1. Introduction
Peripheral nerve injury may trigger neuropathic pain, a type of chronic pain that is usually manifested as ongoing burning pain in the affected extremity and is frequently accompanied by pain evoked by normally innocuous mechanical and thermal stimuli (allodynia) (Bonica, 1990). A number of animal models have been developed in the last decade to study the mechanisms underlying neuropathic pain (Bennett and Xie, 1988; Seltzer et al., 1990; Kim and Chung, 1992; DeLeo et al., 1994). The animals in these models display abnormal behaviors similar to those seen in human patients, behaviors that indicate ongoing and evoked pain. Therefore, these animal models have been very useful in investigating the pathophysiological mechanisms underlying neuropathic pain.
Abnormal ectopic discharges are known to be generated in injured sensory nerve axons and their cell bodies in dorsal root ganglia (DRG) (Wall and Gutnick, 1974; Devor and Jänig, 1981; Scadding, 1981; Korenman and Devor, 1981; Blumberg and Jänig, 1984; Häbler et al., 1987; Welk et al., 1990; Devor et al., 1994). These ectopic discharges enter the spinal cord and alter central sensory processing by sensitizing the dorsal horn neurons. This process, known as central sensitization, is proposed to be the key step for many sensory abnormalities, including neuropathic pain. Since ectopic discharges seem to play an important role in maintaining central sensitization and neuropathic pain, it is important to analyze them systematically in different pain states.
Some patients get relief from their pain by sympathectomy or sympathetic block. This type of neuropathic pain is categorized as sympathetically maintained pain (SMP) to differentiate it from pain that is not relieved by sympathetic manipulations (Roberts, 1986; Campbell et al., 1992). Pain behaviors in animal models of neuropathic pain are reduced by sympathectomy as well (Shir and Seltzer, 1991; Kim et al., 1993; Desmeules et al., 1995). In particular, pain behavior in the segmental spinal nerve injury (SSI) model of neuropathic pain shows a strong sympathetic dependency (Kim and Chung, 1991; Kim et al., 1993, 1997; Choi et al., 1994; Kinnman and Levine, 1995), suggesting that the SSI is a model for SMP. Therefore, sympathectomy should interfere with the generation of ectopic discharge in the SSI model, since ectopic discharges are thought to be a peripheral driving force for the pain behavior.
Using the SSI model of neuropathic pain, the present study was conducted to: (1) quantify the ectopic discharges from injured afferents in the neuropathic rat; (2) correlate the changes in ectopic discharges with those in pain behavior; and (3) examine the changes caused by sympathectomy on the levels of ectopic discharges and pain behavior.
Preliminary data in the present study were presented previously in abstract form (Han et al., 1995).
2. Methods
2.1. Experimental animals and surgical procedures
Adult male Sprague–Dawley rats (obtained from Harlan Sprague–Dawley Inc., Houston) weighing 150–200 g at the time of surgery were used in this study. Experiments were done on four groups of rats:
- Group 1 (1-week group; n=13) – Recordings of ectopic discharges were made at 7–10 days after neuropathic surgery;
- Group 2 (10-week group; n=9) – Recordings of ectopic discharges were made at 10–12 weeks after neuropathic surgery;
- Group 3 (26-week group; n=12) – Recordings of ectopic discharges were made at 26–30 weeks after neuropathic surgery;
- Group 4 (sympathectomy group; n=10) – One week after neuropathic surgery, surgical sympathectomy was performed to remove lumbar sympathetic ganglia bilaterally, along with the sympathetic chains. Recordings of ectopic discharges were made at 1 week after sympathectomy.
Neuropathic surgery was done as previously described in detail (Kim and Chung, 1992; Choi et al., 1994). Briefly, the left L5 and L6 spinal nerves (ventral rami) were tightly ligated at a site distal to the DRG with 6–0 silk thread under general anesthesia with a mixture of halothane (2% for the induction and 0.8% for the maintenance of anesthesia) and a 2:1 flow ratio of N2O and O2. A complete hemostasis was confirmed, and the wound was sutured closed.
