JN Fuel your research with LabChart
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

J Neurophysiol 80: 1383-1390, 1998;
0022-3077/98 $5.00
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Ota, K.
Right arrow Articles by Collins, J. G.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Ota, K.
Right arrow Articles by Collins, J. G.

The Journal of Neurophysiology Vol. 80 No. 3 September 1998, pp. 1383-1390
Copyright ©1998 by the American Physiological Society

Cutaneous Responsiveness of Lumbar Spinal Dorsal Horn Neurons Is Reduced by General Anesthesia, An Effect Dependent in Part on GABAA Mechanisms

Kouichi Ota, Tadao Yanagidani, Kazuhiro Kishikawa, Yuji Yamamori, and J. G. Collins

Department of Anesthesiology, Yale University School of Medicine, New Haven, Connecticut 06520

    ABSTRACT
Abstract
Introduction
Methods
Results
Discussion
References

Ota, Kouichi, Tadao Yanagidani, Kazuhiro Kishikawa, Yuji Yamamori, and J. G. Collins. Cutaneous responsiveness of lumbar spinal dorsal horn neurons is reduced by general anesthesia, an effect dependent in part on GABAA mechanisms. J. Neurophysiol. 80: 1383-1390, 1998. Extracellular activity was recorded from single spinal dorsal horn neurons in both chronic cat and acute rat models. This was done to define the effects of anesthesia on the processing of sensory information elicited by nonnoxious tactile stimulation of peripheral receptive fields (RFs). In the chronic cat model, baseline data were obtained in physiologically intact, awake, drug-free animals before anesthetic administration (halothane 1.0-2.0%). This made it possible to compare and contrast activity of each cell in the drug-free and anesthetized state. Halothane effects were confirmed in the acute rat model (anesthetized, spinally transected, and in some cases decerebrate). In addition, the gamma -aminobutyic acid-A (GABAA)-receptor antagonist picrotoxin (2 mg/kg) was administered intravenously to verify that the observed halothane effect on spinal dorsal horn neurons was mediated by an interaction with GABAA-receptor systems. Halothane effects on three separate measures of response to nonnoxious tactile stimuli were observed in the chronic cat model. Halothane produced a significant, dose-dependent reduction in the low-threshold RF area of the neurons studied. Halothane also caused a significant reduction in neuronal response to RF brushing (dynamic stimulus) and to maintained contact with the RF (static stimulus). A dose dependency was not observed with these latter two effects. Neurons with a predominant rapidly adapting response seemed to be less susceptible to halothane suppression than slowly adapting cells. In the acute rat model an increase in halothane caused a reduction in neuronal response similar to that seen in the cat. The intravenous administration of 2 mg/kg of picrotoxin by itself caused no significant change in RF size or response to brushing. However, the same amount of picrotoxin did cause a 50% reversal of the halothane-induced reduction in RF size without causing a significant change in the halothane effect on response to RF brushing. In contrast to work recently reported in a chronic sheep model, halothane causes a significant reduction in spinal dorsal horn neuronal response to tactile stimulation of peripheral RFs. This effect is caused by, in part, but not exclusively, to GABAA-neurotransmitter systems. However, the relative influence of GABAA systems may vary with the nature of the stimulus.

    INTRODUCTION
Abstract
Introduction
Methods
Results
Discussion
References

The spinal dorsal horn, as one of the sites of initial synaptic communication between first- and second-order neurons, is an important focus of study in efforts to better understand the physiology and pharmacology of sensory processing. Most spinal cord studies, however, were conducted in acute preparations in which anesthesia and/or surgical manipulations have the potential to modulate the very systems under study. Although the profound impact of anesthetic agents on spinal processing of information about noxious stimuli has been appreciated for several decades (Conseilleu et al. 1972; de Jong et al. 1969, 1970; Kitahata et al. 1975; Wall 1967), there has been a tendency to assume that information about nonnoxious stimuli is less susceptible to such modulation. This is in spite of the fact that as early as 1968, de Jong and Wagman proposed that anesthetic action within the spinal cord may contribute to anesthetic-induced loss of sensation. Although Dubner and colleagues (Dubner and Hayes 1979; Hayes et al. 1979) in studies of the trigeminal equivalent of the spinal dorsal horn, the trigeminal nucleus caudalis, suggested that anesthetics are likely to alter ongoing and evoked central neuronal activity, we are unaware of any reports of anesthetic effects on spinal dorsal horn neuronal responses to nonnoxious stimuli being published until our studies of the effects of the intravenous anesthetics pentobarbital sodium, propofol, and ketamine (Collins et al. 1990; Kishikawa et al. 1995; Uchida et al. 1995) and the halothane study of Herrero and Headley 1995. Before that there was evidence from acute studies that anesthetics in general, and halothane in particular, were likely to influence spinal dorsal horn neuronal responses to nonnoxious stimuli. Most compelling was the work by de Jong and Wagman (1968) and de Jong et al. (1969) in which they reported that the cutaneous receptive fields (RFs) of spinal dorsal horn neurons decreased in the presence of anesthesia until it was no longer possible to elicit a response with RF stimulation.

