Application of Point‐of‐Care Testing for Coagulation Tests: A Comparison of Consistency and Interference With a Fully Automatic Coagulation Analysis System

ABSTRACT Background Point‐of‐care testing (POCT) of coagulation is rapid, accurate, and portable. However, because it is difficult to standardize the reference intervals due to differences in metrological traceability, normalizing the results across different testing systems is challenging. The International Organization for Standardization 22870:2016 requires POCT to be compared with large automatic analyzers for medical safety. Thus, the aim of this study was to compare the correlations, consistency, and conformity of a mechanical‐based coagulation POCT system with a fully automated coagulation analyzer, and to explore the common interfering factors for each test. Methods From September 2023 to December 2023, 551 plasma samples were collected consecutively for coagulation testing by the Laboratory Medicine Department of the First Hospital of Jilin University. Prothrombin time/international normalized ratio, thrombin time (TT), activated partial thromboplastin time (APTT), and fibrinogen concentration were determined by a Sysmex CS‐5100 coagulation analyzer to serve as a control group. The coagulation results produced by POCT served as the experimental group. Passing‐Bablok, Bland–Altman, and Fourfold tables were used for correlation, consistency, and conformity, respectively. Logistic regression was used for analysis of interference. Results The correlation coefficients r of PT, INR, TT, APTT, and fibrinogen between the two testing systems were 0.866, 0.863, 0.694, 0.904, and 0.997, respectively ( p < 0.001). The overall deviations were 1.5, 0.08, 0.2, 0.8, and 0.0, respectively ( p < 0.001). The conformity were 91.95% (137/149), 93.96% (140/149), 88.44% (130/147), 89.03% (138/155), and 100.00% (100/100), respectively. The kappa values were 0.827, 0.805, 0.570, 0.768, and 1.000, respectively ( p < 0.001). More than half of the POCT measurements were higher than the analyzer measurements. Logistic regression analysis showed that jaundice (OR = 2.903, p < 0.05) and white blood cell (WBC) count (OR = 1.012, p < 0.05) were statistically significant risk factors for outliers between the two measurement systems being compared. Conclusions We found that a mechanical‐based coagulation POCT testing system showed good correlation, consistency, and conformity with automatic analyzers. The factors of hematocrit, platelet count, hemoglobin, WBC count, and jaundice may interfere with the findings of POCT, causing higher results than those produced by large coagulation analyzers.