Résumé
<p><b>BACKGROUND</b>White blood cell (WBC) counts and differentials performed using an automated cell counter typically require manual microscopic review. However, this last step is time consuming and requires experienced personnel. We evaluated the clinical efficiency of using flow cytometry (FCM) employing a six-antibody/five-color reagent for verifying automated WBC differentials.</p><p><b>METHODS</b>A total of 56 apparently healthy samples were assessed using a five-color flow cytometer to verify the normal reference ranges of WBC differentials. WBC differentials of 622 samples were also determined using both a cell counter and FCM. These results were then confirmed using manual microscopic methods.</p><p><b>RESULTS</b>The probabilities for all of the parameters of WBC differentials exceeded the corresponding normal reference ranges by no more than 7.5%. The resulting WBC differentials were well correlated between FCM and the cell counter (r > 0.88, P < 0.001), except in the case of basophils. Neutrophils, lymphocytes, and eosinophils were well correlated between FCM and standard microscopic cytology assessment (r > 0.80, P < 0.001). The sensitivities of FCM for identification of immature granulocytes and blast cells (72.03% and 22.22%, respectively) were higher than those of the cell counter method (44.92% and 11.11%, respectively). The specificities of FCM were all above 85%, substantially better than those of the cell counter method.</p><p><b>CONCLUSION</b>These five-color FCM assays could be applied to accurately verify abnormal results of automated assessment of WBC differentials.</p>