As pointed out by Herrero and Headley in 1995, the most direct way of assessing the effects of an anesthetic are to establish baseline data for each neuron in the physiologically intact, awake, drug-free state and then maintain single unit recording of each neuron during and subsequent to anesthesia. In that study they reported, in contrast to the finding of de Jong and colleagues, that halothane anesthesia in sheep was associated with an increase, not a decrease, in the size of cutaneous RFs. Unfortunately, they were unable to maintain single unit recordings from the same neurons in both the awake and anesthetized state. Their assumptions were based on examination of populations of neurons in each state. On the basis of previous and current work in our laboratory, we hypothesized that halothane would depress, not excite, spinal dorsal horn neuronal response to low-intensity RF stimulation. The major purpose of this study was to test the hypothesis that halothane effects on neuronal response to low-intensity stimuli in cat and rat are different from those reported in the sheep.

The mechanisms of action by which general anesthetics produce the state referred to as general anesthesia are not yet defined. Evidence is accumulating to support the concept that each agent interacts with selective neurotransmitter systems in a way that causes modulation of sensory transmission. There is great interest in GABAergic mechanisms of anesthesia because many classes of general anesthetics were shown to enhance gamma -aminobutyic acid-A (GABAA)-mediated inhibition (Longoni et al. 1993; Moody et al. 1988; Tanelian et al. 1993; Wakamori et al. 1991; Yeh et al. 1991; Zimmerman et al. 1994). GABA-containing cells (Barber et al. 1982; Carlton and Hayes 1990; Todd and McKenzie 1989) as well as GABAA receptors (Bowery et al. 1987; Persohn et al. 1991; Waldvogel et al. 1990) were shown to exist in large numbers in the spinal dorsal horn. We hypothesized, on the basis of ongoing studies in our laboratory, that the observed effects of halothane on tactile RF stimulation would involve GABAA-receptor systems.

We report on our observations on the effects of halothane anesthesia on single spinal dorsal horn neurons in cats in which we were able to record from the same neurons before, during, and after the induction of anesthesia. As we observed with other anesthetic agents, the response of spinal dorsal horn neurons to nonnoxious tactile stimulation is depressed by halothane anesthesia. These effects, when studied in an acute rat model, were partially reversed by an antagonist of GABAA receptors. Preliminary results were published in abstract form (Ota et al. 1993, 1994).

    METHODS
Abstract
Introduction
Methods
Results
Discussion
References

This research protocol was approved by the Yale Animal Care and Use Committee.

Chronic cat preparation

Experiments were performed on six female cats ranging in weight from 2.5 to 4.5 kg. Each animal was prepared for chronic extracellular recording from single spinal dorsal horn neurons in the physiologically intact, awake, drug-free state.

Details of the chronic recording technique were reported previously (Collins 1985). Animals were adapted to sit quietly in a restraint box while RFs were stimulated. Under general anesthesia and with the use of sterile technique, a recording chamber was surgically attached to the L4 vertebral column over a 6 × 12 mm opening in the bone, leaving the dura intact. An external jugular vein catheter was implanted and externalized on the head. After a minimum 2-wk recovery after chamber implantation, electrophysiological studies were begun.

For each experiment, a tungsten microelectrode (impedance 10 MOmega ) was inserted through the dura into the spinal cord. Dural penetration produced no obvious animal discomfort. Amplitude discrimination was used to select cells of interest.

