| JOURNAL HOME | CME HOME | THIS MONTH | PAST ISSUES | ETOC | COLLECTIONS |
| AUTHORS | REVIEWERS | EDITORIAL BOARD | FEEDBACK | RSS | HELP |
| ||||||||||||||
| ||||||||||||||
|
|
|
|
||||||||||||
|
||||||||||||||
From the *Department of Anesthesiology and Pain Medicine; Section of Neurobiology, Physiology and Behavior, University of California, Davis, California; and Department of Anesthesia and Perioperative Care, University of California, San Francisco, California.
Address correspondence to Joseph F. Antognini, MD, Department of Anesthesiology and Pain Medicine, TB-170, University of California, Davis, CA 95616. Address e-mail to jfantognini{at}ucdavis.edu.
 |
Abstract |
|---|
 Top Abstract IntroductionMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
METHODS: The lumbar spinal cord was exposed in rats anesthetized with isoflurane. Mechanically ventilated rats were placed into a hyperbaric chamber and needle electrodes were inserted into the hindpaws. Isoflurane administration was discontinued and anesthesia converted to N2O by pressurizing the chamber with N2O. A microelectrode was inserted into the lumbar cord using computer-controlled motors and a hydraulic microdrive. Neuronal responses to electrical stimulation of the hindpaw were sought at 1.5, 2, and 2.5 atm N2O (0.8–1.3 minimum alveolar concentration).
RESULTS: Increasing N2O partial pressures variably affected neuronal responses to a 2 s 100-Hz electrical stimulus. Neuronal depth and neuronal response were correlated, with superficial neurons tending to be facilitated, while deeper neurons were depressed; (overall responses were 1331 ± 408, 1594 ± 383, and 1578 ± 500 impulses/min at 1.5, 2, and 2.5 atm N2O, respectively; mean, standard error). N2O did not affect neuronal responses to a repetitive "windup" stimulus. Infusion of the N-methyl-d-aspartate blocker MK-801 into separate rats increased the neuronal response to the 100-Hz stimulus (from 781 ± 216 to 1352 ± 269 impulses/min, P < 0.05).
CONCLUSIONS: N2O facilitated superficial spinal neuronal responses to noxious stimulation while depressing deeper neurons. These results suggest that anesthetic partial pressures of N2O have divergent effects on spinal neuronal responses to noxious stimulation, the specific responses depending on the depth of the spinal neurons.
|
Introduction |
|---|
|
TopAbstractIntroductionMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
The present study describes the effects of anesthetic partial pressures of N2O on spinal neuronal responses to noxious electrical stimulation in a hyperbaric chamber. We hypothesized that N2O would depress neuronal responses at partial pressures (1.5–2.5 atm) that produce immobility (0.8–1.3 minimum alveolar concentration, MAC). Furthermore, we hypothesized that MK-801, which blocks the NMDA receptor and decreases N2O MAC, would depress neuronal responses to noxious stimulation.
|
METHODS |
|---|
|
TopAbstractIntroductionMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
The rat was then placed into a previously described custom built hyperbaric chamber (11,12). The chamber was a 130-cm long acrylic tube (33 cm inside diameter, wall thickness 1.27 cm) that could be closed at each end with two large metal plates (sealed against the tube with rubber gaskets) secured to each other with six long bolts external to the tube. A rodent ventilator was placed inside the chamber. While the chamber was still open and the rat was being secured, gas flow to the ventilator was maintained from the outside using 40%–50% oxygen, 50%–60% N2O and 0.8%–1% isoflurane, which passed through valves in one of the metal plates. A small tube containing soda lime removed carbon dioxide from this rebreathing circuit.
A tungsten electrode (10 M
), attached to a motorized hydraulic microdrive (Kopf Instruments) was inserted into the lumbar cord (approximately at L4–5 level). The ipsilateral hindpaw was stimulated using touching and mild pinching to elicit background neuronal activity. Platinum needle electrodes (E-2, Astro-Med, West Warwick, RI) were inserted into the hindpaw, with one needle placed into the distal aspect of a toe and the other placed into the heel, thereby maximizing the area that would be stimulated by an electrical current. The electrode wires were connected to pass-throughs in one of the metal plates, to permit later application of electrical current. A similar pair of electrodes was placed into the contralateral paw to allow for a search for neurons on that side. The carotid artery catheter was attached to a transducer placed into the chamber, and the jugular catheter was attached to an infusion pump (Stoelting Company, Wood Dale, IL) that administered pancuronium (approximate 0.2–0.3 mg · kg–1 · h–1). The pump was computer-controlled via electrical pass-throughs. The jugular catheter had a Y valve that permitted attachment of another pump for infusion of saline. The rat rested on a warming device controlled from outside the pressure chamber.
