It was suggested recently that cardiac troponins are released as T-I-C complexes and then further degraded to T and I-C. It is not known whether the various affinity to the T-I-C and I-C complex of different troponin I test systems influence the diagnostic and prognostic value of the test results in clinical practice. We studied 162 patients (61.3 S.D. 11.1 years) with suspected acute myocardial infarction (AMI) in a single center study. AMI was confirmed in 109 patients. Blood samples were taken at admission, after 1, 2, 4, 8, 12 and 24 h. Troponin I (TnI) was measured using the OPUS plus (TnI-O, cut-off 1.6 microg/l) and the Stratus II (TnI-S, cut-off 1.5 microg/l) analyzers. TnI-O has high affinity to the binary (I-C) and TnI-S to the ternary (T-I-C) troponin complex. A 6-month follow-up with respect to death and recurrent AMI was performed. The sensitivity (SE) and specificity (SP) for AMI diagnosis were 82.6 and 86.8% for TnI-S; 75.2 and 92.5% for TnI-O 0-2 h after admission. The ROC analysis showed a slightly better curve for TnI-S at 4 h (P<0.05). Logistic regression analysis shows prediction of 6 months outcome by 0-24 h serial TnI-S measurements (odds ratio 5.21, P=0.0356), and serial TnI-O measurements (odds ratio 4.92, P=0.0186). High affinity to the ternary troponin complex enhances the diagnostic but not the prognostic value of a test system. Indeed, the resulting differences are small but underline the need for standardization of biochemical markers.