Controlling nematode infections in sheep: application of HACCP

Gastrointestinal nematode infection in sheep is a significant challenge for production with implications for animal health and welfare, and lamb growth. Flocks rely on the chemical control of worm burdens, but with the increasing challenge of anthelmintic resistance there is a necessity for farmers to uptake SCOPs (Sustainable Control of Parasites) guidance. Vets need to communicate practical strategies to farmers and suggest novel approaches that address these challenges so that holistic and sustainable parasite management can be achieved. This article examines the challenge of nematode infections and anthelmintic resistance and suggests how hazard analysis and critical control point (HACCP) application may facilitate a whole-flock strategy.

every 21 days. Although the performance of Group 4 was better than Group 3, the lambs did not achieve the DLWG performance of the ALC, despite anthelmintic treatment, and their performance was similar to that of Group 2. This highlights the importance of avoiding exposing sheep to high levels of infection, and not just relying on regular anthelmintic treatment in situ.
Although infected lambs may still achieve 110 g/day growth (Kenyon and others 2009) this is below target growth rates of 250 g/day (AHDB 2014) and results in delays to finishing time (Miller and others 2012), delayed onset of puberty in breeding ewe lambs and, in some cases, permanent stunting. Furthermore, PGE can result in a 10.4 per cent reduction in carcase value (Miller and others 2012). Ineffective treatment of PGE has been demonstrated to result in 60 to 100 per cent drops in growth relative to effectively treated animals (Abbott and others 1986).
Historically, anthelmintics have played a central role in the management of PGE (Miller and others 2012). However, effective control now requires a more thoughtful approach in the face of multiple challenges. First and foremost, dependency on anthelmintic products and their misuse has led to the development of anthelmintic resistance. It

Farm Animals
is hoped that the headline figures showing extremely high prevalence of anthelmintic resistance given in Box 1 will drive home the message that this is not simply something that happens to 'other farmers'; anthelmintic resistance is widespread and, despite the fact that it may not be causing visible reductions in productivity on most farms, sustainable use of anthelmintics is required to preserve their efficacy and prevent deterioration to the point where performance and welfare are compromised.
Second, there are broader environmental implications of PGE and anthelmintic resistance beyond the farm, which are becoming increasingly important to both producers and consumers as environmental awareness increases. Sheep production is responsible for an estimated 52 per cent of all agricultural emissions in Scotland, and PGE was identified as one of the 'top three' diseases with a significant impact on greenhouse gas emissions in the livestock sector, based on its impact on growth and feed conversion rates (Bartley and others 2016). In addition, the impact of anthelmintics on dung fauna and knock-on effects at higher trophic levels is well documented (Floate and others 2005). Repeated ineffective treatments, either due to user error or exacerbated by anthelmintic resistance, have the potential to contribute to increased environmental impacts associated with PGE control.

HACCP: risk-based PGE management planning
Overall, these challenges highlight the importance of growth-rate performance as a key driver of environmentally efficient production, and the necessity for PGE management solutions to be practical and accessible by commercial sheep producers. Risk assessment and management planning (RAMP) approaches to disease control are commonly used to identify disease challenge and risks retrospectively, and implement corrective action (McAloon and others 2015). However, RAMP is largely based on recollection by farmers of the risks to their production process, is often confined to a single temporal point of assessment (ie, undertaking an annual review depends on good record-keeping of mortality and growth rates) and is typically applied after disease becomes an issue. Therefore, the conditions are often not in place for effective application of RAMP to sustainably Box 1: Anthelmintic resistance in sheep in the UK In a study by Morgan and others (2012), 10 per cent of farmers surveyed felt there was evidence of failure of anthelmintic efficacy in their flocks, but this is likely to be a gross underestimate. Resistance is currently defined as a post-treatment faecal egg count (FEC) reduction of less than 95 per cent (see Coles and others 1992 for protocol), but clinically obvious failure (ie, poor resolution of clinical signs after treatment) will only become apparent when treatment efficacy is less than 85 per cent (Abbott and others 2012), by which point anthelmintic performance is severely compromised, with little chance of preserving the useful efficacy of that drug class. The true prevalence of anthelmintic resistance in the UK is unknown, but some regional surveys are available (Table 1).
Worryingly, multiple drug resistance was detected on the majority of farms surveyed in Devon (Glover and others 2017) and Wales (Thomas and others 2015). Overall, 19 per cent of Welsh flocks surveyed had evidence of resistance to 1-BZ, 2-LV and both ivermectin and moxidectin within the 3-ML group. Ivermectin and moxidectin do overlap in mechanism of action but resistance mechanisms are not entirely homogenous between the two, hence resistance to each is considered separately (Prichard and others 2012). There have been documented cases of monepantel resistance in New Zealand (Scott and others 2013)   manage anthelmintic resistance. Hazard analysis and critical control points (HACCP) is an alternative risk-based and process-driven approach which involves proactive monitoring with defined end points (Box 2).

