Wavelet transform of heart rate variability to assess autonomic nervous system activity does not predict arousal from general anesthesia.

Abstract

The relationship between autonomic nervous system (ANS) activity and general anesthesia has been explored. Studies have demonstrated partial recovery of heart rate variability (HRV), representative of ANS activity, in the postoperative period, but the arousal period has not been precisely studied. The goals of this study were to analyze modifications of ANS activity during general anesthesia and, more particularly, around the arousal period, to look for predictors of arousal. We analyzed HRV changes using wavelet transform, a time-frequency analysis that, in contrast to Fourier transform, is able to assess abrupt changes of ANS activity. Seventeen patients (mean +/- SD age: 40.9 +/- 16.4 yr) under general anesthesia for hip or knee surgery, were included in the study. The analysis began one hour before anesthesia, focussed on eye opening, and ended three hours after arousal. There was a dramatic decrease in HRV after induction, that extended throughout anesthesia and represented a decrease in global autonomic regulation with, however, a relative predominance of vagal tone. At the moment of eye opening, there was an abrupt change in HRV, representing a sudden shift of ANS balance towards the predominance of sympathetic activity, while none of these indices changed seconds before arousal. Wavelet analysis of HRV appears to be powerful tool to precisely assess instantaneous changes of HRV during anesthesia. Using this method, there were no identifiable precursory HRV indices of arousal.