After preparing the rat and putting it into the acrylic tube, the metal plates closing each end of the tube were secured by bolts to form an air-tight seal. The chamber was then flushed with a mixture of 40%–50% oxygen and 50%–60% N2O for 45–60 min. Samples were taken and analyzed (using a calibrated agent analyzer (Rascal II, Ohmeda, Salt Lake City, UT) to confirm elimination of the nitrogen. The gas flow to the ventilator in the chamber was then switched from the outside source to the ambient gas in the chamber, and the chamber pressurized with N2O to a starting pressure of approximately 2 atm, which resulted in about 0.5 atm oxygen and 1.5 atm N2O.
After waiting at least 30–45 min to permit washout of isoflurane, neurons were sought that responded to electrical stimulation. A brief (250–500 ms) tetanic stimulus (100 Hz, 30 mA; Model NS252, Fisher-Paykel, Auckland, New Zealand) was applied to the hindpaw electrodes during advancement of the tungsten electrode into the spinal cord. We sought neurons that responded to this stimulus. If a neuron could not be found, the electrode was withdrawn and moved using computer-controlled motors (Oriental Motor Company, Torrance, CA) that relocated the microdrive in the X-Y plane. Having found an activated neuron, we applied a tetanic stimulus (100 Hz, 30–80 mA, from 0.25 s to 5 s, depending on the response properties of the neuron, although in almost all neurons we used a 2 s stimulus duration). We also determined whether the neuron would "windup," i.e., increase in discharge frequency in response to repetitive electrical C-fiber strength stimulation. To test for windup, we usually determined the threshold that would elicit one or more action potentials at C-fiber latency (100–400 ms) and used 2–3 x threshold to stimulate the neuron (20 pulses delivered at 1 Hz). We repeated the tetanic and windup stimuli every 2–3 min to obtain 2–3 responses for the tetanic stimulus and 2–3 responses for the windup stimulus.
We then increased the N2O partial pressure by adding N2O to the chamber. Gas samples were obtained from the chamber, passed through the agent analyzer, and the N2O partial pressure determined by multiplying the N2O concentration by the total pressure in the chamber. We waited at least 15 min after achieving the new partial pressure before obtaining another set of neuronal responses. We determined responses at 1.5, 2, and 2.5 atm N2O, which represent a range of 0.8–1.3 MAC (5,11). In about half the experiments, we started at 2 or 2.5 atm. When adjusting the N2O partial pressure to a lower value, we released gas from the chamber until we obtained the total pressure needed. Oxygen was added to maintain oxygen partial pressure >0.3–0.4 atm. The new N2O partial pressure was confirmed with the agent analyzer.
We determined the effects of the NMDA blocker MK-801 on neuronal responses during N2O anesthesia in another group of nine rats prepared as described earlier except that we also attached to the jugular catheter a computer-controlled infusion pump that contained 300–500 µg/mL MK-801. The initial N2O partial pressure was 1.5 atm and neuronal responses to the tetanic and windup stimuli were obtained before and during infusion of MK-801. The MK-801 infusion rate was 11 ± 2 µg · kg–1 · min–1 Neuronal responses were obtained 60–90 min after initiating infusion (13,14). After the rats were killed, MK-801 concentrations in the spinal cords and brains were determined by using high-pressure liquid chromatography as previously described (15). The tissues were frozen until analysis.
To control for time and pressure effects, in another six rats we recorded neuronal responses to the 100-Hz stimulus at 1.5 atm N2O (0.4–0.5 atm O2) and then added nitrogen to increase the chamber pressure to 2.5 and 3 atm without altering the N2O or oxygen partial pressures. The responses were determined at the new total pressures after waiting at least 10 min.