Stimulus presentation

The RF of each neuron responsive to light touch was mapped on the surface of the skin, and the most sensitive area was then stimulated by von Frey filaments, brushing, rubbing, pinching, and heating to evaluate the neuron's response properties. The border of the RF was defined as that area from which a response to light touch could be elicited 50% of the time. When possible, the threshold to mechanical stimulation was determined by using von Frey filaments at three sites within the RF. Animals frequently would not tolerate von Frey stimulation. Brush responses were produced when RFs were stimulated with a 3/4-in. camel hair brush that was moved by hand across the length of the RF. This allowed us to examine the cell's response to a dynamic, nonnoxious mechanical stimulus. Pinch stimuli were produced with forceps that were modified so that a constant contact area (3-mm diam) was stimulated each time. The forceps were also instrumented with strain gauges to monitor stimulus intensity. Pinch was increased in intensity until a reflex withdrawal was initiated by the animal. Withdrawal usually occurred to stimuli in the range considered mildly noxious by the experimenter. Pinch stimuli were separated by a 2-min interstimulus interval.

The pinch stimulus was presented so that we could examine both the response to a noxious mechanical stimulus and the response to a static, nonnoxious mechanical stimulus. We initially gripped a fold of skin with the forceps, but paused for several seconds before beginning to apply increasing pressure. That initial period of contact allowed us to determine if the cell had a slowly adapting (SA) or rapidly adapting (RA) response as well as to examine drug effects on the response to a static mechanical stimulus (in contrast to the response to the dynamic brush stimulus).

Heating was achieved by focusing a radiant heat source (~1 cm diam) on the RF. The skin temperature was increased until a reflex response was elicited (typically 47-49°C). When the RF was large enough to be monitored by a thermocouple on the skin surface, heat stimuli of 45, 47, and 49°C were delivered for a maximum of 8 s. Thermal stimuli were separated by a 2-min interstimulus interval and terminated on reflex withdrawal by the animal.

Halothane administration

All baseline responses for each neuron studied were determined while the animals were drug-free. After determination of baseline responses, glycopyrrolate (0.05 mg) was administered intravenously 5 min before administration of the intravenous general anesthetic propofol. Propofol (10 mg/kg) was administered to induce anesthesia rapidly. Induction of anesthesia was followed immediately by tracheal intubation and ventilation with 1.3% halothane in 100% oxygen. RF area as well as neuronal responses to stimuli used during baseline determinations were again evaluated at 5, 30, 45, and 60 min. From previous work in our laboratory (Uchida et al. 1995), we know that the effects of propofol at the dose used would clear within 30 min. Consequently, statistical analysis of RF size and neuronal responses was based on a comparison between the baseline and the 60-min time points.

End-tidal concentrations of CO2, oxygen, and halothane were monitored during general anesthesia (Datascope MULTINEX). End-tidal CO2 was maintained between 28 and 40 mmHg. An infrared heating lamp was used to maintain body temperature within physiological limits.

In an additional series of experiments baseline data obtained in the presence of 1.5% halothane were used to examine the effect of increasing or decreasing the halothane concentration. These cells were encountered after the animal was anesthetized and therefore do not have drug-free baseline data associated with them. After determination of baseline data, we increased the halothane concentration to 2%, allowed for a minimum of 20 min for equilibration, and repeated the test stimuli. We then reduced the halothane concentration to 1%, allowed for equilibration, and repeated the testing.

Data analysis

The outline of the RF that was mapped on the skin was transferred to tracing paper, digitized, and used to determine and analyze RF areas. Spontaneous firing rates were determined by averaging the activity over 10- to 20-s periods when there was no contact with the RF. Brush responses were averaged over 10 stimulus presentations. A neuron was classified as a low-threshold (LT) neuron based on the fact that the maximum firing frequency was elicited by light touch with no increase in activity when pinch or heat caused animal withdrawal. A neuron was classified as a wide dynamic range (WDR) neuron based on its response profile to increasing intensities of RF stimulation. If the firing frequency increased as the stimulus intensity increased, with maximum activation occurring only with presentation of the most intense stimuli, then the cell was categorized as a WDR neuron.

Statistical analysis of the data was carried out with a paired Student's t-test and Wilcoxon rank sum test or repeated analysis of variance (ANOVA) followed by the Scheffe F test. Differences were considered significant when P < 0.05. All data are presented as means ± SD.

