Diagnostic sampling strategies for virulent ovine footrot: Simulating detection of Dichelobacter nodosus serogroups for bivalent vaccine formulation

Dichelobacter nodosus is a slow-growing anaerobic bacterium that is the causative agent of virulent ovine footrot. Vaccination targeted at up to two specific serogroups can eliminate those serogroups from infected flocks, but requires identification of serogroups present in infected flocks. Serogroups can be identified using slide agglutination or polymerase chain reaction (PCR) methods. The objectives of this project were to use stochastic simulation modeling to estimate the efficacy of sampling strategies encompassing 5-40 sheep per flock and 2-4 colonies per sheep, and to compare efficacies based on slide agglutination or multiplex PCR test results. Foot swabs collected from sheep in 12 flocks were used as the basis for a sampling strategy simulation model. None of the evaluated sampling strategies identified the two most common serogroups in the flock, or all serogroups present in the flock, in 95% of iterations. However, a simulated sample of 22 sheep/flock and 2 colonies/sheep resulted in a simulated vaccine that protected 95% of the sheep that could be protected by a single bivalent vaccine, while a sample of 24 sheep/flock and 2 colonies/sheep resulted in a series of simulated bivalent vaccines that protected 95% of diseased infected sheep. The difference in outcome was due to the distribution and frequency of serogroups within certain flocks where some serogroups were uncommon and others dominant. A sampling strategy (>40 sheep/flock, 4 colonies/sheep) that will identify the two most common serogroups in a flock 95% of the time may not be cost effective. Evaluating efficacy based on the expected effect on the flock may be more useful than one which seeks to determine the most common serogroups. These findings are broadly applicable to diseases where more than one strain or type of pathogen may be present and must be represented in a vaccine.