Antimicrobial Stewardship in Veterinary Medicine

ABSTRACT

While antimicrobial resistance is already a public health crisis in human medicine, therapeutic failure in veterinary medicine due to antimicrobial resistance remains relatively uncommon. However, there are many pathways by which antimicrobial resistance determinants can travel between animals and humans: by close contact, through the food chain, or indirectly via the environment. Antimicrobial stewardship describes measures that can help mitigate the public health crisis and preserve the effectiveness of available antimicrobial agents. Antimicrobial stewardship programs have been principally developed, implemented, and studied in human hospitals but are beginning to be adapted for other applications in human medicine. Key learning from the experiences of antimicrobial stewardship programs in human medicine are summarized in this article—guiding the development of a stewardship framework suitable for adaptation and use in both companion animal and livestock practice. The antimicrobial stewardship program for veterinary use integrates infection prevention and control together with approaches emphasizing avoidance of antimicrobial agents. The 5R framework of continuous improvement that is described recognizes the importance of executive support; highly motivated organizations and teams (responsibility); the need to review the starting position, set objectives, and determine means of measuring progress and success; and a critical focus on reducing, replacing, and refining the use of antimicrobial agents. Significant issues that are currently the focus of intensive research include improved detection and diagnosis of infections, refined dosing regimens that are simultaneously effective while not selecting resistance, searches for alternatives to antimicrobial agents, and development of improved vaccines to enhance immunity and reduce disease.

STEWARDSHIP AND THE USE OF ANTIMICROBIAL DRUGS

Stewardship implies a process of caring and responsible management (1), and as we face the progressive extension of multiresistance amongst bacteria in veterinary and human medicine, and in agriculture, there is a need to apply stewardship efficiently to the use of antimicrobial drugs so that we can preserve and extend their efficacy.

The relationship between the use of antimicrobial drugs and the development of resistant bacteria is now well established (2). The problem of antimicrobial resistance has been extensively reviewed by the O’Neill committee in the United Kingdom, which published its final report and recommendations in May 2016 (3). The committee examined the problem in all fields of antimicrobial use and on a worldwide basis. It concluded that human deaths as a consequence of antimicrobial resistance and inability to treat and control microbial infections were already of the order of 700,000 per year and estimated to rise to 10 million per annum (a figure that includes infections with resistant HIV and malaria, as well as tuberculosis and other bacterial infections), with a cumulative cost to global productivity of $100 trillion to the year 2050, unless appropriate actions are taken to deal with this problem.

An important driver of resistance in bacteria is the use of antimicrobials in agriculture and veterinary medicine. This is rising rapidly, driven by the unprecedented rate of increase in global demand for animal protein for human consumption, particularly in middle-income countries where extensive farming is being replaced by large-scale intensive production systems that routinely use large quantities of antimicrobials, often with no veterinary intervention (4). The O’Neill committee concluded that the total amount of antimicrobial agents being used in agriculture was at least as great as that used in human medicine. Furthermore, using data from the U.S. Food and Drug Administration (FDA), it calculated that, in 2012 in the United States, 70% of the total weight of antibiotics defined by the FDA as medically important in humans was sold for use in animals (5). This situation may no longer apply, because new guidelines for antimicrobial use, including expanded need for veterinary prescription, were introduced by the Center for Veterinary Medicine in the United States on 1 January 2017.

It is recognized that resistant bacteria emerging in farm animals can pose a risk not only to the animals but also to farmers, veterinarians and their families, to workers in meat and milk production and sales, and to consumers; such bacteria include commensal organisms which may not necessarily cause significant disease among farm animals but can pose a significant risk through direct or indirect transfer to members of the public (6). Although antimicrobial use in companion animal practice is much lower than that in farm animals, the close relationship between treated animals, veterinary staff, owners and their families, and the public has increased the risk of transfer and infection among them (7). It is noteworthy that transfer occurs not only from animals to humans but also in the other direction, as evidenced by the occurrence of hospital-associated methicillin-resistant Staphylococcus aureus (MRSA) colonization and infection in dogs and cats (8–10). Furthermore, the greatly increased international movement of companion animals (11), including both pets and horses, and international trade and illegal transportation of foodstuffs (12), coupled with their ability to carry resistant bacteria, increase the risk of global dissemination. Better surveillance of the impact of international trade and movement of these animals and animal products is required so that their impact on the development of antimicrobial resistance can be more accurately assessed, facilitating the establishment of appropriate control measures (13).

We have now reached a situation where bacteria involved in colonization and infection of both farm and companion animals include organisms that can be resistant to all registered veterinary drugs. These include organisms such as MRSA and methicillin-resistant Staphylococcus pseudintermedius, extended-spectrum β-lactamase-producing Escherichia coli, carbapenemase-producing E. coli and Klebsiella pneumoniae, and multidrug-resistant enterococci and Acinetobacter baumannii. These organisms, their resistance factors, and their local and global dissemination have been covered elsewhere and will not be considered here.

The generation of antimicrobial agents and the development of resistance mechanisms is part of the normal biology of bacteria and other microorganisms, a process that has been involved in bacterial evolution over a long period, as demonstrated by the identification of genes encoding resistance to a wide range of modern antibiotics in ancient DNA from 30,000-year-old Beringian permafrost sediments and from the Lechuguilla Cave, New Mexico, which has been isolated for more than 4 million years (14, 15). Indeed, predictions of the time of origin of the B3 subclass of metallo-β-lactamases, based on dating of biosynthetic gene clusters, has been estimated at 2.2 billion years ago (16). Thus, bacteria are very capable of evolving resistance to antimicrobials, and prospects for the development of new agents which will not generate resistance are very low. Implementation of effective antimicrobial stewardship (AMS) strategies is now an urgent necessity if we are to preserve the efficacy of our existing drugs and ensure optimal longevity of any new agents that are developed.

This article focuses on the processes involved in effective AMS, reviewing what has been achieved in the human medicine field, the tools and mechanisms for implementation that are becoming available for veterinary use, evidence for success, and the need for incentives or legislation in promoting compliance.

