Periodic Repolarization Dynamics derived from 10-second ECG recordings predicts mortality in patients after myocardial infarction. (3rd October 2022)

• Record Type: Journal Article
• Authors: Sams, L E; Woerndl, M; Villegas Sierra, L E; Krasniqi, A; Massberg, S; Bauer, A; Rizas, K D
• Is Part Of: European heart journal. Volume 43(2022)Supplement 2
• Publication Date: 2022-10-03
• Subjects: Cardiology -- Periodicals; Heart -- Diseases -- Periodicals; 616.12005
• DOI: 10.1093/eurheartj/ehac544.1350
• Languages: English
• ISSNs: 0195-668X

Abstract

Abstract: Background: Periodic repolarization dynamics (PRD) is an electrocardiographic biomarker that quantifies low-frequency (LF), sympathetic-activity associated instabilities of repolarization. PRD is a strong predictor of mortality in patients after myocardial infarction (MI). The main limitation of PRD is the requirement of ECGs with a duration of ≥20 minutes. Calculation of PRD using 10-second ECGs would be advantageous allowing the implementation in everyday clinical practice. Purpose: We aimed to develop and validate a modified version of PRD, originating from 10-second ECGs, which we called PRDshort. Methods: First, the beat-to-beat change in the direction of repolarization, called dT° was measured for 30-minute ECGs (Figure 1) and PRD was quantified as the amplitude of LF periodicities (≤0.1 Hz) within dT°. We randomly selected segments with a duration of 10 seconds. For each of these segments we calculated several parameters based on dT° and RR-interval. To overcome the issue that the wavelength of PRD is longer than 10 seconds, we performed signal-simulation and machine learning analysis. We simulated 100.000 dT°-signals using different assumptions for the level of PRD, heart rate, respiratory rate, number of premature ventricular contractions and the level of artifacts. Thereafter we used machine learning to calculate PRD from single 10-second ECG recordings (Figure 1). This method was finally validated in a cohort including 455 patients after MI. The primary endpoint was 3-year mortality. The prognostic power of PRD was evaluated using Kaplan-Meier and Cox-regression analyses. Results: The Pearson's correlation coefficients between PRD and PRDshort were 0.80 (0.79–0.80) in the simulated data and 0.75 (0.70–0.78) in the post-MI cohort. In the post-MI cohort 47 patients died within 27±11 months of follow-up. The median left-ventricular ejection fraction (LVEF) was 50±15%. PRDshort was significantly higher in non-survivors (6.8±5.7 deg 2 ) than survivors (4.9±3.0 deg 2 ; p<0.001). Dichotomization of PRDshort at the median value of ≥/<5.0 deg 2 identified a high-risk group with a 3-year mortality rate of 21.0% (13.4–27.9%) compared to a mortality rate of 6.5% (2.7–10.2%; HR=3.2; 1.6–6.2; p<0.001; Figure 2) among patients with PRDshort <5.0 deg 2 . In multivariable analysis, PRDshort was independent from GRACE-score >140 and LVEF ≤35% (HR 2.7; 1.4–5.2; p=0.003). In ROC analysis the predictive value of PRDshort didn't differ significantly from that of the original PRD (p=0.263). PRDshort ≥5.0 deg 2 detected 33 out of the 34 deaths originally identified by PRD. Conclusion: This is the first description of a method to calculate PRD from 10-second ECG recordings. The prognostic value of PRDshort was comparable to that of PRD in post-MI patients with preserved LVEF. As normal 12-lead ECG-recordings are ubiquitous in every hospital and doctor's office this method may allow the wide application of PRD as risk stratification tool in everyday clinical practice. Funding Acknowledgement: Type of funding sources: None.