Survey
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
Human microbiota wikipedia , lookup
Sociality and disease transmission wikipedia , lookup
Hospital-acquired infection wikipedia , lookup
Traveler's diarrhea wikipedia , lookup
Infection control wikipedia , lookup
Carbapenem-resistant enterobacteriaceae wikipedia , lookup
Bacterial morphological plasticity wikipedia , lookup
Antimicrobial copper-alloy touch surfaces wikipedia , lookup
Disinfectant wikipedia , lookup
OIE International Standards on Antimicrobial Resistance, 2003 OIE Headquarters, Paris, 2 to 4 October 2001 OIE publication with the participation of the OIE Collaborative Centre on Veterinary Medicinal Products, Fougères All OIE (World organisation for animal health) publications are protected by international copyright law. Extracts may be copied, reproduced, translated, adapted or published in journals, documents, books, electronic media and any other medium destined for the public, for information, educational or commercial purposes, provided prior written permission has been granted by the OIE. The designations and denominations employed and the presentation of the material in this publication do not imply the expression of any opinion whatsoever on the part of the OIE concerning the legal status of any country, territory, city or area or of its authorities, or concerning the delimitation of its frontiers and boundaries. The views expressed in signed articles are solely the responsibility of the authors. The mention of specific companies or products of manufacturers, whether or not these have been patented, does not imply that these have been endorsed or recommended by the OIE in preference to others of a similar nature that are not mentioned. Toutes les publications de l’OIE (Organisation mondiale de la santé animale) sont protégées par un copyright international. La copie, la reproduction, la traduction, l’adaptation ou la publication d’extraits, dans des journaux, des documents, des ouvrages ou des supports électroniques et tous autres supports destinés au public, à des fins d‘information, didactiques ou commerciales, requièrent l’obtention préalable d’une autorisation écrite de l’OIE. Les désignations et dénominations utilisées et la présentation des données figurant dans cette publication ne reflètent aucune prise de position de l’OIE quant au statut légal de quelque pays, territoire, ville ou zone que ce soit, à leurs autorités, aux délimitations de leur territoire ou au tracé de leurs frontières. Les auteurs sont seuls responsables des opinions exprimées dans les articles signés. La mention de sociétés spécifiques ou de produits enregistrés par un fabriquant, qu'ils soient ou non protégés par une marque, ne signifie pas que ceux-ci sont recommandés ou soutenus par l’OIE par rapport à d’autres similaires qui ne seraient pas mentionnés. Todas las publicaciones de la OIE (Organización mundial de sanidad animal) están protegidas por un Copyright internacional. Extractos pueden copiarse, reproducirse, adaptarse o publicarse en publicaciones periódicas, documentos, libros o medios electrónicos, y en cualquier otro medio destinado al público, con intención informativa, didáctica o comercial, siempre y cuando se obtenga previamente una autorización escrita por parte de la OIE. Las designaciones y nombres utilizados y la presentación de los datos que figuran en esta publicación no constituyen de ningún modo el reflejo de cualquier opinión por parte de la OIE sobre el estatuto legal de los países, territorios, ciudades o zonas ni de sus autoridades, fronteras o limitaciones territoriales. La responsabilidad de las opiniones profesadas en los artículos firmados incumbe exclusivamente a sus autores. La mención de empresas particulares o de productos manufacturados, sean o no patentados, no implica de ningún modo que éstos se beneficien del apoyo o de la recomendación de la OIE, en comparación con otros similares que no hayan sido mencionados. © Copyright OIE (World organisation for animal health), 2003 (reprinted in April and November 2004) 12, rue de Prony, 75017 Paris, France Tel.: 33-(0)1 44 15 18 88 Fax: 33-(0)1 42 67 09 87 http://www.oie.int ISBN 92-9044-601-3 Cover photograph: © Claire Gaudot, AFSSA, Fougères Contents Contents Introduction .............................................................................................................................. 1 OIE International Standards on Antimicrobial Resistance Terrestrial Animal Health Code Guidelines for the harmonisation of antimicrobial resistance surveillance and monitoring programmes .................................................................................................. 5 Guidelines for the monitoring of the quantities of antimicrobials used in animal husbandry..................................................................................................................... 13 Guidelines for the responsible and prudent use of antimicrobial agents in veterinary medicine.................................................................................................................. 17 Manual of Diagnostic Tests and Vaccines for Terrestrial Animals Laboratory methodologies for bacterial antimicrobial susceptibility testing ................ 29 Proceedings of the OIE International Conference 1. General aspects J. Acar & B. Röstel Antimicrobial resistance: an overview ......................................................................... 45 J. Threlfall Antimicrobial drug resistance from salmonellas in humans and food animals: the current situation in relation to foodborne zoonoses in the United Kingdom ................................................................................ 69 S. Benredjeb & A. Hammami Resistance in salmonellae: the situation in developing countries................................. 71 J.-L. Martel Resistant bacteria and their impact on therapy in veterinary medicine ........................................................................................................................... 72 P. Nordmann New resistance mechanisms – review of the diversity.............................................. 73 OIE International Standards on Antimicrobial Resistance, 2003 III Contents M.H. Nicolas-Chanoine & S. Granier Possibilities of characterising resistance genes for use as an epidemiological tool ....................................................................................................... 74 O. Fortineau The perception of veterinary practitioner with regard to the contribution of the use of antimicrobials in animal husbandry to the problems of human health associated with resistant bacteria........................... 79 O.G. Pedersen The pig producer’s position as herd manager following the cessation of the use of antibiotic growth promoters in Denmark.......................... 80 B. Andrews Antimicrobial use in animal husbandry and its relationship to resistant bacteria in human health................................................................................ 85 S. Sirinavin Perception of society with regard to the contribution of the use of antimicrobials in animal husbandry to the problems of human health associated with resistant bacteria: the situation in developing countries........................................................................................................ 89 L.Y Lefferts A consumer perspective: to what extent does antimicrobial use in animal husbandry contribute to resistance associated human health problems? ............................................................................................................. 90 Y. Cheneau Activities of the Food and Agriculture Organization in relation to antimicrobial resistance in humans and animals ................................................... 95 R. Williams The activities of the World Health Organization in antimicrobial resistance .......................................................................................................................... 99 A. Bruno Codex activities in relation to antimicrobial resistance...........................................100 2. Surveillance of antimicrobial consumption †T. Nicholls, J. Acar, F. Anthony, A. Franklin, R. Gupta, Y. Tamura, S. Thompson, E.J. Threlfall, D. Vose, M. van Vuuren, D.G. White, H.C. Wegener & M.L. Costarrica Antimicrobial resistance: monitoring the quantities of antimicrobials used in animal husbandry ............................................................................................109 G. Moulin Surveillance of antimicrobial consumption activities in France ............................118 IV OIE International Standards on Antimicrobial Resistance, 2003 Contents D.L. Monnet, F. Bager & L. Larsen Surveillance of antimicrobial consumption in Denmark........................................122 J.J. Webber Antimicrobial resistance: monitoring the quantity of antimicrobials used in animal husbandry ............................................................................................123 T. Mudd The global usage of antimicrobials for animals health............................................128 3. Risk analysis D. Vose, J. Acar, F. Anthony, A. Franklin, R. Gupta, †T. Nicholls, Y. Tamura, S. Thompson, E.J. Threlfall, M. van Vuuren, D.G. White, H.C. Wegener & M.L. Costarrica Antimicrobial resistance: risk analysis methodology for the potential impact on public health of antimicrobial resistant bacteria of animal origin ............................................................................................................................... 133 M. Wooldridge Risk assessment techniques – and antibiotic resistance.......................................... 162 L. Tollefson Impact of resistant campylobacteriosis in humans due to fluoroquinolone use in chickens................................................................................. 170 M. van Vuuren Antimicrobial resistance and risk analysis: the view of a developing country............................................................................................................................ 174 L.A. Cox Campylobacter risk analysis: a cause-and-effect view ................................................. 177 4. Surveillance of resistance programme A. Franklin, J. Acar, F. Anthony, R. Gupta, †T. Nicholls, Y. Tamura, S. Thompson, E.J. Threlfall, D. Vose, M. van Vuuren, D.G. White, H.C. Wegener & M.L. Costarrica Antimicrobial resistance: harmonisation of national antimicrobial resistance monitoring and surveillance programmes in animals and in animal-derived food......................................................................................................181 P.J. Fedorka-Cray, M.L. Headrick & L. Tollefson The National Antimicrobial Resistance Monitoring System (NARMS) ..............200 V. Jarlier Surveillance of resistance – human/animal coordinated approaches in France ............................................................................................................................205 OIE International Standards on Antimicrobial Resistance, 2003 V Contents Y. Tamura The Japanese Veterinary Antimicrobial Resistance Monitoring System (JVARM) ........................................................................................................................206 T.T.T. Phuong Cases of antimicrobial resistance to some pathogens in Vietnam ........................211 S. Kariuki, G. Revathi & C.A. Hart Antimicrobial resistance surveillance in Kenya: achievements and challenges .......................................................................................................................216 5. Laboratory methods D.G. White, J. Acar, F. Anthony, A. Franklin, R. Gupta, †T. Nicholls, Y. Tamura, S. Thompson, E.J. Threlfall, D. Vose, M. van Vuuren, H.C. Wegener & M.L. Costarrica Antimicrobial resistance: standardisation and harmonisation of laboratory methodologies for the detection and quantification of antimicrobial resistance................................................................................................223 I. Phillips Standardisation of antimicrobial susceptibility testing in Europe: the work of the European Committee for Antimicrobial Susceptibility Testing (EUCAST) .......................................................................................................239 T.R. Shryock National Committee for Clinical Laboratory Standards: a perspective on antimicrobial susceptibility testing methods.......................................................243 I.S.T. Fisher, O.N. Gill, W.J. Reilly, H.R. Smith & E.J. Threlfall Harmonisation of antimicrobial resistance testing results – the outcome of the international Enter-net study ..........................................................................245 6. Prudent use and containment of resistance F. Anthony, J. Acar, A. Franklin, R. Gupta, †T. Nicholls, Y. Tamura, S. Thompson, E.J. Threlfall, D. Vose, M. van Vuuren & D.G. White Antimicrobial resistance: responsible and prudent use of antimicrobial agents in veterinary medicine......................................................................................249 D.L. Smith & J.A. Johnson Antibiotic use in animals and the emergence of antibiotic resistance in human commensal microbes and zoonotic pathogens...........................................267 I.M. Gould Prudent use of antibiotics and containment of antimicrobial resistance: the role of medical associations, guidelines and interventions ..............................273 VI OIE International Standards on Antimicrobial Resistance, 2003 Contents J. Edwards Prudent use of antibiotics and containment of antimicrobial resistance .............276 B. Jennings Prudent use and containment of antimicrobial resistance – the work of the responsible use of medicines in agriculture alliance .........................................279 D.K. Byarugaba Prudent use and containment of antimicrobial resistance in developing countries .........................................................................................................................280 Web links ...............................................................................................................................285 OIE International Standards on Antimicrobial Resistance, 2003 VII Introduction Introduction The increasing antimicrobial resistance of important human pathogenic bacteria, and the spread of such bacteria from the closed environment of hospitals into surrounding communities, are increasingly perceived as threats to public health. Any use of antimicrobials, whether in humans, animals, plants or food-processing, may lead to bacterial resistance. The use of antimicrobials in livestock production is thought to significantly contribute to the phenomenon, but little is known about the true causes of antimicrobial resistance. The lack of relevant scientific data means that risk managers must take precautionary measures, even though the underlying causes of public health risks associated with resistant bacteria may not have been adequately identified. Increasing international travel and international trade in animals and animal products may spread resistance world-wide. The OIE (World organisation for animal health) is developing international standards on antimicrobial resistance to enable Member Countries to protect themselves, without setting up unjustified sanitary barriers. These standards are consistent with the OIE’s missions to: • guarantee the safety of world trade by developing sanitary rules for international trade in animals and animal products, • inform governments on the existence and evolution of animal diseases and zoonoses including their control measures, • coordinate, at the international level, studies and investigations on surveillance and control of animal diseases and zoonoses. The main normative works produced by the OIE are: the Terrestrial Animal Health Code, the Manual of Diagnostic Tests and Vaccines for Terrestrial Animals, the Aquatic Animal Health Code and the Manual of Diagnostic Tests and Vaccines for Aquatic Animals. The OIE normative works are recognised by the World Trade Organization (WTO) as international sanitary standards. They are developed by elected Specialist Commissions and Working Groups which bring together internationally renowned scientists, most of whom are experts within the network of 156 Collaborating Centres and Reference Laboratories which also contribute to the scientific objectives of the OIE. Specific recommendations were elaborated in five subject areas1 by the ‘OIE ad 1 1. Risk analysis and antimicrobial resistance, 2. Prudent use and containment of antimicrobial resistance, 3. Surveillance of antimicrobial consumption in animal husbandry, 4. Resistance surveillance programmes, 5. Laboratory Methods – Standardisation and Harmonisation OIE International Standards on Antimicrobial Resistance, 2003 1 Introduction hoc on Antimicrobial Resistance’ created in 1999. The standards proposed to the OIE International Committee were adopted in May 2003. Countries which import animals and animal products can now legaly use these standards to verify whether or not exporting countries are complying with these new requirements. The second OIE International Conference on Antimicrobial Resistance, held from 2 to 4 October 2001 at the OIE Headquarters in Paris, was a milestone in promoting communication between stakeholders from both human and animal medical fields. This conference reviewed progress achieved since the first OIE International Conference of March 1999, and in particular, focussed on understanding the development of antimicrobial resistance in humans and animals, the problems in human and veterinary medicine and the actions taken for the containment of resistance. It also provided a public forum for the presentation of the results of two years’ work by the OIE and the identification of potential future actions by the OIE. The present publication offers, in the first section, the OIE International Standards on Antimicrobial Resistance as adopted by the OIE International Committee in May 2003, and published in the OIE Terrestrial Animal Health Code, 2003 and in the Manual of Standards for Diagnostic Tests and Vaccines for Terrestrial Animals. The second section contains the Proceedings of the 2nd OIE International Conference where the reports of the OIE ad hoc group on Antimicrobial Resistance were discussed publicly. I would like to express my gratitude to Dr Jean Blancou, Director General of the OIE (1990-2000) and Dr Jacques Boisseau, Director of the OIE Collaborating Centre for Veterinary Medicinal Products2, for initiating this important scientific activity of the OIE. Furthermore, my particular appreciation goes to the OIE ad hoc group on Antimicrobial Resistance, to its Chairman Prof. Jacques Acar and its secretary Dr Barbara Röstel from the OIE Collaborating Centre for the very efficient organisation of this conference. I also would like to thank all national and international experts, who have contributed so generously and eloquently to the success of this second OIE International Conference on Antimicrobial Resistance. Bernard Vallat Director General Dr Jacques Boisseau retired in March 2002. Dr Patrick Dehaumont has been designated as new Director of the OIE Collaborating Centre 2 2 OIE International Standards on Antimicrobial Resistance, 2003 OIE International Standards on Antimicrobial Resistance OIE International Standards on Antimicrobial Resistance, 2003 3 Terrestrial Animal Health Code Terrestrial Animal Health Code Section 3.9. Antimicrobial resistance Appendix 3.9.1. Guidelines for the harmonisation of antimicrobial resistance surveillance and monitoring programmes Article 3.9.1.1. Objective This Appendix provides criteria for the: 1. development of national antimicrobial resistance surveillance and monitoring programmes 2. harmonisation of existing national surveillance and monitoring programmes, in animals and in products of animal origin intended for human consumption. Article 3.9.1.2. Purpose of surveillance and monitoring 1. 2. Surveillance and monitoring of antimicrobial resistance is necessary to: a) follow trends in antimicrobial resistance in bacteria; b) detect the emergence of new antimicrobial resistance mechanisms; c) provide the data necessary for conducting risk analyses with relevance for human and animal health; d) provide a basis for policy recommendations for animal and public health; e) provide information recommendations. for prescribing practices and prudent use Antimicrobial resistance monitoring and surveillance programmes may include the following components: a) scientifically based surveys (including statistically based programmes); OIE International Standards on Antimicrobial Resistance, 2003 5 Terrestrial Animal Health Code b) routine sampling and testing of animals on the farm, at market or at slaughter; c) an organised sentinel programme, sampling animals, herds, flocks, and vectors; d) analysis of veterinary practice and diagnostic laboratory records. 3. Countries should conduct active surveillance and monitoring. Passive surveillance and monitoring may offer additional information. 4. Targeted surveillance is conducted through an active sampling scheme designed to meet programme objectives. Passive surveillance is conducted when samples are submitted to a laboratory for testing from sources outside the programme. Article 3.9.1.3. The development of antimicrobial resistance surveillance and monitoring programmes 1. General aspects Surveillance of antimicrobial resistance at regular intervals or ongoing monitoring of prevalence changes of resistant bacteria of animal, food, environmental and human origin, constitutes a critical part of a strategy aimed at limiting the spread of antimicrobial resistance and optimising the choice of antimicrobials used in therapy. Monitoring of bacteria from products of animal origin intended for human consumption collected at different steps of the food chain, including processing, packing and retailing, should also be considered. 2. Sampling strategies a) General i) Sampling should be conducted on a statistical basis. The sampling strategy should assure: – – ii) The following criteria are to be considered: – – – – – – – 6 the sample representativeness of the population of interest; the robustness of the sampling method. sample size; sample source (animal, food, animal feed); animal species; category of animal within species (age group, production type); stratification within category; health status of the animals (healthy, diseased); random sample (targeted, systematic); OIE International Standards on Antimicrobial Resistance, 2003 Terrestrial Animal Health Code – b) sample specimens (faecal, carcass, processed food). Sample size The sample size should be: i) ii) large enough to allow detection of existing resistance, not excessively large to avoid waste of resources. Details are provided in Table I. Sampling shall follow standard operating procedures. Table I Sample size estimates for prevalence of antimicrobial resistance in a large population Expected prevalence 10% 20% 30% 40% 50% 60% 70% 80% 90% Level of confidence 90% desired precision 95% desired precision 10% 5% 1% 10% 5% 1% 24 43 57 65 68 65 57 43 24 97 173 227 260 270 260 227 173 97 2,429 4,310 5,650 6,451 6,718 6,451 5,650 4,310 2,429 35 61 81 92 96 92 81 61 35 138 246 323 369 384 369 323 246 138 3,445 6,109 8,003 9,135 9,512 9,135 8,003 6,109 3,445 Calculations based on Epi Info v6.04b to c Upgrade, October 1997, Centers for Disease Control (public domain software available at http://www.cdc.gov/epo/epi/epiinfo.htm). 3. Sample sources a) Animals Each Member Country should examine its livestock production systems and decide, after risk analysis, the relative importance of antimicrobial resistance and its impact on animal and human health. Categories of livestock that should be considered for sampling include cattle and calves, slaughter pigs, broiler chickens, layer hens and/or other poultry and farmed fish. b) Food and animal feed Contaminated food is commonly considered to be the principal route for the transfer of antimicrobial resistance from animals to humans. Plants and vegetables of different types may be exposed to manure or sewage from livestock OIE International Standards on Antimicrobial Resistance, 2003 7 Terrestrial Animal Health Code and may thereby become contaminated with resistant bacteria of animal origin. Animal feed, including imported feed, may also be considered in surveillance and monitoring programmes. Table II Examples of sampling sources, sample types and outcome of monitoring Source Sample type Herd of origin Abattoir Prevalence of resistance in bacteria originating from animal populations (of different production types) Relationship resistance – antibiotic use Faecal Prevalence of resistance in bacterial populations originating from animals at slaughter age Intestine As above Carcass Hygiene, contamination during slaughter Processing, packing Meat products Hygiene, contamination during processing and handling Retail Meat products Prevalence of resistance in bacteria originating from food, exposure data for consumers Vegetables Prevalence of resistance in bacteria originating from vegetables, exposure data for consumers Animal feed Prevalence of resistance in bacteria originating from animal feed, exposure data for animals Various origin 4. Outcome Additional information required/additional stratification Per age categories, production types, etc. Antibiotic use over time Sample specimens to be collected Faecal samples should be collected from livestock, and whole caeca should be collected from poultry. In cattle and pigs, a faecal sample size at least of 5 g provides a sufficient sample for isolation of the bacteria of concern. Sampling of the carcasses at the abattoir provides information on slaughter practices, slaughter hygiene and the level of faecal contamination of meat during the slaughter process. Further sampling from the retail chain provides information on prevalence changes before the food reaches the consumer. Existing food-processing microbiological monitoring and ‘hazard analysis and critical control points’ (HACCP) programmes may provide useful samples for surveillance and monitoring of resistance in the food chain after slaughter. 5. Bacterial isolates The following categories of bacteria could be monitored: a) 8 Animal bacterial pathogens OIE International Standards on Antimicrobial Resistance, 2003 Terrestrial Animal Health Code Monitoring of antimicrobial resistance in animal pathogens is important, both to: i) detect emerging resistance that may pose a concern for human and animal health; ii) guide veterinarians in their prescribing decisions. Information on the occurrence of antimicrobial resistance in animal pathogens is in general derived from routine clinical material sent to veterinary diagnostic laboratories. These samples, often derived from severe or recurrent clinical cases including therapy failures, may provide biased information. Table III Examples of animal bacterial pathogens that may be included in resistance surveillance and monitoring Target animals Respiratory pathogens Enteric pathogens Udder pathogens Cattle Pasteurella spp. Escherichia coli Staphylococcus aureus Streptococcus spp. Haemophilus somnus Salmonella spp. Actinobacillus pleuropneumoniae Escherichia coli Brachyspira spp. Salmonella spp. Pigs Poultry Fish b) Other pathogens Streptococcus suis Escherichia coli Vibrio spp. Aeromonas spp. Zoonotic bacteria i) Salmonella Salmonella should be sampled from cattle, pigs, broilers and other poultry. For the purpose of facilitating sampling and reducing the concurrent costs, samples should preferably be taken at the abattoir. Surveillance and monitoring programmes may also use bacterial isolates from designated national laboratories originating from other sources. Isolation and identification of bacteria and bacterial strains should follow internationally accepted procedures. Serovars of epidemiological importance such as S. typhimurium and S. enteritidis should be included. The selection of other relevant serovars will depend on the epidemiological situation in each country. All Salmonella isolates should be serotyped and, when appropriate, phage-typed according to standard methods used at the nationally designated laboratories. Validated methods should be used. OIE International Standards on Antimicrobial Resistance, 2003 9 Terrestrial Animal Health Code ii) Campylobacter Campylobacter jejuni and C. coli can be isolated from the same samples as commensal bacteria. Isolation and identification of these bacteria should follow internationally accepted procedures. Campylobacter isolates should be identified to the species level. Agar or broth micro-dilution methods are recommended for Campylobacter susceptibility testing. Internal and external quality control programmes should be strictly adhered to. Validated methods with appropriate reference strains are expected to become available in the near future. iii) Enterohaemorrhagic Escherichia coli Enterohaemorrhagic Escherichia coli (EHEC), such as the serotype O157, which is pathogenic to humans but not to animals, may be included in resistance surveillance and monitoring programmes. c) Commensal bacteria Escherichia coli and enterococci are common commensal bacteria. These bacteria are considered to constitute a reservoir of antimicrobial resistance genes, which may be transferred to pathogenic bacteria causing disease in animals or humans. It is considered that these bacteria should be isolated from healthy animals, preferably at the abattoir, and be monitored for antimicrobial resistance. Validated methods should be used. 6. Storage of bacterial strains If possible, isolates should be preserved at least until reporting is completed. Preferably, isolates should be permanently stored. Bacterial strain collections, established by storage of all isolates from certain years, will provide the possibility of conducting retrospective studies. 7. Antimicrobials to be used in susceptibility testing Clinically important antimicrobial classes used in human and veterinary medicine should be monitored. However, the number of tested antimicrobials may have to be limited according to the financial resources of the country. 8. Type of data to be recorded and stored Data on antimicrobial susceptibility should be reported quantitatively. Appropriate validated methods should be used in accordance with Chapter I.1.10. of the Terrestrial Manual concerning laboratory methodologies for bacterial antimicrobial susceptibility testing. 10 OIE International Standards on Antimicrobial Resistance, 2003 Terrestrial Animal Health Code 9. Recording, storage and interpretation of results a) b) c) d) Because of the volume and complexity of the information to be stored and the need to keep these data available for an undetermined period of time, careful consideration should be given to database design. The storage of raw (primary, non-interpreted) data is essential to allow the evaluation of the data in response to various kinds of questions, including those arising in the future. Consideration should be given to the technical requirements of computer systems when an exchange of data between different systems (comparability of automatic recording of laboratory data and transfer of these data to resistance monitoring programmes) is envisaged. Results should be collected in a suitable national database. They shall be recorded quantitatively: i) as distribution of minimum inhibitory concentrations (MICs) in milligrams per litre; ii) or inhibition zone diameters in millimetres. The information to be recorded should include at least the following aspects: i) ii) iii) iv) v) vi) vii) e) sampling programme; sampling date; animal species/livestock category; type of sample; purpose of sampling; geographical origin of herd, flock or animal; age of animal. The reporting of laboratory data should include the following information: i) ii) iii) iv) f) g) h) identity of laboratory, isolation date, reporting date, bacterial species, and, where relevant, other typing characteristics, such as: v) serovar, vi) phage-type, vii) antimicrobial susceptibility result/resistance phenotype. The proportion of isolates regarded as resistant should be reported, including the defined breakpoints. In the clinical setting, breakpoints are used to categorise bacterial strains as susceptible, intermediate susceptible or resistant. These breakpoints, often referred to as clinical or pharmacological breakpoints, are elaborated on a national basis and vary between countries. The system of reference used should be recorded. OIE International Standards on Antimicrobial Resistance, 2003 11 Terrestrial Animal Health Code i) j) For surveillance purposes, the microbiological breakpoint, which is based on the distribution of MICs or inhibition zone diameters of the specific bacterial species tested, is preferred. When using microbiological breakpoints, only the bacterial population with acquired resistance that clearly deviates from the distribution of the normal susceptible population will be designated as resistant. If available, the phenotype of the isolates (resistance pattern) should be recorded. 10. Reference laboratory and annual reports a) Countries should designate a national reference centre that assumes the responsibility to: i) coordinate the activities related to the resistance surveillance and monitoring programmes; ii) collect information at a central location within the country; iii) produce an annual report on the resistance situation of the country. b) The national reference centre should have access to the: i) raw data; ii) complete results of quality assurance and inter-laboratory calibration activities; iii) proficiency testing results; iv) information on the structure of the monitoring system; v) 12 information on the chosen laboratory methods. OIE International Standards on Antimicrobial Resistance, 2003 Terrestrial Animal Health Code Appendix 3.9.2. Guidelines for the monitoring of the quantities of antimicrobials used in animal husbandry Article 3.9.2.1. Purpose The purpose of these guidelines is to describe an approach to the monitoring of quantities of antimicrobials used in animal husbandry. These guidelines are intended for use by OIE Member Countries to collect objective and quantitative information to evaluate usage patterns by animal species, antimicrobial class, potency and type of use in order to evaluate antimicrobial exposure. Article 3.9.2.2. Objectives The information provided in these guidelines is essential for risk analyses and planning, can be helpful in interpreting resistance surveillance data and can assist in the ability to respond to problems of antimicrobial resistance in a precise and targeted way. This information may also assist in evaluating the effectiveness of efforts to ensure prudent use and mitigation strategies (for example, by identifying changes in prescribing practices for veterinarians) and to indicate where alteration of antimicrobial prescribing practices might be appropriate, or if changes in prescription practice have altered the pattern of antimicrobial use. The continued collection of this basic information will also help give an indication of trends in the use of animal antimicrobials over time and the role of these trends in the development of antimicrobial resistance in animals. For all OIE Member Countries, the minimum basic information collected should be the annual weight in kilograms of the active ingredient of the antimicrobial(s) used in food animal production. In addition, the type of use (therapeutic or growth promotion) and route of administration (parenteral or oral administration) should be recorded. Member Countries may wish to consider, for reasons of cost and administrative efficiency, collecting medical, food animal, agricultural and other antimicrobial use data in a single programme. A consolidated programme would also facilitate comparisons of animal use with human use data for relative risk analysis and help to promote optimal usage of antimicrobials. Article 3.9.2.3. OIE International Standards on Antimicrobial Resistance, 2003 13 Terrestrial Animal Health Code Development and standardisation of monitoring systems Systems to monitor antimicrobial usage consist of the following elements: 1. Sources of antimicrobial data a) Basic sources Sources of data will vary from country to country. Such sources may include customs, import and export data, manufacturing and manufacturing sales data. b) Direct sources Data from animal drug registration, wholesalers, retailers, pharmacists, veterinarians, feed stores, feed mills and organised industry associations in these countries might be efficient and practical sources. A possible mechanism for the collection of this information is to make the provision of appropriate information by manufacturers to the regulatory authority one of the requirements of antimicrobial registration. c) End-use sources (veterinarians and food animal producers) This may be appropriate when basic or direct sources cannot be used for the routine collection of this information and when more accurate and locally specific information is required. Periodic collection of this type of information may be sufficient. It may be important when writing recommendations on antimicrobial resistance to take into account factors such as seasonality and disease conditions, species affected, agricultural systems (e.g. extensive range conditions and feedlots), dose rate, duration and length of treatment with antimicrobials. Collection, storage and processing of data from end-use sources are likely to be inefficient and expensive processes unless carefully designed and well managed, but should have the advantage of producing accurate and targeted information. 2. Categories of data a) Requirements for data on antimicrobial use The minimal data collected should be the annual weight in kilograms of the active ingredient of the antimicrobial(s) used in food animal production. This should be related to the scale of production (see point 3 below). For active ingredients present in the form of compounds or derivatives, the mass of active entity of the molecule should be recorded. For antibiotics expressed in International Units, the calculation required to convert these units to mass of active entity should be stated. If a Member Country has the infrastructure for capturing basic animal antimicrobial use data for a specific antimicrobial, then additional information 14 OIE International Standards on Antimicrobial Resistance, 2003 Terrestrial Animal Health Code can be considered to cascade from this in a series of subdivisions or levels of detail. Such a cascade of levels should include the following: i) The absolute amount in kilograms of active antimicrobial used per antimicrobial family per year, or for a specific antimicrobial chemical entity when this information is required. ii) Therapeutic and growth promotion use in kilograms of the specific active antimicrobial. iii) Subdivision of antimicrobial use into therapeutic and growth promotion use by animal species. iv) Subdivision of the data into the route of administration, specifically infeed, in-water, injectable, oral, intramammary, intra-uterine and topical. v) Further subdivision of these figures by season and region by a Member Country may be useful (Note: This may be especially management conditions, or where animals are moved from one locality to another during production). vi) Further breakdown of data for analysis of antimicrobial use at the regional, local, herd and individual veterinarian level may be possible using veterinary practice computer management software as part of specific targeted surveys or audits. Analysis of this information within the local or regional context could be useful for individual practitioners and practices where specific antimicrobial resistance has been identified and feedback is required. b) Classes of antimicrobials Nomenclature of antimicrobials should comply with international standards where available. Decisions need to be made on what classes of antimicrobials should be considered and what members of various antimicrobial classes should be included in the data collection programme. These decisions should be based on currently known mechanisms of antimicrobial activity and resistance of the particular antimicrobial and its relative potency. c) Species and production systems Countries should keep a register of all animal use of antimicrobials for individual food animal species (cattle, sheep, goats, pigs, poultry, horses and fish) and for specific diseases. This will help to identify possible nonauthorised usage. 3. Other important information Breakdown of farm livestock into species and production categories, including total live weights, would be most useful in any risk analysis or for comparison of animal antimicrobial use with human medical use within and between countries. OIE International Standards on Antimicrobial Resistance, 2003 15 Terrestrial Animal Health Code For example, the total number of food animals by category and their weight in kilograms for food production per year (meat, dairy and draught cattle, and meat, fibre, poultry and dairy sheep) in the country would be essential basic information. 16 OIE International Standards on Antimicrobial Resistance, 2003 Terrestrial Animal Health Code Appendix 3.9.3. Guidelines for the responsible and prudent use of antimicrobial agents in veterinary medicine Article 3.9.3.1. Purpose These guidelines provide guidance for the responsible and prudent use of antimicrobials in veterinary medicine, with the aim of protecting both animal and human health. The competent authorities responsible for the registration and control of all groups involved in the production, distribution and use of veterinary antimicrobials have specific obligations. Prudent use is principally determined by the outcome of the marketing authorisation procedure and by the implementation of specifications when antimicrobials are administered to animals. Article 3.9.3.2. Objectives of prudent use Prudent use includes a set of practical measures and recommendations intended to prevent and/or reduce the selection of antimicrobial-resistant bacteria in animals to: 1. maintain the efficacy of antimicrobial agents and to ensure the rational use of antimicrobials in animals with the purpose of optimising both their efficacy and safety in animals; 2. comply with the ethical obligation and economic need to keep animals in good health; 3. prevent, or reduce, as far as possible, the transfer of bacteria (with their resistance determinants) within animal populations; 4. maintain the efficacy of antimicrobial agents used in livestock; 5. prevent or reduce the transfer of resistant bacteria or resistance determinants from animals to humans; 6. maintain the efficacy of antimicrobial agents used in human medicine and prolong the usefulness of the antimicrobials; 7. prevent the contamination of animal-derived food with antimicrobial residues that exceed the established maximum residue limit (MRL); 8. protect consumer health by ensuring the safety of food of animal origin. OIE International Standards on Antimicrobial Resistance, 2003 17 Terrestrial Animal Health Code Article 3.9.3.3. Responsibilities of the regulatory authorities 1. Marketing authorisation The national regulatory authorities are responsible for granting marketing authorisation. They have a significant role in specifying the terms of this authorisation and in providing the appropriate information to the veterinarian. 2. Submission of data for the granting of the marketing authorisation The pharmaceutical industry has to submit the data requested for the granting of the marketing authorisation. The marketing authorisation is granted only if the criteria of safety, quality and efficacy are met. An assessment of the potential risk to both the animal and the consumer resulting from the use of antimicrobial agents in food-producing animals must be carried out. The evaluation should focus on each individual antimicrobial product and not be generalised to the class of antimicrobials to which the particular active principle belongs. If dose ranges or different durations of treatment are suggested, guidance on the usage should be provided. 3. Market approval Regulatory authorities should attempt to expedite the market approval process of a new antimicrobial in order to address a specific need for the treatment of disease. 4. Registration procedures Countries lacking the necessary resources to implement an efficient registration procedure for veterinary medicinal products (VMPs), and whose supply principally depends on imports from foreign countries, must undertake the following measures: a) check the efficacy of administrative controls on the import of these VMPs; b) check the validity of the registration procedures of the exporting country; c) develop the necessary technical co-operation with experienced authorities to check the quality of imported VMPs as well as the validity of the recommended conditions of use. Regulatory authorities of importing countries should request the pharmaceutical industry to provide quality certificates prepared by the competent authority of the exporting country. All countries should make every effort to actively combat the trade, distribution and use of unlicensed and counterfeit products. 5. Quality control of antimicrobial agents Quality controls should be performed: 18 OIE International Standards on Antimicrobial Resistance, 2003 Terrestrial Animal Health Code 6. a) in compliance with the provisions of good manufacturing practices; b) to ensure that analysis specifications of antimicrobial agents used as active ingredients comply with the provisions of approved monographs; c) to ensure that the quality and concentration (stability) of antimicrobial agents in the marketed dosage form(s) are maintained until the expiry date, established under the recommended storage conditions; d) to ensure the stability of antimicrobials when mixed with feed or drinking water; e) to ensure that all antimicrobials are manufactured to the appropriate quality and purity in order to guarantee their safety and efficacy. Control of therapeutic efficacy a) Preclinical trials i) Preclinical trials should: – – – ii) establish the range of activity of antimicrobial agents on both pathogens and non-pathogens (commensals); assess the ability of the antimicrobial agent to select for resistant bacteria in vitro and in vivo, taking into consideration pre-existing resistant strains; establish an appropriate dosage regimen necessary to ensure the therapeutic efficacy of the antimicrobial agent and limit the selection of antimicrobial-resistant bacteria. The activity of antimicrobial agents towards the targeted bacteria should be established by pharmacodynamics. The following criteria should be taken into account: – – – – mode of action; minimum inhibitory and bactericidal concentrations; time- or concentration-dependent activity; activity at the site of infection. iii) The dosage regimens allowing maintenance of effective antimicrobial levels should be established by pharmacokinetics. The following criteria should be taken into account: – – – – – bio-availability according to the route of administration; concentration of the antimicrobial at the site of infection and its distribution in the treated animal; metabolism that may lead to the inactivation of antimicrobials; excretion routes; use of combinations of antimicrobial agents should be justified. OIE International Standards on Antimicrobial Resistance, 2003 19 Terrestrial Animal Health Code b) Clinical trials Clinical trials should be performed to confirm the validity of the claimed therapeutic indications and dosage regimens established during the preclinical phase. The following criteria should be taken into account: i) diversity of the clinical cases encountered when performing multicentre trials; ii) compliance of protocols with good clinical practice; iii) eligibility of studied clinical cases, based on appropriate criteria of clinical and bacteriological diagnoses; iv) parameters for qualitatively and quantitatively assessing the efficacy of the treatment. 7. Assessment of the potential of antimicrobials to select for resistant bacteria Other studies may be requested in support of the assessment of the potential of antimicrobials to select for resistant bacteria. The interpretation of their results should be undertaken with great caution. The party applying for market authorisation should, where possible, supply data derived in target animal species under the intended conditions of use. Considerations may include: a) the concentration of active compound in the gut of the animal (where the majority of potential food-borne pathogens reside) at the defined dosage level; b) the level of human exposure to food-borne resistant bacteria; c) the degree of cross-resistance within the class of antimicrobials and between classes of antimicrobials; d) the pre-existing level of resistance in the pathogens of human health concern (baseline determination). Other studies may be requested in support of the assessment of the potential of antimicrobials to select for resistant bacteria. The interpretation of their results should be undertaken with great caution. 8. Establishment of acceptable daily intake, maximum residue level and withdrawal periods for antimicrobial compounds a) 20 When setting the acceptable daily intake (ADI) and MRL for an antimicrobial substance, the safety evaluation should also include the potential biological effects on the intestinal flora of humans. OIE International Standards on Antimicrobial Resistance, 2003 Terrestrial Animal Health Code b) The establishment of an ADI for each antimicrobial agent, and an MRL for each animal-derived food, should be undertaken. c) For each VMP containing antimicrobial agents, withdrawal periods should be established in order to produce food in compliance with the MRL, taking into account: i) ii) iii) iv) v) d) 9. the MRL established for the antimicrobial agent under consideration; the composition of the product and the pharmaceutical form; the target animal species; the dosage regimen and the duration of treatment; the route of administration. The applicant should provide methods for regulatory testing of residues in food. Protection of the environment An assessment of the impact of the proposed antimicrobial use on the environment should be conducted. Efforts should be made to ensure that environmental contamination with antimicrobials is restricted to a minimum. 10. Establishment of a summary of product characteristics for each veterinary medicinal product The summary of product characteristics contains the information necessary for the appropriate use of VMPs and constitutes the official reference for their labelling and package insert. This summary always contains the following items: a) b) c) d) e) f) g) h) i) j) k) pharmacological properties, target animal species, therapeutic indications, target bacteria, dosage and administration route, withdrawal periods, incompatibilities, expiry date, operator safety, particular precautions before use, particular precautions for the proper disposal of un-used products. Antimicrobials that are considered to be important in treating critical diseases in humans should only be used in animals when alternatives are either unavailable or inappropriate. Consideration should be given to providing such guidance by means of the product label and data sheet. The oral route should be used with caution. OIE International Standards on Antimicrobial Resistance, 2003 21 Terrestrial Animal Health Code 11. Post-marketing antimicrobial surveillance The information collected through pharmacovigilance programmes, including lack of efficacy, should form part of the comprehensive strategy to minimise antimicrobial resistance. a) Specific surveillance Specific surveillance to assess the impact of the use of a specific antimicrobial may be implemented after the granting of the marketing authorisation The surveillance programme should evaluate not only resistance development in target animal pathogens, but also in food-borne pathogens and/or commensals. Such surveillance will also contribute to general epidemiological surveillance of antimicrobial resistance. b) General epidemiological surveillance The surveillance of animal bacteria resistant to antimicrobial agents is essential. The relevant authorities should implement a programme according to the Terrestrial Code. 12. Distribution of the antimicrobial agents used in veterinary medicine The relevant authorities should ensure that all the antimicrobial agents used in animals are: a) b) c) d) prescribed by a veterinarian or other suitably trained and authorised person; delivered by an authorised animal health professional; supplied only through licensed/authorised distribution systems; administered to animals by a veterinarian or under the supervision of a veterinarian or by other authorised persons. 13. Control of advertising All advertising of antimicrobials should be controlled by a code of advertising standards, and the relevant authorities must ensure that the advertising of antimicrobial products: a) complies with the marketing authorisation granted, in particular regarding the content of the summary of product characteristics; b) is restricted to authorised professionals, according to national legislation in each country. 14. Training of antibiotic users Training of antibiotic users should focus on: 22 a) information on disease prevention and management strategies, b) the ability of antimicrobials to select for resistant bacteria in food-producing animals, OIE International Standards on Antimicrobial Resistance, 2003 Terrestrial Animal Health Code c) the need to observe responsible use recommendations for the use of antimicrobial agents in animal husbandry in agreement with the provisions of the marketing authorisations 15. Research The relevant authorities should encourage public- and industry-funded research. Article 3.9.3.4. Responsibilities of the veterinary pharmaceutical industry 1. Marketing authorisation of VMPs The veterinary pharmaceutical industry has responsibilities to: 2. a) supply all the information requested by the national regulatory authorities; b) guarantee the quality of this information in compliance with the provisions of good manufacturing, laboratory and clinical practices; c) implement a pharmacovigilance programme and on request, specific surveillance for bacterial susceptibility and resistance. Marketing and export of VMPS For the marketing and export of VMPs: 3. a) only licensed and officially approved VMPs should be sold and supplied, and then only through licensed/authorised distribution systems; b) only VMPs that have been authorised in the (exporting) country in which the product(s) is approved for sale or the quality of which is certified by a regulatory authority should be exported; c) the national regulatory authority should be provided with the information necessary to evaluate the amount of antimicrobial agents marketed. Advertising The veterinary pharmaceutical industry should: 4. a) disseminate information in compliance with the provisions of the granted authorisation; b) ensure that the advertising of antimicrobials directly to the livestock producer is discouraged. Training The veterinary pharmaceutical industry should participate in training programmes as defined in point 14 of Article 3.9.3.3. OIE International Standards on Antimicrobial Resistance, 2003 23 Terrestrial Animal Health Code 5. Research The veterinary pharmaceutical industry should contribute to research as defined in point 15 of Article 3.9.3.3. Article 3.9.3.5. Responsibilities of pharmacists 1. Pharmacists should only distribute veterinary antimicrobials on prescription. All products should be appropriately labelled (see point 5 of Article 3.9.3.6.). 2. The guidelines on the responsible use of antimicrobials should be reinforced by pharmacists who should keep detailed records of: a) b) c) d) e) f) 3. date of supply, name of prescriber, name of user, name of product, batch number, quantity supplied. Pharmacists should also be involved in training programmes on the responsible use of antimicrobials, as defined in point 14 of Article 3.9.3.3. Article 3.9.3.6. Responsibilities of veterinarians The prime concern of the veterinarian is to encourage good farming practice in order to minimise the need for antimicrobial use in livestock. Veterinarians should only prescribe antimicrobials for animals under their care. 1. Use of antimicrobial agents The responsibilities of veterinarians in this area are to carry out a proper clinical examination of the animal(s) and then: a) only prescribe antimicrobials when necessary; b) make an appropriate choice of the antimicrobial based on experience of the efficacy of treatment. On certain occasions, a group of animals that may have been exposed to pathogenic bacteria may need to be treated without recourse to an accurate diagnosis and antimicrobial susceptibility testing to prevent the development of clinical disease and for reasons of animal welfare. 2. Choosing an antimicrobial agent a) 24 The expected efficacy of the treatment is based on: OIE International Standards on Antimicrobial Resistance, 2003 Terrestrial Animal Health Code i) the clinical experience of the veterinarian; ii) the activity towards the pathogenic bacteria involved; iii) the appropriate route of administration; iv) known pharmacokinetics/tissue distribution to ensure that the selected therapeutic agent is active at the site of infection; v) the epidemiological history of the rearing unit, particularly in relation to the antimicrobial resistance profiles of the pathogenic bacteria involved; vi) Should a first line antibiotic treatment fail or should the disease recur, a second line treatment should ideally be based on the results of diagnostic tests. To minimise the likelihood of antimicrobial resistance developing, it is recommended that antimicrobials be targeted to bacteria likely to be the cause of infection. b) Combinations of antimicrobials are used for their synergistic effect to increase therapeutic efficacy or to broaden the spectrum of activity. Furthermore, the use of combinations of antimicrobials can be protective against the selection of resistance in cases in which bacteria exhibit a high mutation rate against a given antimicrobial. Some combinations of antimicrobials may, in certain cases, lead to an increase in the selection of resistance. 3. Appropriate use of the antimicrobial agent chosen A prescription for antimicrobial agents must indicate precisely the treatment regime, the dose, the dosage intervals, the duration of the treatment, the withdrawal period and the amount of drug to be delivered, depending on the dosage and the number of animals to be treated. As far as ‘Off label use’ (extra-label use) of veterinary medicinal products is concerned, although all medicinal products should be prescribed and used in accordance with the specifications of the marketing authorisation, the prescriber should have the discretion to adapt these in exceptional circumstances. 4. Recording All available information should be consolidated into one form or database. This information should: a) allow monitoring of the quantities of medication used; b) contain a list of all medicines supplied to each livestock holding; OIE International Standards on Antimicrobial Resistance, 2003 25 Terrestrial Animal Health Code 5. c) contain a list of medicine withdrawal periods and a system for allowing information to be updated; d) contain a record of antimicrobial susceptibilities; e) provide comments concerning the response of animals to medication; f) allow the investigation of adverse reactions to antimicrobial treatment, including lack of response due to antimicrobial resistance. Suspected adverse reactions should be reported to the appropriate regulatory authorities. Labelling All medicines supplied by a veterinarian should be adequately labelled with the following minimum information: a) b) c) d) e) f) g) h) the name of the owner/keeper or person who has control of the animal(s); the address of the premises where the animal(s) is kept; the name and address of the prescribing veterinarian; identification of the animal or group of animals to which the antimicrobial agent was administered; the date of supply; the indication ‘For animal treatment only’; the warning ‘Keep out of the reach of children’; the relevant withdrawal period, even if this is nil. The label should not obscure the expiry date of the preparation, batch number or other important information supplied by the manufacturer. 6. Training Veterinary professional organisations should participate in the training programmes as defined in point 14 of Article 3.9.3.3. Article 3.9.3.7. Responsibilities of livestock producers 1. Livestock producers with the assistance of a veterinarian, where possible, are responsible for preventing outbreaks of disease and implementing health and welfare programmes on their farms. 2. Livestock producers have to: 26 a) draw up a health plan with the veterinarian in charge of the animals that outlines preventative measures (mastitis plan, worming and vaccination programmes, etc.); b) use antimicrobial agents only on prescription, and according to the provisions of the prescription; OIE International Standards on Antimicrobial Resistance, 2003 Terrestrial Animal Health Code c) use antimicrobial agents in the species, for the uses and at the doses on the approved/registered labels and in accordance with product label instructions or the advice of a veterinarian familiar with the animals and the production site; d) isolate sick animals, when appropriate, to avoid the transfer of resistant bacteria; e) comply with the storage conditions of antimicrobials in the rearing unit, according to the provisions of the leaflet and package insert; f) address hygienic conditions regarding contacts between (veterinarians, breeders, owners, children) and the animals treated; g) comply with the recommended withdrawal periods to ensure that residue levels in animal-derived food do not present a risk for the consumer; h) dispose of surplus antimicrobials under safe conditions for the environment; partially-used medicines should only be used within the expiry date, for the condition for which they were prescribed and, if possible, in consultation with the prescribing veterinarian; i) maintain all the laboratory records of bacteriological and susceptibility tests; these data should be made available to the veterinarian responsible for treating the animals; j) keep adequate records of all medicines used, including the following: people i) ii) iii) iv) name of the product/active substance and batch number, name of supplier, date of administration, identification of the animal or group of animals to which the antimicrobial agent was administered, v) diagnosis/clinical conditions treated, vi) quantity of the antimicrobial agent administered, vii) withdrawal periods, viii) result of laboratory tests, ix) effectiveness of therapy; k) inform the responsible veterinarian of recurrent disease problems. OIE International Standards on Antimicrobial Resistance, 2003 27 Manual of diagnostic tests and vaccines for terrestrial animals Manual of Diagnostic Tests and Vaccines for Terrestrial Animals Chapter I.1.10. Laboratory methodologies for bacterial antimicrobial susceptibility testing Introduction The spread of multiple antibiotic-resistant pathogenic bacteria has been recognised by the OIE and the World Health Organization as a serious global animal and human health problem. The emergence of antimicrobial resistance among many bacterial pathogens makes antimicrobial susceptibility testing (AST) essential when antimicrobials are used in therapy. The resistance of a pathogen to an antimicrobial is highly predictive that the treatment will not be effective and the susceptibility of the pathogen is an excellent base for the choice of the antibacterial treatment. Thus, AST is an important component of effective treatment programme. Additionally, AST of recovered bacterial pathogens is a component of prudent antimicrobial use guidelines in animal husbandry world-wide and the veterinarian should have these data available (1). AST is also the basis of the epidemiological surveillance of bacterial pathogens in animals and humans. Such epidemiological surveillance provides a base to choose properly empirical treatment (first line therapy) and to detect the emergence and/or the dissemination of resistant bacterial strains or resistance determinants in different bacterial species. Standardisation and harmonisation of AST methodologies, used in epidemiological surveillance of antimicrobial drug resistance, are critical if data are to be compared among national or international surveillance/monitoring programmes of OIE Member Countries. It is essential that AST methods provide reproducible results in day-to-day laboratory use and that the data be comparable with those results obtained by an acknowledged ‘gold standard’ reference method. In the absence of standardised methods or reference procedures, susceptibility results from different laboratories cannot be reliably compared. This Chapter provides Guidelines for AST methodologies, and includes procedures to standardise and harmonise interpretation of antimicrobial susceptibility test results. 1. Test requirements In order to achieve standardisation of AST methods and comparability of AST results, the following requirements apply: OIE International Standards on Antimicrobial Resistance, 2003 29 Manual of diagnostic tests and vaccines for terrestrial animals i) the use of standardised AST methods and the harmonisation of susceptibility data (including interpretive criteria) are essential, ii) standardised AST methods and similar interpretive criteria should be accepted and used by all participating laboratories, iii) all AST methods should generate reproducible data, iv) all data should be reported quantitatively, v) establishment of national or regional designated laboratories is essential for the coordination of AST methodologies, interpretations and quality controls, vi) microbiological laboratories should conduct their work within an internal quality assurance system, vii) laboratories should become accredited, where applicable, and participate in external proficiency testing programmes, viii) specific bacterial reference/quality control strains are essential for determining intra- and inter-laboratory quality control, quality assurance and proficiency testing. 2. Antimicrobial susceptibility testing methodologies The following requirements should be respected: i) bacteria subjected to AST must be isolated in pure culture from the submitted sample, ii) the isolation procedure for that particular bacterium should be standardised so that the subject bacteria are consistently and correctly identified to the genus and/or species level, iii) when possible, bacterial isolates should be stored for future analysis (either lyophilisation or cryogenic preservation at –70°C to –80°C), iv) once the bacterium has been isolated in pure culture, the inoculum must be standardised to obtain accurate susceptibility results. The following factors influencing AST methods should be standardised: i) the composition of the agar and broth media used (pH, cations, thymidine or thymine, use of supplemented media), ii) the content of antimicrobial in the carrier (disk, strip, tablet), iii) composition of solvents and diluents for preparation of antimicrobial stock solutions, iv) growth and incubation conditions (time, temperature, atmosphere e.g. CO2), v) 30 agar depth, OIE International Standards on Antimicrobial Resistance, 2003 Manual of diagnostic tests and vaccines for terrestrial animals vi) the subsequent interpretive criteria. For these reasons, special emphasis has to be placed on reference procedures and standardised methods, as sufficient reproducibility can be attained only through the use of standardised methodology. 3. Selection of antimicrobial susceptibility testing methodology The selection of an AST methodology may be based on the following factors: i) ii) iii) iv) v) vi) vii) viii) 4. ease of performance, flexibility, adaptability to automated or semi-automated systems, cost, reproducibility, reliability, accuracy, national preference. Test methods The following three methods are the only ones that consistently provide reproducible and repeatable results: i) disk diffusion, ii) broth dilution, iii) agar dilution. a) Disk diffusion method Disk diffusion refers to the diffusion of an antimicrobial agent of a specified concentration from disks, tablets or strips, into the solid culture media, which has been seeded with a standardised bacterial inoculum. Disk diffusion is based on the determination of an inhibition zone proportional to the bacterial susceptibility to the antimicrobial present in the disk. The diffusion of the antimicrobial agent into the seeded culture media results in a gradient of the antimicrobial. When the concentration of the antimicrobial becomes so diluted that it can no longer inhibit the growth of the test bacterium, the zone of inhibition is demarcated. In theory, the edge of this zone of inhibition correlates with the minimum inhibitory concentration (MIC) for that particular bacterium/antimicrobial combination. In other words, the zone of inhibition correlates inversely with the MIC of the test bacterium. Generally, the larger the zone of inhibition, the lower the concentration of antimicrobial required to inhibit the growth of the organisms. However, this depends on the concentration of antibiotic in the disk and its diffusibility. OIE International Standards on Antimicrobial Resistance, 2003 31 Manual of diagnostic tests and vaccines for terrestrial animals Note: Disk diffusion tests based solely on the presence or absence of a zone of inhibition without regard to the size of the zone of inhibition are not acceptable AST methodology. • Considerations for the use of the disk diffusion methodology Disk diffusion is straightforward to perform, reproducible, and does not require expensive equipment. Its main advantages are: i) ii) low cost, ease in modifying test antimicrobial disks when required. Manual measurement of zones of inhibition may be time-consuming. Automated zone-reading devices are available that can be integrated with laboratory reporting and data-handling systems. A maximum of 12 disks can be placed on one 150 mm agar plate. A maximum of five disks can be placed on a 100 mm plate. Regardless of the number of disks placed on the agar surface, the disks should be distributed evenly so that they are no closer than 24 mm from centre to centre. b) Broth and agar dilution methods The aim of the broth and agar dilution methods is to determine the lowest concentration of the assayed antimicrobial that inhibits the growth of the bacterium being tested (MIC, usually expressed in mcg/ml or mg/litre). However, the MIC does not always represent an absolute value. The ‘true’ MIC is a point between the lowest test concentration that inhibits the growth of the bacterium and the next lower test concentration. Antimicrobial ranges should: i) encompass both the interpretive criteria (susceptible, intermediate and resistant) and quality control reference organisms. ii) take into consideration the antimicrobial concentrations that are achievable in vivo for a specific bacteria/antibiotic combination. Antimicrobial susceptibility dilution methods appear to be more reproducible and quantitative than agar disk diffusion. However, antibiotics are usually tested in doubling dilutions, which can produce inexact MIC data. Any laboratory that intends to use a dilution method and set up its own reagents and antibiotic dilutions should have the ability to obtain, prepare and maintain appropriate stock solutions of reagent-grade antimicrobials and to generate working dilutions on a regular basis. It is then essential that such laboratories use quality control organisms (see below) to assure accuracy and standardisation of their procedures. 32 OIE International Standards on Antimicrobial Resistance, 2003 Manual of diagnostic tests and vaccines for terrestrial animals • Broth dilution Broth dilution is a technique in which a standardised suspension of bacteria is tested against varying concentrations of an antimicrobial agent (usually doubling dilutions) in a standardised liquid medium. The broth dilution method can be performed either in tubes containing a minimum volume of 2 ml (macrodilution) or in smaller volumes using microtitration plates (microdilution). Numerous microtitre plates containing prediluted antibiotics within the wells are commercially available. The use of identical lots of microdilution plates may eliminate potential errors that may arise due to the preparation and dilution of the antimicrobials. The use of these plates with a standardised protocol, including appropriate quality control reference strains, may facilitate harmonisation where sufficient financial resources are available. Due to the fact that most broth microdilution antimicrobial test panels are prepared commercially, they can be considered to be less flexible than agar dilution or disk diffusion in adjusting to the changing needs of the surveillance/monitoring programme. Because the purchase of the equipment and antimicrobial panels may be costly, this methodology may not be the choice for laboratories with limited budgets. • Agar dilution Agar dilution involves the incorporation of an antimicrobial agent into an agar medium in a geometrical progression of concentrations, followed by the application of a defined bacterial inoculum to the agar surface of the plate. These results are often considered as the gold standard for the determination of an MIC for the test bacterium/antimicrobial combination. The advantages of agar dilution methods include: i) a greater control of the purity of the test bacterium, ii) the ability to test multiple bacteria on the same set of agar plates at the same time, iii) the potential to improve the identification of MIC endpoints and extend the antibiotic concentration range, iv) can be adapted to semi-automation. Commercially produced inoculum replicators are available and these can transfer between 32 and 37 different bacterial inocula to each agar plate. Agar dilution methods also have certain disadvantages, for example: i) they are very laborious and require substantial economic and technical resources, ii) once prepared they have to used within a week, OIE International Standards on Antimicrobial Resistance, 2003 33 Manual of diagnostic tests and vaccines for terrestrial animals iii) the endpoints are by no means always easy to read nor is the purity of the inoculum easy to verify. Agar dilution is often recommended as a standardised AST method for fastidious organisms, such as anaerobes, Helicobacter and Campylobacter species. However, at least in veterinary medicine, broth microdilution also works very well for organisms such as Hemophilus, Campylobacter and Brachyspira amongst others. c) Concentration gradient strips methodology Additionally, bacterial antimicrobial MICs can be obtained from commercially available gradient strips that diffuse a pre-formed antibiotic concentration. However, the use of gradient strips can be very expensive and MIC discrepancies can be found when compared with agar dilution results (2). Regardless of the AST method used, the procedures should be standardised to ensure accurate and reproducible results, and appropriate quality control reference organisms need to be tested every time AST is performed in order to ensure accuracy of the data. The appropriate AST choice will ultimately depend on the growth characteristics of the bacterium in question. In special circumstances, novel test methods and assays may be more appropriate for detection of particular resistance phenotypes. For example, chromogenic cephalosporin-based tests (e.g. nitrocefin) or equivalent methods may provide more reliable and rapid results for beta-lactamase determination in certain bacteria. Similarly, extended-spectrum beta-lactamase activity in certain bacteria can also be detected by using standard disk diffusion susceptibility test methods using specific cephalosporins (cefotaxime and ceftazidime) in combination with a beta-lactamase inhibitor (clavulanic acid) and measuring the resulting zones of inhibition. Additionally, chloramphenicol resistance attributed to production of chloramphenicol acetyl transferase can be detected in some bacteria via rapid tube or filter paper tests within 1–2 hours (4). 5. Antimicrobial susceptibility breakpoints and zone of inhibition criteria The objective of in vitro AST is to predict the way in which a bacterial pathogen may respond to the antimicrobial agent in vivo. The results generated by bacterial in vitro antimicrobial susceptibility tests, regardless of whether disk diffusion or dilution methods are used, are generally reported as resistant, susceptible or intermediate to the action of a particular antimicrobial. Antimicrobial susceptibility breakpoints are established by national standards organisations, professional societies or regulatory agencies. The relevant documents 34 OIE International Standards on Antimicrobial Resistance, 2003 Manual of diagnostic tests and vaccines for terrestrial animals should be consulted. However, there can be notable differences in breakpoints among different countries for the same antimicrobial agent. As mentioned previously, antimicrobial susceptibility testing results should be recorded quantitatively: i) as distribution of MICs in milligrams per litre, ii) or as inhibition zone diameters in millimetres. The following two primary factors enable a bacterium to be interpreted as susceptible or resistant to an antimicrobial agent: i) the development and establishment of quality control ranges, using diffusion when possible and dilution testing, for quality control microorganisms. This is essential for validating the specific AST method used. The quality control ranges for the quality control microorganisms should be established prior to determining breakpoints for susceptibility or resistance. ii) the determination of the appropriate interpretive criteria. This involves the generation of three distinct pieces of data: • population distribution of MICs of relevant microorganisms, • pharmacokinetic parameters of the antimicrobial agent, • results of clinical trials and experience. The interpretation of the data involves creating a scattergram from the bacterial population distribution (representative bacterial isolates), by plotting the zone of inhibition against the MIC for each bacterial pathogen. The selection of breakpoints is then based on multiple factors, including regression line analysis, bacterial population distributions, error rate bounding, pharmacokinetics, and ultimately, clinical verification. The development of a concept known as ‘microbiological breakpoints’, which is based on the population distributions of the specific bacterial species tested, may be more appropriate for some antimicrobial surveillance programmes. In this case, bacterial isolates that deviate from the normal susceptible population would be designated as resistant, and shifts in susceptibility to the specific antimicrobial/bacterium combination could be monitored (5). OIE International Standards on Antimicrobial Resistance, 2003 35 Manual of diagnostic tests and vaccines for terrestrial animals 6. Antimicrobial susceptibility testing guidelines A number of guidelines are currently available for antimicrobial susceptibility testing and subsequent interpretive criteria throughout the world. Amongst others, these include standards and guidelines published by: – – – – – – – National Committee for Clinical Laboratory Standards (NCCLS), British Society for Antimicrobial Chemotherapy (BSAC), Comité de l’Antibiogramme de la Société française de Microbiologie (CASFM), Swedish Reference Group for Antibiotics (SIR), Deutsches Institut für Normung (DIN), Japanese Society for Chemotherapy (JSC), Werkgroep richtlijnen gevoeligheidsbepalingen (WRG system, the Netherlands). At this time, only the NCCLS has developed protocols for susceptibility testing of bacteria of animal origin and determination of interpretive criteria (4). However, protocols and guidelines are available from a number of standards organisations and professional societies for susceptibility testing for similar bacterial species that cause infections in humans. It is possible that such guidelines can be adopted for susceptibility testing for bacteria of animal origin, but each country must evaluate its own AST standards and guidelines. Additionally, efforts focusing on harmonisation of susceptibility breakpoints on an international scale are progressing. These efforts have primarily focused on the adoption of the standards and guidelines of the NCCLS, which provide laboratories with standardised methods and quality control values enabling comparisons of AST methods and generated data. For those OIE Member Countries that have not standardised AST methods, the adoption of NCCLS guidelines and standards would be an appropriate initial step. As a first step towards comparability of monitoring and surveillance data, Member Countries should be encouraged to strive for harmonised and standardised programme design (6). Data from countries using different methods and study design may otherwise not be directly comparable (3, 6). Notwithstanding this, data collected over time in a given country may at least allow the detection of emergence of antimicrobial resistance or trends in prevalence of resistance in that particular country. However, if results achieved with different AST methods are to be presented side by side, then comparability of results must be demonstrated and consensus on interpretation achieved. Note: This will be best accomplished by the use of accurate and reliable standardised AST methods in conjunction with monitoring of AST performance with defined quality control bacterial strains among participating laboratories. 36 OIE International Standards on Antimicrobial Resistance, 2003 Manual of diagnostic tests and vaccines for terrestrial animals 7. Comparability of results To determine the comparability of results originating from different surveillance systems, results should be reported quantitatively including information on the methods, quality control organisms and antimicrobial concentration ranges tested and interpretive criteria used. 8. Quality control and quality assurance Adequate quality control/quality assurance systems should be established in AST performing laboratories. The following components should be monitored: i) precision of the AST procedure, ii) accuracy of the AST procedure, iii) performance of the appropriate reagents, iv) the laboratory personnel. The following requirements should be respected: i) Strict adherence to standardised techniques in conjunction with quality control of media and reagents. ii) Record keeping of: • lot numbers of all appropriate materials and reagents, • expiration dates of all appropriate materials and reagents. iii) The appropriate quality control reference bacteria should also be tested to ensure standardisation regardless of the AST method used. iv) Reference bacterial strains should be catalogued and characterised with stable defined antimicrobial susceptibility phenotypes. These quality control strains should also encompass resistant and susceptible ranges of the antimicrobials to be assayed. v) Laboratories involved in AST should use the appropriate quality control reference strains. vi) Reference strains should be kept as stock cultures from which working cultures are derived and should be obtained from national or international culture collections. Reference bacterial strains should be stored at designated centralised or regional laboratories. OIE International Standards on Antimicrobial Resistance, 2003 37 Manual of diagnostic tests and vaccines for terrestrial animals vii) The preferred method for analysing the overall performance of each laboratory is to test the appropriate quality control bacterial strains on each day that susceptibility tests are performed. Because this may not always be practical or economic, the frequency of such quality control tests may be reduced if the laboratory can demonstrate that the susceptibility testing procedures are reproducible. If a laboratory can document the reproducibility of the susceptibility testing methods used, testing may be performed on a weekly basis. If quality control errors emerge, the laboratory has a responsibility to determine the cause(s) and repeat the tests. If the laboratory cannot determine the source of error(s), then quality control testing should be re-initiated on a daily basis. viii) Recognised quality control strains should be tested each time a new batch of medium or plate lot is used and on a regular basis in parallel with the bacterial strains to be assayed. ix) Appropriate biosecurity issues should be addressed in obtaining and dispersing quality control reference strains to participating laboratories. The use of such strains will allow for comparison of antimicrobial susceptibility data (run on). 