and allowing a very low level of permitted residue. A new ranking system is being phased in through 1998 that assesses the health risk through a mathematical function that considers the pharmacokinetic distribution and elimination of drugs in animals, as well as the likelihood of residue consumption, and the inherent toxicological properties of compounds and metabolites. In addition, the evaluation system further refines tolerance levels for different drugs and compounds with regard to slaughter class–compound pairs. The applications and uses of drugs differ and so the chance for residues to occur varies because of differing animal husbandry practices for various species and ages of animals within species. For the purpose of statistical sampling, the numbers of samples and analyses are established to obtain a 95 percent probability of detecting at least 1 violation when 1 percent of the animal population could be theoretically in violation. For example, the number of samples requiring analysis (random sampling) to obtain this 95 percent probability is 299. This

OCR for page 110
The Use of Drugs in Food Animals: Benefits and Risks TABLE 5–1 FSIS Animal Drug Residue Test Results   Samples Violations Violation Rate % Drug Residue Testing 1994 1997 1994 1997 1994 1997 Antibiotics Monitoring program 8,354 7,375 19 9 0.23 0.12 Enforcement testing 211 118 8 10 3.8 8.4 CASTa 65,059 21,045 948 169 1.46 0.803 STOPb 102,521 41,995 1,046 292 1.02 0.69 FASTc 30,343 255 156,078 1,024 0.84 0.65 Sulfonamides Monitoring program 8,098 7,284 23 17 0.28 0.23 Enforcement testing 276 38 98 18 35.5 47.4 SOSd 166,091 15,600 104 24 0.06 0.15 Arsenic Monitoring program 2,223 1,056 5 6 0.22 0.56 Enforcement testing 66 15 1 0 1.52 0.0 Halofuginone Monitoring program 629 1,196 0 0 0.0 0.0 Enforcement testing 10 0 0 0 0.0 0.0 Ivermectin Monitoring program 3,926 3,327 7 6 0.18 0.18 Enforcement testing 7 31 0 0 0.0 0.0 Levamisole Monitoring program 4,077 3,846 6 0 0.15 0.0 Enforcement testing 59 1 3 0 5.08 0.0 Morantel tartrate 2,478 —e 1 — 0.04 — CHC–COPSf Monitoring program 9,109 7,409 9 7 0.09 0.09 Enforcement testing 90 298 0 19 0.0 0.06 aCAST = Calf antibiotic and sulfonamide test. bSTOP = Swab test on premises. cFAST = Fast antimicrobial screen test. dSOS = Sulfa-on-site. eNot done. fCHC–COPS = Chlorinated hydrocarbons and organophosphates. Source: FSIS 1995, 1998.

OCR for page 110
The Use of Drugs in Food Animals: Benefits and Risks increases to 688 if the 99.9 percent probability of detection of one positive in a one percent violative rate and to 13,813 if the desired level of assurance is 99.9 percent in a population with a 0.05 percent violation rate (FSIS 1997). DRUG RESIDUES IN FISH AND SEAFOOD The amount of fish and seafood consumed in the United States is modest compared with meat and poultry (about 15.5 pounds per person annually). Much of that fish and seafood is caught rather than farmed. Nevertheless, aquaculture is a growing industry. Five drugs are approved by the FDA for use in aquaculture. Another 4 are approved by the U.S. Department of the Interior’s Fish and Wildlife Service for use in hatcheries that supply sport fishing (FWS 1994). However, many more drugs are believed to be used in an extra-label fashion in aquaculture. Monitoring of animal drug residues in farmed fish and seafood is the responsibility of the FDA Office of Seafood, which began its small monitoring program in 1991. In 1993, that office analyzed 105 samples of domestic and imported salmon and shrimp. Catfish, the largest aquaculture species, was not tested for animal drug residues. Through 1994, the office tested for the presence of two drugs: chloramphenicol in shrimp and oxolinic acid in salmonids. Both are illegal for use in cultured fish. Of 50 samples taken for chloramphenicol testing, 1 violation was found. There also was 1 violative residue of oxolinic acid in 26 samples. In 1995, monitoring of chloramphenicol in shrimp and oxolinic acid in salmonids was again conducted. No violative residues were detected in 36 samples tested for chloramphenicol or in 66 samples tested for oxolinic acid. The process for choosing which drugs to test involves using the same questions of hazard and exposure as used by FSIS but is much less formal. FDA monitoring of aquaculture products also is constrained by a lack of test methods. DRUG RESIDUES IN MILK FDA has the primary responsibility for regulating milk. Its milk safety program relies on participation by state regulatory agencies. The National Conference on Interstate Milk Shipments (NCIMS) is a cooperative program of the states and U.S. Public Health Service for certification of interstate milk shippers. Its procedures, administration, and enforcement actions provide the framework for the nation’s Grade A milk safety program (FDA 1995b). Every tanker-truck milk load entering a dairy processing plant is tested for drug residues. The only official test for detecting drug residues under the Pasteurized Milk Ordinance (PMO) is the Bacillus stearothermophilus disk assay. That test is effective in detecting 4 drugs in the penicillin family (ß-lactams). However, FDA has approved 53 drugs for use by dairies, including 20 antibiotics. FDA believes that more than 78 drugs might be used in legal and illegal

