[Comparison and evaluation of testing results for two different coagulation analyzers].

OBJECTIVE: To evaluate the performance of a newly installed fully automatic coagulation analyzer and compare the consistency of its testing results with the confirmed clinical automatic coagulation analyzer at our department. METHODS: Precision, linearity, carryover and accuracy of the newly installed coagulation analyzer were evaluated according to the national required standards. Then the testing results were analyzed between the newly installed and confirmed coagulation analyzers according to the EP-5 and EP-9 documents of national committee for clinical laboratory standards (NCCLS). RESULTS: For the newly installed coagulation analyzer, the low, median and high values of relative intra-precision were: 0.93%, 1.32% and 1.27% for prothrombin time (PT); 1.42%, 0.84% and 1.17% for activated partial thromboplastin time (APTT); 1.82%, 3.13% and 3.19% for fibrinogen (FIB); 1.78%, 1.76% and 1.38% for thrombin time (TT) respectively. The linear regression equation of FIB actual and theoretical values was y = 1.012x + 0.0219 (P > 0.05). There was no significant statistical difference between the intercept and 0 (t = 0.2287, P > 0.05) and between linear slope and 1 (t = 0.3221, P > 0.05). The carryover was -2.33%. The testing results of defined acceptable bias of PT and FIB in CLIA'88 for two analyzers were within the acceptable 95% confidence interval of bias. CONCLUSION: The precision, linearity, carryover and accuracy of the newly installed coagulation analyzer meet the requirements of instrument user manual. The performance and the testing results of the same sample from two coagulation analyzers are consistent.