Michael Walker and Kirstin Gray of LGC discuss why and how veterinary residues are controlled in food and describe current problems arising from technical appeals to the Government Chemist in this area.
Humane care of animals includes administration of veterinary medicines for prophylaxis and treatment of specific diseases and control of parasites. This benefits animal welfare through alleviation of pain and suffering and makes the supply chain, where animals are reared to produce food, more secure. Animal husbandry and aquaculture also use veterinary drugs (mainly antibiotics) for generalised prophylaxis with increased intensive rearing and for growth promotion. Antimicrobials in feed improve feed efficiency resulting in a shorter time to achieve market weight.
Prescribed medication to treat human disease, e.g. chloramphenicol eye drops for bacterial conjunctivitis, is generally accepted. But chloramphenicol in our food supply is not – even at concentrations 10 million times lower than the 0.5% in eye drops most of us have used at one time or another. Why is this and what other problems exist with residues of veterinary medicines in food?
Antimicrobial resistance (AMR)
AMR is a global problem, which renders it more difficult, or impossible, to treat an increasing range of infections. Left unchecked, by 2050 AMR could be causing the deaths of 10 million people a year across the world, with $100 trillion in cumulative lost economic output. Addressing AMR demands a holistic approach in human and veterinary medicine and throughout the food supply chain. A recent paper by van Bunnik and Woolhouse[1] challenges the intuitive notion that curtailing the volume of antibiotics consumed by food animals has, as a standalone measure, any impact on the level of AMR in humans.
However, responsible use of antibiotics in agri- and aquaculture is recognised by all stakeholders. In the EU, approval for antibiotic growth promotors has been withdrawn and it is being phased out in the USA. Industry organisations, such as the British Poultry Council, advocate responsible use of antibiotics and antibiotic usage in the poultry industry was reduced by 44% between 2012 and 2015. RUMA (Responsible Use of Medicines in Agriculture) responded to the van Bunnik paper saying the food and farming sector should not dilute its current focus on reducing, refining and replacing antibiotic use across all sectors[2]. An accessible discussion of AMR is provided in the FSA Chief Scientific Adviser’s 4th Science Report on Antimicrobial resistance in the food supply chain[3].
Foods, such as milk, honey, poultry meat, beef and pork, fish and other seafoods, have been found contaminated with chloramphenicol'
Allergic reactions and other toxicity
Some of the other risks of veterinary residues in food are discussed below.
Penicillins
It is estimated that 4–11% of the population are allergic to penicillin and its analogues. IgE (Immunoglobulin E) mediated reactions can range from minor skin rash to severe anaphylaxis risking fatalities. The immunogenicity of penicillins is not based on the drug itself, but on protein adducts formed after the β-lactam ring is opened. There are confirmed cases of excess penicillin residues in beef and pork causing anaphylaxis not attributable to sensitisation to the meat.
Chloramphenicol
Foods, such as milk, honey, poultry meat, beef and pork, fish and other seafoods, have been found contaminated with chloramphenicol. The most significant adverse effect from human chloramphenicol medication is aplastic anaemia, an effect for which no dose relationship or threshold has been identified. Although there appear to be no data implicating chloramphenicol residues as a cause of aplastic anaemia, only a small exposure may be necessary in susceptible individuals. Chloramphenicol is also genotoxic and a possible carcinogen, although the limited number of studies on its carcinogenicity do not allow a definitive classification. Interestingly, chloramphenicol produced by soil bacteria may be transferred to crop plants.
Sulfonamides
There are no reported cases of sulfonamide-related adverse effects from food. However, adverse drug reactions in humans are common, with skin reactions (mild to severe), haemolytic anaemia, other blood disorders and hepatitis being reported. Thyroid cancer is also a potential risk and sulphonamides can be potent contact sensitisers. All of which, coupled with historically high incidences of positive findings in food make sulphonamides a continuing major regulatory concern.
Phenylbutazone
In the 2013 horse meat scandal, phenylbutazone might be regarded as something of a ‘near miss’. An anti-inflammatory legitimately used in horses, phenylbutazone was marketed as a medicine for human use in the United States for the treatment of rheumatoid arthritis and gout in 1952. Accounts of serious and sometimes fatal adverse effects soon appeared in the literature and it was largely withdrawn. Phenylbutazone is now only used in humans for ankylosing spondylitis, a type of arthritis, where other treatments have failed. Although the levels used to treat humans are thousands of times higher than would be expected to be found in horse meat of treated animals, some of the effects of phenylbutazone are reported to be idiosyncratic, not dose-dependent, and in theory might occur at any dose. The assessment of a safe residue level of phenylbutazone in the meat of food producing animals was never completed because key information is not available. Thus, if phenylbutazone is administered, the animal must never enter the food chain.