Acute rat preparation

Because of epileptogenic effects, it was not practical to evaluate a GABAA-receptor antagonist (picrotoxin) in the intact cat model. We therefore conducted complementary experiments with the use of an acute rat model to confirm effects seen in the cat and to examine the role that GABAA systems may play in halothane's ability to alter spinal dorsal horn neuronal response to nonnoxious RF stimulation. We previously demonstrated in an acute rat model that halothane produces effects similar to those observed in the chronic cat model (Yamamori et al. 1995).

Sixteen male Sprague-Dawley rats (320-420 g) were initially deeply anesthetized with halothane (3-4%) in oxygen. Venous, arterial, and tracheal cannulae were inserted followed by artificial ventilation (2.5- to 3.5-ml stroke volume, 40-50 strokes/min) and intravenous fluid administration (10-15 ml·kg-1·h-1). End-tidal CO2 was monitored and maintained at 30-40 mmHg. All rats were paralyzed with pancuronium bromide (0.3 mg iv). Supplemental doses of pancuronium bromide (0.2 mg iv) were administered to all rats to maintain paralysis. Rectal temperature was thermostatically controlled (37.5 ± 0.5C°) with the use of an infrared heat lamp.

In all animals, after high thoracic spinal cord transection, the lumbar enlargement was exposed, the halothane concentration was adjusted to 1%, and a tungsten microelectrode (impedance 4 MOmega , FHC, Brunswick, ME) was inserted through the dura into the spinal cord. Stimulus presentation and data collection were similar to those described above for the chronic cat preparation.

After determination of neuronal responses (spontaneous activity, RF mapping, brushing, heating, von Frey stimulation) under 1% halothane anesthesia, we increased the concentration of halothane from 1.0 to 2.0%, allowed for 20 min of equilibration, and then reevaluated neuronal response. Two milligrams per kilogram of Picrotoxin was then administered intravenously, followed 20 min later by reevaluation of neuronal response. This dose of picrotoxin was chosen because it is a typical maximum seizure-inducing dose in rats reported in the literature. In addition, in other studies in our laboratory, that dose was found to be in the range of doses capable of maximally reversing an anesthetic's effects on LT RF size.

In six additional male Sprague-Dawley rats (380-520 g), we evaluated the impact of picrotoxin alone on neuron response. The animals were prepared as above, but in addition they were rendered decerebrate so that we could examine picrotoxin effects in the absence of any anesthetic agent. The RF area as well as neuronal response to brushing and von Frey stimulation were evaluated before and 20 min after administration of 2 mg/kg of picrotoxin.

    RESULTS
Abstract
Introduction
Methods
Results
Discussion
References

Chronic cat preparation

Data for this study were obtained from six animals. All animals were drug-free for a minimum of 48 h before each experiment. Twenty-seven neurons were analyzed for presentation in this report. All neurons had RFs located on the hindlimbs or hips.

Of the 27 neurons in this study, 26 were classified as LT and 1 as WDR. This low representation of WDR neurons in the spinal dorsal horn of intact, awake, drug-free animals was a consistent finding in our laboratory (Collins et al. 1990).

As in previous chronic experiments, spontaneous activity was low. The mean rate of spontaneous activity in the intact, awake, drug-free state was 0.42 ± 0.9 imp/s. Halothane had no significant effect on spontaneous activity in this study.

Effects of 1.3% halothane

CHANGE IN LT RF SIZE. Sixteen LT neurons are included in this part of the study. The most pronounced effect of halothane on neuronal response to low-intensity stimulation was a significant reduction in the RF area responsive to the low-intensity stimuli. That decreased response was observed in all 15 neurons in which RFs could be accurately mapped. The effects of anesthesia on LT RF size for two neurons is shown in Fig. 1. As seen in Fig. 2, the mean RF size of the 15 neurons was decreased significantly to 58 ± 21% of control (from 11.5 ± 8.8 to 6.8 ± 6.6 cm2) 60 min after the start of administration of 1.3% halothane (P < 0.01, paired Student's t-test).


View larger version (15K):
[in this window]
[in a new window]
  		FIG. 1. Examples of halothane reduction in low threshold (LT) receptive field (RF) size. A: 1.3% halothane produced a 60% reduction in the LT RF of this neuron. B: same partial pressure of halothane caused this neuron's RF to be decreased by 79% from control.


