The precision, accuracy, linearity, comparability and so on of SF-8000 were evaluated.The 200 g/L hemoglobin solution was added to 12 cases of fresh plasma in the proportions of 1:200, 2:200, 3:200, 4:200 and 5:200, and the experimental plasma was prepared. The differences of the results in the experimental plasma, the original plasma and the brine-diluted control plasma were analyzed. The differences of 15 cases of bubble specimens and control specimens' results were analyzed by paired t-test. The 12 fresh specimens were measured with fresh dissolved-reagents and 24h-refrigerated reagents separately, and the differences of the 2 sets of results were analyzed. A total of 13 cases of 4h-refrigerated specimens, 24h-refrigerated specimens and fresh specimens were measured, and the differences were analyzed.

The basic properties of the instrument were all in accordance with the requirements of manufacturers. Hemolysis and bubbles affected the results of SF-8000 in different levels. When hemoglobin concentration reached 1-1.5 g/L, the results of activated partial thromboplastin time(APTT)and prothrombin time(PT) had significantly statistical differences from those of the original plasma group(P<0.01), when it reached 4 g/L, they had statistically significant differences from the control plasma group(P<0.01). In addition to the reagent of APTT(P<0.01), the results of 24 h-refrigerated reagents and fresh refrigerated reagents had no statistical differences with good stability. The results of 4h-refrigerated specimens and fresh specimens had no statistical difference, but the results of 24 h-refrigerated specimens and fresh specimens in APTT, PT and fibrinogen(FIB) had statistical significance(P<0.01).

The performance of SF-8000 meets the national and industrial standards. The manual operation should be strictly carried out to avoid appearing confounding factors, in order to ensure the results with clinical applicability.