A model to determine sampling strategies and milk inoculum volume for detection of intramammary Staphylococcus aureus infections in dairy cattle by bacteriological culture

A model was developed to evaluate the effects that methods of obtaining milk samples and culture inoculum volumes had on the sensitivity of microbiological culture to detect Staphylococcus aureus intramammary infections (IMI). An assumption was made that milk from mammary quarters infected with S. aureus only contains bacteria intermittently, A modified sine wave function was used to model this intermittent shedding pattern. Specifications for the components of the shedding cycle used in this function were based on quantitative culture results from 54 experimentally infected S. aureus quarters, sampled daily for a period of 30-49 days. The components of the shedding cycle were length in days, peak number of CFU shed per milliliter of milk, and length of time in the cycle when no shedding occurred. These components were used to estimate the model's predicted distribution of S. aureus CFU ml(-1) milk when individual quarter milk samples were cultured for S. aureus. The sensitivity of culture for several sampling methods was then calculated, The model predicted that culture of a single quarter milk sample had a sensitivity ranging from 60 to 87% for detection of S. aureus IMI depending on inoculum volume. Quarter milk samples taken on day 1 and repeated either on day 3 or day 4, and cultured separately using 0.1 ml of milk for culture inoculum, were predicted to have sensitivities of 90-95% and 94-99%, respectively, Other milk-sampling strategies examined included culture of a composite milk sample (equal-volume mixture of milk from four separate mammary quarters) and pooled milk samples in which samples from different milkings (either quarter or composite samples) were mixed together and then cultured. The range of predicted sensitivities of these other sampling strategies was 30-97%. Factors having the greatest impact on the sensitivity of culture, in order of importance were: the type of milk sample, the volume of milk cultured, and the time interval between repeated milk sample collection strategies.