The aim of the present work was to evaluate the ProCyte Dx hematology system from IDEXX Laboratories for the sheep species and to verify whether anemia diagnostics in sheep can be improved by using reticulocyte parameters.
EDTA blood samples from 249 sheep were used.
A method comparison was made using the ADVIA 2120, centrifuge hematocrit, and manual differentiation of 200 leukocytes and 1000 erythrocytes.
Statistical analysis was performed using one- and two-factorial analyses of variance, linear regression, Passing-Bablok regression analysis, Bland-Altman analysis, and determination of Spearman's rank correlation coefficient.
The following relevant results were obtained:
- The precision measurement provided coefficients of variation of 0.15-0.84 %
- Leukocyte differentiation yields coefficients of variation of 0.41 - 1.37 %, with the exception of monocytes (coefficient of variation: 4.67 %).
- Reticulocytes are measured with a coefficient of variation of 12.77 %.
- The carryover for the WBC is at 0.007 %.
- Linearity is present for the parameters RBC, Hb, WBC and PLT.
- Anticoagulant comparison shows that the use of Na citrate lowers RBC Hb and Hct levels. Heparin lowers platelet counts.
- In the storage experiment, RBC is stable for 24 hours, after which the values increase. The remaining erythrocyte parameters are distorted by swelling of the cells after 12 hours of storage, platelet concentration decreases after nine hours, WBC values are stable for 48 hours.
- Method comparison between ProCyte Dx and ADVIA 2120 shows rank correlation coefficients of rs = 0.976 - 0.994 for the parameters WBC, RBC, Hb and Hb corrected. The mean difference varies from -1.00 - 1.19. Hct is calculated with an rs = 0.747, bias of -0.01. MCV is presented with an rs = -0.108 and bias of -0.20. MCH and MCH corrected have an rs = 0.732, MCHC and MCHC corrected have an rs = -0.755, regression cannot be calculated. Optically determined platelets correlate with an rs = 0.738 and measured by impedance with rs = 0.548.
- Leukocyte differentiation is for WBC, neutrophil and eosinophil granulocytes and lymphocytes with rs = 0.994, 0.952, 0.994 and 0.929, respectively. The bias is between -1.00 and -0.02. Monocyte differentiation shows an rs = 0.190 with a bias of -0.36. Comparison with manual enumeration yields an rs = 0.903 for neutrophil granulocytes, lymphocytes rs = 0.857, eosinophil granulocytes rs = 0.850, and monocytes rs = 0.503.
- Reticulocyte determination of ProCyte Dx correlates with ADVIA 2120 with an rs = 0.708, bias 4.70 with limits of -3.80 - 13.3. Hemoglobin measurement of reticulocytes does not correlate rs = -0.060. Comparison with manual counting of reticulocytes does not correlate with ProCyte Dx measurement results: rs = 0.068.
- A minor breed influence on the blood parameters could be demonstrated. The extent is so small that attention is not necessary for the evaluation of hematology.
- There is a clear age influence on the hematology of the sheep. For an exact differentiation of the age influence, special attention must be paid to the age of approx. 2 weeks and 3 months.
- An influence of gender on hematology was not demonstrated. Due to the limited number of subjects, the result is not to be considered reliable.
- The reticulocyte parameters provide only low correlations with the other hematological parameters in this work. For the first time, the hemoglobin content of the reticulocytes was measured in sheep. Due to the high variance of the measurement results, this parameter does not allow conclusions in anemia diagnostics.
Christian Schürmann
Doktorarbeit Schafe Tiermedizin