Respiratory sigh associated transient autonomic changes detected with a continuous wavelet method of heart rate variability analysis

Abstract

The measure of heart rate variability (HRV) from an EKG recording is a non-invasive way to evaluate autonomic nervous system (ANS) activity. Traditional methods use either nonparametric (Fourier techniques) or parametric (autoregressive) algorithms. These methods have poor temporal resolution of signal transients and require a stationary signal for accurate evaluation. The short time Fourier transform (STFT) has been used as an attempt to overcome this problem, but still a transient may be poorly localized to time and “blurred” within the signal. We present a Morlet continuous wavelet transform (CWT) algorithm for HRV analysis. A rat model was used of normal respirations with intermittent sighs. Both EKG and respiratory recordings were made. RR interval files were detrended and resampled at 10 Hz using a cubic spline algorithm. Wavelet scales were generated and converted to frequency bands corresponding to sympathetic (SNS) and parasympathetic (PNS) activity. A transient burst in mostly SNS but also PNS activity was noted leading up to a respiratory sigh. To our knowledge, this is the first time brief changes in ANS activity have been detected in association with a respiratory sigh. As demonstrated by this study, this method of HRV analysis may a useful method of investigating transient changes in ANS activity.