Can efficient management of sheep gastro-intestinal nematodes be based on random treatment?

Targeted selective treatment has been proposed as a method to reduce gastro-intestinal nematode infections in ruminants and lower the selective pressure that leads to anthelmintic resistance. Since nematodes are highly aggregated within their host population, treating only the most heavily infected hosts offers an efficient and sustainable strategy to reduce infection within the flock and slow the spread of anthelmintic resistance. Although effective methods to correctly identify such hosts are available, their feasibility is often limited in the field conditions. Instead, treating part of the flock at random may prove to be a useful and practical alternative. This study examined whether such random treatment could be relevant in controlling nematode infections and delaying the selection of anthelmintic resistance compared to targeted selective treatment. Firstly, an individual based model was used to evaluate the sustainability of random treatment according to several parasitic distributions in the host population (negative binomial, uniform and normal distributions). Anthelmintic resistance was modelled based on benzimidazoles, as a monogenic trait. Anthelmintic treatment was done twice a year, week 23 and 41, corresponding to beginning of June and of October. The model was run over a five-year period corresponding to the minimum delay for anthelmintic resistance to be observed following its initial use. The model outputs show an increase in the proportion of treated hosts led to an increase in the frequency of the resistance allele for both treatment regimes. Random treatment was shown to be slightly less efficient than targeted selective treatment in controlling for the infection intensity regardless of the percentage of hosts treated. Random treatment was however more efficient than targeted selective treatment in counter-selecting for anthelmintic resistance in both the aggregated and uniformly distributed models. Secondly, a one grazing season experiment was conducted to compare a random treatment flock (20% of flock was treated at random monthly) against a mass treatment flock (the whole flock was treated monthly). Both treatment regimes produced similar pasture infectivity, similar mean infection intensity and similar final host live weight. This is the first time random treatment of a subset of hosts has been demonstrated to be a sustainable alternative to mass treatment.