Previous Abstract | Next Abstract
Printable Version
EEG, Heart Rate, Pulse Plethysmography and Movement Responses to Skin Incision
E. Seitsonen, M.D.; M. van Gils, Ph.D.; I. Korhonen, Ph.D.; K. Korttila, M.D., F.R.C.A.; A. Yli-Hankala, M.D., Ph.D.
Anesthesia and Intensive Care, Helsinki University Central Hospital, Helsinki, Finland
Background: Evaluation of analgesia and nociception during general anesthesia is challenging. Heart rate (HR) variability has been proposed to indicate level of analgesia, but the role of cortical electroencephalography (EEG) is controversial. Peripheral pulse plethysmography (PPG) may be useful in detecting nociceptive responses mediated by the autonomic nervous system. We investigated the physiological response to skin incision and extracted the most relevant information characterizing probable nociception.

Patients and methods: The study protocol was approved by the local Ethics Committee, and written informed consent was obtained from the subjects. We analyzed data from 13 patients scheduled for abdominal hysterectomy. After induction with fentanyl 1 μg/kg iv and propofol 1 mg/kg iv anesthesia was deepened with sevoflurane 8% in 100% oxygen via facial mask. Controlled ventilation was started after endotracheal intubation. Surgery began 14 min after intubation in stable 0.8 MAC (1.6% end-tidal) sevoflurane anesthesia. Any signs of arousal at skin incision were registered. Propofol 50 mg iv was given as rescue medication if BIS™ exceeded 70. Raw EEG, the bispectral index™ (BIS™ Rev. 3.4), electrocardiography (ECG) and PPG data were collected and analyzed offline. RR-interval (RRI) tachogram was derived from ECG and frontal electromyography (fEMG) from EEG. Several beat-to-beat morphology parameters were derived from PPG signal. Various time and frequency domain parameters were computed from EEG, RRI and PPG data.

Results: Seven patients responded clearly to skin incision with either movement or coughing, while six patients displayed only hemodynamic reactions. When derived variables calculated as ratios or differences between post-incision (+2 min, X1) and pre-incision (-2 min, X0) values were compared, RRI, amplitude of the dicrotic notch in PPG, EEG spectral entropy and fEMG power appeared as primary variables in the optimal linear discriminant function (mover or nonmover). The mean (standard deviation) changes with skin incision for movers and nonmovers, respectively, were: RRI1/ RRI0 0.72 (0.08) vs. 0.87 (0.12); PPG notch amplitude1 - PPG notch amplitude0 -1.37 (0.77) vs. -0.18 (0.81); EEG entropy1/ EEG entropy0 1.13 (0.06) vs. 1.03 (0.08); fEMG1/ fEMG0 12.5 (17.3) vs. 1.0 (0.3). The performance of this model as determined by leave-one-out cross-validation was 77% (10/13) correct classification.

Conclusions: In this study, spectral entropy of EEG, fEMG power and instantaneous HR changed to a greater degree in response to supramaximal noxious stimulation, as indicated by clear arousal during anesthesia, than to submaximal noxious stimulation. The amplitude of the dicrotic notch in PPG tended to decrease with supramaximal and remain unchanged with submaximal stimulation. A combination of these parameters may be useful in assessing the level of analgesia and nociception during anesthesia.

Anesthesiology 2002; 96: A582