Abstract BACKGROUND The release pattern of copeptin during the initial 36 h of spontaneous acute myocardial infarction (AMI) has received relatively little investigation but may provide important information on optimal timing of diagnostic measurements. METHODS We investigated the release pattern of copeptin and cardiac troponin T in patients with suspected acute coronary syndrome (ACS). Blood samples were collected in the ambulance, at admission, and after 2, 4, 6, and 12–36 h. Copeptin and high-sensitivity cardiac troponin T (hs-cTnT) were measured in heparin plasma samples. RESULTS Of 93 patients studied, 37 (39.8%) had ST-elevation myocardial infarction (STEMI), 20 (21.5%) non-STEMI, 20 (21.5%) unstable angina pectoris (UAP), and 16 (17.2%) non-ACS diagnoses. Peak copeptin concentrations were detected during ambulance transport for NSTEMI patients [median 94.0 pmol/L, interquartile range (IQR) 53.3–302.1 pmol/L] and at admission for patients with STEMI (70.0 pmol/L, 22.0–144.8 pmol/L). In patients with AMI, copeptin decreased significantly over time (P < 0.0001). This was true for patients with STEMI (P = 0.005) and non-STEMI (P = 0.021). The diagnostic performance during ambulance transport was similar for hs-cTnT (area under the ROC curve 0.75, 95% CI 0.62–0.88) and copeptin (0.81, 0.69–0.92). In early presenters (n = 52), no patient with AMI was initially (in ambulance or at admission) negative for copeptin, resulting in an area under the ROC curve of 0.963 for ambulance values and a negative predictive value of 100%. In late presenters, the negative predictive value of copeptin was 50% in ambulance and at admission. CONCLUSIONS Our analysis is the first to show a consistent early increase in copeptin at first medical contact in the ambulance and a decrease to routine values within 12–36 h in patients presenting early with spontaneous AMI.