The continuous infusion of Althesin under electroencephalographic (e.e.g.) control provided a constant level of light anaesthesia for periods of 1-5.5 h during ...
Br.J. Anaestk. (1978), 50, 993
E.E.G. MONITORING FOR THE CONTROL OF ANAESTHESIA PRODUCED BY THE INFUSION OF ALTHESIN IN PRIMATES P. F. PRIOR, D. E. MAYNARD AND J. B. BRIERLEY SUMMARY
The continuous infusion of Althesin under electroencephalographic (e.e.g.) control provided a constant level of light anaesthesia for periods of 1-5.5 h during experimental brain hypoxia in spontaneously breathing baboons and Rhesus monkeys. Polygraphic records (respiration, heart rate, arterial pressure, cerebral venous sinus pressure, end-tidal gas concentrations) and also estimation of blood-gas tensions, pH, and concentrations of pyruvate and lactate demonstrated a steady physiological state. Various methods of e.e.g. monitoring were tested to establish an optimal assessment of depth of anaesthesia as a guide to the control of the rate of infusion of Althesin. A purposebuilt modification of the Cerebral Function Monitor was found to give unequivocal recognition of changing depths of anaesthesia.
The continuous infusion of the steroid anaesthetic agent alphaxalone-alphadolone (Althesin, veterinary preparation, Saffan, Glaxo) has been described for the induction and maintenance of anaesthesia in man (Savege et al., 1975). However, in contrast to the large amount of information about bolus injections (Postgraduate Medical Journal, 1972) little has been reported on the systemic or central nervous system effects of continuous infusion either in man or in the experimental primate. During experiments that involve measurements of cerebral blood flow, cerebral metabolism or the electrical activity of the brain, it is essential that anaesthesia is maintained at a steady, and preferably light, level. During studies of profound hypoxia in Rhesus monkeys (Macaca mulatto, (MM)) and baboons (Papio anubis (PA)) (Brierley et al., 1978), light anaesthesia was obligatory for several hours to permit polygraphic physiological recording in spontaneously breathing animals. Anaesthesia had to be kept at a constant depth throughout to allow identification and quantification of alterations in the electroencephalogram (e.e.g.) as a result of hypoxia. Intermittent i.v. injections of pentobarbitone proved unsuitable since they produced sharp decreases which were followed by slow increases in the level of anaesthesia with PAMELA F. PRIOR, M.D., B.S.; D. E. MAYNARD, PH.D.; The
E.E.G. Department, Section of Neurological Sciences, The London Hospital, Whitechapel, London El IBB. J. B. BRIERLEY, M.D., F.R.C.PATH., F.R.C.PSYCH. (and P. F.
Prior); The Medical Research Council Laboratories, Woodmansterne Road, Carshalton, Surrey SM5 4EF. 0007-0912/78/0050-0993 $01.00 82
corresponding e.e.g. changes indicative of major alterations in the functional state of the brain. As a continuous infusion of Althesin appeared to be applicable, a simple method of monitoring cerebral function was deemed necessary to obtain fine control of the level of anaesthesia. Previous studies in MM (Brierley et al., 1977) had shown that a six-point visual rating scale for the background activity of the e.e.g., adapted from that of Scott and Virden (1972), could be used to describe a wide range of levels of pentobarbitone anaesthesia and give minute-by-minute quantification. Although the e.e.g. provides an accurate means of monitoring the depth of anaesthesia (Courtin, Bickford and Faulconer, 1950), visual analysis is too complex and laborious for routine use. Therefore, it was necessary to develop an automatic method to reduce and simplify the e.e.g. data and facilitate the early recognition of changes in the depth of anaesthesia. A simple monitoring device based on the e.e.g. signal, the Cerebral Function Monitor (CFM, Devices Ltd), had been developed (Maynard, Prior and Scott, 1969) and has been in clinical use for several years (Prior et al., 1971; Branthwaite, 1973; Prior, 1973; Schwartz et al., 1973; Dubois, 1975). This device gives a compressed (6 or 30-cm h - 1 ) filtered e.e.g. signal which emphasizes the overall trends of cerebral activity. Adequate function of the recording system is indicated by a continuous monitoring of electrode impedance and, in addition, frequency components likely to be artefactual are rejected or indicated. Previous experience had shown © Macmillan Journals Ltd 1978
BRITISH JOURNAL OF ANAESTHESIA
