To the Editor
Although the article by De Smet et al1 is an interesting evaluation of hematology instrument performance with body fluids, such instruments were designed for, and cleared by regulatory bodies only for, blood samples. There are significant scientific errors or weaknesses in several of the approaches taken to develop performance specifications for cytopenic body fluid samples, as briefly discussed herein.
The authors describe their approach to “background concentration limits,” which simply applies a +2 SD value to the mean of 20 instrument cycles (presumably air aspirations, although not stated). As noted in the EP-17 consensus document from the Clinical and Laboratory Standards Institute,2 this common practice lacks appropriate statistical rigor and leads to falsely low values. EP-17 provides a more …