| JOURNAL HOME | CME HOME | THIS MONTH | PAST ISSUES | ETOC | COLLECTIONS |
| AUTHORS | REVIEWERS | EDITORIAL BOARD | FEEDBACK | RSS | HELP |
 
|
|
|
||
|
|
© 1999 International Anesthesia Research Society
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
Abstract |
|---|
 Top Abstract IntroductionMethodsResultsDiscussionReferences |
|---|
Isoflurane depresses the number and amplitude of I waves of the motor-evoked potential produced by transcranial electrical stimulation of the motor cortex and thus affects components of the corticospinal volley that are believed to arise from Laminae III and V. This study extends these observations to sevoflurane (9 patients) and the two IV anesthetics, thiopental and propofol (10 sets of observations in 10 and 6 patients, respectively). The patients’ ages ranged from 10 to 17 yr. Sevoflurane was administered to achieve end-tidal concentrations of 0.5%–3%. Thiopental and propofol were given as boluses of 5 mg/kg and 2 mg/kg, respectively, to patients anesthetized with nitrous oxide, fentanyl, midazolam, and a muscle relaxant. Sevoflurane had a depressant effect on I waves essentially similar to that of isoflurane; thiopental depressed I wave activity by an average of 33% (95% confidence interval: 20%–46%, P < 0.001) and propofol by 39% (95% confidence interval: 20%–40%, P < 0.001). With all three anesthetics, later I waves showed the most amplitude depression. The three anesthetics had qualitatively similar effects on I waves.
Implications: Sevoflurane, thiopental, and propofol depress components of the corticospinal volley produced by transcranial electrical stimulation of motor cortex in a manner qualitatively similar to isoflurane. The findings indicate that anesthetics with primarily hypnotic actions suppress interneuronal activity in cerebral cortex.
|
Introduction |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
During scoliosis surgery, we monitor spinal cord function using simultaneous transcranial electrical stimulation of the motor cortex and stimulation of the tibial nerves to produce descending corticospinal and ascending somatosensory volleys (1–3). These volleys are recorded from the spinal cord using electrodes in the epidural space and can be distinguished by latency and, with two epidural leads, by their direction of propagation (Fig. 1).
|
The corticospinal volley set up by a single stimulus to motor cortex consists of multiple components, that is, early components caused by direct stimulation of corticospinal axons (D wave) and later waves caused by stimulation of the axons of cortico-cortical interneurons, which have excitatory projections to the corticospinal neurons (4–8). This produces indirect volleys (I waves) in axons that have responded in the D wave. The I waves are generated in the superficial cortical laminae, mainly Laminae III and V (4), and with transcranial electrical stimulation, they are normally small deflections of higher threshold and longer latency than the D wave.
When added to a background anesthetic of fentanyl and nitrous oxide, isoflurane depresses the I waves of the corticospinal volley (motor evoked potential [MEP]), but has minimal effects on the shorter-latency D waves (6,7). Volatile anesthetics primarily produce unconsciousness: they may reduce analgesic requirements in a clinical situation (9), but in standard animal experiments they either have little analgesic effect or antagonize analgesia induced by nitrous oxide and narcotics (10). Propofol and thiopental are also primarily hypnotics, producing little or no analgesia (11). The present studies address the ability of primarily hypnotic anesthetics to depress I waves when added to a baseline regimen that is primarily analgesic and amnesic (12).
The question addressed by the present work was whether I waves may be used as an indication of transmission through oligo- and polysynaptic circuits in the cerebral cortex, thereby providing a model for testing anesthetic actions. If this is a valid model, one might expect different hypnotic/sedative anesthetics to have qualitatively similar effects on I wave generation. This study assesses the effects of increasing concentrations of sevoflurane and of boluses of the standard induction doses of thiopental and propofol, to determine whether I wave depression is a general effect of anesthetics that act primarily by producing unconsciousness rather than analgesia.
|
Methods |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Our subjects were 25 patients, aged from 10 to 17 yr, undergoing surgical correction of scoliosis. The effects of inhaled sevoflurane were studied in 9 patients, those of IV thiopental (5 mg/kg) in 10 patients, and those of IV propofol (2 mg/kg) once in 2 patients and twice in another 4 patients (i.e., 10 observations). Eighteen patients suffered from idiopathic scoliosis, 4 from Duchenne muscular dystrophy, and 3 from either a spinal cord arterio-venous malformation, a spinal cord tumor with previous surgery, or a congenital vertebral anomaly. Patients with Duchenne dystrophy underwent Luque instrumentation and the others Cotrel-Dubousset instrumentation. The protocol was approved by the institutional Research Ethics Committee, and written, informed consent was obtained from the patients and/or their parents, as appropriate.
