Anesthesiology 2006; 104:708–17 © 2006 American Society of Anesthesiologists, Inc. Lippincott Williams & Wilkins, Inc. Spectral Entropy as a Measure of Hypnosis in Children Jaakko G. M. Klockars, M.D.,* Arja Hiller, M.D., Ph.D.,* Seppo Ranta, M.D., Ph.D.,† Pia Talja, M.Sc.,‡ Mark J. van Gils, Ph.D.,§ Tomi Taivainen, M.D., Ph.D.࿣ Background: The Datex-Ohmeda S/5 Entropy Module (Datex- or sedative drugs. These include certain physiologic Ohmeda Division, Instrumentarium Corp., Helsinki, Finland), signs,1 various clinical scales,2,3 and methods based on using time-frequency balanced Spectral Entropy, is a novel tool 4 for monitoring the hypnotic state during anesthesia. The En- the electroencephalogram. In adults, electroencephalo- 5 tropy Module produces two values, State Entropy (SE) and Re- gram-derived indices, such as evoked potential features, sponse Entropy (RE), and in adults, it has been shown to mea- the Bispectral Index,4,6,7 and the Spectral Entropy,8 have sure reliably the hypnotic effects of various drugs. In children, been shown to be indicators of the depth of hypnosis Spectral Entropy has been only preliminary studied. The au- Downloaded from http://pubs.asahq.org/anesthesiology/article-pdf/104/4/708/361152/0000542-200604000-00015.pdf by guest on 26 September 2021 thors’ aim was to study Spectral Entropy as a marker of hyp- during anesthesia. Benefits of monitoring the level of notic state during general anesthesia in infants and children. hypnosis during anesthesia include, e.g., faster recov- Methods: Twenty infants (aged 1 month–1 yr) and 40 children ery,9–11 reduced need for anesthetics,10–12 and better (aged 1–15 yr) were anesthetized for surgery using standardized individual guidance of anesthesia.7,13 sevoflurane-nitrous oxide–based anesthesia. The relationships The Datex-Ohmeda S/5 Entropy Module (M-Entropy; between SE, RE, or Bispectral Index (BIS) and (1) a modified Observer’s Assessment of Alertness/Sedation Scale, (2) non– Datex-Ohmeda Division, Instrumentarium Corp., Hel- steady state end-tidal concentration of sevoflurane, (3) steady sinki, Finland), using Spectral Entropy, is a novel tool for state end-tidal concentration of sevoflurane, and (4) hemody- monitoring the depth of hypnosis. It measures the reg- namic values were calculated using prediction probability, non- ularity of the electroencephalogram signal according to linear regression, and correlation coefficients, as appropriate. 14 The performances of SE, RE, and BIS were compared. an algorithm published previously, providing two val- Results: The prediction probability values (؎ SEM) of SE, RE, ues of entropy, the State Entropy (SE) and the Response and BIS versus the modified Observer’s Assessment of Alert- Entropy (RE). The Entropy Module measures the depth ,ness/Sedation Scale in the induction phase were 0.83 ؎ 0.06 of hypnosis by calculating the spectral entropy of the ؎ ؎ ؎ 0.88 0.06, and 0.87 0.08 for children and 0.76 0.08, 15,16 and 0.73 ؎ 0.10 for infants; values in the emergence electroencephalographic power spectrum. SE, with ,0.08 ؎ 0.79 -phase were 0.68 ؎ 0.05, 0.74 ؎ 0.04, and 0.64 ؎ 0.05 for chil- a scale of 0–91, is computed over the electroencepha ,dren and 0.64 ؎ 0.07, 0.69 ؎ 0.06, and 0.72 ؎ 0.06 for infants, logram-dominant frequency range of 0.8–32 Hz respectively. SE, RE, and BIS values were inversely proportion- whereas RE, with a scale 0–100, is computed over the ally related to the end-tidal concentration of sevoflurane for children, but for infants, the correlation was much less clear. electromyographic-dominant frequency of 0.8–47 Hz. No significant correlations were found between SE, RE, or BIS When there is no electromyographic activity present, SE values and the hemodynamic values. and RE show the same number. The recommended Conclusions: Spectral Entropy may be a useful tool for mea- range for adequate anesthesia for both Spectral Entropy suring the level of hypnosis in anesthetized children and seems to perform as well as BIS. In infants, the clinical usefulness of parameters is from 40 to 60, and zero-values are associ- both these electroencephalogram-derived methods must be ated with a completely suppressed electroencephalo- evaluated in further controlled studies. gram. Spectral Entropy has been validated to measure the depth of hypnosis in adult patients receiving inhaled SEVERAL methods have been introduced to reflect the or intravenous anesthesia.8,17–19 hypnotic state during anesthesia induced by anesthetic In pediatric patients, electroencephalogram-derived monitors, such as BIS,9,20–22 and auditory evoked poten- 23,24 * Staff Anesthesiologist, ࿣ Associate Professor and Head, Department of Anes- tials have been studied for the measurement of the thesiology and Intensive Care Medicine, Hospital