9. External proficiency testing To ensure that reported susceptibility data is accurate, OIE Member Countries should initiate external proficiency testing (e.g. third party testing). External proficiency testing can be carried out on a national basis. Laboratories in Member Countries are encouraged to participate in international inter-laboratory comparisons. All important bacterial species should be included. Countries should appoint or establish designated national laboratories that are responsible to: i) monitor the quality assurance programmes of laboratories participating in surveillance and monitoring of antimicrobial resistance, ii) supply to those laboratories a set of reference strains. References 1. Anthony F., Acar J., Franklin A., Gupta R., Nicholls T., Tamura Y., Thompson S., Threllfall E.J., Vose D., van Vuuren M. & White D.G. (2001). – Antimicrobial resistance: responsible and prudent use of antimicrobial agents in veterinary medicine. Rev. sci. tech. Off. int. Epiz., 20, 829-839. 2. Brown D.F. & Brown L. (1991). – Evaluation of the E-test, a novel method of quantifying antimicrobial activity. J. Antimicrob. Chemother., 26, 185-190. 3. Leegard T.M., Caugant D.A., Froholm L.O. & Hoib E.A. (2000). – Apparent differences in antimicrobial susceptibility as a consequence of national guidelines. Clin. Microbiol. Inf. Dis., 6, 290-293. 38 OIE International Standards on Antimicrobial Resistance, 2003 Manual of diagnostic tests and vaccines for terrestrial animals 4. National Committee for Clinical Laboratory Standards (NCCLS) (2002). – Document M31-A2. Performance standards for antimicrobial disk and dilution susceptibility tests for bacteria isolated from animals, approved standard, 2nd Edition. NCCLS, Wayne, Pennsylvania, 80 pp. 5. Ringertz S., Olsson-Liljequist B., Kahlmeter G. & Kronvall G. (1997). – Antimicrobial susceptibility testing in Sweden II. Species-related zone diameter breakpoints to avoid interpretive errors and guard against unrecognised evolution of resistance. Scand. J. Infect. Dis., 105 (Suppl.), 8-12. 6. Threlfall E.J., Fisher I.S.T., Ward L., Tschape H. & Gerner-Smidt P. (1999). – Harmonisation of antibiotic susceptibility testing for Salmonella: results of a study by 18 national reference laboratories within the European Union-funded Enter-Net group. Microb. Drug Resist., 5, 195–199. __________ OIE International Standards on Antimicrobial Resistance, 2003 39 Proceedings of the OIE International Conference OIE International Standards on Antimicrobial Resistance, 2003 41 1. General aspects OIE International Standards on Antimicrobial Resistance, 2003 43 1. General aspects Antimicrobial resistance: an overview J. Acar (1) & B. Röstel (2) (1) Université Pierre et Marie Curie, Service de Microbiologie Médicale, Fondation Hôpital Saint-Joseph, 185 rue Raymond Losserand, 75674 Paris Cedex 14, France (2) Centre collaborateur de l’OIE (Organization mondiale pour la santé animale) pour les médicaments vétérinaires, Agence française de sécurité sanitaire des aliments (AFSSA) Fougères, Agence nationale du médicament vétérinaire (ANMV), B.P. 90203, 35302 Fougères Cedex, France This report, prepared by the OIE Ad hoc Group of experts on antimicrobial resistance, has not yet received the approval of the International Committee of the OIE Summary Increased antimicrobial resistance in bacteria that are important pathogens of humans, and spread of resistance from the closed environment of hospitals into open communities are increasingly perceived as a threat to public health. Any antimicrobial use, whether in humans, animals, plants or food processing technology, could lead to bacterial resistance. Use of antimicrobials in livestock production is suspected to significantly contribute to this phenomenon in species of bacteria which are common to humans and animals. Further research is required into the specific use conditions that govern the selection and dissemination of resistant bacteria. International travel and trade in animals and food increase the risk of antimicrobial resistance world-wide. Countries are considering import restrictions for products deemed a risk to public health. The World Organisation for Animal Health, a World Trade Organization reference organisation for the Agreement on the Application of Sanitary and Phytosanitary Measures, develops international standards on antimicrobial resistance which, as is the case for national measures, must be based on risk analysis. The scientific background and problems of resistance in human medicine are reviewed. Current knowledge, missing information and actions to be taken are identified. Keywords Agreement on the Application of Sanitary and Phytosanitary Measures – Antimicrobial resistance – Containment of resistance – Food – International standards – National measures – Public health – Resistance mechanisms – Risk analysis – World organisation for animal health. Introduction The existence of antimicrobial resistance, the increase in resistance to a number of antibiotics of bacteria that are important human pathogens, and the spread of resistance from the rather closed environment of hospitals into open communities, are increasingly perceived as a threat to public health. The appearance of new resistance mechanisms, the development of multi-drug resistance or combinations of resistance, and the facility with which genetic material encoding resistance may, in certain cases, spread horizontally between different species of bacteria, all increase the feeling of defencelessness against diseases that were thought to have been controlled when antibiotics were first developed. OIE International Standards on Antimicrobial Resistance, 2003 45 1. General aspects Reports referring to apocalyptic visions of the plague depopulating nations and women dying of puerperal fever, are prone to increase public fears rather than helping to appropriately address important matters of public health. Unfortunately, these kind of publications, such as ‘World leading killers planning their escape’ are rather common and are not only communicated by the kind of media aiming at increasing their sales figures. Any use of antibiotics, may it be for human, animal, plant or food-processing technology, has the potential to lead, at some point in time, to bacterial resistance. Although many publications are beginning to appear, little is known about the different conditions of use under which antibiotics preferably select, or select to a lesser extent, for resistant bacteria. Resistance, once developed, is not bound to borders of different ecological environments or countries. Limited scientific research on resistance (abandoned for several decades and only recently re-established, pushed by growing concerns) and the consequent lack of scientific data leave society and decision makers in the uncomfortable situation of requesting and deciding upon corrective actions when the underlying causes may not have been appropriately identified. This situation, coupled with the slowing, and in certain sectors disappearing, discovery and development of new antibiotics with new mechanisms of action, creates an atmosphere of anxiety calling for immediate action, whether efficient or not. Globalisation, new trade environments and transfer of resistant bacteria through international travel, and trade of animals and food, raise the risk of the spread of resistance world-wide. It also bears the risk of countries closing borders for trade on the basis of inappropriately evaluated risks or perceived risks. Antimicrobial resistance – a responsibility of the OIE Why has the World Organisation for Animal Health taken action on antibiotic resistance? Countries must protect animal and human health. This also includes protecting against risks arising from bacteria resistant to antimicrobial treatment. At the same time, members of the World Trade Organization (WTO) must respect their obligations under the WTO Agreement on the Application of Sanitary and Phytosanitary Measures (SPS), which are to base any sanitary measure on risk assessment and scientific evidence and to restrict measures to the extent necessary to achieve the chosen level of protection. In cases where several potential measures exist, the least trade restrictive measure must be chosen. The OIE (World animal health organisation) is the organisation recognised by the WTO for the elaboration of international standards, guidelines and recommendations on matters of animal health and zoonoses relevant for the trade of animals and animal products. 46 OIE International Standards on Antimicrobial Resistance, 2003 1. General aspects Antibiotic resistance, as it relates to zoonotic bacteria and to resistance determinants (which may be transferred between animals and from animals to humans), and the measures to be taken in view of their control are the responsibility and field of competence of the OIE. The OIE is the appropriate organisation to prepare international recommendations on the detection and control of antimicrobial resistance as they relate to zoonotic bacteria and resistance determinants as they originate from animal bacteria. These standards, when finalised and adopted by the OIE International Committee, will serve as a WTO reference standard, should trade disputes arise. What action has been taken by the World Organisation for Animal Health? A report on existing activities and capacities for the detection and control of antibiotic resistance was made in 1998 to fifty countries of Europe at the OIE Regional Commission for Europe. This report emphasised that additional efforts should be made to develop official antimicrobial resistance surveillance/monitoring programmes, to improve their harmonisation and the harmonisation of laboratory methodologies, which in turn will improve the reliability and comparability of generated resistance data. The report also pointed out that risk analysis was not commonly used when the implementation of sanitary measures was considered by countries. Based on this report, the OIE Regional Commission for Europe recommended to the OIE International Committee that an international Ad hoc Group of experts be formed to address, using a comprehensive and multidisciplinary approach, human and animal health risks related to antimicrobial resistance originating from the use of antimicrobials in veterinary medicine. The OIE International Committee endorsed this recommendation in May 1999 and the OIE Director General appointed the Ad hoc Group of experts on antimicrobial resistance. The OIE Ad hoc Group of experts decided to engage in a three pillar strategy: – immediate measures to contain and reduce antimicrobial resistance (prudent and responsible use of antimicrobials) – development of tools to assess and manage the risks to animal and human health (risk analysis methodology), and harmonisation of surveillance systems and laboratory methodologies – improve knowledge on antimicrobial resistance world-wide (information gathering). The achievements of the Ad hoc Group of the World Organisation for Animal Health As a result of the work of the OIE Ad hoc Group, countries are now gaining access to a set of comprehensive methodologies, assuring that the identification of, and OIE International Standards on Antimicrobial Resistance, 2003 47 1. General aspects decision upon appropriate intervention measures are conducted in an objective, science-based, transparent and defensible way. The specific considerations that were given to the different conditions (geographical, use of antimicrobials, resistance situation) and to the technical capabilities and capacities of countries around the world, open the way for an equal application of these methodologies both to developing and developed countries. The OIE Ad hoc Group emphasises that where animal or human health problems exist throughout the world, without respect to national borders, all countries have an equal need to protect their animal and human populations and their national trade interests. The tools underlying two of the three pillars of the recommendation of the OIE Ad hoc Group (monitoring the quantities of antimicrobials used in animal husbandry and the immediate measures to contain antimicrobial resistance through the prudent and responsible use of antibiotics) and the tools to assess and manage risks to animals and humans (risk analysis methodology, harmonisation of surveillance systems and laboratory methodologies) became available for implementation by country governments and competent authorities. The five respective guidelines, prepared by the OIE Ad hoc Group with the participation of the Food and Agriculture Organization (FAO) and the World Health Organization (WHO) compose the body of this document. The third pillar (to improve knowledge on antimicrobial resistance world-wide) will be constructed as the implementation of the tools constituting the first two pillars proceeds and results are obtained. Future directions of the World Organisation for Animal Health The 69th General Session of the OIE International Committee of May 2001 adopted Resolution No. XXV requesting the OIE Specialist Commissions to develop international standards in the area of antimicrobial resistance. The Specialist Commissions will use the recommendations of the OIE Ad hoc Group to develop these standards. The draft standards proposed by the Specialist Commissions will be circulated for comments to the OIE Member Countries. Revised draft standards will be circulated a second time to the OIE Member Countries and after a second revision, as appropriate, be submitted for adoption to the 70th General Session of the OIE International Committee in May 2002. The standards will be published in the International Animal Health Code and the Manual of Standards for Diagnostic Tests and Vaccines. The OIE is taking steps to encourage Member Countries to make use of the new methodologies in order to establish an objective, science-based view on the subject and consequently contain antimicrobial resistance in animal bacteria. The OIE will undertake the necessary steps to provide assistance, as appropriate, to its Member Countries, on aspects related to the implementation of this standard. 48 OIE International Standards on Antimicrobial Resistance, 2003 1. General aspects The OIE Standards Commission decided during its spring meeting in January 2001 to introduce standards for antimicrobial sensitivity testing into the OIE Manual of Standards for Diagnostic Tests and Vaccines. The Standards Commission also recommended the designation of OIE Reference Laboratories for the detection and quantification of antimicrobial resistance in animal bacteria. These laboratories will, among others, assist OIE Member Countries in setting up microbiological laboratories, where appropriate, and in placing the work of these laboratories under quality assurance. At the 14th Conference of the OIE Regional Commission for Africa, held on 23-26 January 2001, Member Countries decided to actively engage in the promotion of the prudent use of antimicrobials in animals and to undertake efforts to establish national programmes for the management of antimicrobial resistance. The Regional Commission for Africa also recommended that OIE Reference Laboratories assist OIE Member Countries in implementing quality assurance schemes in national microbiological laboratories and in participating in external proficiency testing programmes. The OIE will continue to insist that sanitary measures are based on risk assessment of sound scientific data, and conducted according to appropriate recommended methodologies. Some scientific facts The scientific background What is antimicrobial resistance? Antimicrobial resistance is the capacity of bacteria to survive exposure to a defined concentration of an antimicrobial substance. Antimicrobial resistance has multiple definitions according to the scientific discipline and the goals involved: – clinical definition: the bacteria survive an adequate treatment with an antibiotic – pharmacological definition: the bacteria survive a range of concentrations expressing the various amounts of an antibiotic present in the different compartments of the body when the antibiotic is administered at the recommended dose – microbiological and molecular definition: the bacteria have a mechanism which governs a higher minimum inhibitory concentration (MIC) than the original or wild bacteria – epidemiological definition: any group of bacterial strains which can be distinguished from the normal (Gauss) distribution of MICs to an antibiotic. Bacterial resistance to a particular antibiotic can be a natural property of the bacteria or a secondarily acquired mechanism. Surviving the effect of an antibiotic is a normal reaction of a bacterial cell. When successful, such a reaction gives origin to a clone of bacterial cells able to confront the antibiotic. However, according to the mechanism of resistance, the bacterial clone may confront different amounts of antibiotic, ranging from a small amount, close to the amount formerly able to inhibit the growth (MIC) OIE International Standards on Antimicrobial Resistance, 2003 49 1. General aspects of the bacterial cell, to a very large quantity of antibiotics (e.g. hydrolysing exoenzyme produced by the bacteria). It is a very well known fact that bacteria can resist any antibiotic, and this is a global phenomenon which affects all countries. However, characteristics of the resistance phenomenon relate to the affected bacterial species, the set of antibiotics involved, the distribution of the resistant strains in particular settings in which antibiotics are used (hospital, community, animal husbandry, etc.). The resistant strains are classified according to their identification (genus, species) and to their antibiotic resistance phenotype (sometimes referred to as antibiotype or resistance pattern). The antibiotic resistance phenotype is established by the comparison of the list of antibiotics active on the reference (original, wild) strain of the bacterial species (which may have a natural resistance to some antibiotics) with the list of antibiotics to which the strain tested is resistant. This represents acquired resistance, which is the opposite of natural resistance. It is recommended to group the antibiotics under classes and subclasses according to the mechanism of resistance. This is called cross-resistance (e.g. a β-lactamase Tem type in Escherichia coli governs a cross-resistance to all aminopenicillins, all ureidopenicillins and a few first generation cephalosporins. The cross-resistance for six compounds is expressed by ampicillin resistance). When the sequence of resistance markers concerns different classes of antibiotics, this is referred to as co-resistance. Bacterial resistance to an antibiotic can be considered according to three aspects described below. The mechanism by which the bacteria is able to resist It is important to note that a bacterial cell often possesses more than one mechanism to resist to an antibiotic. Co-operation between several resistance mechanisms often generates high level resistance. A summary of resistance mechanisms, the affected antibiotics and the resulting level of resistance, is presented in Table I. The genetic mechanism governing the proteins involved in the resistance and the origin of resistance Two genetic mechanisms are involved, namely: mutation in an existing gene (chromosome or plasmid), and the de novo acquisition of a gene governing resistance. The location of the altered or the new genes is important (chromosomes, integrons, transposons or plasmids). The most important consequence of the location of the resistance genes concerns the spread of the resistance: – a chromosomal mutation affects a bacterial cell. The clone issued from this cell will multiply and spread. This mode of spread is often called vertical transmission of resistance – a resistance gene located on a transposon or a plasmid can be transmitted horizontally, independently from the spread of the resistant clone. Moreover, the 50 OIE International Standards on Antimicrobial Resistance, 2003 1. General aspects horizontal transmission may occur between different bacterial species. Concomitantly or independently to the expansion of the resistant bacteria, plasmid (gene) epidemics can occur. Many of them have been reported affecting six to eight species of Gramnegative bacteria. Plasmids or transposons are the main systems (genetic material) transferring resistance from bacteria (donor) to bacteria (recipient). They usually carry more than one marker of resistance. Large plasmids may transfer several different mechanisms of resistance against a number of different antibiotics. Their concurrent appearance in the same bacteria explains that one antibiotic may continue to co-select for the whole set of resistance mechanisms (multi-drug resistance). The medical (or therapeutic) aspect Bacterial resistance is recognised in medicine when an antibiotic treatment fails to cure a patient and the bacterial pathogen persists unharmed by the antibiotic prescribed. This is the point in time when human medicine becomes concerned. It is important to note that when a particular resistance emerges in a human bacterial pathogen (except the rapid selection of a chromosomal mutant bacteria), it can be assumed that many bacteria (commensal, environmental and animal) have also acquired the same resistance mechanism. A delay, sometimes very long, elapsed between the emergence of the resistance mechanism and its medical visibility. Table I Resistance mechanisms, the affected antibiotics and the resulting level of resistance Mechanism Affected antibiotics Level of resistance Efflux Tetracyclines Macrolides Quinolones Others in different systems Low Penetration β-lactams Chloramphenicol Trimethroprim Tetracyclines Low Target alteration β-lactams Aminoglycosides Macrolides Quinolones Rifampicin Glycopeptides Sulphonamides Trimethropim Variable By-pass Enzyme detoxification β-lactams Aminoglycisides Macrolides Chloramphenicol Lincosamide OIE International Standards on Antimicrobial Resistance, 2003 High High 51 1. General aspects Why and how does resistance develop? Resistance develops as an answer to the selection pressure exerted by an antibiotic, or by another compound (e.g. antiseptic), provided that they share at least one similar mechanism of resistance. Two conditions should be met. The selecting substance (selector) must be in prolonged contact with the bacterial population. The selector should be at a concentration which allows the bacteria to survive. This is generally referred to as a sub-inhibitory concentration. However, it should be noted that the lower limit of a sub-inhibitory concentration still acting as a selector has been poorly explored. An apparent contradiction exists between the statement that there is a positive correlation between high consumption/usage of antibiotics in a country and resistant bacteria, and the fact that low doses of antibiotics in an individual ecosystem (patient, animal or environment) are more selective for resistance. This stems from the fact that two different systems are compared, which are not directly related. One system is the total human or animal population in a country, the other system is the bacterial populations in a patient or group of patients. High consumption of antibiotics is a surrogate measure of the amount of antibiotics distributed among humans, animals and the environment. The important criteria is not the high concentration of an antibiotic in an individual patient or animal, but the extent of distribution of the antibiotics in the ecosystem. The larger the distribution of antibiotics, the greater the chance that in some place a large population of bacteria will be in contact with the correct selecting concentration of the antibiotic. The different events in bacterial life which can lead to the development of resistance are shown in Table II. Table II Events leading to the development of resistance Events Result Chromosomal mutations (one or more) Induction Derepression (pre-existing mechanism not expressed or barely expressed) Mutation (plamid) Altered target Altered cell wall Efflux system Modifying enzyme Modifying enzyme Inducible trait: constitutive Modifying enzyme Efflux By-passing target Acquisition of genes Plasmids, transposons, phages 52 Spread Clonal (not transferable) Clonal and transferable Transferable Needs donor strains and recipient strain OIE International Standards on Antimicrobial Resistance, 2003 1. General aspects In the first case, the resistance trait is generated from a mutation which occurs in the particular clone in contact with the selector. In the second case, the acquisition of a resistance gene requires the transfer of such a gene from a donor strain to the recipient strain. In that case, two populations of bacteria are required in the presence of the antibiotic, with one resistant (donor) and one susceptible (recipient). It should be noted that in infectious diseases, there is usually one bacterial population at the site of infection. Resistance due to mutation can be acquired by the patient during treatment (rifampin, fluoroquinolones). Only emergence of resistance due to mutation can be easily and rapidly observed in the patient, whereas resistance due to acquisition of genes is only recognised after a delay of months or even years. The primary origin of genes governing mechanism of resistance and able to circulate between bacteria of the same or different species remains partly unknown. However, it is accepted that these genes may originate from the antibiotic-producing organism, which uses the mechanism to survive its own antibiotic production (e.g. enzyme modifying aminoglycoside). This is an event that occurs in nature and is independent of the man-made production and use of antibiotics. There is also evidence that transferable resistance genes can originate from a ‘pick up’ mechanism which mobilises a gene from the chromosome of a naturally resistant bacteria (e.g. plasmid located cephalosporinase). Although no strict separation can be made between chromosomal resistance and transferable resistance, it is useful to keep the distinction as an epidemiological tool. It can be of importance to recognise that the spread of quinolone resistant strains is clonal (resistance to quinolone is a chromosomal mutation) compared to that of ceftriaxon resistant Salmonella. In that case, plasmid-mediated spread is combined with clonal spread, and can be linked to different Salmonella species or other Enterobacteriacae. Where does resistance develop? Very few studies have demonstrated the favourable niche for resistance development. The selector must be mixed with the bacterial population at the right concentration and time needed for the selection and the multiplication of the resistant population. In a unique bacterial population, emerging resistance is necessarily selected through mutations in the pre-existing genes of the bacterial cells. When the niche comprises mixed populations of different bacteria (resistant and susceptible), the emerging resistance may be due either to mutation or to a de novo acquisition of genes from a resistant bacterial cell. An antibiotic entering a niche with several bacterial species will kill the susceptible bacteria while the resistant species will multiply by mere vital advantage. Resistant clones from the susceptible population may be selected provided that: OIE International Standards on Antimicrobial Resistance, 2003 53 1. General aspects a) the antibiotic concentration has declined, allowing the survival of a portion of the susceptible population b) a mutant bacterial cell exists in the bacterial population (the frequency of mutation varies greatly between antibiotics) c) a transfer of genes (plasmids or transposons) has taken place between the preexisting resistant bacterial cells and the surviving susceptible bacterial cells. The most studied location of emergence of resistance is the digestive system of humans and animals. The enormous number of bacteria and species and the obligatory presence of most antibiotics in the gut (oral administration and bile elimination) explain the importance of this essential niche for resistance emergence. Clearly, mutant bacterial cells are theoretically selectable in any site where bacteria and antibiotic are in contact (e.g. abscess, empyema, urine, etc.). However, resistance by chromosomal mutation is not the most frequent system of resistance and does not affect a large number of antibiotics. The only antibiotics affected are those for which a high frequency of mutation (> 10–8) exists (rifampin, fusidic acid, quinolones and phosphomycin). In these cases, mutations are readily selected. Resistant clones are often observed during treatment or shortly afterwards. Such emergence of resistant strains is spectacular and easily observed by medical doctors or veterinarians. In fact, mutations are a small part of the resistance problem. The principal problem is related to the selection and the stabilisation of mechanisms governed by foreign genes acquired by the originally susceptible bacterial cells. As mentioned earlier, the intense circulation of bacteria can either be gene circulation or bacterial cell circulation. The emergence of multiple-resistant pathogens (with plasmids, transposons or integrons) takes a long time, during which numerous bacteria (commensal, environmental) are involved; during this period the phenomenon is not clinically visible. In this case, emergence of the resistant pathogen occurs far from the prescriber of the antibiotic and a long time after the original selection. The second important locations where resistant strains are built and selected are those related to the environment (water, soil, animal litter, sewage, hospital fomites, etc.). Several antibiotics can be present together in a niche. Those antibiotics will select for resistance separately, but also in a co-operative manner, if bacteria exist which are already resistant to them. The multiple resistant strains are favoured, since they can more easily survive the exposure to multiple antibiotics. Those multiple resistant strains are also more likely to acquire a new resistance. 54 OIE International Standards on Antimicrobial Resistance, 2003 1. General aspects What problems are faced? Infectious diseases and resistance in human medicine The difficulties faced in depicting the current situation of antimicrobial resistance are related to the limited, or in many cases lacking, systematic official disease investigations and reporting in this area. Hard data, such as laboratory confirmed cases, are limited even in developed countries where sophisticated disease investigation and reporting systems exist. Total disease burden, morbidity, mortality and economic impact descriptions are based on estimations, which may inherit errors and uncertainties depending on the validity of the underlying assumptions. The few countries that have, in recent years, started official resistance surveillance are beginning to obtain in vitro bacterial susceptibility data. However, systematic reporting of data on clinical outcomes is limited. Therefore, in many instances, in vitro data may have to be interpreted without being able to relate back to clinical outcomes. The reasons for this situation may be related initially to the inherent costs of disease investigation and reporting, but also to political unawareness and the potential negative impact of disease statistics on public opinion. According to the WHO, the emergence and spread of antimicrobial resistance in human pathogens is considered a global problem which increasingly affects the successful treatment of infectious diseases in humans. The WHO has identified six diseases (tuberculosis, malaria, pneumonia, human immunodeficiency virus (HIV)/acquired immune deficiency syndrome (AIDS), diarrhoea and measles) which cause 90% of infectious disease deaths world-wide. A proportion of these diseases is caused by bacteria and a relatively larger proportion by parasite and virus infections. To provide an overview on the resistance situation in medicine, these infections will be briefly reviewed. Tuberculosis Tuberculosis, a disease once thought to be controlled, is currently responsible for the deaths of 1.5 million people a year (a further 0.5 million die from a combination of tuberculosis and HIV/AIDS). Nearly two billion people (one-third of the population of the world) have latent tuberculosis infection. This constitutes a huge potential reservoir for the disease. Tuberculosis is one of the biggest infectious killers of adolescents and adults, and a leading cause of death among women. In addition, infection with HIV weakens the immune system and can activate latent tuberculosis. Infection with HIV is also believed to multiply the risk of contracting tuberculosis. Approximately one-third of all AIDS deaths are currently caused by tuberculosis. Because a patient may have both AIDS and tuberculosis, the reservoir for tuberculosis has increased and threatens more people in the community. OIE International Standards on Antimicrobial Resistance, 2003 55 1. General aspects Moreover, tuberculosis is becoming increasingly resistant to anti-tuberculosis drugs. Researchers assess the approximate number of multi-drug resistant tuberculosis cases at between 1% and 2% of current global tuberculosis figures. However, in some parts of the world, the rates of multi-drug resistant tuberculosis are much higher. China (Henan and Zhejiang), India (Tamil Nadu), Iran, Mozambique and Russia (Tomsk) each reported high levels of multi-drug resistant tuberculosis (over 3%) in new cases. Israel, Italy, Mexico (Baja California, Oaxaca and Sinaloa) reported multi-drug resistant tuberculosis in over 6% of both new and previously treated cases. Malaria Malaria kills over one million people a year, most of them young children. Most malaria deaths occur in sub-Saharan Africa, where malaria accounts for one in five of all childhood deaths. Women are especially vulnerable during pregnancy, suffering miscarriages or giving birth to premature, low-weight babies, and are more likely to die from the disease. An estimated 300 to 400 million people world-wide are infected by this mosquito-borne parasite each year. The development of resistance in the malaria parasite shows similarities to bacterial resistance. Acquired immune deficiency syndrome and sexually transmitted infections At the end of 1999, an estimated 33.6 million individuals were living with HIV worldwide. There is still no cure on the horizon. In some countries, up to one in four of the adult population is now living with HIV/AIDS. The worst affected region is subSaharan Africa. A small but growing number of patients are showing primary resistance to zidovudine (AZT), as opposed to ‘secondary’ resistance where viruses grow increasingly insensitive to antivirals over the course of the illness. This is also true for protease inhibitors which became available only ten years ago. Gonorrhoea and sexually transmitted infections (STIs) are important co-factors in the transmission and spread of HIV. This is because HIV bonds to white blood cells collecting at inflamed sites around the uro-genital tract. Studies show that those coinfected with gonorrhoea and HIV shed HIV at nine times the rate of individuals affected with HIV alone. Of the STIs, including chancroid and chlamydial infections, gonorrhoea is the most resilient, with a resistance rate that continues to outstrip new treatment strategies. Gonorrhoea resistance was first reported in American servicemen during the Vietnam war and is now entrenched around the globe, with multi-drug resistant strains appearing in 60% of those infected each year. In most of South-East Asia, resistance to penicillin has been reported in nearly all strains at an overall rate of 98%. Recent, more expensive drugs, notably ciprofloxacin, are likewise showing an increasing failure rate. Owing to resistance, chronic gonorrhoea has become a driving force in the HIV epidemic. 56 OIE International Standards on Antimicrobial Resistance, 2003 1. General aspects Pneumonia Acute respiratory infections (ARIs) are responsible for 3.5 million deaths each year. Pneumonia, the most dangerous ARI, kills more children than any other infectious disease. Most of these deaths (99%) occur in developing countries, while in industrialised countries childhood deaths from pneumonia are rare. Pneumonia often affects children with low birth weight or those whose immune systems are weakened by malnutrition or other diseases. Without treatment, pneumonia kills quickly. The major causes of pneumonia are the influenza virus and Streptococcus pneumoniae. The development of resistance to penicillin G by S. pneumoniae is now recognised world-wide. However, the prevalence of resistant strains ranges from 5% to 70% of the investigated laboratory samples. Most of these strains are also resistant to several other antibiotics (macrolides, tetracyclines, trimethoprim), dangerously restricting the choice of first-line therapy. Measles Measles is the most contagious disease known to mankind. It is a major childhood killer in developing countries, accounting for approximately 900,000 deaths a year. The measles virus may ultimately be responsible for more child deaths than any other single microbe, due to complications from pneumonia, diarrhoea and malnutrition. Hospital acquired infections Methicillin-resistant Staphylococcus aureus (MRSA), vancomycin-resistant Gram-negative rods (VRE) and enterococci and fermentative Gram-negative Entero-bacteriaceae are the most frequent multi-drug resistant bacteria isolated in hospitals both in developed and developing countries, and are responsible for the most difficult-to-treat hospital infections. Diarrhoeal diseases Diarrhoeal diseases claim nearly two million lives a year among children under five. These diseases are so widespread in developing countries that parents often fail to recognise the danger signs. Children die simply because their bodies are undernourished through lack of food and then are weakened through rapid loss of fluids. Diarrhoeal diseases impose a heavy burden on developing countries, accounting for 1.5 billion cases of illness a year in children under five. The burden is highest in deprived areas where there is poor sanitation, inadequate hygiene and unsafe drinking water. In certain developing countries, epidemics of diarrhoeal diseases such as cholera and dysentery affect both adults and children. Other diarrhoeal diseases include typhoid fever, rotavirus infection, salmonellosis and campylobacteriosis. Multi-drug resistance is also occurring in microbes that cause diarrhoeal diseases. One such agent, Shigella dysenteriae, is a highly virulent microbe that is resistant to almost every available drug. The results of this growing crisis were illustrated most notably in OIE International Standards on Antimicrobial Resistance, 2003 57 1. General aspects the wake of the 1994 civil war in Rwanda when the bacterium spread through vulnerable refugee populations already traumatised by war and loss. Left untreated, death can follow within days of infection. Ten years ago, a shigella epidemic could easily be controlled with co-trimoxazole, a drug available in generic form at low cost. Today, nearly all shigella are non-responsive to the drug, while resistance to ciprofloxacin (the only remaining viable medication) appears to be imminent. The bacteria that cause cholera and typhoid are also revealing the ease with which they acquire resistance. For the treatment of cholera, fluid replacement is paramount, but antibiotics (especially tetracycline) play an important public health role in limiting the spread of epidemics. Salmonella serotype Typhi, like shigella, is adept at accumulating cassettes of resistance genes, producing strains that withstand first-line, second-line and now, third-line drugs. Until 1972, chloramphenicol was the treatment of choice for typhoid fever throughout much of the subcontinent of India. By 1992, two-thirds of reported cases were chloramphenicol-resistant, thereby necessitating treatment with expensive quinolones that are themselves losing effectiveness. Without proper treatment, typhoid is a serious and frequently relapsing disease that kills up to 10% of those infected. Food-borne infections Food- and water-borne pathogens generally cause diarrhoeal diseases. Six major bacterial groups (Salmonella, Campylobacter, E. coli, Yersinia, Clostridia and Listeria) are responsible for these infections. In severe cases, systemic forms of disease may develop. Due to the considerable potential for food and water for human consumption to be contaminated by animal and environmental bacteria, scientists have started to focus attention on this area. Although available scientific data is limited, food and foodborne diseases are considered by many to play a specific role in antimicrobial resistance in humans. When considering antimicrobial resistance in this context, a number of elements should be taken into account, of which a few are mentioned below. If food- or water-borne bacteria cause disease in humans (e.g. Salmonella and Campylobacter), they may directly cause human illnesses, independent of whether they are resistant or susceptible to antibiotics. These food- and water-borne illnesses will, in most cases, result in diarrhoeal diseases. The majority of these diseases are selflimiting, do not require antibiotic treatment and are most appropriately treated by symptomatic treatment. If the illness is caused by a resistant bacteria and does require an antibiotic treatment, the treatment may be prolonged or recourse may have to be taken to another, potentially more expensive, antibiotic. In cases where a bacteria is resistant to all available antibiotics, the infection may become untreatable by antibiotics and eventually a patient may die due to the consequences of a noncontrollable infection. 58 OIE International Standards on Antimicrobial Resistance, 2003 1. General aspects If food- or water-borne bacteria are non-disease causing in humans (enterococci), they may indirectly lead to human illness in those specific cases where the animal or environmental bacteria has become resistant to antibiotics and where the potential exists for a transfer of the resistance genes of these bacteria to human pathogenic bacteria. As a consequence, a completely different human illness, which may not be food- or water-related, may become more difficult or impossible to treat. The evaluation of the impact of the potential transfer of resistance genes from nonpathogenic animal or environmental bacteria to pathogenic human bacteria is a much more complex and difficult undertaking, which currently still resides in the domain of research. Molecular and epidemiological methods are required to demonstrate the identical composition of the resistant gene in both the animal/environment and the human pathogenic bacteria and to trace the transfer of genes from the animal/environment to the human bacterial populations or vice versa. The tracing of the direction of transfer is particularly difficult in those cases where the incriminated antibiotic has been used both in humans and animals or plants. To evaluate the impact and the importance for human health of the non-human use of antimicrobials in animals and plants, data should be systematically collected on the contamination of food and water with resistant bacteria, food-borne infections, the percentage of infections due to resistant bacteria and the clinical outcome of these resistant infections. Food-borne disease surveillance and resistance Although food-borne disease surveillance was launched twenty years ago in some countries (WHO Surveillance Programme for Control of Food-borne Infections and Intoxications in Europe), food-borne disease surveillance appears to be lacking in many countries around the world and requires significant improvement. Where this kind of surveillance does exist, systematic, official collection of information on antimicrobial resistant bacteria in food and water and on human infections due to antimicrobial resistant animal or environmental bacteria appears to be scarce. Some references to food- and water-borne disease reporting are given below, which may to some extent illustrate the complexity in the evaluation of the role of antimicrobial resistance in food-borne disease, and the role of food- and watertransferred resistance of animal or environmental origin in the human resistance problem. The 7th report of the WHO Surveillance Programme for Control of Food-borne Infections and Intoxications in Europe states that ‘the variety and extent of foodborne diseases are such that no country is able to provide accurate data on their incidence and prevalence and surveillance programmes, where they exist, mostly collect information on only a low number of incidences. It is therefore not possible to give an estimate of the real magnitude of the problem. In some cases, the aetiology is multifactorial in nature and disease becomes manifest only after a long period of exposure. Consequently, many of the health problems resulting from food contaminants do not figure in statistics on food-borne diseases.’ OIE International Standards on Antimicrobial Resistance, 2003 59 1. General aspects Indicating that there is direct evidence that antimicrobial use in animals selects for antimicrobial resistant non-typhoid Salmonella serotypes (referencing resistant S. Typhimurium DT 104), and for fluoroquinolone resistant Campylobacter jejuni isolated from humans, poultry and poultry meat, the report indicates however that there ‘is limited information on the prevalence and spread of resistance in zoonotic bacteria. Monitoring programmes in some countries are in the early stage of development, some of these are in parallel with the strengthening of resistance monitoring in hospitals and community settings. Monitoring of antimicrobial resistance of bacteria from food animals and food of animal origin – whether national or international – is still in its infancy.’ With regard to food-borne illness in the United States of America (USA), summarised quantitative data is readily available and data for 1997 are given below. In the USA, food-borne diseases are estimated to cause 76 million illnesses, 325,000 hospitalisations and 1,800 deaths. Among all illnesses attributable to food-borne transmissions, 30% are caused by bacteria, 3% by parasites and 67% by viruses. Six pathogens account for over 90% of the estimated food-related deaths; Salmonella (31%), Listeria (28%), Toxoplasma (21%), Norwalk-like viruses (7%), Campylobacter (5%) and E. coli O157:H7 (3%). In 1997, active surveillance by US FoodNet reported 8,576 laboratory-confirmed cases of food-borne illnesses, of which 3,974 were identified as campylobacteriosis, 2,205 as salmonellosis, 1,273 as shigellosis, 468 as cryptosporidiosis, 340 as E. coli O157:H7, 139 as yersinellosis, 77 as listeriosis, 51 as Vibrio infections and 49 as cyclosporiasis. Overall, 1,270 (15%) of 8,576 patients with laboratory-confirmed infections were hospitalised; the proportion of cases in which people were hospitalised was highest for listeriosis (88%), followed by E. coli O157:H7 infections (29%), and salmonellosis (21%). Thirty-six patients with laboratory-confirmed infections died: fifteen with Listeria, thirteen with Salmonella, four with E. coli O157:H7, two with Cryptosporidium, one with Campylobacter, and one with Shigella. In 1997, the catchment area included 16.1 million people, 6.0% of the population of the USA. Unfortunately, no information is currently available in these publications on the percentage of infections due to resistant micro-organisms. Scientific opinion Current knowledge Antimicrobial resistance is a natural phenomenon. It is the natural response of a bacterium to defend itself against the effects of an antibiotic. The development of antimicrobial resistance is an ecological phenomenon. Any antibiotic use, whether in humans, animals or plants/environment may lead to resistance. In principle, the same molecules and classes of antimicrobials are used in humans, animals and plants. Humans, animals and the environment represent a reservoir in which resistance can develop. As most bacteria can, at least transitionally, contaminate or colonise all 60 OIE International Standards on Antimicrobial Resistance, 2003 1. General aspects possible hosts (humans, animals, plants, environment), there is an exchange between the different hosts and between the hosts and the environment. There is a problem of antimicrobial resistance in human medicine, increasingly perceived around the world as a threat to public health. The major problems have been described; these generally relate to parasites, viruses and human pathogenic bacterial infections and the use of antimicrobials in human medicine. One of the six major human diseases, diarrhoea, is in part related to zoonotic bacteria. The WHO indicates that cholera, typhoid, shigella and rotavirus infections, coupled with undernourishment, poor sanitation, inadequate hygiene and unsafe drinking water, are the principal causes of the heavy diarrhoeal disease burden in developing countries. In the USA, three pathogens, Salmonella, Listeria and Toxoplasma, are considered responsible for more than 75% of deaths caused by known pathogens. Of these three, Salmonella is a zoonotic bacterial pathogen. Concern has increasingly been expressed that, additionally to the resistance existing and emerging in human medicine, the use of antimicrobials in animals and plants will lead to resistance, thereby adding to the existing resistance burden in humans. Two potential mechanisms of resistance transfer from animals or plants/environment to humans are currently under consideration, as follows: a) the transfer of pathogenic bacteria b) the transfer of non-pathogenic bacteria or the transfer of their genes encoding resistance. Infection with resistant zoonotic bacteria may directly lead to human illness. However, the contribution of these bacteria to the overall resistance burden in humans should be carefully evaluated. In view of the very limited existing resistance data in this area, this evaluation might prove to be difficult. Regarding the transfer of non-pathogenic bacteria, the great fear is that resistance mechanisms encoded in mobile, transferable genetic material may be transferred to already multi-drug resistant human pathogenic bacteria, causing a life-threatening infection, which could be impossible to treat. As a less dramatic scenario, it is thought that the transfer of resistance genes could add to multiple resistance in human pathogens. As the treatment of multiple resistant infections is more difficult and more expensive, such infections would result in an increase in public health costs. Studies have been published on resistance gene transfer between bacteria. The impact of the potential transfer of resistance genes is a current area of research. Although a very small number of countries performs surveillance of resistance in enterococci (these commensal bacteria are considered as the appropriate indicator bacteria, as they easily develop or pick up transferable resistance), there is no scientific consensus on how surveillance findings should be interpreted. OIE International Standards on Antimicrobial Resistance, 2003 61 1. General aspects Missing information Where official collection of human infectious disease information exists, information on percentages of resistant infections and clinical outcome data is limited. Food-borne disease surveillance, although present in a number of countries, varies considerably between countries and requires intensification and harmonisation. In other countries, food-borne disease surveillance does not exist. As for zoonotic pathogens, official surveillance of antimicrobial resistance in animal bacteria and food has started only recently in a very small number of countries and in a limited number of bacteria. Recognising the important efforts made by a number of countries, supported by international organisations such as the WHO and the OIE, additional effort must be made by countries world-wide for the collection of the appropriate data. Countries should attempt to establish their priorities in view of the identified public health problems, the inherent costs and the available resources. Continued research to increase knowledge of antimicrobial resistance is vital, as is the integration of scientific knowledge in the decision-making processes to the greatest extent possible. Future research All further research in the subject will be valuable and will add to the knowledge and our understanding of the emergence and the appropriate measures for the containment of antimicrobial resistance. A number of subject areas which urgently require further investigation are presented below: a) The role of different modes of use of antibiotics in animals and humans. This is critical in the emergence and increased number of resistant strains among pathogens, commensals and environmental bacteria. A number of issues should be considered, namely: dose and duration of treatment, route of administration, pharmacokinetic, pharmacodynamic, number of patients/animals treated, stability in the environment (sewage, litter, land and animal housing), bacterial species and animal species. b) The innumerable pathways which enable bacteria and genes to spread between animals and humans. From the living animals to the contaminated food on the table of the consumer, there are many opportunities for contamination and transmission. This calls into question the idea that contamination necessarily originates from the animal reservoir and calls for very careful studies to clarify this question. Such studies are particularly difficult to design and conduct in cases where resistance genes, rather than the bacterial strains, are to be tracked down. c) The colonisation of humans by animal or environmental bacteria. Gram-positive bacteria seem to spread differently to Gram-negative bacteria. The factors in the host specificity of bacteria are to a great extent unknown. 62 OIE International Standards on Antimicrobial Resistance, 2003 1. General aspects d) The factors to be taken into account when trying to measure human health problems through a risk analysis study, specifically when it is applied to resistance traits (transfer of resistance genes). e) The information needed to assess and follow the resistance problems. This depends on surveillance systems and records of the antibiotic consumption in humans and animals. Most countries have not yet established any surveillance system. Moreover, the methods of comparison and interpretation of the results are not yet operational. f) The factors which may explain an increase or a decrease in the prevalence of resistant bacteria. These were recently investigated to determine the strength of their link to the use of a particular antibiotic. Multiple resistance poses a difficult question, since antibiotics other than those immediately concerned can be responsible for coand cross-selection. Many other questions are raised in accordance with the concern for decreasing or containing antibacterial resistance. Environmental ecosystems, animals and humans are very intimately linked through a bacterial circulation that we are only recently beginning to understand. Specific studies, basic and applied, coupled with bacterial surveillance systems and more responsible use of antibiotics should generate feedback and deliver clues for new understanding. It must be understood that not all antibiotics behave in the same way, even those belonging to the same class. Such scientific knowledge will be essential in making public health and political decisions and in adapting and updating guidelines for a better protection of the consumer and the global community. Actions to be taken Immediate actions The OIE Ad hoc Group of experts invites countries to inform themselves about the problem of antimicrobial resistance. As an immediate action, the OIE experts urge countries to implement the prudent and responsible use of antimicrobial agents in veterinary medicine, as laid down in their recommendations which are included in Antimicrobial resistance: responsible and prudent use of antimicrobial agents in veterinary medicine, later in this volume. Concurrent with the implementation of the prudent use of antimicrobials, countries are invited to establish the surveillance of importation, distribution and use of antimicrobials in animals. Countries should undertake all necessary efforts to impede the importation, distribution and use of counterfeit antimicrobial products. The recommendations of the OIE Ad hoc Group of experts are included in Antimicrobial resistance: monitoring the quantities of antimicrobials used in animal husbandry, later in this volume. Furthermore, countries should attempt, as a preliminary evaluation, to obtain an overview of the most important public health and antimicrobial resistance problems in OIE International Standards on Antimicrobial Resistance, 2003 63 1. General aspects their country. To this end, communication between the animal production and the medical field should be established. Countries should prioritise further medium-term actions, including time frames for their implementation, according to the most important problems identified. Requirements for the future The OIE Ad hoc Group encourages the OIE Member Countries to take ownership of the new methodologies and to make use of them in order to establish an objective, science-based view on the resistance situation in their countries. The OIE urges countries to carry out a risk analysis process when establishing sanitary measures relative to antimicrobial resistance. The respective information and recommendations of the OIE Ad hoc Group of experts are included in Antimicrobial resistance: risk analysis methodology for the potential impact on public health of antimicrobial resistant bacteria of animal origin, later in this volume. OIE Member Countries should assure the use of standardised laboratory methods for the detection and identification of antimicrobial resistance. To generate reliable resistance data, microbiological laboratories should implement quality assurance schemes and participate in external proficiency testing programmes. Proficiency testing programmes would preferably be conducted on a regional or sub-regional level. The recommendations of the OIE Ad hoc Group of experts are included in Antimicrobial resistance: standardisation and harmonisation of laboratory methodologies for the detection and quantification of antimicrobial resistance, later in this volume. After a prioritisation of the most important public health and antimicrobial resistance problems, countries should establish antimicrobial resistance surveillance programmes in human pathogenic bacteria, where necessary, and in animal bacteria and food, as appropriate. The recommendations of the OIE Ad hoc Group on how to address the matter are included in Antimicrobial resistance: harmonisation of national antimicrobial resistance monitoring and surveillance programmes in animals and in animal-derived food, later in this volume. Considering the importance of the issue and to foster consistency in decisions taken, the OIE will continue to co-ordinate its work with other international organisations, such as the FAO and WHO, and will also continue to be available for co-operation with other international or regional organisations, as appropriate. Antibiorésistance : une synthèse J. Acar & B. Röstel Résumé L’augmentation de l’antibiorésistance des bactéries pathogènes pour l’homme et la propagation de cette résistance, de l’environnement confiné des hôpitaux à la collectivité apparaissent, de plus en plus, comme une menace pour la santé publique. Tout traitement antimicrobien appliqué aux humains, aux animaux, aux végétaux ou dans les techniques de transformation des aliments peut entraîner une 64 OIE International Standards on Antimicrobial Resistance, 2003 1. General aspects résistance bactérienne. L’utilisation d’antibiotiques dans les élevages pourrait être en grande partie responsable de ce phénomène dans les espèces bactériennes communes à l’homme et aux animaux. Il est important d’explorer les conditions d’utilisation spécifiques qui président à la sélection et à la dissémination des bactéries résistantes. Le commerce et les échanges internationaux d’animaux et de denrées alimentaires élargissent au monde les risques de la résistance bactérienne. Certains pays envisagent des restrictions à l’importation pour les produits jugés à risque pour la santé publique. L’Organisation mondiale pour la santé animale, en tant qu’organisme de référence reconnu par l’Organisation mondiale du commerce aux termes de l’Accord sur l’application des mesures sanitaires et phytosanitaires, élabore des normes internationales sur l’antibiorésistance qui devront s’appuyer sur l’analyse du risque, au même titre que les mesures nationales. Les auteurs examinent les aspects scientifiques de la résistance ainsi que les problèmes qu’elle pose à la médecine humaine. Ils font également le point sur l’état actuel des connaissances, sur les lacunes existantes et sur les mesures qu’il conviendrait de prendre. Mots-clés Accord sur l’application des mesures sanitaires et phytosanitaires – Analyse du risque – Antibiorésistance – Denrées alimentaires – Maîtrise de la résistance – Mécanismes de la résistance – Mesures nationales – Normes internationales – Organisation mondiale pour la santé aniamale – Santé publique. Resistencia a los antimicrobianos: síntesis J. Acar & B. Röstel Resumen El aumento de la resistencia a los antimicrobianos en bacterias que causan importantes afecciones humanas y la salida de esas resistencias del reducto hospitalario al entorno abierto engendran una creciente sensación de amenaza para la salud pública. Cualquier producto antimicrobiano que se utilice en personas, animales, plantas o procesos de transformación de alimentos puede dar lugar a resistencias bacterianas. Hay motivos para pensar que el uso de antimicrobianos en la producción pecuaria contribuye sensiblemente al fenómeno entre especies bacterianas comunes al hombre y a los animales. Será importante investigar acerca de las condiciones específicas de uso que intervienen en la selección y la diseminación de bacterias resistentes. El comercio y movimiento internacional de animales y productos alimentarios confieren una dimensión planetaria a los riesgos de resistencia bacteriana, y no pocos países están estudiando restricciones a la importación de productos considerados peligrosos para la salud pública. La Organización mundial de sanidad animal, en cuanto organismo de referencia reconocido por la Organización Mundial del Comercio para el Acuerdo sobre la aplicación de medidas sanitarias y fitosanitarias, elabora normas internacionales sobre la resistencia a los productos antimicrobianos, cuya aplicación, como ocurre con las medidas de ámbito nacional, debe basarse en el análisis de riesgos. Los autores pasan revista a los antecedentes y problemas científicos de las resistencias en el ámbito de la medicina humana y determinan el estado actual de los conocimientos, las lagunas existentes y las acciones que convendría emprender. OIE International Standards on Antimicrobial Resistance, 2003 65 1. General aspects Palabras clave Acuerdo sobre la aplicación de medidas sanitarias y fitosanitarias – Alimentos – Análisis de riesgos – Contención de las resistencias – Mecanismos de la resistencia – Medidas nacionales – Normas internacionales – Organización mundial de sanidad animal – Resistencia a los productos antimicrobianos – Salud pública. Appendix A World Organisation for Animal Health Ad hoc Group of experts on antimicrobial resistance Members Jacques Acar (Chair), Emeritus Professor of Microbiology, Université Pierre et Marie Curie, Paris, France. E-mail: [email protected] Sharon Thompson (Vice-Chair), Joint Institute for Food Safety Research, Department for Health and Human Services Liaison, 1400 Independence Avenue, SW, Mail Stop 2256, Washington, DC 20250-2256, United States of America. E-mail: [email protected] Francis Anthony, Topic Leader Guideline No. 2, Fresh Acre Veterinary Surgery, Flaggoners Green, Bromyard, Herefordshire, HR7 4QR, United Kingdom. E-mail: [email protected] Anders Franklin, Topic Leader Guideline No. 5, Department of Antibiotics, SVA, SE 751 89 Uppsala, Sweden. E-mail: [email protected] David Vose, Topic Leader Guideline No. 1, David Vose Consulting, Le Bourg, 24400 Les Lèches, France. E-mail: [email protected] †Terry Nicholls, Topic Leader Guideline No. 3, Animal Health Science and Emergency Management Branch, National Offices of Animal and Plant Health and Food Safety, Department of Agriculture, Fisheries and Forestry, P.O. Box 858, Canberra ACT 2601, Australia. R. Gupta, College of Veterinary Sciences, Veterinary Bacteriology, Department of Microbiology, G.B. Pant University of Agriculture and Technology, Pantnagar 263 145 Uttar Pradesh, India. Yutaka Tamura, National Veterinary Assay Laboratory, Ministry of Agriculture, Forestry and Fisheries, 1-51-1 Tokura, Kokubunji, Tokyo 185-8511, Japan. E-mail: [email protected] E. John Threlfall, Laboratory of Enteric Pathogens, PHLS Central Public Health Laboratory, 61 Collindale Avenue, London NW9 5HT, United Kingdom. E-mail: [email protected] 66 OIE International Standards on Antimicrobial Resistance, 2003 1. General aspects Moritz van Vuuren, Department of Veterinary Tropical Diseases, Faculty of Veterinary Science, University of Pretoria, Private Bag X04, Onderstepoort 0110, South Africa. E-mail: [email protected] David G. White, Topic Leader Guideline No. 4, Centre for Veterinary Medicine, Food and Drug Administration, Office of Research, HFV-530, 8401 Muirkirk Rd, Laurel, MD 20708, United States of America. E-mail: [email protected] Observers Maria Lourdes Costarrica, FAO, Food Quality and Standards Service, Via delle Terme di Caracalla, 00100 Rome, Italy. E-mail: [email protected] H.C. Wegener, WHO, Division of Emerging and Transmissible Diseases, Animal and Food-related Public Health Risks, 20 avenue Appia, 1211 Geneva, Switzerland. OIE Barbara Röstel, OIE Collaborating Centre for Veterinary Medicinal Products, ANMV-AFSSA Fougères, B.P. 90203, 35302 Fougères Cedex, France. E-mail: [email protected] Jacques Boisseau, Director of OIE Collaborating Centre for Veterinary Medicinal Products, ANMV-AFSSA Fougères, B.P. 90203, 35302 Fougères Cedex, France. E-mail: [email protected] Jim Pearson, Head, Scientific and Technical Department, World Organisation for Animal Health, 12 rue de Prony, 75017 Paris, France. E-mail: [email protected] Appendix B World Organisation for Animal Health Guidelines on antimicrobial resistance The following documents constitute the work and the recommendations of the OIE Ad hoc Group of experts on antimicrobial resistance. International experts with recognised expertise in the field composed this group. The group was set up to respect and assure a representation of the different regions of the world. It brought together internationally recognised scientific expertise in medical and veterinary medical sciences, microbiology, laboratory sciences and risk analysis. – Risk analysis methodology for the potential impact on public health of antimicrobial resistant bacteria of animal origin – Responsible and prudent use of antimicrobial agents in veterinary medicine – Monitoring the quantities of antimicrobials used in animal husbandry – Standardisation and harmonisation of laboratory methodologies for the detection and quantification of antimicrobial resistance – Harmonisation of national antimicrobial resistance monitoring and surveillance programmes in animals and in animal-derived food OIE International Standards on Antimicrobial Resistance, 2003 67 1. General aspects Appendix C Background literature Boisseau J. & Röstel B. (1999). – The role of international trade in animals, animal products and feed in the spread of transferable antibiotic resistance and possible methods for control of the spread of infectious agent resistance factors. In Comprehensive reports on technical items presented to the International Committee or to Regional Commissions. OIE (World organisation for animal health), Paris, 197234. Centers for Disease Control and Prevention (1998). – Incidence of foodborne illnessess – FoodNet, 1997. MMWR, 47 (37), 782-786. Mead P.S., Slutsker L., Dietz V., McCaig L.F., Bresee J.S., Shapiro C., Griffin P.M. & Tauxe R.V. (1999). – Food-related illness and death in the United States. Emerg. infect. Dis., 5 (6), 840-842. OIE (World organisation for animal health) (1998). – Report of the 18th Conference of the OIE Regional Commission for Europe, 22-25 September, Prague, 11-18, 53-54. OIE (World organisation for animal health) (1999). – Recommendations of the Regional Commissions. In Final report of the 67th General Session of the OIE International Committee, 17-21 May, Paris. OIE, Paris, 35. OIE (World organisation for animal health) (2001). – International animal health code: mammals, birds and bees, 10th Edition. OIE, Paris, 473 pp. OIE (World organisation for animal health) (2001). – Manual of standards for diagnostic tests and vaccines: Lists A and B diseases of mammals, birds and bees, 4th Edition, 2000. OIE, Paris, 957 pp. OIE (World organisation for animal health) (2001). – Report of the 14th Conference of the OIE Regional Commission for Africa, 23-26 January, Arusha, Tanzania. OIE, Paris, 38-40. OIE (World organisation for animal health) (2001). – Resolution XXV. Antimicrobial resistance. In Final report of the 69th General Session of the OIE International Committee, 27 May-1 June, Paris. OIE, Paris, 117-118. OIE (World organisation for animal health) (2001). – Standards Commission meeting, 31 January-2 February, 2001 report. In Final report of the 69th General Session of the OIE International Committee, 27 May-1 June, Paris, 50. __________ 68 OIE International Standards on Antimicrobial Resistance, 2003 1. General aspects Antimicrobial drug resistance from salmonellas in humans and food animals: the current situation in relation to foodborne zoonoses in the United Kingdom E.J. Threlfall Antibiotic Resistance/Molecular Epidemiology Laboratory, Division of Gastrointestinal Infections, Central Public Health Laboratory, 61 Colindale Avenue, London NW9 5HT, United Kingdom There are two distinct epidemiologies for non-typhoidal salmonellosis. In developing countries infections caused by salmonella organisms are characterised by a high incidence of septicaemia with a consequent high mortality. The strains involved are multiple drug-resistant, often with resistance to up to ten antimicrobials. Nosocomial infection is a common means of transmission and food animals do not seem to be an important reservoir of such strains. In contrast, in developed countries the normal presentation is gastro-enteritis and the principal reservoirs of infection for humans are food animals. In such strains resistance is for the most part acquired in the food animal host before transmission to humans through the food chain. In 1999 the incidence of multiple drug resistance (to four or more antimicrobials) in non-typhoidal salmonellas from humans has fallen in isolations of Salmonella enterica serotypes typhimurium, virchow and hadar in comparison to 1996. This fall has been most noticeable in S. typhimurium, where 59% of isolates were multiresistant compared to 81% in 1996. The main reason for this has been a 75% decline in isolations of multiple-resistant (MR) S. typhimurium definitive phage type DT104 since 1996. Nevertheless, MR S. typhimurium DT104 remains second to S. enteritidis phage type 4 as the most common strain in cases of human salmonellosis in England and Wales. Multiple resistance has also remained high in S. hadar, with 49% of isolates resistant to four drugs or more compared to 56% in 1996. Decreased susceptibility to ciprofloxacin has increased in incidence in S. enteritidis, S. virchow and S. hadar; in S. hadar 70% of isolates exhibited decreased susceptibility to this antimicrobial. Use of LightCycler technology has proved invaluable in tracing outbreaks of MR S. typhimurium DT104 with decreased susceptibility to ciprofloxacin through the food chain. The overall decline in resistance has been also evident in isolations from food animals and in 2000 less than 60% of isolations of S. typhimurium from food animals were resistant to chloramphenicol (a marker for MR DT104) compared to 80% in 1996. Although there has been an overall decline in resistance since 1996 outbreaks caused by multiple-resistant strains of S. typhimurium have continued to cause serious problems in 2000. Of particular note have been outbreaks of MR S. typhimurium DT104, S. typhimurium DT204b and S. typhimurium phage type U302. Although for the most part the OIE International Standards on Antimicrobial Resistance, 2003 69 1. General aspects vehicles of infection have not been food animal products, food animals have been implicated as the primary reservoirs of infection. It is hoped that Codes of Practice recently introduced by some pharmaceutical companies, governments, professional organisations and others for the use of specific antimicrobials in animal husbandry, coupled with international agreements on guidelines for resistance monitoring, will now result in a reduction in the incidence of resistance to antimicrobials, such as the fluoroquinolones in salmonella organisms, causing infections in humans. __________ 70 OIE International Standards on Antimicrobial Resistance, 2003 1. General aspects Resistance in salmonellae: the situation in developing countries S. Benredjeb & A. Hammami Laboratoire ‘Résistance aux antibiotiques’, Faculté de Médecine, Tunis, Tunisia The increasing rate of multidrug resistant (MDR) Salmonellae has become a serious public health problem. Salmonella serotype typhi is endemic in developing countries and strains resistant to chloramphenicol, ampicillin and trimethoprim have been reported since 1989 in the Indian subcontinent, Southeast Asia and Africa. In North Africa, the majority of isolates are susceptible to antibiotics of choice for typhoid fever, which include chloramphenicol, ampicillin, cotrimoxazole and fluoroquinolones. Concerning non typhoidal Salmonellae (NTS), an increasing rate of antimicrobial resistance was observed. The majority of human infections is caused by only few serovars. Currently, serotype enteritidis is the most prevalent in developed countries. In North Africa a change was observed from a greater prevalence of serotype wien over the 1980s to a greater prevalence of serotype enteritidis over the 1990s. The distribution of the other serotypes varies between countries but since 1989 S. mbandaka has gained some importance in the epidemiology of salmonellosis in Algeria and Tunisia. MDR was detected in several Salmonellae serotypes particularly serotypes wien, typhimurium, mbandaka with a pattern of resistance to most betalactams, aminoglycosides and cotrimoxazole. They have been reported as producing extended spectrum betalactamases and were responsible for protracted outbreaks of severe paediatric infections. In Tunisia, the susceptibilities to antibiotics of 151 strains of NTS isolated from clinical specimens in 1999-2000 showed high rates of resistance to tetracycline (62.2%), ampicillin (36.4%), cotrimoxazole (30.5%), cephalotin (30.5%), cefotaxime and ceftazidime (28.5%). Most of the MDR Salmonellae belonging to serotypes mbandaka and livingstone were related to nosocomial outbreaks. The increasingly high prevalence of MDR Salmonellae in developing countries in relation with complex socio-economic and behavioural factors contribute to the world-wide spread of resistance. __________ OIE International Standards on Antimicrobial Resistance, 2003 71 1. General aspects Resistant bacteria and their impact on therapy in veterinary medicine J.-L. Martel Head of Bacteriology and Antimicrobial Resistance Unit AFSSA Lyons 31, avenue Tony Garnier F.69364 Lyons Cedex 07, France Since the introduction of antimicrobial agents in medicine, the therapeutic efficacy of the currently available drugs has been increasingly compromised by the development of bacterial resistance in both human and veterinary medicine. As examples in the veterinary field, we will quote some results from the French monitoring network (named ‘RESABO’). Three key factors with regard to the emergence of antimicrobial resistance have to be taken into account: a) the occurrence of resistance genes (present in bacteria before the antibiotic era and cannot be avoided) b) the close contact between bacteria in a polymicrobial environment c) the selective pressure imposed by the use of antimicrobials. This latter aspect is the one which can effectively be influenced by all parties involved in medicine, and particularly by veterinarians who prescribe and use antimicrobial agents. But it is essential not to forget the second aspect and to establish an integrated hygiene management, including not only husbandry, but also food-processing, food storage and consumer handling. The two main objectives of veterinary medicine are to maintain healthy food animals and to prevent hazards to public health. The control and prevention of bacterial infections are achieved by either therapeutic, metaphylactic or prophylactic application of antimicrobials. Substances of mainly the same classes as used in human medicine are available for the treatment of food producing animals. According to the number of animals present and the type of production, these treatments may be individual and given by oral and parenteral route or, when large groups of animals have to be treated, are applied via water or feed. For these purposes, particular galenic forms such as ‘long acting products’, ‘premixes’, intramammary infusions are developed and imply specific pharmacokinetic data. In veterinary medicine, as in human, antibiotics are vital drugs to treat specific bacterial infections. They are crucial for insuring a safe food supply through healthier food producing animals. Thus, veterinarians must have access to a variety of antibiotics for selecting the most effective drug and for alternating compounds to keep resistance at low levels. The prudent use of antimicrobial agents requires the development of and respect for good veterinary practices. __________ 72 OIE International Standards on Antimicrobial Resistance, 2003 1. General aspects New resistance mechanisms – review of the diversity P. Nordmann Head Dept. Microbiology, Hospital Bicêtre, South Paris Medical School, University Paris XI, France Novel mechanisms of antibiotic resistance continue to be recognised as consequences of mutations in house-keeping structural or regulatory genes and of acquisition of foreign genetic elements. These resistance mechanisms, which may become major public health threats, have been reported in nosocomial and community-acquired bacterial isolates. Among Gram-positive nosocomial isolates, the molecular mechanisms involved in vancomycin-intermediate and vancomycin-resistant Staphylococcus aureus and in glycopeptide resistance in Enterococcus sp. and their relationships with peptidoglycan biosynthesis genes are not known with certainty. Among Gram-negative nosocomial isolates, the list of expanded spectrum β-lactamases is still growing. Novel clavulanic-acid inhibited extended-spectrum β-lactamases (Cla-ESBL) have been mostly reported in Enterobacteriaceae and in Pseudomonas aeruginosa, whose genes are, in some cases, part of expression structures named integrons. Plasmid-mediated Cla-ESBLs with an expanded spectrum to carbapenems have been identified in Klebsiella pneumoniae and in P. aeruginosa. Acquired metallo-enzymes (IMP and VIM series) with significant carbapenemase activity are increasingly recognised world-wide, especially in P. aeruginosa. Carbapenem resistance in Acinetobacter baumannii may be the result of expanded-spectrum oxacillinases. Upregulation of naturally-occurring efflux systems is now established as a source of acquired resistance to most antibiotic classes including aminoglycosides in P. aeruginosa and in Enterobacteriaceae. Among community acquired pathogens, mutations in target genes of fluoroquinolones have been reported for quinolone resistance in Streptococcus pneumoniae, S. pyogenes, S. mitis and S. oralis. In S. pneumoniae, cephalosporin-resistant and penicillin-susceptible strains have been reported. Plasmid-mediated cephalosporinases that are derivatives of chromosomally-encoded Ambler class C enzymes are reported increasingly in enterobacterial species including those involved in community-acquired infections. Finally, multiple-antibiotic resistance Salmonella sp., strains have been identified worldwide as a consequence of antibiotic resistance genes located in integrons. __________ OIE International Standards on Antimicrobial Resistance, 2003 73 1. General aspects Possibilities of characterising resistance genes for use as an epidemiological tool M.-H. Nicolas-Chanoine (1) & S. Granier (1, 2) (1) Service de Microbiologie-Hygiène, Hôpital Ambroise Paré (AP-HP), UFR Médicale Paris-Ile-de-France-Ouest, Boulogne-Billancourt, France (2) Laboratoire de Microbiologie Médicale, Fondation Hôpital Saint-Joseph, Paris, France Introduction Epidemiological markers are markers able to discriminate between epidemiologically unrelated isolates of a given species and able to recognise the close-relatedness of isolates derived from the same outbreak or chain of transmission (1). Properties of an epidemiological marker In fact, any marker can be used as an epidemiological marker if and only if it displays the four principal criteria required to type isolates, namely, typeability, stability, reproducibility and discriminatory power. The marker must be present in all isolates of the species, meaning a typeability of 100%. This marker must remain persistent, meaning a stability of 100%. The marker must also be found to be identical to itself following independent and separate analyses, meaning a reproducibility of 100%. The discriminatory power of a marker corresponds to the average probability that this marker will assign a different type to two unrelated strains randomly sampled in the microbial population of a given species. This power is calculated by using the formula of the Simpson index of diversity: 1S D = 1 – ⎯⎯⎯⎯⎯⎯ Σ nj (nj -1) ≥ 0.95 N (N – 1) j=1 where D is the index of discriminatory power, N the number of unrelated strains tested, S the number of different types and nj the number of strains belonging to the jth type, assuming that strains will be classified into mutually exclusive types. To be used as an epidemiological marker, the marker must display a D factor superior or equal to 0.95. Can resistance encoding genes be effectively used as epidemiological markers? Resistance genes are carried on plasmids or chromosomes. If we refer to the first two criteria, typeability and stability, plasmidic genes cannot be used as an epidemiological marker of isolates as all isolates of a given species do not possess resistance coding plasmids, and furthermore, plasmids can also be lost. 74 OIE International Standards on Antimicrobial Resistance, 2003 1. General aspects Concerning the chromosomal genes involved in resistance, there are those which code natural resistance, for example, β-lactamase genes. On the other hand, genes coding antibiotic targets can be the object of genetic events (mutations, insertions and deletions) resulting in acquired resistance. Moreover, a certain allelic diversity exists for all of these genes. The question is: ‘are all of these genetic events involved in resistance, together with allelic diversity, variable enough to reach an index of discriminatory power superior or equal to 0.95?’ Such a question was addressed by Sreevatsan et al. about Mycobacterium tuberculosis (2). Different genes involved in antibiotic resistance in a large number of isolates were analysed: katG for isoniazide resistance, rpoB for rifampin resistance, gyrA for quinolone resistance and pzaA for pyrazinamide resistance. As indicated in Table I, they found very few sites with silent variations; two for katG, two for rpoB, six for gyrA and zero for pzaA. Table I Variability of genes coding antibiotic target in Mycobacterium tuberculosis Gene Resistance katG rpoB gyrA pzaA INH Rifampin Quinolones Pyrazinamide Number of strains analysed Number of sites with silent variations 360 305 629 30 2 2 6 0 By combining the katG and gyrA sequences of a large number of strains, it was only possible to classify all the strains into three groups. Thus, resistance genes cannot be used as an epidemiological marker for M. tuberculosis. We have recently analysed, in my laboratory, the β-lactamase genes of Klebsiella oxytoca. It was previously demonstrated that the β-lactamase genes of this species are divided into two groups, blaOXY-1 and blaOXY-2, which have 87% sequence identity (3). We have demonstrated that the blaOXY gene is able to divide the Klebsiella oxytoca taxon into two genetic groups which are also recognizable by other markers, such as sequence signatures in the 16S rDNA and rpoB genes and characteristic bands in the profiles generated by the ERIC-1R PCR method (4). We sequenced the blaOXY gene of 15 unrelated isolates obtained from 9 centres from 1996 to 2000. Table II which represents the percentage of sequence identity of the blaOXY gene of the fifteen isolates, compared two by two, shows that we found only three pairs of isolates having a blaOXY gene with an identical sequence, whereas the genes of the other isolates were different; 87% representing the sequence identity OIE International Standards on Antimicrobial Resistance, 2003 75 1. General aspects between the oxy-1 and oxy-2 gene groups and 99% representing the sequence identity within each group. Table II blaOXY gene comparison of fifteen K. oxytoca clinical isolates SG SG SG 74 49 56 SG 176 SG 337 SG 344 SG 254 SL 781 SL 911 SG 62 SG 69 SG SG SG 77 81 9 99 99 87 99 99 87 87 99 87 99 99 87 99 99 87 87 99 87 99 99 9 87 99 99 87 87 99 87 99 99 99 99 87 99 99 87 87 99 87 99 99 99 (percentage) SG15 99 SG74S SG49 SG56 SG176 SG337 SG344 SG254 SL781 SL911 SG62 SG69 SG77 SG81 87 87 99 99 87 99 99 87 99 87 87 99 87 87 87 87 99 87 87 99 100 99 87 99 99 87 87 87 87 99 87 99 99 99 87 99 100 87 99 99 87 87 99 87 99 99 99 99 99 99 87 99 99 87 87 99 87 99 99 100 99 99 We calculated the index of diversity and found that it was equal to 0.97, suggesting that blaOXY genes can be used as a marker for isolate typing. By using the ERIC-1R PCR typing system, we confirmed (data not shown) that the isolates found to be different from each other by blaOXY gene sequence, were also found to be different by this method. However, we also found that the pairs of isolates with an identical sequence of blaOXY gene, were different according to the ERIC-1R profiles. Thus, the ERIC-1R PCR method is more discriminatory than blaOXY gene sequence for typing K. oxytoca isolates. The second study that we carried out on Klebsielle oxytoca concerned nine isolates obtained from one hospital over a 4-year period. We found that five out of the nine isolates had an identical blaOXY-2 gene sequence with 3 identical mutations leading, for one of them, to an amino-acid substitution in comparison with the reference blaOXY-2 gene (3). The presence of a single strain over a 4-year period suggested by the blaOXY gene sequence, was confirmed by the ERIC-IR PCR typing system as indicated in Figure 1 (an identical profile A for five isolates). 76 OIE International Standards on Antimicrobial Resistance, 2003 1. General aspects Interestingly, we note that the blaOXY gene of this strain has shown no mutation over a four-year period. Faced with these molecular data, we had to look at the clinical epidemiological data. The first patient concerned by the strain was hospitalised in the gastro-intestinal surgery in August 1996, the second in intensive care unit in March 1998, the third in vascular surgery in December 1998, the fourth consulted in gastrology in March 1999 and the fifth was in fact the fourth patient who was hospitalised in May 1999 in the gastro-intestinal surgery. Additional studies showed that the four patients had one point in common, namely repeated hospitalisations in this hospital with at least one hospital stay in the gastrointestinal surgery ward. Thus, the discovery of the same blaOXY gene sequence in five K. oxytoca isolates led us to carry out an epidemiological study whose results strongly suggested the presence of a K. oxytoca strain in the digestive surgery ward. A A B A A M A C D E Fig. 1 ERIC- 1R PCR of nine clinical isolates of Klebsiella oxytoca These displayed five different profiles (A B C D E), M corresponding to weight marker Conclusion The two examples presented here, strongly suggest that resistance chromosomal genes could be used as an epidemiological marker of isolates for certain species. To confirm OIE International Standards on Antimicrobial Resistance, 2003 77 1. General aspects this suggestion, the analysis of resistance gene sequences of each species must be carried out. References 1. Struelens M.J., Bauernfeind A., Van Belkum A., Blanc D., Cookson B.D., Dijkshoorn L., El Solh N., Etienne J., Garaizar J., Gerner-Smidt P., Legakis N., De Lencastre H., NicolasChanoine M.H., Pitt T.L., Römling U., Rosdahl V. & Witte W. (1996). – Consensus guidelines for appropriate use and evaluation of microbial epidemiologic typing systems. Clin. microbiol. Infect., 2, 2-11. 2. Sreevatsan S., Pan X., Stockbauer K.E., Connell N.D., Kreiswirth B.N., Whittam T.S. & Musser J.M. (1997). – Restricted structural gene polymorphism in the Mycobacterium tuberculosis complex indicates evolutionarily recent global dissemination. Proc. Natl. Acad. Sci. USA, 94, 9869-9874. 3. Fournier B., Roy P.H., Lagrange P.H. & Philippon A. (1996). – Chromosomal betalactamase genes of Klebsiella oxytoca are divided into two main groups, blaOXY-1 and blaOXY-2. Antimicrob. Agents Chemother., 40, 454-459. 4. Granier S.A., Plaisance L., Leflon-Guibout V., Lagier E., Morand S., Goldstein F.W. & Nicolas-Chanoine M.-H. (2003). – Recognition of two genetic groups in Klebsiella oxytoca taxon on the basis of the chromosomal beta-lactamase and housekeeping gene sequences as well as ERIC-1R PCR typing. Int. J. Syst. Evol. Microbiol., 53, 661-668. http://dx.doi.org./10.1099/ijs.0.02408-0 __________ 78 OIE International Standards on Antimicrobial Resistance, 2003 1. General aspects Perception of the veterinary practitioner with regard to the contribution of the use of antimicrobials in animal husbandry to the problems of human health associated with resistant bacteria O. Fortineau Groupement Technique Vétérinaire, 5 rue Moufle, 75011 Paris, France Cross-resistance to avoparcin and vancomycin has highlighted the link between animal health and public health. Whilst restrictions on the use of antibiotics as feed additives in the European Union aim to minimise the risks due to antibioresistance, new questions are now being raised regarding the therapeutic use of antibiotics. The phenomenon of antibioresistance is directly linked to the use of antibiotics. Some bacteria are currently developing resistance, which sometimes limits the actions of the veterinary surgeon. But the main question is linked to the transfer of such resistance acquired in animals to bacteria that may affect humans. Several routes are possible, but the main risk is linked to the ingestion by man of animal intestinal bacteria, such as Coli bacteria, Salmonella, Enterococci and Campylobacter. The strict observance of hygiene rules in force at every level of the food-production chain effectively limits the risk of transferring such bacteria to man. These hygiene rules must, however, be constantly observed, even by the final consumers, who too often forget to wash their hands before eating. However, veterinary surgeons are well aware of the risks linked to their prescription of antibiotics: following the initiatives of the Federation of Veterinarians of Europe, and those of many national organisations, the number of seminars and publications on antibiotic resistance is increasing and more best practice manuals on the use of antibiotics are being published. Veterinary surgeons are working at improving antibiotic usage in animal production: this is in line with public health, animal health and production cost objectives. __________ OIE International Standards on Antimicrobial Resistance, 2003 79 1. General aspects The pig producer’s position as herd manager following the cessation of the use of antibiotic growth promoters in Denmark O.G. Pedersen Copa Cogeca, 23-25 rue de la Science, 1040 Bruxelles, Belgique The National Committee for Pig Production, representing the European farmers on behalf of COPA COGECA The Danish background In Denmark, we produce approximately 23.5 million pigs, 134 million broilers, and we have a population of 1.9 million cattle and 3.7 million hens (over six months). The Danish farmers voluntarily decided to stop using antibiotics as growth promoters from 1 January 2000. In addition, we have a very restrictive system with registration of the use of all veterinary medication at herd level. The basis for the ban on antibiotic growth promoters in Denmark was a double-sided debate. This was partly political and partly due to affirmation by the medical scientists and the veterinary institutes of the risk of the development of resistance and thereby reduced possibility for treating patients in the hospitals. Chain of events in Denmark Antibiotic growth promoters (AGPs) have not been used in animal production in Denmark since 1 January 2000. Figure 1 illustrates the development in antibiotic consumption over the last seven years. The curves are affected by the following events: 1995: National ban on avoparcin 1998: National ban on virginiamycine. Voluntary agreement on stopping the use of antibiotic growth promoters for calves, broilers and for pigs weighing more than 35 kg (sows and finishers) Control and penalty systems were introduced National tax on AGPs 1999: EU ban: tylosine, bacitracine, spiramycine, virginiamycine, olaquindox and carbadox 2000: Voluntary agreement not to use AGPs for pig weighing less than 35 kg (weaners). At the same time control and penalty systems were introduced. 80 OIE International Standards on Antimicrobial Resistance, 2003 1. General aspects Active substance (tonnes) 250 200 150 100 50 0 1994 1996 1997 1998 1999 2000 Year Therapy/medication Growth promoters Total Fig. 1 The development of antibiotic consumption in Denmark Finishers, broilers and calves As mentioned, a voluntary agreement on stopping the use of antibiotic growth promoters for finishers, broilers and calves was made in 1998. ‘The National Committee for Pig Production collected information on the experiences of 62 Danish finisher herds in the period after they stopped using APGs. The majority (63%) of the herds did not experience problems in the form of reduced daily gain or increased frequency of treatments for diarrhoea when AGPs were removed from the feed. 26% of the herds experienced a temporary decrease in the daily gain, while 11% experienced permanent problems, probably as a consequence of the removal of AGPs from the feed. Thus, the removal of AGPs from finisher feed has been fairly unproblematic in the herds participating in this study. Several herds changed the composition of the feed in connection with the ban of antibiotic growth promoters, e.g. reduced crude protein content and increased barley/texture.’ The results from the study are confirmed by the results from the national productivity surveillance in Denmark (cf. Table I). Daily gain and mortality showed no significant changes in the 1998 statement compared with the previous year. It was, however, observed that the increase in daily gain (from 1997/1998 to 1998/1999) was not quite as high as in the previous years, and mortality was marginally higher. We can hereby conclude that antibiotic growth promoters have been removed from the finisher feed without any significant effects on productivity and health. The annual increase in daily gain in 2000/2001 returned to the same level as before antibiotic growth promoters were removed. OIE International Standards on Antimicrobial Resistance, 2003 81 1. General aspects The removal of antibiotic growth promoters from feed for calves on 1 January 1998, also took place without significant problems. It should be noted that the use of antibiotic growth promoters before the removal was very low in feed for calves. In broilers, the voluntary stop resulted in an increased feed consumption (from 1.78 kg feed/kg live broiler to approx. 1.82 units) and a lower gain (the average weight after 42 days dropped from 1,960 g to 1,930 g). Since then, the average weight has increased and is now more than 2,000 g. The moderately negative effect of removing antibiotic growth promoters from the feed for broilers is in particular due to the very high demands to hygiene because of the Salmonella programme the aim of which is zero prevalence. Table I National productivity surveillance – finishers In each column, approximately 1,400 herds are represented Daily gain (g) Mortality (%) April 1995April 1996 April 1996April 1997 April 1997April 1998 April 1998April 1999 April 1999April 2000 April 2000April 2001 744 3.0 762 (+18 g) 3.2 778 (+16 g) 3.2 786 (+8 g) 3.4 798 (+12 g) 3.6 817 (+19 g) 3.5 Weaners The great challenge was the removal of antibiotic growth promoters from weaner feed, which commenced on 1 January 2000, in the form of a voluntary agreement. This means that the three products still approved are not used in Denmark. A large proportion of the pigs (approx. 50%) had already been fed without AGPs from mid-1999 without severe problems. In Table II the results in the weaner period from the national productivity surveillance in Denmark are shown. The statement for 1999/2000, which was the first period without using AGPs for weaners, shows a decrease in daily gain (20 g) and a corresponding increase in the pigs’ age at 30 kg compared with the 1998/1999 statement. Post-weaning mortality also increased (0.7%-units). The latest statement (2000/2001) shows that the negative effect on the weaners’ productivity continues and that it is difficult to reach the level of productivity attained when antibiotic growth promoters were used. We therefore conclude that the removal of AGPs from weaner feed has had significantly negative consequences in the form of reduced gain and higher mortality. It has also exacerbated a number of fundamental problems in many of the herds, such as post-weaning diarrhoea and chronic infections (Lawsonia intracellularis). 82 OIE International Standards on Antimicrobial Resistance, 2003 1. General aspects Table II National average for production efficiency control – weaners Daily gain (g) Mortality (%) Age at 30 kg April 1995April 1996 April 1996April 1997 April 1997April 1998 April 1998April 1999 April 1999April 2000 April 2000April 2001 422 2.7 82.6 420 2.8 82.6 419 2.9 82.8 427 2.9 82.9 407 (-20 g) 3.6 (+ 0.7) 85.3 (+ 2.4) 411 3.5 85.5 Management factors In order to minimise the use of antibiotics, it is important to focus on housing conditions and management in the individual herd. Optimising production systems and management factors in the herds includes: – the use of health promoting housing systems with sectioned batch operation and all in-all out production – the avoidance of overstocking. The reduced growth without AGPs has resulted in an accumulation of weaners in many herds. Overcrowding, increased batch sizes and lack of hospital pens have probably contributed to the negative consequences after removal of AGPs – increased focus on the immediate environment of the pigs in the pen. It is recommended that the pigs are protected against draughts, low temperatures and humidity. Pens fitted with a covered area (two-climate pens) and use of bedding/additional heating are very suitable in that regard – increased focus on weaning weight. For instance, the avoidance of unnecessary transfer of pigs later than 48 hours after birth, since this will reduce the weight at weaning. Nutrition In terms of improving the gastro-intestinal health of weaners through the feeding practices, there is particular focus in Denmark on: – feeding strategy. Restricted feeding in the first 14 days post-weaning can improve the pigs’ health significantly compared with ad libitum feeding. Mortality may also be reduced. – use of additives. The content of copper (Cu) in Danish diets for weaners is close to the allowed maximum content of 175 mg/kg. High doses of zinc (Zn) (2,500 mg/kg) reduce the occurrence of diarrhoea, but the content of Zn in Danish diets is no more than 250 mg/kg. More than 100 commercial feed additives for weaners and finishers have been tested during the last five years in Denmark. For weaners, a group of organic acid products has shown on average to improve the production results at the same level as when AGPs were used. – use of protective diets. These are typically characterised by low protein content, with a high animal protein component (fish meal, whey powder, skimmed milk OIE International Standards on Antimicrobial Resistance, 2003 83 1. General aspects powder), high content of barley, and with the addition of an organic acid product. This type of diet often reduces the incidence of diarrhoea. Optimised production in problem herds The National Committee for Pig Production has performed a number of studies in herds with poor health and a high consumption of medication and low gain in the weaner period. The aim was to improve the health status and production results in the weaner unit by optimising the production conditions in general. In the test groups improvements were made in the following: management procedures, hygiene, pen design, composition of the feed, feeding strategy, etc. The results in the test groups were compared with those from herds whose production conditions had not been improved. The preliminary results show that the optimised system reduced the number of treatments for diarrhoea in most herds. However, it appears to be difficult to improve production results. The optimised system was also able to reduce mortality significantly in two of the herds compared with the control treatment (normal practice). However in two other herds, mortality was unacceptably high in both groups (control and optimised) due to aggressive E. coli and Lawsonia intracellularis infections. Conclusion Overall, we can conclude that the removal of antibiotic growth promoters from the feed in Denmark has only significantly increased mortality and reduced weight gain in weaner production and also increased feed consumption in broilers. Studies in weaner herds with problems show that optimisation of production conditions and feed significantly improves health. However, it is difficult to improve the production results of the weaners. __________ 84 OIE International Standards on Antimicrobial Resistance, 2003 1. General aspects Antimicrobial use in animal husbandry and its relationship to resistant bacteria in human health B. Andrews International Federation for Animal Health, Fort Lee, NJ 07024, United States of America Thank you for this opportunity to address the 2nd Annual OIE International Conference on Antimicrobial Resistance. Representing the International Federation for Animal Health (IFAH) and its member organisations, I am honored by your invitation to offer my comments on society’s perception of this important issue. What I hope to provide is a brief framework for what has shaped society’s, and more specifically the consumer’s perceptions, regarding the use of antibiotics in livestock production, and its relationship to resistant bacteria in humans. There was a time when consumer groups carefully assessed and conducted extensive tests to confirm their position. More recently, they seem to have become primarily lobbying organisations that spend much of their money on campaigns. So how have we moved from the US National Academy of Sciences’ (NAS) serious review of the resistance issue in 1956, to today’s anti-industry, anti-technology PR spin-machines which cleverly reduce one of the most heavily researched and complex scientific discussions of our time, into a handful of sound bytes aimed at scaring the public and ending modern livestock production? I am not referring to the many science-based and dedicated organisations involved in the serious debate over microbial resistance, many of whom are at this conference. I am confident we are equally committed to protecting human health, and it is through our continued collaboration that we will resolve these concerns. I am referring to the more militant consumerists, nonintensive farming and organic farming lobby, and the animal rights activists whose sole agenda is to end the use of animals for food production. When their attempts to gain attention through animal rights issues failed in the 1960s and 1970s, they seized upon the food safety issue and have found the key to the consumer’s conscience – the dinner table. One thing seems clear as we follow the course of events – less science and more supposition have increasingly become the foundation for government and regulatory decisions – the result of increased political pressure by these special interest groups. Respected scientific bodies throughout the world have been unable to substantiate any definite link between antimicrobial use in animals and treatment failure in humans. Yet, the highly politicised Swann Committee report was the foundation for the current ban on a number of products in the EU, and most recently the FDA has moved to withdraw its approval for fluoroquinolones in poultry. By the late 1970s, antagonists of animal agriculture had turned up the volume and enlisted the power of the popular press. What has resulted is a consuming public drawn to the issue through gloom and doom messages and emotive rhetoric regarding OIE International Standards on Antimicrobial Resistance, 2003 85 1. General aspects the future of our food supply. They have perpetuated a ‘conventional wisdom’ with the consumer, based upon misinformation, half-truths, and in some instances patently false messages. IFAH the AHI (Animal Health Industry) and numerous other producer and trade organisations have developed consumer education materials and invested in media relations in an attempt to balance the debate. But as we have all learned, good news is no news in today’s consumer press, and our efforts seldom see the front page. A review of recent literature provides a common thread of consumer perceptions that to date, the scientific community has been unable to thwart. The scientifically defensible facts relative to these myths will be addressed by many of the distinguished speakers at this conference. What I would like to provide is a general framework to establish the existing gap between society’s perception, and the science. Consumers are being told that livestock and poultry producers abuse antibiotics, using more of these products than are being used in human medicine. This statement begs several responses. First, the actual amount used should be put in context with the much larger population of animals relative to humans that may need treatment. Also the amount of antibiotics used in animal feed is minor. For instance, products are used at various inclusion levels from as low as 0.5ppm up to 500ppm per tonne of feed, dependent upon use for prevention or treatment. Second, and possibly more significant, is the fact that these products are highly regulated and used under the supervision of a veterinarian. Prudent use guidelines as well as producer Quality Assurance programmes have been established for proper storage, handling, administration and record-keeping, and these guidelines have been adopted by both national, regional and global animal health organisations. Consumers are being told that the sub-therapeutic use of antibiotics in animal agriculture is eroding physicians’ ability to treat infectious disease in humans, resulting in a related perception that banning the use of these products will fix the problem of treatment failure in humans. The fact is, the only irrefutable scientific evidence available demonstrates that bacterial resistance to antibiotics in humans has been brought about by the over-prescribing and misuse of these products in human medicine. A recent WHO paper on the use of antimicrobials outside human medicine states, ‘there is no doubt that most of the rising antimicrobial resistance problem in human medicine is due to the overuse and misuse of antimicrobials by doctors and other health personnel.’ Improved doctor and patient education on the proper use of antibiotics will be the cornerstone for combating the resistance problems that are emerging today. Another fact that seems to escape the attention of the industry detractors is that the vast majority of antibiotics currently administered through the feed or in the water have little or no significance in human medicine, making the development of resistance to these products largely irrelevant to the treatment of disease in humans. 86 OIE International Standards on Antimicrobial Resistance, 2003 1. General aspects Consumers are being told that antibiotics are routinely added to the feed and water of healthy animals for non-therapeutic purposes such as growth promotion. Non-therapeutic is a term coined by the Union of Concerned Scientists in its recent ‘hogging it’ report. We can only assume it is intended to create disdain if not fear in the consumer, by implying these products serve no purpose other than to sustain the greed of the producer. I would suggest that even the use of the term sub-therapeutic is unfortunate, since low doses clearly have a beneficial effect in the control of subclinical disease. Livestock and poultry producers throughout the world have learned the necessity of maintaining an animal’s health and robust growth for both humanitarian and economic reasons. Only healthy animals can produce healthy meat, milk and eggs. And maintaining this superior health status is, currently at least, our only means of insuring we continue to meet the world’s growing demand for a safe, abundant and affordable supply of animal protein. This requires our unceasing efforts to insure the availability of existing, effective products, while continuing our research into new antimicrobial agents for animal treatment. In 2000, the world produced 283.7 million head of beef cattle and veal calves: 44.7 billion chickens and turkeys and 1.2 billion pigs. Without the prudent use of antibiotics, we would need millions more of these food producing animals to compensate for losses to disease. The cost of animal protein would skyrocket, not to mention the impact on our environment. These seem to be the foundation for the myths that abound in society regarding antibiotics and their use in animal agriculture. Words are changed and put in different contexts, but the message is the same – using antibiotics in animal production is bad. However, in reality, how wide-spread are these perceptions and what level of significance do they hold for the consumer? As an industry, have we succumbed to the gloom and doom as well, believing that the vast majority of society shares these concerns? In Europe, consumers have become afraid of their food for reasons other than antibiotic resistance, such as dioxin contamination and BSE. Research from the International Food Information Council demonstrates that when asked what consumers were most concerned about when it comes to food safety, antibiotics were not mentioned. They are more concerned about package security, how food is being handled and prepared, diseases and pathogen contamination, chemicals and pesticides and altered or engineered foods. This suggests we may be a little too close to the issue to objectively assess what impact anyone’s message will have on society, whether or not it is based upon sound science or sensationalised sound bytes. There is good news to report. In 1996 the National Information Program on Antibiotics, a Canadian coalition of patient, physician and pharmacist organisations, began running advertisements in medical journals, handing out posters and literature to convince doctors to be more prudent in prescribing antibiotics. Since then, Canadian doctors have indeed been writing fewer prescriptions for them and the OIE International Standards on Antimicrobial Resistance, 2003 87 1. General aspects results have been impressive. Two of the so-called drug-resistant superbugs are on the retreat in Canadian communities. The number of Streptococcus pneumoniae resistant to penicillin reached 14% in 1998. Today, for the first time in a decade, it has dropped to 10%. And 40% of Haemophilus influenzae were strains resistant to amoxicillin in 1996. That figure has since dropped to about 25%. At the heart of this issue lies the undeniable truth that food has a profound meaning to people everywhere in the world. It carries significant religious meaning in many countries. Society’s perception of how their food is being produced will never be shaped solely by the scientific assessments we present, but by a careful balancing of messages, both emotional and measurable. As a professional observer of human nature, George Bernard Shaw once noted, ‘there is no love sincerer than the love of food.’ Mr Chairman, distinguished guests, IFAH and its member organisations are committed to continuing the discussion and review of this critical issue, and feel confident a solution lies – not in emotive rhetoric and fear-mongering in the popular media – but through collaborative efforts towards science-based understanding and risk management. __________ 88 OIE International Standards on Antimicrobial Resistance, 2003 1. General aspects Perception of society with regard to the contribution of the use of antimicrobials in animal husbandry to the problems of human health associated with resistant bacteria: the situation in developing countries S. Sirinavin Division of Infectious Disease, Department of Pediatrics, Faculty of Medicine Ramathibodi Hospital, Mahidol University, Bangkok, Thailand Antimicrobial resistance is a major health problem in developing countries. It is an important cause of increasing morbidity, mortality, and cost of treatment of infectious diseases in human beings. The most important mechanism is selection pressure from the use of antimicrobial agents. Humans, animals and microbes share the same world, and microbes do not discriminate between humans and animals. In addition, effective antimicrobial agents which are useful for treating infection in humans and animals are limited. High levels of environmental contamination due to antimicrobial use in animal husbandry is a very important cause of drug resistance, as well as over-thecounter use of the drugs in human beings. In Thailand, overuse of antimicrobial drugs, for both animals and humans, is the starting point. The concept that the use of antimicrobials can lead to resistant microbes is difficult for Thai people to appreciate, since the effects of drug usage are not directly visible. The drugs used appear to be not only useful but also harmless. Antimicrobial drugs can be obtained from drugstores without prescription, either for human or animal use. Antimicrobial drugs are widely used in animal husbandry without adequate guidance or control. Concern from regulatory and relevant authorities is still minimal, especially about the use of antimicrobial drugs in animals. The rapidly growing problem of antimicrobial resistance has been of special concern in academic institutes in Thailand for more than a decade without resulting in effective intervention. Recently the concern seems to have reached the national level. The National Antimicrobial Resistant Surveillance Center for monitoring resistance problems in human beings and the Center for Antimicrobial Resistance Monitoring in Food-Producing Animals were developed a few years ago. Since non-typhoidal salmonellosis is an important endemic disease in Thailand for both humans and animals, a National Committee on Controlling Non-Typhoidal Salmonellosis has been developed. This is an important activity for generating concern about antimicrobial resistance problems in animals and humans and for persuading people to work and to think together in order to reduce them. It is hoped that improved control of antimicrobial use in animal husbandry and in humans will follow, resulting in a reduction in resistance problems. __________ OIE International Standards on Antimicrobial Resistance, 2003 89 1. General aspects A consumer perspective: to what extent does antimicrobial use in animal husbandry contribute to resistance associated human health problems? L.Y. Lefferts Consumers International, 526 Mountain Field Trail, Nellysford, VA 22958, United States of America Most consumers recognise that antimicrobials are vitally important tools in human and animal medicine. They share the concerns of the public health community about the increase in bacteria resistant to antimicrobials. What is the basis for this concern? Infectious disease is the world’s biggest killer of children and young adults – more than 13 million deaths each year, according to the World Health Organization (WHO). Over the next hour alone, 1,500 people will die from an infectious disease – over half of them children under five. Infectious disease has major economic and human costs in both developing and developed countries. And without antimicrobials, many infectious diseases would be untreatable. Multi-resistance in bacteria is widespread and a major problem for the treatment of bacterial diseases. Infections that were once easily cured by antimicrobials are more difficult to treat, and sometimes require lengthy hospitalisation, or cause death. For example, in one retrospective study of 52 Salmonella outbreaks in the USA, it was found that the case mortality rate was higher for patients infected with resistant Salmonella (4.2%) than for those with non-resistant infections (0.2%) (5). Antimicrobial resistance kills almost two people in the US every hour, according to the Alliance for Prudent Use of Antibiotics, a non-profit educational, research, and networking organisation headquartered in the US with chapters in over 25 countries. Few new antimicrobials are being developed, and resistance is showing up in some of the new compounds already. The emergence of resistance in five patients to Linezolid, the first structurally different antibiotic introduced in almost three decades, was recently reported (3). On average, research and development of anti-infective drugs takes ten to twenty years. Consumers believe that antimicrobial use in animal husbandry does contribute to the resistance problem to an extent that could and should be better addressed in many countries. What is the basis for this belief? Resistance to antimicrobials naturally evolves but is greatly amplified by overuse and misuse of antimicrobials. While uses in humans probably are more important than uses in animals, and there is no consensus on how much animal husbandry contributes, it is clear that antimicrobial use in food-producing animals contributes significantly to the problem. A recent study by the Union of Concerned Scientists estimated that 84% of antimicrobial use in the USA is in animals (6). The study challenged the conventional 90 OIE International Standards on Antimicrobial Resistance, 2003 1. General aspects wisdom that use in animals is closer to about 40% of total antimicrobial use in the USA. It also estimated that non-therapeutic livestock use accounts for 70% of total antimicrobial use in the USA (including drugs important in human medicine such as tetracycline, penicillin and erythromycin). A number of resistant pathogens have been found in food animals and foods. For example, scientists at the Swiss Federal Research Institute in Zurich found bacteria in salami resistant to five common antibiotics: chloramphenicol, erythromicin, streptomycin, streptothricin, and kanamycin. A preliminary survey of factory-packaged beef and poultry conducted by the US Food and Drug Administration (FDA) in local supermarkets also found antibiotic-resistant bacteria. Enterococci were found in 67% of chicken samples, 34% of the turkey samples and 66% of the beef samples, and tested for resistance to 29 different types of antibiotics, including six commonly used in animal feed. Strains of enterococci taken from either chicken or turkey were resistant to more drugs than those taken from beef. There have also been reports of antibiotic-resistant bacteria in unpasteurised milk, fresh and frozen seafood, cheese and parsley (9). Resistant pathogens can be transferred indirectly to consumers through the environment or through workers. For example, a child in Nebraska became infected by Rocephin-resistant Salmonella bacteria that had originated from cattle on his farm (2). Animal wastes can contain antimicrobials and antimicrobial-resistant bacteria. As much as 75% of an antimicrobial may pass undigested through the animal into waste. That waste may be spread onto crops, or be stored in lagoons, often unlined lagoons, where it can leach into groundwater or get into surface waters (1, 8). Farm workers may become infected with resistant germs and pass them on to family members or others in their community. Evidence linking use of antimicrobials in animals with illness in humans is mounting. For example: an outbreak of antibiotic-resistant Salmonella in humans was linked to beef cattle that had been fed chlorotetracycline for non-therapeutic purposes (growth promotion) (4); and a multi drug-resistant Salmonella typhimurium outbreak in Denmark affecting 25 and killing 2 was traced to meat from infected pigs (7). The emergence of vancomycin-resistant Enterococcus faecium (VRE) in food can be traced to the widespread use of avoparcin (the animal equivalent of the human antibiotic vancomycin) in livestock (10). Moreover, with livestock production increasing, both in developed and developing countries – particularly intensive (‘factory farm’) production – reliance on antimicrobials is likewise expanding – often without adequate guidelines or requirements for prescriptions. With the trends toward globalisation and the relaxing of trade barriers, inadequate standards and enforcement in one nation means all consumers are vulnerable. What do consumers think should be done, in the context of antimicrobials used in food-producing animals? OIE International Standards on Antimicrobial Resistance, 2003 91 1. General aspects Given the relationship between use of antimicrobials and the development of resistance, consumers support the recommendations from the World Health Organization concerning the phase-out of non-therapeutic use of antimicrobials that are or may become important in human medicine (and structurally related chemicals). They believe that antimicrobials that are or may become important in human medicine (and structurally related chemicals, and those which select for cross-linked resistance) should never be used nontherapeutically. In some instances, consumers may favor restricting certain therapeutic uses in animals, such as fluoroquinolones, to protect the efficacy of vital human drugs, particularly where the drug is administered on a flock-wide or herd-wide basis rather than to individual sick animals. All antimicrobial usage in animals should be subject to veterinary prescription. Many consumers oppose the use of intensive animal production practices that rely heavily on antimicrobials, preferring improved hygiene, animal housing, feed, and other animal management practices that reduce the need to use antimicrobials. Consumers want labeling that indicates whether or not foods have been produced from animals raised using nontherapeutic antimicrobials. Improved data on the therapeutic and non-therapeutic use of antimicrobials in agriculture is needed, as well as improved surveillance of food-borne illness associated with antimicrobial-resistant bacteria and monitoring of foods for antimicrobialresistant bacteria. Consumers believe that the burden of proof of demonstrating that agricultural use of antimicrobials does not contribute to resistance in human therapeutic drugs should be borne by proponents of such use; unless that burden can be met, those uses should be promptly phased out, following a precautionary approach. For example, they support the WHO recommendation that antimicrobials used for growth promotion should be removed from the market in the absence of a risk assessment demonstrating their safety. Consumer organisations are working at numerous national and international fora on these critical public health issues: The Trans-Atlantic Consumer Dialogue (a forum of some 20 consumer groups in the USA and 45 in Europe, see www.tacd.org) in 1999 issued a number of recommendations to the governments of the USA and the EU, including the institution of a total ban on the non-medical use of antibiotics in animals and food. Just this year in April, they updated their recommendations to call for a ban on the use of fluoroquinolone antibiotics in poultry unless the drug is administered by injection. Consumers International, the world-wide federation of consumer organisations, representing more than 270 organisations in over 120 countries (www.consumersinternational.org), has actively participated in debates on this issue at committees of the Codex Alimentarius Commission. 92 OIE International Standards on Antimicrobial Resistance, 2003 1. General aspects The most recent World Congress of Consumers International held in Durban, South Africa issued a statement calling on governments and international institutions to, amongst other things, prohibit the use of antibiotics as feed additives. Consumer-nominated experts have been invited to attend WHO and FAO meetings and consultations on the issue of antimicrobial resistance. Bureau Européen des Unions de Consommateurs (BEUC), Europe’s largest consumer organisation, has a position on antimicrobials (see www.beuc.org/public/xfiles2000/x2000/x007e.pdf). Six Nordic consumer organisations united in their fight to urge their governments to develop restrictions on the use of antibiotics in agriculture. Educational materials such as the on-line guide at www.iatp.org/EatWell/orgResults.cfm helps consumers identify and understand the different labels used for meat raised without antibiotics in the US, and provides information on producers, restaurants, supermarkets, co-ops, and community supported agriculture networks that sell antibiotic-free meats in the US. More than 50 scientists and 40 health and consumer groups in the US petitioned the US Food and Drug Administration to ban the use of certain antibiotics in animal feed if they are also used in (or are related to those used in) human medicine. A coalition of consumer, environmental, health, and agriculture groups in the US sponsored KeepAntibioticsWorking.com, an educational and outreach initiative dedicated to ending the overuse of antibiotics in animal agriculture. These activities provide ample evidence of the concern and viewpoint of consumers on the issue of using antimicrobials in animal husbandry. While there are great benefits to the prudent use of antimicrobials in both human and animal medicine, the risks posed by certain uses and misuses are unacceptable to consumers. References 1. Chee-Sanford J.C., Aminov R..I., Krapac I.J., Garrigues-Jeanjean N. & Mackie R.I. (2001). – Occurrence and diversity of tetracycline resistance genes in lagoons and groundwater underlying two swine production facilities. App. Env. Microbio., 67 (4), 1494-1502. 2. Fey P.D., Safranek T.J., Rupp M.E., Dunne E.F., Ribot E., Iwen P.C., Bradford P.A., Angulo F.J. & Hinrichs S.H. (2000). – Celtriaxone-resistant salmonella infection acquired by a child from cattle. N. Engl. J. Med., 342 (17), 1242-1249. 3. Gonzales R.D., Schreckenberger P.C., Graham M.B., Kelkar S., DenBesten K. & Quinn J.P. (2001). – Infections due to vancomycin-resistant Enterococcus faecium resistant to linezolid. Lancet Apr., 357 (9263), 1179. 4. Holmberg S.D., Osterholm M.T., Senger K.A. & Cohen M.L. (1984). – Drug-resistant Salmonella from animals fed antimicrobials. N. Engl. J. Med., 311 (10), 617-622. OIE International Standards on Antimicrobial Resistance, 2003 93 1. General aspects 5. Holmberg S.D., Wells J.G. & Cohen M.L. (1984). – Animal-to-man transmission of antimicrobial-resistant Salmonella. Investigations of U.S. outbreaks, 1971-1983. Science, 225, 833-835. 6. Mellon M., Benbrook C. & Benbrook K.L. (2001). – Hogging it! Estimates of antimicrobial abuse in livestock. Union Concerned Scientists. 7. Molbak K., Baggesen D.L., Aarestrup F.M., Ebbesen J.M., Engberg J., Frydendahl K., Gerner-Smidt P., Petersen A.M. & Wegener H.C. (1999). – An outbreak of multidrug-resistant, quinolone-resistant Salmonella enterica serotype typhimurium DT104. N. Engl. J. Med., 341 (19), 1420-1425. 8. Raloff J. (1999). – Waterways carry antibiotic resistance. Science News, 155, 356. 9. Raloff J. (2001). – Antibiotic resistance is coming to dinner. Science News, 159, 325. 10. Wegener H.C., Aarestrup F.M., Jensen L.B., Hammerum A.M. & Bager F. (1999). – Use of antimicrobial growth promoters in food animals and Enterococcus faecium resistance to therapeutic antimicrobial drugs in Europe. Emerg. infect. Dis., 5 (3), 329-335. __________ 94 OIE International Standards on Antimicrobial Resistance, 2003 1. General aspects Activities of the Food and Agriculture Organization in relation to antimicrobial resistance in humans and animals Y. Cheneau Animal Health Service, Food and Agriculture Organization of the United Nations, Rome, Italy Following the accidental discovery in the early 1950s that the administration of antibiotics to animals increased their growth rates, the practice of adding such agents to the diet of animals has become a common practice throughout the world. This appears to be the case in the developed world where intensive systems of livestock production are common practice. In addition to the use of antimicrobial agents as growth promoters in feed and water, the therapeutic and sub-therapeutic use of these agents are thought to contribute to the development of resistance in bacteria. A bacterial strain is said to be resistant when a genetic modification allows it to tolerate a significant increase in antibiotic concentration. The problem with antibiotics in feed was recognised by the Government of the United Kingdom when in 1971, the Swann Committee report on the use of antibiotics in animal feed was accepted. The most important recommendation of this committee was that the only antibiotics that could be added to animal feed without a prescription from a veterinarian were those that had little or no application as therapeutic agents in man or animals and did not produce cross resistance against those that were used as therapeutic agents. Thirty years after the acceptance of the Swann Report, the problem of antibiotic resistance is still with us. Screaming headlines in journals and newspapers world-wide recently, seem to echo the continuing concerns of the general public on problems associated with antibiotic resistance in bacteria. Resistant bacteria could be transmitted from food animals to humans primarily via food. The need for containment of the development of resistant bacteria has therefore gained the attention of agencies responsible for food and the general well being of mankind, such as FAO (Food and Agriculture Organization of the United Nations), WHO (World Health Organization) and OIE (World organisation for animal health). Animal products are often contaminated with antimicrobial residues administered through feed or water. Residues, which may give rise to the evolution of resistant bacteria, can also occur through accidental cross contamination in feed mills. Feed contamination with undeclared antimicrobials is now a global problem calling for attention. To assess the magnitude of the problem accurately, utilisation patterns of antimicrobials should be quantified. It is in the context of problems posed world-wide by resistant bacteria to public health that the FAO, together with other agencies that share similar concerns, came together to form committees specifically to address this problem. This paper attempts to present the role of FAO through its various scientific committees and joint committees, in dealing with the problem of antimicrobial OIE International Standards on Antimicrobial Resistance, 2003 95 1. General aspects resistance. Recommendations and proposals are also presented on the way forward in the global efforts to contain antimicrobial resistance. Statement of the problem Compounds used as growth promoters are normally put in feed or water at low concentrations. This on going and often low level dosing for growth promotion and for prophylaxis inevitably results in the development of resistant bacteria in or near livestock. It heightens the fear of new resistant strains ‘jumping’ between species. Vancomycin-resistant Enterococcus faecium (VRE) is a typical example of a resistant bacterium appearing in animals that might have ‘jumped’ into the vulnerable segments of the human population. In 1997, the WHO recommended that antibiotics or their derivatives normally prescribed for humans be prohibited as growth promoters in animals. Secondly, it was recommended that antimicrobials should never be used as a substitute for high quality animal hygiene and management. In 1998, the European Union banned the use of antimicrobials prescribed for the treatment of human infections as growth promoters in animals. In Germany and Denmark, preliminary research appears to offer support to the adoption of this policy. The ban on avoparcin as a growth promoter in poultry and pigs, has led to the decrease in the prevalence of VRE in poultry, pigs and the population at large in these two countries. According to recent data (FEDESA [European Federation of Animal Health], 2001) an estimated 13,216 tonnes of antibiotics were used in the EU member states and Switzerland in 1999. Out of this amount, 8,585 tonnes (65%) were utilised for human health purposes, while animals used 4,688 tonnes (35%). Of the antibiotics that were used in animals, 3,902 tonnes (29% of total usage) were administered for prophylactic or therapeutic reasons, while 786 tonnes (or 6% of the total) were given to farm animals in their feed or water as growth promoters. It has been estimated that the amount of antibiotics used as growth promoters has fallen by 50% since 1997, when the WHO and later the EU recommended the ban of antibiotics used in human therapy as growth promoters. Position of the Food and Agriculture Organization a) The use of veterinary antimicrobial in food producing animals is likely to result in small quantities of residues of the product, or its metabolites, being present in foods of animal origin. It is now possible to detect such residues at extremely low concentrations based on Maximum Residue Limits (MRL) set by the FAO/WHO Codex Alimentarius (Committee on Residues of Veterinary Drugs in Foods). Several existing FAO/WHO Codex Alimentarius standards, guidelines and recommendations that include provisions relating to the quality and safety of animal feeds and foods of animal origin, consider the utilisation of antimicrobials in animal foodstuffs. These include: – List of Codex Maximum Residue Levels (MRL,s) for Veterinary Drugs (Vol. 3) – Recommended International Code of Practice for the control of the use of veterinary drugs (CAC/RCP38-1993, Vol. 3). 96 OIE International Standards on Antimicrobial Resistance, 2003 1. General aspects This code sets out guidelines on the prescriptions, application, distribution and control of drugs used for treating animals and improving animal production. It includes Good Practices in the Use of Veterinary Drugs (GPVD), including premixes for the manufacture of medicated feedstuffs. The question of microbial resistance is also being discussed within the Codex Committee on Residues of Veterinary Drugs in Foods (CCRVDF) and the Codex Committee on Food Hygiene (CCFH). The CCRVDF receives scientific advice from the FAO/WHO Joint Expert Committee on Food Additives which considers the impact of antimicrobial residues on the gut. The CCFH considers antimicrobial resistant bacteria in relation to food hygiene. This committee is already involved in risk assessment associated with microbiological contamination of food and is therefore, well placed to take a risk analysis approach to the question of antimicrobial resistance. b) FAO endorses WHO recommendation to phase out and finally abolish the use of antimicrobials as growth promoters if similar products are also licensed in human medicine. c) FAO advised countries to adopt immediate measures and to follow the available guidelines for the containment of antimicrobial resistance from antimicrobial use in livestock. d) FAO encourages countries to contain the spread of antimicrobial resistance by implementing a combination of measures such as: – establishing, implementing and verifying compliance of policies and legislation for the use of antimicrobials as feed addictives – monitoring the patterns of bacterial resistance and the use (qualitative and quantitative) of antimicrobials in animal production – continuous education, training and awareness of veterinarians, government inspectors, feed producers and farmers on the appropriate utilisation of antimicrobials and the consequences of abuse. Proposed action for the livestock and animal feed industry in all countries Given the direct links between biosecurity, feed safety and safety of foods of animal origin with regards to the prevention of antimicrobial resistance, it is essential that adequate attention be given to feed production and manufacture. Feed production must therefore be subject, in the same way as food production, to quality assurance. Industry is ultimately responsible for the quality and safety of the food and feed that it produces. National authorities should provide guidance to industry, including codes of practice and standards that they must respect. Governments must also establish the necessary controls to ensure that industry consistently meets mandatory quality and safety standards. It is the responsibility of industry and national governments to ensure safety of feed and food. It is important to realise however, that the large OIE International Standards on Antimicrobial Resistance, 2003 97 1. General aspects volume of international trade in foods of animal origin, as well as in feedstuffs, adds an important international dimension to the control of animal feedstuffs. International organisations also have an important role to play in providing information and training that could be used at national level to improve the knowledge on antimicrobial additives to feedstuffs. Recommendations a) The FAO together with the WHO and the OIE and other interested bodies, must develop a global strategy to contain the deteriorating situation with regards to the emergence of resistant bacterial strains. Creating awareness in the general public, especially livestock farmers, on the wiser use of antimicrobial agents is imperative to halt the spread of antimicrobial resistance. The strengthening of legal frameworks on the prescription of drugs and control of growth promoters in feed will all help to curb this potential menace. b) Evidence is gradually emerging that resistant strains of members of the Enterobacteriaceae in the gut flora of domestic animals reach man by the food chain. This calls for the highest degree of hygiene and the implementation, from the farm (stable) to the table, of HACCP (Hazard Analysis and Critical Control Point) standards. c) It is also suggested that a co-ordinated effort be made to survey the incidence of antimicrobial resistance in domestic animal populations world-wide. A distribution map of global microbial resistance to be made from this surveillance, could form the basis of early warning and reaction e.g. enforcement or re-evaluation of regulations in affected countries. References 1. Anon. (1973). – Control of harmful residues in food for human and animal consumption: the public health aspect of antibiotics in feedstuffs. Report of a working group – WHO Regional Office for Europe. 2. Anon. (2000). – Overcoming antimicrobial resistance: world health report on infectious diseases. 3. Anon. (2001). – Antibiotic use in farm animals. Report of the European Federation of Animal Health (FEDESA). 4. Anon. (2001). – Evaluation of certain veterinary drug residues in food. 55th Report of the Joint FAO/WHO Expert Committee on Food Additives. 5. Kidd A.R.M. (1994). – The potential risk of antimicrobial residues on human gastrointestinal microflora. __________ 98 OIE International Standards on Antimicrobial Resistance, 2003 1. General aspects The activities of the World Health Organization in antimicrobial resistance R. Williams Communicable Disease Surveillance and Response, World Health Organization, Geneva, Switzerland The aim of this presentation is to summarise through examples, the four main areas of WHO’s activities in antimicrobial resistance in humans and animals. – Raising awareness: targeted particularly towards Ministries of Health in Member States; professional societies and non-governmental organisations (NGOs); professionals working in infectious disease control programmes; the media and the public. Examples include: Resolutions of the World Health Assembly; expert consultations and reports; participation and presentation at meetings; fact sheets, documents and publications available through the web (www.who.int/emc/amr.html); and information exchange with other WHO programmes through working groups etc. – Providing strategic and technical guidance on interventions to contain resistance: examples include: WHO Global Principles for the Containment of Antimicrobial Resistance in Animals Intended for Food and the WHO Global Strategy for Containment of Antimicrobial Resistance. – Assisting countries to establish surveillance through providing guidelines for surveillance, laboratory manuals, software tools, etc.; laboratory training courses, and the provision of external quality assurance schemes. – Promoting partnership and information sharing: examples include: the WHO Collaborating Centre network which provides key support for national and international capacity strengthening activities and quality assurance; partnerships with NGOs and professional societies in projects to test interventions to contain resistance; web-based databases such as AR Infobank and Global Salm Surv. __________ OIE International Standards on Antimicrobial Resistance, 2003 99 1. General aspects Codex activities in relation to antimicrobial resistance 1 A. Bruno Secretariat, Codex Alimentarius Commission, Joint FAO/WHO Food Standards Programme, Vialle delle Terme di Caracalla, 00100 Rome, Italy The relationship of the use of antimicrobials in food-producing animals and the emergence of resistant bacteria in the food chain is a concern and has been the subject of numerous national and international consultations. The extent to which antimicrobial use in food animals (including aquaculture), horticulture or humans contributes to antimicrobial-resistant bacteria in humans varies between the different bacteria and different regions. Within Codex, there are a number of bodies that currently are considering the public health implications of using antimicrobial agents in food-producing animals: the Codex Committee on Residues of Veterinary Drugs in Foods (CCRVDF) and the Joint FAO/WHO Expert Committee on Food Additives (JECFA), the Codex Committee on Food Hygiene (CCFH), the ad hoc Intergovernmental Task Force on Animal Feeding (TFAF) and others. Approaches to understanding the public health significance of antimicrobial resistance have tended to focus on the professional disciplines reflected by the traditional membership of these groups: safety of residues in CCRVDF and JECFA, microbiological risk profiles in CCFH, and feeding practices and the manufacture of animal feeds in TFAF (Task Force on Animal Feeding). Codex Committee on Residues of Veterinary Drugs in Foods The CCRVDF is responsible for the establishment of maximum levels for residues of veterinary drugs in foods, including drugs used for therapeutic, prophylactic or diagnostic purposes or for modification of physiological functions or behaviour. CCRVDF obtains its scientific advice from JECFA. The CCRVDF was established following the recommendation of a Joint FAO/WHO Expert Consultation on Residues of Veterinary Drugs in Foods 2, (29 October-5 November 1984). The consultation also discussed the problem of sub-therapeutic use of antibiotics in animals and the concern as to the effects on public health. In this context, it clearly recognised that the threat to man due to the sub-therapeutic uses of antibacterials in animals and the development of resistant organisms in animals should not be confused with the threat associated with the ingestion of veterinary drugs residues. However, because of the importance of the problem and the lack of uniformity in 1 Paper based on ‘Discussion Paper on Antimicrobial Resistance and the Use of Antimicrobial in Animal Production’ (CX/RVDF 01/10) prepared for 13th Session of Codex Committee on Residues of Veterinary Drugs in Foods, 4-7 December 2001, Charleston, South Carolina (USA) 2 Residues of Veterinary Drugs in Foods – Report of a Joint FAO/WHO Expert Consultation. FAO Food and Nutrition Paper N. 32, Rome, 1985 100 OIE International Standards on Antimicrobial Resistance, 2003 1. General aspects regulating such uses, the Consultation pointed out that the issue, although unrelated to the drug residues issue may require further action by FAO/WHO and other international organisations with possible further implication for the Codex. The CCRVDF at its First Session (27-31 October 1986) considered the framework of the Committee. At that time, several delegations expressed concern at the consequences of adding antibiotics to feedstuffs in low doses to increase feed efficiency. The CCRVDF agreed that it should deal only with problems related to the residues of veterinary drugs in foods and not to the possibility of transferring resistant strains to human beings; and that the latter was a matter of food hygiene which should be referred to the appropriate Codex Committee 3. The issue of antimicrobial resistance was again considered by the CCRVDF at its 11th Session (15-18 September 1998) as a result of the discussion on the reports on WHO consultation on ‘Use of Antimicrobials in Livestock Production’ and the Joint FAO/WHO Consultation on ‘The Non-Human Medical Use of Antimicrobials’. As the delegations expressed different opinions on how CCRVDF should address issues related to antimicrobial resistance and safety of food of animal origin, the Committee agreed to prepare for its next Session, a paper which takes into account Codex discussion and/or decision on these issues and relevant reports of other international organisations. As per recommendation of the 12th CCRVDF, the paper 4 was further revised for consideration by the 13th CCRVDF, to be held on 4-7 December 2001. This discussion paper entitled ‘Antimicrobial Resistance and the Use of Antimicrobials in Animal Production’ 5 provides an overview of issues and activities concerning antimicrobial resistance relevant to the work of the CCRVDF and proposes a draft ‘Code of Practice to Minimise and Contain Antimicrobial Resistance’, which uses as its starting point the OIE (World organisation for animal health) Guidelines for the Responsible and Prudent Use of Antimicrobial Agents in Veterinary Medicine. The paper highlights the importance of the CCRVDF working closely with other relevant international standard-setting and regulatory bodies and in close coordination with the CCFH and the Codex Committee on Pesticide Residues (CCPR), in accordance with their respective terms of reference. It recognises that the CCRVDF is composed of people having responsibilities in veterinary medicinal product management in general and in veterinary medicinal product registration in particular, and that the CCRVDF has experience with the issue of antimicrobial resistance in establishing MRLs (Maximum Residue Limits) for antimicrobial residues and is currently basing its work on the risk analysis approach, including the setting of science-based risk assessment policy. The paper recommends that the CCRVDF be ALINORM 87/31, paragraph 130,131 CX/RVDF 00/4, July 2001 5 CX/RVDF 01/10 ‘Discussion Paper on Antimicrobial Resistance and the Use of Antimicrobials in Animal Production’, July 2001 3 4 OIE International Standards on Antimicrobial Resistance, 2003 101 1. General aspects involved in reducing the prevalence of bacteria resistant to antimicrobials in animalderived food and that it should be responsible for: a) developing a risk assessment policy regarding animal bacteria resistant to antimicrobials b) identifying the priority risk assessments to be carried out by an appropriate expert group in conjunction with CCFH c) considering the outcome of these risk assessments for proposing recommendations likely to help Codex member states in their risk management responsibilities. Codex Committee on food hygiene The CCFH is responsible for the elaboration of provisions on food hygiene applicable to all foods, and has established principles for the establishment and application of microbiological criteria for foods and principles and guidelines for the conduct of microbiological risk assessment. The CCFH, due to its recognised expertise in food hygiene in general and in food microbiology in particular, has the specific duty to propose any measures likely to improve the microbiological quality of animal-derived food and to decrease the burden of bacteria, sensitive or resistant to antimicrobials. The CCFH should also prioritise pathogens or pathogen-commodity combinations, including antimicrobialresistant pathogens, for microbiological risk assessments. The CCFH addressed the issue of antibiotic resistance in its last three sessions. The 31st CCFH (26-30 October 1998) considered a paper on ‘Antibiotic Resistance Bacteria in Foods’, which outlined the need to evaluate and address the risks associated with the development of drug resistance in bacteria following the use of antibiotics. As there were different opinions expressed at the meeting, the Committee agreed to prepare a discussion paper to clarify the issues involved and their relevance to the work of the Committee, for further consideration at the 32nd session (29 November-4 December 1999). As recommended by the Executive Committee at its 47th session in June 2000, the paper 6 was further revised in the form of a risk profile to determine which subjects fall within the terms of reference of the CCFH. The revised paper 7 presented at the 33rd CCFH (23-28 October 2000) highlighted that: a) antimicrobial resistance contributes to the public health risk of pathogenic bacteria in food because they result in an increase in the morbidity, mortality and costs associated with the infection; and b) antimicrobial-resistant bacteria represent a public health risk via food due to the potential dissemination of resistant genes. 6 7 CX/FH 99/12 CX/FH 00/11 ‘Risk Profile on Antimicrobial-Resistant Bacteria in Food’ 102 OIE International Standards on Antimicrobial Resistance, 2003 1. General aspects It recognises that the presence of antimicrobial-resistant bacteria in food is related to the use of antimicrobials, including growth promoting antimicrobials in food production and in humans, as well as the transmission of bacteria in the various steps of the food chain and environmental spread. It also identifies a number of strategies available to control antimicrobial-resistant bacteria in foods including hygienic measures, prudent use and other efforts to reduce overuse and misuse of antimicrobials. The paper recommended that the health risk associated with antimicrobial-resistant bacteria in the food chain be further addressed in the various committees involved; that the CCFH commission a risk assessment for selected specific scenarios and that quinolone-resistant Salmonella and Campylobacter in poultry should be the top priority. Moreover, it recommended that the principles of ‘reservation for human medicine’ of certain antimicrobial substances need international validation. Recognising the importance of the issue of antimicrobial resistant bacteria in food, the CCFH agreed to forward the document to the Codex Executive Committee to assist the coordination of work between the Committees concerned with the work of other international organisations (e.g. the OIE). In its 34th Session (8-13 October 2001) the CCFH will deliberate on the Executive Committee’s recommendations (see below). The CCFH will explore the possibility of discontinuing or tabling the work until the recommendations made by the Executive Committee are implemented or, continue the work to develop guidelines for incorporation into other Codex documents such as General Principles of Food Hygiene or the Guidelines for Risk Assessment. Ad-Hoc Intergovernmental Task Force on Animal Feeding The ad-hoc Task Force on Animal Feeding was established by the 23rd Session of the Codex Commission with the mandate to: a) complete and extend the work done by the relevant Committee on the Draft Code of Practice for Good Animal Feeding b) address other aspects which are important for food safety, such as problems related to toxic substances, pathogens, microbial resistance, new technologies, storage, control measures, traceability, etc. At its First Session (13-15 June 2000), the Task Force was informed of the current activities of other Codex Committees including the CCFH, the CCRVDF, the CCPR regarding antimicrobial resistance in foods. The Task Force also addressed the issue of antibiotics used for growth promotion purposes. Opinions varied between those delegations that supported a statement in the Code that would prohibit such uses and those delegations that were of the opinion that antibiotics should not be used in the absence of a public health safety risk assessment. Attention was drawn to the report of the Representative of WHO on the outcome of the WHO Consultation on ‘Global Principles for the Containment of Antimicrobial Resistance’. It was agreed that further discussion of this issue should be undertaken in the light of the report and OIE International Standards on Antimicrobial Resistance, 2003 103 1. General aspects recommendations of the Consultation, as well as the reports and guidance of other groups such as the OIE, CCFH and CCRVDF. Codex Committee on Pesticide Residues In its 33rd Session 8 the CCPR considered the issue of the development of antibiotic resistance in humans following the request to add gentamycin and oxytetracycline on the priority list of substances to be evaluated. The request was supported by the fact that these substances complied with the criteria for inclusion on the priority list as they are very effective and important for the control of bacterial diseases of certain commodities; and also, by the fact that, according to GAP, residue levels are very low when these substances are used. Codex Committee on Fish and Fishery Products The issue of the development of resistance is taken into account in the draft Code of Practice for Fish and Fishery Products, Section 16 ‘Aquaculture Production’ 9, not yet been finalised by the CCFFP, as follows: ‘Uncontrolled and unlimited use of medicinal products may lead to the accumulation of undesirable residues in the fish treated and in the environment, and that the continuous use of antibacterial, antiprotozoan or anthelmintic products may favour the development of resistance. It is the responsibility of the veterinarian or other authorised persons to draw up programmes of preventive medicine for the fish farmer and to stress the importance of sound management and good husbandry in order to reduce the likelihood of fish diseases. Every effort should be made to use only those drugs known to be effective in treating the specific disease.’ 