Patients were not given premedication. Induction was with thiopental, approximately 5 mg/kg, fentanyl 1.5 µg/kg, and vecuronium 0.1 mg/kg. The patients were tracheally intubated, and cannulae and probes inserted (a radial artery cannula for blood pressure monitoring, a nasopharyngeal temperature probe, and a second peripheral venous cannula, one for IV fluids and one for the administration of sodium nitroprusside). The patients were then placed in a prone position on a four-posted frame. Temperatures were maintained above 35°C with a hot air blanket over the legs, a heated humidifier in the anesthesia circuit, and a warming blanket under the support frame. A peripheral nerve stimulator was used and repeat boluses of vecuronium given so that fewer than three twitches were present after a train-of-four stimulus.
During patient positioning and the initial stages of the operation, anesthesia was supplemented with isoflurane 0.5%–1% in the patients who received IV study drugs, and sevoflurane in the other patients. Nitrous oxide 60%–70% was administered throughout the procedures. End-tidal carbon dioxide and volatile anesthetic concentrations were monitored using a Datex Capnomac Ultima (Datex Instrumentation, Helsinki, Finland), with the end-tidal carbon dioxide tension maintained at approximately 35 mm Hg. Mean arterial blood pressure was maintained at 50–60 mm Hg using sodium nitroprusside and fentanyl.
Before either of the IV anesthetics was given, the isoflurane administration was ceased, and the end-tidal concentration allowed to decrease to <0.1% for 10 min. Whenever a low concentration of volatile anesthetic was present, midazolam 0.05–0.07 mg/kg and fentanyl approximately 1.5 µg/kg were given, and repeated as necessary. A baseline MEP was recorded, and the study drug was injected as a bolus. Patients therefore acted as their own control. Further recordings were made at 1-min intervals until the MEP returned to baseline. Unless otherwise specified, the data below for the IV anesthetics represent their maximal depressant effects. In four of the patients given propofol, the bolus injection of propofol was repeated after the MEP had stabilized for >10 min. Because of the slow clearance of thiopental, this drug was given only once to each patient. Three patients receiving sevoflurane became hypertensive at low concentrations, and baseline recordings were not obtained. In the other six patients, recordings were obtained between 0% and 3%. The concentration was increased by 0.5% steps to 3%. At each step, the concentration was stabilized for 10 min. In some patients, recordings were made at zero concentration more than once. The average of these was taken as the control, and all points appear in the scattergram of Figure 2.
|
The technique of evoked potential monitoring has been described (1–3). After the patients were placed in the prone position, spiral needle electrodes were placed subdermally in the scalp at the vertex and 7 cm laterally for motor cortex stimulation. This was done using an isolated cortical stimulator with low output impedance, capable of providing a capacitatively coupled discharge of up to 1500 V. The output was regulated by a calibrated potentiometer with 10 equal steps. Stimulus intensity was 375–525 V, adequate to produce volleys containing the D wave and, when the patients received only the baseline anesthetic, multiple I waves (Fig. 3). Corticospinal volleys (the MEP) were recorded at the same time as the ascending somatosensory volleys produced by stimulation of the tibial nerves in the popliteal fossae, as described previously (3). The tibial nerve stimulus was delivered simultaneously with the cortical stimulus, once every 3 s. The ascending somatosensory volley reached the epidural electrodes after the descending corticospinal volley (Fig. 1); it could influence subsequent corticospinal volleys only if its effects on cortical excitability lasted >3 s. In unpublished studies using a comparable anesthetic regimen, we have been unable to demonstrate significant changes in motor cortex excitability within 100–200 ms of the stimulus, and an effect at 3 s is therefore most unlikely.
|
Bipolar recording electrodes were inserted by the surgeon into the epidural space at approximately T1-2 and L1, above and below the operative site respectively, and were pushed cephalad 2–6 cm. A Medelec/Teca Sapphire 4ME EMG/evoked potential machine (Medelec, Old Woking, Surrey, UK) was used with filter settings of 500 Hz to 5 kHz. Each recording was the average of 10 single trials, recorded at 3-s intervals. Only patients in whom recognizable and reproducible I waves (>0.5 µV) could be obtained were studied. For the studies of IV drugs, a baseline recording was made after the electrodes were inserted and the end-tidal isoflurane concentration had been stable at <0.1% for at least 10 min. The drug bolus was given and repeat recordings made for 3–8 min, until the maximal effect was seen to be waning (Figs. 3–5). The number of recordings made in each patient was influenced by the technical and operative factors. When sevoflurane was studied, recordings were made after the end-tidal concentration had been stable for 10 min. The traces were stored on floppy disk and measured off-line.
|
|
Wave identification and measurement followed the same protocol as in previous papers (1–3,6,7). Amplitude was measured peak-to-peak to the nearest 0.25 µV, and all measurements were of traces obtained at the low-cervical electrode (Fig. 3). For each patient, a unitary measure of I wave activity was obtained by summing the amplitudes of the individual I waves. The maximal depression of the sum of the I waves was noted for each IV anesthetic and the time after the bolus when it occurred.