OCR for page 110
The Use of Drugs in Food Animals: Benefits and Risks preparations in dairy cows (GAO 1990). Data from the FDA check ratings (inspections of selected dairy farms and validated state inspection programs) in 1990 and 1991 found 62 drugs not approved for use in dairy cows; 42 drugs were not approved for any use in food animals (GAO 1992). With the support of NCIMS, FDA initiated the National Drug Residue Milk Monitoring Program (NDRMMP) in February 1991 (FDA 1995b). The program has the following objectives: Provide an indication of the animal drug residues that might be present in milk. Provide an indication, through follow-up investigations, of the extent to which farmers, distributors, and veterinarians comply with FDA regulations for the proper sale, distribution, and use of drugs in dairy cattle. Assist federal, state, and local milk officials in designing educational and enforcement programs by providing information on drug residues in milk. Facilitate the transfer of analytical methods and technology from FDA to state and industry laboratories. The number of samples tested within this program is small relative to the milk supply. In 1993, the program analyzed 357 milk samples for 8 sulfa drugs, 3 tetracyclines, 4 ß-lactams, and chloramphenicol (CVM 1993a,b,c; 1994). The tests found only 1 violative residue of a ß-lactam and 4 nonviolative residues of sulfadimethoxine. Violation rates for drugs tested in milk are extremely low. One reason might be that penalties for violations found at the dairies are immediate and severe. Usually, a tanker-truck pools milk from several farms. When a violation is found, further testing is done of samples from each farm. The offending farmer is responsible for finding a site to dump the milk and for reimbursing other farmers for the loss of the load at current prices of $5,000 or more (Carlson 1994). As of October 31, 1994, all 50 states and Puerto Rico participate in the database program. The database includes results of NDRMMP, as reported by the states. Although all 50 states and Puerto Rico participate in the program, it is important to recognize that the samples and tests reported do not necessarily represent 100 percent of the milk supply from every state because the program is voluntary. However, as state participation in the database program has increased, reporting of the number of samples and tests also has increased. Between October 1, 1995, and September 30, 1996, 4,565,600 samples of milk were analyzed for animal drug residues. Of these, 5,404 tested positive for a residue. The breakdown of these results by sample source is shown in Table 5–2. The data show that the rate of occurrence of any residues in pasteurized fluid milk is extremely low.