AMS in Human Medicine

The need for AMS in human medicine is now well recognized, and it is being applied in a variety of ways in human hospitals in an increasing number of countries. Systematic reviews of the effects of stewardship have been carried out by Davey et al., who reported an initial study in 2005 and followed this up in 2017 (17, 18), work corroborated by Schuts et al. (19). Each of these studies demonstrated overall beneficial effects. Davey et al. (17, 18) studied the impact of interventions designed to influence the prescription of antimicrobials to hospital inpatients and reduce antimicrobial resistance or Clostridium difficile associated diarrhea (CDAD) and their effects on clinical outcomes in a total of 155 studies. They found that interventions aiming to reduce excessive prescription were associated with reduction in C. difficile infections and colonization or infection with aminoglycoside- or cephalosporin-resistant Gram-negative bacteria, MRSA, and vancomycin-resistant Enterococcus. Interventions designed to increase effective prescribing were also able to improve clinical outcome. Meta-analysis showed that restrictive interventions such as form filling and the need to obtain approval from an infection specialist prior to antimicrobial prescription were more effective than persuasive interventions (advice on how to prescribe or feedback after prescriptions were made) for up to 6 months, but the two approaches were equally effective after that. Disturbingly, they concluded that prescribing was often suboptimal, and up to 50% of use could be inappropriate. On a positive note, Davey et al. (18) demonstrated that enablement (increasing means or reducing barriers to increase capability or opportunity) consistently increased the effect of interventions and that feedback further increased the positive impact. Schuts et al. (19) focused on four outcomes in 145 studies: clinical responses, adverse events, costs, and rates of bacterial resistance. They found that for the objectives of empirical therapy according to guidelines, de-escalation of therapy, switching from intravenous to oral treatment, therapeutic drug monitoring, use of a list of restricted antimicrobials, and bedside consultation, there were overall significant benefits for one or more of the four outcomes.

These studies show the benefits that can be obtained by the application of AMS in human hospitals. However, implementation of antimicrobial use guidelines has been problematic even in countries with vigorous policies aimed at the control of antimicrobial resistance. An example is the introduction of consensus-based guidelines developed by a multidisciplinary expert group for antimicrobial use in the treatment of meningitis in the Netherlands in 1997, which were disseminated in booklet form. A year later, a prospective study (20) showed that only one-third of patients had been treated in accordance with the guidelines. For the study, patients were divided into four groups according to risk factor status. The largest group was composed of patients with no risk factors, and 39% of these were treated empirically with third-generation cephalosporins, whereas the guidelines, which reflected the very low incidence of local resistance among likely pathogens, recommended penicillin. In a leading article referring to this study, Brown (21) speculated on the reasons why so few clinicians in The Netherlands had adopted the guidelines. These included mistrust, a feeling that the guidelines were too narrow, preference for locally developed guidelines, poor dissemination, and lack of incentives for implementation. He pointed out that such failure to adopt guidelines was far from unique and went on to review the principles underlying the generation of effective guidelines, with a particular emphasis on the need for planning and the allocation of sufficient resources for effective dissemination and implementation.

Although there is broad recognition of the need for AMS in the human field, there continues to be a lack of appropriate guidelines and effective implementation in many medical institutions and disciplines. A notable example is pediatrics, where the need for formal AMS programs has only recently been recognized and there are continuing problems relating to implementation, most importantly, the need for financial resources and for administrative consensus enabling the education of pediatricians and the creation of multidisciplinary interprofessional teams able to prepare guidelines and administer AMS programs (22). The need for AMS to be adapted to different settings is important if it is to be accepted by those involved, and it has become apparent that the impact of behavioral determinants and social norms is not being given sufficient attention (23). Clinicians often emulate the incorrect prescribing behavior of fellow clinicians, and junior doctors are likely to be influenced by their seniors (24, 25). Senior doctors like to have autonomy in decision-making and may be reluctant to interfere with their peers’ prescribing decisions (23). Thus, senior clinicians and consultants need to be involved in the development of tailored guidelines in their disciplines, aligning the guidelines with the evidence base and consultants’ preferences and ensuring that the teaching of junior doctors focuses on adherence while also indicating when deviation is justified (26).

There is particular concern regarding the excessive and indiscriminate use of agents that are regarded as being of special importance in human medicine. These have been listed by the World Health Organization (WHO) (27) under the title “critically important antimicrobials” in a document intended for public health and animal health authorities, physicians, veterinarians, and others involved in managing antimicrobial resistance. The document categorizes a wide range of agents and places them into three categories on the basis of their importance in human medicine: critically important, highly important, and important. These lists are designed to help formulate and prioritize risk assessment and management strategies for containing antimicrobial resistance due to human and nonhuman antimicrobial use when designing guidelines. Fluoroquinolones; third-, fourth-, and fifth-generation cephalosporins; macrolides and ketolides; glycopeptides; and polymyxins have been categorized as being of the highest priority for risk management. The document recommends that carbapenems, glycopeptides, oxazolidinones, and any new classes of antimicrobials developed for human therapy should not be used in animals, plants, or aquaculture. Classes of antimicrobials used in animals but not in humans are also identified and include aminocoumarins, orthosomycins, phosphoglycolipids, polyether ionophores, and quinoxalines. Antimicrobials used in food-producing animals have also been listed and categorized by the World Organisation for Animal Health as critically important, highly important, and important together with the animal species to which they are applied, the indications for their use, and whether they are essential or have few alternatives (28).

While the adoption and implementation of AMS in the human medicine field has moved forward and is being developed in the wealthier countries of the world, this is not the case in low- and middle-income countries where insufficient political commitment, scarcity of funding, and a lack of expertise create major difficulties (29). In such countries, this lack of AMS programs is compounded by high levels of inappropriate antimicrobial drug use, particularly in emerging economies where increasing funds are available for their purchase and where dispensing of antimicrobials without prescription and self-medication are common practices. This problem is exacerbated by the sale of poorly formulated and counterfeit antimicrobials (30, 31) and by the increasing availability of antimicrobials through illicit online pharmacies (32). Such inappropriate antimicrobial use drives resistance and can lead to rapid transfer of resistant microorganisms and resistance genes on a global basis.

