Determination of Reticulocytes: Three Methods Compared

Abstract

Determination of reticulocytes in peripheral blood is a valuable tool for getting information about erythropoiesis of an individual. For many years, reticulocyte numbers were quantified manually by means of a microscope after staining with supravital dyes. However, this method is tedious and shows low reproducibility. Therefore, several methods for the automated determination of reticulocytes have been established in laboratory routine within the last years. The aim of this study was to compare three of these automated methods for reticulocyte analysis. Reticulocytes were determined in 130 subsequent routine samples by means of an ABX Pentra 120 Retic haematological analyser, a Coulter EPICS XL MCL flow cytometer and a Coulter STKS haematology system, using the fluorescent dye thiazole orange or the supravital dye new methylene blue for reticulocyte staining, respectively. The reticulocyte concentrations were slightly lower for the Coulter STKS haematology system (mean ± SD 1.89 ± 1.32%) when compared with the Coulter EPICS XL MCL flow cytometer or the ABX Pentra 120 Retic haematological analyser (2.11 ± 1.25% and 2.12 ± 1.15%, respectively). The correlations between all methods were significant (r_s ≥ 0.843, p < 0.001). Small intercepts were, however, observed in the correlation plots between the values obtained by means of the Coulter STKS haematology system and those obtained by the other two methods. Within-batch coefficients of variation were 6.0%, 6.9% and 7.8% for the ABX Pentra 120 Retic haematological analyser, the Coulter STKS haematology system and the Coulter EPICS XL MCL flow cytometer, respectively. The corresponding between-batch coefficient of variation values were 6.8%, 4.9% and 5.3% as well as 14.1%, 7.6% and 6.1% for the low, medium and high control levels determined by means of the ABX Pentra 120 Retic haematological analyser and the Coulter STKS haematology system, respectively. These data suggest that all three methods allow the efficient and reliable determination of reticulocyte counts under clinical routine conditions. However, although the obtained data are very similar, differences exist which should be taken into account for the normal values of the different methods.