ABSTRACT
<p><b>INTRODUCTION</b>Haemolysis of blood samples is a common problem encountered in the Emergency department (ED). It leads to inaccurate blood results and has cost implications as blood samples very often have to be retaken. The purpose of our study was to determine which factors in blood sampling were associated with higher rates of haemolysis.</p><p><b>MATERIALS AND METHODS</b>An observational convenience sample of all patients presenting to the ED requiring blood urea and electrolyte (UE) analysis were eligible for our study. Questionnaires were distributed to the doctors and nurses conducting blood sampling to determine the method used and outcome data were collected after the samples were processed.</p><p><b>RESULTS</b>Out of 227 UE samples analysed, 45 (19.8%) were haemolysed. Various factors, including method (IV cannulation or venepuncture), system (syringe or vacutainer), operator, rate of blood flow, difficulty of cannulation/venepuncture and source of blood (arterial or venous), were analysed, but their effects on haemolysis were not statistically significant (P >0.05). However, the use of the vacutainer system was associated with the highest rates of haemolysis [adjusted odds ratio (OR), 6.0; 95% confidence interval (CI), 2.3 to 15.1].</p><p><b>CONCLUSION</b>We found blood sampling with the vacutainer system to have increased rates of haemolysis. This could potentially change attitudes towards equipment used for blood sampling in the ED.</p>