SciELO - Scientific Electronic Library Online

vol.51 issue5Paresthesia at the median nerve territory: adverse effect of the peripheral nerve stimulator? Case reportBioethical dilemmas in anesthesia author indexsubject indexarticles search
Home Pagealphabetic serial listing  

Services on Demand




Related links


Revista Brasileira de Anestesiologia

Print version ISSN 0034-7094

Rev. Bras. Anestesiol. vol.51 no.5 Campinas Sept./Oct. 2001 



Monitoring the adequacy of hipnosis by bispectral index*


Monitor de profundidad de la hipnosis. La eletroencefalografia bispectral


Pedro Thadeu Galvão Vianna, TSA, M.D.

Professor Titular do CET/SBA do Departamento de Anestesiologia da FMB - UNESP





BACKGROUND AND OBJECTIVES: Monitoring the "depth" of anesthesia is a complex process. Most approaches to monitoring anesthetic adequacy involve EEG waves or, more recently, some EEG processed form. Bispectral analysis is a method which allows for EEG analysis in different frequency phases.
CONTENTS: Processed EEG starts with the digitalization of the EEG signal. The digitized EEG can then be mathematically transformed by a process known as Fourier analysis, which divides the complex EEG signal in a number of sine wave components, that is, in each portion of different amplitude, but whose sum corresponds to the original EEG waveform. Many parameters are derived from this method. Bispectral index, or simply BIS (100 = awaken to 0= EEG isoelectric) is derived from the best parameters (e.g. spectral edge frequency, median frequency and burst suppression) which are evaluated by statistical analysis.
CONCLUSIONS: Clinical experience has shown that BIS may predict a response to skin incision during anesthesia. However, BIS is not independent of the anesthetic technique. There are different responses depending on which hypnotics or analgesics are used.

Key words: MONITORING: depth of anesthesia, bispectral index


JUSTIFICATIVA Y OBJETIVOS: La supervisión de la profundidad de la hipnosis y de la anestesia es un acto complejo. La mayoría de las propuestas para supervisar los niveles adecuados de la hipnosis, durante la anestesia, envuelven el EEG usando las ondas del EEG, o más recientemente, usando la forma procesada. La análisis bispectral es el método que permite analizar el EEG en las diferentes fases de frecuencias.
CONTENIDO: El EEG procesado es iniciado con la conversión del señal de EEG para la forma digital. El EEG digitalizado puede ser matemáticamente transformado por el proceso conocido como análisis de Fourier, que separa el complejo señal del EEG en varios componentes de la onda, o sea, en cada porción de diferentes amplitudes, mas, cuja suma, corresponde a la forma original de la onda. Con el empleo de este método surgen varios parámetros. El Índice Bispectral, o simplemente BIS (100-acordado hasta 0- isoeléctrico) es derivado de los mejores parámetros (p.ej.: frecuencia de la borda spectral, frecuencia mediana y el "burst supression" o surto de supresión) que fueron evaluados a través de análisis estadística.
CONCLUSIONES: La experiencia clínica ha mostrado que el BIS puede predecir una respuesta a la incisión de la piel durante la anestesia. Entretanto, el BIS no es independiente de la técnica anestésica usada. Hay diferentes respuestas, a depender del hipnótico y analgésico utilizado.




Until recently, anesthesiologists did not count on a monitor able to evaluate depth of anesthesia induced by intravenous and inhalational anesthetics. Anesthesia depth evaluation was limited to autonomic activity measurements, such as blood pressure and heart rate changes.

Anesthesia itself is a blend of several components. It must invariable contain analgesia, which may be induced by peripheral blocks with local anesthetics or with intravenous analgesics which act mainly in central nervous system (CNS) receptors. Opioids are classic intravenous analgesics.

Another anesthetic component is hypnosis, which may be induced by inhalational or intravenous drugs. Neuromuscular block (NMB) may be associated to analgesia and hypnosis by drugs which act in myoneural plate receptors. Currently there are effective monitors to evaluate NMB intensity and it would be ideal if those monitors would be used every time NMBs were used. Hypnosis is more difficult to evaluate, but clearly is related to hypnotic effects on the CNS. EEG can measure these changes in brain activity. This relationship was first suggested in 1937 1.

EEG as such (Chart I) could barely be used to monitor anesthesia depth because it would need miles of paper and the interpretation of results would become a difficult task with very poor practical results. In addition, anesthetic drugs cause different EEG tracing patterns, preventing the creation of a universal monitor and requiring the presence of an EEG expert in the operating room. However, the major difficulty is that each hypnotic drug causes unique changes in EEG wavelengths. Such difficulties were overcome with the introduction of the microcomputer, which made possible the decrease in EEG data, by translating electroencephalographic waves into microprocessed derivates 2 (Figure 1).

