Abstract
Bailey and Thurlow 1 recently studied the negative impact of suboptimal phlebotomy technique on potassium results. They concluded that the contribution of the patient fist clenching to pseudohyperkalaemia is underestimated. Earlier studies at Quest Diagnostics, Inc., a large international medical laboratory, as well as our most recent experiments, support and expand upon these findings.
Following the publication of an article by Don et al., 2 in which they described a patient in whom pseudohyperkalaemia resulted from repeated clenching and unclenching a fist during venepuncture; all phlebotomists at Quest Diagnostics' patient service centres were instructed not to have patients repeatedly clench their fist during venepuncture. This change in practice decreased the proportion of serum potassium concentrations more than 5.3 mmol/L from 3.7% to 2.4%. 3
Finger flexion, however, still remains to be widely unrecognized among physicians and phlebotomists as a leading cause of pseudohyperkalaemia. This was brought to our attention by continuing episodic complaints from physician clients who drew blood specimens in their offices. At times, without any clinical explanation, a patient's serum potassium concentration was reported to be greater than 5.9 mmol/L. In each case we found that the patient had squeezed a hand-held soft rubber ball during venepuncture. Based on this observation we worked with three clinical practices – two general and one oncology – to observe what would happen to the distribution of potassium concentrations if patients stopped using squeeze balls. We retrieved from each of the three practices, the potassium concentrations found in 500 consecutive anonymous blood samples drawn up to the day squeeze balls were discarded. We then recorded the potassium concentrations of 500 consecutive anonymous blood samples drawn after squeeze balls were discarded.
Elimination of squeeze balls led to a striking and statistically significant downward shift in the distribution of potassium concentrations that decreased the overall average proportion of potassium concentrations above the reference range (>5.3 mmol/L) from 10.5% to 2.6% (P ≤ 0.001). The decrease in pseudohyperkalaemia was greater in the two general practices than in the oncology practice. For example, in one general practice the proportion of potassium concentrations above the reference range decreased from 10.4% to 1.0%. In contrast, the decrease in the oncology practice was from 12.0% to 5.4%, presumably because patients being treated for cancer are more likely to have true hyperkalaemia. In addition, and contrary to published comments that sampling techniques are of little importance in the search for hypokalaemia, 4 the elimination of squeeze balls increased the overall average proportion of potassium concentrations below the reference range (<3.5 mmol/L) from 0.3% to 1.4% (P = 0.003), thereby revealing previously unrecognized hypokalaemia in all the three practices.
Haemolysis, a tight tourniquet and small needles are often listed among leading preanalytical causes of pseudohyperkalaemia. 3,5 They are not. Serum samples having cherry red colour before haemolysis has a significant effect on serum potassium concentration, 3 and studies of tight tourniquets and small needles in patients whose veins are difficult to puncture are confounded by the common association of these presumed causes with finger flexion, whether by clenching and unclenching the fist or repetitive flexion of fingers on a squeeze ball.