Effect of milk sample collection strategy on the sensitivity and specificity of bacteriologic culture and somatic cell count for detection of Staphylococcus aureus intramammary infection in dairy cattle

Four milk sample collection strategies for bacteriologic culture and identification of bovine intramammary infection due to Staphylococcus aureus were evaluated, Milk samples were collected at 24 h intervals from 245 lactating mammary quarters of 62 cows from one commercial dairy herd on 6 successive days. A total of 1470 quarter milk samples were available for study, Based on the bacteriologic culture results of all six quarter milk samples, each quarter was classified as infected with or free of S. aureus. The case definition used to establish the 'gold standard' was the isolation of two or more colonies of S. aureus on two or more occasions from the six quarter milk samples obtained from a given mammary quarter, The probability of a false-negative classification of a mammary quarter using all six culture results was estimated to be less than 0.0046, while the probability of a false-positive classification was less than 0.0004. Twenty-two quarters from 16 of 62 cows had S. aureus intramammary infection. Inocula (0.1 ml) for bacteriologic culture were prepared in the laboratory from quarter milk samples to represent alternative strategies for milk sample collection on farms. Sensitivity and specificity of detection of S. aureus-infected mammary quarters and/or cows was then determined, The accuracy of somatic cell counts for the same purpose was also determined for several cut-off values. The range of sensitivity values for bacteriologic culture and SCC were 91-100% and 54-95%, respectively, The range of specificity values for each test method ranged from 97.6 to 100% and from 81 to 83%, respectively. Bacteriologic culture, using any of the sampling strategies examined, had high specificity ( > 98%) and relatively high sensitivity ( > 91%) for identifying S. aureus intra mammary infection (IMI). However, there was a great difference in the number of culture attempts necessary to achieve this accuracy which would influence a dairy farm manager's choice of which type of milk sample collection strategy to use.