View larger version (16K):
[in this window]
[in a new window]
  		FIG. 2. Mean percent change in RF area of LT neurons during halothane anesthesia. All values are means ± SD. At time 0 propofol (10 mg/kg) was administered to induce anesthesia; 1.3% halothane administration (end-tidal concentration) began within 5 min after propofol administration. Previous data indicated that effects of propofol terminated within 30 min. The change in RF size at 5 min (34 ± 19% of baseline) is due almost totally to propofol. The change at 60 min (57 ± 20% of baseline) is assumed to be due only to halothane and is statistically significant.

Animals in the awake state tended to be quite sensitive to stimulation by von Frey hairs. They would often withdraw more briskly from nonnoxious von Frey stimuli than from noxious heat or pinch. This severely limited our ability to detect thresholds across the RFs. We were successful in that attempt for only six neurons, but the results are instructive. For all six neurons halothane increased the threshold in the area of the RF that seemed to disappear. The neuron could still be activated from that area, but a higher intensity of stimulation was required (i.e., the properties of the RF was changed). For those same six neurons the threshold in the area of the RF still apparent in the presence of halothane anesthesia was unchanged in three cases, increased in two cases, and decreased in one case.

Halothane effects on LT response to nonnoxious stimulation

We were able to evaluate neuronal response to nonnoxious stimuli in two different ways: response to brushing (dynamic nonnoxious stimulation) and response to initial forceps contact (static nonnoxious stimulation). Halothane reduced the response to brushing in 13 of 15 neurons tested with a mean decrease of evoked discharge to 48 ± 36% of control (P < 0.05, paired Student's t-test), and 1.3% halothane reduced the mean evoked response to initial contact with the forceps (static nonnoxious stimulus) to 59% of control. However, the effect was not the same for rapidly adapting (RA) neurons as it was for slowly adapting (SA) neurons.

Based on the level of activity during the 3-s period of initial contact, 4 of the 16 neurons analyzed were judged to have an RA response. As seen in Table 1, halothane actually caused an increase in the mean response to initial contact (141% of control) for the four RA neurons. A comparison of the halothane effect on the mean response of the SA neurons (reduction to 26.3% of control) indicated a significant difference between the halothane effects on those responses. A similar difference was observed in the response to brush stimulation. The mean response of the RA neurons to RF brushing was 95% of control in the presence of 1.3% halothane, compared with 31% of control for the SA neurons.

 
View this table:
[in this window] [in a new window]
  		TABLE 1. Halothane effect on response to brush and initial contact

Baseline RF areas of the RA neurons (2.8 ± 0.6 cm2) were significantly smaller than those of the SA neurons (15.4 ± 9.5 cm2).

LT response to noxious stimuli

By definition, LT neurons do not respond to noxious thermal or mechanical stimuli. However, in this study we applied such stimuli to define LT or WDR status and to determine if, in the presence of halothane anesthesia, responses to such noxious stimuli would be unmasked as they were by pentobarbital (Collins and Ren 1987). In the 10 neurons in which noxious thermal stimuli and the 16 in which noxious mechanical stimuli were presented, no unmasking of a response to those stimuli was observed in the presence of 1.3% halothane.

Halothane effects on WDR neurons

Only one neuron had a WDR response profile during baseline studies. This WDR neuron had its response to noxious radiant heat-evoked activity significantly reduced by 1.3% halothane. Halothane also reduced the LT RF size and the activity evoked by brush and pinch stimulation by 30-40% of the control value.

Dose dependency

LT NEURONS. Table 2 summarizes the effects of increasing doses of halothane. For technical reasons, baseline data in the absence of halothane are not available. Although there was a trend toward reduced response in all three measurements, the only measurement in these nine LT neurons that demonstrated a dose-dependent effect was change in RF size. Halothane-induced decreases in the mean response to brushing or contact were not significantly different across the dose range studied.

 
View this table:
[in this window] [in a new window]
  		TABLE 2. RF size and neural responses of LT neurons under halothane anesthesia (1.0-2.0%)

WDR NEURONS. With only two neurons in the sample, statistical analysis was not possible, but there was a trend toward decreased RF size and response to nonnoxious stimuli as halothane concentrations increased.

Acute rat preparation

Twenty-two neurons were studied in the acute rat model, including six in the decerebrate preparation. Spontaneous activity was higher in the decerebrate animals (2.4 ± 1.9 spikes/s) than in the intact rats (0.3 ± 0.3 spikes/s). Neither halothane nor picrotoxin caused a significant change in spontaneous activity.