10 Executive Committee The Executive Committee examined the issues of coordination of work on Antibiotics Used on Agricultural Commodities and on Antimicrobial Resistant Bacteria in Food at its 48th Session 11 at the request of the Committee on Pesticide Residues 12 and the Committee on Food Hygiene 13. In relation to the matter raised by CCPR, the Executive Committee was of the opinion that the use of antimicrobials on agricultural commodities should be subject to evaluation within a risk analysis framework; the question was whether the normal process used for the evaluation of pesticides was the appropriate one. In the second case, the Executive Committee agreed that consideration should be given to antimicrobial resistant micro-organisms in food within a risk analysis framework on a case-by-case basis as microorganism/food combinations were being assessed. ALINORM 01/24, para.222 CX/FFP 00/4 10 CX/FFP 00/4, Section 16.9.1 11 ALINORM 01/4, paras 36-37 12 ALINORM 01/24A, para.122 13 ALINORM 01/13A, paras 132-142 8 9 104 OIE International Standards on Antimicrobial Resistance, 2003 1. General aspects The Executive Committee agreed however, that the issues raised by these Committees required a more general, multidisciplinary and multi-agency response and recommended that FAO and WHO should give consideration to convening, as soon as possible, a multidisciplinary expert consultation in cooperation with the OIE, and if required, the IPPC, to advise the Commission on possible directions to be taken, including the establishment of a new task force if necessary. __________ OIE International Standards on Antimicrobial Resistance, 2003 105 2. Surveillance of antimicrobial consumption OIE International Standards on Antimicrobial Resistance, 2003 107 2. Surveillance of antimicrobial consumption Antimicrobial resistance: monitoring the quantities of antimicrobials used in animal husbandry †T. Nicholls (1), J. Acar (2), F. Anthony (3), A. Franklin (4), R. Gupta (5), Y. Tamura (6), S. Thompson (7), E.J. Threlfall (8), D. Vose (9), M. van Vuuren (10), D.G. White (11), H.C. Wegener (12) & M.L. Costarrica (13) (1) National Offices of Animal and Plant Health and Food Safety, Animal Health Science and Emergency Management Branch, Department of Agriculture, Fisheries and Forestry, P.O. Box 858, Canberra, ACT 2601, Australia (2) Université Pierre et Marie Curie, Service de Microbiologie Médicale, Fondation Hôpital Saint-Joseph, 185 rue Raymond Losserand, 75674 Paris Cedex 14, France (3) Fresh Acre Veterinary Surgery, Flaggoners Green, Bromyard, Herefordshire HR7 4QR, United Kingdom (4) The National Veterinary Institute (SVA), Department of Antibiotics, SE 751 89 Uppsala, Sweden (5) College of Veterinary Sciences, Veterinary Bacteriology, Department of Microbiology, G.B. Pant University of Agriculture and Technology, Pantnagar, 263 145 Uttar Pradesh, India (6) National Veterinary Assay Laboratory, Ministry of Agriculture, Forestry and Fisheries, 1-51-1 Tolura, Kokubunji, Tokyo 185-8511, Japan (7) Joint Institute for Food Safety Research, Department for Health and Human Services Liaison, 1400 Independence Avenue, SW, Mail Stop 2256, Washington, DC 20250-2256, United States of America (8) Public Health Laboratory Service (PHLS), Central Public Health Laboratory, Laboratory of Enteric Pathogens, 61 Collindale Avenue, London NW9 5HT, United Kingdom (9) David Vose Consulting, Le Bourg, 24400 Les Lèches, France (10) University of Pretoria, Faculty of Veterinary Science, Department of Veterinary Tropical Diseases, Private Bag X04, Onderstepoort 0110, South Africa (11) Centre for Veterinary Medicine, Food and Drug Administration, Office of Research, HFV-530, 8401 Muirkirk Road, Laurel, Maryland 20708, United States of America (12) World Health Organization, Detached National Expert, Division of Emerging and Transmissible Diseases, Animal and Food-related Public Health Risks, 20 avenue Appia, 1211 Geneva, Switzerland (13) Food and Agriculture Organization, Food Quality and Standards Service, Senior Officer, via delle Terme di Caracalla, 00100 Rome, Italy This report, prepared by the OIE Ad hoc Group of experts on antimicrobial resistance, has not yet received the approval of the International Committee of the OIE Summary This guideline, developed by the OIE (World organisation for animal health) for the monitoring of the quantities of antimicrobials used in animal husbandry, provides the methodology required to assess the amounts of antimicrobials used, to supply data to be used for risk analysis and to improve guidance on the appropriate use of antimicrobials. Information may be gathered from a number of sources, such as the competent authorities, industry and users. The usefulness of different types of information is discussed and recommendations are given on how to collect detailed information, each year, on the antimicrobial quantities used per class and active substance. Information should also be collected on the route of administration (oral and parenteral) and the animal species. OIE International Standards on Antimicrobial Resistance, 2003 109 2. Surveillance of antimicrobial consumption Keywords Animal health – Antimicrobial resistance – Containment of resistance – Human medicine – Monitoring of antimicrobial use – Public health – Risk analysis – Standards – Veterinary medicine – World Organisation for Animal Health. Introduction There is world-wide concern about antimicrobial resistance in bacteria and about the use of antimicrobials in food-producing animals which may contribute to antimicrobial resistance problems in human and veterinary medicine. Data on antimicrobial use in food animals is essential to identify such problems at the national level and in subsequent risk analysis, planning and execution of programmes where this concern is further defined and addressed. The purpose of this document is to describe an approach for the monitoring of quantities of antimicrobials used in animal husbandry. The objectives of such a monitoring system will be defined, as will indications for the use of the data. The sources and the types of data to be collected will be identified. Attention will be given to the collection of information that most accurately describes the use of antimicrobials in animals, and potential difficulties in the collection of that data. Monitoring programmes will also be useful for local authorities dealing with specific, individual or regional antimicrobial resistance problems. The reporting of data and future directions to facilitate international harmonisation will be addressed. The information presented in this chapter is not designed to be prescriptive for OIE (World organisation for animal health) Member Countries where abilities to monitor the quantities of antimicrobials used in animal husbandry vary greatly. Rather, this chapter outlines a systematic approach that Member Countries can consider when addressing this aspect of antimicrobial resistance management. Reasons for collecting information on the quantities of antimicrobials used in animal husbandry The goal of any programme to monitor the quantities of antimicrobials used in animals is to have objective and quantitative information to evaluate usage patterns by animal species, antimicrobial class, potency and type of use in order to evaluate antimicrobial exposure. These data are essential for risk analyses and planning, can be helpful in interpreting resistance surveillance data and can assist in the ability to respond to problems of antimicrobial resistance in a precise and targeted way. The data may also assist in evaluating the effectiveness of efforts to ensure prudent use and mitigation strategies (for example, by identifying changes in prescribing practices for veterinarians) and to indicate where alteration of antimicrobial prescribing practices might be appropriate, or if changes in prescription practice have altered the pattern of antimicrobial use. 110 OIE International Standards on Antimicrobial Resistance, 2003 2. Surveillance of antimicrobial consumption The continued collection of this basic data will also help give an indication of trends in the use of animal antimicrobials over time and the role thereof in the development of antimicrobial resistance in animals. This information could be compared with medical, agricultural and other antimicrobial use data as part of any risk analysis necessary for the holistic and integrated approach of a Member Country to optimise antimicrobial use. The level of information collected will depend on the perceived or actual concern of a Member Country with the issue of antimicrobial resistance, and the ability of that country to fund the necessary programmes. However, in the consideration of antimicrobial resistance by a Member Country, there will also be a need for data on the medical and agricultural use of the chemicals if meaningful evaluations are to be undertaken. For all OIE Member Countries, the minimum basic information collected should include the total amount of active antimicrobial ingredient used per kilogram by class, or specific formula if there are differences in potency within a class. In addition, the type of use (therapeutic or growth promotion) and route of administration (parenteral or oral administration) should be recorded. Member Countries could explore the possibility of establishing regional or local databases of antimicrobial usage/resistance patterns, since these may be of more practical use to the consulting veterinarian. Such use would require a classification of food animal antimicrobial use. Such classifications need to produce useful data. For example, a simple classification of in-feed and veterinary use would probably be misleading in risk analysis because both in-feed use and veterinary use of antimicrobials can be for the purpose of treatment and growth promotion. The key to understanding the relationship between antimicrobial use in animals and the development of resistance in animal bacteria is likely to be related to the reasons for selection of particular antimicrobials as well as the rate of prescription and the dose and length of treatment regimens. This information is critical if feedback pathways to veterinarians prescribing antimicrobials are to be established so usage patterns can be defined and the development of antimicrobial resistance in animal bacteria can be analysed and acted on by regulatory and other authorities, where appropriate. The total consumption of antimicrobials for human, medical, food animal and other uses is a key factor in any consideration of this issue. While this guideline will only consider animal antimicrobial use, Member Countries may wish to consider, for reasons of cost and administrative efficiency, collecting medical, ood animal, agricultural and other antimicrobial use data in a single programme. A consolidated programme would also facilitate comparisons of animal use with human use data for relative risk analysis. OIE International Standards on Antimicrobial Resistance, 2003 111 2. Surveillance of antimicrobial consumption Sources of antimicrobial use data Basic sources Sources of data will vary from country to country and depend on factors such as whether a Member Country manufactures antimicrobials, exports and/or imports antimicrobials, and whether or not there is any accessible and accurate source of this information from the national regulatory authorities. Such sources may include customs, import and export data, manufacturing and manufacturing sales data. Direct sources Most countries have a legislative infrastructure for the registration, distribution and control of animal antimicrobial use (see Antimicrobial resistance: responsible and prudent use of antimicrobial agents in veterinary medicine, earlier in this volume). Data from animal drug wholesalers, retailers, pharmacists, veterinarians, feed stores, feed mills and organised industry associations in these countries might be an efficient and practical source of data on antimicrobial use in animals. A possible mechanism for the collection of this information is that the provision of appropriate information by manufacturers to the regulatory authority is a requirement of antimicrobial registration, provided commercial confidentiality requirements can be met. End-use sources (veterinarians and food animal producers) Periodic audits and statistically based surveys of either direct sources or end-use sources of animal antimicrobials, rather than ongoing data collection programmes, may be a method of obtaining accurate and detailed information on animal antimicrobial use. This may be appropriate when basic or direct sources cannot be used for the routine collection of this information. Targeted surveys or audits could be used as an adjunct to this information, or when more accurate and locally specific information is required. In addition to assisting in quantifying the extent of use of antimicrobials, end-use (particularly of veterinarians and food animal producers) surveys may be used to identify patterns of antimicrobial, prophylactic, therapeutic and growth promoter use that may have implications in an epidemiological investigation of the development of antimicrobial resistance. Factors such as seasonality and disease conditions, species affected, agricultural systems (e.g. extensive range conditions and feedlots), dose rate, duration and length of treatment with antimicrobials relative to the recommendations for the purpose of the antimicrobial, may be important factors. An issue may be the need to recruit sufficient numbers of veterinarians and farmers to allow robust analysis. Collection, storage and processing of data from end-use sources are likely to be inefficient and expensive processes unless carefully designed and well managed, but should have the advantage of producing accurate and targeted information. Recommendation In consideration of antimicrobial resistance management programmes, the sources of data available and options for the collection of data for individual OIE Member 112 OIE International Standards on Antimicrobial Resistance, 2003 2. Surveillance of antimicrobial consumption Countries need early and careful analysis as well as careful consideration to ensure the cost-effective use of resources to fund national programme objectives. Categories of data Minimal antibiotic use data requirements and data levels It is the opinion of the OIE Ad hoc Group on antimicrobial resistance that, in OIE Member Countries, the minimal data collected should be the annual weight in kilograms of the active ingredient of the antimicrobial(s) used in food animal production. If a Member Country has the infrastructure for capturing basic animal antimicrobial use data for a specific antimicrobial, then additional information can be considered to cascade from this in a series of subdivisions or levels of detail. The relevant authorities within the Member Country should decide on the level of detail required so that the data collected can contribute to the aspirations of the Member Country to limit the development of antimicrobial resistance. Such a cascade of levels could include the following: a) the absolute amount in kilograms of antimicrobial active used per antimicrobial family per year, or for a specific antimicrobial chemical entity when this information is required b) therapeutic and growth promotion use in kilograms of the specific antimicrobial active c) subdivision of antimicrobial use into therapeutic and growth promotion use by species d) subdivision of the data into the route of administration, specifically in-feed, inwater, injectable, oral, intramammary, intra-uterine and topical e) further subdivision of these figures by season and region by a Member Country may be useful (note: this may be especially helpful in countries with large variations in environmental/management conditions, or where animals are moved from one locality to another during production) f) further breakdown of data for analysis of antimicrobial use at the regional, local, herd and individual veterinarian level may be possible using veterinary practice computer management software as part of specific targeted surveys or audits. Analysis of this information within the local or regional context could be useful for individual practitioners and practices where specific antimicrobial resistance has been identified and feedback is required. Registration and regulation All OIE Member Countries should have appropriate veterinary chemical registration standards, either through a national veterinary medicinal product registration authority, or through requirements that imported products comply with the registration system of the exporting country or of another country. This is to ensure that safe, efficacious and quality veterinary products are used in food animals. Many OIE International Standards on Antimicrobial Resistance, 2003 113 2. Surveillance of antimicrobial consumption Member Countries also have agriculture and veterinary chemical residue monitoring and surveillance programmes for measuring chemical residues in food. These two activities can guide Member Countries regarding the level of detail of animal antimicrobial use information required. For example, monitoring the amount of registered antimicrobial residues in food animals at slaughter would be an elementary data collection activity. If the antimicrobial residue programme or other information indicated that non-registered antimicrobials were used in food animals, then provisions could be made to collect animal antimicrobial use information at the end-user level, using targeted surveys or specially designed monitoring programmes. However, if these programmes are not in place, a good starting point for Member Countries may be to utilise customs permit data to quantify imported antimicrobials. The basic regulatory requirements of Member Countries recommended by this OIE Ad hoc Group are discussed in more detail in Antimicrobial resistance: responsible and prudent use of antimicrobial agents in veterinary medicine. Classes of antimicrobials Decisions need to be made on what classes of antimicrobials should be considered and what members of various antimicrobial classes should be included in the data collection programme. These decisions should be based on currently known mechanisms of antimicrobial activity of the particular antimicrobial and its relative potency. For example, individual members of the dichloracetic acid group of antimicrobials (e.g. chloramphenicol and florfenicol) have different mechanisms of action. Other preparations, such as the tetracyclines, have different potency levels, for example, chlortetracycline is not as potent as doxycycline on a mg/kg basis. Ideally, animal-use data should be collected for each individual member of the antimicrobial group registered for use. Where common mechanisms of action exist, this data can be aggregated at a later date, if required. An internationally accepted method of comparing antimicrobials, taking these factors into account, would be useful, as would internationally accepted nomenclature for antimicrobial classes so that future comparison of use data could be facilitated. An international code for the specific identification of medical and veterinary antimicrobials is available in the ATCvet Index (2). It is recommended that this code be used in the identification of specific antimicrobials. Species, production system, regional and seasonal data Most countries register animal use antimicrobials for a specific food animal species (cattle, sheep, goats, pigs, poultry, horses and fish) and often for specific diseases. Frequently, antimicrobial product registration is for multiple species use, such as for cattle, sheep and goats, and this may create difficulties in determining use patterns. In order for a country to effectively analyse animal antimicrobial use patterns, including off-label use, a good understanding of the circumstances of food animal antimicrobial use is required. For example, cattle are raised in extensive range conditions, in feedlots or held in barns during the winter. An understanding of what antimicrobials are used 114 OIE International Standards on Antimicrobial Resistance, 2003 2. Surveillance of antimicrobial consumption in specific animal species and industries in different regions, as well as seasonal influences on disease prevalence are likely to be important information in the risk analysis of this issue. Such general information may, for example, identify a potential problem, such as possible inappropriate animal antimicrobial use. Further investigation could lead to confirmation and suggest corrective action, such as feedback of information to veterinarians and producers. Other important information If an OIE Member Country is considering animal antimicrobial use in food animals, a breakdown of the animal industries may be useful in any risk analysis or for comparison of animal antimicrobial use with human medical use within and between countries. For example, the total number of animals (meat, dairy and draught cattle, and meat, fibre and dairy sheep) in the country would be essential basic information. In addition, the total number of animals raised and their weight in kilograms for food production per year would be essential information in the assessment of animal antimicrobial use figures. Breakdown of the type of production enterprises (for example, extensive versus intensive) would also be useful if accurate industry enterprise antimicrobial use data was not available to give an indication of how animal antimicrobials were being used. Future directions Since the crude amount of antimicrobial (in kilograms) used yearly only indirectly represents antimicrobial exposure, and hence the selective pressure on bacterial populations, more sophisticated measures are needed. Such concepts have been developed in human medicine. Medical monitoring of antibiotic use in community and hospital medicine has led to the evolution of the concept of the defined daily dose (DDD) expressed as the DDDs/1,000 population/day. This approach takes into account the activity and potency of individual antimicrobials and the basic unit of comparison between individual antimicrobials becomes the DDD/1,000/population/day applied to the particular environment. However, a direct comparison between medical and animal use of antimicrobials is difficult and perhaps pointless other than for medical and veterinary authorities to have a rational baseline measure of national antibiotic use for ongoing comparison over a period of years. The management of medical and veterinary antimicrobial registration and use in most OIE Member Countries are separate and independent exercises. In the future, the management of animal antibiotic use may be dependent on risk analysis findings of the contribution of animal antibiotic use to medical antimicrobial resistance problems. Developing a DDD approach for antimicrobials in food animals would be difficult because of the wide range of animal weights (e.g. compare the weight of newly hatched chickens and new-born calves or meat chickens and cattle at slaughter). The reports of the Danish Zoonosis Centre (DANMAPs) use the concept of milligram of antimicrobial used per kilogram of meat produced (1). This concept could be useful in the monitoring and analysis of animal use of antibiotics overall, and within particular OIE International Standards on Antimicrobial Resistance, 2003 115 2. Surveillance of antimicrobial consumption species, although it does not address dose rate and length of treatment regimens in specific animal husbandry circumstances. However, by using milligram of antimicrobial active used per kg of meat produced as a measure of antimicrobial use in food animals it may be possible to undertake a comparison for specific antimicrobials, thus enabling evaluation of the relative selection pressure of, for example, two antibiotics with different activity and potency such as tetracycline and fluoroquinolones. This approach does not take into account the different pharmacological activity of different antibiotics and potency may need to be standardised in some way. Such a system would be a better measure, over time, of the total selection pressure applied to the particular environment under study, and would provide a more accurate measure of the relative importance of the different potential of antimicrobials for generating bacterial resistance. It would also be of value, for example, in measuring the consequence of changes in use patterns, such as the replacement of tetracycline use with fluoroquinolones, or in analyses where lifetime exposures of animals to antimicrobials were important. The development of concepts on the action of antimicrobials reflecting their relative activity through the consideration of potency, in conjunction with the kilograms of food animal product produced using these antimicrobials, is important. It would provide useful baseline information and assist in assessing the possible contribution of the animal use of antimicrobials in the production of antimicrobial resistance of medical and veterinary concern. Conclusions Data on the use of antimicrobials in animals is essential for risk analysis and the design and planning of antimicrobial resistance monitoring and surveillance programmes, as well as for the ongoing management of antimicrobial resistance on the individual farm, district, regional, national and international levels. Antibiorésistance : contrôle des quantités d’antibiotiques utilisées en production animale †T. Nicholls, J. Acar, F. Anthony, A. Franklin, R. Gupta, Y. Tamura, S. Thompson, E.J. Threlfall, D. Vose, M. van Vuuren, D.G. White, H.C. Wegener & M.L. Costarrica Résumé Cette ligne directrice, préparée par l’Organisation mondiale pour la santé animale et applicable au contrôle des quantités d’antibiotiques utilisées en production animale, fixe la méthodologie requise pour évaluer les quantités de produits antimicrobiens utilisés, pour fournir les informations nécessaires à l’analyse du risque et pour améliorer les instructions relatives à l’utilisation appropriée des antibiotiques. Les informations peuvent émaner de plusieurs sources, telles que les autorités compétentes, les professionnels du secteur et les utilisateurs. L’utilité des différents types d’information fait l’objet de la discussion et des recommandations sont données sur la manière de recueillir une information détaillée, chaque année, sur les quantités d’antibiotiques utilisés, par catégorie et par 116 OIE International Standards on Antimicrobial Resistance, 2003 2. Surveillance of antimicrobial consumption substance active. Il convient également de recueillir des informations relatives aux modes d’administration (orale ou parentérale) et aux espèces animales concernées. Mots-clés Analyse du risque – Antibiorésistance – Gestion de l’utilisation des antibiotiques – Maîtrise de la résistance – Médecine humaine – Médecine vétérinaire – Normes – Organisation mondiale pour la santé anbimale – Santé animale – Santé publique. Resistencia a los antimicrobianos: seguimiento del volumen de antimicrobianos utilizados en producción animal †T. Nicholls, J. Acar, F. Anthony, A. Franklin, R. Gupta, Y. Tamura, S. Thompson, E.J. Threlfall, D. Vose, M. van Vuuren, D.G. White, H.C. Wegener & M.L. Costarrica Resumen Esta directriz, elaborada por la Organización mundial de sanidad animal para facilitar el seguimiento de los volúmenes de antimicrobianos utilizados en producción animal, define la metodología indicada para evaluar ese parámetro, obtener datos útiles para el análisis de riesgos y orientar mejor al usuario sobre el empleo adecuado de los antimicrobianos. La información puede proceder simultáneamente de varias fuentes, como las autoridades competentes, los industriales del ramo y los usuarios. Los autores valoran la utilidad de distintos tipos de información y recomiendan métodos para recabar anualmente datos exactos sobre los volúmenes de productos antimicrobianos utilizados (desglosados por clase y principio activo). Conviene también obtener información sobre la vía de administración (oral o parenteral) y la especie animal de que se trate. Palabras clave Análisis de riesgos – Contención de las resistencias – Gestión del uso de antimicrobianos – Medicina humana – Medicina veterinaria – Normativa – Organización mundial de sanidad animal – Resistencia a los productos antimicrobianos – Salud pública – Sanidad animal. References 1. Danish Zoonosis Centre (2001). – DANMAP 2000. Danish Veterinary Laboratory, Ministry of Agriculture and Fisheries, Copenhagen, 56 pp. Website: http://www.svs.dk/uk/organization/lgo_zoo.htm (document accessed on 9 August 2001). 2. Nordic Council on Medicines (NCM) (2001). – Anatomical Therapeutic Chemical (ATC) vet Index. NCM, Uppsala. Website: http://www.nln.se/default.asp (document accessed on 9 August 2001). __________ OIE International Standards on Antimicrobial Resistance, 2003 117 2. Surveillance of antimicrobial consumption Surveillance of antimicrobial consumption activities in France G. Moulin National Agency for Veterinary Products (AFSSA – French Food Safety Agency), La Haute Marche, Javené B.P. 90203, 35302 Fougères, France Introduction In France, different studies have been carried out regarding the surveillance of antimicrobial consumption under the sponsorship of the Ministry of Agriculture: a sales survey of antimicrobials in Veterinary Medicinal Products (VMP), a survey of prescriptions by veterinarians, and an anti-microbial consumption survey in farms. This paper presents the results of the first survey carried out in France and relates to VMP sales in France in 1999. The survey was set up in collaboration with the French association of veterinary pharmaceutical companies (Syndicat de l’industrie du médicament vétérinaire et réactif – SIMV). The survey was only set up for veterinary medicinal products. Additives, growth promoters and coccidiostats were not followed, as they are out of scope of the activities of the National Agency for Veterinary Medicinal Products. Protocol The methodology used is quite simple and based on a very simple questionnaire sent by the agency and completed by the marketing authorisation holder. In March 2000, a letter was sent from the agency to the marketing authorisation holder asking for the return of the enclosed questionnaire for 30 June 2000 for every VMP containing antibiotics. For each medicinal product, the number of sold units should have been given for the period between 1 January 1999 and 31 December 1999. The sales figures for each product were cross-referenced with the data (quantitative and qualitative composition, pharmaceutical form, purpose, target animals) that is available in the National Agency of Veterinary Medicinal Products database. Then, calculations were done to obtain sold quantities in active substance per unit of mass. These figures were then sorted by active constituent and by antibiotics class. Results Sales data were collected for 1938 of medicinal products containing antibiotics. In France, 1,364 tonnes of antibiotics were sold in 1999. 118 OIE International Standards on Antimicrobial Resistance, 2003 2. Surveillance of antimicrobial consumption Sales distribution by antibiotics class The distribution by antibiotics class appears in Figure 1: 700 600 Tonnes 500 400 300 200 100 Am in og lyc os Be id tal es ac tam Ce in ph es alo sp Fl or uo in ro es qu in ol on es Fu ra ne s M ac ro lid es O th er s Ph en ico Po lym ls yx in es Q ui no lo Su ne lfo s na m id Te es tra cy cli Tr ne im s eth op rim 0 Antibiotic class Fig. 1 Veterinary use of antibiotics in France 1999 (tonnes) Four antibiotic classes (Tetracycline, Sulfonamide, Betalactam, Aminoglycoside) represent 83% of sold antibiotics. Tetracycline represents nearly half the total. It can be observed that compounds belonging to newer antibiotic classes represent relatively low volumes (Fluoroquinolone: 0.24%, Cephalosporine: 0.53%). Distribution of antibiotic sales by animal categories It is difficult to give figures for each animal species as the same VMP can be indicated for use in several species. Nevertheless, it is possible to have some information regarding the distribution between companion animals and food producing animals. In volume, sales of antibiotics for cats and dogs represent between 1.3% and 7.7% of the total. Food producing animal use represents the main part of antibiotic sales. There are several reasons why the distribution between food and companion animals is different. For example, furans are not permitted for use in food animal species and can only be used in companion animals. OIE International Standards on Antimicrobial Resistance, 2003 119 2. Surveillance of antimicrobial consumption In addition, for antibiotics such as cephalosporins, differences can be related to, among other parameters, the cost of the VMP. Antibiotic sales distribution by administration route The oral route accounts for 85% of anti-microbial sales, the parenteral route for 13% and other routes for 6.4%. Comparison with data from other countries of the European Union Percentage The distribution by antibiotic class is very similar to that observed in other countries. 60 50 40 30 20 10 0 France United-Kingdom Denmark Country Aminoglycosides Betalactams/Cephalosporines Fluoroquinolones Macrolides Trimethoprim/Sulphonamides Tetracyclines Others Fig. 2 Percentage sales of antibiotics by country in 1999 Discussion Data interpretation should be done with caution and one must take into account different factors such as, the number of animals, their weights, the dosage and the treatment duration. For example, the difference in antibiotic volumes used between food producing animals and companion animals is certainly related to the following factors: number of animals, weight, preferred route of administration, health status. Such surveys have some limitation. In particular, it is not possible to obtain sales data by species due to the fact that the same VMP can be authorised and used in different species. However, this kind of survey remains very interesting for historical comparisons and as a tool for risk analysis. 120 OIE International Standards on Antimicrobial Resistance, 2003 2. Surveillance of antimicrobial consumption Conclusion This was the first time that VMP sales in France had been studied. Antibiotic sales monitoring should be continued in order to follow the evolution of antibiotic use over time. These data could serve as a basis for the interpretation of the evolution of antibiotic resistant bacteria. __________ OIE International Standards on Antimicrobial Resistance, 2003 121 2. Surveillance of antimicrobial consumption Surveillance of antimicrobial consumption in Denmark D.L. Monnet (1), F. Bager (2) & L. Larsen (3) (1) Statens Serum Institut, Department of microbiological research and development, Artillerivej 5, 2300 Copenhagen, Denmark (2) Danish Zoonosis Centre, Danish Veterinary Laboratory, Bülowsvej 27, 1790 Copenhagen, Denmark (3) Danish Medicines Agency for the Danish Integrated Antimicrobial Resistance Monitoring and Research Programme (DANMAP), Axel Heides Gade 1, 2300 Koberhavn’s, Denmark In Denmark, all antimicrobials used in humans and for therapy in food animals are prescription-only medicines and must be distributed through pharmacies. All medicines must be registered by the Danish Medicines Agency (DMA). The DMA has the legal responsibility for monitoring the consumption of all medicinal products in humans. Since 1994, such data are available through monthly electronic reporting by all pharmacies. For each sale, packages are recorded by their specific code, together with information to identify the patient and the prescriber, the date and place of the sale, and possible subsidisation of cost. However, information on the indications for the prescription is not yet available. Data on antimicrobial consumption and the results of specific analyses, e.g. on the effect of changes in subsidisation, are reported at least yearly, as the number of WHO defined daily doses per 1,000 population per day in the DMA newsletter, on its website (www.laegemiddelstyrelsen.dk) and in the DANMAP report (www.svs.dk, ‘Zoonosecentret’). Summarised data on antimicrobial consumption in humans in Denmark are now updated monthly and made available to registered users on a dedicated website. Since 1996, data on the consumption of antimicrobials for therapy of food animals have been available through the yearly reporting by the pharmaceutical industry to the DMA on quantities sold in Denmark. Data on the use of antimicrobials for growth promotion were obtained through compulsory reporting to the Danish Plant Directorate by companies authorised to produce premixes containing antimicrobials. Such data are now starting to be collected through a new monitoring programme called VETSTAT. Pharmacies report on prescriptions by veterinarians, including the identity of the farm, the animal species and age group, the identity of the prescriber and the reason for prescribing, in addition to the name and quantity of the drug. Medicines sold or used by veterinarians must be reported directly to VETSTAT with the same information. Finally, feed mills report all sales of animal feed containing medicines or coccidiostats. Data on the consumption of antimicrobials in food animals in kilograms are presented yearly in the DANMAP report, together with a comparison with consumption in humans. Monthly data should soon be available through VETSTAT. __________ 122 OIE International Standards on Antimicrobial Resistance, 2003 2. Surveillance of antimicrobial consumption Antimicrobial resistance: monitoring the quantity of antimicrobials used in animal husbandry J.J. Webber 1 Agriculture Fisheries and Forestry, G.P.O. Box 858, Canberra ACT 2601, Australia Introduction There is world-wide concern about the potential of antimicrobial use in food animals to contribute to antimicrobial resistance problems in human and veterinary medicine. Data on antimicrobial use in food animals are essential in identifying such problems at the national level and in subsequent risk analysis, planning and execution of programmes to further define and address this concern. Reasons for collecting information on the quantities of antimicrobials used in animal husbandry The goal of any programme to monitor the quantities of antimicrobials used in animals is quantitative information on usage patterns by animal species and antimicrobial class, including potency and type of use that can be used to evaluate antimicrobial exposure. The level of information collected will depend on the member country’s overall concern, perceived or actual, with the issue of antimicrobial resistance, and its ability to fund the necessary programmes. These data can be used for: – risk analysis and planning – interpreting surveillance data on resistance – a precise and targeted response to problems of antimicrobial resistance – evaluating the effectiveness of prudent use guidelines for antimicrobials and strategies for mitigating resistance – tracking trends in animal antimicrobial use over time. The total consumption of antimicrobials for human medical, food animal and other uses is a key factor in any consideration of antimicrobial resistance. This guideline addresses antimicrobial use only in animals. However, for reasons of cost and administrative efficiency, member countries may wish to consider the collection of data on medical, food animal, agricultural and other antimicrobial use in the one programme. 1 Adapted from the OIE Guideline prepared by the OIE Ad hoc Group of Experts on antimicrobial resistance and authored by T. Nicholls, J. Acar, F. Anthony, R. Gupta, Y. Tamura, S. Thompson, E. Threlfall, D. Vose, M. van Vuuren, D. White, H. Wegener & M. Costarrica. OIE International Standards on Antimicrobial Resistance, 2003 123 2. Surveillance of antimicrobial consumption Sources of antimicrobial use data The sources of data available and options for their collection require careful consideration by individual OIE member countries to ensure that resources are used in a cost effective manner to meet the objectives of national programmes. Basic sources of data Basic sources of data will vary from country to country and depend on factors such as whether a member country manufactures antimicrobials, exports, and/or imports antimicrobials, and whether or not there is an accessible and accurate bank of information held by the national regulatory authorities. Sources may include customs, import and export data, manufacturing and sales data. Direct sources of data Animal drug wholesalers, retailers, pharmacists, veterinarians, feed stores, feed mills and organised industry associations in member countries might be efficient and practical sources of data on antimicrobial use in animals. A possible mechanism for the collection of information is that the provision of appropriate information by manufacturers to the regulatory authority is a requirement of antimicrobial registration, provided commercial confidentiality requirements can be met. End use surveys (veterinary surgeons and food animal producers) Periodic audits and statistically based surveys of either direct or end use sources of animal antimicrobials, rather than ongoing data collection programmes, may be a method of obtaining more accurate and detailed information on animal antimicrobial use. In addition to assisting in quantifying the use of antimicrobials, end use surveys, particularly of veterinarians and food animal producers, may be used to identify: – patterns of use that may have implications in epidemiological investigation of the development of antimicrobial resistance – related aspects of antimicrobial use on farms, such as methods of disposal of unused product. Collection, storage and processing of data from end use sources are likely to be inefficient and expensive unless carefully designed and well managed, but should produce accurate and targeted information. Categories of data Minimal antimicrobial use data requirements and data levels In OIE member countries, the minimal data collected should be the annual weight in kilograms of the active ingredient of the antimicrobial(s) used in food animal production. If a member country has the infrastructure for capturing basic animal antimicrobial use data for a specific antimicrobial, then additional information can be considered to cascade from this in a series of subdivisions or levels of detail. The 124 OIE International Standards on Antimicrobial Resistance, 2003 2. Surveillance of antimicrobial consumption relevant authorities within the member country should decide on the level of detail required. Such a cascade of levels could include: – The absolute amount in kilograms of antimicrobial active used per antimicrobial family per year, or for specific antimicrobial chemical entity when this information is required. – Therapeutic and growth promotion use in kilograms of the specific antimicrobial active. – Subdivision of antimicrobial use into therapeutic and growth promotion use by species. – Subdivision of the data into the route of administration, specifically in-feed, inwater, injectable, oral, intramammary, intra-uterine and topical. – Further subdivision of these figures by season and region by a member country may be useful in countries with large variations in environmental/management conditions within its borders, or where animals are moved from one locality to another during production. – Further breakdown of data for analysis of antimicrobial use at the regional, local, herd and individual veterinarian level may be possible using veterinary practice computer management software as part of specific targeted surveys or audits. Registration and regulation All OIE member countries should have appropriate veterinary chemical registration standards to ensure that safe, efficacious and quality veterinary products are used in food animals. Many member countries also have agriculture and veterinary chemical residue monitoring programmes for measuring chemical residues in food. These two activities can guide members on what level of detail of animal antimicrobial use information may be required. If the antimicrobial residue programme or other information indicated that non-registered antimicrobials were used in food animals, then provisions could be made to collect animal antimicrobial use information at the end user level using targeted surveys or specially designed monitoring programmes. Classes of antimicrobials Ideally, animal use data should be collected for each individual member of the antimicrobial group registered for use. Where common mechanisms of action exist data analysis could be enhanced if potency could be standardised in such a way that relative antimicrobial exposure (i.e. selection pressure) can be considered in risk analysis. For example chlortetracycline is not as potent as doxycycline on a mg/kg basis. Future comparison of use data would be facilitated by an internationally accepted method of comparing antimicrobial consumption and of an internationally accepted nomenclature for antimicrobial classes. Species, enterprise, regional and seasonal data Most countries register animal use antimicrobials for a specific food animal species and often for specific diseases. Frequently antimicrobial product registration is for OIE International Standards on Antimicrobial Resistance, 2003 125 2. Surveillance of antimicrobial consumption multiple species use, such as for cattle, sheep and goats, and this may create difficulties in determining use patterns. In order for a country to effectively analyse animal antimicrobial use patterns, including off-label use, a good understanding of the circumstances of food animal antimicrobial use is required e.g. cattle are raised in extensive range conditions, in feedlots or held in barns over winter. An understanding of what antimicrobials are used in specific animal species and industries in different regions, as well as seasonal influences on disease prevalence is likely to be important information in the risk analysis of this issue. Other important information If an OIE member country is considering animal antimicrobial use in food animals, a breakdown of the animal industries may be useful in any risk analysis or for comparison of animal antimicrobial use with human medical use within and between countries. For example, the total number of animals [cattle (meat, dairy, draft), sheep (meat, fibre and dairy)] in the country would be essential information. In addition, the weight in kg for food production per year would be essential information in the assessment of animal antimicrobial use figures. Breakdown of the type of production enterprises (extensive versus intensive for example) would also be useful if accurate industry enterprise antimicrobial use data was not available to give an indication of how animal antimicrobials were being used. Future directions Since the crude amount of antimicrobial (in kilograms) used yearly only indirectly represents antimicrobial exposure, and hence the selective pressure on bacterial populations, more sophisticated measures are needed. Medical monitoring of antimicrobial use in community and hospital medicine has led to the evolution of the concept of the Defined Daily Dose (DDD) 2 expressed as the DDDs/1,000 population/day. This approach takes into account the activity and potency of individual antimicrobials. Developing a DDD approach to antimicrobials in food animals would be difficult because of the wide range of animal weights. The DANMAP 3 reports use the concept of milligram of antimicrobial used per kilogram of meat produced. This concept could be useful in the monitoring and analysis of animal use of antimicrobial overall, and within particular species, although it does not address dose rate and length of treatment regimens in specific animal husbandry circumstances. Direct comparison between medical and animal use of antimicrobials is difficult and perhaps pointless other than for medical and veterinary authorities to have a rational WHO Collaborating Centre for Drug Statistics Methodology. http://www.whocc.no/atcddd/ DANMAP 2001. Use of antimicrobial agents and occurrence of antimicrobial resistance in bacteria from food animals and food and humans in Denmark. ISSN 1600-2032. http://www.vetinst.dk/high_uk.asp?page_id=179 2 3 126 OIE International Standards on Antimicrobial Resistance, 2003 2. Surveillance of antimicrobial consumption baseline measure of national antimicrobial use for ongoing comparison over a period of years. Conclusions Data on the use of antimicrobials in animals is essential for risk analysis and the design and planning of antimicrobial resistance monitoring and surveillance programmes, as well as for the ongoing management of antimicrobial resistance on the individual farm and at district, regional, national and international levels. __________ OIE International Standards on Antimicrobial Resistance, 2003 127 2. Surveillance of antimicrobial consumption The global usage of antimicrobials for animal health Dr T. Mudd IFAH, rue Defacqz 1, 1000 Bruxelles, Belgium The International Federation for Animal Health (IFAH) is the global organisation representing the animal health industry. The primary goal of the organisation is the development of pharmaceuticals and vaccines to ensure animal health and welfare. IFAH has three primary tasks involving Codex Alimentarius, the Veterinary International Committee for Harmonisation of Regulations (VICH) and Antimicrobial Resistance. IFAH has been supportive of the Codex approach to Risk Assessment and agrees that this must be an integral part of the veterinary medicines regulatory process. For this reason, IFAH has participated in the discussions at the level of the Codex Committee on general principles whereby a standard for international use of the risk analysis process is the objective. Over the past twelve years the market for animal health products has evolved from about US$9 billion to a peak of about US$11.5 billion in the late 1990s. Since that time there is evidence that total sales have flattened out or even declined. This suggests that very few new products are coming to the market and it has been essential to ensure that the existing products maintain their efficacy as long as possible. For this reason the members of IFAH have a direct interest in ensuring that resistance to antimicrobials is minimised. By product category biologicals or vaccines comprise 21%, medicinal feed additives 16% and pharmaceuticals, including antimicrobials 64%. This pharmaceuticals category is further split to 18% antiinfectives, 27% parasiticides, 3% performance enhancers and 16% other pharmaceuticals. Global markets are North America 35%, West Europe 26%, Far East 19%, Latin America 13%, East Europe 4% and rest of the world 3%. By species, the largest market now is for companion animals at 34%, next, cattle at 29%, swine 17%, poultry 14% and sheep 6% (2). There are three major areas of usage of antimicrobials in veterinary medicine: therapeutic, prophylactic and digestive/performance enhancement, also known as growth promoters. Regulatory agencies and their independent advisory committees conduct a risk assessment prior to introduction on the market and whenever new information becomes available. It is unfortunate that political interference in recent years has overridden scientific opinion and decisions taken based on the ‘precautionary principle’. IFAH has worked with the WHO in recent years culminating in the publication of the ‘Global Principles for the Containment of Antimicrobial Resistance due to Antimicrobial Use in Animals intended for Food’. This is part of the WHO Global Strategy whose purpose is ‘To minimise the public 128 OIE International Standards on Antimicrobial Resistance, 2003 2. Surveillance of antimicrobial consumption health impact of the use of antimicrobial agents in food-producing animals whilst at the same time providing for their safe and effective use in veterinary medicine’. The quantities of antimicrobial usage are frequently expressed in terms of US$. To obtain some direct comparison with other countries that express quantities as kg active ingredient (kg a.i.) a conversion factor has been used to express all usage in kg a.i. Medicinal feed additives such as ionophores for the control of coccidiosis, rather than as an antimicrobial are not included, unless as stated. In Latin America, as elsewhere, the livestock populations are a good index of the usage of antimicrobials. Thus, Brazil is by far the greatest user at 1,200 tonnes of a.i. Next is Mexico at about 400 tonnes followed by Colombia and Peru at about 150 tonnes and 100 tonnes respectively. Argentina and Venezuela are about 80 tonnes and 50 tonnes respectively, whereas Chile has a lower rate of use at about 20 tonnes. In Asia, values for China are at best an estimate but are in the range around 1,500 tonnes followed by Japan at 1,100 tonnes. The latter is probably an overestimate since the prices of products to the end user are higher than most other countries due to the distribution chain. The quantities for India and Pakistan are 400 tonnes and 200 tonnes respectively, followed by Malaysia and Bangladesh at 150 tonnes and 100 tonnes. Indonesia has a lower rate of use at about 20 tonnes. Other countries in the Far East and Australasia include South Korea at 550 tonnes, Thailand at 420 tonnes, Philippines at 350 tonnes and Taiwan at 180 tonnes. Australia and New Zealand are about 200 tonnes and 50 tonnes respectively. Vietnam usage is similar to New Zealand at about 50 tonnes. Two surveys of usage of antimicrobials in Europe were conducted on behalf of the animal health industry in 1997 and 1999. These data are shown in Table I. Table I European usage of antimicrobials 1997/1999 Year Total tonnes a.i Human use Veterinary therapy Growth promoter 1997 12,752 1999 13,152 7,659 60% 8,525 65% 3,494 27.5% 3,827 29% 1,599 12.5% 800 6% Growth promoter usage has halved in the period but there is a disturbing trend for increased use for veterinary therapy. The data from Denmark (3) show substantial increase of use of therapeutics associated with increased disease in weaner pigs following the withdrawal of growth promoters. A recent survey in the US by the Animal Health Institute estimates a use for disease prevention and treatment at 8,080t and for performance enhancement 1,200 tonnes. Together these account for about one-third the total use of antimicrobials in the US OIE International Standards on Antimicrobial Resistance, 2003 129 2. Surveillance of antimicrobial consumption when human use is considered. Animal usage in Canada is approximately 10% of that in the USA. In conclusion, the usage of antimicrobials for animal populations appears to be around one-third the quantities used for humans. In general, these antimicrobials tend to be those which are not widely used in man. Many, such as the tetracyclines, were developed over fifty years ago and their potency is considerably less per unit of a.i. compared with the modern antimicrobials used in man. This needs to be considered when comparing quantities on a kg or tonne a.i. The regulatory process for veterinary products includes risk assessment and precautionary measures. A comment from the University of Harvard (1) is relevant in this respect ‘Countries that do not invest in scientific enquiry are likely to misuse the precautionary principle to unduly control or restrain technological change’. IFAH worked with the World Veterinary Association and the International Federation of Agricultural Producers in the late 1990s to produce a set of global principles for the prudent use of antimicrobials in animals. It is encouraging that since then these have been adopted by many regional and national agencies. Increasing populations, particularly in developing countries, have a desperate need for adequate supplies of healthy affordable food that can only be supplied from healthy animals. References 1. Anon. (2000). – World chemical news. University of Harvard. 2. Anon. (2001). – IFAH Annual Report 2000. Wood Mackenzie. 3. DANMAP (2001). – Use of antimicrobial agents and occurrence of antimicrobial resistance in bacteria from food animals, foods and humans in Denmark. DANMAP, Cpenhagen, 68 pp. __________ 130 OIE International Standards on Antimicrobial Resistance, 2003 3. Risk analysis OIE International Standards on Antimicrobial Resistance, 2003 131 3. Risk analysis Antimicrobial resistance: risk analysis methodology for the potential impact on public health of antimicrobial resistant bacteria of animal origin D. Vose (1), J. Acar (2), F. Anthony (3), A. Franklin (4), R. Gupta (5), †T. Nicholls (6), Y. Tamura (7), S. Thompson (8), E.J. Threlfall (9), M. van Vuuren (10), D.G. White (11), H.C. Wegener (12) & M.L. Costarrica (13) (1) David Vose Consulting, Le Bourg, 24400 Les Lèches, France (2) Université Pierre et Marie Curie, Service de Microbiologie Médicale, Fondation Hôpital Saint-Joseph, 185 rue Raymond Losserand, 75674 Paris Cedex 14, France (3) Fresh Acre Veterinary Surgery, Flaggoners Green, Bromyard, Herefordshire HR7 4QR, United Kingdom (4) The National Veterinary Institute (SVA), Department of Antibiotics, SE 751 89 Uppsala, Sweden (5) College of Veterinary Sciences, Veterinary Bacteriology, Department of Microbiology, G.B. Pant University of Agriculture and Technology, Pantnagar 263 145 Uttar Pradesh, India (6) National Offices of Animal and Plant Health and Food Safety, Animal Health Science and Emergency Management Branch, Department of Agriculture, Fisheries and Forestry, P.O. Box 858, Canberra ACT 2601, Australia (7) National Veterinary Assay Laboratory, Ministry of Agriculture, Forestry and Fisheries, 1-51-1 Tolura, Kokubunji, Tokyo 185-8511, Japan (8) Joint Institute for Food Safety Research, Department for Health and Human Services Liaison, 1400 Independence Avenue, SW, Mail Stop 2256, Washington, DC 20250-2256, United States of America (9) Public Health Laboratory Service (PHLS), Central Public Health Laboratory, Laboratory of Enteric Pathogens, 61 Collindale Avenue, London NW9 5HT, United Kingdom (10) University of Pretoria, Faculty of Veterinary Science, Department of Veterinary Tropical Diseases, Private Bag X04, Onderstepoort 0110, South Africa (11) Centre for Veterinary Medicine, Food and Drug Administration, Office of Research, HFV-530, 8401 Muirkirk Road, Laurel, Maryland 20708, United States of America (12) World Health Organization, Detached National Expert, Division of Emerging and Transmissible Diseases, Animal and Food-related Public Health Risks, 20 avenue Appia, 1211 Geneva, Switzerland (13) Food and Agriculture Organization, Food Quality and Standards Service, Senior Officer, via delle Terme di Caracalla, 00100 Rome, Italy This report, prepared by the OIE Ad hoc Group of experts on antimicrobial resistance, has not yet received the approval of the International Committee of the OIE Summary The Ad hoc Group of experts on antimicrobial resistance, appointed by the OIE (World organisation for animal health), has developed an objective, transparent and defensible risk analysis process, providing a valid basis for risk management decisions in respect to antimicrobial resistance. The components of risk analysis and of different possible approaches in risk assessment (qualitative, semi-quantitative and quantitative) are defined. The Ad hoc Group recommended the following: an independent risk assessment based on scientific data; an iterative risk analysis process; a qualitative risk assessment systematically undertaken before considering a quantitative approach; the establishment of a risk assessment policy; and the availability of technical assistance for developing countries. OIE International Standards on Antimicrobial Resistance, 2003 133 3. Risk analysis Keywords Antimicrobial resistance – Containment of resistance – Food – Human medicine – International standards – Public health – Risk analysis – Risk assessment – Risk management – Veterinary medicine. Introduction This document presents the concept of risk analysis, comprising the components of hazard identification, risk assessment, risk management and risk communication, as applicable to antimicrobial resistance. The inter-relationship of these components is described and the respective distinct responsibilities of risk assessors and risk managers are identified. An example of a risk analysis methodology is given both in relation to animal health and to human health. Background Use of antimicrobials in animals for therapeutic, preventative and growth promotion purposes can reduce the therapeutic value of antimicrobials used in animal and human medicine because of losses in susceptibility of pathogenic bacteria. This risk may be represented by the loss of therapeutic value of one or several antimicrobial drugs and includes the emergence of multi-resistant bacteria. The principal aim of risk analysis of antimicrobial resistance in bacteria from animals is to provide Member Countries of the OIE (World organisation for animal health) with an objective and defensible method of assessing and managing the human and animal health risks associated with the development of resistance due to the use of antimicrobial drugs in animals, including appropriate communication measures. The procedure should be transparent and clearly separate responsibilities in risk assessment and risk management. Risk assessment should be based on the available scientific data. Transparency is essential because data are often uncertain or incomplete, and without full documentation, the distinction between facts and value judgements may not be clear. Risk management should also be a structured approach so that all stakeholders (for example, agricultural and pharmaceutical industries, healthcare providers and consumer groups) are provided with clear reasons for the imposition of risk management controls (for example, on the animal use of the antimicrobial in question, more stringent slaughtering or processing requirements, or import restrictions on products from animals that have been treated with antimicrobials). A policy framework for the authority regulating antimicrobials should be established to provide risk managers and risk assessors with a consistent set of legal, regulatory and political rules within which risk analyses must be conducted. This Guideline explains the recommendations of the OIE Ad hoc Group on antimicrobial resistance for guidelines and principles for conducting transparent, objective and defensible risk analyses to control the impact of using antimicrobials in animals, and provides recommended definitions of terms used in risk analysis. 134 OIE International Standards on Antimicrobial Resistance, 2003 3. Risk analysis Two principal sets of terminology are currently in use in risk analysis relating to this topic, namely: the United States (US) National Academy of Science (NAS) system on which the Codex Alimentarius Commission (Codex) approach is based, developed for food safety issues, and the Covello-Merkhofer system on which the OIE International Animal Health Code risk analysis is based. Beyond their apparent differences, both systems are very similar and largely contain the same components. The way these components are ordered in each of these two systems has evolved because of the type of risks that are being addressed. The terminology presented in this document follows the Covello-Merkhofer system. Comparison between the two systems and definitions of terms are given in Appendix C. The risk analysis process Risk analysis is defined in the OIE Code as ‘The process composed of hazard identification, risk assessment, risk management and risk communication’. It is a term frequently used to describe the complete process of properly addressing a risk issue. It encompasses assessing and managing the risk together with all the appropriate communication between risk assessors, stakeholders and risk managers. A typical risk analysis proceeds as detailed below. a) A policy framework will previously have been established by risk managers that describes the types of risk that need to be addressed, implying, among other things, the ranking of these risks among the other risk issues. In consultation with technical experts and risk assessors, a strategy for the assessment of the risk is then formulated. The policy framework also provides an explanation of the type of risk management options that can be considered under the legislative and regulatory framework of the country. Finally, the policy framework should explain the risk decision-making process, including methods of evaluating and quantifying risks and the level of risk deemed to be acceptable. b) A risk issue and plausible risk management actions that could be taken to reduce or eliminate the risk are identified by management. c) In consultation with technical experts, risk assessors and other stakeholders, a strategy for a preliminary assessment of the risk is formulated, including precisely how the risk is to be evaluated. d) Risk assessors execute a preliminary qualitative assessment (scoping study) and advise management on the feasibility of assessing quantitatively the risk and on the identified risk management strategies. This report is made public. e) Managers will determine from this scoping study whether the risk is sufficiently severe to warrant further action, including whether resources (which could be very limited) can be dedicated to the issue. If the risk is considered sufficiently important, and if feasible, risk managers may then instruct risk assessors to fully assess the risk (qualitatively, and/or quantitatively) and the reduced level of risk that would exist after each identified risk reduction option. Refining of the risk reduction options and risk assessment may go through several iterations. OIE International Standards on Antimicrobial Resistance, 2003 135 3. Risk analysis f) The risk assessment may be presented for review at various stages until the final risk assessment report has been produced, which is then made public. This aspect of risk communication is particularly helpful in ensuring transparency of the risk analysis as a whole and the efficient collection of data. g) Risk managers use the results of the risk assessment in order to determine, in line with previously defined policy, the appropriate actions to take in order to manage the risk in question in the most efficient manner. h) The risk management decision by a regulatory authority is made public with the greatest possible clarity. i) The risk managers have to implement their decision and to organise the followup of these regulatory and other measures in order to evaluate the impact of these decisions with regard to the expected results. j) The data acquired by the follow-up must be assessed in order to allow a possible amendment of the risk analysis policy, of the assessment strategy, of the outcome of the scientific assessment, and of the regulatory and other actions that have been taken. The following sections elaborate on these stages, categorised into four parts according to the Covello-Merkhofer system. References refer to where in the above bullet points each stage appears: – hazard identification (b) – risk assessment (c, d, e, f) – risk management (b, g, i, j) – risk communication (c, d, f, h). Hazard identification Hazard identification is defined under the OIE system as ‘The process of identifying the pathogenic agents that could potentially be introduced in the Commodity considered for importation’. It is the identification of ‘risk agents’ (hazards) and the conditions under which they might potentially produce adverse consequences. In terms of risk issues related to antimicrobial-resistant bacteria, the risk agent is most generally represented by the resistance determinant that emerges as a result of the use of a specific antimicrobial in animals. This definition then reflects the development of resistance in a species of bacterium that is pathogenic, as well as the development of a resistance determinant that may be passed to other bacteria that are pathogenic. The conditions under which the risk agent might potentially produce adverse consequences include any feasible scenarios via which humans or animals become exposed to pathogens which contain that resistance determinant, fall ill and where the human or animal would be treated with an antimicrobial that is no longer effective because of the resistance. 136 OIE International Standards on Antimicrobial Resistance, 2003 3. Risk analysis Risk management Risk management policy Risk management policy is a new term defined as ‘The regulatory policy framework for monitoring, measuring, assessing and managing risks involved in the use of antimicrobials in food producing animals’. A critical precursor to the risk analysis process is the development and public explanation of such a policy framework. This framework, aimed at providing the guidelines for conducting an appropriate risk assessment, has to be developed by the risk managers with the technical support of the scientific experts in charge of the risk assessment. The policy framework explains the philosophy behind monitoring and controlling risks involved in the use of antimicrobials in food producing animals. It must explain methods for involving risk assessment in the approval of new drug use, the various restrictions of use that might be applied to control and reduce any adverse impact and the procedure for retracting approval of use of the drug. It must also explain how the human or animal impact due to resistance will be measured, what level of impact will be considered unacceptable and how this information is used in the registration of new drugs. The policy framework may also address the additional importance of certain antimicrobial drugs needed to treat infectious diseases in human medicine for which there are no effective alternative therapies. Furthermore, it should explain the range of risk reduction actions that management can select within legislative and regulatory restrictions. The framework should explain the impact of uncertainty on the risk management decision. It should also address what actions will be taken in the event of identifying an unquantifiable risk due to antimicrobial use. The establishment of a population of resistant bacteria as a result of the use of an antimicrobial in animals means that the human or animal health impact may continue long after the animal use of an antimicrobial has ceased. The policy framework should therefore address how to measure a long-term impact, and may include some cut-off period or discount factor that recognises the reduced value of a therapeutic drug as new drugs become available. However, the policy framework should not necessarily restrict risk management from considering potential risk management options that may be outside the current domain of the regulatory authority. Clear explanation of these conditions allows the pharmaceutical and agricultural industries and the veterinary and healthcare professional bodies to plan and test current and future antimicrobial products in a predictable environment and modify their use to achieve clear objectives. Clearly stating the policy framework ensures transparency during the risk management phase of a risk analysis. People react to risk in very different and often emotional ways: a clear policy on how to measure risk and what is deemed acceptable implicitly OIE International Standards on Antimicrobial Resistance, 2003 137 3. Risk analysis recognises that a zero risk policy is unachievable and greatly reduces any suspicion of false argument. Risk management components Risk management is conducted by risk managers who have a comprehensive understanding of policy, and an appropriate level of technical background to communicate effectively with the risk assessors. The OIE defines risk management as consisting of the steps described below. Risk evaluation The process of comparing the risk estimated in the risk assessment with the appropriate level of protection of the Member Country. Option evaluation The process of identifying, evaluating the efficiency and feasibility of, and selecting measures in order to reduce the risk associated with an importation in line with the appropriate level of protection of the Member Country. The efficacy is the degree to which an option reduces the likelihood and/or magnitude of adverse biological and economic consequences. Evaluating the efficacy of the options selected is an iterative process that involves their incorporation into the risk assessment followed by comparison of the resulting level of risk with that considered acceptable. The evaluation for feasibility normally focuses on technical, operational and economic factors affecting the implementation of the risk management options. Implementation The process of following through with the risk management decision and ensuring that the risk management measures are in place. Monitoring and review The ongoing process by which the risk management measures are continually audited to ensure that they are achieving the results intended. Risk decision when data are insufficient or inadequate In the event that insufficient or inadequate data are available to reasonably assess the importance of a potential risk issue, and it is considered that the risk is potentially of such severity that one cannot wait for sufficient data before taking action, it is reasonable for the risk managers to take a temporary risk avoidance action that minimises any exposure to the risk. There are five extremely important considerations when faced with this situation, as follows: a) a risk assessment must first be attempted, and all reasonable efforts made to acquire the necessary data, within the allowable timeframe, before taking the temporary risk avoidance action b) the risk avoidance action must be chosen to provide the required level of protection in the manner least restrictive to trade 138 OIE International Standards on Antimicrobial Resistance, 2003 3. Risk analysis c) the risk avoidance action should be commensurate with the potential severity of the risk d) in all cases, particularly in international trade, the risk avoidance action should be taken in conjunction with a commitment to acquire the necessary data, within a reasonably short and defined time, to help assess the severity of the risk and the most appropriate risk reduction strategy e) the process must remain transparent. Risk assessment Risk assessment is defined in the OIE Code as ‘The evaluation of the likelihood and the biological and economic consequences of entry, establishment, or spread of a pathogenic agent within the territory of an importing country’. There are a number of approaches to assessing the magnitude of a risk and the value of potential risk reduction options. These can be broadly categorised into three types: qualitative, semiquantitative and quantitative risk assessments. Whichever approach is taken, the risk assessment must be designed to address the specific question posed by the risk managers. The risk assessment process is usually sub-divided into four components: risk release assessment; exposure assessment; consequence assessment; and risk estimation. Their meanings are described below and examples of factors that may be considered in each component are listed in Appendices A and B. Release assessment Defined in the OIE Code as ‘Description of the biological pathways necessary for the use of an antimicrobial in animals to release resistant bacteria or resistance determinants into a particular environment, and estimating the probability of that complete process occurring either qualitatively or quantitatively’. Exposure assessment Defined in the OIE Code as ‘Describing the biological pathways necessary for exposure of animals and humans to the hazards released from a given source, and estimating the probability of the exposure occurring, either qualitatively or quantitatively’. Consequence assessment Defined in the OIE Code as ‘Description of the relationship between specified exposures to a biological agent and the consequences of those exposures. A causal process must exist by which exposures produce adverse health or environmental consequences, which may in turn lead to socio-economic consequences. The consequence assessment describes the potential consequences of a given exposure and estimates the probability of them occurring. This estimate may be either qualitative or quantitative’. OIE International Standards on Antimicrobial Resistance, 2003 139 3. Risk analysis Risk estimation Defined in the OIE Code as ‘Integration of the results from the release assessment, exposure assessment, and consequence assessment to produce overall measures of risks associated with the hazards identified at the outset. Thus risk estimation takes into account the whole of the risk pathway from hazard identified to unwanted outcome’. The policy framework will provide guidelines to the risk assessors on how to assess the complete impact of any risk issue and risk reduction strategies. For example, removing an antimicrobial from veterinary use may mean that another antimicrobial is used in its place with potentially worse consequences. Unless these secondary impacts, whether positive or negative, are addressed, the risk management strategy may be suboptimal. The initial planning stages of a risk assessment can be performed as described below. a) The risk issue in question is formally expressed to ensure that all participants agree on the problem to be addressed. The potential mechanisms and pathways via which the hazard can result in an adverse effect are also described. This system, as understood by the risk assessment team, can be explained using one or more flow diagrams. At this point, the diagram is purely conceptual and there is therefore no need for data. The purpose of such diagrams is to focus thought on what data would be useful, what possible risk management options exist, and to integrate and review the level of knowledge about the system in general. It is advisable to involve a broad participation in the exercise and to circulate widely to stakeholders and relevant experts. b) A preliminary data search is conducted to assess what components of the system might be adequately quantified. Components might include, for example: the prevalence of resistant bacteria in faeces, water or carcasses; the distribution by animal species, season and geographical region of use of an antimicrobial; the frequency of the use of the antimicrobial in human medicine and the health status of those receiving the antimicrobial. At this stage, it is sufficient to know of the availability of data. Requests for data that might help quantify the components of the system can also be made to stakeholders and relevant experts. Strong consideration should also be given to useful data that may not be immediately available, but that could become available within a reasonable period, perhaps with some research effort. The interpretation of what constitutes a reasonable period will reflect the imminence and severity of the risk issue in question. It may be appropriate to consider completing a risk assessment rapidly to help decision makers identify the immediate actions to be taken, recognising that a re-evaluation of the risk issue when more data become available may lead the decision makers to alter the preliminary actions that were taken. c) A review of the system, as perceived by the risk assessment team, together with the data available to quantify the components of that system can provide important guidance. It can illustrate which risk management options can be properly assessed for 140 OIE International Standards on Antimicrobial Resistance, 2003 3. Risk analysis their effectiveness. It can also guide the risk assessor regarding the production of a quantitative risk assessment, if required, that would be based on data as well as supplying guidance as to whether such a model could be validated in some way. It is the combination of feasible risk management options, together with the data that could be available to assess those options, that should direct the risk assessment team towards the form of their assessment. If the system is not sufficiently well understood, or insufficient data are available to meaningfully quantify the model, it may only be possible to produce a qualitative risk analysis. However, quantification of certain aspects of the system may also be possible, which could enable the evaluation of a restricted number of risk management options. The risk assessment model can be kept as simple as possible to support the range of risk management decisions being considered. The model structure may not include a complete pathway analysis of the risk scenario if there are limited risk reduction strategies the benefits of which can be addressed in a far simpler model. Flexibility in the approach to modelling will reduce the effort required to produce the assessment and limit the number and type of assumptions that may have to be made in the model. However, the model may not then be useful in addressing other questions that arise over the same risk issue and may not help other stakeholders contribute to efficiently managing the risk. It may also be difficult to demonstrate consistency between models where different model structures have been used together with quite different assumptions. A full assessment of the risk to human and animal health from antimicrobial-resistant bacteria resulting from use of antimicrobials in food-producing animals can be divided into three parts, as follows: a) production of the resistant bacteria of interest as a result of antimicrobial use, or more particularly, production of the resistant determinants if transmission is possible between bacteria. (If it is the use of the antimicrobial in animals that is being considered as the hazard, there may be several different species of bacteria to consider.) b) consideration of the realistic pathways via which humans can become exposed to these resistant bacteria or resistance determinants, together with the possible range of bacterial load ingested at the moment of exposure c) consideration of the response of the person to the exposure. Risk assessment of antimicrobial issues can be technically difficult, and it is essential that the assessment is the work of a team of professionals with broad expertise in risk analysis modelling, microbiology, veterinary medicine and animal husbandry, human healthcare and medicine, chemistry and any other relevant disciplines. Published chemical, microbial and genetic risk assessments can provide useful generic illustrations for modelling components of the risk assessment. Qualitative risk assessment Qualitative risk assessment is defined in the OIE Code as ‘An assessment where the outputs on the likelihood of the outcome or the magnitude of the consequence are expressed in qualitative terms such as high, medium, low or negligible’. A qualitative OIE International Standards on Antimicrobial Resistance, 2003 141 3. Risk analysis risk assessment is always completed first as part of a preliminary evaluation (scoping study), whether or not one progresses to a semi-quantitative or fully quantitative assessment. It is the collation of all available information that will enable the determination of the probability and impact of the risk in question. A qualitative risk assessment discusses the steps necessary for the risk to occur, which pathways are feasible and which can be logically discounted. In a risk assessment of a human health impact due to use of a specific antimicrobial in food producing animals, for example, factors would include patterns of use of the antimicrobial, rates of resistance acquisition in exposed bacteria, the ecology of these resistant bacteria, pathways via which these bacteria may directly or indirectly transfer resistance to pathogens that infect humans, and the rates at which antimicrobials analogue to the animal antimicrobial are prescribed for the infected humans. A qualitative risk assessment would also need to discuss the level of loss of benefit of the human medicine antimicrobial. All of these factors constitute a risk scenario on which one can overlay possible risk reduction strategies and discuss the benefits they might provide. Appendices A and B list factors that may be useful in an assessment. At this stage, a risk may be determined to be logically insignificant because, for example, the biological pathway is not possible or the risk is logically less severe than another for which a full analysis has been completed and determined to be acceptably small. As more risk assessments are conducted on antimicrobial issues, there may be broad agreement concerning the likely risks associated with particular hazards. In such cases, a qualitative assessment may frequently be the sole requirement. Qualitative assessment does not require mathematical modelling skills and so will often be the type of assessment used for routine decision-making. When all easily-obtainable information has been collected, a preliminary report to the risk managers is necessary to advise of any further information that will be needed to complete the picture, or perhaps any additional information that will be necessary to complete a more quantitative analysis. It should also be apparent at this stage whether data are or can be made available to assess each risk reduction strategy and communicating this to the risk managers enables them to assess which risk reduction strategies are worth pursuing in greater depth. Quantitative risk assessment Quantitative risk assessment is defined in the OIE Code as ‘An assessment where the outputs of the risk assessment are expressed numerically’. The purpose of quantitative risk assessment is to numerically evaluate the probability and impact(s) associated with a risk issue. Two principal mathematical approaches are feasible: the most common is to use a Monte Carlo simulation model to describe the risk event (the development of the hazard into an actual impact), together with its uncertainty (lack of knowledge) and variability (inherent randomness); the second method is to use the algebra of probability theory to produce a formulaic model of the risk event. Monte Carlo simulation is almost always preferred over algebraic methods because it is far simpler to execute, particularly with modern software. It offers greater modelling flexibility, is 142 OIE International Standards on Antimicrobial Resistance, 2003 3. Risk analysis easy to understand, check and explain, and less prone to human error in model development. However, Monte Carlo simulation of rare events can become onerous, in which case a combination of calculating some simpler parts of a risk scenario and simulating the remainder may sometimes prove more efficient. A quantitative risk assessment produces a mathematical model that estimates the effect of possible risk management actions. It may be desired that any possible action between and including production of the food animal and the final human health effect be evaluated quantitatively. If so, the quantitative risk assessment model must simulate all important microbial pathways between the farm and the exposed human or animal in sufficient detail to evaluate possible changes in the system as a result of a risk management action. For risk management purposes, it may only be necessary to evaluate changes in the human or animal health impact as a result of a risk management action, not the underlying base health risk, although it may be informative to be able to estimate the base health risk for other purposes. Thus, a full risk assessment model may need to consider a wide range of pathways. For example, Enterococcus faecium is a hardy organism that can survive for long periods outside its original host. Feasible pathways may include, for example, runoff from manure lagoons or fields sprayed with manure entering waterways used by swimmers, or the consumption of vegetables that have been grown in fields sprayed with manure. By contrast, these pathways would not be important for Campylobacter which succumb rapidly to changes in their environment. Failure to appreciate the range of pathwayscould lead to a misevaluation of the effect of some risk management action. For example, irradiation of poultry carcasses may be effective against Campylobacter if consumption of meat were to be considered the primary exposure pathway. However, irradiation might prove ineffective for E. faecium if the primary exposure pathway was from consumption of raw vegetables. Microbial food safety risk assessments have for some time attempted to model very similar risk issues to those posed by antimicrobial resistance. A variety of modelling techniques exists for microbial risk assessments, based around the principles of stochastic simulation of risk scenarios (14, 18, 19, 22). Spreadsheet models are generally used together with Monte Carlo simulation add-ins to create simulations of the entire ‘farm-to-fork’ continuum, finishing with the way in which the consumer is affected by consumption of the bacteria. Other commercially available dynamic simulation applications can achieve much the same effect. There are a variety of formula-based models available from the field of predictive microbiology to estimate the growth and attenuation of various bacteria when exposed for different amounts of time to different environments, particularly level of moisture, temperature and pH. Thus, a quantitative risk assessment combines probability mathematics (11, 17), usually from the binomial and Poisson processes, with empirical curve-fitting equations and sometimes theoretically based formulae from predictive microbiology, to attempt to characterise the exposure events. Microbial food safety models consider the redistribution, growth and attenuation of bacteria during the various actions in slaughtering, processing, food handling and cooking. For example, the microbial load OIE International Standards on Antimicrobial Resistance, 2003 143 3. Risk analysis on contaminated carcasses will be reduced drastically through correct handling, removal of the most contaminated parts of the carcass, scalding and washing. In contrast, cross-contamination between carcasses through aerosols, splashing, workers, etc., may mean that the proportion of contaminated carcasses leaving the slaughter plant is greater than the proportion of contaminated animals entering the plant. Much of the modelling principles necessary in antimicrobial resistance risk assessment parallel those used in microbial food safety risk assessment. At the time of writing (November 2000), very few antimicrobial resistance risk assessments have been published (http://www.fda.gov/cvm/fda/mappgs/antitoc.html; 23) but a significant number of microbial food safety risk assessments have been completed which provide practical illustrations of the techniques employed (2, 8; http://www.fsis.usda.gov/ophs/risk/index.htm; http://www.foodriskclearinghouse.umd.edu/risk_assessments.htm; http://www.fsis.usda.gov/OPHS/ecolrisk/home.htm; http://www.nal.usda.gov/fnic/foodborne/risk.htm). Microbial risk assessments typically use logarithmic scales in estimating the microbial load because of the range of numbers that can be involved and the multiplying nature of bacterial growth and attenuation. Subsequent estimations of the probability of infection, illness or perhaps death from specific exposures are made through doseresponse equations to produce a final estimate of the total human health impact. Risk assessments that model the complete microbial pathway from the farm to final ingestion are sometimes called ‘farm-to-fork’ or ‘farm-to-table’ risk assessments, though these are potentially misleading terms in cases where significant exposure pathways are associated with ingestion via other means (e.g. consumption of vegetables, ingestion through soil or water, and human-to-human or animal-to-human transmission). A full ‘farm-to-fork’ model invariably contains a host of potentially contestable assumptions because of the inherent complexity of the system being modelled and the gaps in knowledge of that system. It also relies a great deal on the validity of a dose-response model, the weaknesses of which are well known (21). In general, a risk assessment model should only be as complex as necessary to evaluate the risk management options available to the regulatory authority, therefore a full ‘farm-to-fork’ model may not be necessary. For example, the risk assessment completed by the United States Food and Drug Administration Center for Veterinary Medicine (USFDA-CVM) on the human health effect of fluoroquinolone-resistant Campylobacter (http://www.fda.gov/cvm/fda/mappgs/antitoc.html) considered only the effect of removal of fluoroquinolone use in poultry. This assessment avoided any modelling of the ‘farm-to-fork’ pathways. It estimated the number of human cases of campylobacteriosis that would have been affected by the fluoroquinolone-resistance from poultry, to provide an estimate of the current risk. The argument was that removing fluoroquinolone from poultry would have the effect of reducing the human impact by this amount, which was supported by the low survivability of Campylobacter outside its host, so resistance would rapidly disappear. The assessment then related this risk to the level of prevalence of fluoroquinolone-resistant Campylobacter 144 OIE International Standards on Antimicrobial Resistance, 2003 3. Risk analysis contaminated broiler carcasses at the end of the slaughter plant. The argument then presented was that changes in that prevalence and/or the load on the contaminated carcasses can be mapped to a corresponding change in the human health impact. The structure of models like this can be used very effectively in other countries, using data appropriate to that country, where similar assumptions would apply. All parameters in a quantitative risk assessment model must be quantified. The most transparent approach, least likely to attract criticism, is to use published data from peer-reviewed papers. However, such data will frequently not be available and reasonable surrogates may be used in their place, together with supporting arguments for the surrogacy. Expert opinion may also be used, but it is more transparent if any data from which the expert has based his or her opinion can be used in its place (12). Unpublished data from reliable sources may also be used. Regardless of the source, all data used in the risk assessment must be critically reviewed. A quantitative risk assessment must explicitly model the uncertainty associated with the model parameters using techniques like the bootstrap (5, 6), Bayesian inference (9, 20) and classical statistics (1, 10, 13). Bayesian inference is particularly useful at explicitly stating the contribution arising from observations, interpretation of those observations and any subjective estimation. Bayesian inference also allows the analyst to combine information from different sources, such as two different random surveys of a population for contamination with different test sensitivities and specificities. The results of the risk assessment are presented as a report to the risk managers, explaining the methods used, characterising the risk in appropriate terms according to policy, together with the benefits of any risk reduction strategies that could be assessed. All quantified terms should be reported with their uncertainties in an easily understandable form. The relative frequency distribution provides an excellent visual representation of the level of uncertainty, whilst cumulative distribution plots allow the risk manager to evaluate the risk at any desired level of confidence. Sensitivity analyses should be performed to determine the key uncertainty parameters of the model and illustrated using techniques such as spider plots and tornado charts. Key assumptions must also be explicitly described, together with a balanced argument of the reasoning for the assumptions and a discussion of the inaccuracy of the predictions of the model should those assumptions be false. This model uncertainty must be keenly analysed, and possible methods of validating assumptions must be considered, perhaps through scientific experiments or comparison with the experience of other nations. Inclusion and discussion of all types of uncertainty