Acta Anaesthesiol Scand 1990: 34: 579-584
The effect of nitrous oxide on EEG spectral power during halothane and isoflurane anaesthesia A. YLI-HANKALA Tampere University of Technology, Tampere, Finland
The effect of N,O on EEG during halothane and isoflurane anaesthesia was studied in 24 elective-surgery patients. The total EEG power and various power bands were analysed with fast Fourier transform power spectra. Anaesthesia was induced by mask. EEG analysis was performed from the data collected before induction and during steady-state halothane and isoflurane anaesthesia with and without N,O. I n both halothane and isoflurane anaesthesia, N,O had a significant effect on EEG. Alpha- and beta-range EEG power and total power decreased during N,O in both groups. Delta- and theta-range power increased during N,O in the halothane group. The study shows that the effect of nitrous oxide should be taken into consideration when EEG is being studied or monitored during anaesthesia. Received 25 November 1989, accepted for publication 24 April 1990
Key words: Anesthesia; EEG; halothane; isoflurane; nitrous oxide; power spectrum.
Various effects of nitrous oxide on the electroencephalogram (EEG) during inhalation anaesthesia have been reported in the literature. In studies by Pichlmayr et al., changes in the EEG specifically due to nitrous oxide could not be established when nitrous oxide was used in combination with other anaesthetics (1). Brandt et al., however, reported that 0.5 minimal alveolar concentration (MAC) of nitrous oxide considerably reduced the total spectral EEG power produced by inhalation anaesthetics (2). This phenomenon was seen in halothane, enflurane and isoflurane anaesthesia. Backman et al. have also described the disappearance of fast halothane rhythm when 75% nitrous oxide was inhaled (3). The purpose of this study was to investigate quantitatively the effect of nitrous oxide on the EEG of patients under pure 1.3 MAC halothane and isoflurane anaesthesia. The specific purpose was to find frequency ranges sensitive to nitrous oxide.
PATIENTS AND METHODS The EEG of 24 elective-surgery patients was studied during inhalation anaesthesia. The study design was approved by the local Ethics Committee, and all patients gave informed consent. The study was performed during various types of surgical procedure (e.g. abdominal, orthopaedic and plastic surgery). Twelve patients were randomly selected for the halothane group (9 female, 3 male, mean age 24.5 years, range 16-37 years) and the remaining 12 patients were given isoflurane anaesthesia (4 female, 8 male, mean age 24.5 years, range 17-38 years). None of the patients had any history or evidence of central nervous system diseases.
Anaesthesia As premedication, each patient received pethidine i.m. I mg/kg body weight and atropine i.m. 0.0075 mg/kg body weight, 45 min before anaesthesia induction. Benzodiazepines were not administered. Anaesthesia was induced by mask with the anaesthetic under study. After deepening of the anaesthesia, the patient was paralysed with succinylcholine, 50 mg i.v., and the trachea was intubated. The intubation tube was connected to a ventilator and monitoring of the anaesthetic under study was carried out from the connection between the endotracheal tube and the ventilator. Positive pressure ventilation was used and the ventilation was adjusted to maintain normocapnia (end-tidal (ET) CO, 4.6-6.0 kPa). ETCO, was measured with a Datex Normocap@ monitor (Datex, Espoo, Finland). A ventilation frequency of 10 cycles per minute was used. Muscle relaxation was continued using alcuronium, 10 mg, after which additional doses of 2.5 mg were given according to the requirements of the operation. ECG lead I1 was monitored continuously. Systolic and diastolic
Induction isoflurane, N*s 'Ir
1.3 MAC isofiuranr,
6 ,tlrnts
> 30 mlnules
> 30 mlnules
Induction isoflurane, air
1.3 MAC isoflurane, air
1.3 MAC Isoflurane,
6 patlrnts
> 30 minutes
-
N20,02
I
I
1.3 MAC Isoflurane, air
N20.02
> 30 minutes
-
I
I
*
-
Fig. 1. The study design in the isoflurane patient group. The same procedure was carried out also with halothane.