OCR for page 110
The Use of Drugs in Food Animals: Benefits and Risks Ground Meat Processed Product Critical Source Mean References Prevalence 46.0 44.3 Animal C1–6 27.8 0.0 —a Animal C7–13 8.6 — — Animal C14 2.0 100.0 12.0 Animal C15–16 78.0 — — Animal C17–19 1.1 65.6 30.0 Environment C20-26 29.0 — — Environment C27–29 18.0 20.0 — Environment C30 17.3 23.0 73.5 Human C31–36 48.7 C18 Griffin and Tauxe 1991 C19 FSIS 1994C C20 Johnson et al. 1990 C21 Grau and Vanderlinde 1992 C22 Wang and Muriana 1994 C23 Vorster et al. 1991 C24 Wang et al. 1993 C25 Pinner et al. 1992 C26 FSIS 1994C C27 Smart et al. 1961 C28 Ternstrom and Molin 1987 C29 FSIS 1994C C30 Konuma et al. 1988 C31 Sokari and Anozie 1990 C32 Farber et al. 1987 C33 Ternstrom and Molin 1987 C34 Nwosu 1985 C35 Vorster et al. 1991 C36 FSIS 1994C significant by itself in contributing to the development of a disease associated with the product. More to the point, the microbial load (the amount of bacteria per gram of product) is critical and is certainly affected by postslaughter practices in storage and handling. Any point in the process of food handling at which the meat or milk can warm to bacteria-proliferating temperatures complicates and escalates the potential for food-borne disease to occur. Similarly, the nature of the finished retail product is a considerable factor in the emergence of food-borne disease. The potential for food infection is greater in ground meat and poultry than it is in an intact muscle product like a steak or chop because of the mixing

OCR for page 110
The Use of Drugs in Food Animals: Benefits and Risks TABLE 5–5 Survey Report of Microbiological Hazards in Lamb   Mean Percentage of Samples Yielding Pathogenic Bacteria in Lamb Human Pathogen Carcass Fresh Meat Meat Organ Salmonella spp. 40.0 0.0 —a Campylobacter jejuni — 15.0 — Aeromonas hydrophila — 53.0 — Escherichia coli O157:H7 — 2.0 — Listeria monocytogenes — 37.0 — Clostridium perfringens — 85.0 — a Not tested. L1 D’Aoust et al. 1992 L2 Farber et al. 1988 L3 Lammerding et al. 1988 L4 Stern 1981 L5 Stern et al. 1984 and spreading of contaminating organisms within and across a greater surface area. Similarly, retail repackaging, transportation to the place of consumption, and cooking habits and preferences all affect the character of disease emergence. Thus, important difficulties arise in tracking down and locating the source of contamination and initiating events that allow the development of food-related disease. Again, the use of antibiotics is thought to be important in reducing the potential bacterial load that might be transferred to meat up to the point of slaughter. Beyond that point, the prior use of antibiotics will not affect proliferation of bacteria in food. A final word should be said about host susceptibility to disease from microorganisms of animal origin. Disease becomes a relative term if one considers or equates the presence of a bacterium or microorganism with the development of disease. Humans and animals can develop the capacity to resist disease from various organisms as well as different microbial loads. For example, children living in India might not be severely affected by some water-borne microorganisms that would be intolerable to visitors from the West. In the animal world, scavengers eat carcasses of dead, diseased animals and appear to suffer no apparent ill effects. This again reflects the complex interplay between biology, ecology, and culture as it affects clinically significant cases of disease.

