Quasi stationary and non-stationary 12 lead electrocardiogram features for improved diagnosis and prognosis of myocardial infarction and myocardial ischemia.

Coronary artery disease (CAD) is the leading cause of death worldwide. CAD is defifined as the narrowing of the coronary arteries due to cholesterol-containing deposits (plaques) on the inner walls of the arteries. A severe narrowing, or occlusion, of a coronary artery leads to a supply-demand imbalance of oxygen. This imbalance can cause irreversible myocardial necrosis, referred to as acute myocardial infarction (AMI). As highly effffective treatments are available, the early and accurate detection of AMI is crucial. The rapid identifification of patients with ST-segment elevation MI (STEMI) based on the ECG followed by immediate coronary revascularisation has resulted in improved survival over the past decade. In contrast, the rapid identifification of non-ST-elevation MI (NSTEMI) is still a major unmet clinical need. ECG criteria provide only very limited sensitivity and specifificity. While the clinical introduction of high-sensitivity cardiac troponin (hs-cTn) assays has substantially improved early diagnostic accuracy and time to diagnosis, their use has multiple limitations including the need for a large central laboratory analyzer, serial sampling, decision delay due to long turn-around times (often 1-2h), and up to 50 % of patients remaining in the observe-zone with rapid hs-cTn-based algorithms. Therefore, the development and evaluation of novel ECG signatures would have enormous clinical value in the assessment of patients with suspected NSTEMI. The present work assesses the diagnostic and prognostic performance of quasi stationary and non-stationary ECG features in patients with suspected NSTEMI in a large prospective multicentre study. Stationary ECG features are considered to be time independent, therefore, do not change between heartbeats of an ECG recording. The features studied are: The Cardiac Electrical Biomarker (CEB), the QRS-T angle, the QRS duration, the QTc interval and the positive T-wave amplitude.
The CEB reflflects the multipolar versus dipolar activity of the heart. The levels of CEB were higher in NSTEMI patients and the combination with standard ischaemic ECG features led to an improved diagnostic performance. The QRS-T angle, a quantifification of depolarisation and repolarisation heterogeneity, is higher in patients with NSTEMI compared to patients with other causes of chest pain, increases diagnostic performance of standard ischaemic ECG features and is an independent predictor for all-cause mortality within 2-years follow-up. The QRS duration, the depolarisation time, and the QTc duration, the heart rate adjusted re- and depolarisation time, were higher in patients with NSTEMI. However, the diagnostic value was very limited. The QTc interval was a signifificant independent predictor for all-cause mortality in 2-years follow-up. Hyperacute T-waves, tall wide T-waves which are an early ECG feature of acute myocardial ischaemia, provide only modest incremental diagnostic value to established ECG-features including ST-segment elevation.
Furthermore, based on the hypothesis that altered electrical de- and repolarisation of myocardial cells lead to small temporal and spatial changes of the surface potential, we implemented and assessed the non stationary instability marker. The instability marker can diagnose NSTEMI with high specifificity and reasonable high sensitivity. Finally, we automated the Selvester-Score algorithm to quantify myocardial scar. The score predicted large scars with high specifificity and was an independent predictor for 1-year mortality