Impaired cardiac autonomic function predicts mortality in patients with cardiogenic shock requiring extracorporeal life support

Background:

Despite advances in intensive care medicine mortality in patients with cardiogenic shock remains high. Extracorporeal life support (ECLS) with use of venoarterial extracorporeal membrane oxygenation (VA-ECMO) is a promising treatment alternative in patients with cardiogenic shock as a bridge to recovery, transplant or mechanical assist device. However, the ECLS-Shock trial showed no reduction of mortality in patients with ECLS compared to usual medical treatment.

Objective: 

To test, whether ECG-based automatic cardiac autonomic risk stratification can help to identify patients that might benefit from treatment with ECLS.

Methods:
In this analysis we retrospectively included patients treated with ECLS between February 1^st 2018 and April 30^th 2020 at the LMU-university hospital (LMU-eICU registry). Inclusion criteria were defined as availability of ECG data between 2:00 am and 2:30 am during the first night of the ICU-stay after ECLS implantation and sinus rhythm. Cardiac autonomic function was quantified using a previously published score (CAF_ICU), based on deceleration capacity of heart rate (DC) and periodic repolarization dynamics (PRD). The primary endpoint of the study was all-cause mortality within 30 days after ECLS implantation. Survival curves were generated using the Kaplan-Maier method and were dichotomized at the optimal value of CAF_ICU derived from ROC analysis.

Results:

During the study period 105 patients were treated in our ICU with ECLS due to cardiogenic shock and out of these 63 met the inclusion criteria. Thirty two (50.8%) died within 30 days after ECLS implantation. CAF_ICU was significantly higher in non-survivors than survivors (20 (13-30) vs. 15 (11-22) units, p = 0.045) and proved to predict 30-day mortality (HR 2.260 (1.13-4.53); p=0.022) after dichotomization at the optimal cut-off value of 23 (Fig. 1). The predictive value of CAF_ICU (HR 1.03 (1.00-1.06); p=0.043) as a continuous variable was independent of the highest lactate (1.07 (1.00-1.16); p=0.062) and catecholamine index (1.00 (1.00-1.00); p=0.099) at admission. CAF_ICU and age over 65 years were the only independent predictors of mortality in the multivariate model with hazard ratios of 2.28 (1.10-4.73, p=0.026) and 2.71 (1.19-6.21, p=0.018) respectively (Tab. 1). 

Conclusion:

ECG-based automatic cardiac autonomic risk stratification is prognostic of mortality in patients with cardiogenic shock and ECLS. Determining cardiac autonomic function might help in identifying patients who might benefit from ECLS implantation.

Table 1: This table shows the results of the multivariate Cox-Regression model. CAF_ICU and age over 65 are the only independent predictors of 30 day mortality in this collective of patients with ECLS due to cardiogenic Shock. 

Figure 1: Cumulative 30-days mortality stratified by cardiac autonomic function score (CAF_ICU) ≥ 23 units (red line) and < 23 units (blue line)