Blood withdrawn through a cannula valve connector does not result in clinically significant haemolysis

To assess the impact of cannula valve connectors on haemolysis of blood samples drawn from newly inserted cannulae.

In a semi-blinded, randomized study paired blood samples, with and without cannula valve connector, were obtained from patients within the ED and tested for haemolysis, defined as haemolysis index of greater than 120 mg/dL. Patients were randomized as to which sample was collected first. Cannula size was standardized and vacutainer systems provided consistent draw pressures. Time taken for the tube to fill was recorded as a measure of blood flow.

Two hundred and ninety patients were randomized, with six subsequently excluded from analysis because of samples being lost or insufficient for testing. Average patient age was 60.8 years and 52.5% were male. There were no significant differences between the randomization groups. The overall rate of haemolysis was 2.6%, being 2.8% in the valve first group and 2.5% in the no valve first group (P = 1.0). Time for collection averaged 7.7 s in the valve first group and 7.5 s in the no-valve first group (P = 0.22). Mean serum potassium level was 4.4 mmol/L in both groups (P = 0.46). The rate of hyperkalaemia was not different between valve first and no-valve first groups (12.7% and 13.7%, respectively, P = 1.0).

The attachment of a cannula connector valve to a peripheral cannula prior to blood sampling is not associated with an increase in the rate of haemolysis or hyperkalaemia.