Using HACCP in practice
HACCP was developed for use in food safety but has been adapted for primary production such as to monitor for bovine Johne's disease (McAloon and others 2015), to control bovine mastitis (Noordhuizen and Frankena 1999) or bovine lameness (Bell and others 2009), for management of trematodes in aquaculture (Clausen and others 2015), to control tapeworms in sheep (Gascoigne and Crilly 2014) and to monitor ovine abortion (Crilly and Gascoigne 2016). In this article, we apply the HACCP model to the process of lamb production with the end goal of minimising losses in DLWG through improved control of PGE (Box 2, Fig 3). The hazard in this case is 'exposure to worms' at a level sufficient to cause slowing of DLWG.
Developing a PGE management plan or discussing improved PGE management with a client using the HACCP model is a stepwise process which aids understanding of the production process, points to where the risk of exposure to infection can be manipulated (critical control points [CCPs]), and where sustainable corrective actions at these CCPs -using the GAME strategy described below -can be implemented (Box 2). Sustainable Control of Parasites in Sheep (SCOPS) guidelines (Abbott and others 2012) and good farming practices should be the cornerstone of corrective actions at CCPs. When prevention cannot be achieved, corrective measures will need to be introduced and this may necessitate revising flock health plans. The GAME approach is based on an adaptation of the four founding SCOPS principles of PGE management: ■ ■ Making sure that any treatment is fully effective; ■ ■ Reducing reliance on anthelmintics using management and monitoring; ■ ■ Quarantine and minimising selection for resistance when treating; ■ ■ Best practice when administering treatment.
Data and record keeping on farm (ie, keeping records in a medicines book for meat withdrawal purposes and sheep movements) is a prerequisite. However, vets should think outside the box for performance data; for example, look at the finishing profile of lambs -what is the average number of days to finish? This may help estimate overall DLWGs retrospectively and assess overall efficacy of the HACCP process.

Corrective actions: GAME
Having identified practical CCPs, there are four key areas which could be addressed to minimise the impact of disease. These four areas, if sequentially considered, place administering anthelmintics as the final option for management, after making maximum use of the alternatives.

G -General health and genetics
Poor general health and nutrition of the flock will undermine immune status, increasing susceptibility to disease. Trace element status, nutritional status and endemic diseases will contribute towards increased worm egg shedding (Coop and Holmes 1996, Wallace and others 1996, Sargison and others 2007. Dietary supplementation with metabolisable protein before lambing has been shown to dampen periparturient worm egg shedding in ewes (Houdijk and others 2000). The SAC (2010) reported that providing 100 g digestible undegradable protein (DUP) per lamb in late pregnancy significantly reduced worm egg shedding in the periparturient period and enhanced ongoing lactational performance; that is, enhancing lamb DLWG. The use of bioactive forages to reduce worm egg counts in ewes has been explored with mixed results (Hoste and others 2006 (Shaw and others 2012). Over time, selecting for resistance could reduce the susceptibility of the flock to PGE while reducing pasture contamination and ongoing exposure to infection, thus reducing the need to use an anthelmintic to maintain growth rates.
The role of vaccination of sheep against GIN is under development, with a commercially available vaccine for Haemonchus contortus available in Australia and South Africa (Barbervax; Wormvax Australia) and published trial work supporting the efficacy of subunit vaccines to reduce ewe shedding of Teladorsagia circumcincta eggs at the time of lambing (Nisbet and others 2013). However, it is unlikely that these vaccines will achieve a greater than 90 per cent reduction in egg counts and, as a consequence, whole-flock strategies are still necessary (Matthews and others 2016).