994 CFM to have advantages over conventional e.e.g. in the earlier recognition of changes resulting from the induction of anaesthesia (Schwartz, Virden and Scott, 1974). It seemed that a greater quantification of data with the addition of frequency analysis might help the anaesthetist to maintain a steady level of anaesthesia. The aims of the study were first, to establish whether the continuous infusion of Althesin under e.e.g. control could provide a steady state of light anaesthesia in experimental primates and also to define the associated physiological variables. Second, to compare the visual rating system for the e.e.g. during anaesthesia with automatic methods of signal analysis. Third, to establish the optimal system for processing the e.e.g. signal using a simple analog device to enable the anaesthetist to control the depth of anaesthesia by adjustment of the rate of administration according to the processed signal. METHODS
Animals Eleven adult MM (3.8-11.5 kg) and 10 adult PA (5.7-9.3 kg) were used to study the physiological effects, including a continuously recorded e.e.g., of the i.v. infusion of Althesin. The more detailed study of the e.e.g. signal analysis was performed on taperecorded data from six of these animals. The animals were subjected subsequently to profound hypoxia in the study of Brierley and others (1978). Anaesthesia and poly graphic recording
About 4 h before the hypoxic stress the 21 animals
were anaesthetized with pentobarbitone 45 mg kg" 1 i.p., given atropine 0.6 mg i.m. and endotracheal intubation was performed. The spontaneously breathing animals were prepared for the polygraphic recording of e.e.g., electrocardiogram (e.e.g.), heart rate (tachometer), respiratory frequency, end-tidal Po2 and Pco 2 , arterial pressure, sagittal sinus pressure and body temperature. Later, blood-gas tensions, pH, and the concentrations of lactate and pyruvate were estimated in arterial and cerebral venous blood samples. During a subsequent 3-6 h experimental period, starting at a mean of 76 min (range 13-207 min) after the last pentobarbitone injection, anaesthesia was maintained by a continuous infusion of Althesin (Palmer infusion apparatus). The standard solution (12 mg ml""1 total steroids) was infused at a mean rate of 0.020 ml min" 1 (range 0.006-0.093) in MM and at a mean rate of 0.070 ml min- 1 (range 0.006-0.187) in PA. When deeper anaesthesia was required briefly for surgical procedures, supplementary bolus injections were given and the effects of various doses were compared. E.e.g. recording and analysis
Bipolar e.e.g. recordings were made from 14 silver extradural ball electrodes (two parasagittal rows) that had been implanted at standard stereotactic sites about 5 days earlier. Recording required gains of 200-450 (J.V cm" 1 and the tracings, with those of e.e.g. and breathing, were made on an eight-channel ' Elema-Schonander Mingograf apparatus. The paper tracings were rated visually on a six-point scale for general background activity (table I), a score being
TABLE I. Visual rating scale for e.e.g. background activity at different levels of anaesthesia {extradural recordings) Level
Clinical state Light anaesthesia May respond to stimuli Light to moderate anaesthesia Moderate surgical anaesthesia Deep surgical anaesthesia Very deep surgical anaesthesia Very deep anaesthesia may be accompanied by failure of spontaneous breathing; may be irreversible state
E.e.g. features Continuous background activity of fairly constant voltage and with any combination of frequencies but no periods of either partial or total suppression Mild burst suppression activity: periods of less than 1 s duration of total or subtotal suppression separated by bursts of activity usually of 100-300 |iV Moderate burst suppression activity: periods of 1-3 s duration of total (occasionally subtotal) type separated by bursts of activity of 100-300 |iV Severe burst suppression activity: periods of at least 3 s duration of total suppression separated by bursts of activity usually of 50-100 |zV Low voltage burst suppression activity: periods of at least 3 s duration of total suppression separated by very brief bursts of low voltage (less than 50 (xV) activity Isoelectric record: no evidence of any cerebral electrical activity even with high gain
E.E.G. CONTROL OF ANAESTHESIA IN PRIMATES
995
allocated and subsequently plotted graphically for (3) The average minimum, mean and average maxievery 1 min in all 21 animals. This assessment was mum voltages and the percentage of activity in the used to determine the appropriate adjustment of five frequency bands of the CFAM. the rate of Althesin infusion for a chosen level of From these the mean value for each of the 19 measures anaesthesia. at each level of anaesthesia was calculated for In addition, in nine PA and six MM, the e.e.g. and each animal (n = 6, except at level 3, where n = 5). time and event signals were tape-recorded on either a After logarithmic transformation of voltage and Thermionics Products T 3000 four-channel tape suppression incidence data and modification of recorder or an Oxford Instruments four-channel frequency data by the addition of a constant, the cassette