Background. Anaesthetics blunt neuronal responses to noxious stimulation, including effects on electroencephalographic (EEG) responses. It is unclear how anaesthetics differ in their ability to modulate noxious stimulation-evoked EEG activation. We investigated the actions of propofol and halothane on EEG responses to noxious stimuli, including repetitive electrical C-fibre stimulation, which normally evokes neuronal wind-up.
Methods. Rats were anaesthetized with halothane (n=8) or propofol (n=8), at 0.8× or 1.2× the amount required to produce immobility in response to tail clamping [minimum alveolar concentration (MAC) for halothane and median effective dose (ED50) for propofol]. We recorded EEG responses to repetitive electrical stimulus trains (delivered to the tail at 0.1, 1 and 3 Hz) as well as supramaximal noxious tail stimulation (clamp; 50 Hz electrical stimulus, each for 30 s).
Results. Under halothane anaesthesia, noxious stimuli evoked an EEG activation response manifested by increased spectral edge frequency (SEF) and median edge frequency (MEF). At 0.8 MAC halothane, the tail clamp increased the MEF from ≈6 to ≈8.5 Hz, and the SEF from ≈25.5 to ≈27 Hz. At both 0.8 and 1.2 MAC halothane, similar patterns of EEG activation were observed with the 1 Hz, 3 Hz and tetanic stimulus trains, but not with 0.1 Hz stimulation, which does not evoke wind-up. Under propofol anaesthesia, noxious stimuli were generally ineffective in causing EEG activation. At 0.8 ED50 propofol, only the tail clamp and 1 Hz stimuli increased MEF (≈8 to ≈10–10.5 Hz). At the higher propofol infusion rate (1.2 ED50) the repetitive electrical stimuli did not evoke an EEG response, but the tetanic stimulus and the tail clamp paradoxically decreased SEF (from ≈23 to ≈21.5 Hz).
Conclusions. Propofol has a more significant blunting effect on EEG responses to noxious stimulation compared with halothane.
The hallmark of general anaesthesia is the blunting of behavioural and neuronal responses to supramaximal noxious stimuli, such as a skin incision or a mechanical stimulus. Although there have been numerous studies which have examined electroencephalographic (EEG) effects of anaesthesia, few studies have examined the effect of anaesthetics on EEG responses to noxious stimulation1–4 and there has been little work comparing anaesthetics.45
Two commonly used anaesthetics (propofol and halothane) have divergent effects on ligand-gated neurotransmitter receptors. Propofol acts almost exclusively at the γ-aminobutyric acid (GABAA) receptor, while halothane acts at numerous receptors, including GABAA receptors, N-methyl-d-aspartate (NMDA) receptors and glycine receptors.6 In a previous study we found that halothane, at concentrations that prevented movement, did not prevent EEG activation in response to noxious stimulation.5 Furthermore, halothane appears to have a major action in the spinal cord, at least for the production of immobility.7 Propofol, however, has a major effect in the brain to produce anaesthesia,8 and blunts EEG activation responses to noxious stimulation.34 These data3–8 suggest that propofol and halothane might differ in terms of their ability to modulate EEG responses to noxious stimulation.
In the present study we examined the effects of halothane and propofol anaesthesia on the ability of supramaximal noxious stimulation to cause EEG activation. We included repetitive electrical C-fibre stimulation because it induces wind-up of nociceptive spinal neurons, a mechanism contributing to temporal summation of pain that is depressed by some anaesthetic agents.9–11 We hypothesized that propofol would blunt EEG responses to supramaximal and repetitive noxious stimulation more than halothane.
The local animal care and use committee approved this study. Adult male Sprague–Dawley rats (≈500 g) were anaesthetized in a chamber with either halothane (n=8) or isoflurane (n=8). The rats were removed and placed on mask anaesthesia and tracheostomy was performed to permit placement of a 12- or 14-gauge tracheostomy tube, followed by mechanical ventilation. A catheter was placed into the internal jugular vein for fluid and drug administration, and a catheter was placed into a carotid artery to measure blood pressure. Ligation of a carotid artery does not affect cerebral metabolism or blood flow.12 Mean arterial pressure was maintained at 75 mmHg or greater using an infusion of lactated Ringer's solution. Rectal temperature was measured with a thermometer and maintained at ≈37–38°C using a heating pad and heating lamp as needed. End-tidal anaesthetic concentration was determined using a calibrated anaesthetic agent analyser (Rascal, Ohmeda, Salt Lake City, UT, USA; or Datex 254, Helsinki, Finland).