As was seen in the chronic cat preparation, the increased partial pressure of inhaled halothane and the resultant increased depth of anesthesia resulted in a reduction in the size of the LT RFs. This is seen for an individual neuron in Fig. 3 along with the partial reversal of the halothane effect by the intravenous administration of 2 mg/kg of picrotoxin. As shown in Fig. 4, increasing the halothane from 1 to 2% caused a significant (48%) reduction (from 358 ± 216 to 174 ± 167 mm2) in the mean RF size of the neurons studied. Subsequent picrotoxin administration only partially reversed the halothane effect.


View larger version (11K):
[in this window]
[in a new window]
  		FIG. 3. A typical change of RF size in the rat. The RF area of the cell in the presence of 1% halothane was 464 mm2. Increasing the concentration of halothane from 1.0 to 2.0% resulted in the RF being reduced to 190 mm2 (41% of the size in the presence of 1% halothane). Picrotoxin (2.0 mg/kg) in the presence of 2.0% halothane significantly increased the RF size to 360 mm2 (77% of the size in the presence of only 1% halothane). However, the RF size after the administration of picrotoxin was still smaller compared with the RF under 1.0% halothane anesthesia.


View larger version (20K):
[in this window]
[in a new window]
  		FIG. 4. Mean RF size (n = 16). All values are means ± SD. The mean RF size in the presence of 2% halothane anesthesia with or without picrotoxin was significantly smaller than the mean RF size under 1% halothane anesthesia. # P < 0.01, repeated analysis of variance (ANOVA) with Scheffe F test. Picrotoxin induced a significant increase of RF size. * P < 0.01, repeated ANOVA with Scheffe F test.

Figure 5 summarizes the effects of halothane on the response to RF brushing for the same cells represented in Fig. 4. An increase in halothane from 1 to 2% significantly reduced the mean response to brush stroke. In spite of the fact that, as seen in Fig. 4, picrotoxin partially reversed the effect of halothane on RF size, it did not cause a significant increase in neuronal response to RF brushing in the presence of 2% halothane. This lack of an effect of picrotoxin was also evident in a lack of effect on the von Frey thresholds within the center of the RFs. Halothane increased the threshold in 10 of 13 neurons, but picrotoxin reversed that effect in only 1 neuron.


View larger version (18K):
[in this window]
[in a new window]
  		FIG. 5. Mean neuronal response to RF brushing in the acute rat (n = 13). All values are means ± SD. The mean neural response to brush in presence of 2% halothane anesthesia with or without picrotoxin was significantly smaller than that under 1% halothane anesthesia. # P < 0.01, repeated ANOVA with Scheffe F test. Unlike effects on RF sizes, picrotoxin did not induce a significant change of neural response to brush.

In six decerebrate animals we studied the effects of 2 mg/kg of picrotoxin in the absence of any anesthetic agent. As seen in Table 3, picrotoxin caused no significant change in either the mean RF size or the mean response to brush stoke. Only one of the six neurons had a >20% increase (28%) in RF area after picrotoxin administration. One of five neurons in which response to brush was studied had a >20% change (53% increase). There was no change in von Frey thresholds in the most sensitive region of the RF after picrotoxin administration.

 
View this table:
[in this window] [in a new window]
  		TABLE 3. Mean RF size and mean neural response to brush in the presence and absence of 2 mg/kg picrotoxin in decerebrate rats

    DISCUSSION
Abstract
Introduction
Methods
Results
Discussion
References

The primary purpose of this study was to define the effects of the inhalation anesthetic halothane on the response of spinal dorsal horn neurons to nonnoxious tactile stimulation of their peripheral RFs. The use of a physiologically intact, awake, drug-free preparation allowed us to compare baseline responses in that state with responses of the same neurons to similar stimuli in the presence of halothane. These results demonstrate a clear effect of halothane anesthesia. Halothane reduced both the RF area responsive to LT stimuli as well as the response to dynamic (brush) and static (initial contact) RF stimulation. Our efforts at defining von Frey thresholds met with limited success but suggest that the halothane reduction in RF area was not caused by a complete block of response from that skin region. Rather, it simply reflected an increased threshold to mechanical stimulation. This was confirmed in current ongoing studies in our laboratory (unpublished observations).

