Abstract 5024: Diagnostic and Prognostic Value of High Sensitive Troponin T in Patients With Acute Coronary Syndromes
Background: Cardiac troponins are the gold standard for diagnosis and risk stratification of patients with ACS. As a threshold for troponins, the serum concentration which is identical to the 99th percentile of a healthy population has been defined. However, this threshold has to measurable with sufficient analytical precision (CV<10%). In the past, troponin assays were not able to measure a troponin concentration at the 99th percentile with sufficient precision. However newly developed highsensitive troponin assays will enable measurement at the 99th percentile with the required precision. The aim of the present study is to evaluate the diagnostic and prognostic value of a new hs TnT assay in comparison to the established TnT assay.
Methods and results: From April 2003 until April 2005 all consecutive patients (n=1023, 30% females, median age 65 years) admitted because of an ACS within the last 48 hours were included. Diagnosis was myocardial infarction in 857 (84%) patients. Clinical follow up after 6 months was available. During this period 72 (7.0%) patients died. Conventional TnT was measured on admission and at day 1 (Elecsys®, Roche, Germany). A TnT > 0.03 ng/ml was considered positive. High sensitive troponin T (hsTnT) was measured on admission (Elecsys®, Roche, Germany). For this assay, a threshold of 0.012 ng/ml, which is the 99th percentile and identical with the 10% CV, has been applied. On admission 694 (68%) of the patients were TnT positive, yielding a sensitivity of 81% for the diagnosis of MI. In comparison, 879 (86%) of the patients were hsTnT positive on admission yielding a sensitivity of 96%. The AUC of the ROC curves for TnT and hsTnT for the diagnosis of MI were comparable (0.929 for TnT and 0.954 for hsTnT). Among the 329 patients who were TnT negative on admission, in 133 patients MI could have been diagnosed correctly by hsTnT (sensitivity of 82%, specificity of 67%). In a multivariate Cox regression model, hsTnT was an independent predictor for mortality. In this model, conventional TnT lost its predictive value.
Conclusion: A new hs TnT assay enables early diagnosis of MI with a higher sensitivity as conventional TnT assays. Furthermore, hsTnT levels provide stronger prognostic information for mortality within 6 months as conventional TnT levels.