recorder. The tape recordings were then differences between levels were analysed by a one-way analysed in one or more of the following ways: analysis of variance. (1) (15 animals) Using a Honeywell DDP 516 computer (C. D. Binnie and D. S. L. Lloyd, St Bartholomew's Hospital, London) producing amplitude histograms of the number of periods of suppressions of each of nine different durations (0-0.25, 0.25-0.5, 0.5-1, 1-2, 2-4, 4-8, 8-16, 16-32, 32-60 s) in every 1-min epoch; (2) (11 animals) By a standard CFM; (3) (6 animals) By a modified CFM, the Cerebral Function Analyser Monitor (CFAM) (Maynard, 1977; D. Maynard, 1978, in preparation) simulated on a Micro 16 computer giving a continuous write-out measuring the mean, average maximum and average minimum voltages of cerebral activity in the CFM trace and also its frequency content. The frequency is given by a baseline crossing interval analysis as the proportion of total activity in each of six bands: < 1 Hz (very low frequency, "VLF", including suppressions), 1-3 Hz (delta), 4-7 Hz (theta), 8-13 Hz (alpha), 14-30 Hz (beta) and 30-40 Hz ("M" for detection of muscle activity when scalp electrodes are used (not relevant in the present study)). The voltage measures are written out as three tracks and the frequencies as variable width bands, otherwise the signal processing and electrode impedance recording are as in the standard CFM.
RESULTS
Steady levels of anaesthesia These were achieved easily for periods of 1-4.75 h (mean 142 min) by adjustment of the rates of infusion of Althesin under e.e.g. control in MM. In PA anaesthesia over 2.5-5.5 h (mean 228 min) required gradually increasing rates and some supplementary bolus injections to achieve a steady state. No anaesthetic complications such as hypersensitivity or neuromuscular reactions or cardiovascular collapse (Clarke, Fee and Dundee, 1977; Evans and Keogh, 1977) were encountered in any of the 21 animals. A steady state was maintained at light levels of anaesthesia (e.e.g. scores 1-3) for a mean of 78 min (range 20-172) before the exposure to hypoxia (table II). Measurements during the infusion of Althesin were similar to those at comparable depths of pentobarbitone anaesthesia in the same species. Certain species differences concerning circulation and blood-gas tensions not attributable to Althesin were evident and are discussed elsewhere (Brierley et al., 1978). The six levels of anaesthesia derived from visual rating of the e.e.g. (table I) have distinctive characteristics that are recognizable easily also on the CFM and CFAM traces (fig. 1). In 12 h 40 min of tape-recorded data from the six animals in which all three forms of analysis were Statistical analysis In the six animals in which all three types of analysis carried out, these differences in cerebral activity were were available 25 samples of the records of a steady confirmed by statistical analysis (tables III and IV). state for each of the six levels of anaesthesia were A one-way analysis of variance showed significant located on .the plots of the visual e.e.g. ratings. These differences between levels for all measures except the 150 sample points were then identified on the com- longest periods of suppression (table IV) on the puter histograms, and CFM and CFAM tracings. computer analysis. In the visual assessment of CFM and CFAM tracings it is the combination of features The following measurements were then made: that defines a particular pattern for each level. (1) The incidence of e.e.g. suppressions of each of the Allowing for the different rates of administration nine durations on the computer histogram. necessary to maintain comparable depths of anaesthe(2) The minimum and maximum voltages on the sia, there was no difference in the e.e.g. features between MM and PA. There was some variation in standard CFM trace.
996
BRITISH JOURNAL OF ANAESTHESIA TABLE II. Physiological state during steady state at light level of anaesthesia (e.e.g. levels 1-3). MAP = mean arterial pressure; CPP = cerebral perfusion pressure (i.e. MAP —cerebral venous (.sagittal sinus) pressure). Venous samples are all of cerebral venous blood. Means ± 1 standard deviation and (n). * * Difference between means significant at 1 % level (Student's t test); n.s. = not significant. Data from Brierley and others (1978) including three MM anaesthetized with pentobarbitone as their values were not significantly different from the 21 animals anaesthetized with Althesin
Frequency of breathing (b.p.m.) Heart rate (beat min"1) MAP (mm Hg) CPP (mm Hg) Pa 0 , (kPa) Pv 0 , (kPa) Paco, (kPa) Pv c o , (kPa) pHa (units) pHv (units) (Ca o ,-Cv 0 i ) (oxygen ml dl"1) Arterial plasma lactate (mmol litre"1) Venous plasma lactate (mmol litre"1) Arterial plasma pyruvate (mmol litre"1) Venous plasma pyruvate (mmol litre"1)
P
MM
PA
30 ± 6 (10)
31+6(14)
n.s.
