Patient State Index: Titration of Delivery and Recovery from Propofol, Alfentanil, and Nitrous Oxide Anesthesia
Drover, David R. M.D.*; Lemmens, Harry J. M.D., Ph.D.†; Pierce, Eric T. M.D., Ph.D.‡; Plourde, Gilles M.Sc., M.D.§; Loyd, Gary M.D.∥; Ornstein, Eugene M.D., Ph.D.#; Prichep, Leslie S. Ph.D.**; Chabot, Robert J. Ph.D.††; Gugino, Laverne M.D., Ph.D.‡‡
Background: The Patient State Index (PSI) uses derived quantitative electroencephalogram features in a multivariate algorithm that varies as a function of hypnotic state. Data are recorded from two anterior, one midline central, and one midline posterior scalp locations. PSI has been demonstrated to have a significant relation to level of hypnosis during intravenous propofol, inhalation, and nitrous oxide–narcotic anesthesia. This multisite study evaluated the utility of PSI monitoring as an adjunct to standard anesthetic practice for guiding the delivery of propofol and alfentanil to accelerate emergence from anesthesia.
: Three hundred six patients were enrolled in this multicenter prospective randomized clinical study. Using continuous monitoring throughout the period of propofol–alfentanil–nitrous oxide anesthesia delivery, PSI guidance was compared with use of standard practice guidelines (both before [historic controls] and after exposure to the PSA 4000 monitor [Physiometrix, Inc., N. Billerica, MA; standard practice controls]). Anesthesia was always administered with the aim of providing hemodynamic stability, with rapid recovery.
: No significant differences were found for demographic variables or for site. The PSI group received significantly less propofol than the standard practice control group (11.9 μg · kg−1
< 0.01) and historic control group (18.2 μg · kg−1
< 0.001). Verbal response time, emergence time, extubation time, and eligibility for operating room discharge time were all significantly shorter for the PSI group compared with the historic control (3.3–3.8 min;P
< 0.001) and standard practice control (1.4–1.5 min;P
< 0.05 or P
< 0.01) groups. No significant differences in the number of unwanted somatic events or hemodynamic instability and no incidences of reported awareness were found.
Conclusions: Patient State Index–directed titration of propofol delivery resulted in faster emergence and recovery from propofol–alfentanil–nitrous oxide anesthesia, with modest decrease in the amount of propofol delivered, without increasing the number of unwanted events.
THE Patient State Index (PSI) was developed as a measure of hypnosis during anesthesia delivery. The PSI is based on quantitative electroencephalogram features, recorded from anterior and posterior scalp sites, as input to a multivariate algorithm that quantifies the most probable level of hypnosis. The PSI has a range from 0 to 100, with decreasing values indicating increasing levels of hypnosis. It has been demonstrated that the PSI has a significant relation to level of hypnosis, as measured by standard scales for quantifying level of alertness and sedation during intravenous propofol anesthesia as well as during inhalation anesthesia using isoflurane, sevoflurane, and desflurane. 1
The PSI algorithm was constructed following a systematic study of the quantitative electroencephalogram changes that accompanied loss and return of consciousness and the most probable underlying sources of those changes, leading to hypotheses about the role of cortical and subcortical structures–systems in the maintenance of the conscious state and the development of the algorithm (Appendix
The sensitivity of this index is in part due to the use of a self-norming technique used in the development. This method, called neurometrics, takes into account differences in individual background electroencephalograms as well as individual variability of the brain's response to anesthetic agents, 5,6
thereby enhancing sensitivity to change by reducing variance within each individual patient.
The use of electroencephalogram-derived multivariate measures of level of hypnosis to guide propofol delivery during surgery has been previously reported using a bispectral index (BIS). 7–9
We describe a multiple-site prospective study of the clinical utility of PSI as a monitor of hypnosis level during surgical procedures using an intravenous propofol-based anesthetic regimen. Specifically, the aim of this study was to determine if monitoring of hypnosis using the PSI could reduce drug dosage and hasten recovery without increasing incidence of adverse events.