OCR for page 110
The Use of Drugs in Food Animals: Benefits and Risks Ground Meat Processed Product Critical Source Mean References Prevalence 75.0 — Animal L1–3 47.5 — — Animal L4–5 12.5 82.5 52.3 Animal L6–8 62.5 — — Animal L9 2.0 — — Environment L10 37.0 — — Environment L11 85.0 L6 Marjeed et al. 1989 L7 Palumbo et al. 1985 L8 Hudson and DeLacy 1991 L9 Doyle and Schoeni 1987 L10 Johnson et al. 1990 L11 Smart et al. 1961 A Nine-Year Survey of Reported Food-Borne Illness Although the individual incidences of food-related illness is in the millions of cases per year, outbreaks are fewer in number but affect a common population of significant size. Of 4,821 outbreaks of food-borne illness reported from 1983 to 1992 to the CDC (Bryan 1980; Bean and Griffin 1990; CDC 1994), 2,114 were traced to the actual food sources. Pathogens present in meat- or poultry-related products were responsible for 334 (15.8 percent) of the outbreaks traced to the source. Of the 310 outbreaks confirmed to be of animal-product origin, 152 (45.5 percent) were of unknown etiology. CDC reports for 1983 to 1992 included 78 outbreaks traced to milk, ice cream, cheese, and other dairy products. There were 1,061 cases of milk- and dairy-product-borne diseases for the 36 outbreaks from 1988 to 1992. One death was recorded. Large outbreaks of food-borne diseases associated with milk and dairy products are reported in the literature; small outbreaks, unless of unusual etiology, are not reported. The committee reviewed published reports of international events for 1980 to 1996. An outbreak of gastrointestinal disease affecting at least 110 people in England caused by Campylobacter jejuni was associated with drinking inadequately pasteurized milk (Fahye et al. 1996). Seventy-two laboratory-confirmed cases of Campylobacter jejuni infection were identified in people who drank unpasteurized milk at a festival in England (Morgan et al. 1994). From Septem-

OCR for page 110
The Use of Drugs in Food Animals: Benefits and Risks TABLE 5–6 Survey Report of Microbiological Hazards in Poultry   Mean Percentage of Samples Yielding Pathogenic Bacteria in Poultry Human Pathogen Carcass Fresh Meat Organ Meat Salmonella spp. 47.4 41.9 52.7 Campylobacter jejuni 66.2 52.7 63.3 Aeromonas hydrophila 98.0 50.0 100.0 Listeria monocytogenes 22.0 23.8 7.0 Clostridium perfringens 79.0 — — Bacillus cereus — 21.5 — Staphylococcus aureus — 40.0 — Escherichia coli O157:H7 — 1.5 — aNot tested. P1 Kanarat et al. 1991 P2 James et al. 1992 P3 Giese 1992 P4 D’Aoust et al. 1992 P5 Vorster et al. 1991 P6 Berndtson et al. 1992 P7 Lillard et al. 1984 P8 Barrel 1987 P9 Duitschaever and Buteau 1979 P10 Farber et al. 1988 P11 Izat et al. 1989 P12 Lammerding et al. 1988 P13 Tokumaru et al. 1991 P14 Jergklinchan et al. 1994 P15 Lillard 1990 P16 Norberg 1981 P17 Castillo-Ayala 1992 P18 Stern and Line 1992 P19 Slavik et al. 1994 P201 Norberg 1981 P21 DeBoer and Hahne 1990 P22 Harris et al. 1986 P23 Hood et al. 1988 P24 Jones et al. 1991 P25 Marinesca et al. 1987 P26 Flynn et al. 1994

OCR for page 110
The Use of Drugs in Food Animals: Benefits and Risks Ground Meat Processed Product Critical Source Mean References Prevalence —a 56.3 Poultry P1–16 46.9 — — Poultry P17–32 61.9 — 6.0 Poultry P33–35 63.5 — 32.5 Environment P36-48 25.4 — — Environment P47 79.0 — — Environment P48–49 21.5 — — Human P50 40.0 — — Human P51 1.5 P27 Gill and Harris 1984 P28 Kinde et al. 1983 P39 Roberts and Murrell 1993 P30 Shanker et al. 1982 P31 Tokumaru et al. 1991 P32 Christopher et al. 1982 P33 Barnhart et al. 1989 P34 Palumbo et al. 1985 P35 Hudson and DeLacy 1991 P36 Johnson et al. 1990 P37 Wang and Muriana 1994 P38 Vorster et al. 1991 P39 Wang et al. 1993 P40 Kerr et al. 1990 P41 Yarabioff 1990 P42 Wenger et al. 1990 P43 Bailey et al. 1989 P44 Genigeorgis et al. 1990 P45 Genigeorgis 1989 P46 Pinner et al. 1992 P47 Lillard et al. 1984 P48 Konuma et al. 1988 P49 Sooltan et al. 1987 P50 Vorster et al. 1991 P51 Doyle and Schoeni 1987

