Interpretation of California mastitis test scores using Staphylococcus aureus culture results for screening of subclinical mastitis in low yielding smallholder dairy cows in the Dar es Salaam region of Tanzania.

Screening of subclinical mastitis under field conditions is done using the California mastitis test (CMT). CMT score of > or = 1 corresponding to > or = 500,000 somatic cells ml(-1) is commonly used as threshold of subclinical mastitis in temperate countries. However, given the innately high physiological level of somatic cells in low yielding dairy cows, this threshold may not apply to low yielding dairy cows. The current study was undertaken to investigate the clinical utility of CMT for screening of Staphylococcus aureus subclinical mastitis in low yielding smallholder dairy cows in Tanzania. A total of 1151 of quarter-milk samples were CMT tested, of these 914-originated from cows with a lactation period of 14-305 days. All samples were screened for subclinical mastitis by the CMT as well as microbiological culture of single, duplicate (two consecutive) and triplicate (three consecutive) samples as a gold standard. For the duplicate and triplicate quarter-samples, cows were considered positive for S. aureus subclinical mastitis if results of microbiologic culture for S. aureus were positive for two of two, and for at least two of the first three consecutive quarter-milk samples collected from that cow, respectively. Using a CMT score of > or = 1 would classify 78.6% of the 940 quarter-samples as positive. Eighty-two percent of the samples in which S. aureus was isolated had CMT scores > or = 2; this would classify 51.6% of the 940 quarter-samples as positive. For the single sample, this cut-off had sensitivity, specificity and likelihood ratio for S. aureus of 0.87, 0.83 and 4.24, respectively. For the duplicate quarter-milk samples this cut-off had sensitivity, specificity, and likelihood ratio of 0.94, 0.86, and 5.19. While, for the triplicate quarter-milk samples this cut-off had sensitivity, specificity and likelihood ratio of 0.97, 0.92 and 7.47, respectively. Based on these results and practical considerations, it is concluded that CMT score of > or = 2 corresponding to > or = 800,000 somatic cells Ml(-1) is the best cut-off to correctly identify S. aureus intramammary infections in low yielding dairy cows in Tanzania.