Surgical sympathectomy was done as previously described in detail (Kim et al., 1993; Choi et al., 1994). Briefly, under gaseous anesthesia with a mixture of halothane, N2O and O2, the sympathetic chain was identified through a transperitoneal approach. All ganglia and the chain below L2 were resected bilaterally.
2.2. Behavioral tests
Methods for the behavioral tests and the justification of interpretations of these tests were provided in previous publications (Kim and Chung, 1991; Kim et al., 1993, 1997; Choi et al., 1994). To avoid redundancy, only a brief description of the behavioral tests is given here. Behavioral tests were done before neuropathic surgery and 3, 5, and 7 days after the surgery. For groups 2 and 3, additional tests were done at least once every 2–4 weeks (including the day of the physiological experiment). The investigator who performed behavioral tests was not aware of the exact procedures carried out on the animals.
To assess mechanical sensitivity of the hindpaw, the frequency of brisk foot withdrawals in response to normally innocuous mechanical stimuli was measured as described previously (Kim and Chung, 1992; Na et al., 1993; Sheen and Chung, 1993). Innocuous mechanical stimuli were applied with von Frey filaments of two different bending forces (8.4 and 54.4 mN). To quantify sensitivity of the foot to cold, the frequency of brisk foot withdrawals in response to the application of acetone was measured as described previously (Choi et al., 1994).
Two different tests for ongoing pain were employed: cold-stress exacerbated ongoing pain and spontaneous pain (Choi et al., 1994). For the measurement of cold-stress exacerbated ongoing pain, the cumulative duration of the time that the rat held its foot off a brass cold plate (5±1°C) over a period of 5 min was recorded. Whether this behavioral test reflects ongoing pain or evoked pain is controversial since it seems logical that lifting the foot from a cold plate may be due to an increased sensitivity to cold. However, as shown in our previous study (Choi et al., 1994), the rat continued to lift its foot off the cold plate after a complete denervation of the foot, suggesting that foot lifting is not an evoked pain behavior. Hence, our interpretation is that these liftings are exaggerated paw withdrawal reflexes and are a form of ongoing pain that is exacerbated by cold stress.
For the measurement of spontaneous pain (ongoing pain without apparent external stimuli), the above procedure was varied by setting the brass plate at a neutral temperature (30±1°C). Such testing has been used as an index of spontaneous pain after the subcutaneous injection of formalin (the formalin test) in both the cat and rat (Dubuisson and Dennis, 1977) and for neuropathic spontaneous pain in the rat (Bennett and Xie, 1988; Attal et al., 1990; Mao et al., 1992; Choi et al., 1994).
2.3. Action potential recordings
After the final behavioral tests, all of the rats were subjected to a single unit action potential recording experiment. The rats were anesthetized with a mixture of halothane (2% for the induction and 0.8% for the maintenance of anesthesia) and a 2:1 flow ratio of N2O and O2. Tracheostomy was performed, and the right jugular vein and carotid artery were cannulated for drug administration and blood pressure monitoring, respectively. Animals were paralyzed by intravenous injection of pancuronium bromide (Pavulon: a single bolus of 1 mg/kg, i.v., followed by a continuous intravenous infusion, 0.4 mg/kg per h). Animals were artificially ventilated, and the end-tidal CO2 was maintained at a level between 4.5 and 5.5% while monitoring with a capnometer. Data were collected only when animals had a mean arterial blood pressure of at least 80 mmHg.
The spinal cord was exposed by a laminectomy at the level of the L1 to L6 vertebrae. The animal was mounted on a spinal investigation frame. A heated mineral oil pool was made over the exposed tissue to prevent it from drying. The dura mater was opened, and the L5 dorsal root was cut close to the spinal cord. The L5 dorsal root was carefully examined and any communicating branches between the L5 and neighboring dorsal roots (L4 and L6) were cut to block any potential interference from other segments. In addition, the dorsal ramus of the L5 spinal nerve, which is located just proximal to the ligated site, was cut to prevent having intact afferent inputs from the dorsal part of the skin and muscle. Single-unit activity was recorded from teased L5 dorsal rootlets. Recorded signals were amplified with an AC-coupled amplifier (WPI, DAM-80). Amplified signals were led to a digital oscilloscope (Nicolet 410) for display and to a window discriminator (Mentor N-750) whose output was used to drive a speaker and to compile time histograms via a data acquisition system (CED 1401). Only signals larger than twice the noise level were accepted for further analysis.