in the risk assessment report allow the risk managers to apply the appropriate level of conservatism in valuing the risk and any risk reduction options. It should be emphasised that failure to properly address uncertainty in the risk assessment report equates to an implicit value judgement of the risk that is not the remit of the risk assessor. OIE International Standards on Antimicrobial Resistance, 2003 145 3. Risk analysis Semi-quantitative risk assessment Semi-quantitative risk assessment is a new term defined as ‘An assessment where estimates of the likelihood of the outcome and the magnitude of the consequences are expressed in semi-quantitative terms via some scoring mechanism’. It will frequently not be possible to perform a complete quantitative risk assessment on each item in a portfolio of risk issues facing risk managers because of lack of appropriate data. In such circumstances, it would nonetheless be useful to have a method for comparing the magnitude of risks and the benefits of risk reduction strategies for those risks. Semi-quantitative risk assessment, when properly executed, is a transparent approach that supports the efficient management of a portfolio of risk issues without requiring complete quantification of the risks or excessive risk avoidance. Semi-quantitative risk assessment techniques are commonly used for risk analysis in commercial projects, but are currently not widely accepted in international risk issues because of the difficulty in retaining transparency and because the process is open to abuse without proper guidelines. The principle of semi-quantitative risk assessment (22) is initially to estimate the probability and size of the potential consequences into broad, but well-defined categories, then convert these estimates using a scoring system to produce a severity score for the risk. Various risk management options can be evaluated according to the degree to which they would reduce the severity score of the risk. The technique has a number of advantages, as follows: – the risks can be compared in a systematic fashion – a severity threshold can be set for unacceptable risk – an efficient and consistent policy framework can be developed which minimises the total severity scores for all risks given the resources available. Risk communication As defined in the OIE Code, ‘Risk communication is the interactive exchange of information on risk among risk assessors, risk managers and other interested parties’. There are many aspects to risk communication. Failure to pay proper attention to risk communication may easily result in failure of the risk analysis process. Both risk managers and risk assessors should be well versed in the concepts of risk analysis. The risk assessors should have a clear understanding of policy. Similarly, the risk managers should be fully conversant with the taxonomy and terminology of risk assessment and appreciate the level of effort and variety of disciplines involved in producing a reliable risk assessment. The goals of risk communication are the following: – to promote awareness and understanding of the specific issues under consideration during the risk analysis process, by all participants – to promote consistency and transparency in arriving at and implementing risk management decisions – to provide a sound basis for understanding the risk management decisions proposed or implemented 146 OIE International Standards on Antimicrobial Resistance, 2003 3. Risk analysis – to improve the overall effectiveness and efficiency of the risk analysis process – to strengthen working relationships and mutual respect among all participants – to promote the appropriate involvement of all stakeholders in the risk communication process – to exchange information on the knowledge, attitudes, values, practices and perceptions of stakeholders concerning the risks in question. The joint Food and Agriculture Organization (FAO)/World Health Organization (WHO) Expert Consultation on the application of risk communication to food standards and safety measures, held in 1998 in Rome, provides an in-depth discussion on the subject (7). Communication between assessors and managers Management must provide clear instructions for the risk issue that is to be analysed, together with the preferred method(s) of characterisation (e.g. person days of illness per year). Assessors must ensure that the managers have reasonable expectations of the assessment and may also advise of other potential information the assessment may provide that would help the management with their decision-making. There should be communication between the risk assessors and risk managers throughout the assessment process to ensure that the assessment is completed in a timely fashion and that the required resources are made available. Communication between assessors and stakeholders It is extremely helpful to widely publicise the intended method of assessment, including model structure and assumptions at the earliest possible opportunity, together with an expression of flexibility in the eventuality of any new information or ideas. This allows stakeholders to provide input, improves transparency of the process and improves support for the assessment and any resultant risk management decision. Communication between managers and stakeholders Risk managers will usually need to advise stakeholders of the intention to perform a risk analysis at the beginning of the project. At this stage, communication with stakeholders is an important opportunity to gather political and scientific support for the risk assessment, as well as a data gathering exercise. When the risk assessment has been completed, it is advisable to make the report publicly available with a reasonable comment period to ensure that there are no large errors in the assessment or additional data available. The World Wide Web is an excellent means for maximising the availability of the assessment and may include downloadable, self-contained versions of the risk assessment. Publishing comments received, together with any responses from the risk assessment and risk management teams, underlines the transparency of the process. These can be included in the final risk analysis document that explains the results of the risk assessment together with the risk management decision that has been made. OIE International Standards on Antimicrobial Resistance, 2003 147 3. Risk analysis Recommendations To effectively manage antimicrobial resistance risk issues, the OIE Ad hoc Group recommends that: – risk analysis should be conducted in an objective and defensible manner – the risk analysis process should be transparent and consistent – risk analysis should be conducted as an iterative and continuous process – risk management and risk assessment functions should be kept separate to ensure the independence of decision-making and evaluation of the risk – risk management should be conducted in reference to a policy framework setting out the domain of the regulator and the range of risk reduction actions that may be considered – the risk assessment should be based on sound science and conducted according to a strategy established by the risk managers in co-operation with the risk assessors – risk assessment requires a multidisciplinary team and should be conducted in broad consultation with available scientific expertise – qualitative risk assessment should always be undertaken, and provides information on whether progression to full quantitative risk assessment is feasible and/or necessary – risk assessment of antimicrobial resistance issues requires very specific, technical skills that may not be available to developing countries. The OIE and its Member Countries should work towards helping these countries to develop or access these skills, to ensure that risk assessment itself does not become a barrier to trade – communication between managers, assessors and stakeholders is essential. Effort should be made to establish such communication early in the process, to allow opportunity for responses, and should be continued throughout the risk analysis process. Antibiorésistance : méthodologie d’analyse du risque appliquée à l’impact potentiel sur la santé publique des bactéries d’origine animale résistantes aux antibiotiques D. Vose, J. Acar, F. Anthony, A. Franklin, R. Gupta, †T. Nicholls, Y. Tamura, S. Thompson, E.J. Threlfall, M. van Vuuren, D.G. White, H.C. Wegener & M.L. Costarrica Résumé Le Groupe ad hoc d’experts sur l’antibiorésistance créé par l’Organisation mondiale pour la santé animale a élaboré une procédure d’analyse du risque à la fois objective, transparente et justifiée, offrant une base valable pour les décisions de gestion du risque relatives à l’antibiorésistance. Les auteurs définissent les éléments constitutifs de l’analyse du risque et les différentes approches possibles de l’évaluation du risque (qualitative, semi-quantitative et quantitative). Les recommandations du 148 OIE International Standards on Antimicrobial Resistance, 2003 3. Risk analysis Groupe ad hoc portent sur les points suivants: évaluation du risque indépendante basée sur des données scientifiques; processus itératif d’analyse du risque; réalisation systématique d’une évaluation qualitative du risque avant toute approche quantitative; élaboration d’une politique d’évaluation du risque; enfin, prestation d’une assistance technique pour les pays en développement. Mots-clés Analyse du risque – Antibiorésistance – Denrées alimentaires – Évaluation du risque – Gestion du risque – Maîtrise de la résistance – Médecine humaine – Médecine vétérinaire – Normes internationales – Santé publique. Resistencia a los antimicrobianos: metodología de análisis de riesgos para determinar la eventual incidencia en la salud pública de bacterias de origen animal resistentes a los antimicrobianos D. Vose, J. Acar, F. Anthony, A. Franklin, R. Gupta, †T. Nicholls, Y. Tamura, S. Thompson, E.J. Threlfall, M. van Vuuren, D.G. White, H.C. Wegener & M.L. Costarrica Resumen El Grupo Ad hoc de expertos sobre la resistencia de las bacterias a los productos antimicrobianos, creado por la Organización mundial de sanidad animal, ha elaborado un proceso de análisis de riesgos objetivo, transparente y defendible, brindando con ello una sólida base para tomar decisiones de gestión de riesgos ligados a la resistencia a los antimicrobianos. Los autores exponen los elementos que configuran el análisis de riesgos y los distintos planteamientos que se pueden aplicar (cualitativo, semicuantitativo y cuantitativo). El Grupo Ad hoc recomendó los siguientes procedimientos: una evaluación de riesgos independiente y basada en datos científicos; un proceso iterativo de análisis de riesgos; una evaluación cualitativa sistemática previa a la eventual aplicación de un método cuantitativo; la definición de una política de evaluación de riesgos; y la prestación de asistencia técnica a los países en desarrollo. Palabras clave Alimentos – Análisis de riesgos – Contención de las resistencias – Evaluación de riesgos – Gestión de riesgos – Medicina humana – Medicina veterinaria – Normas internacionales – Resistencia a los productos antimicrobianos – Salud pública. Appendix A Risk assessment of human health impact due to the use of antimicrobials in animals The following lists, although not exhaustive, describe factors that may need consideration in a risk assessment of human health impact. OIE International Standards on Antimicrobial Resistance, 2003 149 3. Risk analysis Definition of the risk The infection of humans with bacteria that have acquired resistance to the use of a specific antimicrobial in animals, and resulting in the loss of benefit of antimicrobial therapy used to manage the human infection. Hazard identification Two types of hazard exist, as follows: – bacteria that have acquired resistance due to the use of a particular antimicrobial in animals – resistance determinants selected as a result of the use of a particular antimicrobial in animals. The identification of the hazard must include considerations on the class or subclass of antimicrobial. Release assessment Release assessment consists of describing the biological pathways necessary for the use of a specific antimicrobial in animals to lead to the release of resistant bacteria or resistant determinants into a particular environment, and estimating the probability of that complete process occurring either qualitatively or quantitatively. The release assessment describes the probability of the release of each of the potential hazards under each specified set of conditions with respect to amounts and timing, and how these might change as a result of various actions, events or measures. Examples of the kind of inputs that may be required in the release assessment are as follows: – species of animal treated with the antimicrobial in question – number of animals treated, geographical distribution of those animals – variation in methods of administration of the antimicrobial – bacteria developing resistance as a result of the antimicrobial use – mechanism of direct or indirect transfer of resistance – capacity of resistance transfer (chromosomes, plasmids) – cross-resistance and/or co-resistance with other antimicrobials – surveillance of animals, animal products and waste products for the existence of resistant bacteria. Exposure assessment Exposure assessment consists of describing the biological pathways necessary for exposure of humans to the resistant bacteria or resistance determinants released from a given antimicrobial use in animals, and estimating the probability of the exposures occurring, either qualitatively or quantitatively. The probability of exposure to the identified hazards is estimated for specified exposure conditions with respect to amounts, timing, frequency, duration of exposure, routes of exposure and the number, 150 OIE International Standards on Antimicrobial Resistance, 2003 3. Risk analysis species and other characteristics of the human populations exposed. Examples of the kind of inputs that may be required in the exposure assessment are as follows: – human demographics and consumption patterns, including traditions and cultural practices – prevalence of food and/or the animal environment contaminated with resistant bacteria – prevalence of animal feed contaminated with resistant bacteria – microbial load in contaminated food at the point of consumption – survival capacity and redistribution of resistant bacteria during the agrofood process (including slaughtering, processing, storage, transportation and retailing) – disposal practices for waste products and the opportunity for human exposure to resistant bacteria or resistance determinants in those waste products – point of consumption of food derived from the food-producing animal (professional catering, home cooking) – variation in consumption and food-handling methods of sub-populations – capacity of resistant bacteria to settle in human intestinal flora – human-to-human transmission of the bacteria under consideration – capacity of resistant bacteria to transfer resistance to human commensals – exposure to resistance determinants from other sources – amount of antimicrobials used in response to human illness – dose, route of administration (oral, injection) and duration of human treatment – pharmacokinetics (metabolism, bioavailability, access to intestinal flora). Consequence assessment Consequence assessment consists of describing the relationship between specified exposures to resistant bacteria or resistance determinants and the consequences of those exposures. A causal process must be believed to exist by which exposures produce adverse health or environmental consequences, which may in turn lead to socio-economic consequences. The consequence assessment describes the potential consequences of a given exposure and estimates the probability of them occurring. This estimate may be either qualitative or quantitative. Examples of consequences include the following: – dose-response relationships – variation in susceptibility of sub-populations – variation and frequency of human health effects resulting from loss of efficacy of antimicrobials – changes in human medicine practices resulting from reduced confidence in antimicrobials OIE International Standards on Antimicrobial Resistance, 2003 151 3. Risk analysis – changes in food consumption patterns due to loss of confidence in the safety of food products and any associated secondary risks – associated costs – interference with a classical first line antibiotherapy in humans – perceived future of the drug (time reference). Risk estimation Risk estimation consists of integrating the results from the release assessment, exposure assessment and consequence assessment to produce overall measures of risks associated with the hazards identified at the outset. Thus, risk estimation takes into account the whole of the risk pathway from the hazard identified to the unwanted outcome. For a quantitative assessment, the final outputs may include the following: – number of people falling ill – increased severity or duration of disease – number of person/days of illness per year – deaths (total per year; probability per year or lifetime for a random member of the population or a member of a specific more exposed sub-population) – importance of the pathology caused by the bacteria – absence of alternate antibiotherapy – level of resistance observed in humans – some arbitrary scale of impact to allow weighted summation of different risk impacts (e.g. illness and hospitalisation). Risk management options to evaluate The following risk management measures could be implemented: – decision not to grant a licence for use of a new antimicrobial – review of licence authorisation and label indications – revoking of licence – restrict use of antimicrobial (e.g. in particular industries, therapeutic only) – review of prudent use guidelines – establish monitoring of veterinary use of antimicrobials – revision of treatment guidelines. Appendix B Risk assessment of impact on animal health due to the use of antimicrobials in animals The following lists, though not exhaustive, describe factors that may need consideration in a risk assessment of animal health impact. 152 OIE International Standards on Antimicrobial Resistance, 2003 3. Risk analysis Definition of the risk The infection of animals with bacteria that have gained resistance from the use of a specific antimicrobial in animals, and resulting in the loss of benefit of antimicrobial therapy used to manage the animal infection. Hazard identification Possible hazards are as follows: – bacteria that have acquired resistance due to the use of a particular antimicrobial in animals – resistance determinants selected as a result of the use of a particular antimicrobial in animals. The identification of the hazard must include consideration of the class or subclass of antimicrobial. Release assessment Examples of the type of inputs that may be required in the release assessment are as follows: – species of animal treated with the antimicrobial in question – number of animals treated, geographical distribution of those animals – variation in methods of administration of the antimicrobial – bacteria developing resistance as a result of the antimicrobial use – mechanism of direct or indirect transfer of resistance – capacity of resistance transfer (chromosomes, plasmids) – cross-resistance and/or co-resistance with other antimicrobials – surveillance of animals, animal products and waste products for the existence of resistant bacteria. Exposure assessment The following are examples of the type of inputs that may be required in the exposure assessment: – prevalence of resistant bacteria in ill animals – prevalence of food and/or the animal environment contaminated with resistant bacteria – animal-to-animal transmission of the bacteria under consideration – number/percentage of animals treated with the particular antimicrobial – dissemination of resistant bacteria from animals (animal husbandry method, movement of animals) – prevalence of animal feed contaminated with resistant bacteria – amount of antimicrobial used in animals OIE International Standards on Antimicrobial Resistance, 2003 153 3. Risk analysis – treatment (dose, route of administration, duration) – microbial load in contaminated food at point of consumption – survival capacity of resistant bacteria (competition of mixed populations, survival in the environment, contamination cycles including potentially the following elements: animals, humans, animal feed, environment, food, non-food producing animals, wildlife) – dissemination of resistant bacteria and resistance determinants – disposal practices for waste products and the opportunity for human exposure to resistant bacteria or resistance determinants in those waste products – capacity of resistant bacteria to become established in animal intestinal flora – exposure to resistance determinants from other sources – dose, route of administration (oral, injection) and duration of human treatment – pharmacokinetics (metabolism, bioavailability, access to intestinal flora). Consequence assessment Examples of consequences include the following: – dose-response relationships – variation in susceptibility of sub-populations – variation and frequency of animal health effects resulting from loss of efficacy of antimicrobials – changes in veterinary medicine practices resulting from reduced confidence in antimicrobials – associated costs – perceived future of the drug (time reference). Risk estimation For a quantitative assessment, the final outputs may include the following: – number of therapeutic failures due to resistant bacteria – animal suffering (level and increase) – economic cost (treatment with antibiotics, veterinary services, husbandry, reduced income, loss of market) – deaths (total per year; probability per year or lifetime for a random member of the population or a member of a specific more exposed sub-population) – level of resistance observed in animals. Risk management options to evaluate The following risk management measures could be implemented: – decision not to grant a licence for use of a new antimicrobial – review of licence authorisation and label indications 154 OIE International Standards on Antimicrobial Resistance, 2003 3. Risk analysis – – – – – revoking of licence for antimicrobials already used restrict use of antimicrobial (e.g. in particular industries, therapeutic only) review of prudent use guidelines establish monitoring of veterinary use of antimicrobials revision of treatment guidelines. Appendix C Comparison of systems and terms used by the Codex Alimentarius and the World Organisation for Animal Health The terms used in this document comply with the OIE terminology, as defined in Section 1.4. of the Code (16) based on the Covello-Merkhofer system (4). The Codex Alimentarius (3) uses a different, but equally valid system, designed by the US NAS (15). The issue of antimicrobial resistance arising from the use of antimicrobials in food-producing animals bridges the domain of OIE for animal husbandry and that of the FAO for food safety. It is therefore useful to compare these two systems and define terms used in this paper, to help integrate the two approaches. Two risk analysis terminology systems: description Table I summarises the components of risk analysis in the OIE and Codex models. Table I The components of risk analysis: a comparison of the systems used by the Codex Alimentarius and the OIE (World organisation for animal health) Codex Alimentarius Risk assessment Risk management Risk communication Components of risk analysis system OIE Hazard identification Risk assessment Risk management Risk communication Table II summarises the components of risk assessment in the OIE and Codex models. In a system based on the NAS model (called the ‘Codex system’ here), there are only three components of risk analysis, whereas in the system based on the CovelloMerkhofer model (called the ‘OIE system’ here), four components are present. Both systems include risk assessment, risk management and risk communication as components of risk analysis. However, the OIE system also includes hazard identification as a component of risk analysis, whereas the Codex system includes hazard identification as a sub-component of risk assessment. The terms risk management and risk communication are equivalent under both systems. OIE International Standards on Antimicrobial Resistance, 2003 155 3. Risk analysis Table II The components of risk assessment: a comparison of the United States Academy of Science model (used by Codex Alimentarius) and the CovelloMerkhofer model (used by the OIE [World Organisation for Animal Health]) Codex Alimentarius Components of risk assessment model OIE Hazard identification Hazard characterisation Exposure assessment Risk characterisation Risk release assessment Exposure assessment Consequence assessment Risk estimate The NAS system was initially developed to assess the risks to health from exposure to chemicals. Codex has adapted this system for food safety purposes. The CovelloMerkhofer system was initially developed to assess a wide range of risks from any potential hazard. The specific wording of the explanations in Table III reflects those differences. Table III Definition of risk analysis terms: a comparison of the systems used by the Codex Alimentarius and the OIE (World organisation for animal health) Term OIE definition or equivalent Codex Alimentarius definition or equivalent Acceptable risk Risk level judged by Member Countries to be compatible with the protection of animal and public health within their country No equivalent defined Consequence assessment Definition of the relationship between specified Codex equivalent: dose-response exposures to a biological agent and the consequences of assessment those exposures. A causal process must exist by which exposures produce adverse health or environmental consequences, which may in turn lead to socioeconomic consequences. The consequence assessment describes the potential consequences of a given exposure and estimates the probability of these consequences occurring. This estimate may be either qualitative or quantitative OIE equivalent: consequence assessment The determination of the relationship between the magnitude of exposure (dose) to a chemical, biological or physical agent and the severity and/or frequency of associated adverse health effects (response) – see ‘hazard characterization’ Describing the biological pathways necessary for The qualitative and/or quantitative exposure of animals and humans to the hazards evaluation of the likely intake in released from a given source, and estimating the biological, chemical and physical agents probability of the exposure occurring, either via food as well as exposures from qualitatively or quantitatively other sources if relevant …/… Dose-response assessment Exposure assessment 156 OIE International Standards on Antimicrobial Resistance, 2003 3. Risk analysis Term OIE definition or equivalent Codex Alimentarius definition or equivalent Hazard In the context of the Code, any pathogenic agent that could produce adverse consequences on the importation of a commodity A biological, chemical or physical agent in, or condition of, food with the potential to cause an adverse health effect Hazard characterisation Embodied in the ‘consequence assessment’ in the OIE The qualitative and/or quantitative system evaluation of the nature of the adverse health effects associated with biological, chemical and physical agents that may be present in food. For chemical agents, a dose-response assessment should be performed if the data are obtainable Hazard The process of identifying the pathogenic agents which The identification of biological, chelical identification could potentially be introduced to the commodity and physical agents capable of causing considered for importation adverse health affects and which may be present in aparticular food or group of foods Implementation The process of following through with the risk No equivalent defined management decision and ensuring that the risk management measures are in place Monitoring and The ongoing process by which the risk management No equivalent defined review measures are continually audited to ensure that they are achieving the results intended Option The process of identifying, evaluating the efficiency and No equivalent defined evaluation feasibility of, and selecting measures in order to reduce the risk associated with an importation in line with the appropriate level of protection of the Member Country. The efficacity is the degree to which an option reduces the likelihood and/or magnitude of adverse biological and economic consequences. Evaluating the efficacy of the options selected is an iterative process that involves their incorporation into the risk assessment and then comparing the resulting level of risk with that considered acceptable. The evaluation for feasibility normally focuses on technical, operational and economic factors affecting the implementation of the risk management options Qualitative risk An assessment in which the outputs on the likelihood No equivalent defined assessment of the outcome or the magnitude of the consequence are expressed in qualitative termes such as high, medium, low or negligible Quantitative An assessment in which the outputs of the risk No equivalent defined risk assessment assessment are expressed numerically Release Description of the biological pathways necessary for the No equivalent defined assessment use of an antimicrobial in animals to release resistant bacteria or resistance determinants into a particular environment, and estimation of the probability of that complete process occurring, either qualitatively or quantitatively Risk The likelihood of the occurrence and the likely A function of the probability of an magnitude of the consequences of an adverse event to adverse health effect and the severity of animal or human health in the importing countryduring that effect, consequential to a hazard(s) a specified time period in food …/… OIE International Standards on Antimicrobial Resistance, 2003 157 3. Risk analysis Term OIE definition or equivalent Codex Alimentarius definition or equivalent Risk analysis The process composed of hazard identification, risk assessment, risk management and risk communication A process consisting of three components: risk assessment, risk management and risk communication Risk assessment The evaluation of the likelihood and the biological and economic consequences of entry, establishment, or spread of a pathogenic agent within the territory of an importing country Risk characterisation OIE equivalent: risk estimation Risk communication Risk communication is the interactive exchange of information on risk among risk assessors, risk managers and others interested parties Integration of the results from the release assessment, exposure assessment and consequence assessment to produce overall measures of risks associated with the hazards identified at the outset. Thus, risk estimation takes into account the entire risk pathway from the hazrd identified to the unwanted outcome The process of comparing the risk estimate in the risk assessment with the appropriate level of protection of the Member Country Risk estimation Risk evaluation Risk management Sensitivity analysis Transparency Uncertainty Variability 158 The process of identifying, selecting and implementing measures that can be applied to reduce the level of risk The process of examining the impact of the variation in individual model inputs on the model outputs in a quantitative risk assessment Comprehensive documentation of all data, information, assumptions, methods, results, discussion and conclusions used in the risk analysis. Conclusions should be supported by an objective and logical discussion and the document should be fully referenced The lack of of precise knowledge of the input values which is due to measurement error or the lack of knowledge of the steps required, and the pathways from hazrd to risk, when building the scenario being assessed A real-world complexity in which the value of an input is not the same for each case due to natural diversity in a given population A scientifically based process consisting of the following steps: (i) hazard identification, (ii) hazard characterisation, (iii) exposure assessment and (iv) risk characterisation The qualitative and/or quantitative estimation, including attendant uncertainties, of the probability of occurrence and severity of known population based on hazard identification, hazard characterisation and exposure assessment Codex equivalent: risk characterisation Embodied in ‘risk management’ in the Codex system The process, distinct from risk assessment, of weighing policy alternatives, in consultation with all interested parties, considering risk assessment and other factors relevant for the health protection of consumers and for the promotion of fair trade practices, and if needed, selecting appropriate prevention and control options No equivalent defined No equivalent defined No equivalent defined No equivalent defined OIE International Standards on Antimicrobial Resistance, 2003 3. Risk analysis The first difference centres around the place of hazard identification in the models. The initial report of the NAS model (15), describes hazard identification as a major undertaking. The definition relates specifically to chemicals, and even in this case, NAS indicates that it includes weighing the available evidence relevant to cause and effect, as well as evidence relating to the magnitude of effect for the specified chemical. It is essentially a qualitative process of considerable magnitude. Given the number of potential pathogen hazards present in animals and animal products, the OIE risk analysis system, with a separate hazard identification step, is more adapted to pathogenic risk management. The second difference is the presence in the OIE system of a step called release assessment, absent in the Codex system. Covello and Merkhofer argue that this is necessary for describing the probability of a given system (e.g. an industrial complex, a meat processing plant or another risk source) to release risk agents into the environment of interest. They believe this to be an essential step in obtaining an accurate understanding of risk. From a practical standpoint, this is an essential explicit step either to assess the risks due to a particular hazard from a specific source or process, or to undertake a cost-benefit analysis of putting in place release reduction safeguards for that source or process. ‘Release’ comes before the possibility of exposure in actual exposure events. Thus, the Covello-Merkofer system follows release assessment by assessing the probability of exposure for each potential exposure route of interest. The third difference between the models is that the NAS system places exposure assessment after the dose response (hazard characterisation) step. The precise definitions are also slightly different. The fourth difference is in the place and meaning of consequences in the two models. Exposure can then lead to consequences – unwanted consequences when considering a hazard. Thus, the Covello-Merkhofer system places consequence assessment after exposure assessment, and defines it broadly (any consequences that can occur can be considered, and their probability assessed). However, the NAS system looks only at the consequences of variation in dose of the chemical being considered (i.e. a doseresponse assessment, also called hazard characterisation). Table IV Definition of new terms introduced in this document Term Definition Risk management policy The regulatory policy framework for the monitoring, measuring, assessing and managing of risks involved in the use of antimicrobials in foodproducing animals Semi-quantitative risk An assessment where estimates of the likelihood of the outcome and the assesment magnitude of the consequences are expressed in semi-quantitative terms via a scoring mechanism OIE International Standards on Antimicrobial Resistance, 2003 159 3. Risk analysis References 1. Casella G. & Berger R.L. (1990). – Statistical inference. Brooks/Cole Publishing Company, Pacific Grove, California, 650 pp. 2. Cassin M.H., Lammerding A.M., Todd E.C.D., Ross W. & McColl R.S. (1998). – Quantitative risk assessment for Escherichia coli O157:H7 in ground beef hamburgers. Int. J. Food Microbiol., 41 (1), 21-41. 3. Codex Alimentarius Commission (Codex) (1999). – Principles and guidelines for the conduct of microbiological risk assessment, CAC/GL-30. Food and Agriculture Organization, Rome, 6 pp. Website: ftp://ftp.fao.org/codex/standard/volume1b/en/GL_030e.pdf (document accessed on 13 September 2001). 4. Covello V.T. & Merkhofer M.W. (1993). – Risk assessment methods: approaches for assessing health and environmental risks. Plenum Press, New York and London, 334 pp. 5. Davison A.C. & Hinkley D.V. (1997). – Bootstrap methods and their applications. Cambridge University Press, Cambridge, 582 pp. 6. Efron B. & Tibshirani R.J. (1993). – An introduction to the bootstrap. Chapman & Hall, New York, 436 pp. 7. Food and Agriculture Organization (FAO) (1998). – The application of risk communication to food standards and safety matters. Food and Nutrition Paper No. 70. Report of a Joint FAO/World Health Organization Expert Consultation. FAO, Rome, 46 pp. 8. Food and Agriculture Organization (FAO) (2000). – Joint FAO/WHO Expert Consultation on risk assessment of microbiological hazards in foods. Food and Nutrition Paper No. 71. FAO, Rome, 54 pp. 9. Gelman A., Carlin J.B., Stern H.S. & Rubin D.B. (1995). – Bayesian data analysis. Chapman & Hall, London, 526 pp. 10. Groebner D.E. & Shannon P.F. (1993). – Business statistics: a decision-making approach. Macmillan, New York, 1,119 pp. 11. Jeffreys H. (1961). – Theory of probability, 3rd Ed. Oxford University Press, Oxford, 470 pp. 12. Kaplan S. (1999). – Beyond the domain of direct observation: how to specify a probability distribution that represents the state of knowledge about uncertain inputs. Risk Analysis, 19, 131-134. 13. Levin R.I. & Rubin D.S. (1994). – Statistics for management. Prentice Hall, New Jersey, 1,024 pp. 14. Morgan M.G. & Henrion M. (1990). – Uncertainty: a guide to dealing with uncertainty in quantitative risk and policy analysis. Cambridge University Press, New York, 332 pp. 15. National Academy of Sciences National Research Council (NAS-NRC) (1983). – Risk assessment in the Federal Government: managing the process. NAS-NRC Committee on the Institutional Means for the Assessment of Risks to Public Health. National Academy Press, Washington, DC, 191 pp. 160 OIE International Standards on Antimicrobial Resistance, 2003 3. Risk analysis 16. OIE (World organisation for animal health) (1999). – International animal health code: mammals, birds and bees, 8th Ed. OIE, Paris, 468 pp. 17. Olkin I., Glaser L.J. & Derman C. (1980). – Probability models and applications. Macmillan, New York. 18. Ripley B.D. (1987). – Stochastic simulation. Wiley, New York, 252 pp. 19. Rubinstein R. (1981). – Simulation and Monte Carlo methods. Wiley, New York, 278 pp. 20. Sivia D.S. (1996). – Data analysis: a Bayesian tutorial. Oxford University Press, Oxford, 189 pp. 21. Teunis P.S.N. & Havelaar A.H. (2000). – The beta Poisson dose-response model is not a single-hit model. Risk Analysis, 2 (4), 513-520. 22. Vose D. (2000). – Risk analysis: a quantitative guide. Wiley, Chichester, 418 pp. 23. Wooldridge M. (1999). – Risk assessment report: qualitative risk assessment for antibiotic resistance. Case study: Salmonella Typhimurium and quinolone/fluoroquinolone class of antimicrobials. Annex IV. In Antibiotic resistance in the European Union associated with the therapeutic use of veterinary medicines: report and qualitative risk assessment by the Committee for Veterinary Medicinal Products, 14 July 1999. European Agency for the Evaluation of Medicinal Products, London, 41 pp. __________ OIE International Standards on Antimicrobial Resistance, 2003 161 3. Risk analysis Risk assessment techniques – and antibiotic resistance Dr M. Wooldridge Central Veterinary Laboratory, New Raw, Addle Stone, Surrey KT15 3NB, United Kingdom Introduction Risk assessment is one of the components of risk analysis, the others being hazard identification, risk management, and risk communication (OIE Animal Health Code, 2001). Whatever the issue of concern, including that of antibiotic resistance, the purpose of a risk assessment is to supply information to the risk managers, policymakers, and other stakeholders. That information may include estimates of risk, identification of crucial data deficiencies allowing targeted further data collection, and new insights into the processes occurring. Risk assessments may be qualitative, quantitative, and occasionally semi-quantitative. Whilst quantitative risk assessments use numerical data to estimate probabilities in numerical terms, qualitative assessments estimate probabilities by making logical deductions, in words, from the information available. Semi-quantitative assessments, when used, generally involve a system of categorisation, although there is controversy about their usefulness. Quantitative assessments may be deterministic, or stochastic. Deterministic assessments use single numbers as model inputs and therefore give single numbers as the outputs, or risk estimates. However, stochastic assessments use distributions to describe uncertainty in the model inputs, and as a result the model outputs, or risk estimates, are also in the form of distributions. These can be described statistically (mean, mode, median and other percentiles), and illustrated graphically. This method therefore gives much more information to the policymaker about the uncertainty in the risk estimate. Risk assessment methodology applied to antibiotic resistance Basic elements of a risk assessment Briefly, risk assessment methodology, whether qualitative or quantitative, requires: – the selection of one or more ‘risk questions’ – the construction of a ‘risk pathway’ of necessary steps from the initiating event to the outcome of interest (that is, the risk to be estimated) – the collection of data and information to enable the probability of each step in the risk pathway, and thus also the final outcome, to be estimated. Looking specifically at the problem of antibiotic resistance, we now consider what is meant by a ‘risk question’, how the way in which this is framed is crucial to the construction of the ‘risk pathway’, and the effect on the outcome or risk estimate, using three different examples. 162 OIE International Standards on Antimicrobial Resistance, 2003 3. Risk analysis The effect of the ‘risk question’ on the assessment method and resultant output The simplest type of risk question A ‘risk question’ usually concerns the probability of occurrence of a specified event. A simple example would be: ‘What is the probability that a randomly selected piece of uncooked meat on sale in country K is contaminated with pathogen D carrying resistance to antibiotic A?’ In this example, there is no requirement to identify the original source of the pathogen. There is no requirement to identify the further effects, or consequences, of contamination. It is therefore a simple ‘one step’ risk pathway with the initiating event (contamination of the meat) and the outcome of interest (contamination of the meat) being identical. Thus, the question can be answered most simply by a single piece of data – that is data obtained from a statistical survey of contamination of retail meat. Such a survey gives the prevalence of contamination directly, and as it is statistically based, the probability of contamination of a random piece of meat can be estimated directly from the prevalence. Indeed, those answering such a question might not be aware that they are conducting a quantitative risk assessment, so routine is this type of question. Adverse human health effects – the introduction of ‘cases’ into the risk question However, in general, the reason for wanting this information is to determine the adverse effects of such meat on human health. We are interested in human ‘cases’. Therefore, suppose we define a case as: ‘A human who has suffered an adverse health effect due to resistance to an antibiotic of group A present in pathogen D’. Efficient surveillance of the human population will then give the total number of such cases occurring in a given region and time-span, and from this the probability of a random person becoming a case can be estimated. However, there may be many sources of pathogen D and its associated resistance. Surveillance alone cannot give information on the source of this resistance unless pathogen D varies specifically due to its source (for example by strain), and also that there are no other possible sources of that strain of pathogen D. This may sometimes be the case, but it is certainly not always so. So the second example, a much more complex risk question often asked, is of the form: ‘What is the probability of an adverse health effect occurring in a randomly selected person due to the development of antibiotic resistance in pathogen D due in turn to the use of antibiotic A in species S?’ OIE International Standards on Antimicrobial Resistance, 2003 163 3. Risk analysis This risk question is a very complex risk question, with at least one, and possibly more multi-step risk pathways. The most direct pathway is a farm-to-fork pathway with additional steps, ingestion, infection, treatment of antibiotic resistance, and failure of that treatment leading to adverse effects. This pathway is illustrated diagrammatically in Figure 1. Antibiotic A Species S Pathogen D Slaughter Food chain Effect on D Effect on D Processing Retail Effect on D Effect on D Prepare and cook Effect on D Proportion of D develops resistance Adverse health effect Cross-contamination? Multiplication? Death? More D? Less D? Resistance causes failure to work Antibiotic (A or related) used Pathogen D colonises or infects human Human eats Proportion still resistant Fig. 1. Simplified risk pathway illustration for the second example risk question, showing the steps necessary for resistance developed in pathogen D due to the administration of antibiotic A to species S to result in adverse human health effects However, additional risk pathways include for example the pathway via environmental contamination, and the pathway via exchange between pathogens of genetic material coding for resistance – and there are others. The totality of this risk assessment is therefore a very complex exercise, and exceptionally data-hungry. Taking simply the pathway illustrated above, the typical kinds of data necessary are outlined in Table I. 164 OIE International Standards on Antimicrobial Resistance, 2003 3. Risk analysis Table I Examples of the types of data necessary to undertake the risk assessment indicated in the second risk question example Probability of Probable amount of D is random animal of species S Use of antibiotic A in species S D in animal, given infected D at each stage of food chain, given present in defined unit of meat Development of resistance in D when A used Meat consumed per helping D remaining viable through each stage of food D required for illness (i.e. dosechain, and therefore present in defined unit of meat response) Random human eating S-meat Colonisation/infection with D Illness Antibiotic treatment with A (or related) Failure A (or related) to work To summarise, in this example it is necessary to estimate the probability and probable amount of antibiotic A-resistant pathogen D in species S, due to the use of antibiotic A. It is then necessary to trace that pathogen D through the food chain pathway, estimating the probability of the resistant pathogen remaining viable, multiplying or dying, whilst passing through each of those steps, and the total amount present at each step and thus at ingestion, which is dependant upon those effects. Given knowledge of the dose-response effect in humans, this then allows an estimate of the probability that treatment with antibiotic A is required, and will fail. Thus, in theory, an overall estimate of the probability of an adverse human health effect in a random person due to the use of antibiotic A in species S can be ascertained. In practice, the complexity of this pathway is such that currently there are likely to be many areas of data uncertainty and many complete data gaps. It is therefore more likely that the method will currently be useful, and used, specifically to gain insights into the risk pathway, to identify those data gaps which are crucial to the estimation, and to estimate levels of uncertainty in available data. The question is often asked as to whether this method can be simplified, and the answer is yes, under certain circumstances it can be. For example, if there is only one possible source of antibiotic A-resistant pathogen D in meat of species S, then a knowledge of the probability and probable amount of pathogen present at the point of ingestion will give the required estimate. Sometimes, if the pathogen strain or type is very specific to one species of livestock, this may be appropriate, and the assessment may begin part-way along the food chain, for example with retail prevalence data. However, adopting this approach means that no insights into the process at earlier stages are possible, and it can give no information on the effect of control strategies at, for example the farm or early in the processing stages. The OIE International Standards on Antimicrobial Resistance, 2003 165 3. Risk analysis decision to adopt this approach must therefore be considered carefully in the light of all the policy makers’ requirements from the risk assessment process. Which is the problem of interest? – the importance of the correct risk question A third possible risk question format is shown in the following example: ‘Given an adverse health effect due to resistance to antibiotic A in pathogen D occurring in a particular person, what is the probability that it is due to the use of antibiotic A in species S?’ Although at first glance this may appear to be similar to the previous risk question, it is in fact very different. Here, a specific person has actually suffered an adverse health effect. The probability of an adverse health effect therefore does not need to be estimated – it is known, and it is one – a certainty. And the question also specifies that this adverse health effect is known to be due to resistance to antibiotic A in pathogen D. Therefore the probability of antibiotic A-resistant pathogen D being present is also one. The person is not a random member of the total population, but is known to be a member of the sub-population ‘cases’. The question therefore asks not what is the probability that the person will be adversely affected; rather it concerns the source of the development of antibiotic resistance in the affecting organism. To illustrate this risk question more fully, we therefore need to consider what possible sources there are; there may be many. As well as the possible source being treatment with antibiotic A of pathogen D-affected species S (the source in which the risk question is interested, which we will call source 1), other obvious examples of possible source include; treatment of pathogen Daffected species Y with antibiotic A (and species Y may often be humans); previous treatment of any species with a related antibiotic causing cross-resistance to develop in D; or innate resistance mechanisms to antibiotic A present in pathogen D, requiring no prior treatment of any species for development. Other sources may also be possible. Each potential source has a risk pathway, with varying complexities. Each source will result in a certain number of cases due specifically to that source for a given region and time-frame, and the addition of these cases-by-source will give the total number, that is 100%, of cases in that region and time-frame. Fig. 2 illustrates these sources by proportion, and their risk pathways, simplified. The risk assessor presented with this third risk question needs therefore to be able to estimate the proportion of the sub-population of cases caused by the method specified in the risk question; that is, the probability that the effect has come via the specified pathway 1, and this depends directly upon proportion 1. Therefore the assessor needs to estimate proportion 1. The question then becomes ‘how is this proportion estimated?’ In theory this is easily solved. The method used to answer risk question 2 will give an estimate of the probable number of cases from a particular specified source, for example source 1, above. Surveillance data will give the total number of cases from all sources. The estimate from source 1 can then be converted 166 OIE International Standards on Antimicrobial Resistance, 2003 3. Risk analysis to a proportion of total cases, and this then gives a direct estimate of the probability of source 1 as the causal source for the case under consideration. In practice, of course, the estimate of the number of cases due to source 1 is as complex as the estimate for the second risk question example. Proportions 1+2+3+4+5 =100% of cases Species S with D Antibiotic A Proportion 2 Proportion of D develops resistance Human subpopulation cases Proportion 2 Complex pathways contact ? Environment ? Food handling? Antibiotic A (or related) Proportion 5 Complex pathways (as for example 2) Proportion of D develops resistance Humans with D Any other source D from any source Proportion 3 Antibiotic A (or related) previous dose Proportion 5 D with innate resistance Fig. 2 Simplified risk pathway illustration for the third example risk question, showing the possible sources of resistance, by proportion, to antibiotic A present in pathogen D in the sub-population ‘cases’ Cause for confusion – the difference the risk question makes It should now be apparent that the risk estimates for the second and third risk question examples will be very different, and this is illustrated in the following simple numerical example. Suppose that a survey has shown that 0.1% of the whole population suffers adverse health effects due to resistance to antibiotic A present in pathogen D. This sub-population of ‘cases’ is a very small proportion of the total population. The probability of any randomly selected person being within this subpopulation – i.e. a case – is low. Suppose also that there are, say, five different sources for antibiotic A-resistant pathogen D being present in humans, and that each source is equally likely. Then the percentage of cases by source is 20% for each source. This is illustrated in Figure 3. OIE International Standards on Antimicrobial Resistance, 2003 167 3. Risk analysis whole population cases: adverse health effects - any route small sub-group e.g. say: 0.1% low proportion of population low probability cases: adverse health effects proportions by route total =100% proportion 1 e.g. say: 20% of subgroup Fig. 3 The whole population showing the sub-population ‘cases’ as a percentage, and the subpopulation ‘cases’ showing the adverse health effects proportionally by source of antibiotic resistance: a comparison. This means that, for a specified case, the probability that the effect was due to a particular source is 20%, or P=0.2 (a 1 in 20 chance). However, as the cases comprise only 0.1% of the population, then 20% of the cases is equivalent to only 0.02% of the total population. So the probability that a particular source will result in the adverse effect in a random person in the total population is 0.02%, or P= 0.0002 (a 1 in 20,000 chance). Clearly these results are very different, and this is summarised in Table II. Table II A comparison of the difference in outputs, or risk estimates, which might typically be expected from the second and third risk question examples, illustrating the necessity of ensuring precision in the risk question asked, so that it will give the type of information required Question 2 ‘What is the probability of an adverse health effect occuring in a randomly selected person due to the development of antibiotic resistance in pathogen D due in turn to the use of antibiotic A in species S?’ P2 = 0.0002 (1 in 20,000 chance) Question 3 ‘Given an adverse health effect due to antibiotic resistance in pathogen D occuring in P3 = 0.2 a particular person, what is the probability that it is due to the use of antibiotic A in (1 in 20 chance) species S?’ 168 OIE International Standards on Antimicrobial Resistance, 2003 3. Risk analysis Summary All risk assessments are designed to answer ‘risk questions’, and there may be many possible different ‘risk questions’ surrounding a particular issue. One essential element of technique is to select appropriate risk questions, and this is usually done in conjunction with the risk manager and other stakeholders. Inappropriate risk questions lead to inappropriate answers. Worse still, they may lead to misunderstood or misinterpreted answers, which in turn may lead to very poor risk management decisions. One illustrated example of a possible risk question in the field of antibiotic resistance is: ‘What is the probability of an adverse health effect occurring in a randomly selected person due to the development of antibiotic resistance in pathogen D due in turn to the use of antibiotic A in species S?’ The estimated probability for the above question may well be very low. An apparently similar, but in fact very different risk question is: ‘Given an adverse health effect due to antibiotic resistance in pathogen D occurring in a particular person, what is the probability that it is due to the use of antibiotic A in species S?’ The estimated probability for this second question may be very high, particularly if the only significant source of pathogen P is species S. This difference in estimated probability may well be the cause of misunderstanding if the differences in the meaning of the information elicited by the two questions are not appreciated. In the first question, the intention is to estimate the probability of any random person suffering the unwanted effect, from a particular source. Thus, one is looking forward in time from the source action to an event, which may or may not occur. In the second question, the event has occurred – and one is then looking back from this to the action, which may have caused it. The estimated probabilities associated with the two questions are therefore likely to be very different. When undertaking risk assessments, a crucial part of the technique is therefore to decide which ‘risk questions’ are the most appropriate to ask, which are actually being asked, and to ensure that risk assessors and risk managers agree with the risk question, and all parties including other stakeholders understand what the estimate obtained actually does mean. __________ OIE International Standards on Antimicrobial Resistance, 2003 169 3. Risk analysis Impact of resistant campylobacteriosis in humans due to fluoroquinolone use in chickens L. Tollefson United States Food and Drug Administration/Center for Veterinary Medicine, Office of research, HFV-530, 8401 Muirkirk Rd, Laurel, MD 20708, United States of America Introduction and background In 1999 the Food and Drug Administration (FDA) Center for Veterinary Medicine published a framework document that proposed a process for evaluating and managing the human health impact of the microbial effects of antimicrobial animal drugs intended for use in food-producing animals. Stakeholders, particularly the animal pharmaceutical companies and food animal veterinarians, stated that the magnitude of the risk of using antimicrobials in food-producing animals needed to be determined prior to implementing regulatory changes that would impact the industry. In response to these comments, the Center for Veterinary Medicine agreed to conduct a quantitative risk assessment to determine the human health impact in the United States of America (USA) of acquiring fluoroquinolone-resistant Campylobacter infection from exposure to chicken. Process The Center for Veterinary Medicine contracted with a risk analyst to develop a risk assessment model to relate the prevalence of fluoroquinolone-resistant Campylobacter infections in humans associated with the consumption of chicken to the prevalence of fluoroquinolone resistant Campylobacter in chickens. The FDA is particularly concerned about resistant Campylobacter because this organism is one of the most common causes of bacterial foodborne illness in the USA (2, 4). Chicken is known to be a common source, although not the only source, of campylobacteriosis in humans in the United States of America and fluoroquinolones are important drugs in human medical therapy that are often used empirically to treat human enteric infections (3, 4). The risk assessment addressed that portion of the risk that was quantifiable, which is that related to consumption of chicken. The unquantifiable portion, that portion due to spread of the pathogen from chicken to other foods through contamination during food preparation or from secondary spread to other animals, although an important factor in the etiology of resistant Campylobacter infections, was not considered in the risk assessment. The Campylobacter risk assessment is a model for the direct transfer of resistance, which describes the situation where the resistant bacteria are transferred from animals to humans as a food contaminant. The etiology can be simplistically described by the following scenario: a) Poultry contract a disease, e.g. colibacillosis, and all birds in a house are treated with a fluoroquinolone drug. 170 OIE International Standards on Antimicrobial Resistance, 2003 3. Risk analysis b) Selective pressure from fluoroquinolone use leads to proliferation of fluoroquinolone-resistant Campylobacter in the poultry gut. c) Humans are infected by the fluoroquinolone-resistant Campylobacter organisms by consuming poultry. d) Treatment of humans with resistant illness with fluoroquinolones may be less effective. The approach to the risk assessment was straightforward with some initial broad assumptions. The model assumes that the presence of fluoroquinolone-resistant Campylobacter on chicken carcasses results from the use of fluoroquinolones in chickens. This does not mean that every chicken carrying resistant Campylobacter had to have been treated with a fluoroquinolone. Resistant organisms could have been acquired from a contaminated environment due to fluoroquinolone drug use in a previous group of birds, through contact with other chickens during transportation to the slaughter plant and antemortem processing, through contamination in the slaughter plant by other infected chicken carcasses, or through contamination of other foods in the home by the raw chicken meat. The model also assumes that susceptible and resistant Campylobacter have the same virulence characteristics and that susceptible and resistant Campylobacter have the same survival characteristics from slaughter to the point of human exposure. Finally, the model assumes that the presence of resistant Campylobacter on the chicken carcasses was due to antimicrobial drug use. Because of data supporting the linkage between antimicrobial drug use and antimicrobial resistance in animals in studies and in surveillance, this assumption is considered to be scientifically sound. The risk analysis methodology used for the Campylobacter risk assessment is based on that described by the OIE (World organisation for animal health) Ad Hoc Group on Antimicrobial Resistance (8). The risk analysis methodology described in the OIE document is tailored to address antimicrobial resistance in animals and includes hazard identification, risk assessment, risk management, and risk communication. Although it differs somewhat organisationally, the OIE approach includes similar steps in the risk assessment process as the risk analysis paradigm described by the National Academy of Sciences/National Research Council (NAS/NRC) that was developed for the assessment of carcinogenic risks (1). Data The number of Campylobacter culture confirmed human cases in the US population was used to estimate the total burden of campylobacteriosis. These data are collected from state public health laboratories that participate in FoodNet, the Centers for Disease Control and Prevention’s (CDC) Foodborne Disease Active Surveillance Network. FoodNet monitors the incidence of foodborne disease in humans and conducts studies to identify the sources and consequences of infection. Using the data on human Campylobacter cases reported in FoodNet, the risk assessment calculated 1.4 million cases of campylobacteriosis for 1999 (7). OIE International Standards on Antimicrobial Resistance, 2003 171 3. Risk analysis The model also estimates the number of fluoroquinolone-resistant Campylobacter cases attributable to chickens. Data from a Campylobacter case-control study conducted by CDC assisted in the removal of proportions of resistance attributed to other sources. Two other major sources of fluoroquinolone-resistant Campylobacter in humans are foreign travel and human use of fluoroquinolone antimicrobials. We therefore excluded from the estimate cases who had traveled to countries outside the USA in the last seven days, those patients who were prescribed a fluoroquinolone prior to stool culture, and those patients who were unsure of the timing of their treatment in relation to stool culture. For 1999 the mean number of the domestically-acquired fluoroquinolone-resistant Campylobacter cases attributable to chickens was approximately 154,000 (7). The model also estimated the number of the cases with fluoroquinolone-resistant campylobacteriosis due to chickens who actually received a fluoroquinolone drug for therapy. For 1999 the estimated mean number of people infected with fluoroquinolone-resistant Campylobacter from consuming or handling chicken and who subsequently received a fluoroquinolone as therapy was approximately 9,300 (7). These people likely received less effective or ineffective therapy for their infections, resulting in adverse health effects. The adverse health effects also have a negative impact on productivity in terms of lost workdays and increased cost of medical care. However, the risk assessment was limited to resistance development due to use of fluoroquinolones in chickens only and the impact is a mean estimate. The actual risk to humans from fluoroquinolone-resistant Campylobacter infections from all foodborne sources is likely to be higher. To estimate the quantity of chicken with fluoroquinolone-resistant Campylobacter consumed, several sources of data were used. The estimate is based on the per capita consumption of meat, the size of the population of the USA, the prevalence of Campylobacter among carcasses and the prevalence of resistance among contaminated carcasses. To estimate the quantity of chicken consumed, data were obtained from the US Department of Agriculture Economic Research Service with product sent for rendering, diverted for pet food, exports, water added during processing and imports subtracted (6). The proportion of chicken with Campylobacter and the proportion of Campylobacter that were fluoroquinolone-resistant were determined from samples that the Department of Agriculture had analysed (contaminated with Campylobacter) and susceptibility tested by the National Antimicrobial Resistance Monitoring System for fluoroquinolone resistance (5). Discussion One of the advantages of the model used for this risk assessment is its simplicity. Since we used data on actual human cases of campylobacteriosis, it wasn’t necessary to determine infectious dose and then estimate the potential number of human cases based on an average infectious dose. That process is cumbersome and requires complex assumptions with little data available to substantiate the assumptions. Instead, we limited the risk assessment to the human cases of disease. Also, we used 172 OIE International Standards on Antimicrobial Resistance, 2003 3. Risk analysis prevalence of fluoroquinolone-resistant Campylobacter on chicken carcasses at the slaughter plant, a point very close to the consumer, rather than resistance levels on the farm. The farm is remote to what the consumer is actually exposed to through food; testing retail chicken would be an even closer exposure source for consumers. References 1. Committee on the Institutional Means for Assessment of Risks to Public Health, Commission on Life Sciences, and National Research Council (1983). – Risk assessment in the Federal Government: managing the process. National Academy Press, Washington, DC. 2. Mead P.S., Slutsker L., Dietz V., McCraig L.F., Bresee J.S., Shapiro C., Griffin P.M. & Tauxe R.V. (1999). – Food-related illness and death in the Unites States. Emerg. infect. Dis., 5 (5), 607-625. 3. Sande M., Kapusnik-Uner J. & Mandell G. (1996). – Antimicrobial agents general considerations, section XI, chemotherapy of microbial diseases. In The pharmacological basis of therapeutics, 9th Edition (J. Hardman, L. Limbird, P. Molinoff, et al., eds). Goodman & Gilman’s, The McGraw-Hill Companies, New York., 1,039. 4. Tauxe R.V. (1992). – Epidemiology of Campylobacter jejuni infections in the United States and other industrial nations. In Campylobacter jejuni: current and future trends (I. Nachamkin, M.J. Blaser & L.S. Tompkins, eds). American Society for Microbiology, Washington, DC, 9-12. 5. United States Department of Agriculture – Agricultural Research Service. – National antimicrobial resistance monitoring system: enteric bacteria – 1999 animal Campylobacter isolate report. Athens, Georgia. (available at http://www.arru.saa.ars.usda.gov/main.htm) 6. United States Department of Agriculture (USDA) (1999). – Economic Research Service Food Consumption, prices and expenditures, 1970-1997. USDA. (available at: http://www.econ.ag.gov/) 7. United States Food and Drug Administration (USFDA) (2001). – Human health impact of fluoroquinolone resistant Campylobacter associated with consumption of chicken. USFDA. (available at: http://www.fda.gov/cvm/antimicrobial/Risk-assess.htm) 8. Vose D., Acar J., Anthony F., Franklin A., Gupta R., †Nicholls T., Tamura Y., Thompson S., Threlfall E.J., van Vuuren M., White D.G., Wegener H.C. & Costarrica M.L. (2001). – Antimicrobial resistance: risk analysis methodology for the potential impact on public health of antimicrobial resistant bacteria of animal origin. Rev. sci. tech. Off. int. Epiz., 20 (3), 811827. __________ OIE International Standards on Antimicrobial Resistance, 2003 173 3. Risk analysis Antimicrobial resistance and risk analysis: the view of a developing country M. van Vuuren Department of Veterinary Tropical Diseases, Faculty of Veterinary Science, University of Pretoria, South Africa Risk analysis has been used and applied for many years in diverse disciplines such as engineering and veterinary import control. It is currently also used in assessing the risk of food additives and drug residues in food, including toxicity studies in animals for risk assessment of animal products intended for human consumption. However, it was only after the Agreement on the Application of Sanitary and Phytosanitary (SPS) Measures became active during 1995 that regulatory officials from many countries became exposed to concepts such as risk assessment, regionalisation, equivalence and transparency. Under the SPS Agreement, it is expected that World Trade Organization Member Countries adopt risk assessment and risk management systems as part of their obligations to base their SPS protection measures on scientific principles. More specifically, however, in terms of the role that the use of antimicrobial drugs in animals might play as a possible threat to public health, strong sentiments were expressed at the 1999, Paris-based, OIE-sponsored (World organisation for animal health) meeting on The Use of Antibiotics in Animals, that scientific risk analysis should be the vehicle with which to approach this issue. To this effect, the International Committee of the OIE decided in May 1999 to create an ad hoc group to develop guidance documents for member countries that included inter alia risk analysis methodology for managing the potential impact on public health of antimicrobial resistant bacteria of animal origin. The OIE guidance document on risk analysis (1) is the result of continuing efforts (similar to that of other standard-setting organisations such as the International Plant Protection Convention and the Codex Alimentarius Commission in terms of plant health and food safety standards) to develop internationally acceptable standards in the field of animal health. In many developing countries, the application of risk analysis is in its infancy, mainly as a result of a lack of expertise. These countries are confronted with the problem of having to develop the capability to perform risk assessment, which implies that they will have to obtain the services of qualified professionals. To achieve this objective, the establishment of a risk assessment unit within a country is highly desirable, especially in view of the fact that risk assessment is a team effort, requiring experts in several disciplines. Such initiatives will go a long way in satisfying importing countries that the quality of the exporting country’s surveillance systems, laboratory capabilities and approach to quarantine measures are acceptable. Such a capability will also enable developing countries to analyse risk assessment procedures of other countries and to argue against unreasonable import measures. 174 OIE International Standards on Antimicrobial Resistance, 2003 3. Risk analysis The OIE has employed the Covello-Merkhofer method for import risk assessment and its ad hoc group on antimicrobial resistance has also applied this model to develop the risk analysis methodology for the potential impact on public health of antimicrobial resistant bacteria of animal origin (1). Prior to the publication of the latter document, an approach to determine the risk involved in the use of antimicrobial drugs in animals was first described by Wooldridge (2). Some developing countries apply risk assessment to food safety issues. In South Africa (and certain neighbouring countries) for example, risk assessments for pesticides and toxic substances follow the model of the joint Food and Agriculture Organization/World Health Organization (FAO/WHO) Codex Alimentarius Commission. The latter is based on maximum residue limits (MRLs) developed by the Joint FAO/WHO Meeting on Pesticide Residues and the Joint FAO/WHO Expert Committee for Food Additives (JECFA) and published as a Codex document. Briefly, companies apply for registration with the National Department of Agriculture. The dossiers are then forwarded to the Department of Health for risk assessment of the toxicological data. The main objective is to determine the maximum residue limits. The Acceptable Daily Intake is established by either JECFA and published by the Codex Alimentarius, the European Union (EU) or the Food and Drug Administration of the United States Department of Agriculture. ADIs for substances not published by any of these organisations are established locally following full risk assessments. After reviewing the available toxicological profiles of the new active ingredient based on acute toxicity, sub-chronic and chronic toxicity, ecotoxicity and environmental fate, and if there is no toxicological reason for concern, the Department of Health recommends registration thereof to the Department of Agriculture in terms of the appropriate Act. In the case of pesticides, companies are also required to have the MRLs determined for local conditions. For veterinary drug and food additive MRLs, the standards of JECFA, published by Codex are followed. These MRLs are based on the global food basket model and are accepted by African countries. If standards are available from both the EU and JECFA, the lowest value is accepted. Risk analysis for the potential impact on public health of antimicrobial resistant bacteria of animal origin is a formidable challenge for developing countries. It will entail inter alia the gathering of information (data) on the quantities of antimicrobial drugs used, and resistance patterns and trends through a national antimicrobial resistance surveillance and monitoring programme. In the realm of risk management, developing countries must encourage the implementation of prudent use principles, either through the initiative of the national government or relevant professional organisations. Marketing authorisation for antimicrobial drugs should be based on a complete evaluation by the appropriate authorities. Equally important is the responsibility to communicate the risk to all role players including veterinarians, livestock producers and pharmaceutical distributors. This will essentially entail the drafting of guidance documents for all on prudent use principles. Ideally, countries must strive to separate risk assessment from risk management and communication to ensure the independence of decision-making and evaluation of the risk. The OIE International Standards on Antimicrobial Resistance, 2003 175 3. Risk analysis responsibility of the risk assessor is to evaluate the data and to make recommendations on which the risk manager must act. On the other hand, the framework wherein a risk assessor will function in a country, including aspects such as threshold levels accepted by the public is determined and delineated by the risk manager, i.e. politician. These functions in many developing countries are currently performed by the same person. In Africa several initiatives have been launched as part of an effort to implement international standards relating to measures to ensure the protection of public health. The Southern and Eastern African Drug Registration Application Conference (SEAVDRAC) is an annual meeting devoted to the harmonisation of the evaluation and licensing procedures for veterinary medicines. The Southern and Eastern African Regulatory Committee for Harmonisation (SEARCH) is pursuing similar objectives by working towards harmonisation of pesticide registration and MRLs. Although the application of risk analysis in developing countries is absent or minimally implemented in many instances, the ability to do so in many of these countries already exists. The extent to which risk assessment is implemented by developing countries, is limited only by the motivation of management (national governments) to provide human, physical and financial resources. References 1. Vose D., Acar J., Anthony F., Franklin A., Gupta R., Nicholls T., Tamura Y., Thompson S., Threlfal, E.J., Van Vuuren M., White D.G., Wegener H.C. & Costarrica M.L. (2001). – Antimicrobial resistance: risk analysis methodology for the potential impact on public health of antimicrobial resistant bacteria of animal origin. Rev. sci. tech. Off. int. Epiz., 20 (3), 811-827. 