Abstract
The objective of the present study was to propose thresholds of somatic cell counts in bulk-tank milk indicative of the prevalence of subclinical mastitis in a flock. A retrospective analysis was performed on data from a longitudinal survey of subclinical mastitis in Greece, in which the prevalence of subclinical mastitis in 12 flocks sampled four times throughout a milking period was evaluated by collecting milk samples from individual ewes for bacteriological and cytological testing; further, cell counts in the bulk-tanks of the farms were also measured during the visits. Four cohorts were created: A, with cell counts in the bulk-tank milk between 0.100 × 10(6) and 0.400 × 10(6) cells mL(-1), B, with cell counts between 0.400 × 10(6) and 650 × 10(6) cells mL(-1), C, with cell counts between 0.650 × 10(6) and 900 × 10(6) cells mL(-1), and D, with SCC between 0.900 × 10(6) and 1.450 × 10(6) cells mL(-1). There was a significant positive correlation between prevalence of the infection in the flocks and somatic cell counts in bulk-tank milk on the same sampling occasion (p < 0.0001). There was also evidence of significant differences between the four cohorts in the mean prevalence rate of the infection (p < 0.0001). Ninety-five percent confidence intervals of the prevalence of subclinical mastitis according to the somatic cell counts in the bulk-tank milk were calculated as follows: for cohort A, 8.7% to 12.1%, for B, 12.4% to 19.4%, for C, 22.5% to 30.8% and for D, 27.3% to 45.3%. The information that may be thus obtained can be useful in guiding the implementation of various health management procedures for mastitis control in sheep flocks, with no need to perform milk sample collection from ewes and subsequent laboratory examinations.