A systematic sampling method was adapted to compare ectopic discharges in different groups. The L5 dorsal root was divided into four fascicles of approximately equal size. One of these fascicles was randomly chosen and placed on a mirror-based platform; the remaining three were discarded. The chosen fascicle was further divided into 32 fine fascicles of approximately equal size using a pair of finely sharpened forceps under a dissecting microscope. Therefore, each fine fascicle represented approximately 1/128 of the whole L5 dorsal root. Each fine fascicle was placed, in turn, on a unipolar platinum hook recording electrode. When each fascicle was subjected to recording, it usually contained ongoing activity from 0–3 individual units, clearly distinguishable by the amplitude and shape of their signals. For each fine fascicle, the following parameters were measured: (1) whether or not the fascicle contained any spontaneously active unit; (2) the number of distinguishable units showing spontaneous activity; and (3) the average frequency of spontaneous activity of each unit (calculated 5-min recordings). Each fascicle was examined for at least 1 min before concluding that there was no active unit present.
At the end of some of these experiments, a pair of bipolar hook electrodes was placed on the L5 spinal nerve just proximal to the ligated site in order to stimulate it for the purpose of calculating conduction velocities of the recorded units.
2.4. Statistical analysis
Data are expressed as mean±standard deviation (SD). Statistical analysis was done using parametric statistics, including analysis of variance (ANOVA) followed by the Dunnett's test and the Student t test. P values less than 0.05 were considered to be significant.
3. Results
3.1. Preliminary study for ectopic discharge recording
Ectopic discharges were recorded from dorsal root fascicles (a quarter of the L5 dorsal root was divided into 32 pieces), each being 1/128th of the whole L5 dorsal root. We will refer to each small fascicle as a ‘bundle’ in this study. Many bundles contained ongoing discharges of one or more units. Individual units could usually be distinguished from each other based on difference in the amplitudes and shapes of the action potentials.
Although activity from large-sized units could be recorded from each bundle, it was not clear whether any of the activity was from small-diameter fibers since it is much more difficult to record action potentials from small- rather than large-diameter fibers. As a preliminary study, we examined whether or not this size of bundle is suitable for recording small-fiber activity, if present. In two normal (unoperated) rats, the L5 dorsal root was divided (as described above) into small fascicles each comprising 1/128th of the root. These were placed, one by one on a recording electrode. The sciatic nerve was cut, and the proximal stump was electrically stimulated with an intensity that was suprathreshold for C fibers. As shown in Fig. 1, many slowly conducting single-unit responses were recorded, suggesting that activity of at least some Aδ and C fibers can be recorded from this size bundle.
;)
Fig. 1: Capability of the experimental procedure employed in the present study for the recording of action potentials from small-caliber fibers. The L5 dorsal root of the normal rat was cut close to the spinal cord and the distal stump was divided into 4 fascicles of equal sizes. One of these fascicles was further divided into 32 small bundles, and each bundle was placed on a recording electrode while stimulating the proximal cut end of the sciatic nerve. Many individual action potentials evoked by sciatic nerve stimulation could be identified, including Aδ and C fibers, as shown here as an example. Three consecutive tracings are shown. The arrow at the bottom indicates the time of stimulation. Based on the conduction distance and the latency, the first action potential (the first angled arrow) corresponds to an Aδ fiber while the second one corresponds to a C fiber.
3.2. Ectopic discharges and their firing pattern
Fascicles of the L5 dorsal root in rats with ligated spinal nerves were examined at all tested PO times (1, 10, 26 weeks), and many contained spontaneously active units. Since these units were disconnected (due to tight ligation of the spinal nerve [ventral ramus] and additional acute section of the dorsal ramus of the spinal nerve) from their peripheral sensory receptors, which are the normal site of impulse generation, this activity must have been generated from a site that does not normally trigger impulses. That is, this spontaneous activity must represent ectopically generated discharges. The pattern of ectopic discharges varied widely among individual units. Several examples are shown in Fig. 2. The most common pattern was regular firing (Fig. 2A). In this pattern, firing frequency varies very little during the observation period. The other extreme was a bursting pattern, an example of which is shown in Fig. 2D. Fig. 2B,C show two examples of fluctuating discharge patterns. The four patterns, given in Fig. 2, suggest that there are different patterns of ectopic discharges. However, many units displayed a firing pattern that was a mixture of these four examples, making it difficult to categorize their activity into a set of distinctive patterns.
