Harmonization of cut-off values of commercially available D-dimer tests

Abstract

The objective of the D-dimer tests is to exclude venous thromboembolic disorders. The characteristics of various D-dimer tests differ significantly and consequently, their results are difficult to compare. Our goal was to compare three D-dimer tests, analyse their correlation and harmonise their sensitivity and specificity through the optimisation of cut-off values. At Semmelweis University, the D-dimer level was determined with three different reagents in 158 plasma samples, suspected with venous thromboembolism. INNOVANCE D-Dimer was selected as the referent, and in the case of the two other tests (STA-Liatest D-Di; Dia-D-Dimer), the cut-off values were changed between 0.2-1 µg/ml (fibrinogen-equivalent unit - FEU). The optimal cut-off values were estimated by diagnostic parameters and chi-square test. The correlation of the different tests was calculated by regression analysis. Based on the chi-square statistics, there is no significant difference between STA-Liatest D-Di and INNOVANCE D-Dimer tests using the cut-off values 0.3-1 µg/ml (FEU) (STA-Liatest D-Di). In the case of Dia-D-DIMER, there is a significant difference using 0.2-0.3 µg/ml (FEU) cut-off values, 0.4 µg/ml (FEU) is a border-line value and using 0.5-1 µg/ml (FEU) cut-off values, there is no significant difference. The sensitivity of STA-Liatest D-Di changed between 82.7-100% using 0.2-1 µg/ml (FEU) cut-off values, while the sensitivity of Dia-D-DIMER was 92.3-100%. Their specificities ranged between 50-96.3% and 35.2-87%. The optimal cut-off values were estimated as 0.5-0.6 µg/ml (FEU) for the STA-Liatest D-Di test and 0.7 µg/ml (FEU) for the Dia-D-DIMER test. Each diagnostic laboratory should determine the optimal cut-off value of the D-dimer test in use, considering the examined population of the area. Orv Hetil. 2019; 160(15): 585-592.