Beta agonists
In 1990 there was an outbreak in Spain of food poisoning affecting 125 people who had consumed beef liver. Symptoms included muscle tremor, palpations, tachycardia, nervousness, headaches and myalgia. The illness was mild and there were no deaths. Clenbuterol was found in the urine of two patients and in the beef liver consumed (160-291 ppb). Two similar outbreaks occurred the following year in Spain and sporadic outbreaks of clenbuterol toxicity continue to be reported globally. Clenbuterol is a beta adrenergic agonist, or β-agonist, which acts upon the beta adrenoceptors mimicking the action of adrenaline and noradrenaline. In brief, β1 agonists act on the cardiac muscle, leading to increased heart rate and blood pressure, while β2 agonists induce smooth muscle relaxation in the lungs, one of the most common being salbutamol used to treat asthma. However, β-agonists given to cattle late in the feeding period result in consistent increases in the rate of muscle weight gain without an increase in feed consumption. They became notorious for this illegal use, as well as the corresponding (and correspondingly questionable) use by some body-builders, athletes and slimmers.
Risk assessment and management
Given the actual and potential harm caused by veterinary residues, systems of risk assessment and risk management are in place in all the major trading blocs. Figure 1 summarises the basic process.
Risk assessment also includes premarket approval for new veterinary drugs and rules on medicated animal feeds. Risk management includes the imposition of withdrawal periods to ensure residues of the active constituent will not exceed the MRL when the label instructions for the product are followed and that the drug is withdrawn from the animal in good time. Residue monitoring and surveillance for compliance with the applicable MRL also take place (see for example Council Directive 96/23/EC on measures to monitor certain substances and residues in live animals and animal products).
Toxicological evaluation of veterinary medicines is carried out by various organisations, such as JECFA, the FDA and the European Medicines Agency (EMA). EMA has a remit to evaluate both human and veterinary medicines. JECFA is the Joint FAO/WHO Expert Committee on Food Additives, administered jointly by the Food and Agriculture Organization of the United Nations (FAO) and the World Health Organization (WHO). Its remit has extended beyond additives into contaminants, naturally occurring toxicants and residues of veterinary drugs in food. EFSA comments on questions brought to it by the European Commission, with regular opinions on veterinary residue data and special topics, typical examples being the safety of ractopamine (a β-agonist) in feed, hormone residues in beef and chloramphenicol in food and feed[4].
In the EU, Regulation (EC) No 470/2009 sets out the procedures for the establishment of MRLs, which are in turn listed in Commission Regulation (EU) No 37/2010 of 22 December 2009 on veterinary medicinal products in foodstuffs of animal origin. Domestic UK effect is given by the Animals and Animal Products (Examination for Residues and Maximum Residue Limits) (England and Scotland) Regulations 2015 and, in Northern Ireland, by the Animal and Animal Products (Examination for Residues and Maximum Residue Limits) (Northern Ireland) 2016 (SR 54). Wales is still covered by 1997 Regulations and amendments. A useful summary of veterinary medicines and medicated animal feed legislation is available from Defra[5].
Image may be NSFW. Clik here to view.
Figure 1: Risk assessment process
Zero tolerance and analytical performance
For a combination of reasons – serious consequences and/or lack of data – EMA and the Commission took a ‘zero tolerance’ approach to residues in food of a group of pharmacologically active substances (Table 1). Originally, chloroform was included in the list but in 2014 it was permitted as a vaccine excipient for all mammalian food producing species (without an MRL but limited to a dose of 20 mg per animal). The prohibited list can found in Table 2 of the Annex to Regulation 37/2010. Directive 96/22/EC of 29 April 1996 also prohibits in food of animal origin thyrostatic substances, various stilbenes, oestradiol 17β and its ester-like derivatives and β-agonists (subject to limited derogations).
