Instrument-specific reference intervals for coagulation factor XII and antithrombin III: A regional study in Jiangsu, China.
TL;DR
These locally derived RIs may improve the interpretation of FXII and ATIII results and reduce unnecessary follow-up testing caused by inappropriate generic intervals and highlight the importance of local RI verification when applying manufacturer-derived intervals or changing analytical platforms.
Abstract
Objective
This study aimed to establish and verify reference intervals (RIs) for coagulation factor XII (FXII) and antithrombin III (ATIII) in an adult population from the Jiangsu region of China and to evaluate the influence of analytical platform on RI interpretation.
Methods
We retrospectively analyzed data from 301 apparently healthy adults attending the Health Examination Center of Nanjing Drum Tower Hospital. Plasma FXII and ATIII activities were measured in parallel using two automated coagulation analyzers, the Sysmex CS-5100 and CS-6500. Statistical analysis followed CLSI EP28-A3 guidelines. Distribution normality was assessed to determine the appropriate method for RI establishment: the non-parametric percentile method for non-normally distributed data and the parametric method for normally distributed data. Analyzer-related agreement was further assessed using Bland-Altman analysis. The derived RIs were verified using an independent sample set of 20 apparently healthy adults.
Results
FXII showed a non-Gaussian distribution, whereas ATIII showed an approximately Gaussian distribution. For the CS-5100 analyzer, the RI was 32.2%-109.1% for FXII and 84.1%-104.7% for ATIII. For the CS-6500 analyzer, the corresponding RIs were 37.1%-111.1% for FXII and 86.1%-106.7% for ATIII. The newly established RIs differed from the manufacturer-provided ranges currently used in routine practice. No clinically meaningful sex-based partitioning was identified. Although analyzer-related differences were statistically significant, Bland-Altman analysis showed modest mean biases between the CS-6500 and CS-5100 platforms, with mean differences of +5.09% for FXII and + 1.17% for ATIII. All verification samples fell within the proposed RIs.
Conclusion
This study established and verified regional RIs for FXII and ATIII in healthy adults from Jiangsu, China. Although analyzer-related differences between the CS-5100 and CS-6500 platforms were modest, the findings highlight the importance of local RI verification when applying manufacturer-derived intervals or changing analytical platforms. These locally derived RIs may improve the interpretation of FXII and ATIII results and reduce unnecessary follow-up testing caused by inappropriate generic intervals.