Evaluation of rapid somatic cell counters under different analytical conditions in ovine milk

A total of 31 individual ovine milk samples, ranging between 30 and 2600 x 10(3) cells/mL, were divided into 8 aliquots/milk with the objective of studying the overall accuracy of 2 rapid somatic cell count (SCC) counters, one based on cytometry on disk (Fossomatic 360) and the other on flow cytometry (Fossomatic 5000), under 4 types of preservation (without preservation, bronopol, sodium azide, and potassium dichromate) and 2 analytical temperatures (40 and 60 degrees C). All analyses were carried out in duplicate. In addition, each sample was analyzed in quadruplicate by reference microscopic method using Pyronin Y-methyl green as a stain. A second experiment using 13 samples divided into 20 aliquots/ sample, enabled repeatability to be studied depending on the values obtained in the SCC and on the SCC equipment used. Comparison of the methods was based on repeatability and accuracy studies (means comparison and regression studies vs. reference method). Both counters gave adequate repeatability and accuracy values in ovine milk, though the SCC obtained by Fossomatic 5000 was closer to the reference method and was somewhat more repeatable than Fossomatic 360. In the regression study, slope and intercept values were statistically different from their theoretical values (1.00 and 0.00, respectively) in the unpreserved samples but not in the preserved ones. In all cases, correlation coefficients very close to 1.00 were obtained. The preserved milk analyzed by flow cytometry gave optimal repeatability values (s(r) = 16.3 to 19.7 and s(r)% = 1.9 to 2.4), and their logSCC means (5.62 to 5.64) were not different from the reference value (5.63). Bronopol was the optimal preservative for the Fossomatic method. Analytical temperature did not contribute significantly to SCC variation, although disk cytometry gave slightly more repeatable SCC at 40 degrees than at 60 degrees C.