2. Woolridge M. – Risk assessment applied to antibiotic resistance. In Proc. European Scientific Conference on the use of antibiotics in animals, 24-26 March, Paris, 18-28. __________ 176 OIE International Standards on Antimicrobial Resistance, 2003 3. Risk analysis Campylobacter risk analysis: a cause-and-effect view L.A. Cox, Jr. Cox Associates, Denver, Colorado, United States of America Campylobacter (CP) is the most commonly diagnosed cause of bacterial gastroenteritis in the United States. In undercooked chicken and non-poultry meats, raw milk and water and other undercooked contaminated foods or water, CP may cause gastroenteritis and infectious diarrhea lasting a week or more (Friedman et al., 2000). CP-infected patients are sometimes treated with the fluoroquinolone (FQ) antibiotic ciprofloxacin. It seems plausible that an FQ prescription might have diminished effectiveness against FQ-resistant CP strains, leading to excess days of illness. Development of FQ-resistant strains of CP in chickens may be favored by the use of other FQs, such as enrofloxacin, to combat respiratory disease in chicken broilers. Thus, a hypothesis in which eating chicken is a leading cause of domestic sporadic CP cases, and treating chickens with enrofloxacin raises the risk of FQ-resistant CP illness, seems sensible. We call this ‘hypothesis 1’. Although plausible, hypothesis 1 does not explain why several recent data sets indicate that eating chicken (and even touching raw chicken) at home can reduce risk of CP illness. An alternative, hypothesis 2, instead attributes risk of sporadic domestic CP cases primarily to commercial cooking of hamburger, chicken, and other meats. This paper examines evidence for hypothesis 2 from international trend data, a farmto-fork simulation model, and new analysis of case-control data collected by the Centers for Disease Control and Prevention (CDC). First, a simulation model of human exposures to CP via chicken is used to predict probable human health impacts of alternative risk management strategies. It predicts that strategies that reduce microbial load during processing and at the point of food consumption create the largest public health benefits. The case-control data reveal interactions among risk factors. For example, sex and age affect patterns of commercial food consumption, as well as risk of CP illness. We introduce causal graph models based on the data to summarise the main causal relations of interest. These causal models raise policy and methodological problems about how to attribute risks to specific factors (e.g., chicken consumption or use of FQs in chickens) when both direct and indirect effects are present. __________ OIE International Standards on Antimicrobial Resistance, 2003 177 4. Surveillance of resistance programme OIE International Standards on Antimicrobial Resistance, 2003 179 4. Surveillance of resistance programme Antimicrobial resistance: harmonisation of national antimicrobial resistance monitoring and surveillance programmes in animals and in animal-derived food A. Franklin (1), J. Acar (2), F. Anthony (3), R. Gupta (4), †T. Nicholls (5), Y. Tamura (6), S. Thompson (7), E.J. Threlfall (8), D. Vose (9), M. van Vuuren (10), D.G. White (11), H.C. Wegener (12) & M.L. Costarrica (13) (1) The National Veterinary Institute (SVA), Department of Antibiotics, SE 751 89 Uppsala, Sweden (2) Université Pierre et Marie Curie, Service de Microbiologie Médicale, Fondation Hôpital Saint-Joseph, 185 rue Raymond Losserand, 75674 Paris Cedex 14, France (3) Fresh Acre Veterinary Surgery, Flaggoners Green, Bromyard, Herefordshire HR7 4QR, United Kingdom (4) College of Veterinary Sciences, Veterinary Bacteriology, Department of Microbiology, G.B. Pant University of Agriculture and Technology, Pantnagar 263 145 Uttar Pradesh, India (5) National Offices of Animal and Plant Health and Food Safety, Animal Health Science and Emergency Management Branch, Department of Agriculture, Fisheries and Forestry, P.O. Box 858, Canberra ACT 2601, Australia (6) National Veterinary Assay Laboratory, Ministry of Agriculture, Forestry and Fisheries, 1-51-1 Tolura, Kokubunji, Tokyo 185-8511, Japan (7) Joint Institute for Food Safety Research, Department for Health and Human Services Liaison, 1400 Independence Avenue, SW, Mail Stop 2256, Washington, DC 20250-2256, United States of America (8) Public Health Laboratory Service (PHLS), Central Public Health Laboratory, Laboratory of Enteric Pathogens, 61 Collindale Avenue, London NW9 5HT, United Kingdom (9) David Vose Consulting, Le Bourg, 24400 Les Lèches, France (10) University of Pretoria, Faculty of Veterinary Science, Department of Veterinary Tropical Diseases, Private Bag X04, Onderstepoort 0110, South Africa (11) Centre for Veterinary Medicine, Food and Drug Administration, Office of Research, HFV-530, 8401 Muirkirk Road, Laurel, Maryland 20708, United States of America (12) World Health Organization, Detached National Expert, Division of Emerging and Transmissible Diseases, Animal and Food-related Public Health Risks, 20 avenue Appia, 1211 Geneva, Switzerland (13) Food and Agriculture Organization, Food Quality and Standards Service, Senior Officer, via delle Terme di Caracalla, 00100 Rome, Italy This report, prepared by the OIE Ad hoc Group of experts on antimicrobial resistance, has not yet received the approval of the International Committee of the OIE Summary A guideline on the harmonisation of national antimicrobial resistance monitoring and surveillance programmes in animals and animal-derived foods has been developed by the Ad hoc Group of experts on antimicrobial resistance of the OIE (World organisation for animal health). The objective of the guideline is to allow the generation of comparable data from various national surveillance and monitoring systems in order to compare the situations in different regions or countries and to consolidate results at the national, regional and international level. Definitions of surveillance and monitoring are provided. National systems should be able to detect the emergence of resistance, and to determine the prevalence of resistant bacteria. The resulting data should be used in the assessment of risks to public health and should contribute to the establishment of a risk management policy. Specific OIE International Standards on Antimicrobial Resistance, 2003 181 4. Surveillance of resistance programme factors identified for harmonisation include the animal species, food commodities, sampling plans, bacterial species, antimicrobials to be tested, laboratory methods, data reporting, database structure and the structure of reports. Keywords Antimicrobial resistance – Containment of resistance – Harmonisation – Human medicine – International standards – Laboratory methodology – Monitoring – Public health – Risk analysis – Surveillance – Veterinary medicine – World Organisation for Animal Health. Introduction This document describes the objectives of programmes for the monitoring and surveillance of antimicrobial resistance in bacteria of animal origin and animal-derived food products. The programmes will serve as a basis for the detection of national and global trends in the development of antimicrobial resistance in these bacteria. Animal species, food products, bacterial species and antimicrobials to be included in the programmes will be proposed. Sampling strategies, including statistically-based sampling options, data collection, recording, evaluation, and access to the data are considered. Comments are made on programme costs that may be of relevance to Member Countries. All aspects relating to laboratory methodologies are dealt with in Antimicrobial resistance: standardisation and harmonisation of laboratory methodologies for the detection and quantification of antimicrobial resistance, earlier in this volume. Background Antimicrobial susceptibility testing of bacteria has basically aided the clinician in the choice of efficient antimicrobials. Numerous point prevalence studies on antimicrobial resistance in bacteria of animal origin have been reported. Unfortunately, the usefulness of data from published studies is often hampered by inadequacies in study design. The methods and interpretive criteria used vary and comments on drug statistics are rarely included. The number of investigated isolates is generally low and confidence limits are rarely presented. The inclusion and exclusion criteria for the isolates included may be reasonably well described, but not the criteria for sampling. For example, most studies include results from clinical specimens sent to laboratories for routine analysis. It should be borne in mind when designing resistance monitoring and surveillance programmes that results from diagnostic submissions may not reflect the resistance situation in the animal population, as these types of submissions tend to include specimens from severe and/or recurrent clinical cases, including therapy failures. As a first step towards comparability of monitoring and surveillance data, Member Countries of the OIE (World organisation for animal health) should be encouraged to strive for harmonised and standardised programme design (2, 15, 17, 20). Data from countries using different methods and study design may otherwise not be directly comparable (10, 20). Nevertheless, data collected over time in a given 182 OIE International Standards on Antimicrobial Resistance, 2003 4. Surveillance of resistance programme country may at least allow the detection of emergence of antimicrobial resistance or trends in prevalence of resistance in that particular country. A limited number of countries has established national structures for central collection and evaluation of antimicrobial susceptibility data of bacteria isolated from animals (1). In most countries which have already initiated official resistance monitoring and surveillance, these programmes arose from the need to give guidance to practitioners on appropriate clinical therapy. Recently in some countries, programmes have been extended to include knowledge about antimicrobial resistance in food-borne pathogens and commensal bacteria, including evaluating local, regional and national trends (4, 7, 12, 13, 22). Existing systems may include central co-ordination, harmonisation of laboratory methodology, establishment of quality assurance schemes and external proficiency testing by a designated national co-ordinating laboratory. Definition of monitoring and surveillance In the International Animal Health Code, the OIE defines surveillance in animal health as ‘the continuous investigation of a given population to detect the occurrence of disease for control purposes, which may involve testing of a part of the population’. According to the OIE definition, monitoring ‘constitutes on-going programmes directed at the detection of changes in the prevalence of disease in a given population and in its environment’ (16). In the context of this guideline, ‘disease’ can be substituted by ‘antimicrobial resistance’. The chapter of the International Animal Health Code on monitoring and surveillance of animal health describes options for agent detection and disease prevalence. Antimicrobial resistance and prevalence can follow some of the OIE monitoring and surveillance definitions in animal disease guidelines mentioned below: a) scientifically-based surveys (including statistically-based programmes) b) routine sampling and testing of animals on the farm, at market or at slaughter c) an organised sentinel programme, sampling animals, herds, flocks, vectors, and/or collecting diagnostic results from veterinary practice d) the storage of biological specimens for retrospective studies e) analysis of veterinary diagnostic laboratory records. Passive surveillance is conducted when samples are submitted to a laboratory for testing by sources outside the programme. Active surveillance is conducted when the programme develops a sampling scheme based on the objectives of the programme and actively obtains isolates. OIE International Standards on Antimicrobial Resistance, 2003 183 4. Surveillance of resistance programme Reasons for resistance monitoring and surveillance programmes Resistance monitoring and surveillance programmes are intended to generate data that can be used as follows: – in risk analysis to determine risk to human and animal health – to detect emergence of antimicrobial resistance (e.g. particular phenotypes) – to determine the prevalence or trend in prevalence of reduced susceptibility to a certain antimicrobial (or resistance) in a defined population – to provide a basis for policy recommendations for animal and public health – to generate data that may guide the design of further studies – to identify the need for potential interventions – to assess the impact of interventions – to provide information for prescribing practices and prudent use recommendations. General aspects to be considered in resistance monitoring and surveillance When Member Countries are considering their options for the control of antimicrobial resistance arising from the use of antimicrobials in animals, several issues should be examined and analysed. In particular, the resistance situation in humans, including resistance in bacteria of concern to human medicine, and the capacity of countries to undertake resistance surveillance in bacteria of human origin should be taken into consideration. Monitoring of bacteria from animal-derived food collected at different steps of the food chain, including processing, packing and retailing, should also be considered. There are large variations among Member Countries both in the extent of the use of antimicrobials in animals and the public concern over such use. However, for all countries, the basic mechanisms of exposure of humans to resistant bacteria from food are the same. Exposure of humans to resistant bacteria can be either direct through exposure to zoonotic pathogens (Salmonella, Campylobacter), or indirect through exposure to resistance genes potentially transferable from commensal animal bacteria, such as Escherichia coli and Enterococcus spp., to human bacteria (9, 18, 21). Any antimicrobial use will exert selection pressure on exposed bacteria and may result in development of resistance. This should be taken into account when designing monitoring and surveillance programmes. This means that information is required on the antimicrobial substance used, the mode of usage and the quantities used. Although there is no linear relationship between the amount of a certain antimicrobial used and the development of resistance, increased use of an antimicrobial often results in 184 OIE International Standards on Antimicrobial Resistance, 2003 4. Surveillance of resistance programme decreased susceptibility among exposed bacteria. An antimicrobial selective pressure may affect the resistance phenotype of bacteria in different ways, as follows: a) cross-resistance and co-selection of resistance genes may explain how one antimicrobial selects for another antimicrobial b) multiple resistance confers resistance to several antimicrobials c) virulence and lack of hygiene may account for the survival and spread of resistant bacteria, even in the absence of an antimicrobial selection pressure (14). Thus, the rate of development of resistance in bacteria will, amongst others, depend on the character(s) of the resistance gene(s), such as transferability, time of exposure of the micro-organism to the antimicrobial, and not least on the characteristics of the exposed bacterial populations (11). Surveillance of antimicrobial resistance at regular intervals or ongoing monitoring of prevalence changes of resistance bacteria of animal, food, environmental and human origin, constitutes a critical part of a strategy aiming at limiting the spread of antimicrobial resistance and optimising choice of antimicrobials used in therapy. As the situation will vary over time and between countries and regions, data need to be collected at the appropriate regional and national level. Monitoring and surveillance programmes may serve as early warning systems in the sense that even minor shifts in susceptibility may be identified at an early stage. Interventions may then be taken to limit the further shifts in susceptibility or spread of resistance. Despite differences among Member Countries, it is essential that countries consider the collection of certain standardised information and the harmonisation of their surveillance and monitoring programmes to enable the international comparison of data. As bacteria do not respect country boundaries, the ability to evaluate the situation at a global level will enable a better assessment of the potential risks posed by resistant bacteria on human and animal health. The risk for human health from resistant bacteria or resistance genes of animal origin should, as far as possible, be quantified and put into perspective with other human health risks. As recommendations prepared by the OIE will be of global relevance, careful consideration must be given to realistic needs and public and animal health issues of OIE Member Countries in all regions of the world. Specific factors to be considered for the harmonisation of resistance monitoring and surveillance programmes To achieve comparability of results between national monitoring and antimicrobial surveillance programmes, the following factors should be considered by Member Countries in the design of such programmes: a) animal species/categories (including age) to be sampled b) for food sampling, the relative merits of sampling at the abattoir and retail outlet should be considered. In addition to food of domestic origin, food of foreign origin may also be considered, possibly at the port of entry of the products OIE International Standards on Antimicrobial Resistance, 2003 185 4. Surveillance of resistance programme c) sampling strategy to be employed, for example: active or passive collection of samples; random, stratified or systematically collected samples; statistically based sampling or opportunistic sampling d) samples to be collected (faeces, carcass, raw and/or processed food) e) bacterial species to be isolated f) antimicrobials to be used in susceptibility testing g) standardised susceptibility testing (under laboratory methodologies) h) quality control – quality assurance (under laboratory methodologies) i) type of quantitative data to be reported (under laboratory methodologies) j) database design for appropriate data extraction k) analysis and interpretation of data l) reporting (consideration of transparency of reporting and interests of stakeholders). A detailed consideration of specific factors is presented below. Animals Each Member Country should examine its livestock production systems and decide, after risk analysis, the relative importance of antimicrobial resistance for animal and human health. Categories of livestock that should be considered for sampling include cattle and calves, slaughter pigs, broiler chickens, layer hens and/or other poultry and farmed fish. The results of this examination coupled with knowledge of antimicrobial use in animals, where available (see Antimicrobial resistance: monitoring the quantities of antimicrobials used in animal husbandry, earlier in this volume), regional and seasonal factors, as well as the international trading status of the Member Country (e.g. net importer or exporter of livestock products), may influence the design of resistance monitoring and surveillance programmes. Food When considering the transfer of antimicrobial resistance from animals to humans, contaminated food is commonly considered to be the principal route. Antimicrobial resistance can be transferred either by pathogenic bacteria or by transfer of resistance genes carried by commensal bacteria. Raw food of animal origin may be contaminated with resistant enteric pathogens such as Salmonella spp., Campylobacter jejuni and Campylobacter coli or resistant commensal bacteria such as E. coli and Enterococcus spp. Little is known about the prevalence of resistant bacteria in food of animal origin, but it is important that food bacterial isolates (including isolates from food of plant origin) are included in national monitoring and surveillance programmes for antimicrobial resistance (3, 20). Plants and vegetables of different types may be exposed to manure or sewage from livestock and may thereby become contaminated with resistant bacteria of animal origin. 186 OIE International Standards on Antimicrobial Resistance, 2003 4. Surveillance of resistance programme Animal feed, including imported feed, may also be considered in monitoring in surveillance programmes. Sampling strategies General As described in Chapter 1.3.5. of the OIE International Animal Health Code on surveillance and monitoring of animal health (16), Member Countries will have to consider whether to utilise passively collected data from existing sources of information, such as data from veterinary diagnostic laboratories (appreciating the limitations of the data) and/or design new monitoring or surveillance programmes for specific needs, or perhaps modify existing programmes. After deciding on the objectives of the required programme, for example monitoring antimicrobial resistance prevalence changes in bacterial populations of the national pig herd, specific programme design decisions must be taken. The OIE recommends that Member Countries, very early in their consideration of the issue, examine their capacity to undertake such work with regards to financial and human resources. Some Member Countries may need to develop basic scientific antimicrobial resistance expertise in the animal health area before embarking on a resistance monitoring and/or surveillance programme. Other countries may have already implemented comprehensive monitoring and surveillance programmes and may only have to consider the issues related to harmonisation as discussed in this paper. Statistically based sampling strategies for food-borne pathogens and commensal bacteria Sampling strategies are usually based on two basic features: sample representativeness of the population of interest and the robustness of the sample collected. Sampling strategies should be based on addressing the defined objectives of the programme. Samples are typically targeted at representing a specific group or population of interest and may be collected randomly, systematically or stratified within the population of concern. An appropriate sampling strategy provides sample estimates that are accurate for the population of interest. If appropriate sampling strategies have been defined, calculating a statistically based sample size allows programme monitors to determine the precision of the prevalence estimates that will be obtained from the collected sample. Sample size considerations are important, as an inadequate sample size may fail to detect existing resistance and an excessively large sample size is a waste of resources. The source of sample specimens should be determined by the objectives of the monitoring programme. If the objective is to monitor the potential human health impact of antimicrobial resistant bacteria from food of animal origin, then faecal samples from an appropriate sample source, such as the abattoir, may be the most OIE International Standards on Antimicrobial Resistance, 2003 187 4. Surveillance of resistance programme convenient and least costly option for sample collection (2, 4, 13, 15). This would reflect the prevalence of resistance at the first step of the food chain. Sampling of the carcasses at the abattoir would provide information on slaughter practices, slaughter hygiene and the level of faecal contamination of meat during the slaughter process. Further sampling from the retail chain would provide an indication of prevalence changes before the food reaches the consumer (4, 19). However, for studies on the relationship between use of antimicrobials and prevalence of resistance in animal bacterial populations, samples taken from animals with known health status and antimicrobial exposure might be more suitable (Table I). Programmes need to be statistically-based, using random sampling techniques, and need to be stratified for relevant factors. An example of a table and formula to assist Member Countries in programme design considerations is included in Appendix A. The sampling should be stratified by geographic region and run continuously over the year to account for regional and seasonal variations. Depending on, amongst others, the financial resources of a country, sampling may be extended over longer time periods or modified in other ways. Table I Examples of sampling sources, samples types and outcome of monitoring Source Herd of origin Abattoir Processing, packing Retail Various origin Sample type Outcome Additional information required/additional statification Faecal Prevalence in bacteria originating from animal Per age categories, populations (of different age categories and production types, etc. production types) Relationship resistance – antibiotic use Antibiotic use over time Faecal Prevalence in bacterial populations originating from animals at age of slaughter Intestine As above Carcass Hygiene, contamination during slaughter Meat Hygiene, contamination during processing and products handling Meat Prevalence of resistance in bacteria originating products from food, exposure data for consumers Vegetables Prevalence of resistance in bacteria originating from vegetables, exposure data for consumers Animal feed Prevalence of resistance in bacteria originating from animal feed, exposure data for animals As different practices for rearing an animal species might entail different antimicrobial exposure, the category of animal included should be strictly defined. If several categories of the same animal species are included, the sampling should again be stratified for these categories. A single animal should be sampled per herd or flock on 188 OIE International Standards on Antimicrobial Resistance, 2003 4. Surveillance of resistance programme each occasion. It is the prevalence of resistance and trends in the bacterial populations, rather than the specific prevalence on the herd or flock level, that is of interest in antimicrobial resistance monitoring and surveillance programmes (2, 4). Bacterial isolates collected in this way will represent a stratified random sample of the bacterial population of each animal species surveyed. Determining the number of isolates to be tested in order to obtain a statistically robust estimate involves gathering information on the expected prevalence of resistance in the population. The level of precision desired in the prevalence estimate and the degree of confidence that the prevalence estimate would fall within this range are parts of the design parameters of the monitoring or surveillance programme. The total number of samples required to achieve the targeted number of resistant isolates, with the desired confidence in the estimated prevalence level of resistance, should be based on the statistical considerations mentioned above. Additionally, the known frequency with which bacteria may be isolated from animals or food must be taken into account. Furthermore, the actual number of isolates to be tested may need to be adjusted, due to laboratory and other pragmatic resource considerations. However, in the interpretation of data, the concomitant limitations arising from these adjustments must be recognised and taken into consideration. If results of the monitoring programme indicated a prevalence other than that estimated, the programme testing regime would need to be adjusted, or more detailed surveillance and investigation would be required. Sample specimens to be collected (faeces, carcass and retail food) As a rule, faecal samples are collected from livestock and whole caeca are collected from poultry. From cattle and pigs, a faecal sample size of 5 g to 50 g will provide a sufficient sample for isolation of the bacteria of concern. A large sample size will result in a higher number of isolates of the target bacterial species compared to a smaller sample size. The same sample can be used for isolation of both zoonotic and commensal bacteria. Existing food-processing microbiological monitoring and ‘hazard analysis and critical control points’ (HACCP) programmes may provide useful samples for monitoring and surveillance of resistance in the food chain after slaughter. However, experience in the collection of this type of sample is currently rare. Bacteria Three major categories of organisms would be monitored, as follows: a) animal bacterial pathogens b) zoonotic bacteria c) commensal bacteria. OIE International Standards on Antimicrobial Resistance, 2003 189 4. Surveillance of resistance programme If possible, isolates should be preserved at least until reporting is completed. A collection for retrospective studies may be set up by further storing of all isolates from certain years. Isolates should preferably be stored cryogenically. Animal bacterial pathogens Monitoring of resistance in animal pathogens is important, both to detect emerging resistance that may pose a concern for human and animal health and to guide veterinarians in their prescribing decisions. Furthermore, this information will be of value in providing guidance for the prudent use of antimicrobials in veterinary medicine. Animal pathogens have the capacity to rapidly spread between animals and may, in consequence, be repeatedly exposed to antimicrobials. Emergence of new resistance mechanisms and loss of susceptibility in animal bacterial pathogen populations will be detected at its earliest stage by surveillance and monitoring programmes for resistance in these bacterial populations. Furthermore, this type of information is readily available in many countries. Information on the occurrence of antimicrobial resistance in animal pathogens is in general derived from routine clinical material sent to veterinary diagnostic laboratories. These samples are often derived from severe or recurrent clinical cases, including therapy failures. However, because these isolates are likely to represent biased samples, this type of susceptibility data may not show the true prevalence of resistance within the given animal population and the appropriate caution must be exercised in the interpretation of the data. A means of mitigating this bias would be to consider collection of samples from primary clinical cases not previously treated with antibiotics, or isolation of potentially pathogenic bacteria from healthy animals. Examples of animal pathogenic bacteria The range of priority animal bacterial pathogens to be monitored should be determined, taking into account the national animal health situation. Examples of bacterial pathogens which may be considered for inclusion in resistance surveillance or monitoring programmes are presented in Table II. Table II Examples of animal bacterial pathogens which may be included in the resistance surveillance and monitoring Target animals Respiratory pathogens Enteric pathogens Udder pathogens Cattle Pasteurella spp. Haemophilus somnus Actinocacillus pleuropneumoniae Escherichia coli Salmonella spp. Escherichia coli Brachyspira Salmonella spp. Staphylococcus aureus Streptococcus spp. Pigs Poultry Fish 190 Other pathogens Streptococcus suis Escherichia spp. Vibrio spp. Aeromonas spp. OIE International Standards on Antimicrobial Resistance, 2003 4. Surveillance of resistance programme All bacteria should be identified according to internationally recognised standard procedures. Antimicrobial susceptibility testing should be performed with validated methods under internal and external quality assurance (see Antimicrobial resistance: standardisation and harmonisation of laboratory methodologies for the detection and quantification of antimicrobial resistance). Zoonotic bacteria Salmonella Sampling should preferably represent the primary production of cattle, pigs, broilers and other poultry. For the purpose of facilitating sampling and reducing the concurrent costs, samples are preferably taken at the abattoir. However, monitoring and surveillance programmes may also be able to use bacterial isolates from designated national laboratories originating from other sources. A collection of an optimal number of Salmonella isolates should be attempted within the practical and economic constraints of the country. Isolation and identification of bacteria and bacterial strains should follow internationally accepted procedures. Serovars of epidemiological importance such as S. Typhimurium and S. Enteritidis should be included. The selection of other relevant serovars will depend on the epidemiological situation in each country. All Salmonella isolates should be serotyped and when appropriate, phage-typed according to standard methods used at the nationally designated laboratories. Validated methods should be used for antimicrobial susceptibility testing of Salmonella (see Antimicrobial resistance: standardisation and harmonisation of laboratory methodologies for the detection and quantification of antimicrobial resistance). Campylobacter Campylobacter jejuni and C. coli can be isolated from the same samples as commensal bacteria. Isolation and identification of these bacteria should follow internationally accepted procedures. Campylobacter isolates should be identified, but also if possible, typed and characterised. However, this is likely to depend on the technical abilities and resources available in the Member Country. Agar or broth micro-dilution methods are recommended for susceptibility testing of Campylobacter. Internal and external quality control programmes should be strictly adhered to (see Antimicrobial resistance: standardisation and harmonisation of laboratory methodologies for the detection and quantification of antimicrobial resistance). It should be noted that there are no validated methods for susceptibility testing of Campylobacter and no internationally accepted reference strains available for quality control. However, work is currently in progress on validation of methods for susceptibility testing of Campylobacter. Enterohaemorrhagic Escherichia coli Enterohaemorrhagic E. coli, such as the serotype O 157 which is pathogenic to humans but not to animals, may be included in resistance monitoring and surveillance OIE International Standards on Antimicrobial Resistance, 2003 191 4. Surveillance of resistance programme programmes, provided that adequate laboratory security measures are in place. To date, experience from studies of bovine isolates of E. coli O 157 indicates that the prevalence of resistance is similar to that of commensal E. coli (7). Commensal/indicator bacteria Escherichia coli and enterococci are commensal bacteria common to all animals. These bacteria are considered to constitute a reservoir of resistance genes, which may be transferred to pathogenic bacteria causing disease in animals or humans. It is considered that these bacteria should be isolated from healthy animals, preferably at the abattoir, and be monitored for antimicrobial resistance. Escherichia coli and enterococci should be isolated using solid media without antimicrobials. Various enterococcal species may be considered for inclusion in monitoring programmes, but it seems reasonable always to include Enterococcus faecium (4, 14). For antimicrobial resistance traits of special interest, and where prevalence is expected to be very low, more sensitive isolation procedures may be required. In such cases, enrichment in broth containing a selective concentration of the antimicrobial of interest can be used in addition to solid media (8). Identification should follow standard methods used at nationally designated laboratories (2, 5). For susceptibility testing of commensal bacteria, validated methods should be used (see Antimicrobial resistance: standardisation and harmonisation of laboratory methodologies for the detection and quantification of antimicrobial resistance). Antimicrobials to be used in susceptibility testing All clinically important antimicrobial classes used in human and veterinary medicine should be monitored. However, the number of tested antimicrobials may have to be limited according to the financial resources of the country in question. A suggested selection of antimicrobials that may be considered for inclusion in national monitoring programmes is presented in Appendix B. The proposed list includes almost all major classes of antimicrobials used to treat both animal and human bacterial infections. In susceptibility testing, some of the proposed antimicrobials are also commonly used as representatives for other antimicrobials belonging to the same class. In general, bacteria that are for example, resistant to erythromycin or tetracycline, are also resistant to most other macrolides or tetracyclines, respectively. Standardised susceptibility testing See Antimicrobial resistance: standardisation and harmonisation of laboratory methodologies for the detection and quantification of antimicrobial resistance. Quality control – quality assurance See Antimicrobial resistance: standardisation and harmonisation of laboratory methodologies for the detection and quantification of antimicrobial resistance. 192 OIE International Standards on Antimicrobial Resistance, 2003 4. Surveillance of resistance programme Type of quantitative data to be reported See Antimicrobial resistance: standardisation and harmonisation of laboratory methodologies for the detection and quantification of antimicrobial resistance. Database design and recording of results Member Countries should give careful consideration to database design for antimicrobial monitoring and surveillance programmes. This is because of the volume and complexity of the information and the probable need for access over a long period of time. The storage of raw (primary, non interpreted) data is essential in order to allow for the evaluation of the data in response to various kinds of questions, including those arising in the future. However, it is strongly recommended that the strains are stored for an even longer period for future analysis. Consideration may need to be given to technical requirements of computer systems when an exchange of data between those different systems (comparability of automatic recording of laboratory data and transfer of these data to resistance monitoring programmes) should be envisaged. Results should be entered into a suitable national database and recorded quantitatively, for example as distribution of minimum inhibitory concentrations (MICs) in milligrams per litre or inhibition zone diameters in millimetres (2, 4, 12, 15, 20, 21). The information should include at least the following aspects: a) sampling programme b) sampling date c) animal species/livestock category d) type of sample e) purpose of sampling f) geographic origin of herd, flock or animal g) age of animal. The reporting of laboratory data should, where relevant, include the following information: a) identity of laboratory b) isolation date c) reporting date d) bacterial species e) serovar f) phage-type g) antimicrobial susceptibility result/resistance phenotype. The proportion of isolates regarded as resistant should be reported, with defined breakpoints. In the clinical setting, breakpoints are used to categorise bacterial strains OIE International Standards on Antimicrobial Resistance, 2003 193 4. Surveillance of resistance programme as susceptible, intermediate susceptible or resistant (6). These breakpoints, often referred to as clinical or pharmacological breakpoints, are elaborated on a national basis and vary between countries (10, 17, 20). The system of reference used should be recorded. For surveillance purposes, another type of breakpoint, the microbiological breakpoint, based only on the distribution of MICs or inhibition zone diameters of the specific bacterial species tested is preferred. When using microbiological breakpoints, only the bacterial population with acquired resistance that clearly deviates from the distribution of the normal susceptible population will be designated as resistant (17). Furthermore, the recording of the phenotype (resistance pattern) of isolates is also very important. Reporting and analysis of results Countries should give consideration to the designation of a national centre, which should assume the responsibilities to co-ordinate the activities related to the resistance surveillance and monitoring programmes, to collect information at a central location within the country and to produce an annual report on the resistance situation of the country. Participating laboratories should report results periodically to the national centre. The national centre should have access to the raw data and the complete results of quality assurance and inter-laboratory calibration activities and proficiency testing results (see Antimicrobial resistance: standardisation and harmonisation of laboratory methodologies for the detection and quantification of antimicrobial resistance). The annual report should include information on the structure of the monitoring system and on the chosen laboratory methods. It is of critical importance that quantitative results are reported in a harmonised way, as MICs or inhibition zone diameters in the form of histograms or as tables on frequency distributions. Additional information of value includes statistics on the number of animals produced, antibiotic use data and antimicrobials authorised for use. If possible, trends in prevalence of resistance should be related to antimicrobial usage data and also to the disease situation in each country. For the purpose of a risk assessment addressing a particular question, it may be necessary to generate specific information which would be relevant to the model that has been developed for these purposes. In such cases, special reports might be produced in co-operation with the persons responsible for conducting a specific risk assessment. If countries should envisage the sharing of raw data, the questions of ownership of the data, access to raw data, interpretation of data and publication of reports should be addressed. Conclusions and recommendations In many countries, antimicrobial resistance monitoring and surveillance in animal husbandry have recently become a targeted area. Monitoring or surveillance of 194 OIE International Standards on Antimicrobial Resistance, 2003 4. Surveillance of resistance programme antimicrobial resistance in bacteria from food is conducted only by a few countries. Monitoring of resistance in commensal bacteria of human origin is conducted by even fewer countries. Acknowledging the different resources available in different countries, co-ordination with other programmes, such as residue monitoring programmes, should be considered. The extensive experience of the OIE in animal disease monitoring and surveillance may form an important foundation for Member Countries in the consideration of approaches to the monitoring of antimicrobial resistance. However, as this is a new area for most OIE Member Countries, each country should evaluate the overall issue of antimicrobial resistance in animals and animal-derived food and carefully assess its needs. The practical issues of existing technical expertise, economic and resource requirements are important factors to be considered. Antibiorésistance : harmonisation des programmes nationaux de suivi et de surveillance de l’antibiorésistance chez les animaux et dans les aliments d’origine animale A. Franklin, J. Acar, F. Anthony, R. Gupta, †T. Nicholls, Y. Tamura, S. Thompson, E.J. Threlfall, D. Vose, M. van Vuuren, D.G. White, H.C. Wegener & M.L. Costarrica Résumé Le Groupe ad hoc d’experts sur l’antibiorésistance créé par l’Organisation mondiale pour la santé animale a élaboré une ligne directrice sur l’harmonisation des programmes nationaux de suivi et de surveillance de l’antibiorésistance chez les animaux et dans les aliments d’origine animale. Cette ligne directrice a pour objet de permettre l’obtention de données comparables dans les systèmes nationaux de suivi et de surveillance afin de pouvoir comparer les situations dans différents pays et régions et d’obtenir ensuite des résultats agrégés aux niveaux national, régional et international. Les auteurs donnent une définition de la surveillance et du suivi. Les systèmes nationaux devraient être en mesure de déceler l’apparition d’une résistance et de déterminer la prévalence de bactéries résistantes. Les données obtenues devraient être utilisées lors de l’évaluation des risques pour la santé publique et contribuer à la mise en œuvre d’une politique de gestion du risque. Plusieurs facteurs spécifiques ont été identifiés pour les besoins d’une telle harmonisation: l’espèce animale, les produits alimentaires, les programmes d’échantillonnage, les espèces bactériennes, les antibiotiques soumis aux tests, les méthodes de laboratoire, la communication des données, ainsi que la structure des bases de données et des rapports. Mots-clés Analyse du risque – Antibiorésistance – Harmonisation – Maîtrise de la résistance – Médecine humaine – Médecine vétérinaire – Méthodologie de laboratoire – Normes internationales – Oragnisation mondiale pour la santé aniamale – Santé publique – Suivi – Surveillance. OIE International Standards on Antimicrobial Resistance, 2003 195 4. Surveillance of resistance programme Resistencia a los antimicrobianos: armonización de programas nacionales de seguimiento y vigilancia de la resistencia a los antimicrobianos en animales y alimentos de origen animal A. Franklin, J. Acar, F. Anthony, R. Gupta, †T. Nicholls, Y. Tamura, S. Thompson, E.J. Threlfall, D. Vose, M. van Vuuren, D.G. White, H.C. Wegener & M.L. Costarrica Resumen El Grupo Ad hoc de expertos sobre la resistencia de las bacterias a los productos antimicrobianos de la Organización mundial de sanidad animal ha elaborado una directriz sobre la armonización de programas nacionales de seguimiento y vigilancia de la resistencia a los antimicrobianos en animales y alimentos de origen animal, pensada para que puedan obtenerse datos comparables a partir de distintos sistemas nacionales de vigilancia, lo que a su vez serviría para comparar la situación en diferentes países o regiones y elaborar datos agregados a escala nacional, regional e internacional. Los autores ofrecen la definición de ‘vigilancia’ y ‘seguimiento’. Los sistemas nacionales deben ser capaces de detectar la aparición de resistencias y determinar la prevalencia de bacterias resistentes. Esa información debe servir después para evaluar los riesgos para la salud pública y ayudar a definir programas de gestión de riesgos. A juicio de los autores, los principales elementos que conviene armonizar son: las especies animales, los productos alimentarios, los programas de muestreo, las especies bacterianas, los antimicrobianos analizados, los métodos de laboratorio, la forma de presentar los datos y la estructura de bases de datos e informes. Palabras clave Análisis de riesgos – Armonización – Contención de las resistencias – Medicina humana – Medicina veterinaria – Métodos de laboratorio – Normas internacionales – Organización mundial de sanidad animal – Resistencia a los productos antimicrobianos – Salud pública – Seguimiento – Vigilancia. Appendix A Sample size estimates for prevalence of antimicrobial resistance in a large population Expected prevalence 10% 20% 30% 40% 50% 60% 70% 80% 90% Level of confidence 90% desired precision 95% desired precision 10% 5% 1% 10% 5% 1% 24 43 57 65 68 65 57 43 24 97 173 227 260 270 260 227 173 97 2,429 4,310 5,650 6,451 6,718 6,451 5,650 4,310 2,429 35 61 81 92 96 92 81 61 35 138 246 323 369 384 369 323 246 138 3,445 6,109 8,003 9,135 9,512 9,135 8,003 6,109 3,445 Calculations based upon Epi Info v6.04b to c Upgrade, October 1997, Centers for Disease Control (public domain software available at http://www.cdc.gov/epo/epi/epiinfo.htm) 196 OIE International Standards on Antimicrobial Resistance, 2003 4. Surveillance of resistance programme Appendix B Proposed list of antimicrobials which, as a first step may be included in antimicrobial resistance surveillance and monitoring programmes Antimicrobial Salmonella/ Escherichia Campylobacter Enterococcus coli Animal pathogens, Grampoitive Animal pathogens, Gramnegative Beta-lactams Penicilin G Ampicillin Oxacillin Amoxi/Clav Cephalosporins Ceftiofur Ceftriaxone Cephalothin Macrolides Erythromycin Lincosamides Clindamycin Streptogramins Virginiamycin Quinupristin/Dalfopristin Tetracyclines Tetracycline Aminoglycosides Streptomycin Neomycin Kanamycin Gentamicin Apramicin Amikacin Amphenicols Chloramphenicol Florfenicol Potentiated sulphonamides Trimethroprim/Tmp-Sul Sulphonamides Quinolones Nalidixic Acid Enrofloxacin/ Ciprofloxacin Glycopteptides Vancomycin + + + + + + (Staph) + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + OIE International Standards on Antimicrobial Resistance, 2003 (+) + + + + 197 4. Surveillance of resistance programme References 1. Boisseau J. & Röstel B. (1998). – The role of international trade in animals, animal products and feed in the spread of transferable antibiotic resistance and possible methods for control of the spread of infectious agent resistance factors. In Comprehensive reports on technical items presented to the International Committee or to Regional Commissions. OIE (World organisation for animal health), Paris, 197-234. 2. Caprioli A., Busani L., Martel J.L. & Helmuth R. (2000). – Monitoring of antibiotic resistance in bacteria of animal origin: epidemiological and microbiological methodologies. Int. J. antimicrob. Agents, 14 (4), 295-301. 3. Corpet E.E. (1988). – Antibiotic resistance from food (letter). N. Engl. J. Med., 318, 1206-1207. 4. Danish Zoonosis Centre (2000). – DANMAP 99. Consumption of antimicrobial agents and occurrence of antimicrobial resistance in bacteria from food animals, food and humans in Denmark. Danish Veterinary Laboratory, Ministry of Agriculture and Fisheries, Copenhagen, 52 pp. Website: http://www.SVS.dk (document accessed on 22 August 2001). 5. Devriese L.A., Pot B. & Collins M.D. (1993). – Phenotypic identification of the genus Enterococcus and differentiation of phylogenetically distinct enterococcal species and species groups. J. appl. Bacteriol., 75, 399-408. 6. European Committee for Antimicrobial Susceptibility Testing (EUCAST) of the European Society of Clinical Microbiology and Infectious Diseases (ESCMID) (2000). – EUCAST Definitive Document E.Def 1.2, May 2000: terminology relating to methods for the determination of susceptibility of bacteria to antimicrobial agents. Clin. Microbiol. Infect., 6, 503-508. 7. Food and Drug Administration (FDA)/United States Department of Agriculture (USDA)/Centers for Disease Control (CDC) (1998). – National antimicrobial susceptibility monitoring program – veterinary isolates. Website: www.cdc.gov./ncidod/dbmd/narms (document accessed on 4 September 2001). 8. Ieven M., Vercauteren E., van Descheemaker P., Laer F. & Goosens H. (1999). – Comparison of direct plating and broth enrichment culture for the detection of intestinal colonization by glycopeptide-resistant enterococci among hospitalized patients. J. clin. Microbiol., 37, 1436-1440. 9. Joint Expert Technical Advisory Committee on Antibiotic Resistance (JETACAR) (1999). – The use of antibiotics in food producing animals: antibiotic resistant bacteria in animals and humans. Report of the Joint Expert Advisory Committee on Antibiotic Resistance. Commonwealth Department of Health and Aged Care and Department of Agriculture, Fisheries and Forestry, Canberra, 238 pp. http://www.affa.gov.au/docs/quarantine/fopolicy/jetacar.htm (document accessed on 22 August 2001). 10. Leegaard T.M., Caugant, D.A., Froholm L.O. & Hoiby E.A. (2000). – Apparent differences in antimicrobial susceptibility as a consequence of national guidelines. Clin. Microbiol. Infect., 6, 290-293. 198 OIE International Standards on Antimicrobial Resistance, 2003 4. Surveillance of resistance programme 11. Levy S.B. (1997). – Antibiotic resistance: an ecological imbalance. In Antibiotic resistance. Origins, evolution, selection and spread (D.J. Chadwick & J. Goode, eds). John Wiley & Sons, Chichester, 1-14. 12. Martel J.-L., Tardy F., Brisabois A., Lailler R., Coudert M. & Chaslus-Dancla E. (2000). – The French antibiotic resistance monitoring programs. Int. J. antimicrob. Agents, 14 (4), 275-283. 13. Moreno M.A., Dominguez L., Teshager T., Herrero I.A. & Porrero M.C. (2000). – Antibiotic resistance monitoring: the Spanish programme. Int. J. antimicrob. Agents, 14 (4), 285-290. 14. Mundy L.M., Sahm D.F. & Gilmore M. (2000). – Relationships between enterococcal virulence and antimicrobial resistance. Clin. Microbiol. Rev., 13, 513-522. 15. OIE (World organisation for animal health) (1999). – Report of Working Group 3: resistance monitoring and surveillance. In European Scientific Conference. The use of antibiotics in animals. Ensuring the protection of public health, 24-26 March, Paris. OIE, Paris, 96-135. Website: http://www.anmv.afssa.fr/ccoie/documents/cdrom1999/Antibiotics/index.htm (document accessed on 22 August 2001). 16. OIE (World organisation for animal health) (2000). – International animal health code: mammals, birds and bees, 9th Ed. OIE, Paris, 473 pp. 17. Ringertz S., Olsson-Liljequist B., Kahlmeter G. & Kronvall G. (1997). – Antimicrobial susceptibility testing in Sweden II. Species-related zone diameter breakpoints to avoid interpretive errors and guard against unrecognised evolution of resistance. Scand. J. infect. Dis. Suppl., 105, 8-12. 18. Scientific Steering Committee of the European Commission (1999). – Opinion of the Scientific Steering Committee on antimicrobial resistance. European Commission, DirectorateGeneral XXIV, Brussels, 121 pp. Website: http://europa.eu.int/comm/food/fs/sc/ssc/out50_en.pdf (document accessed on 22 August 2001). 19. Teuber M. & Perreten V. (2000). – Role of milk and meat products as vehicles for antibiotic-resistant bacteria. Acta vet. scand. Suppl., 93, 75-87. 20. Threlfall E.J., Fisher I.S.T., Ward L.R., Tschäpe H. & Gerner-Smidt P. (1999). – Harmonization of antibiotic susceptibility testing for Salmonella: results of a study by 18 national reference laboratories within the European Union-funded Enter-Net group. Microb. Drug Resist., 5, 195-200. 21. World Health Organization (WHO) (1997). – The medical impact of the use of antibiotics in food animals. Report of a WHO meeting, 13-17 October, Berlin. WHO, Geneva, 281 pp. 22. Wray C., McLaren I.M. & Beedel Y.E. (1993). – Bacterial resistance monitoring of salmonellas isolated from animals, national experience of surveillance schemes in the United Kingdom. Vet. Microbiol., 35, 313-319. __________ OIE International Standards on Antimicrobial Resistance, 2003 199 4. Surveillance of resistance programme The National Antimicrobial Resistance Monitoring System (NARMS) P.J. Fedorka-Cray (1), M.L. Headrick (2) & L. Tollefson (2) (1) PhD, USDA-ARS-Antimicrobial Resistance Research Unit, Athens, GA, United States of America (2) FDA Center for Veterinary Medicine, Rockville, MD, United States of America Antibiotic resistance in foodborne pathogens is an increasingly important health issue. There is a programme in place in the United States of America (USA) to monitor changes in susceptibility of enteric bacteria to antimicrobial drugs used in animals and humans. That programme is the National Antimicrobial Resistance Monitoring System – Enteric Bacteria (NARMS). Background Resistant bacteria are not as susceptible to antibiotics or other antimicrobial drugs as non-resistant bacteria. The use of antibiotics may eliminate susceptible bacteria, leaving resistant bacteria behind. If resistant bacteria spread, a person or animal with this infection may not be able to be treated with the usual antibiotics, or an increased dose may be required. As a result they may be sick for a longer time than if they had an infection caused by bacteria that were easily treatable with antibiotics. The increase in bacterial resistance to antimicrobial drugs is a natural phenomenon, an outcome of evolution. Any population of organisms, including bacteria, naturally includes variants with unusual traits. In this case, some bacteria have the ability to fend off the action of an antimicrobial. The use of antimicrobial drugs in humans and animals over the past fifty years has inadvertently accelerated the development of resistance by increasing the selection pressure exerted on these organisms. Once antimicrobial pressure has been introduced into an environment, resistance may be spread to other microbes. Food animals such as cattle, pigs, turkeys, or chickens may receive antimicrobial drugs for growth promotion and control or treatment of infectious diseases. Food animals can carry organisms that can make people sick, but may not necessarily make the animal sick. For example, Salmonella, Campylobacter, and E. coli O157 are common bacteria found in the intestines of various food animals. These bacteria may not cause disease in the animal, however, all three bacteria may cause foodborne illness in humans. These organisms may develop resistance when exposed to antimicrobial drugs given to the animal. These resistant organisms can contaminate food products at slaughter and then infect humans who eat the food, particularly if the food is undercooked or cross-contaminated after cooking. Evidence of increasing resistance to antimicrobial drug treatment in bacteria that infect humans has raised questions about the role that antimicrobial drug use in food animals plays in the emergence of antimicrobial drug resistant bacteria. The link 200 OIE International Standards on Antimicrobial Resistance, 2003 4. Surveillance of resistance programme between antimicrobial resistance in foodborne pathogenic bacteria and use of antimicrobials in food animals has been reported in a number of studies. For foodborne pathogens, particularly those such as Salmonella that are rarely transferred from person to person in the United States of America, food (such as meat or eggs) from food animals is considered a likely source of human exposure to resistant organisms. The National Antimicrobial Resistance Monitoring System The United States of America now has a programme in place that allows the Food and Drug Administration (FDA) to monitor resistance to antimicrobial drugs used in humans and food animals. The programme is called the National Antimicrobial Resistance Monitoring System – Enteric Bacteria (NARMS). It combines the activities of FDA, the Centers for Disease Control and Prevention (CDC), and the U.S. Department of Agriculture (USDA) to create a nationwide monitoring system. NARMS was started (and has been expanded) because of the human health concerns related to the use of antimicrobial drugs in food animals. As a part of NARMS, isolates of foodborne bacteria such as E. coli, Salmonella, Enterococci, and Campylobacter from humans and food animals are collected and tested to determine changes in bacterial susceptibility to antimicrobial drugs. Each year, samples are taken and tested to determine whether there have been changes over time in the resistance (or susceptibility) of certain enteric bacteria to a collection of antimicrobial drugs. The antimicrobial drugs tested are selected based on their importance in human and animal medicine. The food animal specimens are gathered from healthy farm animals, animal clinical specimens, from carcasses of food animals at slaughter, ground product at processing plants, and from retail meats. The human-origin isolates are sent to the CDC in Atlanta, Georgia, by state and/or local health departments in all fifty states. Animal-origin isolates are collected from sites across the U.S. and submitted for susceptibility testing conducted by the Agricultural Research Service (ARS) of USDA in Athens, Georgia. Animal isolates are received from a number of sources including federally inspected slaughter and processing facilities, the USDA National Animal Health Monitoring System, studies on farms, sentinel sites, which are Veterinary Diagnostic Laboratories, and the USDA National Veterinary Services Laboratories. The NARMS programme was begun in 1996 with Salmonella as the sentinel organism. Other enteric bacteria were added as the programme was expanded. NARMS results for Salmonella are available since 1997. Links to the summary data are posted on the FDA Center for Veterinary Medicine (CVM) web site (see below). These data can provide useful information about patterns of emerging resistance, which in turn can help guide treatment decisions. NARMS data are an asset to outbreak investigations. Antimicrobial resistance patterns are useful in identifying the source and magnitude of resistance. Antimicrobial resistance data from humans and animals are important for the development of public health recommendations for the use of drugs in humans and food animals. OIE International Standards on Antimicrobial Resistance, 2003 201 4. Surveillance of resistance programme The NARMS programme was expanded in 2001 and 2002 to include a Retail Arm of NARMS. This aspect of NARMS began with a pilot study that included the susceptibility testing of enteric bacteria isolated from retail meat samples collected from grocery stores in Iowa. The success of this pilot project led to the implementation of the collection of retail meat samples in multiple states in collaboration with the CDC and FoodNet sites. A study that includes susceptibility testing of enteric bacteria isolated from animal feed ingredients is also being conducted. The FDA CVM Office of Research Laboratory in Laurel, Maryland, is conducting the susceptibility testing of these isolates. CDC, USDA, and FDA test Salmonella, E. coli, Campylobacter, Enterococci, and other bacterial isolates for susceptibility to designated panels of selected antimicrobial drugs. The results of these tests are compared with data from previous years to look for changes in resistance patterns of the organisms to these drugs. Public health officials, animal producers, drug manufacturers, physicians, and veterinarians can use the information from NARMS to control and prevent harm from the use of antimicrobial drugs in food animals. The National Antimicrobial Resistance Monitoring System methods Human arm Participating state or local public health laboratories systematically select every 20th non-typhi Salmonella isolate, Shigella, and E. coli O157:H7 submitted to their laboratory and send the isolates, at the end of each month, to CDC. All Salmonella typhi, Listeria monocytogenes, and non-cholerae Vibrio isolates received by the participating laboratories are forwarded to CDC. Additionally, health department partners that also participate in the FoodNet Program, submit one Campylobacter isolate each week to the CDC Foodborne and Diarrheal Diseases Laboratory for susceptibility testing. States that participate in this programme continue to increase as more health department partners join the FoodNet Program. In 2003, FoodNet is expanding to ten sites. The FoodNet Web Site at: http://www.cdc.gov/foodnet/ contains a listing of the current participants. The antimicrobial susceptibility testing results are sent from the CDC laboratory to NARMS epidemiologists at CDC, where data are entered and analysed. Animal arm The USDA, ARS, Antimicrobial Resistance Research Unit (ARRU) laboratory in Athens, Georgia, receives Salmonella, Campylobacter, E. coli and Enterococci isolates from animals for antimicrobial susceptibility testing. Isolates are received at the Athens laboratory from the sources described in ‘How We Monitor.’ Poultry carcass rinses are sent to ARRU from the Food Safety Inspection Service (FSIS) laboratories for culture, isolation, and susceptibility testing of Campylobacter, E. coli, and Enterococci 202 OIE International Standards on Antimicrobial Resistance, 2003 4. Surveillance of resistance programme organisms. Salmonella isolates are received from FSIS, the Animal and Plant Inspection Service, the National Veterinary Services Laboratories (NVSL), and sentinel site laboratories. Salmonella serotyping is conducted at the NVSL. E. coli and Enterococci isolates are also isolated from on-farm faecal samples as part of the NAHMS or other on-farm epidemiologic investigations. Retail arm The Iowa Retail Meat Pilot Survey included collection and antimicrobial susceptibility testing of bacterial isolates from retail meats purchased from Iowa retail grocery stores. A total of 870 samples of ground beef, ground turkey, pork chops, and chicken breasts were collected from 300 randomly selected sites. These samples were cultured for Salmonella, Campylobacter, E. coli, and Enterococci. The collection phase of the Iowa Retail Meat Pilot Survey was completed in June 2002. A FoodNet Retail Meat Surveillance study began in January 2002. Samples of ground beef, ground turkey, pork chops and chicken breasts are being collected from grocery stores in participating FoodNet States. Enteric bacterial isolates from these samples are being sent from the FoodNet laboratories to FDA/CVM Office of Research for antimicrobial drug susceptibility testing of Salmonella, Campylobacter, E. coli, and Enterococci. An animal feed ingredient survey collected samples of meat meal, meat and bone meal, fish meal, blood meal, and poultry meal at rendering plants in the USA during 2002. These samples are tested for Salmonella, E. coli, Campylobacter and Enterococci. Additional components, including a study of plant-origin animal feed ingredients such as soybean or cottonseed meal in 2003, are being added. The NARMS programme is designed with comparable methodology between the human, animal, and retail arms. For all isolates, susceptibility testing currently involves the determination of the minimum inhibition concentration (MIC) for a panel of selected antimicrobial agents. These antimicrobial drugs are evaluated each year for their continued importance in human and animal medicine. The antimicrobial drugs tested can be modified to meet monitoring needs. Susceptibility testing of Campylobacter is performed to determine the MICs for eight antimicrobial agents: azithromycin, chloramphenicol, ciprofloxacin, clindamycin, erythromycin, gentamicin, nalidixic acid and tetracycline. National Committee for Clinical Laboratory Standards (NCCLS) guidelines are followed, when possible, throughout the testing procedure. Using NARMS as a template, FDA CVM and Mexico are working on a cooperative project known as ‘ResistVet’ to monitor trends in antimicrobial resistance in humans, animals, and retail meats at four sites in Mexico. A pilot project was begun in 2001 and a three-year cooperative agreement was signed in 2002. To further support antimicrobial resistance monitoring in Mexico, FDA CVM collaborated with the World Health Organization to conduct a training course in 2001 on the surveillance of Salmonella and antimicrobial resistance in foodborne pathogens. The training took place at a participating ResistVet site in Mexico. FDA’s goal is to protect the public health by ensuring that significant human antimicrobial therapies are not lost due to use of antimicrobial drugs in food- OIE International Standards on Antimicrobial Resistance, 2003 203 4. Surveillance of resistance programme producing animals, while providing for the safe use of antimicrobial drugs in foodproducing animals. Information from NARMS will allow evaluation of trends in the susceptibility of the organisms causing disease to the drugs used to treat them. For more information on antimicrobial resistance issues, visit CVM’s web site at www.fda.gov/cvm. NARMS reports are published annually. To review the NARMS Annual Reports and other NARMS information use the following Internet addresses: FDA/CVM NARMS site: – http://www.fda.gov/cvm/index/narms/narms_pg.html – CDC NARMS site: http://www.cdc.gov/narms/ – USDA/ARS web site: http://www.arru.saa.ars.usda.gov/narms.html For additional information on the NARMS programme, contact Dr Marcia Headrick, FDA CVM NARMS Coordinator, at [email protected] or (706) 546-3689. References for this article and the NARMS programme are also available from Dr Headrick. __________ 204 OIE International Standards on Antimicrobial Resistance, 2003 4. Surveillance of resistance programme Surveillance of resistance – human/animal coordinated approaches in France V. Jarlier Hôpital Pitié Salpêtrière, laboratoire central de bactériologie, 91 bd de l’Hôpital, 75634 Paris Cedex 13, France L’Observatoire National de l’Epidémiologie de la résistance Bacterienne aux Antibiotiques (ONERBA) (Scientific Committee: P. Allouch, G. Antoniotti, O. BajoletLaudinat, O. Bellon, J.D. Cavallo, H. Chardon, E. Chaslus-Dancla, H. Dabernat, F. Grobost, V. Jarlier [coordinator], N. Marty, M.H. Nicolas-Chanoine, Y. Péan, B. Perichon, J. Robert, M. Roussel-Delvallez, F. Tardy, E. Varon, Ph. Weber.) was created in 1997 as a non profit organisation, with the following missions: a) to gather and to analyse available information on bacterial resistance in France and to provide this to health authorities and professionals b) to advise on conditions of data collection c) to set up studies designed to obtain data in areas not yet covered and d) to participate in training programmes. ONERBA links fourteen networks organised by medical and veterinary labs that have been monitoring antibiotic resistance in France for several years, independently of the pharmaceutical industry. In order to reach its objectives, such a ‘network of networks’ requires methodological support and recommendations concerning the following: a) principles, aims and presentation of the different types of information b) definitions and terms c) data management (e.g. duplicate isolates) and data stratification d) quality controls. Efforts have been made to adopt a common body of definitions and terms used for humans and animals. A guide containing recommendations on the methodology and implementation of bacterial resistance monitoring in laboratories has been edited by ONERBA. Moreover it has been decided to include in drug susceptibility tests antibiotics allowing, as markers, comparisons between bacteria of animal and human origin, such as: ampicillin, amoxicillin-clavulanate, gentamicin, cotrimoxazole, tetracycline for Enterobacteriaceae, and penicillin G, oxacillin, kanamycin, gentamicin, erythromycin, tetracycline for Staphylococcus. The resistance data from private medical laboratories, hospital laboratories and veterinary laboratories are being presented under standardised formats on a website (www.onerba.org). __________ OIE International Standards on Antimicrobial Resistance, 2003 205 4. Surveillance of resistance programme The Japanese Veterinary Antimicrobial Resistance Monitoring System (JVARM) Y. Tamura National Veterinary Assay Laboratory, Ministry of Agriculture, Forestry and Fisheries, 1-15-1, Tokura, Kokubunji, Tokyo 185-8511, Japan With the rapid development of intensive systems for rearing food-producing animals, bacterial infection has caused serious economic losses in animal husbandry. As a result, antimicrobials have been widely used for the control of infection. Some reports indicate that many bacteria of animal origin have become resistant to these antimicrobials. An increasing incidence of antimicrobial-resistant bacteria could pose serious problems not only to animal hygiene, but also to public health. However, until recently there was a lack of nationwide information available on the antimicrobial resistance of bacteria of animal origin in Japan. Consequently, the Japanese Veterinary Antimicrobial Resistance Monitoring System (JVARM) was established in 1999 to replace the former monitoring system that specialised in animal hygiene. Background In 1969, the Swann Committee (4) reviewed the agricultural use of antimicrobials. Among their recommendations was that regular and much wider surveillance should be made of the bacteria of animals, animal products and man, including their antimicrobial resistance. Recently, the relationship between the use of antimicrobials in food-producing animals and the emergence of resistant bacteria in the food chain has become of great concern and has been the subject of numerous international meetings (2, 6, 7). In Japan, basic legislation on food, agriculture and rural areas was established in 1999 to stabilise and improve people’s lifestyle and to develop the national economy. This legislation aimed to improve the management of food in order to ensure food safety and improve food quality. Objectives The objectives of JVARM are to monitor the occurrence of antimicrobial resistance in bacteria in food-producing animals and monitor the consumption of antimicrobials for animal use. Moreover, other objectives are to identify the efficacy of antimicrobials in food-producing animals, to promote prudent use of such antimicrobials, and to ascertain the public health problem. Outline of the Japanese veterinary antimicrobial resistance monitoring system The JVARM (summarised in Fig. 1) is composed of three parts: monitoring the quantities of antimicrobials used in animals; resistance monitoring in zoonotic and indicator bacteria isolated from healthy animals; and resistance monitoring in animal pathogens isolated from diseased animals. In Japan, the Ministry of Agriculture, Forestry and Fisheries (MAFF) is responsible for the field of animal husbandry, but 206 OIE International Standards on Antimicrobial Resistance, 2003 4. Surveillance of resistance programme not food hygiene. Thus, test bacteria are isolated on the farm from food-producing animals, but not in food products. Ph am aceu tical com panies JV A R M C o n su m p tio n of An tim icrob ials R esistance in zo ono tic and indicato r b acteria R esistan ce in an im al patho g ens H ealth y an im als D iseased an im als Fig. 1 Outline of the Japanese veterinary antimicrobial resistance monitoring system. Monitoring of antimicrobial consumption The monitoring of antimicrobial consumption is shown in Figure 2. Pharmaceutical companies that produce and import antimicrobials for animals are required to submit data to the National Veterinary Assay Laboratory (NVAL) annually in accordance with pharmaceutical legislation. The NVAL subsequently collects, analyses and evaluates such data and MAFF headquarters publishes this data in a yearly report entitled the ‘Amount of medicines and quasi-drugs for animal use’. The annual weight in kilograms of the active ingredients of approved antimicrobials used in animals is collected. This includes only therapeutic antimicrobials for animal use and the data are subdivided by animal species. However, this only provides an estimate of the consumption for each target species, because one antimicrobial is frequently used for multiple animal species. P h a r m a c e u t ic a l C o . 1 2 ...… F o rm at ( M ic r o s o f t E x c e l) M AFF P u b lic a t io n ( y e a r l y ) 5 3 4 R ep o rt N a t io n a l V e t e r in a r y A s s a y L a b o r a t o r y C o lle c t io n , A n a l y s is , E v a lu a t io n Fig. 2 Monitoring of antimicrobial consumption OIE International Standards on Antimicrobial Resistance, 2003 207 4. Surveillance of resistance programme Bacteria for resistance testing are collected continuously and include: zoonotic bacteria and indicator bacteria isolated from healthy animals; and pathogenic bacteria isolated from diseased animals. Zoonotic bacteria include: Salmonella species, and Campylobacter jejuni or C. coli; indicator bacteria include Escherichia coli including O157 and Enterococcus faecium or E. fecalis, including Vancomycin-Resistant Enteroccoci. Animal pathogens included at present are Salmonella species, Staphylococcus aureus, Actinobacillus pleuropneumoniae, Actinobacillus pyogenes, Pasteurella multocida, Streptococcus species and Klebsiella species. The zoonotic and indicator bacteria are isolated from faecal samples collected from cattle, pigs, broilers and layers. Six samples from animals are collected in each prefecture every year with a limit of one sample per farm. Two strains per sample are collected for antimicrobial susceptibility testing. Animal pathogens are isolated from samples submitted for diagnosis. Minimum Inhibitory Concentrations (MICs) of test bacteria are determined for antimicrobials mainly by the agar dilution method as described by the National Committee for Clinical Laboratory Standards (3). The Japanese veterinary antimicrobial resistance monitoring implementation system The JVARM implementation system is shown in Figure 3. A total of one hundred and ninety-five Livestock Hygiene Services Centers (LHSC), which belong to prefecture offices, participate in JVARM. The LHSC function as participating laboratories of JVARM, and are responsible for the isolation and identification of target bacteria, as well as MIC measurement. They send results and resistant bacteria to NVAL, which functions as the reference laboratory of JVARM, and is responsible for preservation of resistant bacteria, collection and analysing all data and reporting to MAFF headquarters. In addition, NVAL conducts research into the molecular epidemiology and resistance mechanisms of the bacteria. MAFF Administrative action Report National Veterinary Assay Laboratory Announcement Preservation of resistant bacteria Distribution of reference strains Molecular epidemiology, resistance mechanisms Collection, analysis and evaluation of prefecture data Livestock Hygiene Services Center Sampling Isolation/Identification MIC measurement Food-producing Animal Cattle, Swine, Broiler, Layer Fig. 3 Monitoring of antimicrobial resistant bacteria 208 OIE International Standards on Antimicrobial Resistance, 2003 4. Surveillance of resistance programme Quality assurance/quality control systems Quality control procedures are implemented in participating laboratories that perform antimicrobial susceptibility testing to help monitor the precision and accuracy of the test procedure, the performance of the appropriate reagents and the personnel involved. Strict adherence to standardised techniques is necessary for the collection of reliable and reproducible data from participating laboratories. Quality control reference bacteria are also tested in each participating laboratory to ensure standardisation. Moreover, NVAL holds a national training course on antimicrobial resistance every year to provide training in standardised laboratory methods for the isolation, identification and antimicrobial susceptibility testing of target bacteria. Recently, proficiency testing of participating laboratories has been initiated for the major bacterial species included in JVARM. The participating laboratories test these strains using the same conditions as the antimicrobial susceptibility test. Proficiency testing is one of the foundations of quality assurance for participating laboratories in JVARM and ensures that reported MIC data are accurate without question. Announcement of data Since a problem with antimicrobial resistance directly influences animal and human health, it is of paramount importance to distribute information on antimicrobial resistance as soon as possible. We have officially taken three steps to publicise such information; initially through the MAFF weekly newspaper called ‘Animal Hygiene News’, then by publication in scientific journals and via the NVAL website (URL http://www.nval.go.jp/taisei/taisei.html). Although JVARM was started in 1999 and conforms to the OIE report on antimicrobial resistance (1, 5), further steps could be taken to ensure animal and public health in Japan. In particular, several countries have initiated national monitoring systems that include both animal and public health, but at present there is neither a global monitoring system in Japan nor coordination between these areas. Joint efforts are now needed to establish a national antimicrobial monitoring system that includes both animal and public health to solve the emerging problem of antimicrobial resistance. References 1. Franklin A., Acar J., Anthony F., Gupta R., Nicholls T., Tamura Y., Thompson S., Threlfall E.J., Vose D., van Vuuren M., White D.G., Wegener H.C. & Costarrica M.L. (2001). – Antimicrobial resistance: harmonisation of national antimicrobial resistance monitoring and surveillance programmes in animals and in animal-derived food. Rev. sci. tech. Off. int. Epiz., 20 (3), 859-870. 2. OIE (World organisation for animal health) (1999). – The use of antibiotics in animals ensuring the protection of public health. In Proceeding of European Scientific Conference 2426 March, Paris. 3. Shryock T.R., Apley M., Jones R.N., Lein D.H., Thornsberry C., Walker R.D., Watts J.L., White D.G. & Wu C.C. (2002). – Performance standards for antimicrobial disk and dilution susceptibility tests for bacteria isolated from animals. Approved standard, 2nd Ed.. NCCLS M31-A2. OIE International Standards on Antimicrobial Resistance, 2003 209 4. Surveillance of resistance programme 4. Swann M.M. (1969). – Report of the joint committee on the use of antibiotics in animal husbandry and veterinary medicine. HM Stationary Office. 5. White D.G., Acar J., Anthony F., Franklin A., Gupta R., Nicholls T., Tamura Y., Thompson S., Threlfall E.J., Vose D., van Vuuren M., Wegener H.C. & Costarrica M.L. (2001). – Antimicrobial resistance: standardisation and harmonisation of laboratory methodologies for the detection and quantification of antimicrobial resistance. Rev. sci. tech. Off. int. Epiz., 20 (3), 849-858. 6. World Health Organization (WHO) (1997). – The medical impact of the use of antimicrobials in food animals. Report of WHO meeting, 13-17 October, Berlin. 7. World Health Organization (WHO) (1998). – Use of quinolones in food animals and potential impact of human health. Report of WHO meeting, 2-5 June, Geneva. __________ 210 OIE International Standards on Antimicrobial Resistance, 2003 4. Surveillance of resistance programme Cases of antimicrobial resistance to some pathogens in Vietnam T.T.T. Phuong 1 SVSV Project supported by European Community, Ho Chi Minh City, Vietnam Introduction Reports from all over the world, indicating a steady increase in antimicrobial resistance of micro-organisms pathogenic for both humans and animals, are causing concern within the human and veterinary medical professions. Finding answers to this important problem presents challenges. In Vietnam, little research has been carried out so far in this field, but the fact that this country is also confronted with the problem of antimicrobial resistance is demonstrated by the few reported cases that are presented in this paper. They should be a stern warning to us and a good reason to step up our national pharmaco-vigilance. The use of antibiotics in animal production in Vietnam Market statistics of veterinary drugs in recent years indicate widespread use of antibiotics in Vietnam. Their share of the total veterinary drug sales is around 50%. There is a great variety of different antibiotics produced in and imported into Vietnam. Drugs often contain a cocktail of different antibiotics and little is known about their therapeutic effect. They are used for therapeutic treatment and also as animal feed additives with two functions: sub-therapeutic prophylaxis and as a growth promoter. Results from some investigations on antimicrobial resistance Sensitivity to antibiotics of some pathogenic bacteria in animal health Example 1 Sensitivity of mastitis causing bacteria to antimicrobial drugs In 1998, the University of Agriculture and Forestry in Ho Chi Minh City carried out an investigation (2) on the sensitivity to several antibiotic substances of bacteria (Staphylococcus aureus, Streptococcus agalactiae and E. coli). The samples were derived from cows with subclinical mastitis (Table I). 1 Representative Office of Department of Animal Health, Ho Chi Minh City, Vietnam OIE International Standards on Antimicrobial Resistance, 2003 211 4. Surveillance of resistance programme Example 2 Sensitivity to antimicrobial drugs of pathogenic bacteria isolated from pigs A sensitivity investigation on a total of 675 samples from pigs was carried out in 1998-2000 by the University of Agriculture and Forestry in Ho Chi Minh City (3) (Table II). Example 3 Sensitivity to antimicrobial drugs of pathogenic bacteria isolated from poultry The University of Agriculture and Forestry in Ho Chi Minh City also carried out in 1998-2000 a sensitivity investigation on a total of 237 samples from poultry (3). The effectiveness of antimicrobial substances against pathogenic bacteria in poultry also varies widely and is generally low (Table III). Table I Sensitivity (as a percentage of the relevant milk samples) of Staphylococci, Streptococci and E. coli to different antimicrobial substances Antibiotic Cephalexin Gentamycin Chloramphenicol Norfloxacin Bactrim Staphylococci (n = 50) Streptococci (n = 67) E. coli (n = 28) 70 68 56 56 56 35.82 32 50 29 75 50 33 10 21 Penicillin, Ampicillin, Amoxicillin, Neomycin, Kanamycin, Streptomycin, Erythromycin, Tetracyclin, Colistin were less than 40% effective against Staphylococci, Streptococci, E. coli. Table II Sensitivity (as a percentage of the relevant samples from pigs) of different pathogenic bacteria to different antimicrobial substances Antibiotic Staphylococcus spp Streptococcus spp n = 38 n = 99 n = 320 E. coli Enterobacter spp n = 14 Cephalexin Gentamycin Norfloxacin Bactrim 47.1 32.5 45 32.5 49.5 0 17.7 23.9 23.6 46.4 48.8 19.7 9 63.6 90.9 54.5 Flumequin Chloramphenicol 0 32.5 0 36.3 9.4 19.7 45.5 90.9 212 OIE International Standards on Antimicrobial Resistance, 2003 4. Surveillance of resistance programme Ampicillin, Amoxycillin, Penicillin, Neomycin, Kanamycin, Colistin, Streptomycin, Erythromycin, Tetracyclin were from 0% to 36.4% effective against Staphylococcus spp, Streptococcus spp, E. coli, Enterobacter spp. Table III Sensitivity (as a percentage of the relevant samples from poultry) of different pathogenic bacteria to different antimicrobial substances Antibiotic Gentamycin Neomycin Kanamycin Norfloxacin Bactrim Chloramphenicol Colistin Tetramycin E. coli (n = 105) Salmonella spp (n = 35) 43.8 9.5 7.6 20 13.3 18.1 25.7 2.9 78.8 63.6 51.5 69.7 48.5 57.6 48.5 45.5 The effectiveness of Ampicillin, Amoxycillin, Penicillin, Cephalexin, Flumequin, Streptomycin, Erythromycin against pathogenic bacteria was below 40%. Sensitivity to antibiotics of some pathogenic bacteria in human health Over the last few years in Vietnam, a number of clinical and basic laboratory studies have been jointly carried out by the Center of Tropical Diseases (CTD) in Ho Chi Minh City and the University of Oxford supported by the Wellcome Trust. They focused on diseases that have a considerable impact on the health of the community, such as malaria, pneumonic disease and typhoid fever. Antimicrobial resistance has been detected in a variety of pathogenic bacteria in some provinces of Southern Vietnam. According to the CTD experts’ 1997 report (4, 7), more than 90% of Salmonella typhi bacteria were multi-drug resistant and the sensitivity to fluoroquinolons in some areas had decreased to only 50%. As in many other countries, penicillin-resistant Streptococcus pneumoniae (PRP) is also a significant problem in Vietnam. 50% of Streptococcus pneumonia (thirty-four strains) isolated from blood and cerebrospinal fluid (CSF) exhibited a low level of resistance to penicillin, 50% were tetracycline-resistant and 26% carried by healthy children were highly resistant (5). 15% of Corynebacterium diptheriae were found to be resistant to erythromycin, and some strains were multi-drug resistant (4). 10% of Mycobacterium tuberculosis isolates were tested as resistant to two or more antituberculosis drugs (4). OIE International Standards on Antimicrobial Resistance, 2003 213 4. Surveillance of resistance programme The resistance to several antimicrobial substances of sixty-two strains of Staphylococcus aureus isolated from blood cultures at the CTD (6) found that Staphylococcus aureus was highly resistant to penicillin (97%) and quite resistant to erythromycin. Residues of antibiotics in animal products Research results obtained from some provinces in South Vietnam, by the Veterinary Faculty of the University of Agriculture and Forestry in Ho Chi Minh City between March 1999 and April 2000, showed that 19.54% of the 87 samples taken from chicken meat, pork and beef contained residues of antibiotics (1). The following quantities of different antimicrobial substances were measured in seventeen samples: Erythromycin: 2.12 mg/kg Streptomycin: 0.14 mg/kg Gentamycin: 0.75 mg/kg Kanamycin: 5.17 mg/kg Chloramphenicol: 3.04 mg/kg (chicken meat), 3.04 – 111.5 mg/kg (pork) Sulfamethazol: 119.47 mg/kg (chicken meat), 35.07 – 119.47 mg/kg (pork) Current difficulties in regulating the use of veterinary drugs Since 1993, the legal framework for regulating the production, trade and application of veterinary medicinal products in Vietnam has been the Veterinary Ordinance on Veterinary Services. The Government of Vietnam is determined to improve the quality and the control of veterinary medicinal products, including implications for the safety of food of animal origin. However, this process is taking time and the public veterinary services face many difficulties. In this context it should be mentioned that the situation in Vietnam with regard to medicinal products for application in humans is not much better than in the veterinary sector. Anybody can buy, for instance, antimicrobial drugs from pharmacies without prescription. Quality control during production, distribution and storage is deficient, and improper application widespread. Therefore, it would be prudent if national efforts were harmonised in regulating medicinal products for both the medical and veterinary sectors. Hopefully the pharmaceutical industry will come up with new drugs potent against drug-resistant microbes. This will inevitably take time and will predictably be expensive. Therefore, it is of paramount importance to make prudent use of the existing antimicrobial products to avoid the development of antimicrobial resistance. References 1. Do C.D. & Nguyen T.T.G. (2000). – Bacterial infection and residues of antibiotics in animal products. In Summary of Science Researching Conference of Univeristy of Agriculture and Forestry. Univeristy of Agriculture and Forestry, Ho Chi Minh City, 4-9. 