Prohibition posed some obvious questions on analytical limits of detection and how to deal with potential false positive and false negative findings – type 1 (alpha) and type 2 (beta) errors. ‘An attempt to answer these questions was made by Commission Decision 2002/657/EC implementing Council Directive 96/23/EC on the performance of analytical methods and the interpretation of results. Detailed criteria are given in 2002/657/EC – e.g. confirmatory methods must provide chemical structural information. The desired specificity must be achieved by suitable analytical combinations of clean-up, chromatographic separation(s) and spectrometric detection. Maximum permitted tolerances are set out for relative ion intensities using a range of mass spectrometric techniques and method validation is prescribed in great detail. In addition to identification criteria, 2002/657/EC defines the Decision Limit (CCα) as the measured concentration at which it can be said with an error probability of α (a 99% confidence interval) that a prohibited substance is truly present and the Detection Capability (CCβ) as the lowest concentration at which a method is able to detect truly contaminated samples with a statistical certainty of 1 – β (a 95% confidence interval). Both CCα and CCβ are derived from the method validation data as multiples of the standard deviation of measured responses of blank samples under within-laboratory reproducibility conditions. By definition, CCα is a lower concentration than CCβ. But this left open the possibility that different laboratories could certify against a food at different concentrations since CCα and CCβ can differ from one lab to another. Thus in 2003, the European Commission introduced a further criterion – the Minimum Required Performance Limit (MRPL)[6] and established MRPLs of 0.3 µg kg-1 for chloramphenicol and 1 µg kg-1 for nitrofurans. Eventually these lab criteria became de facto limits above which a food must not be allowed into the supply chain for human consumption. Suppliers who repeatedly offer consignments for import into the EU with concentrations of prohibited compounds above the CCα but below the MRPL must be investigated.
Pharmacologically active substances or groups
Background
Aristolochia spp. and preparations thereof
Historically used in herbal medicine; aristolochic acid, a major active constituent, is now known to be a carcinogen and a potent nephrotoxin and abortifacient.
Chloramphenicol
Can cause aplastic anaemia in a non-dose-dependent manner; genotoxic and a possible carcinogen.
Chlorpromazine
A phenothiazine antipsychotic in human medicine but with potential adverse effects such as hypotension, jaundice, leukocytosis, leukopenia and dermatological reactions. JECFA were unable to set an ADI owing to lack of available data.
Colchicine
An alkaloid of plant origin; has been used for treatment and prevention of gout but has a narrow therapeutic index, with no clear-cut distinction between nontoxic, toxic, and lethal doses. Although colchicine poisoning is sometimes intentional, unintentional toxicity is common and often associated with a poor outcome. It is genotoxic and teratogenic; no ADI could be established.
Dapsone
A sulphonamide antibiotic used for treatment of leprosy and malaria, originally included in the prohibited list owing to suspected genotoxic carcinogenicity; later studies suggested a non-genotoxic mechanism. Dapsone can occur as an impurity in other sulphonamides from their chemical synthesis with residue findings via this route. Has been suggested as a candidate for a reference point for action.
Dimetridazole
No ADI could be established as a NOAEL could not be identified.
Metronidazole
Antiprotozoal and antibiotic in human medicine but prohibited in food owing to genotoxic carcinogenicity.
Nitrofurans (including furazolidone)
Concerns about carcinogenicity and mutagenicity of nitrofurans and their metabolites.
Ronidazole
Insufficient data to establish a MRL
Table 1: Pharmacologically active compounds prohibited in food
Referee cases
Veterinary residue referee cases are a regular feature of the work of the Government Chemist with technical appeals on official findings of prohibited nitrofuran and chloramphenicol residues and issues with, for example, albendazole, which has a MRL.
Nitrofurans
Nitrofuran[7] antibiotics were first synthesised in the 1950’s for human use, with veterinary uses following soon afterwards, and found widespread use. For example, by the 1980’s furazolidone was an extremely common feed additive for pig husbandry in Europe. Nitrofurans were the treatment of choice for everything from fowl cholera to parasitic mites in honeybees and aquaculture. Five of the most common veterinary nitrofurans are furaltadone, furazolidone, nifursol, nitrofurantoin and nitrofurazone. Studies in the 1980’s began to raise concerns about carcinogenicity and mutagenicity of nitrofurans and their metabolites. Nitrofurans are now prohibited for use in food-producing animals in most jurisdictions in the world. However, they are still authorised and popular in human medicine and for the treatment of non-food animals and are widely manufactured and sold worldwide.
Nitrofurans are rapidly metabolised in the animal and residues of the parent molecule can no longer be detected within days, if not hours, of administration. But protein-bound metabolites of four of the five most common veterinary nitrofurans have been identified, which are stable for many weeks.
Analytical test methods, mainly developed by Bob McCracken, Glenn Kennedy and colleagues, are therefore based upon measuring these protein-bound metabolites. Unfortunately, semicarbazide (SEM), the marker metabolite for nitrofurazone, is not unique to the use of the parent drug. SEM can arise from the use of breadcrumbs and other bread products probably from azodicarbonamide, a flour treatment agent (now banned in the EU). But because azodicarbonamide gives food safety benefits, its use is retained in many other jurisdictions. SEM can also arise from the hypochlorite treatment of foods. But, most surprisingly, SEM occurs naturally in crustacean shells. A confirmatory analysis has been in place for many years based on washing out free SEM so that only SEM bound to tissue as a result of metabolism of the administered drug is determined.