OCR for page 110
The Use of Drugs in Food Animals: Benefits and Risks ber 6 through October 10, 1994, 142 cases of Salmonella enteritidis infections in a 3-state area were associated with raw egg contamination of ice-cream mix supplied to a large manufacturer (CDC 1994). In 1989, 164 cases of Salmonella javiana and Salmonella oranienburg infections in 4 states were associated with mozzarella from a single cheese plant. Those cases were identified in retrospective epidemiological studies (Hedberg et al. 1992). In 1985, a multiple-antibiotic-resistant strain of Salmonella typhimurium bypassed faulty pasteurization in a large dairy-processing plant, resulting in an outbreak of more than 1,600 cases traced by the unusual plasmid profile of the organisms (Schuman et al. 1989). From 1980 to 1983 in California, more than one-third of reported cases of Salmonella dublin infections were attributed to consumption of raw milk. Among consumers of raw milk, more than 95 percent of reported S. dublin infections were associated with drinking contaminated raw milk. A single, major certified dairy in California was associated epidemiologically with that incidence (Richwald et al. 1988). The shifting patterns of disease outbreak over time as a function of demographics and cultural practices are exemplified by the epidemiology of infections of people with Brucella melitensis. Thirty-one cases of infection in residents of a community in Houston, Texas, were traced to unpasteurized goat milk cheese imported from Mexico (Thapar and Young 1986). In a larger study, of the 332 laboratory-confirmed cases of human brucellosis in Texas between 1977 and 1986 (Taylor and Perdue 1989), the patterns of emergence were different between 1977 and 1981 than they were between 1982 and 1986. In the earlier period, 82 percent of the cases were in males and 52 percent of infections were in white people. For the next five years, only 55 percent of the cases were in males and 72 percent of the cases were in people of Hispanic origin. Ingestion of unpasteurized goat milk was reported in 67 percent of the cases between 1982 and 1986. From the early 1980s to the present, the trend shows that the incidence of microbially related food-borne illness has increased. Generalizations regarding apparent trends need to be made cautiously, because many factors other than the actual number of cases are relevant. For example, reporting of food illnesses has not always been as uniform or widespread as it is now. CDC estimates that only 10 percent of cases are actually reported. Education, awareness, and observation and diagnosis have certainly assisted in increasing the number of food-borne illnesses reported. Similarly, conclusions on a cause-and-effect relationship between antibiotic use in food animals, on-farm food-animal production practices, and the incidence of disease should not be made quickly or in the absence of definitive, trackable proof and data. As seen in Figure 5–1, the incidence of reported cases of food-related illness in all major sentinel organisms has a seasonal component, and the largest number of reported cases of illness occur in July and August. In the United States, ambient environmental conditions are more amenable to increased bacterial proliferation in a shorter time. Those data can be

OCR for page 110
The Use of Drugs in Food Animals: Benefits and Risks FIGURE 5–1 Seasonality of Reported Cases of Food-Borne Disease as Monitored Across the Seven Sentinel Organisms at All Locations. Source: FSIS 1997. used as indirect evidence that postproduction practices have an important effect on the number of bacteria entering the food supply. In Minnesota, a state with a relatively high incidence of Campylobacter infections, the seasonality of reported cases shows a secondary increased incidence of Campylobacter reported during winter months, when the outbreak trends would be expected to be low. These cases of Campylobacter infections were traced to Minnesotans’ vacations in Latin America during the cold Minnesota winter (Smith et al. 1997). INTEGRATING ISSUES OF RESIDUES AND MICROBIAL CONTAMINATION Microbial contamination of food is a principal area of food safety for which hazard analysis and critical control points (HACCP) will have a major potential to curb the transfer of microorganisms to humans via food. Issues related to drug residue contamination of animal-derived food products are largely related to the introduction of drugs and compounds into the live animal prior to slaughter or milking. In contrast, microbial contamination of foods, especially where the contamination is animal-feces-related, is an issue at the time of product harvest, and it has tremendous potential for risk to increase with improper food handling prior to consumption. Practical considerations affect the further propensity for organisms to cause disease, including poorly defined interactions of contaminating bacteria with the environment found in the food product (acidity, salinity, cold-storage effects). The magnitude and severity of transference of human