A -Avoidance
Understanding the epidemiology of infection and development of immunity is essential for avoidancebased strategies on farm, which rely on grazing high-risk animals on the lowest risk grazing and forward planning to enable ongoing avoidance of such pastures. Strongyle eggs can survive on pasture for more than 12 months (Rose 1961). Pre-existing worm burdens should be considered when assigning risk to pastures. Knowing where previous high-risk shedding classes of stock have been grazed enables risk to be assigned to pastures (Fig 4).
Recording where and when strongyle burdens are greatest on farm by monitoring movements enables avoidance of infection (Box 3). Grazing vulnerable lambs on high-risk pasture will undermine productivity and growth, necessitating high-drench use for production to continue. FECs and weather-based disease alerts can also feed into pasture-mapping and risk monitoring (Fig 5, Box 3). For example, pastures assessed as high risk based on contamination by growing lambs shedding Nematodirus eggs (identified in FECs) the previous year can be avoided when Nematodirus risk is predicted to be highest and lambs are most vulnerable to infection.
Avoidance strategies also extend to buying in anthelminticresistant GINs with new stock. Morgan and others (2012) found that 35 per cent of flocks surveyed were buying in sheep, but only 17 per cent were quarantined as a standard procedure, and in many cases this was done ineffectively. Treatment with a new derivative and a sheep scab control is 'gold standard' quarantine practice (SCOPS 2016). It is imperative that all advisers are consistent in delivery of this advice. Worryingly, total new derivative use in 2015/2016 accounted for 1 per cent of total drench use (SCOPS 2017), a fraction of the likely number of sheep moving premises in the same period.

M -Monitoring
Having planned a strategy based on having high-risk lambs on the lowest risk category of grazing available (Fig 4, Box 3), ongoing monitoring and review of avoidance strategies is essential to evaluate efficacy, ongoing performance and to establish if treatment is necessary.
Monitoring DLWG (Fig 6) in lambs, although not very specific for parasitic disease, is by far the most sensitive indicator of PGE-related impact. If performed regularly it will identify early drops in performance before significant damage to the gut lining and irreparable stunting has occurred. The disadvantages to such a method include the potential high labour inputs, especially if laboursaving technology is not used for weighing and data recording. Regular monitoring will identify drops in group performance as well as in individuals, and generate opportunities for targeted selective treatment protocols, further reducing overall drench usage (Kenyon and others 2009, Stear and others 2009, Busin and others 2014. A review of key phases of growth is described by Gascoigne and Lovatt (2015).
When used alongside other diagnostic tools, such as FECs (Fig 7), the specificity of DLWG monitoring will be improved, which may highlight other challenges for performance Pastures can be referred to as high-risk, medium-risk or lowrisk pastures and practical application could be assigning risks to ordnance surveys maps of the farm at flock planning meetings (Fig A) .
The pasture risk assessment approach can be extended by incorporating faecal egg counts (FECs) to assess the magnitude of contamination, and therefore risk, on each pasture. These risk assessments also evaluate the scale of the on-pasture refugia population which should highlight and emphasise the importance and risks associated with dose and move (Abbott and others 2012) and be used to adjust treatment schedules where appropriate. For example, leaving a proportion of the flock untreated is especially important if they are to be moved to low-risk pasture to ensure a population of worms in refugia within the sheep are available to mitigate the lack of significant on-pasture refugia. Furthermore, mathematical models predicting gastrointestinal nematode population dynamics and risk of infection, such as the Nematodirus forecast (Fig 5) (Gethings and others 2015, www.scops.org.uk) and Haemonchus contortus/Teladorsagia circumcincta models (Rose and others 2015) can be integrated with the pasture-mapping exercise and regularly reviewed in conjunction with FECs and grazing history to inform pasture-level risk assessments. and correct interpretation requires engagement with a veterinarian or qualified adviser. FECs will vary across groups and pastures, and care should be taken when applying single FEC counts across an age category or a whole farming enterprise. Also, it is important to note that there will be a delay between infection, gut inflammation Farm Animals and egg production, and prepatent disease costs will be missed by FECs (eg, in nematodirosis).
Monitoring can incorporate clinical signs (eg, dag scoring [Fig 1]), evidence of illthrift, and visible drops in body score in lambs. We suggest that it is undesirable to rely on clinical monitoring alone because of high levels of subjectivity and substantial drops in DLWG before such signs are recognised. Relying solely on clinical indicators is likely to be highly inefficient with both economic and environmental costs accumulating in the interim, leading to increased days to finishing and, in instances of severe disease burdens, mortality (Sargison 2009). Evidence of clinical disease indicates failure of the checks and balances already in place.

