Association between quarter-level milk somatic cell count and intramammary bacterial infection in late-lactation Irish grazing dairy cows.

The objectives of our study were to describe quarter-level prevalence of intrammamary infection (IMI), to evaluate the performance of commonly used somatic cell count (SCC) thresholds for the diagnosis of quarter-level IMI, and to determine those with maximized sensitivity (Se) and specificity (Sp) for identifying quarter-level IMI as defined by positive aerobic culture in late-lactation grazing dairy cows. In this observational study, quarter milk samples were collected from all cows in 21 commercial spring-calving, pasture-based Irish dairy herds. Total SCC determination and aerobic bacterial culture were performed in 8,177 quarter milk samples obtained between 238 and 268 d in milk from 465 primiparous and 1,609 multiparous cows. The Se and Sp of SCC thresholds used for diagnosis of IMI were evaluated against the gold standard aerobic culture separately for all, primiparous, and multiparous cows. The overall prevalence of bacteriologically infected quarters was 6.3%, and it was higher among primiparous (11.3%) than multiparous cows (5.5%). However, considering all samples, quarter-level SCC was higher for multiparous than for primiparous cows (195,250 ± 21,422 vs. 115,940 ± 26,260 cells/mL). Associated Se and Sp for the 200,000 cells/mL threshold were 59.2% and 88.0% for all, 52.7% and 95.4% for primiparous, and 62.9% and 85.9% for multiparous cows, respectively. Receiver operating characteristic curve analyses determined the thresholds that optimized the Se and Sp of a positive bacterial culture: 101,000 cells/mL for all cows [Se = 80.0%; Sp = 76.4%; area under the curve (AUC) = 0.84], 61,000 cells/mL for primiparous (Se = 87.1%; Sp = 84.0%; AUC = 0.90), and 101,000 cells/mL for multiparous (Se = 80.9%; Sp = 72.6%; AUC = 0.83). The results indicate that the 200,000 cells/mL threshold was inefficient in identifying late-lactation quarter-level IMI (low Se) in the studied herds where the main etiological agent was Staphylococcus aureus. Suggested quarter-level SCC thresholds have the potential of serving as a supporting tool for dry cow therapy decisions and warrant further study in late-lactation cows from spring-calving, pasture-based herds with S. aureus as the main pathogen causing IMI.