Anesth Analg 2005;100:413-436
© 2005 International Anesthesia Research Society
doi: 10.1213/01.ANE.0000142128.29660.AE
ANESTHETIC PHARMACOLOGY
Hyperresponsiveness on Washout of Volatile Anesthetics from Isolated Spinal Cord Compared to Withdrawal from Ethanol
Shirley M.E. Wong, MSc,
Sarah M. Sweitzer, PhD ,
Michael C. Peters, BS, and
Joan J. Kendig, PhD
Department of Anesthesia, Stanford University School of Medicine, Stanford, California.
 |
Abstract
|
|---|
We performed experiments in spinal cords isolated from neonatal rats to probe the mechanisms responsible for hyperresponsiveness of the population excitatory evoked potential (pEPSP) observed on washout of the volatile anesthetics halothane and isoflurane (1 minimal alveolar anesthetic concentration equivalent, MAC) compared with that observed after an anesthetic concentration of ethanol. After 30 min exposure to each anesthetic and washout, pEPSP area increased to levels significantly more than control (P < 0.01–0.001). Exposure to a very small (0.025 MAC) concentration of isoflurane over the same period itself produced a similarly exaggerated pEPSP (P < 0.05) in the continued presence of the drug, suggesting that the phenomenon is a direct excitatory effect of the small concentrations of anesthetic on washout, unlike the true withdrawal observed with ethanol. Isoflurane, but not halothane, significantly increased the amount of potassium-stimulated release of the excitatory neurotransmitters glutamate, aspartate, and substance P, suggesting the hyperresponsiveness for that drug is the result of a presynaptically mediated increase in transmitter release. A broad spectrum specific protein kinase C inhibitor, GF109203X, blocked ethanol withdrawal hyperresponsiveness but not hyperresponsiveness after halothane. If the behavioral symptoms of emergence from anesthesia are based on excitatory actions similar to those observed in the spinal cord, the results show that they represent direct excitatory actions rather than withdrawal and are attributable to direct actions on ion channels or receptors, rather than indirect effects mediated by protein kinase C.
|
Introduction
|
|---|
Emergence from inhaled general anesthesia is associated with a constellation of symptoms that indicate heightened excitability of the nervous system. These include hyperalgesia (1), agitation (2–4), hyperreflexia, and spasticity (5). The present study used a motor neuron response in isolated neonatal rat spinal cord to test whether heightened excitability on withdrawal from volatile anesthetics corresponded to an exaggerated motor neuron response to dorsal root stimulation, to test whether the phenomenon represented withdrawal or an excitatory effect of low anesthetic levels, and to test whether an increase in excitatory transmitter release contributed. The spinal cord is the appropriate part of the nervous system to probe anesthetic actions at the concentrations associated with an important anesthetic end-point, lack of movement in response to noxious stimulation (MAC), because the spinal cord has been shown to be the site at which MAC is predominantly determined (6,7). We have previously described, and partially characterized, ethanol withdrawal hyperresponsiveness in the same preparation (8). In spinal cord slice we have characterized ethanol withdrawal hyperresponsiveness in motor neurons (9) and shown it to be dependent on protein kinase C  (PKC) modulation of currents through the N-methyl-d-aspartic acid (NMDA) receptor (9,10). The present study compared volatile anesthetics and ethanol in intact isolated spinal cord with respect to the question of whether postexposure hyperresponsiveness after general anesthetic exposure is true withdrawal, as postethanol hyperresponsiveness is, and tested the involvement of PKC in the hyperresponsiveness associated with each class of drug.
|
Methods
|
|---|
As we have previously described (8), spinal cords were isolated from halothane-anesthetized Sprague-Dawley rat pups 1–4 days old in a protocol approved by Stanford University’s Institutional Animal Care and Use Committee. Cords were superfused with artificial cerebrospinal fluid (ACSF) of the following composition (in mm): NaCl 123, KCl 5, NaH2PO4 1.2, MgSO4 1.3, NaHCO3 26, CaCl2 2, dextrose 30. The ACSF was equilibrated with 95% O2-5% CO2, yielding a stable pH of 7.3–7.4. Temperature was recorded by a thermistor in the chamber close to the cord and maintained at 27°C–28°C, a temperature in the physiological range for rats of this age when not incubated by the mother.
