Autonomic Risk Score (ARS) from 10-second ECGs predicts 30-day intrahospital mortality post-ACS, with highest ARS group showing 9.2% mortality vs 0.8% lowest group.
Does the Autonomic Risk Score (ARS) derived from 10-second ECGs predict 30-day intrahospital mortality in patients undergoing coronary angiography for suspected ACS?
An automated Autonomic Risk Score derived from standard 10-second 12-lead ECGs strongly stratifies 30-day intrahospital mortality risk in patients with suspected ACS undergoing coronary angiography.
Absolute Event Rate: 0% vs 0%
Abstract Background Patients with suspected acute coronary syndrome (ACS) and indication for coronary angiography (CA) can frequently challenge the capacities of modern hospitals. Periodic repolarization Dynamics (PRD), a marker of sympathetic overactivity, is a strong predictor of mortality after ACS, but it’s application in everyday clinical practice is limited, as it requires 20-minute ECGs. Purpose To develop an automatic autonomic risk stratification tool from standard 12-lead ECG recordings that can guide treatment in patients undergoing CA because of suspected ACS. Methods Between 1/2014 and 11/2021 we retrospectively identified patients (positive ethics vote 21-1180)who underwent CA due to suspected ACS at two tertiary centers in our city. Inclusion criterion was availability of raw ECG-data. Exclusion criteria were STEMI and pacemaker stimulation. For development of the Autonomic Risk Score (ARS) three parameters were prospectively identified: heart rate (HR), QTc-interval and PRD. As calculation of PRD requires 20-min ECG-recordings we simulated 106 recordings and randomly truncated 10-second segments. We finally applied artificial intelligence to calculate a new parameter called PRDshort from the 10-second segments. We used established cut-off values (HR ≥/ 100bpm, QTc ≥/ 480ms and PRDshort ≥/ 5.75 deg2) to divide patients in 4 groups based on ARS. The primary endpoint was 30-day intrahospital mortality. Survival curves were estimated by the Kaplan-Meier method. Predictors of mortality were analyzed using Cox-regression analysis. Multivariable models were adjusted for age, sex and maximum creatine kinase (CKmax). Results We retrospectively identified 10,266 patients undergoing CA due to suspected ACS. Of these, 4,790 patients (age 73; IQR 62-80 years, 30.3% females; Tabl. 1) had available ECG raw data and were included in the study. The median ARS was 1 (IQR 0-2). ARS was significantly higher in non-survivors than survivors (2; IQR 1-2 vs. 1; IQR 0-2; p0.001). Figure 1 illustrates intrahospital mortality stratified by ARS. A step-up increase in ARS was associated with a significant two-fold increase in intrahospital mortality (HR 2.07; 95% CI 1.76-2.45; p 0.001). Patients with the highest ARS (HR ≥ 100 bpm and QTc ≥ 480ms and PRDshort ≥ 5.75 deg2) showed a 9.2% (95% CI 5.4–12.9%) intrahospital mortality rate, compared to 0.8% (95% CI 0.3–1.3%) among patients with the lowest ARS (HR 100bpm and QTC 480ms and PRDshort 5.75 deg2; HR 12.30; 95% CI 5.79–26.12; p0.001). The prognostic value of ARS remained statistically significant in multivariable analysis (HR per ARS-increase 1.88; 95% CI 1.58–2.23 and highest vs. lowest ARS HR 8.36; 95% CI 3.80–18.37; p0.001 for both). Conclusion ARS is a fully automated autonomic risk tool that can be routinely assessed using standard 12-lead ECGs. Broad implementation of ARS could lead to hospital decongestion by guiding triage and treatment strategies.Table 1 Figure 1
Sams et al. (Sat,) reported a other. Autonomic Risk Score (ARS) from 10-second ECGs predicts 30-day intrahospital mortality post-ACS, with highest ARS group showing 9.2% mortality vs 0.8% lowest group.