Emergence of resistance among livestock and companion animals and the risk of its transfer to humans is an issue that is causing increasing concern. This is a true One Health issue because transfer may occur in either direction, a situation that is exemplified by the occurrence of human hospital-associated MRSA infections in animals, particularly in domestic pets (33). The emerging problem of MRSA CC398 colonization of pigs and other farm animals illustrates the complexity of the situation. CC398 has become established with high rates of carriage in pig farms in Denmark and, while typically causing little pathology in the pigs, it has now become the dominant MRSA clone found in humans in Denmark (34) and a cause of infection in those having no direct contact with the pigs (35). Although CC398 is believed to have originated as a human strain of S. aureus which became adapted to pigs (36), it is now evolving to produce distinct animal-adapted and human-adapted strains, as well as strains with increased invasive capacity for both humans and animals (37). This has led to the recommendation that human hospitals with MRSA exclusion policies should screen farmers and veterinarians with livestock contacts for carriage of CC398 prior to admission. Recognition of this problem in Norway has led to a search and destroy policy on pig farms aimed at keeping pig populations MRSA free and preventing them from becoming reservoirs for transmission of CC398 to humans (38), but this approach will be more difficult to establish in countries with larger pig populations. Public health risks posed by enterobacterial species producing extended-spectrum and AmpC β-lactamases in food and food-producing animals have also been highlighted by the European Food Safety Authority, with the identification of cephalosporin use and international trade in animals as risk factors (39).

More recently, the identification of the plasmid-mediated colistin resistance gene, mcr-1, from E. coli from pigs, poultry, and hospital patients in China (40) and the subsequent recognition that it is present worldwide with more frequent isolation from animals, coupled with the much greater use of colistin in livestock, has caused the European Medicines Agency to issue new advice on its use in animals on the basis that this gene has probably arisen in animals and transferred to humans (41, 42). European Union member states are recommended to minimize sales of colistin for use in animals to achieve a 65% reduction, and colistin is to be added to the critical category of medicines reserved for treating clinical conditions for which there are no effective alternative treatments for the respective target species and indication (42), while WHO has added colistin (a polymyxin) to the list of highest-priority critically important antimicrobials (27).

AMS in the human medicine field has been principally focused on reduction of antimicrobial use and more critical choice of appropriate antimicrobial therapy. However, there is increasing attention on the unintended consequences which may occur following the use of antimicrobial drugs (43), and particularly the effects of intestinal dysbiosis (44). The effect of antimicrobial therapy on the occurrence of C. difficile-associated diarrhea is well recognized, but there is now increasing evidence associating antimicrobial exposure with inflammatory bowel disease and childhood obesity (45–47) and with juvenile idiopathic arthritis occurring in children aged 1–15 years (48). Disturbance of the microbial ecology of the gut has also been associated with neurodevelopmental disorders (49). More study is required in these areas, and it seems likely that other diseases associated with disturbance of the microbiome by antimicrobial drugs will emerge.

AMS in Veterinary Medicine

The development of AMS in veterinary medicine has lagged behind that in the human medicine field but has gained impetus with increasing evidence of the worldwide multiple-resistance crisis in human medicine and the possible contributions from antimicrobial use and resistance selection in animals. Recommendations on the establishment of AMS programs and guidelines on prudent antimicrobial use and disease prevention have been published by a number of authors and by international and national organizations, and there is an accumulating literature providing a framework for their implementation in the companion animal and farming sectors.

At the international governmental level, the European Union moved to control the use of antimicrobial drugs in food production at an early stage. Chloramphenicol use for therapeutic purposes was banned in food production animals in the European Community in 1994 because of public health toxicological concerns. Owing to concerns about antimicrobial resistance selection, antimicrobial drugs were progressively withdrawn from use as growth promoters in animal feed in Europe beginning in 1972, with a complete ban in 2006 (50). The WHO adopted a global action plan on antimicrobial resistance in 2015 with the principal goal of ensuring treatment and prevention of infectious diseases with quality-assured, safe, and effective medicines. The plan outlines five strategic objectives (Table 1), all of which relate to the development and implementation of effective AMS (51). These objectives are being implemented in collaboration with the United Nations Food and Agriculture Organization (FAO) and World Organisation for Animal Health in a coordinated “One Health” approach involving a wide range of sectors including human and veterinary medicine, agriculture, finance, environment, and consumers. The aim was to have multisectorial national plans in place in 2017. In February 2016, WHO, the FAO, and the World Organisation for Animal Health jointly released a manual and toolkit for developing national action plans (52), and the FAO released its action plan, which deals with the food and agricultural sectors, including terrestrial and aquatic animal health and production, crop production, food safety standard-setting, and legal aspects, in September 2016 (53). The FAO plan focuses on four components: (i) awareness of antimicrobial resistance and related threats, (ii) surveillance and monitoring of antimicrobial resistance and antimicrobial use, (iii) strengthening governance related to antimicrobial use and antimicrobial resistance, including implementation of international guidelines/standards on antimicrobial resistance such as the Codex Alimentarius (54), and (iv) promoting good practice in food and agricultural systems and prudent antimicrobial use at the country level, including the capacity for implementation of international standards and guidelines relating to antimicrobial resistance and use, and consideration of antimicrobial resistance issues in the development of voluntary guidelines for sustainable agricultural production.

TABLE 1.

Objectives of the WHO Global Action Plan on Antimicrobial Resistance^a

1	To improve awareness and understanding of antimicrobial resistance through effective communication, education, and training
2	To strengthen the knowledge and evidence base through surveillance and research
3	To reduce the incidence of infection through effective sanitation, hygiene, and infection prevention measures
4	To optimize the use of antimicrobial medicines in human and animal health
5	To develop the economic case for sustainable investment that takes account of the needs of all countries and to increase investment in new medicines, diagnostic tools, vaccines, and other interventions

^a Source: reference 51.