A suction stimulating electrode was placed on a lumbar dorsal root 4–5 mm distal to the entrance of the root into the cord. A suction recording electrode was placed on an ipsilateral ventral root 1–3 segments away to record the population excitatory postsynaptic potential (pEPSP), which underlies the monosynaptic reflex without contamination by the compound action potential. Stimuli to the dorsal root were square wave pulses 0.4 ms in duration, nominal intensity 9.8 V, given at a constant frequency of 0.02 s–1. After the responses of the preparation were observed to be stable, control recordings were taken over a 30-min period. Responses were averaged in groups of 5 and digitally recorded.
Volatile anesthetics were prepared as saturated solutions and diluted to the desired concentration in the ACSF; ethanol was diluted to the desired concentration (100 mm) in the ACSF. This is an anesthetic concentration for adult rats (11). Most experiments with volatile anesthetics were also done at 1 minimal alveolar anesthetic concentration (MAC) equivalents. These were in accord with published MAC equivalents for rats as free aqueous concentrations as reported in an article that finds these an appropriate benchmark for in vitro experiments even uncorrected for temperature (12). To probe for an excitatory action of small anesthetic concentrations, some isoflurane experiments were done at 0.025 MAC. Concentrations in the superfusate were measured by gas chromatography of the vapor phase in equilibrium with the solution. Anesthetics were administered for 30 min followed by washout for 1 h. The broad spectrum PKC antagonist GF109203x (Sigma, St. Louis, MO) was dissolved in the ACSF and applied to the preparation for 30–60 min before and during anesthetic or drug exposure.
Area under the curve of the pEPSP was measured and normalized to the average area in the control period. The time course of anesthetic and ethanol actions was followed, and comparisons of withdrawal hyperresponsiveness were made at 60 min washout. Control studies were made with untreated preparations over the same time period as the experimental studies. Significance of differences was assessed by analysis of variance followed by a Bonferroni posttest for multiple comparisons, with significance set at a P level of 0.05.
To test the role of enhanced excitatory transmitter release in exaggerated responses, superfusate was collected at the end of the washout period and also after 10-min exposure to large potassium concentrations (30 mm). Superfusate samples were obtained from 4–5 different cords for each transmitter, and samples were run in duplicate. Glutamate and aspartate were assayed by high-performance liquid chromatography. Substance P and calcitonin gene-related peptide (CGRP) were assayed by enzyme-linked immunosorbent assay.
High-performance liquid chromatography methods were derived from Graser et al. (13). Samples (20 µL) were reacted with 1.0 µL of o-phthaldialdehyde/3-mercaptoproprionic acid reagent (Sigma) for derivitization. After 1 min, a 20-µL sample was injected into the high-performance liquid chromatography, which uses a reversed-phase C18 column (3.9 x 150 mm, Waters, Milford, MA). Glutamate and aspartate were quantitated at absorbance of 340 nm (Waters 2487). The samples were separated with a mobile phase gradient using the following solutions: 12.5 mm NaH2PO4, pH 7.2, and a 50% v/v mixture of 12.5 mm NaH2PO4 and acetonitrile. The aspartate peak appears at 3.3 min and the glutamate peak at 4.9 min after sample application when the mobile phase is isocratic phosphate buffer. The column was flushed by using a steep gradient (ending in 50% acetonitrile) before the next sample. This procedure allowed for the detection of glutamate and aspartate at a sensitivity of 0.5–2.0 ng/sample. A dose-response curve was generated on each day of sampling. A personal computer-based data acquisition and analysis system (Millennium, Waters) was used to quantify amino acid content based on the area under the curves.