Materials and Methods
This study was a multicenter (six centers) prospective randomized clinical study in which PSIs were recorded continuously throughout the period of anesthesia delivery. The efficacy of guiding the conduct of general anesthesia using the PSI was compared with anesthesia administered according to standard practice guidelines. A total of 306 American Society of Anesthesiologists physical status I–III consenting adult patients of either gender were enrolled and completed the study. Institutional review board approval and signed informed consent were obtained at each institution. Inclusion criteria were as follows: males and females between the ages of 18 and 80 yr, American Society of Anesthesiologists physical status I–III, who were undergoing elective surgical procedures scheduled for at least 30 min and who were able to read the consent form. Patients were excluded if they had scalp or skull abnormalities (psoriasis, eczema, angioma, scar tissue, burr holes, cranial implants), a history of head injury with loss of consciousness, psychiatric or supraspinal neurologic disorders, insulin-dependent diabetes mellitus, renal or hepatic disease, uncontrolled hypertension (systolic blood pressure > 160 mmHg, or diastolic pressure > 105 mmHg), a body weight greater than 150% of ideal weight for height, or if they were scheduled to undergo cardiac, vascular, or cranial neurosurgical procedures. Each site enrolled between 35 and 70 patients.
The PSA 4000 (Physiometrix, Inc., N. Billerica, MA) was used for acquiring electroencephalogram and displaying the PSI values. A disposable electroencephalogram electrode appliance (PSArray 1
) was affixed after arrival in the preoperative holding unit. The appliance is composed of a self-adjusting flexible head strap which serves to position and hold the Ag/AgCl electroencephalogram recording electrodes at the internationally defined electroencephalogram recording locations of FP1, FPZ, 1
Cz, Pz, and ground (FP2), referenced to linked ears. The data were bandpass filtered to 0.5–70 Hz and sampled at a rate of 250 Hz. Recording electrode impedance was continuously monitored and alarmed when values were greater than 15°K ohms (average values were < 7°K ohms for all electrode sites). Ultra-isolation, optimized filtering, oversampling, and shielding were used to minimize sensitivity to electromagnetic interference in the operating room (OR) environment. The PSA 4000 can display the electroencephalogram traces acquired from all recording sites, continuously updated current PSI values, and color-coded trajectories of PSI values across time.
A total of 347 patients were enrolled in the study. The initial 44 patients were assigned to a historical control (HC) group. This group was used to establish standard clinical practice of the participating anesthesiologists prior to any experience with intraoperative electroencephalogram guidance using the PSA 4000. The anesthesiologists were blinded to the PSI values obtained during the administration of anesthesia to this group of patients. The next 58 patients were enrolled in the training group. For this group, anesthesiologists administered anesthesia using standard practice guidelines but were allowed to view the PSA 4000 display. The purpose of this group was to familiarize the anesthesiologists with the PSA 4000 and to demonstrate how to use the PSI for guidance as defined in the research protocol. The remaining 245 patients were randomly allocated to either a standard practice control (SPC) group (n = 122) or a PSI-monitored group (n = 123). For the SPC group, anesthesiologists were instructed to adjust the dose of propofol according to standard practice guidelines and were blinded to the PSI information. Anesthetics administered to patients enrolled in the PSI group were guided by the PSI measure, as described below. Table 1
shows the distribution of these patients by site for enrollment and completion of the protocol. A total of 306 patients (HC, n = 35; training group, n = 47; SPC, n = 112; PSI, n = 112) completed the protocol and are reported on herein. Table 1
gives the reasons for noncompletion by site. Demographic data for the 259 patients in the 3 groups analyzed in the study (HC, SPC, and PSI) are presented in table 2
All patients studied received a propofol–alfentanil–nitrous oxide anesthetic, following the protocol described by Gan et al
Propofol and alfentanil were administered using a Harvard Medical dual infusion pump (Harvard Apparatus Inc., South Natick, MA). Noninvasive blood pressure, electrocardiogram, and any other surgically indicated vital signs were obtained.
All patients received 1–2 mg intravenous midazolam and a ≥ 500-ml bolus of lactated Ringer's solution. Continuous electroencephalogram monitoring was initiated after the patient was transferred to the OR; electroencephalogram recording was performed for about 5 min prior to induction of anesthesia and continued until the patient responded to command at the end of the anesthetic (and further when possible).