The European Union has also been active in working with WHO to develop plans dealing with antimicrobial resistance and covering both human and veterinary medicine and agriculture. WHO Europe published the European strategic action plan on antibiotic resistance in 2011, which aimed to promote coordination and development of European national action plans. These are listed by the European Centre for Disease Prevention and Control (55), which shows 13 countries with programs initiated between 2011 and 2016. Other countries have also published strategies which aim to support development of the WHO One Health global plan including the Public Health Agency of Canada, the Australian Government (Departments of Health and of Agriculture) and the U.S. government (56–58). Although the United States has relied on guidance documents and voluntary action to regulate antimicrobial use in farm animals (59), it is now increasingly restricting the use of antimicrobials in feed and water for food animals (60).

The development of effective AMS programs in both human and veterinary medicine in low-income and medium-income countries continues to be problematic owing to insufficient political commitment and the lack of funding and expertise (29, 61–63), as discussed earlier.

At the institutional level, veterinary AMS and responsible antimicrobial use guidelines have been developed by a wide range of national and international organizations. In the United Kingdom, the Responsible Use of Medicines in Agriculture Alliance (RUMA, http://www.ruma.org.uk/) brings together organizations involved in all stages of the food chain with the objective of promoting food safety, animal health, and animal welfare. Established in 1997, it publishes guidelines on antimicrobial use and vaccination in poultry, pigs, cattle, sheep, and fish while also promoting and providing links to AMS and disease control schemes by producers’ organizations. The European Platform for the Responsible Use of Medicines in Animals (EPRUMA, http://www.epruma.eu/) has similar objectives and broad international membership. Established in 2005, it works at the European Union level facilitating and promoting coordinated and integrated action, including the production of framework documents providing guidance on responsible antimicrobial use in food-producing animals.

While the focus on AMS among international and national governmental organizations has principally been directed at farming and food production, recommendations and guidelines for companion animals have been chiefly developed by veterinary societies and associations. In North America, the American Veterinary Medical Association and its constituent allied organizations, the American College of Veterinary Internal Medicine, and the Canadian Veterinary Medical Association have been particularly active (64–66). Within Europe, national veterinary organizations have also produced guidelines and posters on AMS and responsible use of antimicrobials, and the Federation of European Companion Animal Veterinary Associations, which brings together European national associations, has produced a series of posters for veterinary surgeons and owners.

More specific guidelines for responsible antimicrobial use in particular diseases are also being increasingly made available. The International Society for Companion Animal Infectious Disease (ISCAID) Antimicrobial Working Group is working to produce a range of guidelines and has already published on the treatment of urinary tract infections, respiratory diseases, and superficial bacterial folliculitis (67–69). The ISCAID website (http://www.iscaid.org/guidelines) also lists a range of guidelines produced by other organizations.

In Europe, The Netherlands has been particularly successful in promoting AMS and reducing antimicrobial use. It recorded a 58.4% decrease in sales between 2009 and 2015, and this has been associated with a clear reduction in levels of antimicrobial resistance in broilers, veal calves, and pigs. Its policy has been proposed by the European Union as a model of good practice for other European community states (70). The prudent use policies were set up as a public-private partnership which took responsibility for the institution of effective measures based on expert scientific advice. The partnership involved stakeholders in the major livestock production sectors together with the Royal Netherlands Veterinary Association, facilitated and supervised by the national government, and led to the establishment of the independent Netherlands Veterinary Medicine Authority, which analyzed data on antimicrobial use at the farm level and set benchmarks. Veterinarians that are found to be noncompliant with specified prudent use policies may be subject to a range of sanctions including official warnings, application of administrative or criminal law, and reference to the veterinary disciplinary board, leading to suspension or the application of fines.

DEVELOPING AND IMPLEMENTING GOOD STEWARDSHIP PRACTICE

Good stewardship practice (GSP) describes the active, dynamic, and motivated approach to antimicrobial use reinforced by a mindset for continuous improvement (71). However, despite the many calls for prudent antimicrobial use and the wide range of guidelines that is now available, the benefits of effective implementation of GSP both in human and in animal medicine are only just starting to be recognized. A review of key AMS literature (72–81) that continues to evolve and influence the development and implementation of AMS plans in human medicine has revealed a number of important messages to guide the establishment of a more effective veterinary AMS framework, the core elements of which—goals, key strategies, and success factors—are presented in Table 2.

TABLE 2.

Principal elements in the establishment of an effective veterinary AMS framework

Core elements	Generate enthusiasm, commitment, and support among senior management Identify resistance patterns and antimicrobial use in treated animals Select priority areas and devise intervention plans Determine how progress and success will be measured Implement at least one readily achievable policy or practice of improved antimicrobial use to allow the group to experience early success Implement quantitative and qualitative measurement of prescribing practice, including self-audit, if appropriate Customize the AMS plan to serve the special needs of each practice or other operation Provide educational resources and access to expertise on optimizing antimicrobial prescribing
AMS goals	Ensure that each patient receives the most appropriate treatment: the right drug, at the right time, at the right dose for the right duration by the right route of delivery (5 rights) Eliminate antimicrobial overuse and misuse Minimize the selection, maintenance, and dissemination of antimicrobial resistance
Key strategies	Implement clinical guidelines that take into account local microbiology and susceptibility patterns Establish formulary restriction and approval systems that restrict later generation and critically important antimicrobials to patients where clinical need is justified Review antimicrobial prescribing with intervention and direct feedback to the prescriber Ensure that the clinical microbiology laboratory uses selective reporting of susceptibility results that is consistent with clinical guidelines
Success factors	Presence of a motivated team leader Motivated team sharing responsibility for GSP AMS program tailored to local context Credibility established by demonstrating early success

The 5R Antimicrobial Framework

There is a growing experience with a collaborative and participatory approach to antimicrobial use that sets out to bring about change by recognizing the importance of sociology and using practical knowledge that is useful in local practice (82–84). The participatory approach works particularly well at the farm level, where changes in antimicrobial use can be discussed by the veterinarian and the farmer, and a jointly owned plan can be prepared and implemented (85–89).