For substance P and CGRP ELISA analysis, samples were analyzed using the rat CGRP enzyme immunoassay kit (SPI-Bio, Massy Cedex, France) or the Substance P enzyme immunoassay kit (Assay Designs, Inc., Ann Arbor, MI). Each sample was performed in duplicate as directed in the instructions. CGRP or Substance P concentrations were determined by reference to the sample absorbances of the standard curves generated using the recombinant rat CGRP or Substance P supplied with the kits.
|
Results
|
|---|
An example of halothane’s effects on the pEPSP is shown in Figure 1A, and the time courses of halothane, isoflurane, and ethanol depression and recovery on washout are shown in Figure 1B. In each case, there is a large depression of pEPSP area during anesthetic exposure followed by a marked increase to a level well above control on washout. The increase persists for at least 1 h; after this period declines may be the result of loss of viability of the preparation or to actual decrease in the effect.
View larger version (14K):
[in this window]
[in a new window]
|
Figure 1. A, An example of the population excitatory postsynaptic potential (pEPSP) in control conditions, depressed by 30 min exposure to an anesthetic concentration of halothane, and increasing to a level above control when the halothane is washed off for 60 min. B, Graphs of the time courses of depression and washout for anesthetic concentrations of halothane, isoflurane, and ethanol. Data points are means of 5–6 experiments; error bars are sem. C, preparations maintained under control conditions for the same duration as the anesthetic experiments show no increase in pEPSP area (n = 5).
|
|
To exclude the possibility that pEPSP area increased spontaneously over time sufficiently to account for the apparent rebound above control on washout, control preparations were maintained without anesthetic for a comparable period (Fig. 1C). Under control conditions there was no significant difference between the pEPSP area at 25 and 125 min. Depression at 30 min anesthetic exposure and washout 60 min after anesthetic withdrawal are illustrated graphically for the three anesthetics in Figure 2. For halothane, depression was to 37.8% ± 3.4% of control (mean ± sem), followed by a rebound on wash to 130.1% ± 9.4% (n = 7). Comparable figures for isoflurane depression and wash are 49.1% ± 6.3% and 120.3% ± 4.3% (n = 5), and comparable figures for ethanol depression and wash are 49.9% ± 3.1% and 124.2% ± 3.6% (n = 5). In each case both depression and washout areas are significantly different from each anesthetic’s own control (Fig. 2) and from untreated preparations at the same time point.
View larger version (18K):
[in this window]
[in a new window]
|
Figure 2. Graphs of the same data as shown in Figure 1B, showing control, anesthetic depression at 30 min, and recovery to levels above control at 60 min wash. **Significantly different from control at P < 0.01; ***significantly different from control at P < 0.001. Error bars are sem.
|
|
Two possibilities were considered for the event underlying the rebound to levels above control on washout. The first was that as anesthetic concentrations declined to low levels on wash, small anesthetic concentrations might have an excitatory effect. The second was that hyperresponsiveness is true withdrawal: a phenomenon induced by and requiring anesthetic exposure. To discriminate between these possibilities, preparations were exposed to a small (0.025 MAC) concentration of isoflurane and the area of the pEPSP was followed for 90 min, the same time as for anesthetic exposure and washout at the larger concentrations. This concentration was chosen because our earlier studies had indicated a possible excitatory effect (14). The results are shown in Figure 3. There was, in fact, a significant (P < 0.05) increase in pEPSP area after 90 min exposure to the small anesthetic concentration to 125.1% ± 7.4% of control (n = 5) suggesting that the rebound after volatile anesthetic depression and washout is in large part the result of an excitatory effect of the volatile anesthetic observed at low levels, rather than to a true withdrawal. This appears not to be the case for ethanol. In our previous studies there was no suggestion of any increase in pEPSPs in intact spinal cord (8,15) or in excitatory postsynaptic currents in motor neurons in spinal cord slices (16,17) in response to low levels of ethanol.