580
A. YLI-HANKALA
awake
Odair
02/N20
TrOi
Halothane Fz-Cz
T3-01
lsoflurane Fz-Cz Fig. 2. The effect of nitrous oxide on the EEG during halothane and isoflurane anaesthesia. Time constant is 0.1 s. For the six isoflurane patients whose anaesthesia was induced with nitrous oxide, the anaesthesia was continued with isoflurane in 35% oxygen and 65% nitrous oxide for a period of at least 30 min during the surgical procedure. After that, the delivery of nitrous oxide was interrupted and isoflurane was further supplied along with oxygen in air for another period of 30 rnin or longer, the oxygen content of the gas mixture being 35%. For those six isoflurane patients whose anaesthesia was induced with isoflurane in oxygen-air mixture and without nitrous oxide, the anaesthesia was continued with 1.5 v01.y~E T isoflurane in oxygenair mixture for a period of at least 30 min during the surgical procedure. After that, the isoflurane was further supplied along with oxygen in 65% nitrous oxide for another period of 30 min or longer. The same anaesthesia procedure was carried out in the halothane
arterial pressures were measured with a stethoscope and a sphygmomanometer. The EEG recordings were started 5 min before the induction, and control tracings were obtained while patients were awake and had their eyes closed. In six randomly selected cases in both groups, anaesthesia induction was carried out and the anaesthesia continued with the volatile anaesthetic under study (i.e. isoflurane or halothane), and 35% oxygen in nitrous oxide. In the other six, the anaesthesia was induced, and at first further supported with the volatile anaesthetic under study and oxygen-air mixture containing 35% oxygen, but without N,O. For the isoflurane patient group, an ET isoflurane concentration of 1.5 v01.y~was maintained throughout anaesthesia, monitored with a Datex Normac" monitor. This concentration is presumed to be 1.3 minimal alveolar concentration (1.3 MAC) of isoflurane (4).
Halothane
lsoflurane
TsOi Awake
FZGZ
hh--rhcI-
I
1-1
0
8
16
24
32
0
Fz-Cz
Ts-01
8
16
24
32
I
0
8
16
24
1
32
Weir
0
8
16
24
32
0
8
16
24
32
WNZO 0
8
16
24
32
0
8
16
24
32
Fig. 3. The effect of nitrous oxide on the power spectra of EEG during halothane and isoflurane anaesthesia. The domination of low EEG frequencies caused by nitrous oxide is suppressed by the equaliser high-pas filter (4 Hz, 6 dB/octave). Power spectra curves are redrawn from the EEG analyser plotter curves.
58 1
EEG, HALOTHANE, ISOFLURANE AND N,O patient group, 1.3 MAC halothane being 1.0 vol.% ET halothane (5). Sixty-five per cent nitrous oxide in oxygen, which was used in the study, is reported to be equivalent to 0.6 MAC (6). From the end of each period of 30 min, a 5-min sample of EEG was used for analysis. The patient had thus received at least 25 min
of constant gas concentration before the EEG sample. This was considered to be sufficient time for the anaesthesia to reach a steady state. The study design in the isoflurane patient group is shown in Fig. 1. The same procedure was performed also in the halothane group.