Anesthesia was induced with 1–2 mg/kg of propofol and ≤ 30 μg/kg of alfentanil. During induction, the patient's arousal level was assessed every 15 s using a modified Observer Assessment of Alertness and Sedation scale (OAA/S), 10
shown in table 3
. After loss of consciousness, patients breathed via
a laryngeal mask airway or a muscle relaxant was administered and an endotracheal tube placed.
The maintenance phase of anesthesia was initiated using 140 μg · kg−1 · min−1 of propofol, 0.5 μg · kg−1 · min−1 of alfentanil, and nitrous oxide at 50%. Muscle relaxation, used at the discretion of the anesthesiologist, was adjusted to maintain a minimum of two out of four twitches as judged by a train-of-four monitor. The further conduct of each anesthetic depended on the study group to which the patient was assigned.
In the SPC group, hypertension, tachycardia, or somatic signs of inadequate anesthesia during the maintenance phase were managed with increased dosages of alfentanil, propofol, or an antihypertensive at the discretion of the anesthesiologist. Hypotension and bradycardia were managed by appropriate dose reductions, adjustments of fluid status, or other needed pharmacologic agents.
In the PSI group, the anesthesiologists were instructed to adjust the infusion of propofol to maintain a PSI target range between 25 and 50. Signs of inadequate analgesia were to be treated with increased doses of alfentanil. Hypotension was treated by decreasing the dose of alfentanil. The administration of other pharmacologic agents was the same as in the SPC group.
Approximately 15 min prior to the anticipated completion of the surgical procedure, anesthesia was reduced in both the SPC and the PSI groups, at the anesthesiologist's discretion, to achieve rapid recovery. At this time, in the PSI group, the alfentanil was discontinued and the propofol infusion rate was adjusted to maintain a PSI of 50–60. This period of time was defined as the emergence phase and extended to 5 min prior to the completion of surgery. At this time, both propofol and nitrous oxide were discontinued for both the SPC and PSI groups in order to facilitate patient arousal and extubation.
In the HC group, the anesthesiologist adjusted propofol, nitrous oxide, and alfentanil delivery based on standard clinical assessments and practice for induction, maintenance, and emergence.
Intraoperative Events and Endpoints
During the maintenance and emergence phases of anesthesia for both the SPC and PSI groups, unwanted somatic and autonomic responses were recorded as defined in table 4
. After noting each occurrence, unwanted responses were treated with changes in the alfentanil or propofol infusion rates as described in the protocol above. The use of nonanesthetic agents was permitted only after changes in the anesthetic agents failed to return the unwanted autonomic responses to the accepted range while maintaining the PSI within the prescribed limits.
The recovery phase of anesthesia was defined as beginning after the discontinuation of propofol and nitrous oxide and extended to eligibility for discharge from the postanesthetic care unit (PACU). Anesthetic recovery was assessed by recording the following efficacy endpoints: (1) verbal command response time, defined as the length of time between the end of propofol infusion and when the patient first responds to simple verbal commands (such as, “squeeze my fingers”); (2) emergence time, defined as the length of time between the end of propofol infusion and when the patient responded to the command to open eyes; (3) extubation time, defined as the length of time between the end of propofol infusion and tracheal extubation; (4) OR recovery time, defined as the time interval between the end of propofol infusion and initial eligibility for transfer to the PACU (based on achieving a score of 9 or better using the Aldrete scoring system 11
); and (5) PACU recovery time, defined as the interval between the end of propofol infusion and eligibility for PACU discharge. Eligibility for PACU discharge was based upon reaching acceptable levels of pulse rate, blood pressure, respiration rate, oxygen saturation, cardiac rhythm, temperature, alertness, nausea, and pain. Memories of intraoperative events or reports of awareness were queried at follow-up performed at 24 h (± 12 h). Total amounts of alfentanil and propofol delivered to each patient during the maintenance phase of the anesthetic delivery and numbers of unwanted somatic and hemodynamic events were also recorded.