Fig. 2: Various firing patterns of ectopic discharges. (A–D) Time histograms of ectopic discharges recorded from four different units. The inset in (A) shows a sample recording of action potentials taken while compiling the time histogram shown in (A).
3.3. Time course of changes in ectopic discharges and behaviors
Fig. 3A–C show changes in ectopic discharges at different PO time points. As a sampling procedure, we counted the number of bundles containing at least one spontaneously active unit (we refer to these as ‘active bundles’). About 70% of the sampled 32 bundles contained active units at 1 week PO time, and the percentage gradually declined as PO time progressed (Fig. 3A). Since individual units could be distinguished in the recordings, the total number of recorded units during the entire sampling procedure (sampling from 32 bundles) was counted. At 1 week PO time, the total number of active units in the 32 bundles was about 50, and that number gradually declined as PO time progressed (Fig. 3B). The average discharge frequency of all units was about 10 Hz at 1 week PO, and this also declined at later PO times (Fig. 3C). In summary, as PO time progressed, fewer spontaneously active units were found in fewer bundles, and the remaining active units tended to discharge at a lower frequency.
Fig. 3: Changes in ectopic discharges and pain behaviors at different PO time periods. (A–C) Three aspects of ectopic discharges. (A), ACTIVE BUNDLES refers to the percent of 32 sampled dorsal root fascicles (each represents about 1/128 of the whole L5 dorsal root) that contained at least one spontaneously active unit. (B), TOTAL UNITS represents the cumulative number of distinguishable units in recordings from 32 bundles. (C), FREQUENCY plots the average frequency of the total units. D-F show three components of neuropathic pain behavior. (D), VF 54.4 mN refers to the frequency of the foot lifts elicited by the stronger of the two von Frey filaments used (bending force 54.4 mN), and the values represent the magnitude of mechanical allodynia. (E) The frequency of foot lifts elicited by acetone application, and the values represent the magnitude of cold allodynia. (F) Cumulative foot lift duration during 5 minutes of observation when the rat was placed on a cold plate (5°C), and the values represent the degree of cold-stress exacerbated ongoing pain. Asterisks indicate values significantly different from the value at 1 week (1 W) PO (P<0.05 by ANOVA followed by the Dunnett's test). The n values for 1, 10 and 26 W were 13, 9, and 12, respectively. Preoperative behavioral data for the pooled three groups were: 3.9±9.3% for VF 54.4 mN; 7.3±16.4% for acetone; and 0.16±0.42 s for cold plate.
Fig. 3D–F show changes in test scores for various neuropathic pain behaviors at different PO time points. These behavioral data were obtained on the same day as, and just prior to, the ectopic discharge recording session. The plotted test scores included behavioral signs indicative of mechanical allodynia (VF 54.4 mN; test scores for VF 8.4 mN were slightly smaller in magnitude, but had a similar pattern to VF 54.4 mN), cold allodynia (acetone), and cold-stress exacerbated ongoing pain (cold plate). As PO time progressed, the magnitudes of all tested behavioral signs declined.
An attempt was made to correlate the characteristics of ectopic discharges and the various neuropathic pain behaviors. Data were pooled from all 3 PO groups (1, 10 and 26 weeks). Then, the five behavioral response variables (responses to applications of two strengths of von Frey filaments; responses to acetone; and foot lift duration on the neutral temperature and cold plates) were correlated with three ectopic discharge variables: (1) % active bundles, (2) the total number of sampled units displaying ectopic discharges, and (3) the average frequency of discharge rate. The results are shown in Table 1. All three discharge variables were significantly correlated with the 5 behavior variables, except in three cases (one for neutral temperature and two for cold plate tests).
