Microscopic differential cell counting to identify inflammatory reactions in dairy cow quarter milk samples

The diagnosis of intramammary infections is mostly based on somatic cell count (SCC) and bacteriological analysis. As an alternative, differential cell counting (DCC) could be a useful method, because it identifies changes in the relative cell populations before the increase in total cell number occurs. The aim of the study was to identify cytological parameters that could be used in the field to classify mammary quarters as healthy or diseased, comparing cyto-bacteriological results with DCC. Overall, 48 cows were randomly selected from 3 herds in Lombardy region of Italy. Herd A was characterized by the absence of contagious microorganisms; in herds B and C, the prevalence of Staphylococcus aureus was 20 and 50%, respectively. Foremilk samples were aseptically collected from 188 quarters and submitted to bacteriological analysis, SCC, and DCC. For statistical analysis, the samples were clustered into 4 health groups, and DCC results were compared in each group. Ninety-six samples were classified as normal secretions (N), 30 as mastitis (M), 15 as latent mastitis (LM), and 47 as unspecific mastitis (UM) based on SCC and bacteriological results. Single percentages of lymphocytes, polymorphonuclear neutrophilic leukocytes (PMNL), or macrophages were first evaluated to established variables capable of identifying healthy and inflamed quarters. Then, combinations of cell populations were tested to increase the discrimination power of DCC: phagocytes, logarithmic PMNL:lymphocyte ratio, and logarithmic phagocyte:lymphocyte ratio. The mean percentage of lymphocytes was significantly higher in group N than in groups LM, UM, and M. The mean percentage of PMNL was significantly lower in group N than in groups UM and M, but not LM. Mean percentages of macrophages were not significantly influenced by the 4 groups. The mean value of phagocytes was significantly lower in group N than in the other groups. Both the logarithmic PMNL:lymphocyte and the logarithmic phagocyte:lymphocyte ratios were significantly lower in group N than in groups LM, UM, and M. Fisher (F-)values were calculated, and the highest F-value was that of log PMNL:lymphocytes ratio (48.23). The explanation for this could be that log PMNL:Lym is the only variable that involved both cell populations statistically influenced by health groups but excluded macrophages. Microscopic DCC has potential as a tool to identify cows affected by any inflammatory process of the mammary gland, with the best results being achieved using log PMNL:lymphocyte as variable.