Sample size estimates were based upon the reduction in the duration between the end of surgery (discontinuation of anesthesia) and the first time the patient responds to commands, in patients actively monitored with the device, compared with patients whose device results were not monitored. With 200 total patients (100 patients per treatment arm), the power to detect a 4-min reduction in time to respond to commands (SD of 6 min 40 s) between the two treatment arms would be greater than 90%, assuming a two-sided hypothesis testing and a type I error rate of 5%. Given a potential nonevaluability–dropout rate of 20%, 240 patients (120 per treatment arm) were randomized in the study. Mean values and standard deviations were presented for each of the demographic values for the HC, SPC, and PSI groups of patients. Three-way analyses of variance were used to test for differences in demographic features between the three groups of patients. Three- and two-way analyses of variance were used for group comparisons for all anesthesia efficacy endpoints defined for this study. Duncan multiple comparisons were used to compare pairs of groups when three-way analyses of variance were significant. For all comparisons, P < 0.05 was considered statistically significant.
No significant site differences or group by site interactions were found for any of the endpoint measures. Therefore, all analyses were done collapsed across sites.
Demographic data for the 259 study patients used in the statistical analyses are presented in table 2
. No significant differences in any of these measures were found across the three groups of patients.
shows the efficacy and recovery endpoints by patient group. Verbal response time, emergence time, extubation time, and eligibility for OR discharge time were all found to be significantly shorter for the PSI-monitored group as compared with both the HC (P
< 0.001) and SPC (P
< 0.05 or P
< 0.01) groups. In comparing patients in the SPC and HC groups only, extubation time reached significance (P
< 0.05). The normalized propofol infusion rates were less in the PSI group than in the HC (P
< 0.01) or SPC (P
< 0.01) groups, which did not differ significantly from each other (P
≤ 0.37). The alfentanil infusion rates used within each group were not significantly different from each other (P
≤ 0.72). PACU discharge times did not reach statistical significance (P
≤ 0.72). Figure 1
shows the cumulative percentage of patients who reached verbal response endpoint as a function of time since the end of propofol infusion, separated by each group. The PSI group reached 100% (by 10 min) before the other two groups, and the SPC group reached 100% (by 20 min) before the HC group (more than 45 min).
As can be seen in table 6
, there were no significant differences found between the three patient groups in the number of unwanted somatic or hemodynamic events that occurred during the course of anesthesia. There was no incidence of reported awareness or memories in any patient in any group.
This study demonstrates that guidance for the titration of propofol infusion resulted in significantly decreased propofol use when compared with both HC and SPC patient groups. Further, PSI monitoring resulted in significantly faster recovery from anesthesia, defined in terms of the amount of time needed before consciousness was regained and when the patient was responsive enough to be extubated and eligible for discharge from the OR. Thirty-seven percent of the PSI group had verbal response times less than 5 min after end of propofol infusion, compared with 26% in the SPC group and 9% in the HC group. The differences between groups can more clearly be seen in figure 1
. These endpoints were achieved without increasing unwanted somatic events or hemodynamic instability. In fact, although not significant, it can be seen in table 6
that the PSI-monitored patients showed a tendency toward fewer “unwanted movements” (38% less) during maintenance than the SPC group, suggesting that guided delivery of anesthesia contributed to reduction of such unwanted events during maintenance. In general, these findings replicate earlier results utilizing the BIS index described by Gan et al
and indicate that multivariate descriptors of electroencephalogram-derived features other than BIS can be used to achieve this goal.
The significant, though modest, group difference seen in this study should be considered in the context of “learning contamination bias.” Previous studies 7,12
have emphasized that prior learning can play a significant role in the introduction and testing of new monitoring technologies. Several of the investigators in the present study had experience in anesthetic delivery monitoring utilizing the BIS index prior to initiating the present study. This may be reflected in comparisons of emergence time endpoints found for HC, SPC, and BIS–PSI-monitored groups of patients and also between the present study and data presented by Gan et al
). It can be seen in the present study that HC emergence times were substantially less than those obtained by Gan et al
. and were essentially equivalent to their SPC time endpoints. The SPC emergence time endpoints in the present study were less than those Gan et al
reported, while the emergence endpoints obtained after either PSI of BIS monitoring appear to be equivalent. (Note: statistical significance could not be assessed for this table since the standard deviations were not available for the BIS group.) However, the fact that significant findings were obtained despite possible prior learning effects suggests the possible clinical utility of such monitoring.
