The present study evaluated the prognosis at different troponin concentrations regardless of the underlying diagnosis in an unselected group of patients admitted to the coronary care unit (CCU) and whether the prognosis is dependent on the cause of the troponin increase. Troponin I was measured with the Stratus CS analyzer (Dade Behring) in 468 consecutively and unselected patients admitted to the CCU at Karolinska University Hospital with possible myocardial ischemia lasting <12 hours before arrival. A troponin I concentration below (0.02 to 0.10 microg/L) the diagnostic cut-off level for myocardial infarction predicted mortality after 40 months. In patients with a troponin I level above the diagnostic cut-off level for acute myocardial infarction, the underlying cause of the troponin increase did not affect the prognosis. In conclusion, we report that a troponin I increase below the new cut-off level for myocardial necrosis was associated with increased mortality in unselected patients with possible myocardial ischemia admitted to the CCU at a university hospital. We also observed that the underlying cause of the troponin increase appears to be less important for outcome.