In the halothane group, the minimum alveolar concentration (MAC) needed to prevent movement was determined in each individual rat. The halothane concentration was equilibrated at 0.9–1% and maintained for at least 15–20 min. A clamp was applied at the base of the tail and oscillated at 1–2 Hz for up to 1 min or until the rat displayed gross and purposeful movement. Depending on the initial response, the halothane concentration was increased or decreased by 0.2%, stabilized for 15 min, and the tail clamp was applied again. This process was continued until two halothane concentrations were found that just prevented and just permitted gross and purposeful movement. MAC was the average of these.
In one of the isoflurane-anaesthetized rats and in an additional five rats, we determined the propofol requirements to prevent movement in response to supramaximal noxious stimulation. In brief, the six rats were anaesthetized with isoflurane and ventilated via tracheostomy. A jugular catheter was inserted and a propofol infusion initiated at ≈400 µg kg−1 min−1. The isoflurane was discontinued and after 30–45 min, when the expired isoflurane was <0.2%, a clamp was applied to the base of the tail and oscillated at 1–2 Hz for up to 1 min or until the rat displayed gross and purposeful movement. Depending on the response, the infusion rate was increased or decreased 20%, and after 15–20 min the tail clamp was reapplied. This process was repeated until two infusion rates were found that just permitted and just prevented movement; the median effective dose (ED50) for the infusion rate was the average of these rates. The average of these values for all six rats was used as the population ED50 for the propofol EEG studies.
The electroencephalogram was recorded via four stainless steel screws placed into the skull. Two screws were placed 0.5 cm from the midline on each side near lambda. One screw was placed near the midline in the frontal region while the fourth screw was placed near the base of the skull. Leads from an Aspect 1050 EEG machine (Aspect Medical Systems, Newton, MA, USA) were attached to the screws to record EEG responses. The EEG signals were digitized at 256 Hz and filtered at 2–70 Hz. The A-1050 monitor performed a power analysis to generate the median edge frequency (MEF) and spectral edge frequency (SEF), which are the frequencies below 50 and 95% of the EEG power, respectively. MEF and SEF were downloaded every 5 s to a computer. The EEG monitor used a rolling average of the previous 30 s when generating these numbers. In addition, the raw EEG was recorded onto a computer hard drive using Chart5 (AD Instruments, Colorado Springs, CO, USA).
In the halothane rats, the halothane concentration was stabilized at 0.8 or 1.2 MAC for 15–20 min before application of noxious stimuli. The animals anaesthetized with isoflurane were administered propofol via infusion, starting at 0.8 or 1.2 ED50 (≈480 µg kg−1 min−1 or ≈720 µg kg−1 min−1), and the isoflurane was discontinued. We waited at least 30–45 min to permit the expired isoflurane concentration to decrease to less than 0.2% before beginning the noxious stimulation.
Noxious stimuli were applied in the following manner. After two needle electrodes (E-2; Grass Instruments, West Warwick, RI, USA) had been inserted into the skin at the base of the tail, trains of 20 C-fibre strength electrical stimuli (40 V, 0.5 ms pulse duration) were delivered at 0.1, 1 and 3 Hz, with 3–4 min between each train. In addition, two supramaximal stimuli were used: a tetanic stimulus (50 Hz, 60 mA current passed via the electrodes) and a tail clamp, each applied for 30 s. Pancuronium (0.2–0.3 mg kg−1 every 1–2 h) was administered intravenously to eliminate electromyographic artefacts. Once EEG responses had been recorded at one anaesthetic concentration (or propofol infusion rate), the anaesthesia was switched to the other concentration (or infusion rate) and stabilized for 15–20 min, and the noxious stimuli were applied as described above. The order in which the anaesthetic concentrations were administered was alternated between experiments. When data collection was complete, the animals were euthanized with additional anaesthesia and i.v. potassium chloride.
The MEF and SEF data were evaluated using repeated measures analysis of variance for the 30s period before stimulation and the 200 s period after initiation of stimulation. Post hoc testing was performed using the Student–Newman–Keuls test. Baseline MEF and SEF at 0.8 MAC (average of 30 s before stimulation) were compared with the respective values at 1.2 MAC using a paired t-test when comparing within an anaesthetic or an unpaired t-test when comparing between anaesthetics. P<0.05 was considered significant.