In both our study and that by Gan et al
specific criteria for decreasing propofol infusion rates were followed in the guided but not the control group. During a fixed infusion rate, propofol plasma concentrations will increase for several hours until steady state is achieved. Systematic failure to reduce the infusion rate during prolonged periods of adequate anesthesia in the SPC group may have resulted in higher propofol concentrations than necessary and consequently a delayed emergence. The introduction of electroencephalogram-based monitoring systems to provide such guidance may reduce the amount of anesthetic used and decrease recovery time when used as an adjunct to current standard clinical practice.
The authors acknowledge the efforts of Lori Gilmartin, R.N. (Boston University Medical Center, Boston, Massachusetts), Sujata Reast, M.S. (Stanford Medical Center, Stanford, California), Louise Ullyatt, R.N. (Royal Victoria Hospital, Montreal, Quebec, Canada), and all of the other site coordinators, in the acquisition of the data. We thank Dawn Frazer (V.P., Regulatory Affairs and Quality Assurance, Physiometrix, Inc., Five Billerica Park, North Billerica, Massachusetts) for her efforts in implementing the protocol, coordination of study centers, and the coordination of collection and tabulation of the data. We also gratefully acknowledge the contribution of E. Roy John, Ph.D. (Professor of Psychiatry, New York University School of Medicine, NY, New York) for his work in the development of this methodology. Dr. E. Roy John holds patents related to certain aspects of the quantitative electroencephalogram analysis techniques used in the data processing. These patents are assigned to NYU School of Medicine (NYUSM) and licensed by them to Physiometrix, Inc. NYUSM receives royalties under this agreement, part of which are shared with the inventor (as specified in the Faculty handbook).
Appendix: PSA 4000 Algorithm
is a schematic of the primary steps of PSA algorithm development. Three databases were fundamental to the development of the algorithm: EEG Library (Database I), Surgical Cases (Database II), and Volunteer Calibration Study (Database III).
Database I was the electrophysiological database previously established in the Brain Research Laboratories of New York University School of Medicine. This database contained approximately 20,000 raw and quantitative electroencephalograms recorded from normal, psychiatric, neurologic, and surgically monitored patients. 13
Statistical descriptors have been computed for the age-dependent normative distributions of numerous quantitative measures (quantitative electroencephalogram) extracted from these recordings. These normative values have been shown to have high reliability, specificity, sensitivity, and significantly high test–retest replicability, as well as no ethnic or cultural bias. The Brain Research Laboratories database also provided a library of the characteristics of artifact, nonelectroencephalogram signals that were used to develop algorithms for automatic identification and real-time exclusion of artifact from subsequent data processing.
Database II was derived from continuous electroencephalogram recordings made in a large population of patients undergoing general anesthesia with a variety of agents, administered under standard clinical practice. Each procedure was carefully documented for anesthetic, hemodynamic, and surgical events. In each case, quantitative electroencephalogram features were extracted under defined states and conditions. These features (approximately 2,500 per session) included spectral and bispectral measures of power and coherence and were used to form a new database documented for state of consciousness. Systematic exploration of this database was used to identify a subset of features that changed in an invariant way with deepening levels of sedation–hypnosis and loss of consciousness and reversed with return of consciousness. 3
Candidate measures were further explored using multivariate analysis to form mathematical classifier functions estimating the most probable level of consciousness. Candidate classifier functions were evaluated retrospectively for correlates of depth of sedation–hypnosis, as defined by the attending anesthesiologist. Details of the construction of this database and the retrospective analysis are given elsewhere. 1
Database III was derived from electroencephalogram recordings made during 64 procedures in which various anesthetics were administered incrementally (0.1 minimum alveolar concentration steps to loss of consciousness and return of consciousness) to healthy volunteer subjects. Quantitative electroencephalogram and clinical measures extracted from this database were used to assist in the calibration of the PSI.
The basic steps in the computation of the PSI are shown in figure 3
. Electroencephalogram recordings are collected from two anterior (FPI and FPZ’), a midline central (Cz), and a midline posterior (Pz) scalp location, referenced to linked ears, utilizing circuitry optimized to exclude electrical contamination from the environment. The signals are sampled at 2,500 samples per second per channel, bandpass filtered to 0.5–70 Hz, and then decimated to 250 samples per second per channel, in accordance with the Nyquist theorem.
