Abstract
AIMS: Heart failure may occur following acute myocardial infarction, but with the use of high-sensitivity cardiac troponin assays we increasingly diagnose patients with minor myocardial injury. Whether troponin concentrations remain a useful predictor of heart failure in patients with acute coronary syndrome is uncertain. METHODS AND RESULTS: We identified all consecutive patients (n = 4748) with suspected acute coronary syndrome (61 ± 16 years, 57% male) presenting to three secondary and tertiary care hospitals. Cox-regression models were used to evaluate the association between high-sensitivity cardiac troponin I concentration and subsequent heart failure hospitalization. C-statistics were estimated to evaluate the predictive value of troponin for heart failure hospitalization. Over 2071 years of follow-up there were 83 heart failure hospitalizations. Patients with troponin concentrations above the upper reference limit (URL) were more likely to be hospitalized with heart failure than patients below the URL (118/1000 vs. 17/1000 person years, adjusted hazard ratio: 7.0). Among patients with troponin concentrations