A novel 12-lead standard ECG derived scoring system for risk stratification in the emergency department

Abstract

Abstract Background/Aims The ECG represents a routinely used diagnostic tool in the assessment of patients in the emergency department. Beside specific alterations within the ECG indicative of underlying pathologies, such as ST-elevation in myocardial infarction, several other parameters have been developed over the past decades that proofed themselves as strong and independent predictors of worsened outcome independent of a specific underlying pathology. These include basic metrics such as heart rate and QRS duration, but also more advanced biomarkers such as QRS microfragmentation (QRSµf) [1] and total cosine R to T (TCRT) [2] . In this work, we aimed to summarize all these markers derived from 12-lead standard ECGs into one novel risk score and evaluate its performance as mortality predictor in the emergency department. Methods We retrospectively collected 10- second 12- lead ECG recordings unselected patients presenting to the emergency department of a large university hospital between September 2019 and June 2022. For each patient a risk score was calculated from the first retrieved ECG using the criteria displayed in Figure 1. Calculation of all parameters was done as previously published. Patients with a total score of 0-1 points, 2-3 points, 4-6 points and 7 points were considered as low, intermediate, high and very high risk, respectively. The performance of the risk score was evaluated using Cox regression analysis and Kaplan-Meier curves with all-cause mortality within one year as primary endpoint. Results In total, 10 781 patients were included in our analyses. Median age was 69 (IQR 54-80) years, 5 588 (52%) were male and 5 193 (48%) were female. 6 966 (65%) patients were considered as low risk, 2 656 (25%) patients as intermediate risk, 1 098 (10%) patients as high risk and 61 (1%) patients as very high risk according to the ECG risk score with one year mortality rates of 8%, 17%, 26% and 43%, respectively (p < 0.001 for difference). In cox regression analysis the novel ECG risk score was a strong and independent predictor of one-year all-cause mortality with a hazard ratio of 1.35 (95% CI 1.31-1.39, p < 0.001). Kaplan-Meier survival curves for different risk categories are displayed in Figure 2. Conclusion In an unselected population of patients presenting to the emergency department of a large university hospital a novel risk score calculated from standard 12-lead ECGs recorded during clinical routine was a strong and independent mortality predictor. Further studies are needed to evaluate how incorporation of the novel score into existing triage systems can improve patient care.Figure 1:Calculation of ECG risk scoreFigure 2:Kaplan-Meier survival curves