Related to the participatory approach is that of handshake stewardship (90, 91), which is defined by the lack of a restriction and preauthorization but includes a collaborative review of all prescribed antimicrobials and an in-person approach to feedback. Handshake stewardship is ironically named, because handshakes are associated with pathogen transmission and may be replaced with alternative greetings (92).

This less intrusive approach to modifying antimicrobial use is likely to be more acceptable to veterinarians who may be similar to family physicians who have been reported to oppose any measure aimed at restricting freedom of prescription (93). Fortunately, several authoritative studies have demonstrated that postprescription review with feedback can have a greater impact on modifying antimicrobial prescribing than restrictive approaches (18, 94).

The value of feedback in improving antimicrobial use is the subject of many studies (18, 95–98). Dunn and Dunn (99) described the most striking effects of clinical audit and feedback in small animal practice: “with a modicum of work, the standards within practice have been improved, and a plan to improve them further demonstrates that a simple audit process works.”

From this background, a veterinary framework of AMS has evolved (100, 101), with each core element mapping to the major features of currently described AMS programs. The framework is customized to meet the varying requirements across the veterinary profession, which encompasses everything from single animal treatment to herd and flock health management. The framework is summarized in Fig. 1, which describes GSP and the application of five core elements (the 5Rs: responsibility, reduce, replace, refine, review). Responsibility is the fundamentally important and essential starting point; AMS programs will not succeed without senior management commitment. Under the aegis of corporate support, GSP requires that all uses of antimicrobials are examined under the multiple lenses of the potential to reduce, replace, or refine each use. The final element, review, involves the measurement and assessment of use, antimicrobial resistance, and resources needed, which may often include continuing professional development. The process is self-motivating and a continuous form of improvement.

FIGURE 1.

Key elements of antimicrobial stewardship. 1. GSP requires embedded thinking and action to improve antimicrobial use and minimize resistance selection and impact. 2. Responsibility implies high-level commitment, with everybody taking and sharing responsibility 3. The 3Rs of responsible use—reduce, refine, and replace—should be applied wherever possible. 4. Review antimicrobial use and infection control and develop objectives to improve current practice and implementation of the stewardship plan. 5. Every cycle of 5R stewardship reflects continuous improvement (kaizen) and leads to best practices in infection prevention and control and antimicrobial use. (Figure graphics by Ed Hewson.)

Two important features distinguish this model from that of current human AMS programs. First, the GSP 5R model includes consideration of both improved antimicrobial use and infection prevention and control, which necessarily work in tandem and cannot be naturally separated. Second, a fundamental objective of the framework is to consider ways not to use antimicrobials. It is only when antimicrobials are necessary that optimal use is considered.

Responsibility

It is widely acknowledged that a fundamental requirement for a successful and sustained AMS program is to have executive or senior management support (102–105). This element of the 5R framework maps to the first core element of AMS programs (106), outpatient AMS (81) and hospital AMS (107), described as ‘leadership commitment” and requiring dedication to and accountability for optimizing antimicrobial prescribing and patient safety. There are many enabling mechanisms to ensure that a collaborative and participatory team approach is taken with effective communication with all stakeholders. A particularly powerful positive influence is associated with skilled leadership and social cohesion (108), which has been observed to transform even conservative clinicians into early adopters.

Reduction

A common theme of veterinary science for more than 100 years that remains pivotal to the reduction of antimicrobial resistance is the implementation of improved infection prevention and control measures (109, 110), a process described by the concept of biosecurity, which includes the set of preventive measures designed to reduce the risk of introduction, development, and spread of infectious disease within an animal population (111). An excellent review of how biosecurity can reduce antimicrobial use and antimicrobial resistance selection has been published (112), and pivotal features and examples of biosecurity actions are summarized in Table 3.

TABLE 3.

Biosecurity and disease prevention

Biosecurity phase	Examples of effective actions
Primary prevention: external biosecurity (bioexclusion)	Minimize the introduction of animals Minimize the number of sources of introduced animals Clean and disinfect transport vehicles and containers Isolate sick animals before introduction Provide clean water, feed, air Exclude pests from and control human access to housing, filter exhaust to reduce pathogen load
Secondary prevention: internal biosecurity (biocontainment)	All-in-all-out production system Hygiene, infection control protocols Housing design: ventilation, drainage Litter/bedding materials Early diagnosis of disease Once pathogen is present, introduce measures to eliminate or reduce transmission—guided by on-farm microbiological risk assessment Reduce stocking density, segregation, sick pens
Tertiary prevention: individual animal resilience (adaptive capacity to changing environment)	Genetic selection Vaccination Management (handling, low stress, enrichment) Nutrition Housing (ventilation, temperature, stocking rate, hygiene)

Modern precision farming could not take place in the absence of high levels of biosecurity, and it underpins production practices in poultry (113, 114), dairy cattle (115), feedlot cattle (116), pigs (117), and aquaculture (118). A quantitative tool to measure biosecurity on broiler farms has been developed (119), making self-assessment much easier. There can be competing interests that have to be balanced. The increasing growth of outdoor livestock production systems can be incompatible with the maintenance of biosecurity, as well illustrated by the introduction of a requirement to enhance biosecurity in United Kingdom bird enterprises to help reduce the risk of avian influenza introduction (120).

Biosecurity is equally important in small animal (121) and equine practices and hospitals (122), and practice-specific infection control plans have been widely advocated (123), with model plans available for adaptation (124–126).

GSP requires consideration of the entire spectrum of possible reduction approaches, which also include genetic selection for disease resistance (127–132), use of vaccines (which have repeatedly been shown to reduce antimicrobial use in a variety of food animal species, including fish [133], calves [134], pigs [135–138], and poultry [139]), identifying modifiable risk factors (140–143), and of course, measuring current practice. The importance of this has been demonstrated by Elbers et al. (144, 145), who identified otherwise inapparent opportunities to reduce antimicrobial use, and by Greko (146), who demonstrated that feedback on personal or practice antimicrobial use can lead to reductions.

