Prognostic Value of Resting Heart Rate and Heart Rate Variability in the 12-lead Electrocardiogram: Mortality Data from the CODE Database

Abstract

Resting heart rate (HR) and heart rate variability (HRV) reflect cardiovascular autonomic control and are implicated as prognostic factors. We aimed to evaluate the prognostic value of HR and HRV in a tele-ECG cohort network. We assessed unique 12-lead ECGs recorded from patients ≥16 years, from a tele-ECG database in Brazil, between 2010 and 2017. Variables of interest were HR and standard deviation of normal RR intervals (SDNN). Four Cox models were adjusted to evaluate the association between HR and HRV and the outcomes of interest (all-cause and cardiovascular mortality, assessed by ICD codes from death certificates): model (1) Unadjusted; (2) Adjusted for sex and age; (3) Model 2 + risk factors and clinical comorbidities; (4) Model 3 + adjustment for HRV or HR, respectively. At total 992,611 individuals were included, median age of 55 years. In 6 years, there were 33,292 (3.37%) deaths, 21% due to cardiovascular causes. After adjustments (model 4), all HR quartiles were independently associated with a progressively increased risk of all-cause mortality, being 88% higher for the fourth quartile (HR = 1.88, 95% CI 1.77 to 1.89). Similarly, the first and second HRV quartiles remained associated with increased all-cause mortality (first quartile [HR] = 1.42, 95% CI 1.37 to 1.47) in the final model. HR (fourth quartile: [HR] = 1.77, 95% CI 1.65 to 1.91) and HRV (first quartile: [HR] = 1.33, 95% CI 1.23 to 1.44) were also independent predictors (model 4) of cardiovascular mortality. In conclusion, in a large cohort of Brazilian adults, baseline HR and HRV were independent predictors of all-cause and cardiovascular mortality, even when adjusted for each other.