View larger version (23K):
[in this window]
[in a new window]
|
Figure 3. A, exposure to a low (0.025 minimal alveolar anesthetic concentration, MAC) level of isoflurane for a period equivalent to that for anesthetic exposure and washout shows a slow increase in the area of the population excitatory postsynaptic potential (pEPSP) in the continued presence of the anesthetic. B, the increase in pEPSP area is significant at 90 min but not at 30 min isoflurane exposure. n = 5, error bars are sem; *P < 0.05.
|
|
To test whether some of the volatile anesthetic hyperresponsiveness is the result of presynaptically-mediated increase in transmitter release, the excitatory neurotransmitters glutamate, aspartate, substance P, and CGRP were measured in the superfusate during the washout period and again after stimulation with KCl as described in Methods. The results are shown in Tables 1 and 2. Isoflurane, but not halothane, significantly increased the amount of K-stimulated release of glutamate, aspartate and substance P, suggesting that for this drug presynaptic actions on transmitter release may contribute to hyperresponsiveness.
View this table:
[in this window]
[in a new window]
|
Table 1. Glutamate and aspartate release during the washout period alone and after exposure to 30 mm KCl using high performance liquid chromatography
|
|
View this table:
[in this window]
[in a new window]
|
Table 2. Substance P and calcitonin gene-related peptide (CGRP) release during the washout period alone and after exposure to 30 mm KCl using ELISA
|
|
To probe the role of PKC in hyperresponsiveness on washout, preparations were treated with the broad spectrum specific PKC antagonist GF109203x (0.3 µm), then exposed to halothane or ethanol as in the experiments above. The PKC inhibitor was not tested with isoflurane. This concentration was chosen because preliminary experiments with ethanol showed it completely effective in blocking withdrawal hyperresponsiveness for this drug. The results are shown in Figure 4. GF109203x completely prevented ethanol withdrawal hyperresponsiveness but had no effect on halothane withdrawal. A limited number of experiments with 0.6 µm of the PKC inhibitor showed no indication that larger concentrations would inhibit halothane hyperresponsiveness. The inhibitor had no effect when applied alone (Fig. 4).
View larger version (10K):
[in this window]
[in a new window]
|
Figure 4. The broad spectrum protein kinase C (PKC) inhibitor GF109203x blocks the increase to levels above control on washout for ethanol but not halothane. A, in the presence of the inhibitor throughout the experiment there is no increase in population excitatory postsynaptic potential (pEPSP) area in the washout period after ethanol. B, In the presence of the inhibitor there is still an increase to levels above control after halothane exposure. C, the inhibitor alone has no effect. n = 4, error bars are sem.
|
|
|
Discussion
|
|---|
Although preparations exposed to either volatile anesthetics or ethanol recover to levels above control on washout, its basis for volatile anesthetics is attributable in large part to an excitatory effect of prolonged exposure to a low level of anesthetic, whereas ethanol exposure induces a true withdrawal when the ethanol is removed. Low levels of isoflurane produce an increase in response during exposure comparable in magnitude to that observed on washout, whereas no such excitation is observed with low levels of ethanol in intact cord (15). Isoflurane and halothane differ in the contribution of increased neurotransmitter release to the increased response; K-stimulated release of glutamate, aspartate, and substance P in the washout period is increased after isoflurane but not halothane, suggesting a presynaptic site for the excitatory actions of the former but not the latter anesthetic. Ethanol withdrawal and volatile anesthetic-induced hyperexcitability differ in dependence on PKC. A PKC inhibitor blocks the increase to levels more than control after exposure to ethanol but not to halothane. We did not explore the PKC dependence of isoflurane excitation; it is possible that the two anesthetics may differ in this respect.
Early in the history of inhaled anesthetics several stages were described during induction, including analgesia, followed by an excitement phase, then surgical anesthesia (18). The excitement phase consisted of several signs of nervous system hyperexcitability. More recently, a detailed analysis of the behavioral manifestations of hyperexcitability during emergence has been reported, as outlined in the Introduction (1–5). Both induction and emergence are associated with relatively low subanesthetic levels of inhaled drug. If the hyperresponsiveness we observe during washout or exposure to low levels of a volatile anesthetic underlies the behavioral manifestations of excitement, then PKC plays no role in the latter. Nor does PKC appear to affect the anesthetic properties of inhaled anesthetics as measured by the end-point MAC, immobility after a noxious stimulus. Isozyme-specific inhibitors of PKC and PKC, applied intrathecally in quantities that block hyperalgesic responses to formalin, did not alter halothane MAC (19). Thus, for halothane, neither the excitement to low levels of exposure nor immobility at anesthetic levels is attributable to an indirect action mediated by PKC. By implication, these actions are rather the result of direct effects on receptors or ion channels.