for Children and Adolescents, depth of hypnosis during anesthesia. BIS has been vali- University of Helsinki. † Unit Manager, Healthcare Development Unit, HAUS 20–22 Finnish Institute of Public Management Ltd., Helsinki, Finland. ‡ Research dated in older children, but there are conflicting Engineer, Datex-Ohmeda Division, Instrumentarium Corp., Helsinki, Finland. data about the validation of BIS in infants.7,9,21,25–27 Until § Senior Research Scientist, VTT Information Technology, Tampere, Finland. now, Spectral Entropy has been only preliminarily stud- Received from the Department of Anesthesiology and Intensive Care Medi- 25,28 cine, Hospital for Children and Adolescents, University of Helsinki, Helsinki, ied in pediatric patients. Finland. Submitted for publication April 16, 2005. Accepted for publication January 4, 2006. Supported by the Biomedicum Helsinki Foundation, Helsinki, The aim of our study was to evaluate whether Spectral Finland, and Datex-Ohmeda Division, Instrumentarium Corp., Helsinki, Finland. Entropy reliably measures the depth of hypnosis in in- The authors wish to express sincere thanks to Warren D. Smith, Ph.D. (Professor, Department of Bioengineering, California State University, Sacramento, Califor- fants and children during sevoflurane-based general an- nia), for providing the PKMACRO software used in this study. Presented in part esthesia. The level of hypnosis was assessed with a at the Annual Meeting of the American Society of Anesthesiologists, Las Vegas, Nevada, October 23–27, 2004. Dr. Ranta is a medical advisor of GE Healthcare modified responsiveness category of the Observer’s As- Finland, Helsinki, Finland. Ms. Talja is a full-time paid employee of GE Healthcare sessment of Alertness/Sedation Scale (mOAA/S).2,3 We Finland, Helsinki, Finland. Dr. van Gils is a statistical consultant paid by GE Healthcare Finland, Helsinki, Finland. hypothesized that there should be, first, a correlation Address correspondence to Dr. Klockars: Department of Anesthesiology and between SE or RE values and the mOAA/S and, second, Intensive Care Medicine, Hospital for Children and Adolescents, University of Hel- a correlation between SE or RE values and the end-tidal sinki, PL 281, 00029 HUS, Finland. jaakko.klockars@hus.fi. Individual article reprints may be purchased through the Journal Web site, www.anesthesiology.org. concentration of sevoflurane. Secondarily, we also com- Anesthesiology, V 104, No 4, Apr 2006 708 SPECTRAL ENTROPY DURING PEDIATRIC ANESTHESIA 709 pared the performance of Spectral Entropy with BIS in 40 ␮g/kg intravenous glycopyrrolate and 10 ␮g/kg children of various ages. neostigmine was administered when needed (i.e., when a fade in the train-of-four stimulation was observed visu- ally) for reversal of residual neuromuscular blockade. Materials and Methods At the end of surgery, the inhalational agents nitrous oxide and sevoflurane were discontinued without any Patients and Study Design tapering, and the patient was extubated when sufficient After institutional ethics committee approval (Hospital spontaneous breathing had returned. Thereafter, the pa- for Children and Adolescents, Helsinki University, Hel- tient was transferred to the postanesthesia care unit. In sinki, Finland) and written informed consent of parents case of postoperative pain, 0.1 mg/kg morphine was were obtained, 20 infants and 40 children were studied. given repeatedly when needed. Downloaded from http://pubs.asahq.org/anesthesiology/article-pdf/104/4/708/361152/0000542-200604000-00015.pdf by guest on 26 September 2021 All patients had an American Society of Anesthesiologists In addition to Spectral Entropy and BIS values, end- physical status classification of I or II and were sched- tidal sevoflurane and nitrous oxide concentrations, elec- uled to undergo elective surgery during general anesthe- trocardiogram, heart rate, noninvasive blood pressure sia. The inclusion criteria were an age between 1 month (measured at 5-min intervals), capnogram, pulse oxime- and 15 yr and an elective surgical operation with an try saturation, and temperature (S/5 Anesthesia Monitor) estimated duration between 0.5 and 5 h, which required were monitored continuously during anesthesia and col- general anesthesia without regional anesthetics. Chil- lected with 5-s intervals on a laptop computer (Toshiba dren were excluded if they had a disease or medication Satellite; Toshiba Corp., Tokyo, Japan). affecting the central nervous system or if the surgery affected the head or the neck of the child. Electroencephalogram Acquisition The anesthesiologist in charge was blinded to Spectral Anesthetic Regimen Entropy and BIS monitoring throughout the study pe- Premedication with 0.3 mg/kg oral midazolam (maxi- riod, and an independent study nurse