We investigated the difference in the number of myocardial infarction (MI) diagnoses based on troponin T compared with clinical and epidemiologic (modified FINnish Multinational MONItoring of trends and determinants in CArdiovascular diseases) diagnoses, and the prognosis of patients with discordant diagnoses. Five hundred fifty-nine consecutive patients (315 men and 244 women, median age 69 years) were admitted to the hospital with a suspected acute coronary syndrome. Median follow-up time was 17 months. Of the 559 patients, 127 had a clinical and 137 an epidemiologic diagnosis of MI. When a diagnosis of MI was primarily based on troponin T (>0.10 microg/L), the number of MIs was 169, which increased by 33% compared with the number of MIs by clinical diagnosis, and by 23% compared with those by epidemiologic diagnosis. However, troponin T was not elevated in 13% of the 127 patients with the clinical diagnosis and in 14% of the 137 patients with the epidemiologic diagnosis of MI. Among patients in whom clinical diagnosis of MI was not made, the prognosis with regard to coronary death or nonfatal MI was not significantly worse in patients with troponin T >0.10 microg/L than < or =0.10 microg/L (hazard ratio 1.07; 95% confidence interval 0.62 to 1.84). In patients with a suspected acute coronary syndrome, troponin T-based diagnostics leads to an increase in the number of patients diagnosed with MI compared with clinical or epidemiologic diagnosis. The prognostic impact of troponin T in patients without clinical diagnosis of MI based on elevations in conventional enzyme activities needs further study in larger series of patients.