Abstract Analysis of beat-to-beat heart rate variability (HRV) provides information of cardiac vagal outflow to the sinus node. Some methodological problems might, however, be involved in the analysis of cardiac vagal outflow from ambulatory Holter recordings, such as saturation, physical activity, and abrubt prolongations of R-R intervals unrelated to respiration. The purpose of this thesis was to assess the physiological basis of beat-to-beat HRV and to develop and assess new methods for the quantification of cardiac autonomic modulation from ambulatory Holter recordings.

The study population consisted of 89 healthy volunteers (age 24 ± 4 years) and 590 patients with a recent acute myocardial infarction (AMI, age 61 ± 10 years). The relationship between R-R interval length and the high-frequency (HF) spectral power of the R-R intervals was assessed in 76 healthy subjects and 82 post-AMI patients. The effects of aerobic exercise training on the dynamics between R-R interval and HF power were evaluated by means of a controlled 8-week training intervention (n = 17). The effects of sympathetic activation and concomitant sympathetic and vagal outflow on beat-to-beat HR dynamics were studied in laboratory conditions (n = 13). A new method for quantifying beat-to-beat HRV from the R-R interval lengths where the relationship between HF power and R-R interval is most linear was developed to avoid the confounding effects of possible saturation, physical activity, and random R-R interval dynamics. The clinical significance of the new method was assessed in a series of 590 post-AMI patients.

Saturated HF R-R interval dynamics, expressed as a lack of increase in HF power despite an increased R-R interval, was observed in 35 healthy subjects and 9 post-AMI patients. In the training study, 7 subjects out of a total of 17 had saturated HF power before the intervention. After the training period, 5 new cases of saturated HF power were observed. In laboratory conditions, co-activation of sympathetic and vagal outflow resulted in random R-R interval dynamics. In post-AMI patients, HF power analyzed exclusively from the R-R intervals where the relationship between the R-R interval and HF power was most linear (Vindex) predicted independently the occurrence of SCD among post-AMI patients, while traditionally analyzed HF power did not.

In conclusion, the saturation of beat-to-beat HRV in ambulatory conditions is a common phenomenon. The prevalence of saturated HF power increases due to enhanced cardiac vagal outflow induced by aerobic training. Finally, the novel analysis of vagally mediated HRV (Vindex) provides unique information that cannot be obtained by traditional analysis of HF R-R interval dynamics.