Electroencephalogram signals are then processed by a series of artifact detection algorithms, allowing the identification of artifact-free “epochs” of electroencephalogram data. An additional algorithm detects electroencephalogram suppression, excludes these data from further algorithmic processing, and is used to compute a suppression ratio. This ratio can be monitored via the instrument's user interface and is taken into account by the PSI algorithm.
Electroencephalogram data in 1.25-s epochs is frequency transformed into subbands, including δ, θ, α, β, γ, and total power (0.5–50 Hz).
The set of features found to account for most statistical variance related to hypnotic state (selected from studies reviewed above) were derived for input to the discriminant algorithm (proprietary). These features included measures such as:
* absolute power gradient between frontopolar and vertex regions in γ
* absolute power changes between midline frontal and central regions in β and between midline frontal and parietal regions in α
* total spectral power in the frontopolar region
* mean frequency of the total spectrum in midline frontal region
* absolute power in δ at the vertex
* posterior relative power in slow δ
All features used in the computation of the PSI are transformed to obtain Gaussianity, an essential step in conforming to the statistical assumptions necessary for legitimate interpretation of the multivariate statistics utilized in computation of the PSI.
Every element in the set of critical features is transformed to a standard score (Z-score) relative to its distribution in a specific reference state and expressed as the probability of deviation from that state. The current values of these standardized scores are the inputs to the constantly updating calculation of the PSI value. The PSI is the ratio of the probability that the observation belongs to the reference state versus the sum of the probabilities that the observation belongs to either the reference state or to a different level of arousal. Thus, the PSI value can range from 0 to 100.
Two “observer” functions, sensitive to the suppression ratio and to abrupt changes in level of sedation, modulate the PSI algorithm. For optimal sensitivity, these observers are “self-normed” relative to the individual patient's baseline. These functions and the adjusted PSI are updated every 1.25 s. These updates are filtered and decimated into a sliding window, in a manner appropriate for optimal viewing characteristics, leading to an update of the global PSI trend every 6.4 s. Cited Here...
1. Prichep LS, John ER, Gugino LD, Kox W, Chabot R: Quantitative EEG assessment of changes in the level of the sedation/hypnosis during surgery under general anesthesia: I. The Patient State Index (PSI), II. Variable Resolution Electromagnetic Tomography (VARETA), Memory and Awareness in Anesthesia, IV edition. Edited by Jordan C, Vaughan DJA, Newton DEF. London, Imperial College Press, 2000, pp 97–107
2. John ER: A field theory of consciousness. Conscious Cogn 2001; 10: 184–213
3. John ER, Prichep LS, Kox W, Valdes-Sosa P, Bosch-Bayard J, Aubert E, Tom M, di Michele F, Gugino LD: Invariant reversible QEEG effects of anesthetics. Conscious Cogn 2001; 10: 165–83
4. Gugino L, Chabot RJ, Prichep LS, John ER, Formanek V, Aglio LS: Quantitative EEG changes associated with loss and return of consciousness in healthy adult volunteers anesthetized with propofol or sevoflurane. Br J Anaesth 2001; 87: 421–8
5. John ER, Prichep LS, Friedman J, Easton P: Neurometrics: Computer assisted differential diagnosis of brain dysfunctions. Science 1988; 239: 162–9
6. Chabot RJ, Gugino LD, Aglio LS, Maddi R, Cote W: QEEG and Neuropsychological profiles of patients after cardiopulmonary bypass surgical procedures. Clin EEG 1997; 28: 98–105
7. Gan TJ, Glass PS, Windsor A, Payne F, Rosow C, Sebel P, Manberg P, BIS Utility Study Group: Bispectral index monitoring allows faster emergence and improved recovery from propofol, alfentanil, and nitrous oxide anesthesia. A nesthesiology 1997; 87: 808–15
8. Glass PS, Bloom M, Kearse L, Rosow C, Sebel P, Manberg P: Bispectral analysis measures sedation and memory effects propofol, midazolam, isoflurane, and alfentanil in healthy volunteers. A nesthesiology 1997; 86: 836–47
9. Sebel PS, Lang E, Rampil IJ, White PF, Cork R, Jopling M, Smith NT, Glass PS, Manberg P: A multicenter study of bispectral electroencephalogram analysis monitoring anesthetic effect. Anesth Analg 1997; 84: 891–9
10. Chernik DA, Gillings D, Laine H, Hendler J, Silver JM, Davidson AB, Schwam EM, Siegel JL: Validity and reliability of the Observer's Assessment of Alertness/Sedation scale: Study with intravenous midazolam. J Clin Psychopharmacol 1990; 10: 244–51
11. Aldrete JA, Kroulik DA: A postanesthetic recovery score. Anesth Analg 1970; 49: 924–34
12. Roizen MF, Toledano A: Technology assessment and the “learning contamination” bias. Anesth Analg 1994; 79: 410–2
13. John ER, Prichep LS, Easton P: Normative data banks and neurometrics: Basic concepts, methods and results of norm construction, Handbook of Electroencephalography and Clinical Neurophysiology, vol. I. Edited by Geveins AS, Remond A. Amsterdam, Elsevier, 1987, pp 449–95
This article has been cited 50 time(s).
