STEMI versus occlusion MI - a retrospective challenge to an established paradigm (SORC study)

Abstract Background At least 20% of patients with acute coronary occlusion do not manifest an electrocardiogram (ECG) suggestive of ST-elevation myocardial infarction (STEMI). These patients face greater morbidity and mortality, primarily due to delayed diagnosis and intervention. The concept of occlusive myocardial infarction (OMI) has gained traction and drives a paradigm shift towards a more comprehensive diagnostic approach, supported by the use of artificial intelligence (AI). Purpose This study synthesises evidence on diagnosing OMI, highlighting the insensitivity of the current STEMI criteria and untrained clinicians at detecting OMI. This aims to propagate adoption of OMI nomenclature, to reduce the burden on high-risk patients. Methods A retrospective analysis was conducted on a 750-patient cohort discharged from a UK tertiary cardiac centre with a diagnosis of myocardial infarction, over an 8-month period in 2022. Initial ECGs were independently and blindly evaluated by two experienced interventional cardiologists and two medical students trained in OMI recognition ("OMI experts"). The primary definition of occlusive pathology was: (1) culprit vessel with Thrombolysis in Myocardial Infarction (TIMI) 0-2 flow, (2) TIMI 3 flow with a high-sensitivity cardiac Troponin I (Hs-cTnI) value 5,000ng/L and a new regional wall motion abnormality, or (3) TIMI 3 flow with a raised Hs-cTnI and evidence of acute plaque rupture/erosion on intravascular imaging. In patients without invasive angiography, secondary definitions incorporated troponin assays alongside diagnostic imaging. Results Preliminary results suggest that European Society of Cardiology (ESC) STEMI criteria were met in 36.9% of initial ECGs (46.6% of patients with OMI). Of the 63.1% of OMI patients that presented without STE, 36.2% later developed STEMI. At discharge, 33.6% of patients diagnosed with NSTEMI had OMI pathology. OMI patients had higher rates of cardiac arrest, cardiogenic shock and 30-day mortality (23.6%) than STEMI patients alone (19.4%). When interpreting the initial ECGs, OMI experts demonstrated a greater sensitivity and diagnostic accuracy than both standard ESC STEMI criteria and interventional cardiologists. OMI experts also exhibited a higher degree of inter-rater concordance. STEMI criteria exhibited a low sensitivity and diagnostic accuracy for identifying OMI but retained a high specificity. Interventional cardiologists outperformed STEMI criteria, achieving a higher sensitivity and diagnostic accuracy, albeit with reduced specificity. Conclusion Established ESC STEMI criteria provide poor sensitivity and accuracy in comparison to trained OMI experts. A vast proportion of patients with OMI present without STEMI and may never develop STEMI ECG changes, yet their ECG is consistent with transmural ischaemia. There is a need for broader recognition of frequently presenting OMI patterns. This could improve prognosis, reduce false negative and false positive diagnoses.Methods Overview Diagnostic Performance of Experts