K2- or K3-EDTA: the anticoagulant of choice in routine haematology?

Summary The choice of K2- or K3-EDTA as the preferred anticoagulant for blood count remains controversial. We compared the effect of different concentrations of both anticoagulants on normal blood. In optimal conditions (appropriate anticoagulant concentration and measurements done between 1 and 4 h after phlebotomy), no marked differences are seen between either EDTA salt. Important discrepancies appear, however, in less optimal conditions, as often happens in day to day practice. The packed cell volume, when measured on centrifuged blood, decreases with increasing anticoagulant concentrations and this is most pronounced with the K3 salt. This phenomenon has been reported by different authors and is ascribed to shrinking of erythrocytes in an hypertonic medium. Automated instruments react in a different way, their MCV is not influenced by K3-EDTA concentrations up to ten times normal, while K2-EDTA, at high concentrations, results in a slight increase in MCV, as measured with three of the instruments. With most instruments, the accuracy of the white cell count is not markedly influenced. However, when measured with the Unipath CD 3000 all tested blood samples, taken in high K3 (but not in K2) concentrations, showed an appreciable decrease in the leucocyte count (to less than 50% of the original value, at a concentration of 15g/l, 24 h after blood collection). Measurement of RDW and automated differentials is also influenced by the choice of anticoagulant when determinations are done in less than optimal conditions. We conclude that the choice of anticoagulant, its use at an appropriate concentration and the age of the blood sample are important matters and should be given due consideration. However, generally it is only in less than optimal conditions (too small blood samples for the amount of anticoagulant and delayed measurement) that marked divergencies between results obtained with different systems are observed.