OCR for page 110
The Use of Drugs in Food Animals: Benefits and Risks pathogens are not inconsequential, and still they are relatively difficult to assess because of a paucity of validated evidence and difficulty in tracking. The extent of the threat might increase or decrease with the identification of solid data to evaluate and draw conclusions. These data will take some time to amass, and recommendations based on the new data might be substantially different from those made today. A major concern that arises when food-borne bacteria cause illness relates to the potential for a bacterium to become invasive and, therefore, to require medical intervention through the use of antibiotics. Within the context of microbial contamination of food, antibiotic-resistant bacteria constitute a subpopulation of organisms that, when present, can be carried within the food product to pose a formidable challenge to treatment and remediation of disease in humans. Based on the number of reported cases, the threat of disease and illness occurring through food contaminated with microorganisms is much greater than the threat of resistance transfer from animals to humans. To put the comparison of risks into perspective, there are few published reports of resistance transfer from animals to humans, and there are thousands of reported cases and outbreaks of foodrelated microbial-derived illness in the United States. A significant gap exists in data that could be used to track the occurrence of food-borne bacterial disease and correlate incidents with the presence of antibiotic resistant bacteria. More precisely, the relationship between the actual disease caused by a specific antibiotic-resistant organism needs to be tracked as well as the occurrence where a genetic element of resistance was transferred to bacteria resulting in disease. This is a formidable task, requiring significant financial input and scientific effort. However, if an accurate assessment of the relationship between antibiotic use in food animals, the occurrence of resistance and disease in humans, and the risk to human health are to be established, such data must be collected and analyzed. According to FDA (CFSAN 1997) and USDA (FSIS 1997), as many as 33 million illnesses and 9,000 deaths occur each year in the United States as a result of food contamination by only 7 pathogens. The yearly medical and productivity costs for those illnesses are estimated to be almost $35 billion. Most of the illnesses arising from contaminated food result in diseases of “inconvenience”: Vomiting, diarrhea, and muscle cramping are the symptoms most often observed. Recently, concern about the many microorganisms that infect food is increasing, because new strains of bacteria are being discovered that confound and complicate treatment. Those include the multidrug-resistant forms of Salmonella (definitive type DT-104) and forms of E. coli not only that have increased intestinal tissue adherence properties but also have the ability to secrete Shiga-like toxins (E. coli O157:H7). Achieving the goal of safe food requires a multifocal planning and entails cooperation between individuals and agencies, extending from the farm to the consumer. Achieving the goal was addressed as a national priority by President

OCR for page 110
The Use of Drugs in Food Animals: Benefits and Risks Clinton in a report entitled “Food Safety from Farm to Table: A National Food-Safety Initiative,” (CFSAN 1997). As summarized by this report, many levels of responsibility must be functioning simultaneously to keep the food supply safe: On the Farm State agencies and EPA oversee the use of pesticides. FDA oversees the use of animal drugs. Animal Plant Health Inspection Service oversees animal and plant disease. The Clean Water Act lessens the environmental burden of animal waste. At Processing FDA and USDA monitor the processing of foods to detect residue contamination and microbial pathogens. Transport and Import FDA and USDA regulate pertaining to interstate and international food transportation and importation. Food Services State and local agencies, FDA, and USDA develop laws for the safe handling and preparation of foods by schools, governing hospitals, restaurants, and so forth. Water Standards EPA establishes and maintains water standards, and state and local agencies oversee local standards. Education USDA Cooperative State Research, Education and Extension Service (CSREES). Illness Surveillance CDC and state health departments. Food Research CDC, FDA, ARS, CSREES, and EPA research ways to improve quality control and pathogen and residue detection; HACCP strategies for by-product rendering, slaughter practices, and preparation; and disease prevention strategies. Consumer Largely responsible for the safety of the final food as served. Education and awareness can help preserve the integrity of foods. The consumer has the last chance to identify potentially tainted food before it is eaten. Decreasing the extent and severity of food-related illness is predicated on the establishment of an early-warning system to identify and track the sources of illness. That is a formidable goal and difficult enough to oversee within the U.S. production–consumer system, but it is further complicated by increasing global trade and, perhaps, by inadvertent entry into the United States of contaminated produce from other countries. To accomplish the task, an aggressive plan has been proposed in the National Food Safety Initiative to increase resources for the following:

