Results from a multicenter evaluation of the 4th generation Elecsys troponin T assay

This multicenter study evaluated the performance of an improved Elecsys Troponin T (cTnT) immunoassay with a special focus on the serum to heparin-plasma comparability. The evaluation was performed in 10 laboratories according to a standardized protocol (Roche Diagnostics, study no. B05P008). The cTnT assay was performed on Modular E170 and Elecsys 2010 systems. The cTnI immunoassays (Vidas® Biomerieux; Dimension® Dade Behring; Immulite® 2000 DPC) were performed according to the manufacturer’s instructions. Intraassay (n=21) and interassay (2 runs/d, 10d, triplicate measurements) imprecisions were evaluated using two commercial controls and 6 serum pools (cTnT, 0.014 to 4.102 μg/L). Intraassay CVs ranged from 0.7% to 3.2%. Interassay CVs ranged from 3.6% to 35.4% (cTnT <0.10 μg/L) and 1.8% to 9.1% (cTnT >0.10 μg/L), respectively. The cutoff for myocardial necrosis was determined to be 0.028 μg/L using the 10% interassay CV criteria. Linearity was assessed by serial dilutions of 5 serum samples using cTnT negative pools. Linearity was proven up to 21.3 μg/L (recoveries: 90-110%). The comparison between the 4th generation and the commercially available cTnT assay showed highly similar results across the whole measuring range: y=1.02x-0.001, r=0.998; n=988. Using the commercial cTnT reagent, the serum to heparin-plasma comparison yielded a systematic bias to ~10% lower cTnT results in heparin-plasma. Suitable comparability was obtained using the 4th generation assay. The regression analysis (serum vs. heparin) across the studied range (0.01 to 16 μg/L) yielded the following equation: y=0.993x-0.001; r=0.986; n=441. The regression data were comparable to the serum/heparinplasma comparability of the studied cTnI immunoassays (slopes: 0.939-1.241; intercepts: 0.015-0.054, r: 0.981- 0.998, n: 33-49). However, individual serum to matched heparin-plasma samples still yielded poor comparability (bias >20%) using the 4th generation cTnT as well as cTnI assays. Our data show an excellent analytical performance of the improved cTnT mmunoassay, with no systematic bias between serum and heparin-plasma cTnT results. Further studies are necessary to investigate the cause of poor comparability of troponin results in rare individual serum to matched heparin-plasma samples.