To examine factors affecting heart rate variability (HRV) and the role of HRV in heart disease. Specifically, to examine the role of HRV: as a predictor of fatal and nonfatal ischemic heart disease over a six year follow-up of the Atherosclerosis Risk in Communities (ARIC) population based, bi-ethnic cohort; on the six year progression of carotid atherosclerosis measured by B-mode ultrasound; and on the incidence of hypertension. Also, to study the effect of elevated fasting insulin, glucose, diabetes mellitus, and other metabolic abnormalities on changes in HRV over nine years of follow-up.

BACKGROUND:

Heart rate variability analysis has been widely used in clinical research as a noninvasive measurement of autonomic function. It has been found to be associated with post-myocardial infarction mortality, hypertension, sudden cardiac death, atherosclerosis, and diabetes. However, little epidemiologic research on HRV had been reported prior to this study in 1996. Almost no data were available on the population distribution of HRV, its correlates in populations, the factors associated with changes in HRV over time, or on the cardiovascular sequelae of impaired autonomic function assessed by HRV obtained from population-based prospective studies.

DESIGN NARRATIVE:

The study was ancillary to ARIC, a population-based, longitudinal study of cardiovascular and pulmonary diseases. The baseline exam was completed in 1987 to 1988, followed by yearly contacts and re-examinations every three years. The present study built on the data collected by the ARIC investigators by retrieving and processing beat-to-beat heart rate data collected during the baseline exam. Five minutes of beat-to-beat heart rate data were obtained from the ARIC cohort participants during their third follow-up visit (Visit) 4 in 1996 through 1998. Time and frequency domain HRV indices were derived for an assessment of autonomic function. The following HRV indices were computed both for the baseline and the nine-year follow-up exam (1996 through 1998) on the 13,000 members of the ARIC cohort: time domain indices; mean heart rate, minimum and maximum heart rate, standard deviation of all normal R-R intervals, the coefficient of variation of all normal R-R intervals, root mean square of the differences of successive R-R intervals, and the proportion of adjacent R-R intervals. Frequency domain indices were also computed, including high frequency power, low frequency power, and the high/low frequency power ratio.

Residents of 4 geographically defined areas

Population-based sample of adults ages 45-64 years