Replacement

Replacement of the use of antimicrobials with alternative, nonantimicrobial measures, wherever possible and appropriate, is another critical AMS tenet. Key considerations include whether or not the alternative approach will select for antimicrobial resistance and the quality and strength of the evidence supporting the use of the selected approach. The most comprehensive recent review of the use of alternative products in livestock (112) concluded that “due to limitation in data availability, the potential impact of the alternative measures on the occurrence of antimicrobial resistance in bacteria from food-producing animals and food cannot be established.” High-quality research is needed to overcome this critical information gap.

While the impact on antimicrobial resistance may not be clear, the subject is attracting substantial interest (147–150); several alternatives have been demonstrated to have positive impacts on animal health in particular circumstances, and their use is increasing in livestock production. These alternatives include the use of dietary acidifiers or organic acids in broilers (151) and pigs (152); the use of probiotic yeasts (153–155) and probiotic bacteria (156–158) in fish, monogastric, and ruminant species; and the use of prebiotics (nondigestible food ingredients that beneficially affect the host by selectively stimulating the growth and/or activity of one or a limited number of bacteria in the colon) and their combination with probiotics (known as synbiotics) (159–162) in livestock and companion animal species.

The literature is also providing support for a number of other options that include antimicrobial peptides (163, 164), bacteriophages (165–171), essential oils (172, 173), honey (174), nitric oxide (175), predatory bacteria (176, 177), and immunoglobulins, both IgY from hyperimmune chicken egg yolk (178–180) and spray dried immune plasma (181, 182).

Two alternative products that replace antimicrobials have already received regulatory approval in major markets worldwide and are having a significant impact in replacing antimicrobial use in dairy cattle, where mastitis is a major indication for antimicrobial treatment. Teat-sealing pastes containing an inert heavy metal, such a bismuth subnitrate, are available for use in suitably selected dairy cattle at the end of each lactation, where they provide a physical barrier that prevents new infections from ascending the teat canal during the dry period (183). More recently, pegbovigrastim (184), a modified form of the naturally occurring immunoregulatory cytokine bovine granulocyte colony-stimulating factor, which restores normal neutrophil function to cattle during the periparturient period, thereby reducing susceptibility to clinical mastitis infections, has become available and also has the potential to replace the need for later antimicrobial use.

Although the movement away from the use of antimicrobial growth promoters has been a strong driver for research and development of antimicrobial replacement approaches in livestock, there is also keen interest in companion animal and equine practice to identify and use replacements (185). However, experience with probiotic use in horses suggests that the beneficial effects seen in livestock species cannot always be expected in foals and horses (186, 187) and highlights the need for careful selection and testing of probiotic strains.

Where infections can be reached with topical therapy, this can be a very effective way of replacing systemic antimicrobial drugs (68, 188–190). Topical antimicrobials are commonly active in the face of resistance to systemically administered agents because they can be used at much higher concentrations. It should be noted that antimicrobial susceptibility tests are normally based on the expectation of systemic therapy and their results cannot therefore be applied to the assessment of resistance when topical therapy is being considered.

The potential benefits of antimicrobial replacements are increasingly being studied and reported, but it cannot be assumed that their use is innocuous. A growing list of unintended consequences, some extremely serious, is being reported. Examples relating to animal feed supplementation in pigs include apparent coselection of MRSA, and of tetracycline and sulfonamide resistance in Gram-negative bacteria, by high concentrations of dietary ZnO (191–193) and apparent coselection of macrolide and glycopeptide resistance by diets with high concentrations of copper (194, 195). In cattle, selection of tetracycline-resistant E. coli by dietary menthol has been reported (196). Clays routinely added to animal feed to improve growth and animal product quality appear to facilitate horizontal transfer of resistance determinants in the digestive tracts of farm animals (197). Worryingly, in vitro serial passage studies of MRSA and host-derived antimicrobial peptides have demonstrated evolution of stable virulent mutants with cross-resistance to human innate immunity as well as antimicrobial therapy (198).

Refinement

Opportunities for refinement of antimicrobial use are identified in Fig. 2, which summarizes the steps in initiating and implementing an antimicrobial therapeutic plan. The process begins with diagnosis, often the most tenuous link in the chain as identified by O’Neill (199), who further noted that uncertainty of diagnosis of bacterial infection is a major driver of antimicrobial overuse and antimicrobial resistance selection in humans. It is likely that in many situations this is the same in veterinary practice. For example, it is notoriously difficult to establish an accurate diagnosis of bovine respiratory disease in feedlot cattle, especially early in the pathogenesis (200). These issues have been reviewed by Griffin (201), who concludes that a major factor that is seldom considered in the treatment response in bovine respiratory disease is correctness of the clinical diagnosis; misdiagnosis as a cause of treatment failure is a common necropsy finding.

FIGURE 2.

Decision-making and application of GSP in suspected bacterial infection.

The pivotal role that diagnosis plays in ensuring appropriate antimicrobial use has led to a massive investment in research to find better diagnostic tests. Substantial rewards are available to those that can meet the diagnostic needs. The European Commission (202) awarded the €1 million Horizon Prize to the innovators of a breakthrough test which distinguishes between viral and bacterial infections in humans. The United Kingdom Longitude Prize (203) is a challenge with a £10 million prize fund to reward a diagnostic test that helps solve the problem of global antimicrobial resistance by identifying when antimicrobials are needed and, if they are, which ones to use. Criteria that the test must satisfy are that it is needed, accurate, affordable, rapid, easy to use, scalable, safe, connected, and available to anyone, anywhere in the world. These criteria are very similar to those enunciated by WHO as the ASSURED (affordable, sensitive, specific, user-friendly, rapid and robust, equipment-free, and deliverable) criteria (204) for novel diagnostic tests suitable for global use.

