Abstract
We read with interest two recent related case reports regarding false potassium results in patients with very high white cell counts. 1,2 In our laboratory, we have also seen both pseudohyperkalaemia and pseudohypokalaemia in the presence of major leukocytosis. In 2004, we reported a study in seven leukaemic children with white cell counts >300 × 109/L. 3 In four of these, the plasma potassium concentrations, measured after vacuum tube transport and centrifugation, were markedly higher than corresponding plasma potassium concentrations measured on a blood gas machine within 60 min of the samples being taken. For example, one patient with a white cell count of 812 × 109/L had a plasma potassium concentration of 9.8 mmol/L after centrifugation with a corresponding potassium concentration measured using whole blood of 4.0 mmol/L. In 2006, we reported pseudohypokalaemia in a boy with myeloid leukaemia and a white cell count >200 × 109/L. 4 His centrifuged plasma potassium results were 1.4 and 1.2 mmol/L, with corresponding potassium concentrations measured using whole blood of 3.0 and 4.1 mmol/L. He had a marked metabolic alkalosis, so potassium may have moved into his white cells ex vivo at room temperature before his blood was centrifuged.
Leukaemic cells can be very fragile and easily break down either in the process of transport through a vacuum tube or while being centrifuged in the laboratory. Therefore, we feel that the most appropriate way of measuring potassium in patients with very high white cell counts (>100 × 109/L) is to measure plasma potassium concentration using whole blood on a blood gas machine or a similar device. If there are any concerns about ex vivo haemolysis, this can be checked by centrifuging the sample after measurement. Although serum samples have been shown also to be free from this problem, 1 we are of the opinion that using whole blood is a much better and simpler approach, given the wide availability of blood gas machines in hospitals that admit such patients. In addition, the turnaround time using a blood gas machine is much quicker – an important consideration in this clinical situation.
For a number of years now, we have had this policy operating in our hospital, to good effect. It requires close cooperation between biochemistry and haematology departments to make sure that laboratory staff are quickly made aware of patients with white cell counts >100 × 109/L. We feel that if such a policy were more widely adopted, the problem of false potassium results in the presence of major leukocytosis would go a long way to being solved.
DECLARATIONS