In order to prepare the substitution of a commercially available diagnostic kit, ProCOUNT (Becton Dickinson) or Stem-Kit (Coulter Immunotech), for our institutional protocol, we compared the three techniques for the numeration of CD34+ progenitor cells in 50 peripheral blood and 51 apheresis samples, obtained from cancer patients or healthy donors. We show here that the three techniques yield results of the same order of magnitude. Although statistical analyses demonstrate significant differences between the three methods, these differences turned out to be clinically insignificant in most situations. Observed differences mostly affect samples with the highest content of CD34+ cells, while the three assays provide equivalent results for values that are close to clinically relevant thresholds (20 x 10(3) CD34+ cells/ml in peripheral blood to start apheresis, and accumulated number above 3 x 10(6) CD34+ cells/kg to stop apheresis). This study also supports the view that institutional protocols can provide a highly reliable determination of CD34+ cells counts and percentages. However, because institutional protocols often use research reagents and vary from institution to institution, the use of diagnostic kits may be prefered as one way to improve quality assurance in the practice of cell therapy.