The effects observed in this study are in general agreement with the 1968 report by de Jong and Wagman that halothane could cause a reduction in RF size. However, de Jong and Wagman reported that 2% halothane eliminated neuronal response to stroking or pinching of the skin and that 1% halothane left a reduced RF that only responded to strong stimulation. The more profound effect in the earlier study is likely caused by the use of two very different animal models. The current model has the advantage of providing a measure of true baseline activity in an animal that is physiologically intact and did not recently undergo significant surgery. In spite of differences, both studies provide clear evidence that in the cat nonnoxiously evoked activity within the spinal dorsal horn is susceptible to depression by halothane.

The ability of general anesthetic agents to decrease the response of spinal dorsal horn neurons to nonnoxious tactile stimulation of peripheral RFs is not limited to halothane. As we previously reported, the intravenous anesthetics pentobarbital (Collins et al. 1990) and propofol (Uchida et al. 1995) cause a significant reduction in those responses. We also demonstrated that the proportional to 2-adrenergic agonist dexmedetomidine (Kishikawa et al. 1991), which is capable of causing loss of consciousness in the chronic cat model (drugs in this class are used as veterinary anesthetics), and the inhalation anesthetics isoflurane (Kaneko et al. 1994) and enflurane (Yanagidani, Ota, and Collins, unpublished manuscript) cause similar effects.

Not only were we able to demonstrate that most general anesthetics produce this effect in the chronic cat model, we also showed in the acute rat model that the halothane effect is seen within the spinal cord in spinally transected animals (Yamamori 1995). In previous work by one of the authors using a chronic monkey model, it was shown that general anesthetics cause reduction of tactile RFs associated with single neurons in the somatosensory cortex (Collins and Roppolo 1980). Taken together these studies provide evidence that, in at least three species, the presence of general anesthesia is likely to reduce the sensitivity of somatosensory neurons to nonnoxious tactile stimulation of their peripheral RFs. That conclusion is directly counter to the results reported in a chronic sheep model (Herrero and Headley 1995).

Contrary results in cat and sheep models could be attributed to differences in preparation, pharmacological differences, and species differences. There are two specific differences between the chronic sheep and cat preparations that could explain the conflicting results. The first, as recognized by Herrero and Headley, is that they did not record from the same neurons in both the conscious and anesthetized states. The absence of such recordings in the sheep make it impossible to know if similar populations were studied in each state. However, it is unlikely that Herrero and Headley studied different neuronal populations in awake and anesthetized animals.

A second difference in preparation relates to the training and state of the animals. The sheep in the Herrero and Headley study were maintained in an environment that the animals would commonly encounter. The sheep therefore required no training and could be assumed to be at ease during the experiments (J. F. Herrero and P. M. Headley, personal communications). In our studies the animals were highly trained to sit quietly in a very artificial environment. One could argue that such an environment was a stressor that caused changes in tonic modulation of spinal dorsal horn neurons and that the pharmacology we observed was mainly the impact of halothane on that abnormal modulation. Our study of the effects of sleep and propofol on spinal dorsal horn neurons suggests that that was not the case (Kishikawa et al. 1995). If anesthetic-induced removal of stress was responsible for the observed change in the cat model, we would expect that natural sleep, by removing the stress, would reduce, if not eliminate, the anesthetic effects on RF size. However, that was not the case. When propofol was administered to sleeping animals, LT RFs were significantly reduced in size. On the basis of those findings the likely higher levels of stress in our preparation did not cause the apparent difference in halothane effects between sheep and other species studied.

The potential for pharmacological differences are many. The most obvious one is a difference in dosing. Herrero and Headley (1995) describe the depth of anesthesia as one in which they maintained "... a constant level of weak palpebral reflexes, very weak pedal withdrawal reflexes, mildly dilated pupils and constant heart rate. ..." Our animals were more deeply anesthetized. In our pentobarbital study (Collins et al. 1990) we demonstrated that, as the level of anesthesia lightened, the RFs became significantly larger than the awake baseline value. However, we did not see that in the propofol study (Uchida et al. 1995) in which we were able to monitor RF size through the process of awakening from anesthesia. In addition, in the acute rat model (decerebrate, spinal cord transected) halothane as low as 0.5% did not cause RFs to enlarge. Therefore differences in depth of anesthesia do not necessarily explain differences in halothane effects on cats and sheep.