Ten pairs of observations were obtained for each of the IV drugs, and 44 observations were taken at 0.5%, 1%, 1.5%, 2%, 2.5%, and 3% in the sevoflurane group. Mean, SD, and 95% confidence intervals were calculated and graphs drawn, using the SPSS computer package (SPSS Inc., Chicago, IL). Probabilities were calculated using the one-sample t-test (13). The significance level was taken as 5%. Each patient acted as his or her own control.
|
Results |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Introduction of each of the anesthetics depressed the number of I waves and the amplitudes of remaining I waves. With sevoflurane, these effects were dose-dependent (Fig. 3B). Propofol and thiopental were administered as bolus injections, but the gradual attenuation of I waves and then their gradual recovery after the peak effect are consistent with similar dose dependency.
In the control recordings for thiopental and propofol (10 for each), I waves were recorded consistently: I1, I2, and I3 in all 20 recordings, with mean amplitudes of 2.2, 3.1, and 4.0 µV respectively; I4 in 18 recordings, with mean amplitude of 3.0 µV; I5 in 12 recordings, with mean amplitude of 2.55 µV; I6 in 5 recordings and I7 in 2 recordings, both with a mean amplitude of 1.74 µV. The sum of I wave amplitudes was suppressed by the administration of thiopental or propofol in 9 of 10 recordings for each (Figs. 4 and 5).
The time to the maximal recorded depression of I-wave activity was variable in different patients, from 0.5 to 7 min and had a median value of 2 min for thiopental and 2.75 min for propofol. Nevertheless, in the four patients in whom bolus injections of propofol were studied twice, the times to maximal I wave depression were reasonably reproducible: 1 and 2.5 min, 0.5 and 1.5 min, 3 and 4 min, and 5 and 7 min. For each IV drug, the line of best fit for the pooled data on the I wave amplitude versus time graph was consistent with the expected time course of its action (Figs. 4 and 5). This suggests a drug level/effect relationship similar to that for sevoflurane (see below).
With propofol, the 10 sets of observations showed a mean decrease in I wave activity of 39% (range, 6.2% to -79.1%; Fig. 4A). The 95% confidence interval for the mean ranges from -20% to -59% (P < 0.001 by one-sample t-test). After thiopental, the maximal decrease in the I wave activity was, on average, 33% (range, 13% to -54%; Fig. 4B). The 95% confidence interval for the mean was from -20% to -46% (P < 0.001 by one-sample t-test).
After the administration of both drugs, I1, I2, and I3 underwent minimal change or even a slight increase in amplitude that was not statistically significant (Fig. 6). However, after propofol administration, the mean depression of the amplitude of I4 and I5 was 61% and 65%, respectively, and after thiopental administration, the mean depression of the amplitude of I4 and I5 was 24% and 62%, respectively. The changes in amplitude of I4 and I5 after propofol, and of I5 after thiopental, were statistically significant (all P < 0.01 by one-sample t-test).
|
In addition to attenuating I wave amplitudes, both IV anesthetics reduced the number of I waves: in the 10 recordings, from 45 to 37 in the case of propofol, and from 52 to 46 in the case of thiopental. This decrease was a result of the loss of the later I waves (I4 and I5).
Sevoflurane depressed total I wave activity in a similar manner to that previously reported for isoflurane (6), at equipotent doses (Fig. 2). The maximal concentration studied for which there was a matching control at zero concentration was 3% (1.5 minimum alveolar anesthetic concentration [MAC]). At this concentration, the amplitude of the summed I waves was depressed by approximately 70%. The relationship between the end-tidal concentration of sevoflurane and I wave activity was approximately linear, and projection of the linear regression line suggests that 100% depression would have been reached at 4% sevoflurane (2 MAC). A similar result was found when the data points of our previous study of isoflurane (6) were superimposed (Fig. 2).