OCR for page 110
The Use of Drugs in Food Animals: Benefits and Risks Increase food-borne disease surveillance with increased site participation in the National Antimicrobial Susceptibility Monitoring Program–Veterinary Isolates and the Foodborne Diseases Active Surveillance Network (FoodNet, a cooperative project of federal, state, and local agencies with oversight by USDA, FDA, and CDC, that monitors incidences of food-borne illness and the effectiveness of food safety programs to diminish the incidence of food-borne illness). Enhance outbreak detection by supporting the development of electronic instrumentation and increased communication between agencies. Modernize public health laboratories to increase diagnostic capacity. Create a “pathogen fingerprint” database. Increase national surveillance of antibiotic resistance, initially concentrating on Salmonella and E. coli contamination. Increase surveillance of human pathogens in food-animal populations, feed, and manure, with special emphasis on identifying the mechanisms of antibiotic resistance. Enhance local and state infrastructures to improve detection, evaluation, and response to disease outbreaks. Establish a risk assessment consortium to set priorities for research and data collection, serve as a clearinghouse for data, assess effectiveness of current measures, and recommend modifications and amendments to plans. Fund the research necessary to establish directives in pathogen avoidance and reduction in food animals, food handling, and storage and cooking, and charge the Office of Science and Technology Policy with coordinating the implementation of these directives in federal programs. SUMMARY OF FINDINGS AND RECOMMENDATIONS The committee finds that the general public health is adequately protected by the current system of drug residue surveillance. However, the drug-residue-monitoring system could be improved. There are gaps in testing programs. The number of samples tested is large for meat, poultry, and milk but small for fish. Current incidence of opportunistic pathogen blooms and the consequences for aquatic and human health suggest that putting additional effort into monitoring programs for fish and shellfish would be beneficial. A serious limiting factor for test programs is the lack of rapid and specific test methods. In general, FDA lacks the analytical methods to test for many extra-label uses of drugs (including human drugs) in food animals. In addition, imported foods might need to be monitored more carefully because of the potential introduction of drug residues and microbial pathogens from countries with farm production practices and quality-control measures that are less stringent than those in the United States. This is a concern because the pharmacokinetics of drug elimination could be different in diseased or recovering animals. Few data

OCR for page 110
The Use of Drugs in Food Animals: Benefits and Risks are collected that allow comparison of drug metabolism in healthy and diseased animals. FSIS is planning to devote more resources to the detection and prevention of microbial contamination of meat and poultry products. Recommendations The committee recommends that more resources be made available for developing appropriate analytical methods for a wider range of drugs. In addition, the committee recommends increasing the specificity of certain tests to reduce the inordinate number of false-positive results, particularly in the inspection of milk. The high false-positive rate can cause unnecessary dumping of large quantities of milk and it causes unnecessary concern among consumers about the safety of the milk supply. The committee recommends that resources be identified to support the initiatives of programs that monitor residues and microbial contamination. The private sector should be a partner in this initiative and share in the management and funding of the programs. The committee recommends that further research be pursued to develop more rapid analytical procedures and increased specificity for residue analysis as well as microbial contamination. Additional research should focus on residue depletion and the pharmacokinetics of drugs in diseased animals.