Halothane MAC was 1.0 (0.1)% and propofol ED50 was 600 (130) µg kg−1 min−1. During halothane anaesthesia, the prestimulus SEF decreased [from 24.3 (2.2) to 21.8 (2.7) Hz, P<0.01], as did MEF [from 5.8 (0.9) to 4.5 (1.0) Hz, P<0.002], in the transition from 0.8 to 1.2 MAC. During propofol anaesthesia, prestimulus MEF decreased [from 9.0 (1.3) to 7.8 (1.0) Hz, P<0.05] in the transition from 0.8 to 1.2 ED50, while the SEF was unchanged [23.2 (1.5) to 23.3 (1.3) Hz, P>0.05]. Prestimulus MEF values at 0.8 and 1.2 MAC for halothane were significantly different from the MEF values for propofol at 0.8 and 1.2 ED50, respectively (P<0.001). Prestimulus SEF values for halothane were not significantly different from the SEF values for propofol.
Figure 1 shows individual examples of the electroencephalogram under propofol anaesthesia. At 0.8 ED50 propofol (Fig. 1a), both repetitive electrical stimulation (upper trace) and the noxious tail clamp (lower trace) produced limited activation responses in the EEG. At 1.2 ED50 propofol, the same noxious stimuli did not evoke EEG changes (Fig. 1b). The EEG pattern during propofol anaesthesia included large spikes (Fig. 1c). Data are summarized in Fig. 2, where filled symbols represent EEG responses at 0.8 ED50 propofol and the open symbols 1.2 ED50 propofol. At 0.8 ED50, significant EEG activation in the MEF occurred for the 1 Hz stimulus train (Fig. 2b) and the tail clamp (Fig. 2e). At 1.2 ED50 propofol, none of the noxious stimuli resulted in EEG activation; in fact, the tetanic electrical stimulus and the tail clamp evoked a paradoxical decrease in SEF (Figs 2d and e).
In contrast to the depressant effect of propofol, halothane had very little effect on EEG activation evoked by noxious stimulation. Figure 3 shows examples of EEG traces at 1.2 MAC halothane. Electrical stimulation at 3 Hz produced EEG activation (upper trace in Fig. 3), and the supramaximal tail clamp (middle trace) and tetanic stimulus (lower trace) both resulted in strong EEG activation. Data with halothane are summarized in Fig. 4, where it can be seen that the noxious stimuli produced significant increases in SEF and MEF at both 0.8 and 1.2 MAC halothane (Fig. 4b–e), except for the 0.1 Hz stimulus train at 0.8 and 1.2 MAC (Fig. 4a) and the MEF for the 1 Hz train at 0.8 MAC (Fig. 4b).
The main finding of the present study is that propofol, given at infusion rates bracketing the ED50 for movement in response to supramaximal noxious stimulation, blunted EEG responses to repetitive and supramaximal noxious stimulation. In contrast, halothane given at peri-MAC concentrations did not prevent EEG activation elicited by noxious stimuli.