Table 1: Correlation between ectopic discharges and various behavioral outcomes in 3 groups of animals representing three different PO times (n=34; 1, 10, and 26 W). Pearson correlation coefficients are showna
3.4. Conduction velocity
Conduction velocity was not measured routinely because the part of the L5 spinal nerve proximal to the ligation was too short to allow the placement of stimulating electrodes without causing mechanical disturbance to the neuroma, which may influence the characteristics of the ectopic discharges. However, attempts to record conduction velocity were made at the end of the period of sampling the ectopic discharges. After completing the sampling, a pair of stimulating hook electrodes was placed on the spinal nerve in some experiments. Spontaneously active units found during the sampling procedure were reexamined. The activity was evoked by stimulating the spinal nerve, and three criteria were used to include the unit in the conduction velocity sample: (1) the action potentials of the unit could be clearly separable from others as an individual unit activity; (2) the action potentials showed a clear all-or-none response to varying stimulus intensity; and (3) the amplitude and shape of action potentials were the same as the spontaneously active unit. The conduction velocity was calculated by dividing the conduction distance by the latency.
Conduction velocities of 50 spontaneously active units were obtained from 13 rats in group 1 (1 week PO). Axons conducting > 14 m/s are considered to be Aβ fibers; those conducting 2–14 m/s are Aδ fibers; and axons conducting slower than 2 m/s are C fibers (Waddell and Lawson, 1990; Ritter and Mendell, 1992). As shown in Fig. 4A, 29 (58%) of the axons were Aβ and 19 (38%) were Aδ fibers. Only two units were found to conduct at a velocity lower than 2 m/s and hence were categorized as C fibers. As shown in Fig. 4B, 20 additional units were obtained from groups 2 and 3 (10 and 26 weeks PO). Interestingly, the population of active fibers shifted to larger-sized fibers so that the majority of these fibers (85%) were Aβ and the remaining (15%) were Aδ fibers. No C fibers were included. However, the small number of observations makes it difficult to draw a firm conclusion about population shift toward larger-sized fibers at the later PO times.
Fig. 4: Distributions of conduction velocities. (A) Histogram of the conduction velocities of fibers displaying ectopic discharges at 1 W PO time. (B) Histogram for fibers recorded at later PO times (10 and 26 W).
3.5. Sympathectomy
The effects of surgical sympathectomy on ectopic discharges and neuropathic pain behaviors were examined by comparing them in two groups of rats: a 1 week PO neuropathic surgery group (group 1) and a 1 week PO sympathectomy + neuropathic surgery group (group 4). As shown in Fig. 5A–C, the various measures of ectopic discharges were significantly lower in group 4 than in group 1. As shown in Fig. 5D–F, the levels of all examined behaviors, except sensitivity to cold (response to acetone application), were also significantly lower in group 4. The data suggest that the amelioration of neuropathic pain behaviors by sympathectomy in neuropathic animals is accompanied by a reduction in ectopic discharges from injured sensory neurons.
Fig. 5: Changes in ectopic discharges and pain behaviors after sympathectomy (SYMX). A–C show three aspects of ectopic discharges, as in
Fig. 3D,F show three components of neuropathic pain behavior, as in
Fig. 3. Asterisks indicate values significantly different from the value at 1 week (1 W) PO (
P<0.05 by Student's
t-test). The
n values for 1 W and SYMX were 13 and 10, respectively.
4. Discussion
The present study examined changes in ectopic discharges over time after injury in the SSI model of neuropathic pain. All aspects of measured ectopic discharges declined as PO time passed. Neuropathic pain behaviors also declined in similar fashion over the same time period. Surgical sympathectomy on neuropathic animals lowered ectopic discharges along with neuropathic behaviors. The data suggest that the amount of ectopic discharge is generally well correlated with the degree of pain behavior in a variety of conditions in neuropathic animals.
Shortly after injury to a peripheral nerve, many damaged sensory neurons produce spontaneous discharges from a site away from that of normal impulse generation (Wall and Gutnick, 1974; Devor and Jänig, 1981; Scadding, 1981; Korenman and Devor, 1981; Kajander and Bennett, 1992; Devor et al., 1994). These ectopic discharges enter the spinal cord and sensitize dorsal horn neurons. Therefore, the sensitization of dorsal horn neurons by ectopic discharge input is the critical step in the development and maintenance of neuropathic pain. In fact, blocking such input from entering the spinal cord by dorsal rhizotomy abolishes pain behaviors in neuropathic rats (Sheen and Chung, 1993; Yoon et al., 1996). Furthermore, blocking ectopic discharges by applying local anesthetics to trigger points in neuropathic pain patients produces immediate relief from pain (Gracely et al., 1992).