EEG
n
0.64 4A
2.6
c8
6-10
812
10-14
A Hilothane + air Halothane + NzO + Oz
12-16 16-20 2424 13-16 18-22
HZ
340
-
320
-
A Iroflurme + alr 0 Iaoflurane + Nz0 +Oz
300280
-
280-
With the use of an O.T.E. Biomedicaa (Italy) analyser and silver/ silver chloride dome electrodes, EEG derived from electrode pairs T,O, and F;C, (international 1&20 electrode system) were recorded on a RacaP (England) tape recorder. A power-spectrum analysis of the 5-min samples of EEG was performed off-line by repeatedly playing back the EEG from the tape recorder to the 1244 analyser of the O.T.E. In this connection the EEG was also recorded on paper. On the basis of these paper recordings, only artefact-free samples of EEG were used in the spectral analysis. As the analyser had only six adjustable analysing bands, the tape was replayed to get a variety of frequency bands. The analyser calculated the average spectra of 1-min epochs and the 1246 data printer printed the powers of the frequency bands selected. In the analysis, ordinary fast Fourier transform was used. The sampling frequency of the analyser was 128 samples/second/channel. An equaliser high-pass filter (4 Hz, 6 dB/octave) was used to suppress the lower frequencies, which would otherwise have a disproportionate weighting in the spectral analysis. The principles of filtering in EEG monitoring, particularly of suppressing the low frequencies, are thoroughly discussed by Prior & Maynard (7), and are not further presented here. A power-spectrum analysis was performed using partially overlapping 8-s epochs, producing power spectra with a resolution of 0.5 Hz. The spectral bands of 0.5-4 Hz, 2-6 Hz, 4-8 Hz, 6-10 Hz, 8-12 Hz, 10-14 Hz, 12-16 Hz, 13-18 Hz, 16-20 Hz, 18-22 Hz, and 20-24 Hz, and the total power of 0.5-32 Hz, were analysed. The analyser printed the spectral power of the frequency bands listed above. The values were averages from I-min samples of EEG. From five 1-min averages, a grand average was calculated. In order to approach a Gaussian distribution, the square root of these grand averages was used in the statistical analysis of the data and in the illustrations. As a result, the interpretation of the values is that they
Halothaneanlleathaala
0.532 HZ
400
300
812
c8
0.54 2-6
48
6-10
10-14
0 Ta.01
lsoflunne anaaithoala
0 Fz-Cz
f
12-16 15-20 20-24 13-18 18-22
Hr
Fig. 4. A. The effect of nitrous oxide in F,-C, EEG recording during halothane anaesthesia. Square root of power (mean s.d.) is presented in various spectral bands. Symbols for Wilcoxon's matchedpairs signed-ranks tests: * =P
The effect of nitrous oxide on EEG spectral power during halothane and isoflurane anaesthesia A. YLI-HANKALA Tampere University of Technology, Tampere, Finland
The effect of N,O on EEG during halothane and isoflurane anaesthesia was studied in 24 elective-surgery patients. The total EEG power and various power bands were analysed with fast Fourier transform power spectra. Anaesthesia was induced by mask. EEG analysis was performed from the data collected before induction and during steady-state halothane and isoflurane anaesthesia with and without N,O. I n both halothane and isoflurane anaesthesia, N,O had a significant effect on EEG. Alpha- and beta-range EEG power and total power decreased during N,O in both groups. Delta- and theta-range power increased during N,O in the halothane group. The study shows that the effect of nitrous oxide should be taken into consideration when EEG is being studied or monitored during anaesthesia. Received 25 November 1989, accepted for publication 24 April 1990
Key words: Anesthesia; EEG; halothane; isoflurane; nitrous oxide; power spectrum.
Various effects of nitrous oxide on the electroencephalogram (EEG) during inhalation anaesthesia have been reported in the literature. In studies by Pichlmayr et al., changes in the EEG specifically due to nitrous oxide could not be established when nitrous oxide was used in combination with other anaesthetics (1). Brandt et al., however, reported that 0.5 minimal alveolar concentration (MAC) of nitrous oxide considerably reduced the total spectral EEG power produced by inhalation anaesthetics (2). This phenomenon was seen in halothane, enflurane and isoflurane anaesthesia. Backman et al. have also described the disappearance of fast halothane rhythm when 75% nitrous oxide was inhaled (3). The purpose of this study was to investigate quantitatively the effect of nitrous oxide on the EEG of patients under pure 1.3 MAC halothane and isoflurane anaesthesia. The specific purpose was to find frequency ranges sensitive to nitrous oxide.
