Short communication: Evaluation of an automated method for assessing white blood cell concentrations in Holstein dairy cows

Assessing white blood cell (WBC) differentials is one way to assess cow health and well-being. The objective of this study was to determine the agreement between WBC populations identified by manual evaluation via microscopy and an automated approach. Data were collected from mid-to-late lactation dairy cows sampled at 6-h intervals starting at 2100 over a 48 period (n = 192). The agreement between the two methods was calculated using a regression model in SAS (v9.4, SAS Institute Inc., Cary, NC) to construct limits of agreement. Data were analyzed by comparing the mean response of the methods for each sample against the difference between the methods The maximum allowable difference (MAD) was set at ±10% for each response variable. Agreement between methods was evident for neutrophils and lymphocytes, but not for monocytes and eosinophils. Agreement for these factors was established from an insignificant P-value, a low R2 value, and concentration of the data within the MAD. This data indicates that the automated method is appropriate for analysis of neutrophil and lymphocyte concentrations. However, accuracy needs to be improved for analysis of monocytes and eosinophils if differentiation of all WBC populations is necessary.