The recording depth was estimated from the movement of the microdrive. We visualized the surface of the cord (via a microscope placed outside the chamber) as the zero point for the electrode entrance to the cord. In those cases where we could not observe the electrode entering the cord because of blood or fluid collecting on the cord, we used zero point as the depth where the sound of background neuronal activity was first heard. We did not make spinal lesions using direct current while the animal was still in the chamber because of safety reasons and the possible introduction of electrical noise during recording (12).
Neuronal responses to tetanic stimulation were evaluated by counting the number of action potentials occurring for 1 min after cessation of the stimulus. The artifact from stimulation concealed action potentials during stimulus application. We also recorded the spontaneous activity for the 30 s period before stimulation.
Neuronal responses to the windup stimuli were evaluated by counting the number of action potentials occurring in the C-fiber latency range (100–400 ms) and combined C-fiber and afterdischarge (100–1000 ms) range after each of the 20 stimuli. The responses to each stimulus were summed across all 20 stimuli (i.e., "area under the curve") (16).
Neuronal responses (windup, response to 100 Hz stimulus) were compared across anesthetic conditions (1.5, 2, and 2.5 atm N2O) using repeated measures analysis of variance followed by the Student-Newman-Keuls test. Greater N2O partial pressures appeared to have variable effects on neuronal responses. Accordingly, we evaluated the responses with the F-test to determine if the amount of variation increased at the greater N2O partial pressures. We correlated (linear regression) neuronal depth in the spinal cord with the effect of N2O on neuronal responses. Responses for the effects of MK-801 were compared using a paired t-test. Responses for the effects of nitrogen were compared using repeated measures analysis of variance followed by the Student-Newman-Keuls test. A P < 0.05 was considered significant.
|
RESULTS |
|---|
|
TopAbstractIntroductionMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
|
Summary data for the responses to the 100-Hz stimulus are shown in Figure 3A. The average responses at the three N2O partial pressures did not differ significantly. Figure 3B shows the individual responses to the 100-Hz stimulus; most neuronal responses were minimally affected by changing the N2O partial pressure, although some responses were increased and some were decreased. Figures 3C and D show summary windup responses; the C-fiber response (100–400 ms) was not different among the anesthetic conditions (Fig. 3C). The combined C-fiber and after-discharge responses (100–1000 ms) were not significantly different at the three N2O partial pressures (Fig. 3D). The after-discharge response was also not significantly different among the three N2O partial pressures (data not shown).
|
Although increasing the partial pressure of N2O did not produce a difference in the average responses to the 100-Hz stimulus (Fig. 3A), a comparison of the responses at 2 and 2.5 atm to that at 1.5 atm revealed increased variation (F-test P < 0.001), suggesting that the divergent neuronal responses did not result from chance alone. Comparison of the neuronal recording depth to the effect of increasing N2O partial pressure on the neuronal response revealed a significant correlation (r = 0.70, P < 0.05; Fig. 4). The deeper the neuronal depth, the more likely the neuronal response was depressed. Thus, neurons that were depressed by increasing N2O from 1.5 to 2.5 atm tended to be deeper than the more superficial neurons that were activated by the same N2O partial pressure change. There was not a similar correlation of recording depth and neuronal response in the neurons that exhibited windup.
|
In a separate group of animals (n = 9) anesthetized at 1.5 atm N2O, administration of MK-801 increased the neuronal response to the 100-Hz stimulus (P < 0.05; Fig. 5A) but MK-801 did not significantly affect neuronal windup (Fig. 5B). The recording depth of these neurons was 822 ± 515 µm. The effect of MK-801 on the neuronal response did not correlate significantly with recording depth. The total MK-801 dose was 504 ± 84 µg. The brain and spinal cord concentrations of MK-801 were 3.2 ± 1.0 and 1.6 ± 0.1 µg/g tissue, respectively.
|
Changing the total pressure in the chamber, while keeping the N2O partial pressure at 1.5 atm, did not affect neuronal responses to the 100-Hz stimulus, with responses at 1109 ± 282, 1109 ± 290, and 1121 ± 375 impulses/min at total pressures of 2, 2.5, and 3 atm, respectively (repeated measures ANOVA). The recording depth of these neurons was 1267 ± 560 µm.
