Short communication: Evaluation of an automated in-house hematology analyzer for bovine blood

摘要

The objective of this study was to evaluate the suitability of the V-Sight hematology analyzer (A. Menarini Pharma GmbH, Vienna, Austria) for bovine blood by a comparison with a reference device (Ad-via 2120i, Siemens AG, Erlangen, Germany). In total, 97 blood samples were obtained from 75 dairy cows. Analyzed parameters included counts of white blood cells (WBC), lymphocytes, monocytes, granulocytes, red blood cells (RBC), and platelets (PLT), as well as hemoglobin concentration (HGB), hematocrit (HCT), mean corpuscular volume, mean corpuscular hemoglobin (MCH), MCH concentration (MCHC), mean platelet volume (MPV), and plateletcrit (PCT). Based on Passing-Bablok regression, the V-Sight provided accurate and precise results for MCH and MCHC only. The PCT results were comparable to the reference method, but precision was inconclusive. Significant proportional differences were detected for monocytes, granulocytes, HCT, and PLT. For all other analytes, significant proportional and systemic differences were observed. The WBC and lymphocyte results from the V-Sight were characterized by poor accuracy, poor precision, and a high number of false positive outliers. Bland-Altman analysis indicated negative biases for all WBC parameters, the erythrocyte indices, and PLT. Positive biases were observed for RBC, HGB, HCT, MPV, and PCT. Correlation coefficients of ＞0.9 between the V-Sight and the reference method were found only for RBC, HGB, HCT, and MPV. Intraassay precision of the V-Sight analyzer was acceptable (coefficient of variation ＜5%) for granulocytes, the erythrocyte indices, and MPV. It was unacceptable (coefficient of variation ＞5%) for WBC, lymphocytes, monocytes, as well as RBC, and inconclusive for HGB, HCT, PLT, and PCT. Sensitivity was high for all RBC counts and indices as well as PLT, but low for monocytes, granulocytes, and MPV. Specificity was high for monocytes and granu- locytes, but low for RBC, HCT, MCH, and MCHC. With accurate and precise results for only 2 out of 13 parameters, the V-Sight cannot be recommended for analysis of bovine blood.