Biological variation data and analytical specification goal estimates of the thrombin generation assay with and without thrombomodulin in healthy individuals.

Abstract

Evaluation of an individual's thrombin-generating capacity enables a global assessment of the coagulation cascade and is therefore thought to better reflect the clotting function of blood. However, the lack of standardization still hampers the use in routine clinical practice. Nineteen healthy subjects were sampled once a week for 5 consecutive weeks. Thrombin generation assay (TGA) was performed in duplicate by calibrated automated thrombogram (CAT) on platelet poor plasma with and without thrombomodulin. After exclusion of outliers, a nested analysis of variance (ANOVA) was performed to evaluate the biological variability (BV) results. Analytical variation (CVA ), within-individual variation (CVI ), between-individual variation (CVG ), index of individuality (II), and reference change value (RCV) were calculated. All parameters taken together, the CVA, CVI , and CVG without TM, ranged from 2.8% to 6.5%, from 4.1% to 13.3% and from 10.4% to 28.4%, respectively. For TG with TM, CVI and CVG were higher and ranged from 5.0% to 18.1% and from 14.9% to 35.3%, respectively. For endogenous thrombin potential (ETP), a CVI of 4.1% and CVG of 10.4% were obtained without addition of thrombomodulin (TM). With addition of TM, both CVI and CVG were higher: 14.0% and 34.8%, respectively. The II was low and the RCV ranged from 17.2% to 50.4%. CAT parameters are highly individualized and population-based reference values could be called into question. The assessment of BV and RCV for thrombin generation assays could optimize interpretation of serial patient results and guide setting of analytical specification goals.