214 OIE International Standards on Antimicrobial Resistance, 2003 4. Surveillance of resistance programme 2. Nguyen V.P. et al. (2000). – Isolating pathogen bacteria in milk samples from cows with latent mastitis. In Summary 3rd Science Conference of Husbandry and Veterinary Faculty. University of Agriculture and Forestry, Ho Chi Minh City, 113-116. 3. Nguyen V.P. et al. (2000). – Isolating pathogen bacteria from pig and poultry specimens at the Veterinary Clinic. In Summary 3rd Science Conference of Husbandry and Veterinary Faculty. University of Agriculture and Forestry, Ho Chi Minh City, 124-127. 4. Parry C.M. (1997). – Antimicrobial resistance: the challenge of the 21th Century. In Symposium on antimicrobial resistance in Southern Viet Nam. Center of Tropical Diseases, Ho Chi Minh City, 5-6. 5. Parry C.M. et al. (1997). – Penicilline resistance in Streptococcus pneumonia isolated from aldults and children in HCMC. In Symposium on antimicrobial resistance in Southern Viet Nam. Center of Tropical Diseases, Ho Chi Minh City, 9-10. 6. To S.D. et al. (1997). – Antibiotic resistance of Staphylococcus aureus at CTD 1993-1997. In Symposium on antimicrobial resistance in Southern Viet Nam. Center of Tropical Diseases, Ho Chi Minh City, 7-8. 7. Tran T.H. (1997). – Treatment options for typhoid fever due to multidrug-resistant Salmonella typhi. In Symposium on antimicrobial resistance in Southern Viet Nam. Center of Tropical Diseases, Ho Chi Minh City, 1-4. __________ OIE International Standards on Antimicrobial Resistance, 2003 215 4. Surveillance of resistance programme Antimicrobial resistance surveillance in Kenya: achievements and challenges S. Kariuki (1), G. Revathi (2) & C.A. Hart (3) (1) Centre for Microbiology Research, KEMRI, P.O. Box 43640, Nairobi, Kenya (2) Department of Medical Microbiology, Kenyatta National Hospital, P.O. Box 20723, Nairobi, Kenya (3) Department of Medical Microbiology and Genito-Urinary Medicine, University of Liverpool, Liverpool, L69 3GA, United Kingdom General introduction Although progress has been achieved in the therapeutic management of many infectious diseases, the issue of antimicrobial resistance continues to be a major threat to this achievement. Low resource countries in particular are at greatest risk of drugresistant infections, as paucity of resources makes the purchasing of newer and more effective treatment agents difficult. Thus, it would be important to carefully manage available therapeutic choices in order to ensure their continued use in the treatment of infections. Emerging antimicrobial resistance is an enormous health problem and more so in low resource countries, where the bulk of communicable diseases are found and the resources to combat them are meagre. Left unchecked, this problem will adversely affect our ability to treat and control infectious diseases. In Kenya, the problem of antimicrobial resistance has been recognised for a number of years, and this led to the first ever workshop on surveillance of antimicrobial resistance and rational use of antimicrobial agents held in 1997, with a follow up meeting in 1999. Background Several issues concerning antimicrobial resistance were raised during these meetings. Many laboratories at the Provincial and District level used obsolete methods such as the modified Stokes in susceptibility testing. There was therefore need to adopt the widely recommended Kirby-Bauer method and recommendations by the National Committee for Clinical Laboratory Standards (NCCLS), which should be supported by regular updates on NCCLS protocols (NCCLS, 2000). In addition, interlaboratory communication was poor and laboratory-to-clinician rapport was sub-optimal, thus contributing to the lack of utilisation of available laboratory capabilities. Participants also raised other issues concerning the practicalities of carrying out susceptibility testing. Among these concerns were inadequate numbers of suitably trained personnel and the lack of a credible internal and external quality assurance programme. Many laboratories, particularly at District level, lacked access to basic media, antibiotic discs, petri dishes and other materials essential for susceptibility testing of common bacterial pathogens. In addition, equipment such as autoclaves, incubators and microscopes were either lacking or broken down. As the collection of specimens was done by laboratory technical personnel this needed to be well 216 OIE International Standards on Antimicrobial Resistance, 2003 4. Surveillance of resistance programme supervised to ensure that specimens getting to the laboratory for examination were of high quality. Several matters touching on communication were also debated. There was a consensus that information on antibiotic susceptibility testing needed to be shared between laboratories and epidemiologists; microbiologists, clinicians, pharmacists and laboratories in various hospitals; clinicians, microbiologists and policy makers. To make this recommendation viable hospitals and laboratories were requested to find a forum for exchanging information on antibiotic susceptibility testing and surveillance. In addition continuing medical education for microbiologists, laboratory technologists and technicians would ensure that they are equipped with new and updated information on antimicrobial susceptibility testing, surveillance and control measures. In order to improve susceptibility testing, a number of immediate measures were suggested. Firstly, there was a need to define a national focal point and to develop a situation analysis. Secondly, there was urgent need to ensure quality laboratory training and participation in external quality assurance programmes provided by the World Health Organization (WHO). On data collection, analysis and dissemination, sentinel laboratories should be encouraged to publish data locally in newsletters that should be circulated to interested groups. Other recommendations included the establishment of a full-time programme coordinator, to be based at the reference laboratory, who would offer support to all sentinel sites in internal and external Quality Assurance (QA) programmes. In addition, antibiotic susceptibility testing protocols, media and reagents, and data storage, processing and reporting should be standardised to give reliable data. Achievements in antimicrobial susceptibility testing and monitoring Due to limitations in funding, only a few selected laboratories (based on available manpower, materials and data entry and computational hardware) could begin the susceptibility testing and surveillance programme. Among them were the Centre for Microbiology Research, part of the Kenya Medical Research Institute (Reference laboratory), the Kenyatta referral hospital, two other public hospitals and three private hospitals. Presently, laboratories in these sites participate in the external Quality Assurance programme coordinated through a World Health Organization/Centers for Disease Control and Prevention (WHO/CDC) programme, twice annually. There is also regular informal consultation between the laboratories, particularly in sharing antimicrobial susceptibility testing and surveillance data, and updates on pathogens of importance that would be included in the surveillance programme. In addition, internal quality assurance for each laboratory has been set up ensuring that all use NCCLS recommended standards for antimicrobial susceptibility testing, including using American Type Culture Collection (ATCC) Quality Control strains (NCCLS, 2000). OIE International Standards on Antimicrobial Resistance, 2003 217 4. Surveillance of resistance programme Some data are obtained through the Antimicrobial Susceptibility Testing and Surveillance Programme. The surveillance programme was aimed at monitoring antimicrobial resistance in some commonly isolated pathogens, including, Klebsiella, Staphylococci, Shigella and Salmonella spp. Cefotaxime-hydrolysing Klebsiella spp All twenty-two K. pneumoniae isolates (1999-2000) obtained from outbreaks in neonatal wards were uniformly resistant to ampicillin, cephradine, cefuroxime, cefotaxime, carbenicillin, ceftazidime and tetracycline. However, they were susceptible to coamoxyclav, ceftazidime, aztreonam, streptomycin, co-trimoxazole, gentamicin and nalidixic acid. Isolates had minimal inhibitory concentrations (MICs) of 24 and 1 µg/ml for cefotaxime and ceftazidime, respectively. The presence of clavulanic acid decreased the MIC of cefotaxime 750-fold to 0.032 µg/ml, indicating that resistance was a result of the production of extended-spectrum β-lactamases (Kariuki et al., 2001). Table I Multidrug-resistant Shigella spp. (1999-2000) Antimicrobial agent Ampicilin Co-amoxyclav Piperacilin Cephradine Cefuroxime Ceftazidime Imipenen Gentamicin Cotrimoxazole Chloramphenicol Sonnei (n = 32) 84 5 100 8 6 0 0 0 80 30 Percentage of resistance Flexneri (n = 34) Dysenteriae (n = 6) 76 0 64 12 6 0 0 0 80 10 100 0 100 0 0 0 0 0 100 0 Emergence of multidrug resistant Salmonella typhi, Kenya Between 1998 and 2000, we studied 87 S. typhi isolated from blood cultures of adults admitted to various hospitals in Nairobi. Only 15.3% were fully sensitive while the rest (84.7%) were resistant to all five commonly available drugs – ampicillin, chloramphenicol, tetracycline (MICs > 256 µg/ml), streptomycin (MIC > 1,024 µg/ml) and cotrimoxazole (MIC > 32µg/ml). For resistant S. typhi MICs for nalidixic 218 OIE International Standards on Antimicrobial Resistance, 2003 4. Surveillance of resistance programme acid and ciprofloxacin were respectively 5- and 10-fold higher than for sensitive strains (Kariuki et al., 2000). Multidrug-resistant non-typhi Salmonella from bacteraemic cases Non-typhi Salmonella from HIV-seropositive patients has recently been one of the major opportunistic pathogens to be isolated from blood. Our surveillance data indicates that up to 64% of these isolates are multi-resistant to two or more of the commonly used antimicrobials (Kariuki et al., 2000). Table II MIC using the E-Test of eleven antimicrobial agents for 151 non-typhi Salmonella isolates from medical wards at two hospitals in Nairobi (1997-2000) Antimicrobial agent Ampicillin Co-amoxyclav Cefuromixime Ceftazidime Co-trimozadole Chloramphenicol Ciprofloxacin Gentamicin Nalidixic acid Streptomycin Tetracycline Range Minimum inhibitory concentration (µg/ml) MIC90 % resistant Mode MIC50 0.75->256 0.5-32 2-128 0.125-16 0.032->32 1.5->256 0.006-0.25 0.19-64 1->256 3->1,024 0.75-192 >256 0.75 3 0.25 >32 >256 0.023 0.75 3 32 1 >256 6 8 0.5 >32 32 0.023 1 3 >1,024 16 >256 16 12 2 >32 >256 0.125 8 >256 >1,024 64 65 8 18 1 60 40 0 9 11 90 48 Limitations to achieving a nation-wide antimicrobial susceptibility testing and surveillance programme As there is no formal funding for the programme, distribution of materials from the external quality assurance programme is limited to the selected sentinel sites. Indeed several other laboratories interested in participating are locked out due to paucity of funding. Even for the participating laboratories there is a need for training support for their staff in order to undertake quality antimicrobial susceptibility testing and surveillance. References 1. Kariuki S., Corkill J.E., Revathi G., Musoke R. & Hart C.A. (2001). – Molecular characterization of a novel plasmid-encoded cefotaximase (CTX-M-12) found in clinical Klebsiella pneumoniae isolates from Kenya. Antimicrob. Agents Chemother., 45, 2141-2143. OIE International Standards on Antimicrobial Resistance, 2003 219 4. Surveillance of resistance programme 2. Kariuki S., Gilks, Revathi G. & Hart C.A. (2000). – Genotypic analysis of multidrugresistant Salmonella enterica serovar Typhi, Kenya. Emerg. infect. Dis., 6 (6), 649-651. 3. Kariuki S., Oundo J.O., Muyodi J., Lowe B., Threlfall E.J. & Hart C.A. (2000). – Genotypes of multidrug-resistant Salmonella enterica serotype typhimurium from two regions of Kenya. FEMS Immun. Med. Microbiol., 29, 9-13. 4. National Committee for Clinical Laboratory Standards (NCCLS) (2000). – Methods for dilution antimicrobial susceptibility tests for bacteria that grow aerobically. In Approved standard, 5th Ed. NCCLS document M7-A5, Wayne, Pa. 5. National Committee for Clinical Laboratory Standards (NCCLS) (2000). – Performance standards for antimicrobial disk susceptibility tests. In Approved standard, 7th Ed. NCCLS document M2-A7, Wayne, Pa. __________ 220 OIE International Standards on Antimicrobial Resistance, 2003 5. Laboratory methods OIE International Standards on Antimicrobial Resistance, 2003 221 5. Laboratory method Antimicrobial resistance: standardisation and harmonisation of laboratory methodologies for the detection and quantification of antimicrobial resistance D.G. White (1), J. Acar (2), F. Anthony (3), A. Franklin (4), R. Gupta (5), †T. Nicholls (6), Y. Tamura (7), S. Thompson (8), E.J. Threlfall (9), D. Vose (10), M. van Vuuren (11), H.C. Wegener (12) & M.L. Costarrica (13) (1) Centre for Veterinary Medicine, Food and Drug Administration, Office of Research, HFV-530, 8401 Muirkirk Road, Laurel, Maryland 20708, United States of America (2) Université Pierre et Marie Curie, Service de Microbiologie Médicale, Fondation Hôpital Saint-Joseph, 185 rue Raymond Losserand, 75674 Paris Cedex 14, France (3) Fresh Acre Veterinary Surgery, Flaggoners Green, Bromyard, Herefordshire HR7 4QR, United Kingdom (4) The National Veterinary Institute (SVA), Department of Antibiotics, SE 751 89 Uppsala, Sweden (5) College of Veterinary Sciences, Veterinary Bacteriology, Department of Microbiology, G.B. Pant University of Agriculture and Technology, Pantnagar 263 145 Uttar Pradesh, India (6) National Offices of Animal and Plant Health and Food Safety, Animal Health Science and Emergency Management Branch, Department of Agriculture, Fisheries and Forestry, P.O. Box 858, Canberra ACT 2601, Australia (7) National Veterinary Assay Laboratory, Ministry of Agriculture, Forestry and Fisheries, 1-51-1 Tolura, Kokubunji, Tokyo 185-8511, Japan (8) Joint Institute for Food Safety Research, Department for Health and Human Services Liaison, 1400 Independence Avenue, SW, Mail Stop 2256, Washington, DC 20250-2256, United States of America (9) Public Health Laboratory Service, Central Public Health Laboratory, Laboratory of Enteric Pathogens, 61 Collindale Avenue, London NW9 5HT, United Kingdom (10) David Vose Consulting, Le Bourg, 24400 Les Lèches, France (11) University of Pretoria, Faculty of Veterinary Science, Department of Veterinary Tropical Diseases, Private Bag X04, Onderstepoort 0110, South Africa (12) World Health Organization, Detached National Expert, Division of Emerging and Transmissible Diseases, Animal and Food-related Public Health Risks, 20 avenue Appia, 1211 Geneva, Switzerland (13) Food and Agriculture Organization, Food Quality and Standards Service, Senior Officer, via delle Terme di Caracalla, 00100 Rome, Italy This report, prepared by the OIE Ad hoc Group of experts on antimicrobial resistance, has not yet received the approval of the International Committee of the OIE Summary The Ad hoc Group of experts on antimicrobial resistance of the OIE (World organisation for animal health) has developed a guideline on the standardisation and harmonisation of laboratory methodologies used for the detection and quantification of antimicrobial resistance. The existing methods (disk diffusion [including concentration gradient strips], agar dilution and broth dilution) are reviewed, including a comparison of their advantages and disadvantages. The definitions of resistance characteristics of bacteria (susceptible, intermediate and resistant) are addressed and the criteria for the establishment of breakpoints are discussed. Due consideration has to be given to these aspects in the OIE International Standards on Antimicrobial Resistance, 2003 223 5. Laboratory method interpretation and comparison of resistance monitoring or surveillance data. The use of validated laboratory methods and the establishment of quality assurance (internal and external) for microbiological laboratory work and the reporting of quantitative test results is recommended. Equivalence of different methods and laboratory test results is also recommended to be established by external proficiency testing, which should be achieved by the means of a reference laboratory system. This approach allows the comparison of test results obtained using different methods generated by laboratories in different countries. Keywords Antimicrobial resistance – Breakpoints – Containment of resistance – Harmonisation – International standards – Laboratory methodology – Public health – Risk analysis – Standardisation – Threshold – World Organisation for Animal Health. Introduction The objective of this document is to review currently used antimicrobial susceptibility testing methodologies and protocols and to encourage the Member Countries of the OIE (World organisation for animal health) to initiate standardisation and harmonisation of bacterial antimicrobial susceptibility testing and results. The similarities, differences, advantages and disadvantages of accepted standardised antimicrobial susceptibility testing methods are described. Additionally, the requirements of each antimicrobial susceptibility testing method are discussed (equipment, training, resources and quality assurance). The need for internal quality control and external proficiency testing is emphasised. Standardisation and harmonisation of antimicrobial susceptibility testing methodologies are critical if data is to be compared among the international surveillance/monitoring programmes of OIE Member Countries. Background There is increasing international concern regarding both the potential transfer of antimicrobial resistant bacteria between animals and humans and of resistance genes from animal strains of bacteria to human bacterial pathogens. Concern about antimicrobial resistance in relation to animal health is also growing. In response to these concerns, antimicrobial resistance testing initiatives, together with surveillance and monitoring programmes focusing on zoonotic bacterial pathogens and enteric commensals in animals have been initiated in numerous countries throughout the world (3, 5, 21). Data generated from these surveillance and monitoring programmes will eventually play a key role in the development of national, and perhaps international polices for the containment of antimicrobial resistant bacterial pathogens from animals and their immediate environments. The need to compare susceptibility testing data between laboratories in different countries necessitates a re-examination of the standardisation and harmonisation of the antimicrobial susceptibility testing (AST) methods currently in use world-wide (12). 224 OIE International Standards on Antimicrobial Resistance, 2003 5. Laboratory method Historically, veterinarians and medical practitioners selected effective antimicrobials based on past clinical experiences. However, with an observed increase in bacterial resistance to regularly used antimicrobials, it has become gradually more difficult for clinicians to empirically select an appropriate antimicrobial agent (13, 24). As a result, laboratory in vitro AST of the relevant bacterial pathogens from properly collected specimens is currently standard procedure (13, 17, 25). Antimicrobial susceptibility testing was initiated in many countries world-wide soon after the introduction of antimicrobials for treatment of bacterial diseases (12). Rapid bacterial identification systems and subsequent improvements in AST in both human and veterinary clinical laboratories were primarily driven by the need to identify the appropriate antimicrobials for successful clinical use. Additionally, the need for laboratory reproducibility of AST methods arose to ensure that data generated was technically accurate and consistent. This required that AST laboratories adopt quality control measures to guarantee the reporting of reliable and reproducible susceptibility data (9, 17). Although protocols for bacterial identification, AST and data analysis developed very rapidly, standardisation and validation of these three procedures is relatively recent compared to the progress achieved in analytical chemistry. Historically, most laboratories have employed disk diffusion methods for AST. Reported results can be quantitative if zone diameters are recorded, but they are generally reported qualitatively as either susceptible, intermediate, or resistant (9, 14, 17, 25). In the past few years, many laboratories have adopted either broth microdilution or agar dilution methods (9, 11). Results from these assays may be quantitative, in that they provide the minimal concentration of an antimicrobial required to inhibit the growth of the test organism (minimum inhibitory concentration [MIC]), as well as providing a qualitative description (susceptible, intermediate and resistant). Some laboratories have not been as successful in adopting these methods, primarily due to training and financial limitations. Additionally, the need to develop and implement quality assurance programmes for bacterial identification and AST is a fairly new concept which may take time to implement. However, some initiatives have been introduced and are currently underway in an attempt to standardise and/or harmonise AST. In veterinary and human medicine, antimicrobial resistance data is being shared between a number of laboratories through the creation of antimicrobial resistance surveillance networks. Some of these networks are linked internationally. This has resulted in the standardisation and harmonisation of AST methods between participating laboratories. Participating laboratories adhere to strict standards of AST and quality control monitoring to ensure accuracy and comparability of the data. Examples of international and national surveillance systems employing standardised methods include the European Antimicrobial Resistance Surveillance System (EARSS), the Alexander Project for Respiratory Pathogens, Antibiotic Resistance in Bacteria of Animal Origin (ARBAO), SENTRY, the Surveillance Network (TSN), the Danish Integrated Antimicrobial Resistance Monitoring and Research Program (DANMAP), the World Health Organization Network on Antimicrobial Resistance OIE International Standards on Antimicrobial Resistance, 2003 225 5. Laboratory method Monitoring (WHONET), Enter-Net and the National Antimicrobial Resistance Monitoring System (NARMS) (3, 15, 16, 21, 22). The success of these surveillance and monitoring programmes suggests that standardisation and harmonisation of AST methods are both conceivable and progressing globally. Although in the past few years there has been a move to standardise AST methods within countries, the move to harmonise methods and susceptibility data among countries has not been initiated on a global scale for bacteria originating from animals. One significant obstacle is the fact that there is no international monitoring system for AST that utilises a single methodology with identical quality control organisms. To obtain comparable antimicrobial susceptibility data from different laboratories in the same country, or in different countries, laboratory methodologies need to be standardised and harmonised. This can be best accomplished if the antimicrobial susceptibility data collected is quantitative (i.e. MIC, zone diameters), rather than qualitative for comparison purposes. Data to be used for epidemiological surveillance purposes must be reported quantitatively in order to both detect shifts in antimicrobial susceptibility in bacterial strains and be comparable with other surveillance programmes. Quantitative in vitro bacterial antimicrobial susceptibility testing is essential for the purpose of monitoring shifts in susceptibility to antimicrobial agents. However, to achieve its aim, testing must be performed according to standardised testing methods. Comparison of the frequency of antimicrobial resistance in bacterial pathogens among the many countries that have surveillance systems in place is difficult for many reasons. Antimicrobial susceptibility testing currently serves two purposes, firstly to provide meaningful results to the clinician and secondly to monitor shifts in susceptibility of targeted bacterial populations (12). Historically, laboratories have been restricted in reporting bacterial AST data as ‘susceptible, intermediate or resistant’. Bacterial antimicrobial susceptibilities reported this way are primarily for the immediate needs of physicians or veterinarians as guidelines for appropriate antimicrobial therapies. Taking into account the different AST protocols and interpretive criteria among the numerous testing methods and guidelines available, it is evident that this type of reporting excludes any possibility for the comparison of susceptibility data. Unfortunately, there is no world-wide consensus on interpretive criteria for susceptibility testing. Additionally, the emphasis of many surveillance programmes is to monitor shifts in antimicrobial susceptibilities in target bacterial pathogens. Since there are no standardised dilution schemes available world-wide, it becomes difficult to compare susceptibility profiles of bacterial pathogens from different countries. Standardisation and harmonisation of AST methods are needed for meaningful comparisons of quality and accurate susceptibility data between individual OIE Member Countries involved in both national and international surveillance programmes. This will be best accomplished by the use of accurate and reliable standardised AST methods in conjunction with monitoring of AST performance with defined quality control bacterial strains among participating laboratories. If results 226 OIE International Standards on Antimicrobial Resistance, 2003 5. Laboratory method achieved with different AST methods are to be presented side by side, then comparability of results must be demonstrated and consensus on interpretation achieved. It is essential that AST methods provide reproducible results in day-to-day laboratory use and that the generated data be comparable to those results obtained by an acknowledged ‘gold standard’ reference method. In the absence of standardised methods or reference procedures, susceptibility results from different laboratories cannot be adequately compared with assurance. Antimicrobial susceptibility testing methodologies It is essential that the bacteria subjected to AST be isolated in pure culture from the submitted sample. The isolation procedure for that particular bacterium should be standardised so that the subject bacteria are consistently and correctly identified to the genus and/or species level. When possible, bacterial isolates should be stored for future analysis via either lyophilisation or cryogenic preservation at –70°C to –80°C. Once the bacterium has been isolated in pure culture, the inoculum must be standardised to obtain accurate susceptibility results, since variations may substantially affect both the qualitative and quantitative endpoint determinations. Other factors influencing AST methods that require standardisation and harmonisation include the composition of the agar and broth media used (pH, cations, thymidine or thymine, use of supplemented media), content of antimicrobial agent in the carrier (disk, strip, tablet), growth and incubation conditions (time, temperature, oxygen), agar depth, and the subsequent interpretive criteria (17, 18, 24). For these reasons, special emphasis needs to be placed on reference procedures and standardised methods, as sufficient reproducibility can be attained only through standardisation. The decisions regarding which antimicrobials to test can be difficult given the vast numbers of antimicrobials available. Testing all antimicrobial agents is neither necessary (since numerous antimicrobials have similar, if not identical, in vitro activities), nor practical (given the economic restraints faced by laboratories). This is further discussed in Antimicrobial resistance: harmonisation of national antimicrobial resistance monitoring and surveillance programmes in animals and in animal-derived food, later in this volume). A wide variety of bacterial AST methodologies are being used by microbiological laboratories around the world. The selection of an AST methodology may be based on numerous factors, such as ease of performance, flexibility, adaptability to automated or semi-automated systems, cost, reproducibility, reliability, accuracy and national preference. However, only three primary methods have been shown to be reproducible and repeatable. These are disk diffusion (including concentration gradient strips), broth dilution and agar dilution. Disk diffusion refers to the diffusion of an antimicrobial agent of a specified concentration from disks, tablets or strips, into solid culture media seeded with a standardised bacterial inoculum. The diffusion of the antimicrobial agent into the seeded culture media results in an antimicrobial gradient. When the concentration of the antimicrobial becomes so dilute that it can no longer inhibit the growth of the test OIE International Standards on Antimicrobial Resistance, 2003 227 5. Laboratory method bacterium, a zone of inhibition is formed. The edge of this zone of inhibition correlates with the MIC for that particular bacterium/antimicrobial combination. In other words, the zone of inhibition correlates inversely with the MIC of the test bacterium. The larger the zone of inhibition, the lower the concentration of antimicrobial required to inhibit the growth of the organisms. However, the MIC cannot always be easily determined using disk diffusion methods, due to the variation of the tested antimicrobial agent concentration at the edge of the zone of inhibition for each drug-bacterium combination (9, 13). It should be emphasised that disk diffusion tests based solely on the presence or absence of a zone of inhibition without regard to the size of the zone of inhibition are not acceptable. Disk diffusion is technically straightforward to perform, reproducible, and does not require expensive equipment. The main advantages of the disk diffusion method are the low cost and the ease in modifying test formats when needed. Although disk diffusion is the simplest and most cost- effective method for AST, many aspects of this method require standardisation, as mentioned previously. Additionally, manual measurement of zones of inhibition may be time-consuming, making this method impractical for some laboratories (2). However, automated zone reading devices are available which can be integrated with laboratory reporting and data handling systems (2, 13). It is important to remember that no more than twelve disks should be placed on one 150 mm agar plate, and no more than five disks on a 100 mm plate (18). Regardless of the number of disks placed on the agar surface, the disks should be distributed evenly so that they are no closer than 24 mm from centre to centre (18). Additionally, bacterial antimicrobial MICs can be obtained from commercially available gradient strips which diffuse a pre-formed antibiotic concentration (4). However, the use of strips containing antimicrobials at predefined concentrations can be very expensive and MIC discrepancies can be found when compared with agar dilution results (4). The aim of the broth and agar dilution methods is to determine the lowest concentration of the assayed antimicrobial that inhibits the growth of the bacterium being tested (MIC, usually expressed in mg/ml or mg/l). However, the MIC does not always represent an absolute value. The ‘true’ MIC is a point between the lowest test concentration that inhibits the growth of the bacterium and the next lower test concentration (18). Antimicrobial ranges should be utilised that encompass both the interpretive criteria (susceptible, intermediate and resistant) and quality control reference organisms. Additionally, laboratory results should take into consideration the antimicrobial concentrations that are achievable in vivo for a specific bacteria/antibiotic combination. Antimicrobial susceptibility dilution methods appear to be more reproducible and quantitative than agar disk diffusion, although antibiotics are usually tested in doubling dilutions which can produce inexact MIC data (11). Any laboratory that intends to use a dilution method and set up its own reagents and antibiotic dilutions must have the ability to obtain, prepare and maintain appropriate stock solutions of reagent grade antimicrobials and to generate working dilutions on a regular basis. It is then essential that such laboratories utilise quality control organisms to assure accuracy and standardisation of their procedures. 228 OIE International Standards on Antimicrobial Resistance, 2003 5. Laboratory method Broth dilution is a technique in which a standardised suspension of bacteria is tested against varying concentrations of an antimicrobial agent (usually doubling dilutions) in a standardised liquid medium. The broth dilution method can be performed either in tubes containing a minimum volume of 2 ml (macrodilution) or in smaller volumes using microtitration plates (microdilution) (18). Numerous microtitre plates containing prediluted antibiotics within the wells are commercially available. The use of these plates with a standardised protocol, including appropriate quality control reference strains, is the most likely choice to achieve standardisation of AST world-wide. Additionally, the use of identical lots of microdilution plates may eliminate potential errors that may arise due to preparation and dilution of the antimicrobials in participating laboratories (18). However, due to the fact that most broth microdilution test panels are prepared commercially, they can be considered less flexible than agar dilution or disk diffusion in adjusting to the changing needs of the surveillance/monitoring programme. Additionally, purchasing the equipment and antimicrobial panels can be quite costly and may not be a choice for laboratories with limited budgets. Agar dilution involves the incorporation of an antimicrobial agent into an agar medium in a geometrical progression of concentrations, followed by the application of a defined bacterial inoculum to the agar surface of the plate. This results in the accurate determination of a MIC for the test bacterium/antimicrobial combination. Agar dilution methods offer several advantages; these include a greater control of the purity of the test bacterium and the ability to test multiple bacteria on the same set of agar plates and at the same time. Another attractive benefit of this technique is the potential to improve the identification of MIC endpoints and extend the antibiotic concentration range as far as necessary. Additionally, it is the only recommended standardised antimicrobial susceptibility testing method for many fastidious organisms, such as anaerobes, Helicobacter and Campylobacter species (14). Agar dilution can also be adapted to semi-automation. Commercially available inoculum-replicators are available and these can transfer between thirty-two and thirty-seven different bacterial inocula to each agar plate (18). Agar dilution is referred to as the ‘gold standard’ of AST; however, the technique requires extensive training of personnel and may be more expensive and labour-intensive than other testing methods. Routine AST methods are best standardised for aerobic and facultative bacteria and antimicrobial agents that are intended for systemic use. These methods have not been standardised and in some cases are not recommended for uncommon or fastidious bacteria, due to potential inaccurate results. Regardless of the AST method used, the procedures must be standardised to ensure accurate and reproducible results. Additionally, appropriate quality control reference organisms need to be tested every time AST is performed, to ensure accuracy of the data. Clearly, the appropriate AST choice will ultimately depend on the growth characteristics of the bacterium in question, for example, disk diffusion should not be used to test anaerobes, Campylobacter, or other bacteria with considerable strain-to-strain variability in growth rates (14, 25). Given the many biological and technical variables that may influence OIE International Standards on Antimicrobial Resistance, 2003 229 5. Laboratory method AST, standardisation is essential for the correct interpretation of generated results. Lastly, if one of these standardised AST methods is to be adopted by a laboratory of an OIE Member Country where it has not been previously used, programmes should be developed to educate and train the appropriate technical staff. In special circumstances, novel test methods and assays may be more appropriate for detection of particular resistance phenotypes than the standardised AST methods described above. For example, chromogenic cephalosporin-based tests (e.g. nitrocefin) or equivalent methods may provide more reliable and rapid results for beta-lactamase determination in certain bacteria compared to traditional AST methods (18). Extended-spectrum beta-lactamase (ESBL) activity in certain bacteria can also be detected by using standard disk diffusion susceptibility test methods utilising specific cephalosporins (cefotaxime and ceftazidime) in combination with a beta-lactamase inhibitor (clavulanic acid) and measuring the resulting zones of inhibition (20). Additionally, chloramphenicol resistance attributed to production of chloramphenicol acetyl transferase (CAT) can be detected in some bacteria via rapid tube or filter paper tests within 1 h to 2 h (18). Interpretation of antimicrobial susceptibility testing results The objective of in vitro AST is to predict the way in which a bacterial pathogen may respond to the antimicrobial agent in vivo. The results generated by bacterial in vitro antimicrobial susceptibility tests, regardless of whether disk diffusion or dilution methods are used, are generally reported as resistant, susceptible or intermediate to the action of a particular antimicrobial. Resistant implies that the bacterium would not respond to treatment with that particular antimicrobial agent at the usually achievable systemic concentrations and/or possesses a specific resistance mechanism. Susceptible implies that the antimicrobial agent should be successful in treating the bacterial infection with the recommended dosage. Intermediate indicates that the antimicrobial agent may be successful in treating the bacterial infection if high levels of the agent can be achieved at the site of infection. The term intermediate also indicates a buffer zone, which prevents bacterial strains exhibiting borderline susceptibility from being misconstrued as resistant. Similarly, it can serve to indicate that treatment failure may occur, even though the strains exhibit MICs which are below the theoretical treatment levels for a particular antimicrobial. These designations are obtained by determining in vitro breakpoints, those MICs or zones of inhibition at which a bacterium is considered to be susceptible, intermediate or resistant, based on both obtainable serum concentrations of the antimicrobial agent administered at therapeutic doses and through clinical trials (17, 24). A susceptible breakpoint implies that the recommended dosage of the antimicrobial agent will attain serum or tissue concentrations adequate to inhibit the growth of the bacterium in vivo. Intermediate breakpoints represent ‘buffer zones’, in which unforeseen laboratory technical problems inadvertently categorise a susceptible bacterium as resistant, or vice versa. Resistant breakpoints represent those antimicrobial concentrations that cannot be achieved in the host using normal dosing regimes. 230 OIE International Standards on Antimicrobial Resistance, 2003 5. Laboratory method Two primary factors enable a bacterium to be interpreted as susceptible or resistant to an antimicrobial agent. The first factor is the development and establishment of quality control (QC) ranges, using diffusion when possible and dilution testing, for QC micro-organisms. This is essential for validating the specific AST method used. The QC ranges for the QC micro-organisms must be established prior to the development of the second factor, which is the determination of the appropriate interpretive criteria. The determination of the interpretive criteria involves the generation of three distinct pieces of data, population distribution of relevant microorganisms, pharmacokinetic parameters of the antimicrobial agent, and results of clinical trials and experience (1, 24). Interpretation of the data involves creating a scattergram from the bacterial population distribution (300-600 representative bacterial isolates), by plotting the zone of inhibition against the MIC for each bacterial pathogen and calculating a linear regression line (19). The selection of breakpoints is then based on multiple factors, including regression line analysis, bacterial population distributions, error rate bounding, pharmacokinetics, and ultimately, clinical verification (1, 18, 24). Antimicrobial susceptibility breakpoints derived by professional societies or regulatory agencies in various countries are often very similar. However, there can be notable breakpoint differences among different countries for the same antimicrobial agent. These differences may be due to many factors, such as variation in technical AST factors (inoculum density, test media and test method), and the fact that different countries use different dosages or administration intervals for some antimicrobials. Some countries are also more conservative in setting interpretive criteria for specific antimicrobials. Additionally, it is important to remember that interpretive criteria developed for human clinical medicine are not always relevant for veterinary use, as pharmacokinetics, pharmacodynamics, and relevant infectious agents may differ significantly (18). The development of a concept known as ‘microbiological breakpoints’, which is based on the population distributions of the specific bacterial species tested, may be more appropriate for some antimicrobial surveillance programmes. In this case, bacterial isolates that deviate from the normal susceptible population would be designated as resistant, and shifts in susceptibility to the specific antimicrobial/bacterium combination could be monitored. Standardisation and harmonisation of antimicrobial susceptibility testing methodologies The most effective approach for the local, national and international surveillance of antimicrobial resistance would be for all participating OIE Member Country laboratories to use a common AST method, including similar quality control reference organisms and ranges. However, since there are several variations in methodologies, techniques and interpretive criteria currently being used, this will not be an easy task. A number of guidelines are currently available for antimicrobial susceptibility testing and subsequent interpretive criteria throughout the world. These include standards and guidelines published by the National Committee for Clinical Laboratory Standards OIE International Standards on Antimicrobial Resistance, 2003 231 5. Laboratory method (NCCLS), the Japan Society for Chemotherapy (JSC), the Swedish Reference Group for Antibiotics (SIR), Deutsches Institut für Normung (DIN), Comité de l’Antibiogramme de la Société française de Microbiologie (CASFM), Werkgroep richtlijnen gevoeligheidsbepalingen (WRG system, the Netherlands), the British Society for Antimicrobial Chemotherapy (BSAC) and others (6, 7, 8, 10, 18, 26). Because of the variations in diffusion and dilution AST methods and the differing interpretive criteria among the many countries (i.e. choice of agar medium, inoculum size, growth conditions and susceptibility breakpoints), comparison of susceptibility data from one system to another is difficult. Additionally, as mentioned earlier, the majority of the interpretive criteria was developed by AST of bacteria and antimicrobials relevant to human medical pharmacokinetics and there are few breakpoints for many veterinary antimicrobials that may be included in surveillance and monitoring programmes. These data may also not be directly applicable to veterinary medicine in terms of standardisation of testing of animal bacterial isolates. It appears that only the NCCLS has developed protocols for susceptibility testing of bacteria of animal origin and determination of interpretive criteria (18, 19). However, protocols and guidelines are available for susceptibility testing for similar bacterial species which cause infections in humans. It is possible that such guidelines can be adopted for susceptibility testing for bacteria of animal origin, but each country must evaluate its own AST standards and guidelines. Additionally, efforts focusing on harmonisation of susceptibility breakpoints on an international scale are progressing. These efforts have primarily focused on the adoption of the standards and guidelines of the NCCLS, which provide laboratories with standardised methods and quality control values enabling comparisons of AST methods and generated data. For those OIE Member Countries that have not standardised AST methods, the adoption of NCCLS guidelines and standards would be an appropriate initial step. To determine the comparability of results originating from different surveillance systems from OIE Member Countries, antimicrobial susceptibility test results must be reported quantitatively, including information on the methods, quality control organisms and ranges tested. Also essential is agreement upon which micro-organisms are to be susceptibility tested (e.g. Campylobacter species, for which no susceptibility testing methods have currently been published and the choice of antimicrobials to be tested is under discussion). Minimum inhibitory concentration values or zone diameters should be the desired outcome of AST testing to be able to determine shifts in antimicrobial susceptibility among the target bacterial pathogens. This can be achieved by either broth or agar dilution methods, or by statistical transformation of the zone of inhibition diameters obtained by disk diffusion methods to MICs (1). Quantitative data can then be transformed into contingency tables or histograms for comparative purposes and analysis. Regardless of the AST method used, laboratories engaged in antimicrobial susceptibility testing must give high priority to both producing and reporting technically accurate data. Additionally, susceptibility data should be stored electronically in databases, when possible, with additional descriptive 232 OIE International Standards on Antimicrobial Resistance, 2003 5. Laboratory method information regarding the origin of bacterial strains tested and other appropriate details. Quality control and quality assurance of antimicrobial susceptibility testing The implementation of quality control in laboratories that perform AST aims to help to monitor the precision and accuracy of the AST procedure, the performance of the appropriate reagents, and the personnel involved (18). Strict adherence to standardised techniques in conjunction with quality control of media and reagents is necessary for the collection of reliable and reproducible antimicrobial susceptibility data from OIE Member Country laboratories. Records should be kept regarding lot numbers and expiration dates of all appropriate materials and reagents used in AST. The appropriate quality control reference bacteria must also be tested to ensure standardisation regardless of the AST method used. Reference bacterial strains to be used for quality control should be catalogued and characterised with stable defined antimicrobial susceptibility phenotypes (18). These quality control strains should also encompass resistant and susceptible ranges of the antimicrobials to be assayed. Laboratories involved in AST need to use identical or similar quality control reference strains. Reference strains should be kept as stock cultures from which working cultures are derived and should be obtained from national or international culture collections (e.g. American type culture collection [ATCC]). If possible, the preferred method for analysing the overall performance of each laboratory is to test the appropriate quality control bacterial strains on each day that susceptibility tests are performed (18). Because this may not always be practical or economic, the frequency of such quality control tests may be reduced if the laboratory can demonstrate that the susceptibility testing procedures are reproducible. If a laboratory can document the reproducibility of the susceptibility testing methods used, testing may be performed on a weekly basis (18). If quality control errors emerge, the laboratory has a responsibility to determine the cause(s) and repeat the tests. If the laboratory cannot determine the source of error(s), then quality control testing should be re-initiated on a daily basis (18). Recognised quality control strains should be tested each time a new batch of medium or plate lot is used and on a regular basis in parallel with the bacterial strains to be assayed. Reference bacterial strains should be stored at designated centralised or regional laboratories. Appropriate biosecurity issues should be addressed in obtaining and dispersing quality control reference strains to participating laboratories. The use of such strains will allow for comparison of antimicrobial susceptibility data among the many surveillance systems in place among OIE Member Countries. OIE Member Country laboratories should ultimately base quality control testing on factors and circumstances specific to their needs and within reason. However, without the appropriate quality control testing, susceptibility data derived from antimicrobial surveillance and monitoring systems will be of limited value. OIE International Standards on Antimicrobial Resistance, 2003 233 5. Laboratory method External proficiency testing (e.g. third party testing) of participating laboratories should be initiated for major bacterial species included in national surveillance systems and should be mandatory. Designated national laboratories should be appointed or established to monitor quality assurance of the participating surveillance laboratories. The responsibilities of the reference laboratory may include the development of a set of reference bacterial strains with varying antimicrobial susceptibilities to be sent to the participating laboratories to ensure the accuracy and precision of the AST methods and results. The participating laboratories will test these strains under their normal AST conditions. Proficiency testing on a regular basis would become one of the foundations of quality assurance for participating laboratories in a surveillance programme and ensure that reported susceptibility data is accurate (21). Future directions in antimicrobial resistance detection The most recent and perhaps the state-of-the-art approach for detection of certain bacterial antimicrobial resistance phenotypes is via identification and characterisation of the known genes that encode specific resistance mechanisms. Methods that employ the use of genetic probes, nucleic acid amplification techniques (e.g. polymerase chain reaction [PCR]), and deoxyribonucleic acid (DNA) sequencing offer the promise of increased sensitivity, specificity and speed in the detection of specific known resistance genes (14, 17). These genotypic methods are important supplements to traditional phenotypic methods, e.g. for the verification of methicillin resistance in staphylococci, vancomycin resistance in enterococci, and detection of fluoroquinolone resistance mutations (14, 17, 23). Additionally, recent technological advances may facilitate the ability to probe bacterial species for large numbers of antimicrobial resistance genes rapidly and at low cost, thereby providing additional relevant data for surveillance and monitoring programmes. Recommendations To standardise AST methods and achieve comparability of antimicrobial susceptibility test results between OIE Member Countries, the following recommendations are presented: – standardised antimicrobial susceptibility testing methods and harmonisation of susceptibility data (including interpretive criteria) are essential for national and international surveillance comparisons in OIE Member Countries – standardised AST methods and similar interpretive criteria must be accepted and used by all participating laboratories in surveillance and monitoring programmes – it is essential that all data, regardless of the AST method, be reproducible and reported quantitatively if comparisons are to be drawn on a world-wide scale between surveillance programmes – establishment of national or regional designated laboratories is essential for coordination of AST methodologies, interpretations and quality controls – microbiological laboratories must conduct their work under internal quality assurance 234 OIE International Standards on Antimicrobial Resistance, 2003 5. Laboratory method – it is desirable for laboratories to become accredited, where applicable, and to participate in external proficiency testing programmes – specific bacterial reference/quality control strains, with varying susceptibility ranges (susceptible, intermediate and resistant), are essential for determining intra- and inter-laboratory quality assurance and proficiency testing – interpretive criteria should be determined, developed and internationally agreed upon for commonly encountered bacteria, especially zoonotic pathogens such as Salmonella and Campylobacter – co-ordination, where appropriate, with other international organisations (Food and Agriculture Organization, World Health Organization) and/or regional organisations (e.g. European Committee on Antimicrobial Susceptibility Testing, NCCLS) may be important in providing support for standardisation and harmonisation of AST methodologies and data among OIE Member Countries. Antibiorésistance : standardisation et harmonisation des méthodes de laboratoire pour la détection et la quantification de l’antibiorésistance D.G. White, J. Acar, F. Anthony, A. Franklin, R. Gupta, †T. Nicholls, Y. Tamura, S. Thompson, E.J. Threlfall, D. Vose, M. van Vuuren, H.C. Wegener & M.L. Costarrica Résumé Le Groupe ad hoc d’experts sur l’antibiorésistance créé par l’Organisation mondiale pour la santé animale a élaboré une ligne directrice sur la standardisation et l’harmonisation des méthodologies de laboratoire appliquées à la détection et à la quantification de l’antibiorésistance. Les auteurs analysent les méthodes existantes (diffusion sur disque [avec les bandes de gradients de concentration], dilution en gélose et dilution en bouillon de culture) et comparent leurs avantages et inconvénients respectifs. Ils définissent les caractéristiques correspondant au classement des bactéries du point de vue de leur résistance (sensibles, intermédiaires et résistantes) et discutent les critères relatifs à la détermination des valeurs critiques. Tous ces aspects doivent être pris en compte lors de l’interprétation et de la comparaison des données de suivi ou de surveillance de la résistance. Il est recommandé de recourir à des méthodes de laboratoire validées, d’établir des programmes d’assurance qualité (interne et externe) pour les travaux microbiologiques en laboratoire et de communiquer les résultats des épreuves quantitatives. Il est également conseillé de faire établir l’équivalence entre les différentes méthodes et résultats de tests de laboratoire par des évaluations externes des performances qui devraient être conduites par un réseau de laboratoires de référence. Cette approche permettrait de comparer les résultats des tests obtenus par différentes méthodes dans des laboratoires de divers pays. Mots-clés Analyse du risque – Antibiorésistance – Harmonisation – Maîtrise de la résistance – Méthodologie de laboratoire – Normes internationales – Organisation mondiale pour la santé animale – Santé publique – Standardisation – Valeur critique. OIE International Standards on Antimicrobial Resistance, 2003 235 5. Laboratory method Resistencia a los antimicrobianos: normalización y armonización de los métodos de laboratorio para detectar y cuantificar la resistencia a los antimicrobianos D.G. White, J. Acar, F. Anthony, A. Franklin, R. Gupta, †T. Nicholls, Y. Tamura, S. Thompson, E.J. Threlfall, D. Vose, M. van Vuuren, H.C. Wegener & M.L. Costarrica Resumen El Grupo Ad hoc de expertos sobre la resistencia de las bacterias a los productos antimicrobianos, creado por la Organización mundial de sanidad animal, ha elaborado una directriz sobre la normalización y armonización de los métodos de laboratorio para detectar y cuantificar la resistencia a los productos antimicrobianos. Los autores pasan revista a las técnicas existentes (difusión en disco [incluidas las tiras de gradiente de concentración], dilución en agar y dilución en caldo) y comparan sus respectivas ventajas e inconvenientes. También definen las categorías de bacterias en función de su resistencia (susceptibles, intermedias y resistentes) y examinan los criterios para determinar los valores críticos, aspectos que conviene tener en cuenta a la hora de interpretar y comparar datos procedentes del seguimiento o la vigilancia de las resistencias. Los autores recomiendan utilizar métodos de laboratorio validados y someter a procesos (internos y externos) de garantía de calidad tanto el trabajo microbiológico como los informes sobre resultados de pruebas cuantitativas. Recomiendan asimismo que se establezcan equivalencias entre distintos métodos y resultados de laboratorio mediante pruebas externas de eficiencia, proceso en el que ha de intervenir un sistema de laboratorios de referencia. Esta fórmula serviría para comparar los resultados obtenidos mediante métodos diversos y por laboratorios de países distintos. Palabras clave Análisis de riesgos – Armonización – Contención de las resistencias – Métodos de laboratorio – Normalización – Normas internacionales – Organización mundial de sanidad animal – Resistencia a los productos antimicrobianos – Salud pública – Valores críticos. References 1. Acar J. & Goldstein F.W. (1995). – Disk susceptibility test (V. Lorian, ed.). In Antibiotics in laboratory medicine, 4th Ed. Williams and Wilkins, Baltimore, 1-51. 2. Andrews J.M., Boswell F.J. & Wise R. (2000). – Evaluation of the Oxoid Aura image system for measuring zones of inhibition with the disk diffusion technique. J. antimicrob. Chemother., 46, 535-540. 3. Bager F. (2000). – DANMAP: monitoring antimicrobial resistance in Denmark. Int. J. antimicrob. Agents, 14, 271-274. 4. Brown D.F. & Brown L. (1991). – Evaluation of the E-test, a novel method of quantifying antimicrobial activity. J. antimicrob. Chemother., 27, 185-190. 236 OIE International Standards on Antimicrobial Resistance, 2003 5. Laboratory method 5. Caprioli A., Busani L., Martel J.L. & Helmuth R. (2000). – Monitoring of antibiotic resistance in bacteria of animal origin: epidemiological and microbiological methodologies. Int. J. antimicrob. Agents, 14, 295-301. 6. Cars O. (ed.) (1997). – Antimicrobial susceptibility testing in Sweden. Scand. J. infect. Dis., 105 (Suppl.), 5-31. 7. Comité de l’Antibiogramme de la Société française de Microbiologie (CA-SFM) (1993). – Définition des catégories thérapeutiques et méthode de détermination de la concentration minimale inhibitrice en milieu solide pour les bactéries aérobies à croissance rapide. Bull. Soc. fr. Microbiol., 8, 156-166. 8. Courvalin P. & Soussy C.J. (eds) (1996). – Report of the Comité de l’Antibiogramme de la Société française de Microbiologie. Clin. Microbiol. Infect., 2 (Suppl. 1), S3-S34. 9. Craig W. (1993). – Qualitative susceptibility tests versus quantitative MIC tests. Diagn. Microbiol. infect. Dis., 16, 231-236. 10. Deutsch Industrie Norm-Medizinsche Mikrobiologie (1994). – Methoden zur Empfindlichkeitsprunfung von bakteriellen Krankheitserregen (ausser Mykobakterien) gegen Chemotherapeutika, Geschaftsstelle des NAMeds im DIN. Dtsch Ind. norm-med. Mikrobiol., 58, 940. 11. Gould I.M. (2000). – Towards a common susceptibility testing method? J. antimicrob. Chemother., 45, 757-762. 12. Greenwood D. (2000). – Detection of antibiotic resistance in vitro. Int. J. antimicrob. Agents, 14, 303-306. 13. Hubert S., Nguyen P.D. & Walker R.D. (1998). – Evaluation of a computerized antimicrobial susceptibility system with bacteria isolated from animals. J. vet. diagn. Invest., 10, 164-168. 14. Jorgensen J.H. & Ferraro M.J. (2000). – Antimicrobial susceptibility testing: special needs for fastidious organisms and difficult-to-detect resistance mechanisms. Clin. infect. Dis., 30, 799-808. 15. Livermore D.M. & Wale M.C.J. (1998). – Surveillance of antimicrobial resistance. Br. med. J., 317, 614-615. 16. Marano N.N., Rossiter S., Stamey K., Joyce K., Barrett T.J., Tollefson L.K. & Angulo F.J. (2000). – The national antimicrobial resistance monitoring system (NARMS) for enteric bacteria, 1996-1999: surveillance for action. J. Am. vet. med. Assoc., 217, 1829-1830. 17. Murray P.R., Baron E.J., Pfaller M.A., Tenover F.C. & Yolken R.H. (1999). – Antimicrobial agents and susceptibility testing. In Manual of clinical microbiology, 7th Ed. American Society for Microbiology, Washington, DC, 1469-1592. 18. National Committee for Clinical Laboratory Standards (NCCLS) (1999). – Document M31-A. Performance standards for antimicrobial disk and dilution susceptibility tests for bacteria isolated from animals, approved standard. NCCLS, Villanova, 57 pp. 19. National Committee for Clinical Laboratory Standards (NCCLS) (1999). – Document M37-A. Development of in vitro susceptibility testing criteria and quality control parameters for veterinary antimicrobial agents, approved guideline. NCCLS, Villanova, 17 pp. OIE International Standards on Antimicrobial Resistance, 2003 237 5. Laboratory method 20. National Committee for Clinical Laboratory Standards (NCCLS) (2001). – Document M100-S11. Performance standards for antimicrobial susceptibility testing, 9th informational supplement. NCCLS, Wayne, 122 pp. 21. Threlfall E.J., Fisher I.S.T., Ward L.R., Tschäpe H. & Gerner-Smidt P. (1999). – Harmonization of antibiotic susceptibility testing for Salmonella: results of a study by 18 national reference laboratories within the European Union-funded Enter-Net group. Microb. Drug Resist., 5, 195-200. 22. Trevino S. (2000). – Antibiotic resistance monitoring: a laboratory perspective. Military Med., 165, 40-42. 23. Walker R.A., Lawson A.J., Lindsay E.A., Ward L.R., Wright P.A., Bolton F.J., Wareing D.R.A., Corkish J.D., Davies R.H. & Threlfall E.J. (2000). – Decreased susceptibility to ciprofloxacin in outbreak-associated multiresistant Salmonella Typhimurium DT104. Vet. Rec., 147, 395-396. 24. Walker R.D. (2000). – Antimicrobial susceptibility testing and interpretation of results. In Antimicrobial therapy in veterinary medicine, 3rd Ed. Iowa State University Press, Ames, 12-26. 25. Woods G.L. (1995). – In vitro testing of antimicrobial agents. In Antibacterial therapy: in vitro testing, pharmacodynamics, pharmacology, new agents. Infectious Disease Clinics of North America, Vol. 9. W.B. Saunders, Philadelphia, 463-495. 26. Working Party on Antibiotic Sensitivity Testing of the British Society for Antimicrobial Chemotherapy (1991). – A guide to sensitivity testing. J. antimicrob. Chemother., 27 (Suppl. D), 1-50. __________ 238 OIE International Standards on Antimicrobial Resistance, 2003 5. Laboratory method Standardisation of antimicrobial susceptibility testing in Europe: the work of the European Committee for Antimicrobial Susceptibility Testing (EUCAST) I. Phillips Department of Infection, KCL, st Thomas’Hospital Campus, London SE1 7EH, United Kingdom The European Committee for Antimicrobial Susceptibility Testing (EUCAST) is a standing committee of the European Society for Clinical Microbiology and Infectious Diseases (ESCMID), and was formed in 1997 as the successor of an earlier working group. Under the Chairmanship of Professor I Phillips until 2001, when Dr G. Kahlmeter succeeded him, its main aim has been the rationalisation of antimicrobial susceptibility testing in Europe. Rationalisation has become necessary because there has developed a great diversity of methodology in different countries and in different laboratories within countries since the work of the International Collaborative Study in the 1960s. Different media, different inocula and different breakpoints are used, including National Committee for Clinical Laboratory Standards (NCCLS) methodology and interpretive criteria originating in the United States of America. This has meant that it has been difficult if not impossible to compare results of tests, and particularly reports, on the prevalence of resistance both within the European Union and with the rest of the world. I shall describe the work of the Committee up to the change of Chairmanship. Until 2001 the Committee was made up of representatives appointed by each European country – thirty-four in all – plus two representatives for the Pharmaceutical Industry and two for the Manufacturers of Susceptibility-testing Devices. All of these representatives were to be appointed for two years and were charged with acting as intermediaries between the Committee and those who appointed them. In the event, it took longer to set up the Committee than had been visualised, and most members served for longer periods until there were changes to the constitution on the appointment of the new Chairman. The Committee attempted to make progress towards rationalisation by the agreement of reference methodology, but not, as has sometimes been understood, by the agreement of a single standard methodology, since it was quite clear that those countries that had developed their own methodology – for example, France, Sweden and the United Kingdom – as well as those who used NCCLS methods, would be loth to change. If correlations with the results of a reference methodology could be achieved, there was actually no reason why they should change. This is a point that has not always been understood by those not directly involved in the practice of susceptibility testing, who appeared to believe that comparability could be achieved only by standardisation and the use of a single method by all. The initial aim was to achieve comparability of quantitative data minimum inhibitory concentrations (MICs) OIE International Standards on Antimicrobial Resistance, 2003 239 5. Laboratory method and subsequently of interpretations (susceptible, intermediate and resistant). In all this we were attempting to produce results that were of value to the prescribing clinician, taking into account not only antimicrobial susceptibility but also antibiotic pharmacology and the results of therapy. The Committee worked via sub-committees appointed for specific purposes. These sub-committees produced discussion documents (the E.Dis series), and, after consultation, definitive documents (the E.Def series) published in the Society’s official journal, Clinical Microbiology and Infection (CMI). The first subcommittee to produce a definitive report was the Terminology group, which had actually started work under the earlier working party. By 2000, it had produced a second edition of its document, E.Def 1.2 (2). It was hoped that this document would aid international understanding, but it was not intended to endorse any particular methodology. Two sub-committees worked on quantitative susceptibility testing. The first produced its definitive report on agar-dilution MIC determination, E.Def 3.1, following consultation, in 2000 (3). The reference method agreed is fully compatible with the NCCLS reference methodology. The second group, dealing with broth microdilution, took longer but its report is expected to be pubished this year in CMI. Again, it is compatible with the NCCLS reference method. In order to aid the progress of the setting of breakpoints for specific agents, a further sub-committee produced a document in 2000 on the criteria to be considered by those undertaking the process, E.Def 2.1 (4). As an experiment, an effort was made to follow the layout of a similar document produced by the NCCLS, but this fell foul of copyright and intellectual property considerations – despite the fact that the actual practice of breakpoint setting had developed along similar but independent lines in Europe and the USA since the logic was common. The methodology was used to set breakpoints for the new antibiotic linezolid, reported in E.Def 4 (5). In an effort to define the problem of agreeing to standard breakpoints for existing antibiotics, a further document was produced listing current breakpoints used in different countries in Europe (1). It has to be said that the differences are not so great, at least in relation to the identification of susceptible populations of bacteria, as to be a real threat to consensus, although amour propre could be! Because of the particular diversity of disc susceptibility testing methods, it was decided to defer consideration until MIC reference methodology had been agreed. By then it was time for the new chairman to take action, and it was not by chance that his main area of expertise is the achievement of comparable results from diverse disc methodology. The results of his initiatives are awaited. Still within the ambit of main-line susceptibility testing, working groups on automation, on molecular methodology and on quality assurance, were unable to achieve consensus documents for consultation within the timeframe. 240 OIE International Standards on Antimicrobial Resistance, 2003 5. Laboratory method Outside main-line susceptibility testing, discussion documents were produced on Intracellular Pathogens, E.Dis 6.1 (6), and mycobacteria, E.Dis 8.1 (7), and another was planned on antifungals. The EUCAST tried at all times to co-ordinate its activities with other bodies involved in the methodology of susceptibility testing. Perhaps least successful was coordination with national bodies who were naturally wary of its activities but at the same time critical of the time taken to attain consensus. The new Chairman has taken the solution of this problem to be one of his main priorities. Much more successful was co-ordination with NCCLS, and for a period of four years Professor Phillips acted as a formal advisor to its Antimicrobial Susceptibility Testing (AST) Subcommittee, attending its twice-yearly meetings in the USA. The NCCLS AST sent an observer to annual EUCAST meetings. The Committee also co-ordinated with the European Standards Organisation in the hope that when the time was ripe the Reference Methods might help in the production of legal standards. Finally, the Committee attempted to work with the European Medicines Evaluation Agency: although they gave some support to our work, the power of the national groups that make up the Agency’s committees was a tempering influence. It is still hoped that they will accept the results of surveillance of antibiotic resistance based on demonstrated agreement with our reference methods. In its work during the four years leading up to the change of chairmanship in 2001, the EUCAST took a number of important steps towards the achievement of comparability of the results of quantitative antibiotic susceptibility testing and their interpretation, outlined in this paper. On the table are reference methods for agardilution and broth-microdilution testing, and an agreed method of setting breakpoints. It is for official bodies in Europe to accept them and encourage their use. All of the publications may be obtained from Ms Cornelia Hasselmann, Martin-BuberWeg 17, D-81245 Munich, by e-mail via [email protected] or on the net via http://www.escmid.org. References 1. Degener J. E. & Phillips I. (2001). – Comparison of antimicrobial susceptibility test breakpoints of national societies. Clin. microbiol. Infect., 7, 51-54. 2. European Committee for Antimicrobial Susceptibility Testing (EUCAST) (2000). – Terminology relating to methods for the determination of susceptibility of bacteria to antimicrobial agents: E.Def 1.2. Clin. microbiol. Infect., 6, 503-508. 3. European Committee for Antimicrobial Susceptibility Testing (EUCAST) (2000). – Determination of minimum inhibitory concentrations of antibacterial agents by agar dilution: E.Def 3.1. Clin. microbiol. Infect., 6, 509-515. 4. European Committee for Antimicrobial Susceptibility Testing (EUCAST) (2000). – Determination of antimicrobial susceptibility test breakpoints: E.Def 2.1. Clin. microbiol. Infect., 6, 570-572. OIE International Standards on Antimicrobial Resistance, 2003 241 5. Laboratory method 5. European Committee for Antimicrobial Susceptibility Testing (EUCAST) (2001). – Linezolid breakpoints: E.Def 4. Clin. microbiol. Infect., 7, 1-3. 6. European Committee for Antimicrobial Susceptibility Testing (EUCAST) (2001). – Antimicrobial susceptibility testing of intracellular and cell-associated pathogens: E.Dis 6.1. Clin. microbiol. Infect., 12. 7. European Committee for Antimicrobial Susceptibility Testing (EUCAST) (2000). – Antimicrobial susceptibility testing of Mycobacterium tuberculosis: E.Dis 8.1. __________ 242 OIE International Standards on Antimicrobial Resistance, 2003 5. Laboratory method National Committee for Clinical Laboratory Standards: a perspective on antimicrobial susceptibility testing methods T.R. Shryock, Ph.D Elanco Animal Health, 2001 W. Main St., GL21, Greenfield, IN, USA 46140 The vision, mission, and organisational goals of the National Committee for Clinical Laboratory Standards (NCCLS) are fundamentally grounded in the ideals of service and leadership. The NCCLS achieves this through member support, volunteer commitment, organisational partnerships, and collaborative efforts. Global Consensus Standardization for Health Technologies, the NCCLS tagline, creates the context for, and a declaration of, NCCLS’s commitment to providing a global forum for the development of harmonised standards and guidelines that facilitate safety, best practices, and quality patient care in the world’s medical testing and healthcare services community. This work is conducted by volunteers in academia/professions, industry, and government. NCCLS provides clinical laboratory Standards on clinical chemistry, hematology, molecular methods, parasitology, lab safety, healthcare services, immunology, automation, virology, and microbiology. The documents produced by NCCLS cover: – bacteria, mycobacteria, fungi, viruses, etc. – disk and dilution testing (media, inoculum, incubation conditions, antibiotic selection, endpoint determination and interpretation, reporting) – quality control (QC) (protocols, flowcharts) – sponsor development of QC (multilaboratory studies) and interpretive criteria (based on pharmacokinetics-efficacy-epidemiology). The main objective of both the subcommittee on Antimicrobial Susceptibility Testing and the subcommittee on Veterinary Antimicrobial Susceptibility Testing is to provide information that enables laboratories to assist the clinician in the selection of appropriate antimicrobial therapy for patient care. Their mission is to foster appropriate antimicrobial susceptibility testing methodology, sanction quality control data, establish interpretive criteria, provide suggestions to users for routine laboratory testing and reporting, and to educate users. NCCLS Standards may be used by outside organisations for demonstrating conformance with accrediting or proficiency testing requirements or as purchase specifications. The NCCLS provides standardised methodology that is considered a reference method for both animal and human bacterial pathogens. The clinical application of these methods guides the practitioner in the selection of the appropriate antimicrobial agent to treat the patient (animal or human), consistent with the implementation of judicious use guidelines. OIE International Standards on Antimicrobial Resistance, 2003 243 5. Laboratory method If the results of susceptibility tests done by non-NCCLS methods are to be compared with results from tests done by NCCLS methods, then a study for equivalency should be done. Differences in media formulation, inoculum density, etc., could influence the outcome of susceptibility testing. The application of NCCLS interpretive criteria to results generated by non-NCCLS methods may lead to improper antibiotic selections. The similarity of NCCLS testing methods for bacteria isolated from both animals and humans (e.g. foodborne bacteria) allows a direct comparison of results. This is useful for the design and/or application of monitoring programme data when conducting risk assessments on the potential impact of antibiotic use in food animals on public health. In conclusion, the NCCLS provides Standards for those interested in applying antimicrobial susceptibility testing methodology, validated through quality control; with interpretive criteria for the results. With the global application of NCCLS methods, comparisons of results of antimicrobial resistance monitoring programmes are possible. References Specific NCCLS documents for each topic area can be purchased on www.nccls.org __________ 244 OIE International Standards on Antimicrobial Resistance, 2003 5. Laboratory method Harmonisation of antimicrobial resistance testing results – the outcome of the international Enter-net study I.S.T. Fisher (1), O.N. Gill, W.J. Reilly, H.R. Smith & E.J. Threlfall on behalf of the Enter-net participants (1) Enter-net Scientific Co-ordinator, PHLS Communicable Disease Surveillance Centre, 61 Colindale Avenue, London, NW9 5EQ, United Kingdom There has been, and still is, much discussion about the requirement for standardising antimicrobial susceptibility testing results. Which standards should be adopted and which method should be used? The importance of doing this is to allow the meaningful comparison of resistance patterns between laboratories and even countries. The problem with standardisation of methods is that every laboratory wishing to be involved would have to conform to a single method. Enter-net – the international surveillance network for enteric pathogens – took the radical step of looking at whether it would be possible to harmonise the results of antimicrobial testing rather than the methods behind them. Forty-eight salmonella strains, with resistance patterns ranging from fully sensitive to being resistant to twelve different antimicrobial agents, were sent to eighteen national salmonella reference laboratories in Western Europe for antimicrobial tests to be performed. These tests were performed under each laboratory’s own standards and methods. The qualitative results (resistant, intermediate or sensitive) were returned to the Laboratory of Enteric Pathogens for comparison and analysis. The results showed that, with the exception of low-level resistance to ciprofloxacin, there is a very high level of agreement between laboratories. Therefore, data on antimicrobial resistance results were incorporated in the international database that had already been created. This has provided a definitive method for the meaningful international surveillance of antimicrobial susceptibility results. Already in the EU this has contributed to an important assessment of the role of antimicrobials in humans and food animals in development of drug resistance in zoonotic Salmonella species. Similarly, it has resulted in the exchange of harmonised antimicrobial resistance data in addition to phage typing results in the investigation of international outbreaks of multi-resistant S. typhimurium DTs 104 and 204b in several European countries. __________ OIE International Standards on Antimicrobial Resistance, 2003 245 6. Prudent use and containment of resistance OIE International Standards on Antimicrobial Resistance, 2003 247 6. Prudent use and containment of resistance Antimicrobial resistance: responsible and prudent use of antimicrobial agents in veterinary medicine F. Anthony (1), J. Acar (2), A. Franklin (3), R. Gupta (4), †T. Nicholls (5), Y. Tamura (6), S. Thompson (7), E.J. Threlfall (8), D. Vose (9), M. van Vuuren (10) & D.G. White (11) (1) Fresh Acre Veterinary Surgery, Flaggoners Green, Bromyard, Herefordshire HR7 4QR, United Kingdom (2) Université Pierre et Marie Curie, Service de Microbiologie Médicale, Fondation Hôpital Saint-Joseph, 185 rue Raymond Losserand, 75674 Paris Cedex 14, France (3) The National Veterinary Institute (SVA), Department of Antibiotics, SE 751 89 Uppsala, Sweden (4) College of Veterinary Sciences, Veterinary Bacteriology, Department of Microbiology, G.B. Pant University of Agriculture and Technology, Pantnagar 263 145 Uttar Pradesh, India (5) National Offices of Animal and Plant Health and Food Safety, Animal Health Science and Emergency Management Branch, Department of Agriculture, Fisheries and Forestry, P.O. Box 858, Canberra ACT 2601, Australia (6) National Veterinary Assay Laboratory, Ministry of Agriculture, Forestry and Fisheries, 1-51-1 Tolura, Kokubunji, Tokyo 185-8511, Japan (7) Joint Institute for Food Safety Research, Department for Health and Human Services Liaison, 1400 Independence Avenue, SW, Mail Stop 2256, Washington, DC 20250-2256, United States of America (8) Public Health Laboratory Service, Central Public Health Laboratory, Laboratory of Enteric Pathogens, 61 Collindale Avenue, London NW9 5HT, United Kingdom (9) David Vose Consulting, Le Bourg, 24400 Les Lèches, France (10) University of Pretoria, Faculty of Veterinary Science, Department of Veterinary Tropical Diseases, Private Bag X04, Onderstepoort 0110, South Africa (11) Centre for Veterinary Medicine, Food and Drug Administration, Office of Research, HFV-530, 8401 Muirkirk Road, Laurel, Maryland 20708, United States of America This report, prepared by the OIE Ad hoc Group of experts on antimicrobial resistance, has not yet received the approval of the International Committee of the OIE Summary A guideline on the responsible and prudent use of antimicrobials in animal husbandry has been developed by the Ad hoc Group of experts on antimicrobial resistance, created by the OIE (World organisation for animal health). The objectives of responsible use are to maintain antibiotic efficacy, to avoid the dissemination of resistant bacteria or resistance determinants and to avoid the exposure of humans to resistance through food. The guideline attributes a central role to the competent authorities responsible for granting marketing authorisations for antimicrobial substances. Requirements before and after granting of marketing authorisations are defined. Important aspects include the control of the pharmaceutical product quality and the therapeutic efficacy, the assessment of the selection pressure, the protection of the environment, specific and non-specific antimicrobial resistance surveillance. The guideline is also addressed to the veterinary pharmaceutical industry, veterinary practitioners, dispensing pharmacists and farmers. The respective roles and responsibilities of these groups are defined. OIE International Standards on Antimicrobial Resistance, 2003 249 6. Prudent use and containment of resistance Keywords Antimicrobial resistance – Competent authorities – Containment of resistance – Food – Human medicine – International standards – Marketing authorisation – Public health – Veterinary medicine – World Organisation for Animal Health. Introduction This document provides guidance for the responsible and prudent use of antimicrobials in veterinary medicine, with the aim of protecting both animal and human health. The authors define the respective responsibilities of authorities and groups involved in the registration, production, control, distribution and use of veterinary antimicrobials, such as national competent authorities, the veterinary pharmaceutical industry, veterinarians, pharmacists and livestock producers. Prudent use is principally determined by the outcome of the marketing authorisation procedure and by the implementation of specifications when antimicrobials are administered to animals. A number of codes of practice, relating to the use of antimicrobials and the conditions thereof have been developed by different organisations. These codes were taken into consideration and some elements were included in the preparation of this guideline. Aims and objectives It is imperative that all who are involved in the authorisation, manufacture, sale and supply, prescription and use of antimicrobials in livestock act legally, responsibly and with the utmost care, in order to limit the spread of resistant bacteria among animals and to protect the health of consumers. Antimicrobial agents: powerful tools for treating and preventing/controlling bacterial diseases in animals Guidelines for the responsible use of antimicrobial agents in veterinary medicine include a set of practical measures and recommendations intended to prevent and/or reduce the selection of antimicrobial resistant bacteria in animals, with the following aims: a) to maintain the efficacy of antimicrobial agents and to ensure the rational use of antimicrobials in animals with the purpose of optimising both their efficacy and safety in animals b) to comply with the ethical obligation and economic need to keep animals in good health c) to prevent, or reduce as far as possible, the transfer of bacteria (with their resistance determinants) within animal populations, to maintain the efficacy of antimicrobial agents used in livestock 250 OIE International Standards on Antimicrobial Resistance, 2003 6. Prudent use and containment of resistance d) to prevent or reduce the transfer of resistant bacteria or resistance determinants from animals to humans, to maintain the efficacy of antimicrobial agents used in human medicine e) to prevent the contamination of animal-derived food with antimicrobial residues which exceed the established maximum residue limit (MRL) f) to protect consumer health by ensuring the safety of food of animal origin intended for human consumption. The responsible use of antimicrobials in veterinary medicine The Ad hoc Group described responsible use as follows: a) represents the scientific and technically directed use of these compounds that are the responsibility of professionals with the required expertise b) is part of good veterinary and animal husbandry practice and takes into consideration disease prevention practices such as the use of vaccination and improvements in husbandry conditions when disease problems become evident c) aims to reduce the use of antimicrobial agents to their approved and intended uses d) takes into consideration on-farm sampling and testing of isolates from foodproducing animals during their production (where appropriate), and makes adjustments to therapy when problems become evident e) should be based on the results of resistance surveillance and monitoring (bacterial cultures and antimicrobial sensitivity testing) f) is aimed at all the relevant professionals, including the following: – administrative and scientific authorities – the veterinary pharmaceutical industry – distributors and others handling antimicrobials – veterinarians, pharmacists and livestock producers. Responsibilities of the regulatory authorities The national regulatory authorities, which are responsible for granting the marketing authorisation, have a significant role in specifying the terms of this authorisation and in providing the appropriate information to the veterinarian through product labelling in support of the prudent use of antimicrobials in veterinary medicine. It is the responsibility of the pharmaceutical industry to submit the data requested for the granting of the marketing authorisation. The use of an antimicrobial agent in veterinary medicine requires a marketing authorisation, which is granted by the competent authorities only if the criteria of safety, quality and efficacy are met. The examination of applications for drug authorisation must include an assessment of the risks to both the animal and the consumer resulting from the use of antimicrobial agents in food-producing animals. OIE International Standards on Antimicrobial Resistance, 2003 251 6. Prudent use and containment of resistance The evaluation should focus on each individual antimicrobial product and not be generalised to the class of antimicrobials to which the particular active principle belongs. The safety evaluation should include consideration of the potential impact on human health of the proposed use in food-producing animals. If dose ranges or different durations of treatment are suggested, the national authorities should give guidance on the approved product labelling regarding the conditions that will minimise the development of resistance. Regulatory authorities should, where possible, expedite the market approval process of new antimicrobial molecule formulation, which is considered to have the potential to help the control of resistance. The preparation of internationally accepted guidelines would assist in this regard. Countries lacking the necessary resources to implement an efficient registration procedure for veterinary medicinal products and whose supply of veterinary medicinal products principally depends on imports from foreign countries must undertake the following measures: – check the efficacy of administrative controls on the import of these veterinary medicinal products – check the validity of the registration procedures of the exporting country – develop the necessary technical co-operation with experienced authorities to check the quality of imported veterinary medicinal products as well as the validity of the recommended conditions of use. Regulatory authorities of importing countries could request the pharmaceutical industry to provide quality certificates prepared by the competent authority of the exporting country. All countries should make every effort to actively combat the trade, distribution and use of illegal and counterfeit products. Quality control of antimicrobial agents Quality controls should be performed as follows: – in compliance with the provisions of good manufacturing practices – to ensure that analysis specifications of antimicrobial agents used as active ingredients comply with the provisions of approved monographs – to ensure that the quality and concentration (stability) of antimicrobial agents in the marketed dosage form(s) is maintained until the expiry date, established under the recommended storage conditions – to ensure the stability of antimicrobials when mixed with feed or drinking water – to ensure that all antimicrobials are manufactured to the appropriate quality and purity in order to guarantee safety and efficacy. 