While it is expected that any technological breakthroughs in diagnostic tests for medical use will find ready application in the animal health world, in the meantime there has been substantial research into improved methods of disease diagnosis in livestock, driven by the desire to detect infections early in individual animals to allow more targeted treatment with more rapid responses. Examples of new approaches that are actively being investigated include remote automatic sensing of animal behavior, including feeding and drinking, in chickens (205–207), pigs (208), calves (209), and cattle (210–212), with evidence of the possibility of earlier detection of clinical illness from 1 day to 14 days in advance of observed disease. Remote detection of cough sounds in pigs (208, 213) and cattle (214) also presents opportunities for earlier diagnosis and intervention. A variety of other promising approaches have also been reported, including computer-aided lung auscultation in cattle for bovine respiratory disease diagnosis (215), infrared thermography of cattle (216) and pigs (217), and a diversity of acute phase protein (218–220), bacteriological (221, 222), genomic (223–226), proteomic (227), and immunological (228, 229) tests.

Already, improved culture-based diagnostic tests are allowing selective treatment of dairy cattle with purulent vaginal discharge (222) or clinical mastitis (230–232), and improved use of somatic cell count data is guiding selective dry cow treatment of dairy cattle (233, 234), each decreasing antimicrobial use.

Depending on the strength of the diagnosis, prophylactic (risk factors for infection present), empiric (bacterial infection suspected), or directed (bacterial infection and site of infection known) treatment can be planned. It is then important to be guided by local treatment guidelines incorporating the judicious use principles set out in Table 4, which have been derived from an analysis of existing guidelines and present a taxonomy divided into stage of treatment.

TABLE 4.

Core principles of judicious use of antimicrobial agents

Category	Principles
Pretreatment principles	Disease prevention Apply appropriate biosecurity, husbandry, hygiene, health monitoring, vaccination, nutrition, housing, and environmental controls. Use codes of practice, quality assurance programs, flock or herd health surveillance programs, and education programs that promote responsible and prudent use of antimicrobial agents.
	Professional intervention Ensure that use (labelled and extra-label) of antimicrobials meets all the requirements of a valid veterinarian-client-patient relationship.
	Alternatives to antimicrobial agents Efficacious, scientific, evidence-based alternatives to antimicrobial agents can be an important adjunct to good husbandry practices.
Diagnosis	Accurate diagnosis Make clinical diagnosis of bacterial infection with appropriate point of care and laboratory tests and epidemiological information.
Therapeutic objective and plan	Develop outcome objectives (for example, clinical or microbiological cure) and implementation plan (including consideration of therapeutic choices, supportive therapy, host, environment, infectious agent, and other factors).
Drug selection	Justification of antimicrobial use Consider other options first; antimicrobials should not be used to compensate for or mask poor farm or veterinary practices. Use informed professional judgment balancing the risks (especially the risk of antimicrobial resistance selection and dissemination) and benefits to humans, animals, and the environment.
	Guidelines for antimicrobial use Consult disease- and species-specific guidelines to inform antimicrobial selection and use.
	Critically important antimicrobial agents Use all antimicrobial agents, including those considered important in treating refractory infections in human or veterinary medicine, only after careful review and reasonable justification.
	Culture and sensitivity testing Utilize culture and susceptibility (or equivalent) testing when clinically relevant to aid selection of antimicrobials, especially if initial treatment has failed.
	Spectrum of activity Use narrow-spectrum in preference to broad-spectrum antimicrobials whenever appropriate.
	Extra-label (off-label) antimicrobial therapy Must be prescribed only in accordance with prevailing laws and regulations. Confine use to situations where medications used according to label instructions have been ineffective or are unavailable and where there is scientific evidence, including residue data if appropriate, supporting the off-label use pattern.
Drug use	Dosage regimens Where possible, optimize regimens for therapeutic antimicrobial use following current pharmacokinetic and pharmacodynamic guidance.
	Duration of treatment Minimize therapeutic exposure to antimicrobials by treating for only as long as needed to meet the therapeutic objective.
	Labelling and instructions Ensure that the veterinarian gives the end user written instructions on drug use, with clear details for method of administration, dose rate, frequency and duration of treatment, precautions, and withholding period.
	Target animals Limit therapeutic antimicrobial treatment to ill or at-risk animals, treating the fewest animals possible.
	Record keeping Keep accurate records of diagnosis (indication), treatment, and outcome to allow therapeutic regimens to be evaluated by the prescriber and permit benchmarking as a guide for continuous improvement.
	Compliance Encourage and ensure that instructions for drug use are implemented appropriately
	Monitor response to treatment Report to appropriate authorities any reasonable suspicion of an adverse reaction to the medicine in either treated animals or farm staff having contact with the medicine, including any unexpected failure to respond to the medication. Thoroughly investigate every treated case that fails to respond as expected.
Posttreatment activities	Environmental contamination Minimize environmental contamination with antimicrobials whenever possible.
	Surveillance of antimicrobial resistance Undertake susceptibility surveillance periodically and provide the results to the prescriber, supervising veterinarians, and other relevant parties.
	Continuous evaluation Evaluate veterinarians’ prescribing practices continually, based on such information as the main indications and types of antimicrobials used in different animal species and their relation to available data on antimicrobial resistance and current use guidelines.

While the therapeutic objective of local antimicrobial prescribing guidelines is frequently stated as attaining optimal use where effectiveness is maximized and adverse effects, including antimicrobial resistance, are minimized, the recommended dosage regimens of existing antimicrobial products have been developed to demonstrate efficacy, not to minimize resistance.

