Leukocyte contamination of various blood products is implicated in feb
rile transfusion reactions, in alloimmunization phenomena immunosuppre
ssion, and in the transmission of viral infections. The biggest diffic
ulty in interpreting published reports lies in the methods used to qua
ntify residual leukocytes. As automated cell counters are not precise
and sensitive enough to detect low cell counts, we compared two techni
ques: i) the classical method using the Nageotte haemocytometer; and i
i) fluorescent staining with propidium iodide (PI). We performed these
two techniques to evaluate the residual white cell count in 41 platel
et units collected by cytapheresis. The difference in results obtained
with the two techniques was borderline significant (P = 0.05). To eva
luate the accuracy of both techniques, we added a known number of lymp
hocytes to platelet concentrates that had been filtered three times. T
here was a good correlation between the values obtained using the two
techniques (r > 0.99). However, though the difference between the know
n value and PI results was not significant, the difference for Nageott
e cell counter results was significant (P < 0.0001). From a practical
point of view, the PI fluorescent technique is simple, rapid, and easy
to carry out. Our results demonstrate the reliability of the PI techn
ique for the evaluation of residual leukocytes.