Traditional and advanced electrocardiographic measures of autonomic function in the population-based KORA-F3 study.

AIMS

Heart-rate variability (HRV) measures are surrogates of autonomic function at the level of the sinus node and have evolved as markers of cardiovascular mortality in patients after myocardial infarction (MI). Traditionally, HRV is assessed in time-domain and frequency domain. Advanced measures of autonomic function include deceleration capacity (DC) and periodic repolarization dynamics (PRD). DC predominantly quantifies the influence of parasympathetic tone. PRD captures low-frequency oscillations of repolarization instability and is considered to reflect sympathetic activity at the level of the left ventricular myocardium. However, population-based reference values are missing.

METHODS AND RESULTS

In 505 participants of the population-based KORA F3 study (Cooperative Health Research in the Region of Augsburg) with extant digital 24-h Holter electrocardiograms we assessed markers of HRV in time and frequency domains. Additionally, we determined advanced measures of autonomic function including DC and PRD applying previously established technologies. We used standard, pre-defined cut-off values to define high-risk groups. The cohort's mean age was 63.6 ± 5.5 years, and 256 (50.1%) were women. Among HRV measures, exemplarily the median standard deviation of all normal-to-normal intervals (SDNN) was 141 ms [119;165] and the median low frequency to high frequency ratio (LF/HF-ratio) was 3.92 [2.69;6.18]. Regarding autonomic function, median DC was 5.32 ms [2.69;6.18], and median PRD was 2.92 ms [2.06;4.14]. Among these measures LF/HF-ratio was significantly higher among men (5.15 [3.23; 7.20]) than women (3.37 [2.36;4.53], p < 0.001). Measured distribution is also provided in a cohort subset without overt cardiovascular conditions. While DC decreased with age, SDNN, LF/HF-ratio, and PRD were stable across age-groups. For participants with comorbidities including hypertension, intake of betablockers, history of MI, stroke, or diabetes mellitus significantly lower SDNN, LF/HF-ratio, and DC were observed.

CONCLUSION

In a large population-based cohort, we systematically present traditional and advanced measures of HRV of cardiac autonomic function. We report reference values in the overall cohort, as well as stratified by sex, age, and concomitant cardiovascular conditions.