British Journal of AnaesthesiaDetection of awareness in surgical patients with EEG-based indices - bispectral index and patient state indexBritish Journal of Anaesthesia
Canadian Journal of Anaesthesia-Journal Canadien D Anesthesie
Bispectral index monitoring does not improve early recovery of geriatric outpatients undergoing brief surgical procedures
Canadian Journal of Anaesthesia-Journal Canadien D Anesthesie, 53(1):
Guidelines for the use of deep sedation and anesthesia for GI endoscopy
Gastrointestinal Endoscopy, 56(5):
Anesthesia and AnalgesiaPeripheral nerve blocks result in superior recovery profile compared with general anesthesia in outpatient knee arthroscopyAnesthesia and Analgesia
Clinical NeurophysiologyRelation of EEG spectrum progression to loss of responsiveness during induction of anesthesia with propofolClinical Neurophysiology
Training guideline for use of propofol in gastrointestinal endoscopy
Gastrointestinal Endoscopy, 60(2):
Anesthesia and AnalgesiaA comparison of state and response entropy versus bispectral index values during the perloperative periodAnesthesia and Analgesia
Anasthesiologie Intensivmedizin Notfallmedizin Schmerztherapie
Neuromonitoring for estimating the depth of narcosis
Anasthesiologie Intensivmedizin Notfallmedizin Schmerztherapie, 39(1):
Anesthesia and AnalgesiaThe effect of cerebral monitoring on recovery after general anesthesia: A comparison of the auditory evoked potential and bispectral index devices with standard clinical practiceAnesthesia and Analgesia
Fluctuations and Noise in Biological, Biophysical, and Biomedical Systems IIConsciousness from neurons and wavesFluctuations and Noise in Biological, Biophysical, and Biomedical Systems II
Acta Anaesthesiologica ScandinavicaPitfalls and challenges when assessing the depth of hypnosis during general anaesthesia by clinical signs and electronic indicesActa Anaesthesiologica Scandinavica
Icsp: 2008 9Th International Conference on Signal Processing, Vols 1-5, Proceedings
Extraction of Bis (Tm) Index Sub-Parameters in Different Anesthetic and Sedative Levels
Icsp: 2008 9Th International Conference on Signal Processing, Vols 1-5, Proceedings, ():
2009 9Th International Conference on Information Technolology and Applications in Biomedicine
Estimating the Depth of Anesthesia by Applying Sub Parameters to an Artificial Neural Network during General Anesthesia
2009 9Th International Conference on Information Technolology and Applications in Biomedicine, ():
International Journal of PsychophysiologyThe sometimes pernicious role of theory in scienceInternational Journal of Psychophysiology
Anesthesia and AnalgesiaAssessing propofol induction of anesthesia dose using bispectral index analysisAnesthesia and Analgesia
Anaesthesia Pain Intensive Care and Emergency Medicine - A.P.I.C.E, Vol 1 and 2
Awareness during anaesthesia: incidence and clinical and legal aspects
Anaesthesia Pain Intensive Care and Emergency Medicine - A.P.I.C.E, Vol 1 and 2, ():
Canadian Journal of Anaesthesia-Journal Canadien D Anesthesie
Sedation after spinal anesthesia in elderly patients: a preliminary observational study with the PSA-4000
Canadian Journal of Anaesthesia-Journal Canadien D Anesthesie, 51(6):
Fractals-Complex Geometry Patterns and Scaling in Nature and SocietyQuantification of Depth of Anesthesia By Means of Adaptive Calculation of Correlation Dimension ParametersFractals-Complex Geometry Patterns and Scaling in Nature and Society
Anesthesia and AnalgesiaPatient state index during a cardiac arrest in the operating roomAnesthesia and Analgesia
