ABSTRACT
Virulent ovine foot rot is a contagious foot disease. Given the development and validation of a real-time PCR to detect Dichelobacter nodosus isolates that contain the virulence-associated protease genes aprV2 and aprB2, the diagnosis of foot rot has made considerable progress. We evaluated pooling methods to reduce the number of samples during a foot rot control program. Samples of individual feet were compared to a 4-feet sample of the same sheep. All further analyses based on 4-feet samples (pools-of-5 and pools-of-10 4-feet samples) were compared to samples of individual sheep, and a risk-based herd sampling was evaluated and compared to the whole flock. The sensitivity and specificity of the 4-feet samples for detection of aprV2-positive strains was 93.8% (CI: 87.6-97.5%) and 98.3% (CI: 96.5-99.3%), respectively. The sensitivity and specificity of the pools-of-10 was 86.7% (CI: 78.4-92.7%) and 100.0% (CI: 97.4-100%), respectively. Pools-of-5 were not significantly more sensitive than pools-of-10. The pooling of 4 individual foot samples into one 4-feet sample is an adequate method to reduce the number of samples of individual sheep. The sensitivity of pools-of-5 and pools-of-10 is too imprecise for a control program. Risk-based sampling allowed for a substantial reduction of samples to be tested, had a sensitivity of 95.8% (CI: 78.9-99.9%) and specificity of 100.0% (CI: 88.1-100.0%) when determining the foot rot flock status, and represents an adequate methodology to predict within-flock freedom from infection.
