Plasma or serum samples: measurements of cardiac troponin T and of other analytes compared.

Conflicting data in the literature concern possible differences in the immunochemical measurement of cardiac troponins, either in plasma or in serum. In order to address this specific point, 96 serum and heparin-plasma pairs were obtained for cardiac marker measurement [cardiac troponin T (cTnT); myoglobin (Myo) and creatine kinase-MB isoenzyme (CK-MB)]; 29 additional "common" analytes were measured in 77 such samples. The cardiac markers were measured by electrochemiluminescence (Elecsys 2010, Roche); the other analytes by established automated methods (Modular, Roche). Mean plasma/serum ratios for cTnT (0.95), creatine kinase-MB (1.01) and myoglobin (0.99) were comparable with those of the 29 common analytes (interval of means 0.83-1.05). The distribution of the plasma-serum differences also showed similarities between cardiac markers and other analytes. A few outlier plasma-serum differences (3-5%) were measured for both categories of analytes. Addition of heparin to serum (51 samples) caused decreased cTnT (mean ratio 0.92). In 3 of 51 such samples the cTnT decrease was more marked, but in a second sample from the same subjects (1 week later) such a prominent, heparin-induced loss of cTnT no longer appeared. In conclusion, plasma-serum differences in immuno-reactive cTnT compare with those observed for other analytes. In occasional heparin-plasma samples immunochemical measurement of cTnT may give exceptionally low values. However, in our sample group of 96 patients (cTnT lower or higher than the cut-off in, respectively, 24 and 72 patients), no misclassification occurred if plasma instead of serum cTnT values were considered.