Interference of blood cell lysis on routine coagulation testing.

CONTEXT Preanalytical factors influencing the reliability of laboratory testing are commonplace. It is traditionally accepted that hemolytic samples are unsuitable for coagulation assays because of the release of hemoglobin, intracellular components, and thromboplastic substances from damaged blood cells. OBJECTIVE To evaluate the influence of blood cell lysis on routine coagulation testing. DESIGN Twelve aliquots prepared by serial dilutions of homologous lysated samples collected from 10 different subjects, and displaying a final percentage of lysis ranging from 0% to 9.1%, were tested for prothrombin time, activated partial thromboplastin time, fibrinogen, and dimerized plasmin fragment D. Lysis was achieved by subjecting citrated whole blood to a freeze-thaw cycle. OUTCOME MEASURES Interference from blood cell lysis on routine coagulation testing. RESULTS Statistically significant increases in prothrombin time and dimerized plasmin fragment D were observed in samples containing final lysate concentrations of 0.5% and 2.7% respectively, whereas significant decreases were observed in activated partial thromboplastin time and fibrinogen in samples containing a final lysate concentration of 0.9%. The current analytical quality specifications for desirable bias are +/-2.0% for prothrombin time, +/-2.3% for activated partial thromboplastin time, and +/-4.8% for fibrinogen. Percent variations from the baseline values exceeding the current analytical quality specifications for desirable bias were achieved for lysate concentrations of 0.9% (prothrombin time and activated partial thromboplastin time) and 1.8% (fibrinogen), corresponding to average free plasma hemoglobin concentrations of 1.7 and 3.4 g/L, respectively. CONCLUSION Our results confirm that, although slightly hemolyzed specimens might still be analyzable, a moderate blood cell lysis, as low as 0.9%, influences the reliability of routine coagulation testing. Because the interference in coagulation assays has a wide interindividual bias, we do not recommend lysis correction and we suggest that the most appropriate corrective measure should be free hemoglobin quantification and sample recollection.