A dosage regimen for a particular route of systemic administration consists of a dose (mg/kg), a dosing interval (either continuous via feed, water or infusion, or one or more administrations each day) and a duration (days). These are the only parameters that can be changed, though it is a reasonable expectation that in the majority of cases it is likely that the label dose recommendation will lead to satisfactory therapeutic efficacy. However, it seems unlikely that a single dosage regimen could apply to all situations all the time (especially when pathogen MICs are forever increasing), and even optimal use of systemic drugs will delay but not prevent the spread of resistance (235). The father of antimicrobial chemotherapy, Nobel laureate Paul Ehrlich, famously advocated antimicrobial use regimens characterized by “frapper fort et frapper vite”—hit hard and hit quickly (236). Interestingly, this approach is again finding favor as it is realized that appropriate doses early in the course of a bacterial infection have greater efficacy and lower propensity to select for antimicrobial resistance (237, 238). In addition, the possibility of using topical or anatomically targeted therapy should always be considered. This is likely to be applicable most often for surface wounds and infections involving mucosae and skin and for accessible sites of infection such as the mammary gland, uterus, and eye. Topical and targeted therapy has the advantage that the antimicrobial agents can be administered at much higher concentrations than those used systemically. This can overcome existing resistance to systemic drugs and is less likely to facilitate development of resistance among susceptible organisms. It is being increasingly used and recommended in small-animal dermatology, particularly in the face of methicillin-resistant staphylococcal infection (68, 239).

Review: the basis of continuous improvement

Review of the AMS program is a fundamental core principle. At the outset, stock-taking or audit of current antimicrobial use practice, infection prevention and control measures, and the antimicrobial resistance status defines the starting point and allows an examination of those areas where improvements can inform AMS objectives and lead to early gains (99). Review remains an essential element of AMS as each cycle of continuous improvement is achieved. Review includes the measurement of progress toward each objective. Information on the use of antimicrobials can be obtained from both quantitative and qualitative assessments. There are many indices describing the quantity of antimicrobial agents used (240), each related to the objective, whether to assess trends in use over time, to compare use in different populations of animals (for example, in different countries), to benchmark, or to assess the relationship of antimicrobial use and AMS. Unfortunately, there is no consensus on the most appropriate index. However, all indices rely on limited data (see Fig. 3), which include species of animal treated, number of animals, bodyweight, daily dose, and duration of use. If this information is captured, then any of the many indices of quantitative use can be derived. When quantities of antimicrobial use are reduced, there is much to learn from an assessment of the quality of antimicrobial use.

FIGURE 3.

Interventions that can guide enhanced AMS.

While it is generally not difficult to measure the quantity of antimicrobials used, the ability to measure the quality of use can be challenging. However, it can be argued that achieving a high level of quality use of antimicrobials is an important AMS goal.

One approach that facilitates measuring quality of use is the development of antimicrobial prescribing survey tools (241), an example of which is provided in Table 5. The basic principle of assessment of quality is whether or not the prescription or use of antimicrobials was compliant with the most appropriate local or national antimicrobial use guideline. To be able to assess quality of use, it is necessary that the indication or reason for antimicrobial use is recorded, and there must be an antimicrobial use guideline available as the reference standard for quality determination. There is a growing literature focused on how best to determine quality of use (242–245) which cautions against the dogmatic use of guidelines because clinical factors may mean that deviations are warranted (246) and emphasizes the need to validate quality indicators based on quality of use (247).

TABLE 5.

Assessment of quality of use

Appropriate
1	Optimal	Antimicrobial prescription optimally follows a national or endorsed local guideline, including indication, antimicrobial choice, dosage, route, and duration (including for surgical prophylaxis).
2	Adequate	Antimicrobial prescription does not optimally follow the national or endorsed local guideline, including antimicrobial choice, dosage, route, or duration, but is a reasonable alternative choice for the likely causative or cultured pathogens; or for surgical prophylaxis, as above and duration is less than 24 hours.
Inappropriate
3	Suboptimal	Antimicrobial prescription including antimicrobial choice, dosage, route, and duration, is an unreasonable choice for the likely causative or cultured pathogens, including spectrum excessively broad or an unnecessary overlap in spectrum of activity; and/or failure to appropriately de-escalate when microbiological results are available.
4	Inadequate	Antimicrobial prescription including indication, antimicrobial choice, dosage, route, or duration is unlikely to treat the likely causative or cultured pathogens; or an antimicrobial is not indicated for the documented or presumed indication; or there may be the potential risk of toxicity due to drug interaction; or for surgical prophylaxis, the duration is greater than 24 hours (except where guidelines endorse this).
Unknown
5	Not assessable	The indication is not documented and cannot be determined from the clinical case notes; or the case notes are not comprehensive enough to assess appropriateness; or the patient is too complex, due to multiple comorbidities, microbiology results, etc.

Review also includes the assessment of educational needs, review of the literature and other sources of information on antimicrobial resistance and AMS, use of benchmarking and infection prevention and control measures, seeking external assessments to broaden the basis of decision making, determination of risk factors for infectious disease to identify risk management interventions, and setting of new objectives for the next cycle of improvement.

A summary of interventions that can guide enhanced AMS is presented in Fig. 3, which identifies the steps from diagnosis to clinical outcomes, highlighting opportunities for improvement.

MAINTAINING GSP

Maintaining and extending GSP implies continual revision and improvement. This needs to take into account not only development of new therapeutic agents and methods of treating and preventing infections but also increasing demand for antimicrobials created by the extension of infectious disease and changes in the livestock industry as global population growth drives increasing intensification. Allied to this are the threats posed by the increasing availability of counterfeit products and substandard generic drugs, rising demand for antimicrobials for human medicine and both livestock and companion animal medicine in emerging countries with increasingly wealthy populations, and the lack of effective legislation and enforcement of regulations controlling the prescription of antimicrobial drugs in much of the world. These threats will exacerbate the problem of inappropriate antimicrobial use and continue to drive the development of antimicrobial resistance. Global travel, migration, and the international market in animals and animal products will transport resistant organisms and continue to threaten stewardship even in countries with effective antimicrobial usage policies.

It is very unlikely that the situation will be resolved by the development of new and powerful antimicrobial agents. However, increased biosecurity and a focus on the development of disease prevention and control methods are likely to be highly effective and will need to be increasingly prominent components of GSP in the future.