This study aimed to determine whether myocardial infarction (MI) could be safely diagnosed or excluded within 30 min instead of 1 h.

This single-center, prospective, observational study included patients presenting with non-traumatic chest pain. Patients underwent a thorough evaluation, including medical history, physical exams, ECG, and serial hs-cTn T measurements at 0, 30, and 60 min. Patients were classified into STEMI or further evaluated for NSTEMI-ACS based on ECG results. The hs-cTn T tests placed patients into rule-in, observation, or rule-out groups. Diagnostic performance was assessed using sensitivity and negative predictive value (NPV) to rule out MI (primary endpoints) and specificity and positive predictive value (PPV) to rule in MI (secondary endpoints).

809 patients were analyzed, with a median age of 53 ± 15.9 years, 36.1 % of whom were women. MI was confirmed in 15.6 % of patients. The 0-h/30-min algorithm placed 457 patients in the rule-out group, 222 in observation, and 188 in the rule-in, while the 0-h/1-h algorithm placed 507, 141, and 161 patients, respectively. The 0-h/30-min and 0-h/1-h algorithms showed identical sensitivity [100 % (96.11 %–100 %) and 100 % (99.61 %–100.00 %), respectively] and NPV [100 %] for excluding MI. Both had high specificity in the rule-in group [94.83 % (92.95 %–96.34 %) and 92.31 % (90.08 %–94.17 %)]. The 0-h/30-min algorithm had a superior PPV [71.54 % (64.75 %–77.48 %)] compared to the 0-h/1-h algorithm [66.46 % (60.53 %–71.91 %)].

The 0-h/30-min algorithm is as effective as the 0-h/1-h algorithm in safely ruling out MI and may offer improved diagnostic efficiency in ruling in MI.