PATIENTS AND METHODS The EEG of 24 elective-surgery patients was studied during inhalation anaesthesia. The study design was approved by the local Ethics Committee, and all patients gave informed consent. The study was performed during various types of surgical procedure (e.g. abdominal, orthopaedic and plastic surgery). Twelve patients were randomly selected for the halothane group (9 female, 3 male, mean age 24.5 years, range 16-37 years) and the remaining 12 patients were given isoflurane anaesthesia (4 female, 8 male, mean age 24.5 years, range 17-38 years). None of the patients had any history or evidence of central nervous system diseases.
Anaesthesia As premedication, each patient received pethidine i.m. I mg/kg body weight and atropine i.m. 0.0075 mg/kg body weight, 45 min before anaesthesia induction. Benzodiazepines were not administered. Anaesthesia was induced by mask with the anaesthetic under study. After deepening of the anaesthesia, the patient was paralysed with succinylcholine, 50 mg i.v., and the trachea was intubated. The intubation tube was connected to a ventilator and monitoring of the anaesthetic under study was carried out from the connection between the endotracheal tube and the ventilator. Positive pressure ventilation was used and the ventilation was adjusted to maintain normocapnia (end-tidal (ET) CO, 4.6-6.0 kPa). ETCO, was measured with a Datex Normocap@ monitor (Datex, Espoo, Finland). A ventilation frequency of 10 cycles per minute was used. Muscle relaxation was continued using alcuronium, 10 mg, after which additional doses of 2.5 mg were given according to the requirements of the operation. ECG lead I1 was monitored continuously. Systolic and diastolic
Induction isoflurane, N*s 'Ir
1.3 MAC isofiuranr,
6 ,tlrnts
> 30 mlnules
> 30 mlnules
Induction isoflurane, air
1.3 MAC isoflurane, air
1.3 MAC Isoflurane,
6 patlrnts
> 30 minutes
-
N20,02
I
I
1.3 MAC Isoflurane, air
N20.02
> 30 minutes
-
I
I
*
-
Fig. 1. The study design in the isoflurane patient group. The same procedure was carried out also with halothane.
580
A. YLI-HANKALA
awake
Odair
02/N20
TrOi
Halothane Fz-Cz
T3-01
lsoflurane Fz-Cz Fig. 2. The effect of nitrous oxide on the EEG during halothane and isoflurane anaesthesia. Time constant is 0.1 s. For the six isoflurane patients whose anaesthesia was induced with nitrous oxide, the anaesthesia was continued with isoflurane in 35% oxygen and 65% nitrous oxide for a period of at least 30 min during the surgical procedure. After that, the delivery of nitrous oxide was interrupted and isoflurane was further supplied along with oxygen in air for another period of 30 rnin or longer, the oxygen content of the gas mixture being 35%. For those six isoflurane patients whose anaesthesia was induced with isoflurane in oxygen-air mixture and without nitrous oxide, the anaesthesia was continued with 1.5 v01.y~E T isoflurane in oxygenair mixture for a period of at least 30 min during the surgical procedure. After that, the isoflurane was further supplied along with oxygen in 65% nitrous oxide for another period of 30 min or longer. The same anaesthesia procedure was carried out in the halothane
arterial pressures were measured with a stethoscope and a sphygmomanometer. The EEG recordings were started 5 min before the induction, and control tracings were obtained while patients were awake and had their eyes closed. In six randomly selected cases in both groups, anaesthesia induction was carried out and the anaesthesia continued with the volatile anaesthetic under study (i.e. isoflurane or halothane), and 35% oxygen in nitrous oxide. In the other six, the anaesthesia was induced, and at first further supported with the volatile anaesthetic under study and oxygen-air mixture containing 35% oxygen, but without N,O. For the isoflurane patient group, an ET isoflurane concentration of 1.5 v01.y~was maintained throughout anaesthesia, monitored with a Datex Normac" monitor. This concentration is presumed to be 1.3 minimal alveolar concentration (1.3 MAC) of isoflurane (4).