E -Effective and efficacious treatment
Once PGE has been confirmed in a group, despite avoidance, monitoring and attention to general health, targeted treatment is warranted. Treatments should adhere to the SCOPS principles (Abbott and others 2012) and farmers should be made aware of the 5 Rs for effective treatment: Choosing the right product includes using a product that is effective (established through regular testing for anthelmintic resistance) and suitable for the parasitic burden of the flock. Treating the right animal can include, for example, avoiding unnecessary treatment of ewes and, by extension, targeted selective treatment of lambs based on DLWG and/or FECs. Treating at the right time includes monitoring to identify the optimal timing for treatment. Administering the right dose rate includes dosing to the correct weight of the individual (or heaviest in the group) and calibrating equipment. Administering treatment in the right way includes the correct drenching technique or injection site.
Drench use can be reduced further by using targeted selective treatments; that is, treating those individuals with evidence of disease such as drops in DLWG, without having a negative effect on overall performance (Kenyon and others 2009, Busin and others 2014, Learmount and others 2015. This is a significant change in rhetoric for many farmers and vets, deviating as it does from the need for whole-group sanitary measures against many other diseases, and may be impractical for some flocks to achieve without accurate record keeping. Given the high prevalence of anthelmintic resistance (Box 1), the anthelmintic resistance status of a farm should be checked at intervals to ensure that only efficacious anthelmintic classes of drugs are used. This can be assessed using pre-and post-treatment FECs and checking for a greater than 95 per cent reduction in eggs, or by checking post-drench samples for the presence of fewer than 50 eggs per g. However, the limitation of the latter method is the inability to assess percentage reduction and reliance on a high pretreatment FEC. Nevertheless, routine post-drench checks should be encouraged, even if the pretreatment FEC is unknown. Readers also need to be aware that resistance profiles may vary between strongyle species on farm and therefore within season due to the seasonal succession of nematode species within the hosts (Thomas and others 2015) and even between areas on a farm. Speciation and repetition will enable practitioners to decipher if variation occurs between different times of year, and equally if drenches retain efficacy depending on species. This understanding should be built up over time, by integrating drench checks into routine flock health monitoring.
Equally, the new derivatives monepantel (Zolvix oral solution for sheep; Elanco Animal Health) and deraquantel/ abamectin (Startect dual active oral solution for sheep; Zoetis Animal Health) have a role to play as mid-to lateseason knock-out/break-dose drenches. Every time any anthelmintic is used on an animal, the gut strongyle population is skewed so that resistant worms survive to contaminate pasture. Over the season and as multiple drenches are used there will be a selection pressure for and accumulation of the resistant population. Mid-to late-season treatment of remaining lambs on farm aims to remove these worms, preventing continued pasture contamination by resistant strains. At that time of year, there is likely to be a substantial on-pasture refugia, hence reducing the risk of development of new derivative resistance due to dilution of any resistant strains that have developed. In a scenario where a low-drench dependency management system has been developed and a targeted minority of lambs have been treated, break-dosing/midlate season knock-out drenching may not be necessary, but where sequential treatments have been used, modelling has demonstrated that this targeted treatment can slow resistance (Leathwick and Hosking 2009). Combination anthelmintics may also slow the development of resistance if the pre-existing levels of resistance are low and there are sufficient parasites in refugia (Bartram and others 2012, Leathwick and others 2012).

Summary
PGE and anthelmintic-resistant GIN are a significant threat to the productivity of commercial sheep flocks in the UK. This risk is rapidly escalating and veterinary surgeons have a central role to play in its management.
The necessity for engagement is emphasised by other bodies in the sector, such as the National Sheep Association and the Farm Animal Welfare Council (FAWC). Given the broader implications of anthelmintic resistance, FAWC'S 2016 report stresses the importance of knowledge sharing and improving performance (FAWC 2016).
As flock planning advisers we need to be clear about the aims of management plans on farm (ie, usually efficient and rapid growth of lambs on farm), and PGE is a significant threat to that. We have described a framework for flockwide discussion of PGE and key strategies which should be used at each CCP, highlighting that chemical control should be used only after a series of avoidance strategies have been put in place, monitored and treatment found to be necessary.
The veterinary surgeon's ability to understand and deal with this integral limiter of performance by using a flockwide approach including nutrition, infectious disease, trace element status, best practice and evidencebased medicine places them in a unique position to facilitate management of PGE. The challenge for many practitioners is the initial development of a proactive working relationship, but the evidence for necessity is compelling and the threat of uncontrolled anthelmintic resistance is rapidly becoming a reality.

Declaration of conflict of interest
Hannah Vineer and Eric Morgan are funded by the Biotechnology and Biological Sciences Research Council BUG consortium project 'Building Upon the Genome: using Haemonchus contortus genomic resources to develop novel interventions to control endemic GI parasites' (grant BB/M003949/1).