We must seriously consider possible species differences as an explanation for the apparent opposite effects of halothane. If we compare the general description of neuronal activity in the two chronic recording models we see that, as was reported also by Sorkin et al. in 1988, the number of WDR neurons observed in the chronic cat model was quite low. We typically classify only 10% of the neurons as WDR. However, in contrast to our work and that of Sorkin (1988), Herrero and Headley (1995) report that 76% of the neurons they studied in the dorsal horn were of the WDR type. That difference and the contrasting effects of halothane suggest that species differences may underlie the observed halothane-induced increase in RF area in the sheep, whereas in cat and rat the area becomes smaller.

An additional point of interest in the results of this study is the apparent difference in halothane effects on RA and SA neurons. We must be cautious in interpreting these results because the sample size is small and the total neuronal response of SAs is limited. However, the results do suggest the possibility of a selective effect on inputs from two different types of primary afferents. The likelihood of differential sensitivity for different types of LT input is even more evident in recent studies conducted in our laboratory. In one of those studies (Yanagidani, Ota, and Collins, unpublished manuscript) 2.1% enflurane reduced LT RF size in chronic cats, but the response to brushing the remaining RF area was significantly increased. Differential effects are also apparent in Table 2, where we see that the response of some neurons to static and dynamic stimuli are influenced in opposite directions. The possibility of differential effects is further emphasized by the picrotoxin portion of this study. It appears that a picrotoxin-reversible presumably GABAA mechanism is involved in some but not all of the halothane effect on neuronal response to LT RF stimulation. In recent unpublished work, we observed that serotonin and glycine but not nitric oxide, adenosine, or nicotinic cholinergic systems may also participate in the effect of inhalation anesthetics on neuronal responses to LT stimuli. We previously reported that GABAB did not seem to be involved (Yamamori et al. 1995).

There is increasing evidence that general anesthetics influence multiple transmitter systems in the spinal cord and, importantly, that the systems may vary depending on the anesthetic. As Kendig summarized in a recent review (Collins et al. 1995), not all general anesthetics influence the same transmitters within the spinal cord. On the basis of the results of this study, we hypothesize that GABAA is one of only several transmitter systems that helps to mediate the observed effects of halothane on spinal sensory processing.

de Jong and Wagman (1968) proposed that spinal actions of anesthetics may contribute to loss of sensation. The current study demonstrates a profound and complex effect of halothane anesthesia on nonnoxiously evoked activity within the spinal dorsal horn. It is likely that such actions are generalized across many anesthetics. Future studies of processing of information by the somatosensory system must take into account the possible effect that the anesthetic state of the animal may have on results obtained.

    ACKNOWLEDGEMENTS

  The technical assistance and animal care provided by A. Hinds and S. Canon are gratefully acknowledged.

  Work in J. G. Collins' laboratory was supported in part by National Institutes of Health Grants GM-44954 and NS-10174.

    FOOTNOTES

  Address for reprint requests: J. G. Collins, Dept. of Anesthesiology, Yale University School of Medicine, 333 Cedar St., PO Box 208051, New Haven, CT 06520-8051.

  Received 15 July 1997; accepted in final form 1 June 1998.

    REFERENCES
Abstract
Introduction
Methods
Results
Discussion
References
0022-3077/98 $5.00 Copyright ©1998 The American Physiological Society



This article has been cited by other articles:


Home page
 Br J AnaesthHome page
J. J. Kendig
In vitro networks: subcortical mechanisms of anaesthetic action
Br. J. Anaesth., July 1, 2002; 89(1): 91 - 101.
[Abstract] [Full Text] [PDF]

Home page
 Br J AnaesthHome page
C. Vahle-Hinz and O. Detsch
What can in vivo electrophysiology in animal models tell us about mechanisms of anaesthesia?
Br. J. Anaesth., July 1, 2002; 89(1): 123 - 142.
[Abstract] [Full Text] [PDF]

Home page
 Anesth. Analg.Home page
C. Vahle-Hinz, O. Detsch, M. Siemers, E. Kochs, and B. Bromm
Local GABAA Receptor Blockade Reverses Isoflurane's Suppressive Effects on Thalamic Neurons In Vivo
Anesth. Analg., June 1, 2001; 92(6): 1578 - 1584.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Ota, K.
Right arrow Articles by Collins, J. G.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Ota, K.
Right arrow Articles by Collins, J. G.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Visit Other APS Journals Online