Blood gas analysis represents an important and frequently used clinical decision support. Spurious haemolysis is common at the emergency department (ED) and causes clinically relevant bias to many analyses. Blood gas analysers inability to identify hemolysis can therefore lead to clinical misinterpretations. The objective of this study was to assess the proportion of haemolytic blood gas samples performed at the ED and calculate the predictive values of a novel point-of-care solution for haemolysis detection. Approximately 100 µl of blood from anonymized syringes including venous or arterial blood was, directly after performed blood gas analysis, tested with the POC-Method (Helge H10 s-system, Hemcheck, Karlstad, Sweden) at the ED and result recorded. Thereafter the blood was centrifuged (2000 g) for 10 minutes, 100 µl of plasma put into vials and run in batches of 40 daily on an AU680 analyzer (Beckman Coulter, Brea, California, USA) used as reference method. The threshold for clinically relevant hemolysis was defined as ≥50 mg/dL free haemoglobin. Out of 1270 samples, haemolysis was detected in 7.9% (n = 100). The POC-method identified haemolytic samples with a sensitivity of 80%, specificity of 99%, positive and negative predictive values was 89% and 98% respectively. Spurious hemolysis is frequently present in blood gas samples at the ED and should be taken into consideration when assessed in patients. Errors could be detected directly at the point-of-care with a novel technology for hemolysis detection.
Keywords: Blood gas analysis; emergency service; hemolysis; hospital; point-of-care systems; pre-analytical phase; quality improvement.