Discipline: udder health; Key words: SCC, quarter milk samples, composite samples, threshold, udder infection. 

Somatic cell counts (SCCs) from quarter samples have been generally accepted and used as the operational measure of inflammation in the udder, as an indicator of the severity of udder infection and as an indicator of economic losses. However, the SCC threshold level used to describe normal milk, infection and subclinical mastitis has often been and still is a controversial subject. In an IDF bulletin of the sixties udder quarters were considered normal when no pathogens were isolated and the SCC was less than 500 000 cells per mL milk. More recently, quarter milk samples with an SCC of equal or less than 100 000 cells per mL from which no pathogens have been isolated and without a history of recent infection are considered to be normal, whereas those with an SCC of less than 200 000 cells per mL have been regarded as an indication of an inflammatory response and that the quarter is likely to be infected. The current quarter milk SCC threshold of 200 000 cells per mL does not distinguish between principal udder pathogens although pathogens are known to differ in their pathogenic effects. Therefore, information on SCC threshold levels for specific pathogens causing infection is needed because SCC is known to differ between udder pathogens and over time. 

As the average number of lactating cows per herd in South Africa has increased substantially in the last decade, quarter milk samples for routine udder health herd examination have been replaced by composite milk samples because of practical and financial reasons. In composite samples, milk from the four udder quarters is combined, with a consequent dilution effect that has required different interpretation of the results from quarter milk samples. However, an international SCC threshold has not been agreed for composite milk samples. Therefore, in the article of the authors referenced below, the incidence of udder infection in composite milk samples at various SCC levels was studied further. 

The objective of their study was to establish a SCC threshold to predict the presence of udder infection in composite milk samples and to compare the results with those in quarter milk samples. A retrospective study was conducted on 345 461 composite and 89 638 quarter milk samples from South African herds. Udder infection at SCC thresholds of 150 000 cells per mL and 200 000 cells per mL differed only by 3.26% in composite milk samples. The optimum SCC thresholds for composite and quarter milk samples was estimated to be 150 000 cells per mL and 200 000 cells per mL, respectively. The likelihood of infection for udders and quarters, respectively, was 1.034 and 1.327 at an SCC threshold of 150 000 cells per mL and 0.864 cells per mL and 1.177 cells per mL at 200 000 cells per mL.

The SCC test was found to be good but not excellent as an indicator of IMI in both quarter and composite sample types. The current threshold for SCC of 200 000 cells/mL used to detect only IMI in quarter milk samples was reconfirmed to be optimal. A SCC threshold of 150 000 cells/mL milk can therefore be recommended for use in composite milk samples and could be used in practice as a selection criteria to select samples for culturing in large herds. Although less true culture-negative milk samples can be expected to be identified for composite milk compared to quarter milk samples because of a lower sensitivity of the SCC test in composite samples, it was indicated that SCC could be used at the lower threshold level to indicate udder infection in composite milk samples. Thus, using the SCC as the only test to predict the presence of IMI is helpful as a screening test to select samples for microbiological determinations. However, it is not ideal as a stand-alone test to indicate udder infection. 


I-M. Petzer, J. Karzis,  E. F. Donkin, E. C. Webb & E. M.C. Etter, 2018. Somatic cell count thresholds in composite and quarter milk samples as indicator of bovine intramammary infection status.  Onderstepoort Journal of Veterinary Research 84(1), a1269. https://doi. org/10.4102/ojvr.v84i1.1269.