Heart rate turbulence-based predictors of fatal and nonfatal cardiac arrest (The Autonomic Tone and Reflexes After Myocardial Infarction substudy)

Abstract

A previous report on heart rate (HR) turbulence showed its value in postinfarction risk stratification. The present study determines the predictive value of HR turbulence in a low-risk population after acute myocardial infarction and provides insight into its pathophysiologic correlates. With use of the database of the The Autonomic Tone and Reflexes After Myocardial Infarction (ATRAMI) study, data were obtained from 1,212 survivors with a mean duration of follow-up of 20.3 months. The a priori end point was defined as the combination of fatal cardiac arrest and nonfatal cardiac arrest. HR turbulence characterized by turbulence onset (TO) and turbulence slope (TS) was calculated and correlated with baroreflex sensitivity (BRS) and the SD of the normal-to-normal RR intervals (SDNN). A composite index of cardiac autonomic function was assessed by combining HR turbulence (TO and TS), BRS, and SDNN. Both TO and TS correlated moderately but significantly with BRS and SDNN (r = 0.26 to 0.44, p <0.001 for all correlations). On Cox's univariate regression analysis, the RRs for abnormal values of TO, TS, and the combination of abnormal TO and TS were 1.86 (95% confidence interval [CI] 0.96 to 3.61, p = 0.065), 4.08 (95% CI 2.11 to 7.89, p <0.0001), and 6.87 (95% CI 3.06 to 15.45, p <0.0001), respectively. The composite autonomic index (combined TO, TS, BRS, and SDNN) was the strongest risk predictor: for all 4 abnormal factors, RR 16.79 (95% CI 6.01 to 46.89, p <0.0001). On multivariate analysis, abnormal TO and TS, and left ventricular ejection fraction remained as independent predictors: RRs 4.07 (95% CI 1.70 to 9.77, p = 0.0017) and 3.53 (95% CI 1.76 to 7.06, p = 0.0004), respectively. In a separate model, the composite autonomic index was the strongest multivariate risk predictor: RR 8.67 (95% CI 2.72 to 7.65, p = 0.0003) for all abnormal factors, and adjusted for left ventricular ejection fraction. Thus, this study confirms the independent value of HR turbulence in predicting fatal cardiac arrest and nonfatal cardiac arrest in a low-risk post-acute myocardial infarction population. By combining HR turbulence, BRS, and SDNN, a comprehensive assessment of cardiac autonomic reflexes and modulation can be obtained.