To minimize those inconveniences, computer-processed EEG has been proposed. Processed EEG starts by digitalizing small EEG signal tracings (known as epoch).

Digitized EEG may be mathematically transformed by a process known as Fourier analysis, which divides the complex EEG signal in integers representing EEG wavelengths 2 (Figure 2 and 3).

Each processed EEG segment "power spectrum" or "epoch" may be graphically represented. More information may be obtained if consecutive "epochs" are placed parallel and close to each other, producing what is called "compressed spectral array" (CSA), which produces the typical "peak and valley" pattern (Figure 4).

CSA peaks represent EEG frequencies in which the highest amplitude (or power) is located. An alternative method is to calculate the frequency in which 95% of total power (highest amplitude) are located. Such parameter is known as "spectral edge frequency 95%" (SEF-95%) meaning that 95% of the waves have their peak within this limit. The major advantage of SEF-95% is to be a simple integer which was believed to be related to adequate anesthesia depth 3-6. Another total power-derived parameter is median frequency. It must also be reminded that, in certain clinical conditions (such as deep hypnosis and hypothermia), "burst suppression" may be present. This parameter is defined as EEG tracing with amplitude lower than 5 µV and duration longer than 0.5 milliseconds.

All those variables make difficult the evaluation of hypnosis intensity.

To reduce interferences and obtain an accurate evaluation of the level of hypnosis, bispectral analysis was created, which is a statistical technique to study the phenomenon with non-linear features. The first study using this technique was published in 1971 7. Bispectral analysis is an alternative to study EEG waves and replaces the conventional spectral power analysis derived from Fourier's rapid transformation 8,9. Bispectral analysis quantifies and relates sinusoidal components of the EEG tracing 2,10. Fourier's analysis and bispectral analysis data are used to create the bispectral index parameter, or simply BIS index. BIS is a numeric scale from 100 to 0 where 100 or a number close to it represents the awaken patient and 0 represents EEG parameters in a single variable. This was obtained from the accumulated experience of more than 2000 EEGs in awaken and anesthetized patients using different hypnotic drugs (inhalational and intravenous) 11.

After excluding artifacts, EEG tracings were submitted to spectral calculation to arrive at several sub-parameters which were then submitted to statistical analysis to obtain the "best" parameters, resulting in the bispectral index.



This equipment uses signals originated from the frontal region, which are digitalized and also filtered to avoid artifact interference. Then, the signal is analyzed to detect suppression burst. The signal undergoes Fourier's rapid transformation to reach Bispectrum (Chart II). The sum of such parameters gives origin to the Bispectral Index, or simply BIS. This is made up of a scale from 100 to 0. Figures close to 100 indicate an awaken individual, that is, one not influenced by any type of hypnotic drug. As the scale decreases, sedation levels increase being 70 considered mild sedation and below 60, high levels of hypnosis (Table I). This is true both for anesthetic induction and recovery.



Early clinical studies were combined to form the basis for the BIS monitor model 1.1 12,13. In this study, BIS was compared to median frequency and spectral edge frequency as to its ability to forecast movement at skin incision in patients submitted to anesthesia with thiopental and isoflurane. The result was significantly different between BIS levels (65 ± 15, mean ± SD) of patients reacting to skin incision and those not reacting to such stimuli (BIS 40). Spectral edge frequency 95% and median frequency were not statistically significant 14,15.



Ketamine, which in the pharmacological concept is not a hypnotic drug, produces dissociative anesthesia with excitatory effects on EEG. Ketamine doses of 0.25 to 0.5 are sufficient to produce absence of responses, but BIS remains unchanged 16,17. The association of ketamine to intravenous propofol has an additive effect, but BIS remains unchanged 18,19.

Nitrous oxide (N2O) at 70% leads to no response to commands, but BIS remains unchanged 20. The addition of N2O to propofol decreases the response to several stimuli 21.

There are no sufficient data to evaluate the use of BIS in neurological patients.



BIS is the first sedation and hypnosis depth monitor and only a deeper clinical experience will prove its usefulness in Anesthesiology. Our 4-year experience with this equipment has shown the possibility of inducing adequate anesthesia at lower sedation levels, which are, in general, lower than those obtained by clinical evaluation. Another important and very relevant factor in the clinical practice is the possibility of distinguishing between hypnotic drugs (intravenous or inhalational agents) and the indirect need for analgesia.