Halothane
lsoflurane
TsOi Awake
FZGZ
hh--rhcI-
I
1-1
0
8
16
24
32
0
Fz-Cz
Ts-01
8
16
24
32
I
0
8
16
24
1
32
Weir
0
8
16
24
32
0
8
16
24
32
WNZO 0
8
16
24
32
0
8
16
24
32
Fig. 3. The effect of nitrous oxide on the power spectra of EEG during halothane and isoflurane anaesthesia. The domination of low EEG frequencies caused by nitrous oxide is suppressed by the equaliser high-pas filter (4 Hz, 6 dB/octave). Power spectra curves are redrawn from the EEG analyser plotter curves.
58 1
EEG, HALOTHANE, ISOFLURANE AND N,O patient group, 1.3 MAC halothane being 1.0 vol.% ET halothane (5). Sixty-five per cent nitrous oxide in oxygen, which was used in the study, is reported to be equivalent to 0.6 MAC (6). From the end of each period of 30 min, a 5-min sample of EEG was used for analysis. The patient had thus received at least 25 min
of constant gas concentration before the EEG sample. This was considered to be sufficient time for the anaesthesia to reach a steady state. The study design in the isoflurane patient group is shown in Fig. 1. The same procedure was performed also in the halothane group.
EEG
n
0.64 4A
2.6
c8
6-10
812
10-14
A Hilothane + air Halothane + NzO + Oz
12-16 16-20 2424 13-16 18-22
HZ
340
-
320
-
A Iroflurme + alr 0 Iaoflurane + Nz0 +Oz
300280
-
280-
With the use of an O.T.E. Biomedicaa (Italy) analyser and silver/ silver chloride dome electrodes, EEG derived from electrode pairs T,O, and F;C, (international 1&20 electrode system) were recorded on a RacaP (England) tape recorder. A power-spectrum analysis of the 5-min samples of EEG was performed off-line by repeatedly playing back the EEG from the tape recorder to the 1244 analyser of the O.T.E. In this connection the EEG was also recorded on paper. On the basis of these paper recordings, only artefact-free samples of EEG were used in the spectral analysis. As the analyser had only six adjustable analysing bands, the tape was replayed to get a variety of frequency bands. The analyser calculated the average spectra of 1-min epochs and the 1246 data printer printed the powers of the frequency bands selected. In the analysis, ordinary fast Fourier transform was used. The sampling frequency of the analyser was 128 samples/second/channel. An equaliser high-pass filter (4 Hz, 6 dB/octave) was used to suppress the lower frequencies, which would otherwise have a disproportionate weighting in the spectral analysis. The principles of filtering in EEG monitoring, particularly of suppressing the low frequencies, are thoroughly discussed by Prior & Maynard (7), and are not further presented here. A power-spectrum analysis was performed using partially overlapping 8-s epochs, producing power spectra with a resolution of 0.5 Hz. The spectral bands of 0.5-4 Hz, 2-6 Hz, 4-8 Hz, 6-10 Hz, 8-12 Hz, 10-14 Hz, 12-16 Hz, 13-18 Hz, 16-20 Hz, 18-22 Hz, and 20-24 Hz, and the total power of 0.5-32 Hz, were analysed. The analyser printed the spectral power of the frequency bands listed above. The values were averages from I-min samples of EEG. From five 1-min averages, a grand average was calculated. In order to approach a Gaussian distribution, the square root of these grand averages was used in the statistical analysis of the data and in the illustrations. As a result, the interpretation of the values is that they
Halothaneanlleathaala
0.532 HZ
400
300
812
c8
0.54 2-6
48
6-10
10-14
0 Ta.01
lsoflunne anaaithoala
0 Fz-Cz
f
12-16 15-20 20-24 13-18 18-22
Hr
Fig. 4. A. The effect of nitrous oxide in F,-C, EEG recording during halothane anaesthesia. Square root of power (mean s.d.) is presented in various spectral bands. Symbols for Wilcoxon's matchedpairs signed-ranks tests: * =P