Mayo Clinic Proceedings
Comparison of electrophysiologic monitors with clinical assessment of level of sedation
Mayo Clinic Proceedings, 81(1):
Journal of Clinical AnesthesiaTracheal resection in the morbidly obese patient: the role of dexmedetomidineJournal of Clinical Anesthesia
Journal of AnesthesiaMechanisms of anesthetic actions and the brainJournal of Anesthesia
Gastrointestinal EndoscopySedation and anesthesia in GI endoscopyGastrointestinal Endoscopy
Anesthesia and AnalgesiaIs the patient state analyzer with the PSArray(2) a cost-effective alternative to the bispectral index monitor during the perioperative period?Anesthesia and Analgesia
British Journal of AnaesthesiaThe Patient State Index as an indicator of the level of hypnosis under general anaesthesiaBritish Journal of Anaesthesia
Anaesthesia and Intensive Care
Recovery from bispectral index-guided anaesthesia in a large randomized controlled trial of patients at high risk of awareness
Anaesthesia and Intensive Care, 33(4):
Pediatric AnesthesiaMeasuring anesthesia in children using the EEGPediatric Anesthesia
Ieee Transactions on Biomedical EngineeringForecasting the unresponsiveness to verbal command on the basis of EEG frequency progression during anesthetic induction with propofolIeee Transactions on Biomedical Engineering
Gastrointestinal EndoscopySedation and anesthesia in GI endoscopy (Reprinted from Gastroenterology, vol 133, pg 675-501, 2007)Gastrointestinal Endoscopy
2008 3Rd International Conference on Information and Communication Technologies: From Theory to Applications, Vols 1-5
Electroencephalogram Fractral Dimension as a Measure of Depth of Anesthesia
2008 3Rd International Conference on Information and Communication Technologies: From Theory to Applications, Vols 1-5, ():
Proceedings of the 26Th Annual International Conference of the Ieee Engineering in Medicine and Biology Society, Vols 1-7
Higuchi fractal dimension and spectral entropy as measures of depth of sedation in intensive care unit
Proceedings of the 26Th Annual International Conference of the Ieee Engineering in Medicine and Biology Society, Vols 1-7, 26():
British Journal of AnaesthesiaDepth of anaesthesia monitoring: what's available, what's validated and what's next?British Journal of Anaesthesia
AnaesthesistMeasurement of the depth of anaesthesiaAnaesthesist
Neuroscience LettersTime-frequency properties of electroencephalogram during induction of anesthesiaNeuroscience Letters
Neural PlasticityA Plastic Temporal Brain Code for Conscious State GenerationNeural Plasticity
Acta Anaesthesiologica Scandinavica
Time-frequency balanced spectral entropy as a measure of anesthetic drug effect in central nervous system during sevoflurane, propofol, and thiopental anesthesia
Acta Anaesthesiologica Scandinavica, 48(2):
European Journal of Anaesthesiology (EJA)Light or deep general anaesthesia: does it matter and how to assess it?European Journal of Anaesthesiology (EJA)
European Journal of Anaesthesiology (EJA)EEG-based indices of anaesthesia: correlation between bispectral index and patient state index?European Journal of Anaesthesiology (EJA)
Journal of Neurosurgical AnesthesiologyDoes Using the Bispectral Index (BIS) During Craniotomy Affect the Quality of Recovery?Journal of Neurosurgical Anesthesiology
© 2002 American Society of Anesthesiologists, Inc.
Publication of an advertisement in Anesthesiology Online does not constitute endorsement by the American Society of Anesthesiologists, Inc. or Lippincott Williams & Wilkins, Inc. of the product or service being advertised.