During the administration of sevoflurane anesthesia, the number of I waves recorded varied between patients. The least sensitive I wave was I3, which was present in all recordings made with sevoflurane, although its amplitude was greatly attenuated by the larger sevoflurane concentrations. I1 and I2 were recordable during sevoflurane anesthesia in all but two patients. The later I waves, I4 to I6, were less frequent in control recordings without sevoflurane and, when present, proved to be very sensitive to sevoflurane. I6 was recorded in only one patient, at 0% sevoflurane. In this patient, an I5 was detectable with sevoflurane concentrations up to 3%. In all other patients, I5 was abolished by 2% sevoflurane or less.
Of the individual I waves, I3 was the largest at 0% sevoflurane and decreased rapidly in amplitude as the sevoflurane concentration was increased, so that, with 3% sevoflurane, the first three I waves were of similar size (Fig. 7).
|
|
Discussion |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Each of the anesthetics studied decreased the number of I waves and depressed their amplitude. The effects were greater for late I waves (I4–I6) and sometimes spared the early I waves (I1 and I2), possibly because the cortical interneuronal chain is longer for the later I waves. The doses of the IV anesthetics were the standard induction doses, and at these dosages, both propofol and thiopental reduced the summed amplitudes of I waves by approximately one third. Sevoflurane had similar effects on I waves to isoflurane (6). These volatile anesthetics each attenuated I wave activity by approximately 50% at a concentration of one MAC. These data confirm that, when monitoring spinal cord function using epidural recordings of corticospinal volleys (3), the D wave is relatively resistant to commonly used anesthetics.
Our previous study showed no effect of 1% or 2% isoflurane on the mean amplitude of I1 or I2, but dramatic reduction or elimination of I3 and later waves. Likewise, the IV anesthetics had a greater effect on the later I waves, but the changes were significant at the 5% level (one-sample t-test) only for I4 and I5 after propofol and for I5 after thiopental. Some of these minor differences in the effects of each anesthetic are probably because the subjects studied were, of necessity, different for each anesthetic. We conclude that the three anesthetics we studied, and isoflurane studied previously, have qualitatively similar effects on the processes responsible for I waves.
Studies in the cat and monkey have shown that I waves produced by surface stimulation probably result from the combined excitation of many different neural elements. However, deep tangential cortico-cortical fibers in Laminae III and V play a central role in their generation (4,5,14,15). It is therefore possible that the susceptibility of I waves to thiopental, propofol, isoflurane, and sevoflurane reflects actions on cortico-cortical interneurons in Laminae III and V. This is consistent with the results of experiments on rats using depth electrodes, in which the responses of neurons in Laminae III and V of sensory cortex to a peripheral stimulus were particularly sensitive to depression by general anesthetics, including volatile anesthetics and barbiturates. In these experiments midazolam had little effect, and propofol showed less effect than the barbiturates (16). Admittedly, these data are for the sensory cortex of a rodent, but the parallel with the present findings remains.
The anesthetics studied here (and isoflurane) are primarily hypnotics and were added to a baseline anesthetic of nitrous oxide, midazolam, and fentanyl, which relies heavily on a combination of analgesia and amnesia (12,17,18). Their usage is a potentially confounding factor, but without them the studies would not have been possible. While producing acceptable clinical anesthesia, nitrous oxide, midazolam, and fentanyl do not prevent the recording of I waves (6,7; present study), and we have not noted I wave depression after bolus injections of midazolam or fentanyl. Moreover, although we recognize that anesthetics cannot be rigidly classified, there are differences between the study drugs and the drugs used for baseline anesthesia. Under experimental conditions, using the isolated forearm technique and an anesthetic of alfentanil and midazolam, patients are sometimes able to respond purposefully to verbal stimuli, demonstrating that central processing is able to occur (12). No such processing has been demonstrated with the study drugs, at more than minimal doses. Others have drawn the distinction between "general anesthetics" that have specific receptors and sites of action, including narcotics and benzodiazepines and those that do not (19). As the analgesia produced by nitrous oxide is antagonised by naloxone and 
2 antagonists (20), it shares some of the properties of the former group, at least in clinical concentrations.
We conclude that thiopental, propofol, sevoflurane, and isoflurane depress interneurons in Laminae III and V of the motor cortex, thus attenuating I waves produced by transcranial electrical stimulation. This does not imply that only cortical interneurons in these laminae are affected by general anesthetics. However, if consciousness depends on functioning interneuronal circuits, this I wave model may provide insights into anesthetic actions on processes related to consciousness.
|
Acknowledgments |
|---|
Supported by the National Health and Medical Research Council of Australia.
|
References |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
- Hicks RG, Burke D, Stephen J. Monitoring spinal cord function during scoliosis surgery with Cotrel-Dubousset instrumentation. Med J Aust 1991;154:82–6.[Web of Science][Medline]
- Boyd SG, Rothwell JC, Cowan JMA, et al. A method of monitoring function in corticospinal pathways during scoliosis surgery with a note on motor conduction velocities. Psychiatry 1986;49:251–7.
