Saturation of high-frequency oscillations of R-R intervals in healthy subjects and patients after acute myocardial infarction during ambulatory conditions

This study was designed to assess the relationship between R-R interval length and heart rate (HR) variability in healthy subjects and patients after an acute myocardial infarction (AMI). Twenty-four-hour ambulatory ECG recordings were obtained for 76 healthy subjects and 82 post-AMI patients. The high-frequency (HF, 0.15-0.4 Hz) spectral power of R-R intervals was analyzed in 5-min sequences over 24 h and plotted as a function of the corresponding mean R-R interval length. Quadratic regression model was used to study the relationship between R-R interval length and HF power. If a distinct deflection point (R-R0) occurred in the quadratic regression (r >0.50) model before maximum R-R interval, indicating the plateau of HF power, the relationship between R-R interval and HF power was defined as saturated. Otherwise, the relationship was defined as linear (r >0.50) or low correlated (r >0.50). The relationship was saturated in 35, linear in 38, and low correlated in 3 healthy subjects. In post-AMI patients, the relationship was saturated in 9 subjects, linear in 44 subjects, and low correlated in 29 patients. The HF power analyzed from the 24-h period did not differ between the saturated and linear groups, but when analyzed from the linear portion only, HF spectral power was smaller in the linear than the saturated group both among healthy subjects (P <0.05) and post-AMI patients (P <0.05). Saturation of the HF oscillations of R-R intervals is a common phenomenon in healthy subjects and also present in post-AMI patients during ambulatory conditions. This saturation effect may bias the quantification of cardiac vagal function when HR variability is analyzed from Holter recordings.