252 OIE International Standards on Antimicrobial Resistance, 2003 6. Prudent use and containment of resistance Control of the therapeutic efficacy Preclinical trials Preclinical trials should be undertaken, with the following aims: – to assess the ability of the antimicrobial agent to select for resistant bacteria in vitro and in vivo. The design of in vivo studies is currently under development. In certain cases, preclinical trials should evaluate not only the bacteria of the target animals for resistance, but also the impact of the antimicrobial use on food-borne and/or commensal bacteria – to establish an appropriate dosage regimen necessary to ensure the therapeutic efficacy of the antimicrobial agent and limit the selection of antimicrobial resistant bacteria. Pharmacodynamics and the establishment of the activity of antimicrobial agents towards the targeted bacteria The following criteria should be taken into account: – mode of action – minimum inhibitory and bactericidal concentrations – time- or concentration-dependent activity – activity at the site of infection. Pharmacokinetics and the establishment of the dosage regimens allowing maintenance of effective antimicrobial levels The following criteria should be taken into account: – bio-availability according to the route of administration – concentration of the antimicrobial at the site of infection and its distribution in the treated animal – metabolism which may lead to the inactivation of antimicrobials – excretion routes. The use of combinations of antimicrobial agents should be justified, taking into account the following: – pharmacodynamics (additive or synergistic effects towards the target bacteria) – pharmacokinetics (maintenance of the levels of associated antibiotics responsible for additive or synergistic effects at the site of infection throughout the treatment period). Clinical trials Clinical trials should be performed to confirm the validity of the claimed therapeutic indications and dosage regimens established during the preclinical phase. The following criteria should be taken into account: OIE International Standards on Antimicrobial Resistance, 2003 253 6. Prudent use and containment of resistance – diversity of the clinical cases encountered when performing multi-centre trials – compliance of the protocols of clinical trials with good clinical practice – eligibility of the studied clinical cases, based on appropriate criteria of clinical and bacteriological diagnoses – parameters for qualitatively and quantitatively assessing the efficacy of the treatment. Assessment of the potential of antimicrobials to select for resistant bacteria Studies may be appropriate and requested in support of the assessment of the potential of antimicrobials to select for resistant bacteria. However, it should be noted that the results from these in vivo studies may be very different from the resistance that develops under normal conditions. Therefore, the interpretation should be undertaken with great caution. The party applying for market authorisation for antimicrobials for veterinary use should, where possible, supply data derived from the testing of antimicrobials for the development of antimicrobial resistance in target animal species under the intended conditions of use. To reduce the potential selection of resistance, preclinical and clinical trials should, in certain cases, evaluate not only pathogenic bacteria of target animals for resistance, but also the impact of the antimicrobial use on food-borne and/or commensal (indicator) bacteria. In these cases, considerations may include the following: – the concentration of active compound in the gut of the animal (where the majority of potential food-borne pathogens reside) at the defined dosage level – the level of human exposure to food-borne or other resistant bacteria – the degree of cross-resistance within the class of antimicrobials and between classes of antimicrobials – the pre-existing level of resistance in the pathogens of human health concern (baseline determination). Establishment of acceptable daily intake, maximum residue limit and withdrawal periods for antimicrobial compounds a) When setting the acceptable daily intake (ADI) and MRL for an antimicrobial substance, the safety evaluation should, for this class of substances, also include the potential biological effects on the intestinal flora of humans. Using in vitro and/or in vivo tests and/or data originating from human medicine, an assessment should be undertaken regarding the capability of antimicrobial residues, ingested by the consumer, to disturb the intestinal flora of humans by selecting resistant bacteria and/or weakening the barrier effect against the colonisation of pathogenic bacteria. 254 OIE International Standards on Antimicrobial Resistance, 2003 6. Prudent use and containment of resistance b) The establishment of an ADI for each antimicrobial agent, and an MRL for each animal-derived food, should be undertaken. An MRL is necessary in order that officially approved control laboratories can verify that all foods comply with the safety standards. c) For each veterinary medicinal product containing antimicrobial agents, withdrawal periods should be established which make it possible to produce safe food in compliance with the MRL. Withdrawal periods should be established for each veterinary medicinal product by taking into account the following: – the MRL established for the antimicrobial agent under consideration – the pharmaceutical form – the target animal species – the dosage regimen and the duration of treatment – the route of administration. The applicant should provide methods for regulatory testing of residues in food. Protection of the environment An assessment of the impact of the proposed antimicrobial use on the environment should be conducted. Efforts should be made to ensure that environmental contamination with antimicrobials is restricted to a minimum. Establishment of a summary of product characteristics for each veterinary medicinal product The summary of product characteristics contains the information necessary for the appropriate use of veterinary medicinal products containing antimicrobial agents. It constitutes, for each veterinary medicinal product, the official reference of the content of its labelling and package insert. This summary contains the following items: – pharmacological properties – target animal species – therapeutic indications – target bacteria – dosage and administration route – withdrawal periods – incompatibilities – expiry date – operator safety – particular precautions before use – particular precautions for the proper disposal of un-used products. OIE International Standards on Antimicrobial Resistance, 2003 255 6. Prudent use and containment of resistance The conditions of prudent use of an antimicrobial agent in veterinary medicine should be based on a safety evaluation, which takes into particular consideration the importance of the drug, or other antimicrobial agents belonging to the same therapeutic class, in human and/or veterinary medicine. Antimicrobials which are considered important in treating critical diseases in humans should only be used in animals when alternatives are either unavailable or inappropriate. Consideration should be given to providing such guidance to the veterinarian by means of the product label. The oral route, which enhances the access of antimicrobial agents to the complex intestinal flora, and hence the possibility of the selection and the transfer of resistance genes, should be used with caution. For certain antimicrobial classes, other administration routes may also cause similar selection of resistance. Specific mention should be made on the product label. Post-marketing antimicrobial surveillance A structured approach is required to the investigation and reporting of the incidence and prevalence of resistance. Regulatory authorities should have implemented a pharmacovigilance programme for the monitoring, reporting and recording of adverse reactions to antimicrobials, including the lack of efficacy related to antimicrobial resistance. The information collected through the pharmacovigilance programme should form part of the comprehensive strategy to minimise antimicrobial resistance. A surveillance programme A specific surveillance programme to assess the impact of the use of an authorised antimicrobial agent on the selection of antimicrobial resistant bacteria in foodproducing animals may be implemented after the granting of the marketing authorisation. In certain cases, the surveillance programme should evaluate not only resistance development in target animal pathogens, but also in food-borne pathogens and/or commensals. This protocol of surveillance should be implemented if justified by the safety evaluation performed during the registration process. Specific surveillance The surveillance of animal bacteria resistant to antimicrobial agents is essential. The relevant authorities should implement a programme, established from the results of a risk analysis, which allows the ranking of priorities regarding antimicrobials and animal bacteria, whether or not they are pathogenic for animals and humans. For reasons of efficiency, the methods used to establish such programmes (laboratory techniques, sampling, choice of antimicrobial agents and bacteria, etc.) should be harmonised as much as possible at the international level (see Antimicrobial resistance: standardisation and harmonisation of laboratory methodologies for the detection and quantification of antimicrobial resistance and Antimicrobial resistance: harmonisation of national antimicrobial resistance 256 OIE International Standards on Antimicrobial Resistance, 2003 6. Prudent use and containment of resistance monitoring and surveillance programmes in animals and in animal-derived food, later in this volume). This epidemiological surveillance of antimicrobial resistance should be accompanied by a continuous survey on the amounts of antimicrobial agents used by veterinarians and other authorised users, in order to encourage the most appropriate prescription of these medicinal products. If justified by the results of this post-registration surveillance of antimicrobial resistance, whether specific or not, the conditions of use of the antimicrobial agents in veterinary medicine should be modified. Distribution of the antimicrobial agents used in veterinary medicine The relevant authorities should, where possible, ensure that all the antimicrobial agents used in food animals fulfil the following criteria: – are prescribed by a veterinarian or other suitably trained and authorised person – are delivered by an authorised animal health professional – are supplied only through licensed/authorised distribution systems – are administered to animals by a veterinarian or under the supervision of a veterinarian or by his/her agent. Control of advertising All advertising of antimicrobials should be controlled by a code of advertising standards, and the relevant authorities must ensure that the advertising of antimicrobial products fulfils the following criteria: – compliance with the marketing authorisation granted, in particular regarding the content of the summary of product characteristics – restriction to authorised professionals, according to national legislation in each country. Training of antibiotic users Training of antibiotic users, involving all the relevant professional organisations, including regulatory authorities, the pharmaceutical industry, veterinary schools, research institutes and professional associations, should focus on the following: – information on disease prevention and management strategies to reduce the need to prescribe antimicrobials – the ability of antimicrobials to select for resistant bacteria in food-producing animals, which may cause animal and/or human health problems – the need to observe responsible use recommendations and the use of antimicrobial agents in animal husbandry in agreement with the provisions of the marketing authorisations, and veterinary advice, in order to assure the safety to the consumer of animal-derived food, and therefore the protection of public health OIE International Standards on Antimicrobial Resistance, 2003 257 6. Prudent use and containment of resistance – relevant pharmacokinetic and pharmacodynamic information to enable the veterinarian to use antimicrobials prudently. Development of research The relevant authorities should encourage public and private research with the following aims: – to improve knowledge regarding the mechanisms of action of antimicrobials, to optimise the dosage regimens and the therapeutic activity of these medicinal products – to improve knowledge about the mechanisms of selection, emergence and dissemination of bacterial genes encoding resistance against antimicrobial agents – to develop practical models for applying the concept of risk analysis to assess the public health concern precipitated by the development of resistant bacteria – to further develop protocols to predict, during the registration process, the impact of the proposed use of the antimicrobials on the rate and extent of resistance development – to develop alternative methods to control bacterial diseases (vaccines, changes in husbandry practices, etc.). Responsibilities of the veterinary pharmaceutical industry Marketing authorisation of veterinary medicinal products The veterinary pharmaceutical industry has responsibilities in the following areas: – to supply all the information requested by the national regulatory authority in order to establish objectively the quality, safety and efficacy of veterinary medicinal products – to guarantee the quality of this information on the basis of the implementation of procedures, tests and trials in compliance with the provisions of good manufacturing, laboratory and clinical practices. The pharmaceutical industry should be encouraged to perform post-approval studies, as practised for human medicinal products, in order to seek an extension of the authorised indications in the light of practical experience. This would limit the need for off-label use. Marketing and export of veterinary medicinal products In regard to marketing and export of veterinary medicinal products, the following suggestions are presented: – only officially licensed and approved veterinary medicinal products should be sold and supplied, and then only through licensed/authorised distribution systems – only veterinary medicinal products which have been authorised in the (exporting) country in which the product(s) is approved for sale or the quality of which is certified by a regulatory authority should be exported 258 OIE International Standards on Antimicrobial Resistance, 2003 6. Prudent use and containment of resistance – the national regulatory authority should be provided with the information necessary to evaluate the amount of antimicrobial agents marketed. Advertising The following are the responsibilities of the veterinary pharmaceutical industry: – to disseminate information in compliance with the provisions of the granted authorisation and to ensure that this dissemination reaches only those authorised professionals involved in the prescription and distribution of the products – to ensure that the advertising of antimicrobials directly to the livestock producer is discouraged. Training The veterinary pharmaceutical industry is responsible for participation in training programmes as defined in the earlier section entitled ‘Training of antibiotic users’. Research It is the responsibility of the veterinary pharmaceutical industry to contribute to the research effort as defined in the earlier section entitled ‘Development of research’. Responsibilities of pharmacists Pharmacists distributing veterinary antimicrobials should only do so on the prescription of a veterinarian, and all products should be appropriately labelled (see later section entitled ‘Labelling’). The guidelines on the responsible use of antimicrobials should be reinforced by pharmacists, who should keep detailed records of all antimicrobials supplied, including the following: – date of supply – name of prescribing veterinarian – name of user – name of product – batch number – quantity supplied. Pharmacists should also be involved in training programmes on the responsible use of antimicrobials. Responsibilities of veterinarians The use of antimicrobials is no substitute for good management practices and the prime concern of the veterinarian is to encourage good farming practice in order to minimise the need for antimicrobial use in livestock. OIE International Standards on Antimicrobial Resistance, 2003 259 6. Prudent use and containment of resistance In the frame of good management practice, the veterinarian is responsible for identifying recurrent disease problems and developing alternative strategies to prevent or control disease. These may include changes in husbandry conditions and vaccination programmes where vaccines are available. Veterinarians should only prescribe antimicrobials for animals under their care, which means that: – the veterinarian must have been assigned responsibility for the health of the animal or the herd/flock by the producer or an agent of the producer – that responsibility must be real and not merely nominal – that the animal(s) or herd/flock must have been examined immediately before the prescription and supply or sufficiently recently or frequently for the veterinarian to have personal knowledge of the condition of the animal(s) or current health status of the herd or flock to make a diagnosis and prescribe – the veterinarian should maintain clinical records of the animal(s)/herd/flock. It is recommended that veterinary professional organisations develop for their members, species-specific clinical practice guidelines on the responsible use of antimicrobials, with particular reference to the choice of product, disease prevention strategies and treatment protocols. The responsibilities of veterinarians in this area are described below. Use of antimicrobial agents when necessary The appropriate use of antimicrobials in practice is a critical decision which, where possible, should be based on the following: – the experience and local expertise of the prescribing veterinarian – an accurate diagnosis, based on adequate diagnostic procedures. On certain occasions, a group of animals which may have been exposed to pathogenic bacteria may need to be treated without recourse to an accurate diagnosis and antimicrobial susceptibility testing, to prevent the development of clinical disease and for reasons of animal welfare. Determination of the choice of an antimicrobial The expected efficacy of the treatment The expected efficacy of the treatment is based on the following: – the clinical experience of the veterinarian – the activity towards the pathogenic bacteria involved – the epidemiological history of the rearing unit, particularly in relation to the antimicrobial resistance profiles of the pathogenic bacteria involved. Ideally, the antibiotic profiles should be established before the commencement of treatment. Should a first line antibiotic treatment fail or should the disease recur, the use of a 260 OIE International Standards on Antimicrobial Resistance, 2003 6. Prudent use and containment of resistance second line antimicrobial agent should be based on the results of the microbiological tests – the appropriate route of administration – results of initial treatment – known pharmacokinetics/tissue distribution to ensure that the selected therapeutic agent is active at the site of infection – prognosis. To minimise the likelihood of antimicrobial resistance developing, it is recommended that antimicrobials be targeted to bacteria likely to be the cause of infection. Absence of selection or limited selection of antimicrobial resistant bacteria The absence of selection or limited selection of antimicrobial resistant bacteria is influenced by the following: – the choice of the activity spectrum of the antimicrobial – the targeting of specific bacteria – known or predictable susceptibilities using antimicrobial susceptibility testing – the correct dosing regimens – the use of combinations of antimicrobial agents – the importance of the drug to human and/or veterinary medicine. Antimicrobials which are considered important to treat critical diseases in humans and/or animals, should be used only when other therapies are unavailable or inappropriate – the route of administration. Combinations of antimicrobials Combinations of antimicrobials are used for their synergistic effect to increase therapeutic efficacy or to broaden the spectrum of activity. Furthermore, the use of combinations of antimicrobials can be protective against the selection of resistance in cases in which bacteria exhibit a high mutation rate against a given antimicrobial. However, a bad choice of a combination of antimicrobials may, in certain cases, lead to an increase of the selection of resistance. If the use of a combination of antimicrobials is justified, the veterinarian should ensure that there is no antagonism between the chosen antimicrobials and should check the ability of these antibiotics to reach the infection site under similar time and concentration conditions, to maintain effective therapeutic concentrations as long as required. OIE International Standards on Antimicrobial Resistance, 2003 261 6. Prudent use and containment of resistance Appropriate use of the antimicrobial agent chosen A prescription for antimicrobial agents must precisely indicate the treatment regime, the dose, the dosage intervals, the duration of the treatment, the withdrawal period and the amount of drug to be delivered, depending on the dosage and the number of animals to be treated. All medicinal products should be prescribed and used according to the conditions of the marketing authorisation, which are reflected in the summary of product characteristics provided by the manufacturer. If the label conditions allow for some flexibility, the veterinarian should consider a therapeutic regimen that is sufficiently long to allow the effective recovery of the animal, but sufficiently short to limit the selection of resistance in food-borne and/or commensal bacteria. ‘Off label use’(extra-label use) of veterinary medicinal products Although all medicinal products should be prescribed and used in accordance with the specifications of the marketing authorisation, the prescribing veterinarian should have the discretion to adapt these in exceptional circumstances. The ‘off label use’ of an antimicrobial agent may be permitted in appropriate circumstances and should be in agreement with the national legislation in force. The veterinarian has the responsibility to define the conditions of responsible use in such a case, including the therapeutic regimen, the route of administration and the duration of the treatment. Recording All available information should be consolidated into one form or database, such that this information should: – allow monitoring of the quantities of medication used – contain a list of all medicines supplied to each livestock holding – contain a list of medicine withdrawal periods and a system for allowing information to be updated – contain a record of antimicrobial susceptibilities – provide comments concerning the response of animals to medication – allow the investigation of adverse reactions to antimicrobial treatment, including lack of response due to antimicrobial resistance. Suspected adverse reactions should be reported to the appropriate regulatory authorities. Labelling All medicines supplied by a veterinarian should be adequately labelled with the following minimum information: – the name of the owner/keeper or person who has control of the animal(s) 262 OIE International Standards on Antimicrobial Resistance, 2003 6. Prudent use and containment of resistance – – – – – – the address of the premises where the animal(s) is kept the name and address of the prescribing veterinarian the date of supply the indication ‘For animal treatment only’ the warning ‘Keep out of the reach of children’ the relevant withdrawal period, even if this is nil. The label should not obscure the expiry date of the preparation or any important information supplied by the manufacturer. Training Veterinary professional organisations should participate in the training programmes as defined in the earlier section entitled ‘Training of antibiotic users’. Responsibilities of producers Producers are responsible for preventing outbreaks of disease and implementing health and welfare programmes on their farms. They may, as appropriate, call on the assistance of their veterinarian in undertaking these duties. All those involved with the livestock on the farm have an important role to play in ensuring the responsible use of antimicrobials. Therapeutic antimicrobial products should be regarded as complementing good management, vaccination and farm hygiene. Efforts should be made to ensure that environmental contamination both by antimicrobials and by resistant bacteria is kept to a minimum. Livestock producers have the following responsibilities: a) to draw up a health plan with the veterinarian in charge of the animals that outlines preventative measures (mastitis plan, worming and vaccination programmes, etc.) b) to use antimicrobial agents only on veterinary prescription and according to the provisions of the prescription c) to use antimicrobial agents in the species, for the uses and at the doses on the approved/registered labels and in accordance with product label instructions or the advice of a veterinarian familiar with the animals and the production site d) to isolate sick animals, when appropriate, to avoid the transfer of resistant bacteria e) to comply with the storage conditions of antimicrobials in the rearing unit, according to the provisions of the leaflet and package insert f) to address hygienic conditions regarding contacts between people (veterinarians, breeders, owners, children) and the animals treated OIE International Standards on Antimicrobial Resistance, 2003 263 6. Prudent use and containment of resistance g) to comply with the recommended withdrawal periods to ensure that residue levels in animal-derived food do not present a risk for the consumer h) to dispose of surplus antimicrobials under safe conditions for the environment. Partially-used medicines should only be used within the expiry date, for the condition for which they were prescribed and, if possible, in consultation with the prescribing veterinarian i) to maintain all the laboratory records of bacteriological and susceptibility tests. These data should be made available to the veterinarian responsible for treating the animals to optimise the use of antimicrobials in that unit j) to keep adequate records of all medicines used, including the following: – name of the product/active substance and batch number – name of supplier – date of administration – identification of the animal or group of animals to which the antimicrobial agent was administered – diagnosis/clinical conditions treated – quantity of the antimicrobial agent administered – withdrawal periods – result of laboratory tests – effectiveness of therapy k) to inform the veterinarian responsible for the unit of recurrent disease problems. Conclusion Antimicrobial agents are very important tools for controlling a great number of bacterial diseases in both animals and humans. It is vital that all countries implement the appropriate systems to ensure that antimicrobials are manufactured, marketed, distributed, prescribed, supplied and used responsibly, and that these systems are adequately audited. The OIE Ad hoc Group of experts on antimicrobial resistance is well aware of the difficulties that a number of countries may face in the immediate implementation of all elements of this guideline. This document is designed to provide the framework which countries should implement in accordance with their capabilities and resources, but within a reasonable period of time. A step-by-step approach may be appropriate for a number of countries, to properly implement all of the elements. The continued availability of veterinary medicines, which are essential for animal welfare and health, and consequently for human health, will ultimately depend on the responsible use of these products by all those involved in the authorisation, production, control, distribution and use of antimicrobials in animals. 264 OIE International Standards on Antimicrobial Resistance, 2003 6. Prudent use and containment of resistance Antibiorésistance : utilisation responsable et prudente des antibiotiques en médecine vétérinaire F. Anthony, J. Acar, A. Franklin, R. Gupta, †T. Nicholls, Y. Tamura, S. Thompson, E.J. Threlfall, D. Vose, M. van Vuuren & D.G. White Résumé Le Groupe ad hoc d’experts sur l’antibiorésistance mis en place par l’Organisation mondiale pour la santé animale a mis au point une ligne directrice sur l’utilisation prudente et responsable des antibiotiques dans la production animale. L’utilisation responsable a pour objectif de perpétuer l’activité des antibiotiques, d’éviter la dissémination de bactéries résistantes ou des facteurs favorisant la résistance ainsi que l’exposition de l’homme à celle-ci au travers des aliments. La ligne directrice attribue un rôle majeur aux autorités compétentes chargées de la délivrance des autorisations de mise sur le marché (AMM) des substances antimicrobiennes. Les auteurs définissent les conditions préalables et consécutives à la délivrance de ces AMM. L’accent est mis sur le contrôle de la qualité et de l’efficacité thérapeutique des produits pharmaceutiques, sur l’évaluation de la pression sélective, sur la protection de l’environnement ainsi que sur la surveillance de l’antibiorésistance, spécifique et non spécifique. La ligne directrice s’adresse également à l’industrie des médicaments vétérinaires, aux praticiens, aux pharmaciens et aux éleveurs. Les rôles et responsabilités respectifs de ces groupes sont également définis. Mots-clés Antibiorésistance – Autorisation de mise sur le marché – Autorités compétentes – Denrées alimentaires – Maîtrise de la résistance – Médecine humaine – Médecine vétérinaire – Normes internationales – Organisation mondiale pour la santé animale – Santé publique. Resistencia a los antimicrobianos: uso prudente y responsable de productos antimicrobianos en medicina veterinaria F. Anthony, J. Acar, A. Franklin, R. Gupta, †T. Nicholls, Y. Tamura, S. Thompson, E.J. Threlfall, D. Vose, M. van Vuuren & D.G. White Resumen El Grupo Ad hoc de expertos sobre la resistencia de las bacterias a los productos antimicrobianos, creado por la Organización mundial de sanidad animal, ha elaborado una directriz sobre el uso prudente y responsable de productos antimicrobianos en producción animal. El uso responsable ha de servir para: mantener la eficacia antibiótica de los productos; evitar la diseminación de bacterias resistentes o de determinantes de resistencia; y evitar que el ser humano se vea expuesto por vía alimentaria a organismos resistentes. Esta directriz asigna un papel básico a las autoridades responsables de conceder las licencias de comercialización de sustancias antimicrobianas y define los requisitos que éstas deben cumplir (antes y después de la autorización de comercialización). Entre los aspectos más importantes cabe destacar: el control de la calidad y eficacia terapéutica de los productos farmacéuticos; la evaluación del grado de presión selectiva; la necesidad de proteger el medio ambiente; OIE International Standards on Antimicrobial Resistance, 2003 265 6. Prudent use and containment of resistance y la vigilancia de la aparición de resistencias específicas e inespecíficas a los antimicrobianos. La directriz establece también las respectivas funciones y responsabilidades de la industria farmacéutica veterinaria, los veterinarios, los farmacéuticos y los productores agropecuarios. Palabras clave Alimentos – Autoridades competentes – Autorización de comercialización – Contención de las resistencias – Medicina humana – Medicina veterinaria – Normas internacionales – Organización mundial de sanidad animal – Resistencia a los productos antimicrobianos – Salud pública. __________ 266 OIE International Standards on Antimicrobial Resistance, 2003 6. Prudent use and containment of resistance Antibiotic use in animals and the emergence of antibiotic resistance in human commensal microbes and zoonotic pathogens D.L. Smith (1) & J.A. Johnson (2) (1) Department of Epidemiology and Preventive Medicine, University of Maryland School of Medicine, 660 West Redwood St., Baltimore, MD 21201, United States of America (2) Veterans Affairs Maryland Health Care System and Department of Pathology, University of Maryland School of Medicine, 660 West Redwood St., Baltimore, MD 21201, United States of America Introduction Policy makers have the unenviable task of making a decision about the use of antibiotics for animal growth promotion. There is very limited scientific data on which to base such a decision, and its interpretation is highly controversial. Mathematical models can be an extremely useful policy tool by generating reasonable expectations about the relationship between cause and effect, and the likely impact of a policy. I discuss mathematical models for the chain of events leading from antibiotic use in animals to a nosocomial infection, and make some general observations about the kind of impact that should be expected. Risk assessment As antimicrobial resistance has increased in the clinical setting, the use of avoparcin and virginiamycin in agriculture for growth promotion and fluoroquinolones for treating infection in poultry has become the focus of intense debate. Agricultural use of antibiotics has potential negative public health impacts because similar antibiotics are important in human medicine. In response to calls for a science-based policy, governments have recently commissioned risk assessments. Risk assessment provides well-developed methodology and a common framework to synthesise vast amounts of information from multiple academic disciplines. A risk assessment generates three kinds of output: a best guess, an assessment of how close the best guess would approximate an outcome if reality is exactly like the model, and a guess about how different reality and the risk model might be. Risk assessment methodology is transparent, but the mathematical language and methods are not familiar to many stakeholders. Despite a familiarity with the concepts, they remain excluded from the debate about the merits or failings of the model. Most critically, the availability of high quality data is often limited; scientists do not usually collect data that are useful for a risk assessment. When they do collect useful data, they are rarely reported in a form that can be integrated into a risk assessment. The current lack of good quantitative data about antibiotic resistant bacteria at each step severely limits the usefulness of the standard risk assessment in making policy decisions regarding antibiotic use. OIE International Standards on Antimicrobial Resistance, 2003 267 6. Prudent use and containment of resistance The virtues of simple models Simple mathematical models are an important companion to risk assessment, and a useful tool for studying the spread of antibiotic resistance (1). They provide a clearer picture of basic concepts than do highly detailed risk assessments because they are based on concise but vastly simplified assumptions about the underlying processes. Simple models are a much easier way to relate cause and effect in long chains of causation by linking several flexible links in a chain into a single stiff one. The strength of simple models is that the relationships between major assumptions and the output become much easier to understand because several quantitative relationships are reduced to one. The weakness is that the estimates are probably biased by the stiffness of the single relationship, and sometimes difficult to parameterise with data. On the other hand, a simple model may be a more useful representation than a perfect model that includes more detail (3). Each step in a mathematical representation of the world is a set of assumptions and parameters; these must be estimated. Each parameter estimate and assumption has uncertainty associated with it. The uncertainty from one step is propagated through each subsequent step, and the conclusion of the complicated model may be far more variable than the simple one. The trick is to find a model that is an appropriate tradeoff between simple but biased and complex but variable. This is the essence of good model building and parsimony, a fundamental principle governing all of science. Farm-to-fork For example, it is widely recognised that the use of antibiotics for growth promotion has selected for antibiotic resistance in bacteria in the gut of food animals. These resistant bacteria contaminate meat during production, and can be recovered on retail products. Consumers are exposed to resistant bacteria through improper handling of the meat. Farm-to-fork risk assessments provide detailed descriptions of each step in this process, from the pen, through transport and processing, to the time that the bacteria colonise the consumer. Thus, the use of antibiotics in farm animals contributes to an increased prevalence of antibiotic resistant bacteria in the gut of humans through a long chain of events. It follows that if antibiotic use in agriculture is stopped, the rate of exposure to resistant bacteria on food will eventually decline. One important question is, ‘How fast will exposure decline?’ Since resistance to antibiotics is often naturally occurring or comes from clinical antibiotic exposure, a second question is ‘What fraction of exposure to resistant bacteria is caused by the use of antibiotics in agriculture?’ A simple farm-to-fork model can be reduced to two parameters and a function that prescribes the ‘shape’ of the curve. Very simple farmto-fork models of this sort are descriptive, but they can be used to ask how well a policy would work when other factors also affect the emergence of resistance to antibiotics. The choice of a model depends on what happens after exposure. 268 OIE International Standards on Antimicrobial Resistance, 2003 6. Prudent use and containment of resistance Bacteria life-history variability Different types of models are needed for different types of resistant bacteria. The bacteria that cause human infections vary widely in their ecology, epidemiology and life-history traits. At two extremes are Campylobacter jejuni and Enterococcus faecium. According to common wisdom, C. jejuni are zoonotic pathogens exclusively acquired through contaminated food. Following exposure, they may establish transient populations that are usually associated with disease, and they are never or rarely transmitted from human-to-human. In contrast, E. faecium are human commensals (harmless or beneficial bacteria). Colonisation is often transient, but occasionally persists for months or possibly years. Human-to-human transmission of E. faecium is common. More accurately stated, colonised humans frequently shed E. faecium and others are exposed, but colonisation is less common, and poorly understood. C. jejuni are rarely found in humans, except when they cause disease. Although E. faecium are commonly found in humans, they may not be the same strains that circulate among farm animals. One hypothesis is that E. faecium are host specific; strains isolated from animal populations are adapted to the unique environment of that host species’ gut, and may be unlikely to colonise a different species. Bacterial species vary in their degree of host specificity, their propensity for human-to-human transmission, and their propensity to colonise without causing disease. Antibiotic resistance in zoonotic infections Assuming the common wisdom about C. jejuni is correct, it follows that campylobacteriosis is always the direct result of exposure to C. jejuni on contaminated food. Computing the fraction of fluoroquinalone (FQ) resistance that is attributable to FQ use in animals is straightforward. If an infection is FQ resistant, resistance either preceded infection, or it evolved in response to FQ use in humans. It follows that an increase in the fraction of FQ resistance among human cases of campylobacteriosis is due to some change in the C. jejuni populations in animals, since there is no human reservoir. Therefore, all the increases in FQ resistance from a baseline rate are directly attributable to FQ use in farm animals. One alternative is that FQ resistance accumulates in some human commensal bacteria, and is subsequently acquired by C. jejuni. If this is true, the attributable fraction must account for the more complex dynamics associated with commensal bacteria. Antibiotic resistance in commensal pathogens Resistance to antibiotics among human commensals may occur in two ways. Direct exposure to antibiotic resistant bacteria on food may lead to increased prevalence of resistant bacteria in humans if the bacteria from animals colonise the human gut or transfer mobile genetic elements including resistance genes. Once resistant bacteria colonise humans, they may spread to other humans. Farm-to-fork models describe only half the problem; they must be coupled to models of the infectious process from household-to-hospital (1). Evidence suggests that transmission rates are higher in places where antibiotics are heavily used in human medicine. Simple models predict OIE International Standards on Antimicrobial Resistance, 2003 269 6. Prudent use and containment of resistance that prevalence of resistance over time follows a sigmoidal curve. At any point, prevalence depends on the previous history of antibiotic use in farm animals and subsequent transmission aided by its use in human medicine. A simple model, simulating a ‘counter-factual’ population where growth promoters were never used, can illustrate the impact of antibiotic use, but the non-linearity in the model leads to counter-intuitive conclusions. Basic reproductive number One key term is called the basic reproductive number, usually denoted R0, and defined in this case as the number of new people who would be colonised by an average individual, carrying resistant bacteria, introduced into a population with no resistance. The concept plays a central role in the understanding of infectious processes in populations by providing a simple criterion for determining when an infectious agent will spread: if the first case generates more than one new case (R0 > 1), resistance persists. When resistance is being introduced into a population from external sources, R0 and the rate of introduction co-determine the prevalence. When R0 is low (as with C. jejuni), the rate of exposure from animal antibiotics is the main determinant of prevalence. For intermediate values of R0, the two interact in complicated ways. When R0 is high, exposure from animal antibiotics may initiate an epidemic, but thereafter plays a minor role (4). Rare but important events A key insight from this simple model is that rare events may be enormously important, especially for intermediate and high values of R0 that are characteristic of human commensal microbes. For example, the first case of vancomycin resistant Staphylococcus aureus (VRSA) was a product of peculiar circumstances (2), but if it had not been discovered then, it might have triggered an epidemic that would be 270 OIE International Standards on Antimicrobial Resistance, 2003 6. Prudent use and containment of resistance devastating. The value of delaying that epidemic is the sum of all cases prevented between the actual start and the counterfactual start of the epidemic (Figure). The concurrent use of an antibiotic in a hospital and on a farm provides ideal conditions for new strains of resistance to appear in hospitals. Farms and hospitals are probably ideal environments for genes conferring high-level resistance to move from environmental bacteria species into the nosocomially important bacteria; from the perspective of resistance, they are evolutionary incubators. With heavy antibiotic use in hospitals, the emergence of nosocomially important strains seems inevitable, but they may happen much earlier if a drug is concurrently used in animals. Such effects are not studied from the paradigm of ‘experimental,’ but must be examined using the paradigm of a ‘historical’ science, such as evolutionary biology. For policy makers, it represents a real and credible threat, though such a threat may be difficult to estimate. How does one estimate R0 for a nosocomial pathogen that has not yet emerged? How does one attribute an impact that is historical? These are open questions, but the simple models illustrate that they are the relevant ones for this debate. Prudent use The potential medical impact of using an antibiotic for animal growth promotion is most critical in the honeymoon period, when antibiotic resistance is virtually absent. It follows that prudent use of antibiotics in animals can occur only after medically important resistant bacteria have become common in human populations. What about the antibiotics that have already been used for animal growth promotion, such as quinupristin/dalfopristin (QD)? The answer depends on how QD (Synercid) is used in humans, how fast QD resistance declines in animals after a ban, and how rare QD resistance becomes. The approval of oxazolidinone (Linezolid) may have reduced the demands on QD use in humans, opened a window of opportunity and increased the positive effect of a ban. On the other hand, it is possible that the impact of QD use for animal growth promotion for nearly three decades is not reversible in the short term since QD is approved for treating vancomycin-resistant enterococci. Conclusion The conclusions of mathematical models are only as good as the assumptions they are based on, but models should also be judged by their ability to help us understand what we are studying. The assumptions of our model are based on well-established general principles in biology; the conclusions depend upon the values of key parameters. Substantial disagreement exists about the value of these parameters, and the associated conclusions. Making complicated risk assessment models can’t eliminate disagreement about uncertain scientific principles, but they can certainly disguise the central issues. When little is known, it is probably best to keep the models simple. References 1. Bonten M.J.M., Austin D.J. & Lipsitch M. (2001). – Understanding the spread of antibiotic resistant pathogens in hospitals: mathematical models as tools for control. Clin. infect. Dis., 33, 1739-1746. OIE International Standards on Antimicrobial Resistance, 2003 271 6. Prudent use and containment of resistance 2. Centers for Disease Control (2002). – Staphylococcus aureus resistant to vancomycin – United States. MMWR, 51, 565-567. 3. Ludwig D. & Walters C.J. (1985). – Are age-structured models appropriate for catcheffort data? Can. J. Fish. aquat. Sci., 42, 1066-1072. 4. Smith D.L., Harris A.D., Johnson J.A., Silbergeld E.K. & Morris J.G. Jr. (2002). – Animal antibiotic use has an early but important impact on the emergence of antibiotic resistance in human commensal bacteria. Proc. natl Acad. Sci. USA., 94, 1152-1156. __________ 272 OIE International Standards on Antimicrobial Resistance, 2003 6. Prudent use and containment of resistance Prudent use of antibiotics and containment of antimicrobial resistance: the role of medical associations, guidelines and interventions I.M. Gould Department of Medical Microbiology, Aberdeen Royal Infirmary, Foresterhill, Aberdeen, AB25 2ZN, United Kingdom Introduction Prudent use of antibiotics is a responsibility of all people, not just all health professionals and especially not just medical specialists in the field of infection. This makes the problem all the more difficult to address. Antibiotics are unlike any other group of pharmaceutical products in that all areas of medical practice have cause to use them and thus they are usually prescribed by non-specialists, often with very little knowledge of antibiotics or the consequences of their actions. In addition, they are often available (legally or illegally) as over-the-counter purchases, with or without trained medical or pharmaceutical advice. This makes education, which must be the main responsibility of Medical Societies, a daunting task. Education Public education is not a traditional area that medical societies have been involved in and in the area of antibiotic prescribing, great efforts have recently been made by societies specifically formed for that purpose such as, Stichting Werkgroep Antibioticabeleid ([email protected]) (Holland), Strategigruppen för Rationell Antibiotikaanvändning och Minskad Antibiotikaresistens (www.strama.org) (Sweden) and Alliance for the Prudent Use of Antibiotics (www.healthsci.tufts.edu/apua/apua.html). In the United Kingdom (UK) this has been the responsibility of the Government and there is some soft evidence that it has been beneficial. Certainly, the levels of community antibiotic prescribing have decreased substantially in the past five years and are now back to 1991 levels. This still leaves them at twice the level of corresponding figures from Holland, so there is scope for further improvement. Some of this improvement may be due to improved professional education of General Practitioners (primary care prescribers) using advice prepared by government advisory committees which had medical society representation. There is clearly a role for improved post-graduate education by medical societies. Here there is an obvious need for better links between specialist societies such as the British Society for Antimicrobial Chemotherapy (BSAC – www.bsac.org) and primary care medical associations. BSAC has produced CD-roms on treatment of community respiratory infections. In the UK, the Royal Colleges and in Europe the European Society of Clinical Microbiology & Infectious Diseases are the main players in continuous professional education but other specialist infection societies, such as the Hospital OIE International Standards on Antimicrobial Resistance, 2003 273 6. Prudent use and containment of resistance Infection Society (www.his.org.uk) and the British Infection Society, also have an important role to play. Again, liaison between the specialist societies such as the European Society of Clinical Microbiology and Infectious Diseases and the general medical societies in each country should become a priority to ensure that they have access to the correct educational material. When it comes to undergraduate medical education, a BSAC working group recommended, in 1993, that antibiotic prescribing be given a higher priority. At the request of the UK government, that working group has been re-formed and has written to the Deans of the medical schools to ascertain any progress in their areas, which over the past fifteen years have suffered keen competition from other areas. Clinical practice guidelines These are the preserve of the specialist societies and this is a rapidly expanding area with the potential for great benefit. The legal basis of guidelines and the ability to implement them successfully are, however, an area of great debate. Computerisation of prescribing allows for much easier implementation of guidelines, as there are too many available for non-specialists to be knowledgeable on them all or to retain access to paper copies. These days, evidence based guidelines, drawn up according to strict methodological regulations, are considered best, as opposed to previously popular ‘expert opinion’ guidelines. To those involved in drawing up these guidelines it is evident that much of modern medical practice has a deficient evidence base. The Scottish Intercollegiate Guideline Network (SIGN) are leaders in the field and are based at the Royal College of Physicians in Edinburgh. They have several guidelines on areas of antibiotic prescribing as do many specialist societies, although currently, not many of those from other societies are truly evidence-based due to the resources needed to produce such documents. Currently BSAC, with the help of Cochrane are producing evidence-based guidelines on antibiotic control measures in hospitals and a Canadian group are doing the same for community prescribing. (1). Evidence based clinical guidelines can form the basis for standards of clinical governance and Minimum Standards of Antibiotic Stewardship for hospital and community health-care administration can be set from recommendations made originally by the Infectious Diseases Society of America (IDSA) and BSAC and more recently by the European Society of Clinical Microbiology and Infectious Diseases (ESCMID). There are also more detailed practical recommendations set at two levels, one for countries with little established infrastructure in this area and another for those countries with well developed systems of antibiotic stewardship (2). 274 OIE International Standards on Antimicrobial Resistance, 2003 6. Prudent use and containment of resistance Evidence based interventions A recent evidence based review of the literature, using the criteria of the Cochrane Effective Practice and Organisation of Care Group, has identified only about 10% of studies on hospital antibiotic prescribing intervention as qualifying, mainly as timeseries analysis, for inclusion in an evidence based review when outcomes other than reduced prescribing are needed. Such outcome measurements include reduction in resistance, clinical outcome measures and colonisation with C. difficile. Nevertheless, if one includes a broader literature on intervention to change prescriber behaviour (non-antibiotic), then there are six Cochrane reviews identifying education, guidelines, academic detailing and audit/review as being effective, although trying to change the habits of doctors has been described as an exercise in futility designed to induce premature ageing! Education can obviously take on many forms: it can be either persuasive or restrictive and must be appropriate to the local situation, taking account of cultural and social factors and financial aspects. Barriers to change will mostly be physician based but may, of course, involve the system and patient. The attitude, knowledge and behaviour of physicians must be addressed, always looking for hidden agendas and making clear the benefits of changing practices. There are 2 new initiatives from the European Commission: Antibiotic Resistance Prevention and Control (ARPAC) and European Surveillance of Antibiotic Consumption (ESAC). The overall aim of ARPAC (www.abdn.ac.uk/arpac) is to lay the foundations for a better understanding of the emergence and epidemiology of antibiotic resistance in human pathogens and to evaluate and harmonise strategies for prevention and control of antibiotic resistant pathogens in European hospitals. The main aims of ESAC ([email protected]) are to collect, validate, and present in a standardised and meaningful manner, data pertaining to the use of antimicrobials in human medicine in all Member States of the European Community, in countries signatories to the Agreement on the European Economic Area and in associated countries of Central and Eastern Europe. In the future it is hoped that Medical Associations will be more involved in a two way liaison process with industry, regulating authorities, politicians and healthcare strategists to advise on key issues and formalise a greater role, not only in education, but also in clinical governance. References 1. Gould I.M. (2001). – Minimum antibiotic stewardship measures. Clin. Microbiol., 7 (Suppl. 6), 22-26. 2. Keuleyan E. & Gould I.M. (2001). – Key issues in developing antibiotic policies: from an institutional level to Europe-wide. European study group on antibiotic policy (ESCAP), subgroup III. Clin microbiol. Infect., 7 (Suppl. 6), 16-21. __________ OIE International Standards on Antimicrobial Resistance, 2003 275 6. Prudent use and containment of resistance Prudent use of antibiotics and containment of antimicrobial resistance J. Edwards Kakariki Grove, Waikanze 6454, New Zealand It was not until the last decade of the 20th Century that antimicrobial resistance became recognised as a serious public health issue. In just sixty years after the first parenteral use of benzyl penicillin, a nanosecond in evolutionary time, real increases in antimicrobial resistance have been reported from all corners of the globe. Antimicrobial resistance is now acknowledged as a serious public health issue, not only in hospital settings, but also in the community. The World Veterinary Association (WVA) recognises the importance of emerging antimicrobial resistance in both human and veterinary medicine Recent attention has focused on antimicrobial use in agriculture. There is limited but conclusive evidence that resistant bacteria in food producing animals have spread to humans, either directly or through the food chain. In the case of E. coli and enterococci the transfer of genetic material that confers resistance is equally or more important. The risk and rate of transfer differ between bacteria and the location of resistant genes. There is general agreement that monitoring of resistance is needed in agriculture. Surveillance information on the prevalence and trends in antimicrobial resistance will contribute to: – the formation of local and national antimicrobial treatment guidelines – infection control policies – the development of strategies to contain the emergence and spread of resistance – the measurement of the effectiveness of intervention strategies. Over the last two to three years, various international and national expert groups have considered and reported on the problem of increasing antimicrobial resistance. This issue has been subject to on-going development of international polices, and highlights the need for increasing collaboration between the major international organisations. We need to develop a cohesive international policy in a timely manner and plan for future collaboration where more than one organisation has an interest. Veterinary involvement is critically important in the formulation and implementation of measures to control antimicrobial resistance. This includes non-government veterinarians because of their role in pre-harvest food safety and the control and responsible prescription of medicines intended for use in animals. It is essential that 276 OIE International Standards on Antimicrobial Resistance, 2003 6. Prudent use and containment of resistance veterinarians in private practice be involved in general and world-wide policy formulation. The OIE (World organisation for animal health) is the forum for government veterinary services. The WVA is the only non-governmental organization representing the veterinary profession globally and is working to strengthen its role. The WVA appreciates its involvement with the OIE and offers to assist the OIE in all matters of veterinary concern. The veterinary profession: – takes its responsibilities very seriously – will support initiatives as part of meeting it’s societal obligation to assist in the drive for food security and food safety. The WVA has a communication network to distribute policies, adopted conclusions and information on new developments to the veterinary profession around the world. In 1999 the WVA joined the International Federation for Animal Health (IFAH) and the International Federation of Agricultural Producers (IFAP) in the proactive development of guidelines for the prudent use of antibiotics. Recognising that antibiotics are health management tools that enhance good husbandry practices for the purpose of disease prevention, disease treatment and production enhancement, the following were promoted: – the responsible and prudent use of antibiotics – codes of good practice – quality assurance programmes – herd health surveillance programmes – education programmes. Antibiotics shall be used under the supervision of a veterinarian. Therapeutic antibiotics should be used when it is known or suspected that an infectious agent is present which will be susceptible to therapy. It is the responsibility of the veterinarian to choose the antibiotic on the basis of his/her informed professional judgement, balancing the risks and benefits for humans and animals. When antibiotics need to be used for therapy, sensitivity testing should, whenever possible, be part of the informed professional clinical judgement. There should be careful attention paid to the species and disease indications and contra-indications, withdrawal periods and storage instructions. Off-label use of antibiotics should only be exceptional and always be under the professional responsibility of a veterinarian. Antibiotics used for therapy should be used: – for as long as needed – over as short a dosage period as possible – at the appropriate dosage regimen. Records should be kept of all antibiotic administrations. OIE International Standards on Antimicrobial Resistance, 2003 277 6. Prudent use and containment of resistance Co-ordinated susceptibility surveillance should be conducted. Efficacious, scientifically proven alternatives to antibiotics are needed as an important part of good husbandry practices. In the drive for food security and food safety; globalisation and opening of borders will put animal production industries under a lot of pressure. The economic aspects of animal production will continue to be of major importance in the future. The WVA encourages all parties involved in the development of animal management systems, especially those dealing with intensive animal production, to respect all available data on the basic and essential needs of animals. The WVA encourages the pharmaceutical industry to continue to research the improvement of existing feed additives, as well as the development of new compounds of non-antibiotic origin, such as bacteria, binders and organic acids which can replace the use of antibiotics as feed additives. Alternatives to antibiotics are a necessary tool to assist the economic feasibility of intensive animal production. The WVA urges regulatory bodies globally to: – implement regulations to prevent misuse of antibiotics and reduce the possibility of development of resistant strains. – implement viable structures to monitor the development of resistance and adopt measures to prevent such development. The WVA encourages national veterinary organisations to adopt clear policies and guidelines about the use and misuse of antibiotics. Conclusion The WVA will continue to: – collaborate with global organisations in the development of policies and programmes to restrict and contain antimicrobial resistance – advise veterinarians around the world about the risks of antimicrobial resistance and encourage veterinarians to take all reasonable safeguards to minimise the development and spread of antimicrobial resistance. The World Veterinary Association: http://www.worldvet.org/. __________ 278 OIE International Standards on Antimicrobial Resistance, 2003 6. Prudent use and containment of resistance Prudent use and containment of antimicrobial resistance – the work of the responsible use of medicines in agriculture alliance B. Jennings NFU, 164 Shaftesbury Avenue, London WC2 H8HL, United Kingdom Antimicrobials have made a major contribution to farm animal health and welfare for several decades. They are vital for the treatment and control of animal disease. The use of a limited group of them at low levels as digestive enhancers has also made them a useful tool for farmers. The Responsible Use of Medicines in Agriculture (RUMA) alliance was established in November 1997. Its aim was to facilitate and promote, by means of a co-ordinated and integrated approach involving all stakeholders, best practice in the use of veterinary medicines, beginning with antimicrobials. This was in direct response to concerns about the crossover of resistant bacteria from livestock to the human population, and the associated possibility of medical antimicrobial treatments becoming less effective. The RUMA alliance is a coalition of agricultural, veterinary, welfare, pharmaceutical, retail and consumer interests which aims to keep under review the use of antimicrobials in food animals, and establish practical strategies to enable farmers to reduce the need for their use. It is an independent voice, based on science. Best practice guidelines have now been produced for all major farmed species, and in some cases they are already being used as part of farm assurance schemes. Part of the future work of RUMA will be to monitor uptake, and modify the guidelines if necessary in the light of experience. RUMA is not a political organisation, but will always encourage a rational and scientific approach to the availability of antimicrobial availability, and help farmers comply with any legislative changes. The future aims of the RUMA alliance are: – to identify issues of scientific and public concern in the areas of public health, animal health, animal welfare, and the environment – to provide an informed consensus view on the identified issues which will be developed by discussion and consultation – to communicate as widely as possible the guidelines which describe best practice in the use of medicines – to advise industry in the implementation of ‘best practice’, especially in the development of Codes of Practice and Assurance schemes – to influence the way medicines are used and the way in which that use is perceived by consumers, the public health authorities, the media, and others. __________ OIE International Standards on Antimicrobial Resistance, 2003 279 6. Prudent use and containment of resistance Prudent use and containment of antimicrobial resistance in developing countries D.K. Byarugaba Department of Veterinary Microbiology and Parasitology, Faculty of Veterinary Medicine, Makerere University, P.O. Box 7062, Kampala, Uganda Introduction When antimicrobials are used for too short a time, at too low a dose, at inadequate potency or when poorly indicated, the likelihood that bacteria and other microbes will adapt and replicate rather than be killed is greatly enhanced. Much evidence supports the view that the high consumption of antimicrobials in animals and man is the critical factor in selecting resistance (3). While overuse is responsible for much of the antimicrobial resistance in developed nations, paradoxically it is underuse that is responsible for resistance in developing countries (DCs). Antimicrobial use whether for therapy, prevention of infectious diseases or as performance enhancers leads to selection for antibiotic resistant micro-organisms, not only among pathogens but also among bacteria of the endogenous microflora of animals and man. Many other interconnected factors fuel the emergence of antimicrobial resistance. In DCs, where poverty, lack of commitment by governments, greed, corruption, hunger and, recently, the acquired immune deficiency syndrome (AIDS) pandemic, limit access to antimicrobials and lead to underuse, misuse, and use of poor quality counterfeit drugs leading to more rapid selection of resistance. It is therefore important that interventions that can be used to slow the emergence and reduce the spread of resistance be used. Particular attention to interventions that encourage prudent use of the available antimicrobials is very important in the strategy for controlling antimicrobial resistance. Situational analysis Some studies that have been carried out in DCs have revealed that antimicrobial agents are grossly misused (2). The most commonly available antimicrobial agents which have been used for many years: penicillins and tetracyclines, have suffered the greatest abuse. The factors that influence antimicrobial abuse can be categorised into three: a) those related to the overall regulatory framework and policies regarding health systems b) those related to the service providers c) those related to the users. These factors are interrelated and their interactions make the problem very complex. The majority of people in many DCs suffer from chronic poverty, socio-economic marginalisation, food insecurity and, most recently, the devastating impact of the 280 OIE International Standards on Antimicrobial Resistance, 2003 6. Prudent use and containment of resistance human immunodeficency virus (HIV)/AIDS pandemic. Poverty is the most important underlying factor that leads to misuse of antimicrobial agents and resistance in DCs. Lack of purchasing power for full doses or for seeking professional advice, lack of access to proper facilities, lack of education etc. lead to problems such as self prescription and under dosing, which may precipitate microbial resistance (summarised in Figure 1 below). Related to poverty and lack of resources, is the lack of reliable diagnostic facilities in general and, especially for culture and sensitivity testing to make differential diagnosis, organism identification and to obtain reliable data on antimicrobial susceptibility patterns on which to base prescription. This means that greater amounts of antimicrobials are often prescribed to cover any possible infection without supporting reliable laboratory data. In many DCs, HIV/AIDS has caused severe economic constraints that have had negative impacts on various sectors and have induced governments to refocus their attention and resources to other critical areas. The problem of inadequate resources compounded with lack of relevant laws and their implementation, as well as service provider behavior, make the situation of antimicrobial resistance in DCs precarious. Poverty ‘Unrestricted availability’ Over or under use and self prescription Wrong conceptions Stocking excess drugs on farms Underdevelopment Imprudent use Regulatory related factors Advertisment pressure by drug cos. Treatment expectations Services proivider related factors Antimicrobial resistance Fig. 1 Summary of some of the factors responsible for irrational use of antimicrobial agents in DCs In many institutions in DCs, antimicrobial resistance is never emphasised during the training of students and prudent drug use is never taught. The practical significance of the problem is never presented to students. Thus, they leave school un-informed and without practical understanding of antimicrobial resistance and its implications. This lack of practical knowledge may play a role in irrational or no prescribing at all. OIE International Standards on Antimicrobial Resistance, 2003 281 6. Prudent use and containment of resistance Most pharmacies are manned by attendants who are not the licence holders and do not have the qualifications to enable them to handle drugs professionally. This factor is extremely important as antimicrobials and other drugs are frequently purchased without prescription. Antimicrobials are often sold at subtherapeutic doses because the users insist on a specific quantity. Also, the smaller pharmacies keep a limited range of antimicrobial drugs and even if users have a prescription they may be advised to buy available alternatives. Users may demand antimicrobials of their choice either based on their own or somebody else’s experience or from advertisements on TV, radio or very attractive posters. Others demand specific antibiotics for simple ailments which do not require antibiotics. Pharmaceutical companies market some of the drugs available on prescription directly to the public by very attractive direct-to-user advertising through multi-media such as television and radio. This has the potential to stimulate demand by playing on the users’ relative lack of education about the evidence supporting the use of one treatment over another. These advertising methods are very effective. Many pharmaceutical companies also give incentives or commissions to drug agents and outlets which may encourage crude methods of advertising. In addition, pharmacists and dispensers gain financially from over-dispensing and dispensing more expensive broad-spectrum agents when cheaper narrow spectrum agents would suffice. The regulatory mechanisms for control of antimicrobial manufacture, market authorisation, distribution and use are still very weak. Many DCs still depend on ancient laws but even where new laws have been made, their implementation and enforcement are weak. Some reports have indicated the sale of expired drugs not only in urban centres but also in rural settings. Antibiotics are sold in cattle markets next to cowsheds in direct sunlight despite laws or guidelines. Counterfeit products also enter DCs, especially through smuggling or corruption. Structural adjustment policies such as privatisation and liberalisation have also affected the procurement and supply of antimicrobial agents and as DCs adjust to these changes, new challenges emerge. Improving antimicrobial drug use The improvement of antimicrobial use in order to contain antimicrobial resistance is crucial and requires world-wide concerted efforts. Several recommendations have been made and these revolve around a combination of educational measures, regulatory and managerial issues. Integrated mechanisms will be required, to educate, train and sensitise all stakeholders, such as health service providers, policy makers, and users to understand and appreciate the problems and the consequences of failing to institute measures to contain the problem. With proper sensitisation of all the stakeholders, containment will be much easier as proper policies will be formulated, services will be delivered professionally and users will utilise drugs properly. These sensitisation programmes may involve massive campaigns to educate users about prudent antimicrobial use. Increased interaction and planning among informed policy makers, service providers and users will go a long way in improving antimicrobial use. 282 OIE International Standards on Antimicrobial Resistance, 2003 6. Prudent use and containment of resistance This partnership is essential for influencing policies such as the incorporation of proper antimicrobial use in education systems. There are opportunities for the promotion of prudent use of antimicrobial agents in DCs. These include: – optimal utilisation of available resources such as facilities, equipment, technical personnel – incorporation of prudent drug use and antimicrobial resistance issues in programmes such as in schools or universities, farmer education, extension and training programmes – a role for professionals societies to sensitise and regulate their members – the formation of a collaborative partnership with other developed nations – support from international agencies such as, the United Nations Development Programme, the OIE (World organisation for animal health) and the World Health Organization (WHO) etc. Conclusion Imprudent use of antimicrobial agents is the major factor responsible for the emergence of antimicrobial resistance world-wide and requires immediate action. Both the OIE and the WHO have published guidelines for containing the problem of antimicrobial resistance in veterinary medicine (1, 4). Developing countries, therefore, need to carry out situational analyses of the problems and opportunities in their countries and introduce those interventions that are feasible within the limits of available resources (financial, human and physical) in order to improve the use of antimicrobials. References 1. Anthony F., Acar J., Franklin A., Gupta R., †Nicholls T., Tamura Y., Thompson S., Threlfall E.J., Vose D., van Vuuren M. & White D.G. (2001). – Antimicrobial resistance: responsible and prudent use of antimicrobial agents in veterinary medicine. Rev. sci. tech. Off. int. Epiz., 20 (3), 829-839. 2. Byarugaba D.K. et al. (2001). – Development of sustainable strategies for the management of antimicrobial resistance in man and animals at district and national level in Uganda. Feasibility Study Report. May, Makerere University, Kampala. 3. Danish Integrated Resistance Monitoring and Research Programme (2000). – DANMAP 99 – consumption of antimicrobial agents and occurrence of antimicrobial resistance in bacteria from food animals, food and humans in Denmark. Statens Serum Institut, Danish Veterinary and Food Administration, Danish Medicines Agency and Danish Veterinary Laboratory, July. OIE International Standards on Antimicrobial Resistance, 2003 283 6. Prudent use and containment of resistance 4. World Health Organization (WHO) (2000). – WHO global principles for the containment of antimicrobial resistance in animals intended for food. WHO/CDS/CSR/APH/2000.4. __________ 284 OIE International Standards on Antimicrobial Resistance, 2003 Web links http://212.3.246.142/1/CINLADPBFCOCOANBLLJOGIFNPDBY9DBW EW9DW3F71KM/BEUC/docs/DLS/2002-00868-01-F.pdf http://dx.doi.org./10.1099/ijs.0.02408-0 http://europa.eu.int/comm/food/fs/sc/ssc/out50_en.pdf www.abdn.ac.uk/arpac www.anmv.afssa.fr/ccoie/documents/cdrom1999/Antibiotics/index.htm www.arru.saa.ars.usda.gov/main.htm www.bsac.org www.cdc.gov/epiinfo/Epi6/EI6dnjp.htm www.cdc.gov/narms/ www.consumersinternational.org www.ers.usda.gov/data/sdp/view.asp?f=food/89015/ www.escmid.org www.fda.gov/cvm/antimicrobial/antimicrobial.html www.fda.gov/cvm/antimicrobial/Risk_asses.htm www.foodriskclearinghouse.umd.edu/risk_assessments.htm www.fsis.usda.gov/OPHS/ecolrisk/home.htm www.fsis.usda.gov/ophs/risk/index.htm www.health.gov.au/pubs/jetacar.pdf www.his.org.uk www.iatp.org/EatWell/orgResults.cfm www.KeepAntibioticsWorking.com www.laegemiddelstyrelsen.dk www.nal.usda.gov/fnic/foodborne/risk.htm www.nccls.org www.nval.go.jp/taisei/taisei.html www.oie.int www.onerba.org www.strama.org www.SVS.dk www.tacd.org www.tufts.edu/med/apua/ www.vetinst.dk/file/danmap2000.pdf www.who.int/emc/amr.html www.whocc.no/atcddd/atcsystem.html www.worldvet.org/ OIE International Standards on Antimicrobial Resistance, 2003 285