163 ±18 (10) 123 ±17 (10) 118 ±19 (10) 10.6 + 0.9(10) 6.7 ±0.7 (10) 4.9 ±0.4 (10) 5.5 ±0.5 (10) 7.405 ±0.037 (10) 7.369 ±0.032 (10) 3.38±1.18(9) 0.78 ±0.11(10)
174 ±21 (14) 105 ±15 (13) 102 ±15 (12) 11.7±1.2(13) 6.1 ±0.5 (13) 4.4 ±0.5 (13) 5.2 ±0.5 (13) 7.444 ±0.039 (13) 7.403 ±0.032 (13) 3.67±1.13(13) 0.66 ±0.24 (5)
n.s. •* »* •* ** ** »*
0.90 ±0.22 (10)
0.67 ±0.17 (5)
n.s.
0.039 ±0.018 (10)
0.029 ±0.027 (4)
n.s.
0.046 ±0.018 (10)
0.035 ±0.022 (4)
n.s.
*• »* »*
n.s.
TABLE \\\}Differences between cerebral activity at six levels of anaesthesia. Five PA and one MM: one-way analysis of variance using logarithmic transformation of voltage and adding one to frequency data (actual values for voltage and frequency given) n = 6 except for level'3 where n = 5; d.f. = 29; ** Significant at 1% level. Note that peak-to-peak voltages for CFM and CFAM data are recorded from extradural electrodes and are thus about three times higher than would be the case from scalp electrodes Depths of anaesthesia (e.e.g. levels) F ratio and significance CFM voltage ((iV peak-to-peak) Maximum Minimum CFAM voltage ((iV peak-to-peak) Average maximum Mean Average minimum CFAM frequency content (% at Hz) 13-30 8-13 4-7 1-3
Error variance
118 60
83 41
66 16
72 5
28 2
4 2
32.45** 51.19**
0.245 0.200
79 66 57
61 49 40
44 33 21
50 30 9
27 8 1
3 0.5 0
32.24** 54.44** 36.45**
0.042 0.200 0.265
14 20 49 17 0
13 15 46 25
13 14 34 32 8
10 15 22 34 19
3 6 9 26 56
0 0 0 0 0
43.55** 26.56** 113.03** 23.27** 15.8**
2.229 3.553 4.590 6.200 14.53
0
E.E.G. CONTROL OF ANAESTHESIA IN PRIMATES
997
BH8 E.e.g. 300pv|
CFM
34 12
CFAM
FIG. 1. Steady levels of anaesthesia. From above downwards compressed e.e.g. samples (left frontocentral region) from extradural recording in PA; time marker at 5-s intervals. E.e.g. rating scale. CFM and CFAM samples taken from simultaneous points on the tape recordings; time markers at 1-min interals. Upper part of CFAM trace shows from above downwards, average maximum, mean and average minimum voltages peak-to-peak of smoothed signal calibrated at 10 Hz. Lower part shows frequency as the proportion of activity in each of the six relevant bands. The heavy black band below frequency is the electrode impedance in kO (not in use as e.e.g. tape-recorded).
voltage between individuals of both species at the lighter levels (1-3) of anaesthesia. These individual and species differences were also evident during pentobarbitone anaesthesia. Supplementary bolus injections
Bolus injections led to a rapid, minor and brief decrease in mean arterial pressure of about 5 mm Hg returning to normal within 30 s.
In individual animals a relationship between dose per kg body weight and the effect on cerebral electrical activity was evident from all forms of analysis. The degree and rate of change in the e.e.g. were related also to the rate of injection of the bolus, infusion rate (if any) and to the interval since the previous bolus injection (fig. 2A-F). Althesin was noted to have a very rapid, often marked and short-lived effect (fig. 2G)
BRITISH JOURNAL OF ANAESTHESIA
998
TABLE IV. Differences in incidence of suppressions at six levels of anaesthesia (number of suppressions/min of each duration Five PA and one MM: one-way analysis of variance using logarithmic transformation of data (incidences of suppressions given as actual values), n = 6 for all levels except for level 3 where n = 5; d.f. = 29; ** significant at 1% level; * significant at 5% level; | not significant E.e.g. levels and depths
