Multichannel spectral analysis of R-R and respiratory series from 24-hour Holter recordings

The influence of respiration on power spectral analysis of heart rate variability is often ignored. The availability of simultaneous 24-hour recordings of ECG and respiratory activity can provide important information on the possible correlation between the two signals. A spectral multichannel autoregressive method (Nuttall-Strand algorithm) was applied to R-R interval and respiratory time series obtained from 24-hour recordings. Fifteen normal subjects and seven climbers, who reached the altitude of 5000 m, were submitted to the study. The coherence between R-R series and respiration in the LF and HF band was computed and compared in the two groups. As compared to normals, climbers showed a significantly higher coherence in the LF band-possibly because of the presence of periodic breathing