Short- and long-term reproducibility of time and frequency domain heart rate variability measurements in normal subjects

Objective: To obtain data relating to the reproducibility of the time and frequency domain measurements obtained from 10-min ECG recordings. Methods: Eighteen normal volunteers underwent evaluations of time and frequency domain heart rate variability 2 weeks and 7 months after baseline analysis. The time domain parameters were mean NN, the standard deviation of NN intervals, the percentage-of successive NN intervals > 50 ms and the root mean square successive difference of NN intervals. The frequency domain evaluations (total power, low frequency, and high frequency) were made by means of both the Fast Fourier Transform algorithm (FFT) and the E autoregressive method (AR) from 10-min ECG recordings made under three different conditions: rest, controlled respiration, and after a passive head-up tilt test. Reproducibility was evaluated by means of the interclass correlation coefficient (ICC), comparing baseline values with the results obtained at the second week and the seventh month. Time domain evaluation were also made from 10-min EGG. Results: All of the time domain measurements had an ICC ≤ 0.75, except for the standard deviation of NN intervals, which had an ICC of 0.57. The frequency domain parameters obtained by means of either FFT or AR showed similar reproducibility. Low frequency was reproducible under all three conditions, total power only at rest, and high frequency only during controlled respiration. Conclusion: The reproducibility of frequency domain parameters depends on the analysed condition. These results are of primary importance when the effects of drugs or other interventions on heart rate variability are under investigation.