The confirmatory analysis is laborious enough but to avoid SEM from the crustacean shell, excised core flesh must be analysed. This presents problems in sampling and homogenisation of shellfish consignments. Efforts continue to find a more suitable marker for nitrofurazone use in animals, especially shellfish[8].
Image may be NSFW. Clik here to view.
Giant freshwater prawns
Albendazole
The Government Chemist was called in when an importer disputed official results for the concentration of the veterinary medicine ‘albendazole’ in corned beef[9]. The Public Analyst had certified against the consignment for excess albendazole. A laboratory acting for the importer reported data below the MRL, including a finding of the parent drug, which is not included in the residue definition. Albendazole, one of the benzimidazole anthelmintics, is authorised for use in ruminants. However, owing to reported teratogenicity, there are MRLs for various tissues as the sum of albendazole sulfoxide, albendazole sulfone and albendazole 2-aminosulfone, expressed as albendazole. The case commenced with further sampling and we received a range of samples from the detained consignment, which consisted of product from two batch production dates: two portions of homogenised corned beef originally analysed by the laboratories previously involved, two unopened cans from the original sampling exercise and a further 40 unopened cans chosen at random from the consignment, of which four cans were separately analysed. The analysis consisted of acetonitrile extraction, liquid/liquid partitioning and solid phase extraction clean-up followed by liquid chromatography separation and tandem mass spectrometry detection and quantification (LC-MS/MS). Isotopically labelled albendazole-D3 and albendazole sulphoxide-D3 were used as internal standards.
Our findings did not confirm the presence of the parent drug, as only the sulfoxide metabolite was detected. We confirmed an exceedance of the MRL in both samples originally analysed and also in two previously unopened cans of the product. One of these contained over 15 times the maximum permitted amount of residue. The consignment was rejected and did not enter the UK.
Image may be NSFW. Clik here to view.
Liquid chromatography of veterinary residues
Conclusions
Veterinary medicines are essential for animal welfare but a stringent control system is necessary to prevent harm to consumers from residues in treated animals, allay consumer fears and prevent incidents that forfeit trust in food safety.
By and large such control systems are in place legislatively. Their practical implementation differs in the main trading blocs although some global harmonisation is evident [10]. It remains to be seen how such harmonisation will cope with global geopolitical changes especially with regard to trade. Sampling and analysis methods for veterinary medicines continue to evolve.
Michael Walker and Kirstin Gray,
Laboratory of the Government Chemist, LGC, Teddington, TW11 0LY, UK.
The Government Chemist continues to take a very active interest in veterinary residue analytical and sampling issues to safeguard consumers, industry, regulators and the courts from unwitting measurement errors. For more information see: https://www.gov.uk/government/organisations/government-chemist
References
1. Van Bunnik BAD, & Woolhouse MEJ, 2017, Modelling the impact of curtailing antibiotic usage in food animals on antibiotic resistance in humans, R Soc open sci. 4:161067. http://dx.doi.org/10.1098/rsos.161067
4. EFSA CONTAM Panel (EFSA Panel on Contaminants in the Food Chain), 2014. Scientific Opinion on Chloramphenicol in food and feed. EFSA Journal 2014; 12(11): 3907, 145 pp. doi:10.2903/j.efsa.2014.3907
6. Commission Decision 2003/181/EC of 13 March 2003 amending Decision 2002/657/EC as regards the setting of minimum required performance limits (MRPLs) for certain residues in food of animal origin.
7. EFSA CONTAM Panel (EFSA Panel on Contaminants in the Food Chain), 2015. Scientific Opinion on nitrofurans and their metabolites in food. EFSA Journal 2015; 13(6): 4140, 217 pp. doi:10.2903/j.efsa.2015.4140
8. John Points, D. Thorburn Burns, Michael J. Walker, 2014, Forensic issues in the analysis of trace nitrofuran veterinary residues in food of animal origin, Food Control, 50, 92-103.
9. Walker, M., Gray, K., Hopley, C., Mussell, C., Clifford, L., Meinerikandathevan, J., Firpo, L., Topping, J., and Santacruz, D., 2016, Resolution of a disputed albendazole result in the UK Official Control System – time for more guidance?, Food Additives & Contaminants: Part A, DOI: 10.1080/19440049.2016.1243807 http://dx.doi.org/10.1080/19440049.2016.1243807
10. Baynes, R.E., Dedonder, K., Kissell, L., Mzyk, D., Marmulak, T., Smith, G., Tell, L., Gehring, R., Davis, J. and Riviere, J.E., 2016. Health concerns and management of select veterinary drug residues. Food and Chemical Toxicology, 88, pp.112-122.