Halothane differs from isoflurane in that the excitatory effects in spinal cord are mediated in part by presynaptic increases in transmitter release for the latter but not the former drug. Halothane and isoflurane are from different classes of compounds, halothane being a fluorinated ethane, isoflurane a fluorinated ether. In studies at levels just less than and just more than MAC, isoflurane, but not halothane, depressed spinal levels of c-fos induced by noxious stimulation, which might be attributable to a presynaptic action (20). On the other hand, halothane, but not isoflurane MAC, was associated with reduction in levels of activity in dorsal horn neurons (21), suggesting that halothane might bring about immobility by actions in the dorsal horn and that isoflurane might act more ventrally. Thus the two drugs appear to differ in their mechanisms and sites of action in spinal cord.
In previous studies we have reported the withdrawal after ethanol to be a postsynaptic phenomenon in motor neurons, mediated by NMDA receptors and dependent on PKC (9). More recently, we have shown that withdrawal in motor neurons is dependent on translocation of the specific PKC isozyme PKC (10). In whole animal behavioral studies of ethanol withdrawal, we have shown that hyperalgesia during the withdrawal period after a single injection of ethanol is the result of both PKC and PKC isozymes (Shumilla et al., unpublished data). In early studies in isolated spinal cord, we observed an increase in evoked response at the smallest isoflurane concentration tested (14) but the actions of ethanol appear purely depressant (8,15). However, in the case of ethanol there is an acute tolerance that is manifested as a partial recovery toward control levels in the continued presence of ethanol and differing from ethanol withdrawal in that it is dependent on metabotropic glutamate receptors (22).
The results of the present study show that inhaled anesthetics differ from ethanol in several respects. On wash, recovery to levels above control represents withdrawal from ethanol, whereas, for halothane, it is a direct excitatory action of the anesthetic. Furthermore, ethanol withdrawal, but not halothane excitation, is attributable to actions mediated by PKC. Halothane and isoflurane differ in the role played by presynaptic increase in excitatory neurotransmitter release in hyperresponsiveness.
We are indebted to the laboratory of E. I. Eger II at the University of California, San Francisco, for the measurement of ethanol and volatile anesthetic concentrations.
|
Footnotes
|
|---|
Dr. Peters died August 22, 2004.
Supported by National Institutes of Health grants NS13108 and GM47817 (to JJK), and by National Institutes of Health grant NS4472901 and the Alejandro and Lida Zaffaroni Innovation Fund for Addiction Research (to SMS).
Accepted for publication July 22, 2004.
Address correspondence to Joan J. Kendig, Department of Anesthesia, Stanford University School of Medicine, Stanford, CA 94305-5117. Address e-mail to kendig{at}stanford.edu.