Various studies have examined propofol and halothane EEG effects. Propofol is associated with progressive EEG depression, including burst suppression and isoelectricity at higher doses.13 We examined a narrow dose range (0.8–1.2 MAC or ED50) and cannot comment on any possible effects outside that range. Furthermore, although propofol depressed the EEG response to noxious stimulation, the baseline electroencephalogram was active. Our finding that the baseline SEF was unaffected by the increased propofol dose is consistent with the data reported by Antunes and colleagues,14 although they reported that MEF was also unchanged, while we observed a slight decrease in MEF at the greater propofol dose. Some studies have described EEG activation after propofol administration in low doses associated with the transition from consciousness to unconsciousness.1516 Halothane also appears to cause EEG activation, followed by depression.1718 Unlike other more commonly used volatile anaesthetics, such as isoflurane, halothane will usually induce burst suppression only at concentrations that exceed the clinically relevant range.719 Interestingly, at 1.2 ED50 for propofol, we observed a paradoxical decrease in the SEF during application of the tetanic stimulus and the tail clamp. The mechanism by which this occurs is unknown, but has been reported before with noxious stimuli applied during isoflurane anaesthesia.2021 In the present study with halothane, the 1 and 3 Hz stimuli increased SEF and MEF in a manner similar to those occurring with the tail clamp and electrical tetanic stimulus. The 0.1 Hz stimulus does not normally cause neuronal wind-up and did not evoke significant EEG changes, probably reflecting much lower temporal summation compared with the 1 Hz, 3 Hz, tail clamp and tetanic stimuli.522
While many previous studies have reported anaesthetic effects on spontaneous EEG activity, few studies have examined the effect of noxious stimulation on cerebral activity and the electroencephalogram during anaesthesia.13–520 In general, when clinically relevant concentrations of anaesthetic are administered, noxious stimulation causes EEG activation,1520 although less EEG activation occurs during propofol anaesthesia.34 We found that propofol blunted EEG activation resulting from noxious stimulation.2 Hofbauer and colleagues23 investigated in humans the relationship between propofol administration, subjective pain ratings and cerebral activation, as determined by positron emission tomography. When propofol was infused at doses that caused mild sedation, pain ratings of noxious heat increased, as did neural activity in the anterior cingulate cortex and thalamus. As the propofol dose was increased, the pain rating decreased, as did the neural responses; however, even when unconsciousness occurred, noxious heat evoked increased activity in the cingulate cortex and thalamus.23 Greater propofol concentrations, however, are associated with blunted EEG responses to noxious stimulation,34 suggesting that this depressant effect occurs between propofol concentrations that produce unconsciousness and those needed to produce immobility.
Although the exact mechanisms by which propofol and halothane produce anaesthesia are unclear, emerging evidence suggests that action at specific ligand-gated receptors might be critically involved.6 Propofol acts at the GABAA receptor to enhance the effect of GABA. A mutation in the β3 subunit of the GABAA receptor renders mice resistant to propofol, but only slightly increases halothane requirements.24 Halothane probably acts at several receptors, including GABAA, glycine and NMDA receptors.6 Furthermore, propofol and halothane might have different modes of action with respect to sites within the central nervous system. The sedative effect of propofol probably occurs by actions at discrete supraspinal sites, including a sleep-promoting pathway in the tuberomamillary nucleus.8 In addition, propofol and halothane may act in the septohippocampal system to induce anaesthesia.25 Halothane appears to have an action in the spinal cord to ablate movement that occurs in response to noxious stimulation.7 Furthermore, propofol and volatile anaesthetics such as isoflurane can suppress nociception in the spinal cord and thereby affect EEG responses to noxious stimulation.226 Thus, in the present study, propofol and halothane could have acted in the brain directly to suppress the EEG response, and in the spinal cord to indirectly suppress the response.
The lower propofol infusion rate may have already surpassed the infusion rate needed to blunt the EEG responses. If so, this was not because of excessive EEG depression. We used the same fractions (0.8 and 1.2) of the amount needed to produce immobility, and our values for halothane MAC and propofol ED50 are similar to those previously published.2728 We found that SEF and MEF during propofol anaesthesia were greater than those during halothane anaesthesia. At the greater propofol infusion rate (720 µg kg−1 min−1), the MEF was ≈8 Hz. Tzabazis and colleagues29 used a modified MEF that incorporated the occurrence of spikes and burst suppression; these authors maintained a modified MEF of 3 Hz by infusing propofol at a dose similar to the higher dose used in the present study [730 (200) µg kg−1 min−1]. Although we did observe spikes in the EEG during propofol infusion, we did not routinely observe burst suppression. Antunes and colleagues14 observed burst suppression at infusion rates greater than those used in our study (1000 µg kg−1 min−1). Because of the modified MEF used by Antunes and colleagues, it is difficult to make a direct comparison with our data. Nonetheless, we believe that excessive EEG depression does not explain the stronger blunting effect of propofol on EEG activation by noxious stimuli.
In summary, we found that propofol, in a dose range that prevents movement, caused significant depression of EEG responses to noxious electrical and mechanical stimulation, while halothane did not.
This work was supported in part by grants NIH GM61283, GM57970 and a subcontract via P01-GM47818 to J.F.A.
1Department of Anesthesiology and Pain Medicine and 2Section of Neurobiology, Physiology and Behavior, University of California, Davis, CA, USA. 3Department of Anesthesiology, University of Medicine and Dentistry of New Jersey, Newark, NJ, USA