OBJECTIVE To determine if high sensitivity troponin T (hs-TnT) measurements performed during the 'plateau phase' of troponin release (≥48 h) following ST-segment elevation myocardial infarction (STEMI) can predict major adverse cardiovascular endpoints (MACE), and to evaluate its prognostic value compared with cardiac magnetic resonance imaging (CMRI) parameters. RESULTS We prospectively recruited 201 first presentation STEMI patients. Serial hs-TnT levels were measured at admission, peak (highest), 24, 48 and 72 h. CMRI and transthoracic echocardiography were performed (4 days median) post-STEMI, evaluating infarct scar characteristics and left ventricular ejection fraction (LVEF). Associations were determined between hs-TnT levels and CMRI parameters early after STEMI with MACE (comprising mortality, re-infarction, new or worsening of heart failure, cerebrovascular accident, and sustained ventricular arrhythmias) at medium-term follow-up. After 602 days (median), 33 (17%) patients had MACE. Upper tertile hs-TnT levels at 48 and 72 h were associated with MACE (Kaplan-Meier P = 0.002 and P = 0.012, respectively). Multivariate Cox analyses, incorporating diabetes, CMRI scar size, LVEF and hs-TnT levels (applied at a single hs-TnT time point) showed that 48 and 72 h hs-TnT levels were independent predictors for MACE (HR = 1.20, P = 0.002, and HR = 1.21, P = 0.035 respectively). CONCLUSIONS Measurement of hs-TnT in the plateau phase after STEMI is an inexpensive method of prognostic risk assessment.