|
References
|
|---|
- Flood P, Daniel D. Pronociceptive actions of isoflurane: a protective role for estrogen. Anesthesiology 2003;99:476–9.[Web of Science][Medline]
- Cohen IT, Hannallah RS, Hummer KA. The incidence of emergence agitation associated with desflurane anesthesia in children is reduced by fentanyl. Anesth Analg 2001;93:88–91.[Abstract/Free Full Text]
- Cravero JP, Beach M, Dodge CP, Whalen K. Emergence characteristics of sevoflurane compared to halothane in pediatric patients undergoing bilateral pressure equalization tube insertion. J Clin Anesth 2000;12:397–401.[Web of Science][Medline]
- Voepel-Lewis T, Malviya S, Tait AR. A prospective cohort study of emergence agitation in the pediatric postanesthesia care unit. Anesth Analg 2003;96:1625–30.[Abstract/Free Full Text]
- McCulloch PR, Milne B. Neurological phenomena during emergence from enflurane or isoflurane anaesthesia. Can J Anaesth 1990;37:739–42.[Web of Science][Medline]
- Rampil IJ, Mason P, Singh H. Anesthetic potency (MAC) is independent of forebrain structures in the rat. Anesthesiology 1993;78:707–12.[Web of Science][Medline]
- Antognini JF, Schwartz K. Exaggerated anesthetic requirements in the preferentially anesthetized brain. Anesthesiology 1993;79:1244–9.[Web of Science][Medline]
- Wong SM, Tauck DL, Fong EG, Kendig JJ. Glutamate receptor-mediated hyperexcitability after ethanol exposure in isolated neonatal rat spinal cord. J Pharmacol Exp Ther 1998;285:201–7.[Abstract/Free Full Text]
- Li HF, Kendig JJ. Ethanol withdrawal hyper-responsiveness mediated by NMDA receptors in spinal cord motor neurons. Br J Pharmacol 2003;139:73–80.[Web of Science][Medline]
- Li HF, Kendig JJ. Protein kinase C mediates ethanol withdrawal hyperresponsiveness of NMDA receptor currents in spinal cord motor neurons. Br J Pharmacol. In press.
- Fang Z, Gong D, Ionescu P, et al. Maturation decreases ethanol minimum alveolar anesthetic concentration (MAC) more than desflurane MAC in rats. Anesth Analg 1997;84:852–8.[Abstract]
- Franks NP, Lieb WR. Temperature dependence of the potency of volatile general anesthetics: implications for in vitro experiments. Anesthesiology 1996;84:716–20.[Web of Science][Medline]
- Graser TA, Godel HG, Albers S, et al. An ultra rapid and sensitive high-performance liquid chromatographic method for determination of tissue and plasma free amino acids. Anal Biochem 1985;151:142–52.[Web of Science][Medline]
- Savola MK, Woodley SJ, Maze M, Kendig JJ. Isoflurane and an alpha 2-adrenoceptor agonist suppress nociceptive neurotransmission in neonatal rat spinal cord. Anesthesiology 1991;75:489–98.[Web of Science][Medline]
- Wong SM, Fong E, Tauck DL, Kendig JJ. Ethanol as a general anesthetic: actions in spinal cord. Eur J Pharmacol 1997;329:121–7.[Web of Science][Medline]
- Wang MY, Rampil IJ, Kendig JJ. Ethanol directly depresses AMPA and NMDA glutamate currents in spinal cord motor neurons independent of actions on GABAA or glycine receptors. J Pharmacol Exp Ther 1999;290:362–7.[Abstract/Free Full Text]
- Wang MY, Kendig JJ. Patch clamp studies of motor neurons in spinal cord slices: a tool for high-resolution analysis of drug actions. Acta Pharmacol Sin 2000;21:507–15.[Web of Science][Medline]
- Guedel AE. Inhalation anesthesia: a fundamental guide. New York: The MacMillan Company, 1937.
- Shumilla JA, Sweitzer SM, Eger EI II, et al. Inhibition of spinal protein kinase cepsilon or gamma isozymes does not affect halothane MAC in rats. Anesth Analg 2004;99:82–4.[Abstract/Free Full Text]
- Jinks SL, Antognini JF, Martin JT, et al. Isoflurane, but not halothane, depresses c-fos expression in rat spinal cord at concentrations that suppress reflex movement after supramaximal noxious stimulation. Anesth Analg 2002;95:1622–8.[Abstract/Free Full Text]
- Jinks SL, Martin JT, Carstens E, et al. Peri-MAC depression of a nociceptive withdrawal reflex is accompanied by reduced dorsal horn activity with halothane but not isoflurane. Anesthesiology 2003;98:1128–38.[Web of Science][Medline]
- Li HF, Wang MY, Knape J, Kendig JJ. Ethanol tachyphylaxis in spinal cord motorneurons: role of metabotropic glutamate receptors. Br J Pharmacol 2003;138:1417–24.[Web of Science][Medline]
|
Top
Abstract
Introduction