|
DISCUSSION |
|---|
|
TopAbstractIntroductionMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
At the neuronal network level, N2O acts at a supraspinal site (the periaqueductal gray) to cause the release of endogenous opiates onto GABAergic (
-amino butyric acid) neurons that have synaptic input to pontine noradrenergic neurons (1). These latter neurons project to the spinal cord dorsal horn to release norepinephrine which directly inhibits spinal nociceptive neurons, while parallel descending projections excite inhibitory GABAergic interneurons which, in turn, release GABA to also inhibit spinal nociceptive neurons (1). The overall result is to depress nociceptive transmission in the spinal cord. Even so, we found that the overall spinal neuronal responses were unaffected by increasing partial pressures of N2O; however, some neurons were facilitated, while others were depressed. Our results are not necessarily inconsistent with the paradigm developed by others (1) and described earlier. Possibly, the neurons facilitated by N2O were GABAergic, and might thus be expected to have increased responses to noxious stimulation. Furthermore, the correlation of neuronal depth to effect of N2O could reflect the progressive depression of spinal neurons in a dorsal horn-to-ventral horn neural circuit, although the response variability is only partly explained by recording depth. The recording depths in the present study encompass laminae 2–3 to lamina 9 (17). Obviously, not all neurons are depressed according to their location, and this might simply reflect the likelihood that the neurons selected in the present study were divergent with respect to their role in nociceptive processing. Nonetheless, in general, partial pressures of N2O approximating 1 MAC minimally depress (or even excite) dorsal horn neurons but inhibit neurons in the ventral horn containing the "final common pathway" for movement: central pattern generators (which generate rhythmic movement), premotor interneurons and motoneurons.
Many, but not all, of the neurons we studied developed windup, a condition in which neurons have increased responses to repetitive stimulation. The mechanism of windup is not completely understood, but is probably due to several factors, including increased calcium in the presynaptic terminal which leads to increased neurotransmitter release, activation of NMDA and neurokinin receptors, causing persistent cumulative depolarization and increased calcium in the postsynaptic neuron (18). It is unclear why some nociceptive neurons do not develop windup, but we have observed this before (19). In the present study, N2O did not depress windup; in fact, there was a trend towards enhancement. The reasons for this are unclear. Because N2O blocks the NMDA receptor, the increase in N2O partial pressure might be akin to administration of MK-801, which also enhanced nociceptive responses (Fig. 5A).
Tolerance to N2O occurs within 30–60 min and is complete by 90 min (4). In the present study, the rats were exposed to N2O for >120 min before testing neuronal responses and the analgesic effect N2O was likely minimal. Thus, the immobilizing action of N2O does not likely involve the analgesia neuronal circuits previously described (1). Whether N2O produces immobility by a separate supraspinal action or by a spinal action is unknown.
Administration of MK-801 increased neuronal responses to noxious stimulation. Modulation of the NMDA receptor has divergent effects on spinal neuronal responses. Luccarini et al. (20) observed that superficial neurons were facilitated, while deep neurons were depressed, by NMDA. Thus, if N2O is acting to block the NMDA receptor, then neuronal location might be a factor with regard to neuronal responses to noxious stimulation, as was presently observed. Based on our MK-801 infusion rates, the total dose, and the tissue concentrations of MK-801, the administration of MK-801 would be expected to decrease N2O anesthetic requirements (13). For example, a MK-801 total dose of approximately 900 µg results in a spinal cord MK-801 concentration of about 2.2 µg/g tissue (15). Our lower spinal cord concentration of MK-801 (1.6 µg/g tissue) is consistent with a lower total MK-801 dose that decreases N2O MAC by 30%–40% (13). Thus, we expected the neuronal responses to be depressed by MK-801.
