How repeatable is a faecal egg count reduction test?

Aim: To gather information on the repeatability of a faecal nematode egg count (FEC) reduction (FECR) test (FECRT), evaluating both different methods of calculating efficacy and variations within a method, in order to supply veterinarians and other advisors with sufficient information to apply some level of confidence around a diagnosis of anthelmintic resistance based on FECRT results. Methods: Two commercial sheep farms were selected on the basis of having previously recorded FECR < 95% after treatment with ivermectin (Farm 1) or albendazole (Farm 2). On each farm at least 250 lambs, managed as a single mob, were individually ear-tagged and sampled for FEC. The resulting counts were used, 3-4 days later, to sort the lambs into 24 groups of 10. First, the animals were split into three groups of 80, having high, medium or low FEC. Second, within each of these groups the 80 animals were further divided into four replicate mobs of 20 (each with the same mean count). Third, each of these replicates was further split into two groups of 10: those that would be drenched and those that would remain as untreated controls. All animals were again faecal-sampled and those in the drenched groups were dosed, using a syringe, to their individual liveweight, with ivermectin (Farm 1) or albendazole (Farm 2). Ten days after treatment all animals were individually faecal-sampled again. FEC and larval cultures were undertaken for all 24 groups from both pre- and post-treatment samples. Efficacy (FECR) of the undifferentiated FECRT was calculated using three different equations, and efficacy by genus was also calculated. Results: Calculated efficacies differed between equations, and the equation which did not utilise an untreated control yielded significantly lower efficacy estimates on both farms. Faecal cultures varied considerably in the proportions of parasite genera recovered. In general, this did not differ between FEC groups, except on Farm 1 where Haemonchus spp were more common and Cooperia spp less common in high-FEC samples. Estimated efficacies against individual genera varied considerably or very little, depending on the level of resistance. On both farms, differing proportions of tests against some genera passed or failed FECRTs based on a threshold pass mark of >= 95% FECR. Conclusion: There was considerable variability in the outcomes of FECRTs and in larval culture results. Caution is warranted in interpreting the results of FECRTs when efficacy values fall into the 90-95% range. Further, the possibility of a test returning a false-negative result is raised, indicating that even an efficacy estimated >= 95% may not guarantee the absence of resistant parasites.