- Burke D, Hicks R, Stephen J, et al. Assessment of corticospinal and somatosensory conduction simultaneously during scoliosis surgery. Electroencephalogr Clin Neurophysiol 1992;85:388–96.[Web of Science][Medline]
- Patton HD, Amassian VE. Single- and multiple-unit analysis of cortical stage of pyramidal tract activation. J Neurophysiol 1954;17:343–63.
- Rothwell JC, Thompson PD, Day BL, et al. Stimulation of the human motor cortex through the scalp. Exp Physiol 1991;76:159–200.[Web of Science][Medline]
-
Hicks RG, Woodforth IJ, Crawford MR, et al. Some effects of isoflurane on the motor evoked potential. Br J Anaesth 1992;69:130–6.
[Abstract/Free Full Text] -
Hicks R, Burke D, Stephen J, et al. Corticospinal volleys evoked by electrical stimulation of human motor cortex after withdrawal of volatile anaesthetics. J Physiol (Lond) 1992;456:393–404.
[Abstract/Free Full Text] -
Rothwell JC, Burke D, Hicks RG, et al. Transcranial electrical stimulation of the motor cortex in man: further evidence for the site of activation. J Physiol (Lond) 1994;481:243–50.
[Abstract/Free Full Text] - Wee MY, Hasan MA, Thomas TA. Isoflurane in labour. Anaesthesia 1993;48:369–72.[Web of Science][Medline]
-
Goto T, Marota JJ, Crosby G. Volatile anaesthetics antagonise nitrous oxide and morphine-induced analgesia in the rat. Br J Anaesth 1996;76:702–6.
[Abstract/Free Full Text] - Miller R , ed. Anesthesia. New York:Churchill Livingston, 1990:263.
-
Russell IF. Midazolam-alfentanyl: an anaesthetic? An investigation using the isolated forearm technique. Br J Anaesth 1993;70:42–6.
[Abstract/Free Full Text] - Bland M. An introduction to medical statistics. 2nd ed. Oxford:Oxford Medical Publications, 1995:155–9.
-
Landau WM, Bishop GH, Clare MH. Site of excitation in stimulation of the motor cortex. J Neurophysiol 1965;28:1206–22.
[Free Full Text] - Amassian VA, Stewart M, Quirk GJ, Rosenthal JL. Physiological basis of motor effects of a transient stimulus to cerebral cortex. Neurosurgery 1987;20:74–92.[Web of Science][Medline]
-
Angel A. Central neuronal pathways and the process of anaesthesia. Br J Anaesth 1993;71:148–63.
[Free Full Text] -
Cormack RS. Conscious levels during anaesthesia. Br J Anaesth 1993;71:469–71.
[Free Full Text] - Hug CC. Does opioid "anesthesia" exist? Anesthesiology 1990;73:1–4.[Web of Science][Medline]
- Newton DEF. Depth of anaesthesia. Anaesthesia 1993;48:367–8.[Web of Science][Medline]
-
Guo TZ, Poree L, Golden W, et al. Antinociceptive response to nitrous oxide is mediated by supraspinal opiate and spinal 2 adrenergic receptors in the rat. Anesthesiology 1996;85:846–52.[Web of Science][Medline]
This article has been cited by other articles:
|
|
 |
|
  P. C. Reinacher, H.-J. Priebe, W. Blumrich, J. Zentner, and K. M. Scheufler The effects of stimulation pattern and sevoflurane concentration on intraoperative motor-evoked potentials. Anesth. Analg., March 1, 2006; 102(3): 888 - 895. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. Nathan, F. Tabaraud, F. Lacroix, D. Moulies, X. Viviand, A. Lansade, G. Terrier, and P. Feiss Influence of propofol concentrations on multipulse transcranial motor evoked potentials Br. J. Anaesth., October 1, 2003; 91(4): 493 - 497. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
N. Boisseau, M. Madany, P. Staccini, G. Armando, F. Martin, D. Grimaud, and M. Raucoules-Aime Comparison of the effects of sevoflurane and propofol on cortical somatosensory evoked potentials Br. J. Anaesth., June 1, 2002; 88(6): 785 - 789. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Mori, X. Zhao, Y. Zuo, G. L. Aistrup, K. Nishikawa, W. Marszalec, J. Z. Yeh, and T. Narahashi Modulation of Neuronal Nicotinic Acetylcholine Receptors by Halothane in Rat Cortical Neurons Mol. Pharmacol., April 1, 2001; 59(4): 732 - 743. [Abstract] [Full Text]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