01. Gibbs FA, Gibbs EL, Lennox WG - Effect on the electroencephalogram of certain drugs which influence nervous activity. Arch Intern Med, 1937;60:154-166.         [ Links ]

02. Rampil IJ - A primer for EEG signal processing in anesthesia. Anesthesiology, 1998;89:980 -1002.         [ Links ]

03. Schwilden H, Schüttler J, Stoeckl H - Quantitation of the EEG and pharma-codynamic modelling of hypnotic drugs: etomidate as an example. Eur J Anaesth, 1985;2:121-130.         [ Links ]

04. Levy WJ, Shapiro HM, Maruchak G et al - Automated EEG processing for intraoperative monitoring: a comparison of techniques. Anesthesiology, 1980;53:223-236.         [ Links ]

05. Rampil IJ, Matteo RS - Changes in EEG spectral edge frequency correlate with the hemodynamic response to laryngoscopy and intubation. Anesthesiology, 1987;67:139-142.         [ Links ]

06. Schwilden H, Stoeckl H - Effective therapeutic infusions produced by closed-loop feedback control of methohexital administration during total intravenous anesthesia with fentanyl. Anesthesiology, 1990;73:225-229.         [ Links ]

07 Barnett TP, Johnson LC, Naitoh P et al - Bispectrum analysis of electroencephalogram signals during waking and sleeping. Science, 1971;172:401-402.         [ Links ]

08. Rosenblatt M, Van Ness JW - Estimation of the bispectrum. Ann Math Stat, 1972;36:1120-1136.         [ Links ]

09. Nikias CL, Raghuveer MR - Bispectrum estimation: a digital signal processing framework. IEEE Proc, 1987;75:869-891.         [ Links ]

10. Sigl J, Chamoun N - An introduction to bispectral analysis for the electroencephalogram. J Clin Monit, 1994;10:392-404.         [ Links ]

11. Glass PSA, Bloom M, Kearse L et al. - Bispectral analysis measures sedation and memory effects of propofol, midazolam, isoflurane, and alfentanil in healthy volunteers. Anesthesiology, 1997;86:836-847.         [ Links ]

12. Dutton RC, Smith WD, Smith NT - The Use of EEG to Predict Movement during Anesthesia, Consciousness, Awareness and Pain em: Rosen M, Lunn JN - General Anaesthesia. London, Butterworth, 1987;72-82.         [ Links ]

13. Kearse LA, Manberg PJ, de Bros F et al - Bispectral analysis of the electroencephalogram during induction of anesthesia may predict hemodynamic responses to laryngoscopy and intubation. Electroenceph Clin Neurophl, 1994;90:194-200.         [ Links ]

14. Sebel PS, Bowles SM, Saini V et al - EEG bispectrum predicts movement during thiopental isoflurane anesthesia. J Clin Monit, 1995;11:83-91.         [ Links ]

15. Dwyer RC, Rampil IJ, Eger IEI et al - The electroencephalogram does not predict depth of isoflurane anesthesia. Anesthesiology, 1994;81:403-409.         [ Links ]

16. Moriaka N, Ozaki M, Matsukawa T et al - Ketamine causes a paradoxical increase in the bispectral index. Anesthesiology, 1997;87:A502.         [ Links ]

17. Suzuki M, Edmonds HL, Tsueda K et al - Ketamine on bispectral index and levels of sedation. J Clin Monit, 1998;14:373-377.         [ Links ]

18. Sakai T, Singh WD, Kudo T et al - The effect of ketamine on clinical endpoints of hypnosis and EEG variables during propofol infusion. Acta Anaesthesiol Scand, 1999;43:212-216.         [ Links ]

19. Avramov MN, Badrinath S, Shadrick M et al - The effect of ketamine on EEG-bispectral index (BIS) during propofol sedation. Anesthesiology, 1997;87:A501.         [ Links ]

20. Barr G, Jakobsson JG, Owall A et al - Nitrous oxide does not alter bispectral index: study with nitrous oxide as sole agent and as adjunct to i. v. anaesthesia. Br J Anaesth, 1999;82:827-830.         [ Links ]

21. Kearse LA, Rosow CE, Zaslavsky A et al - Bispectral analysis of the electroencephalogram predicts conscious processing of information during propofol sedation and hypnosis. Anesthesiology, 1998;88:25-34.         [ Links ]



Correspondence to:
Dr. Pedro Thadeu Galvão Vianna
Departº de Anestesiologia da FMB - UNESP
Distrito de Rubião Junior
ZIP: 18618-970 City: Botucatu, Brazil

Submitted for publication February 20, 2001
Accepted for publication April 16, 2001



* Received from CET/SBA da Faculdade de Medicina de Botucatu (FMB - UNESP)
This article has received corrections in agreement with the ERRATUM published in Volume 51 Number 6.

Creative Commons License All the contents of this journal, except where otherwise noted, is licensed under a Creative Commons Attribution License