Potential pitfalls of comparative measurements of reticulocytes with flow cytometry and microscopy in prematures and infants.

Reticulocyte counting by flow cytometry (Bayer H*3, ADVIA 120) in blood of prematures, infants and children > 1 year of age was compared with microscopic counting under research conditions (9000 counted red blood cells per slide). While in children >1 year a good concordance of both methods was observed, 2.3-2.4-fold higher values were obtained in neonates by microscopy (Brilliant Cresyl Blue stain, 0.5%). However, another laboratory found good agreement between H*3-counting and microscopy in samples also obtained from neonates using the same methods. Despite very similar results for all age groups in comparative flow cytometry measurements in both laboratories, counting of smears from neonates differed, showing an approximately 2.3-fold larger amount of reticulocytes in our laboratory. The reason for these observations was a greater enlargement (1250-fold) used routinely in our laboratory compared with 800-fold in the other one. Thus very mature reticulocytes frequently found in neonates could only be detected using a 1250-fold enlargement. Similarly, the low concentration of the colouring matter used in the H*3 (0.0005% oxazin or 0.001% ADVIA 120) is obviously not sufficient for detection of mature reticulocytes. Therefore, it is important to consider this phenomenon and to standardise microscopic enlargement, especially for comparisons in multicentre studies.