How can these disparate findings be reconciled? We speculate that administration of MK-801 might have blocked glutamatergic excitatory input onto inhibitory interneurons, and the subsequent diminished inhibitory drive would cause increased responses to noxious stimulation. NMDA has divergent effects on C-fiber evoked activity, and a similar theory has been advanced to explain the inhibitory effects of NMDA on nociceptive stimuli, i.e., modulation of inhibitory interneurons (20). Nonetheless, as mentioned above, given the MAC-sparing effect of MK-801, the effect of MK-801 on the neuronal responses suggests that MK-801 decreases N2O MAC by an action at a more ventral site in the cord, perhaps the motoneuron or pre-motor interneurons, similar to our prior observations with isoflurane (21). Thus, for reasons that are unclear, though MK-801 might enhance neuronal responses at superficial and intermediate laminae, it likely depresses neurons in the ventral horn, including the motoneuron.
Prior studies that examined N2O actions in spinal cord have reported variable results, with activation, depression or both, or with minimal effect on spinal neurons. Komatsu et al. (22) reported that N2O activated a supraspinal inhibitory system. In decerebrate and spinalized cats, Kitahata and coworkers (7,8) found that N2O facilitated some spinal neurons, while other neurons were depressed. In spinalized anesthetized cats, Miyazaki et al. (10) determined that N2O had no significant effect on responses of wide-dynamic range neurons to noxious stimulation. In goats anesthetized with isoflurane, we found that N2O has variable effects on nociceptive responses of spinal neurons (9). These various studies used subanesthetic partial pressures of N2O and different species, with some animals intact and others either decerebrate, spinalized or both. To our knowledge, the present study is the first to examine spinal neuronal responses to noxious stimulation in intact animals with N2O as the sole anesthetic delivered under hyperbaric conditions.
Obtaining stable and reliable neuronal responses in a hyperbaric chamber presented challenging obstacles. During the course of an electrophysiological experiment in an intact animal various problems can occur that normally would be easily solved if direct access were possible. However, once the animal was in the chamber and the chamber was pressurized, these problems could not be solved. For example, if the tracheal tube became blocked, the rat would die before anything could be done. In addition, we needed to use a search paradigm that differs from the normal way in which neurons are sought. We had to place the needle electrodes across a large area of the rat's hindpaw so that we could maximize our success for finding nociceptive neurons. Because we were unable to apply other stimuli, such as tactile stimuli, we could not reliably characterize the neurons, i.e., whether the neurons were wide-dynamic range or nociceptive-specific. Electrical stimulation activates nociceptors, as well as other primary afferents that are not involved in nociception; however, electrical stimulation is supramaximal with respect to anesthetic requirements for immobility (23), and thus can be considered an appropriate stimulus. Lastly, we used large rats, which facilitated some of the surgical and technical procedures. These rats were likely older than those normally used in pain research, and thus we cannot exclude the possibility that our results were affected by this factor.
In conclusion, we found that N2O, in the 0.8–1.3 MAC range had variable effects on spinal neuronal responses, with most responses unaffected, but with some being facilitated and others depressed. Neurons that were depressed tended to be located more ventrally in the spinal cord, suggesting that the anesthetic effects of N2O converge to cause depression of a final common pathway, such as the motoneuron or pre-motor interneuron.
|
ACKNOWLEDGMENTS |
|---|
|
Footnotes |
|---|
Supported by NIH grants GM-57970, GM-61283, and 1PO1GM-47818.
Reprints will not be available from the author.
|
REFERENCES |
|---|
|
TopAbstractIntroductionMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
]2B adrenoceptors. J Neurosci 2000;20:9242–51.This article has been cited by other articles:
|
|
 |
|
  E. I. Eger II, D. E. Raines, S. L. Shafer, H. C. Hemmings Jr, and J. M. Sonner Is a New Paradigm Needed to Explain How Inhaled Anesthetics Produce Immobility? Anesth. Analg., September 1, 2008; 107(3): 832 - 848. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. C. Dutton, J. M. Cuellar, E. I. Eger II, J. F. Antognini, and E. Carstens Temporal and Spatial Determinants of Sacral Dorsal Horn Neuronal Windup in Relation to Isoflurane-Induced Immobility Anesth. Analg., December 1, 2007; 105(6): 1665 - 1674. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Kim, A. Yao, R. Atherley, E. Carstens, S. L. Jinks, and J. F. Antognini Neurons in the Ventral Spinal Cord Are More Depressed by Isoflurane, Halothane, and Propofol Than Are Neurons in the Dorsal Spinal Cord Anesth. Analg., October 1, 2007; 105(4): 1020 - 1026. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
