Long-term association of ultra-short heart rate variability with cardiovascular events

Abstract

Heart rate variability (HRV) is a cardiac autonomic marker with predictive value in cardiac patients. Ultra-short HRV (usHRV) can be measured at scale using standard and wearable ECGs, but its association with cardiovascular events in the general population is undetermined. We aimed to validate usHRV measured using ≤ 15-s ECGs (using RMSSD, SDSD and PHF indices) and investigate its association with atrial fibrillation, major adverse cardiac events, stroke and mortality in individuals without cardiovascular disease. In the National Survey for Health and Development (n = 1337 participants), agreement between 15-s and 6-min HRV, assessed with correlation analysis and Bland-Altman plots, was very good for RMSSD and SDSD and good for PHF. In the UK Biobank (n = 51,628 participants, 64% male, median age 58), after a median follow-up of 11.5 (11.4-11.7) years, incidence of outcomes ranged between 1.7% and 4.3%. Non-linear Cox regression analysis showed that reduced usHRV from 15-, 10- and 5-s ECGs was associated with all outcomes. Individuals with low usHRV (< 20th percentile) had hazard ratios for outcomes between 1.16 and 1.29, p < 0.05, with respect to the reference group. In conclusion, usHRV from ≤ 15-s ECGs correlates with standard short-term HRV and predicts increased risk of cardiovascular events in a large population-representative cohort.

Figure 1

Flow diagram of the study. HRV: Heart rate variability; usHRV: Ultra-short HRV; PAC/PVC: Premature atrial/ventricular contractions; CVD: Cardiovascular disease. ECG abnormality includes bundle branch block morphology, sinus node disfunction, atrial fibrillation and premature atrial and ventricular contractions. MACE: Major adverse cardiac disease.

Figure 2

Heart rate variability indices. From top to bottom: a representative example of 15-s ECG, RR-interval (RRI), successive differences of RRI, power spectral density of RRI (after pre-processing, see text). The standard deviation of successive differences (SDSD), the root mean square of successive differences (RMSSD) and high-frequency power spectral density (PHF) for this example are shown in the figure.

Figure 3

Method agreement. Bland–Altman (top) and scatter (top) plot comparing RMSSD and PHF from 6-min versus 15-s ECGs in NSHD. Data from 20 non-overlapping 15-s intervals are plotted against data from 6-min ECG recordings. Mean estimation error, limits of agreement and correlation coefficients were estimated for each non-overlapping 15-s intervals. In the Bland–Altman plot, solid red lines represent the minimum and maximum bias (mean estimation error) across the 20 15-s intervals, while dashed lines represent minimum and maximum limits of agreement across the 20 15-s intervals. In the correlation plots the red line represents the diagonal x = y.

Figure 4

Cross-sectional associations between risk factors and usHRV. Distributions are shown as median (interquartile range). Forest plots show the estimated coefficients and confidence intervals of a linear multivariable regression models. Continuous variables, including usHRV indices, were normalised to mean = 0 and standard deviation = 1. BMI: Body mass index; HTN: Hypertension; β-Block: Use of beta-blockers; Diab: Type 2 diabetes. LDL: low-density lipoprotein cholesterol; RHR; resting heart rate.

Figure 5

Long-term associations between usHRV and cardiovascular outcomes. Bars and whiskers indicate adjusted hazard ration and 95% confidence interval for 1 SD decrease in the continuous exposure unless specifically stated (+ 1SD indicates 1 SD increase). Models for ultra-short HRV indices (RMSSD, SDSD and PHF) and model for heart rate recovery (HRR) were adjusted for resting heart rate, age, sex, body mass index, hypertension, smoking, LDL cholesterol, diabetes, and use of beta-blockers.

Figure 6

Non-linear relationships for long-term associations between usHRV and cardiovascular outcomes: Hazard ratio and 95% confidence interval (shaded area) for atrial fibrillation (AF), major adverse cardiac events (MACE), stroke and mortality are shown as a function of ultra-short heart rate variability (usHRV, RMSSD, SDSD, PHF), resting heart rate, and heart rate recovery (HRR). Boxplots represent exposures’ distribution, with number indicating percentiles. All models were adjusted for standard risk factors (see text) and usHRV and HRR models were further adjusted for resting heart rate.

Figure 7

Associations between 10-s and 5-s HRV and outcomes. Hazard ratio and 95% confidence interval (shaded area) for atrial fibrillation (AF), major adverse cardiac events (MACE), stroke and mortality are shown as a function of ultra-short heart rate variability. Boxplots represent exposures’ distribution, with number indicating percentiles. All models were adjusted for standard risk factors (see text). The root mean square of successive differences (RMSSD) and the standard deviation of successive differences (SDSD) were measured during the first 10 s (left, green) and 5 s (right, orange) of the resting phase.