We found a significant correlation between many variables of ectopic discharges and pain behaviors. This positive correlation suggests that ectopic discharges are important for maintaining neuropathic pain behaviors. However, correlation alone does not sufficiently prove the causal relationship, since numerous unrelated factors could possibly have influenced the correlation. For example, the level of ectopic discharges may simply decline with age in rats and may not necessarily depend on the time after nerve injury. However, the positive correlation as shown in this study, along with the results of blocking ectopic discharges, as mentioned above, strongly suggests that ectopic discharges are a critically important factor for maintaining neuropathic pain behaviors.
Although it is clear that ectopic discharges play a critical role in the generation of neuropathic pain, it is not clear what aspect of ectopic discharges is most important. Potentially important variables include: (1) the average firing frequency, (2) the total number of active units, (3) a particular firing pattern, and/or (4) firings of a particular functional group of afferents. We measured the average firing frequency in this study. In addition, we attempted to examine other aspects of ectopic discharges. It is a difficult task to measure the total number of active units in any direct way. Furthermore, we may have underestimated the total number of active units, since the only criteria used for distinguishing an individual unit were the amplitude and shape of action potentials. Although we used these criteria for practical reasons, we may have missed counting multiple units, which have same amplitude and shape of action potentials. The total number of active units is needed to calculate the total discharge input (average frequency times total number of active units), which may be an important variable. Therefore, we used a sampling method for counting the number of active ‘bundles’ and then counted the number of units in each bundle. This sampling method will not result in an accurate count of the total number of active units, but we thought that it was a practical way to estimate the relative population of active units for a comparison between different neuropathic conditions. When we examined the firing patterns, we found that although different firing patterns were observed, many units could not be categorized into a distinctive pattern. We found it equally difficult to identify a functional group of afferents displaying ectopic discharges because these afferents were disconnected from their sensory receptors. As an alternative to functional identification of afferents, it is common to classify them according to their conduction velocity, since sensory fibers with similar function tend to have a similar conduction velocity. However, we could not measure conduction velocity routinely because placing the stimulating electrodes on the short length of the spinal nerve proximal to the ligation would have disturbed the neuroma in the SSI model.
Population variables (the number of bundles and/or units) correlated particularly well with behavioral components representing evoked pain (two strengths of von Frey filaments for mechanical allodynia and acetone for cold allodynia). Average frequency was also reasonably well correlated with various behavioral variables. In most experimental conditions, it is much easier to sample frequency than to estimate the population of units displaying ectopic discharges. Therefore, average frequency, which is a more practical measurement under most experimental conditions, might have some predictive value for behavioral outcome. However, none of the measured ectopic discharge variables, were perfectly correlated with behavioral variables. There are several possible reasons for this: (1) we failed to record the most important variable; (2) a combination of several variables (recorded and/or not recorded), rather than a single individual variable, was critically important; or (3) the inaccuracy of our sampling method prevented us from making a better correlation.
The conduction velocities of primary afferent fibers are much slower in the rat than in the cat. Hence, axons conducting > 14 m/s are considered to be Aβ fibers, those conducting 2–14 m/s are Aδ fibers, and axons that conduct slower than 2 m/s are C fibers (Harper and Lawson, 1985; Waddell and Lawson, 1990; Ritter and Mendell, 1992). We did not find many C fibers displaying ectopic discharges. In fact, the predominant population of axons with ectopic discharges was Aβ fibers. Since it is generally thought that inputs from small-caliber fibers are important for a long-term change in spinal dorsal horn neurons, which is an important mechanism underlying many chronic pain conditions, the lack of such input is difficult to explain. One obvious possibility is that we may have missed a large number of C fibers since recording from small-caliber fibers is much more difficult than from large fibers when the teased fiber recording technique is used. However, we did not find ectopic discharges in many small-sized fibers in neuropathic rats, despite the fact that we were able to record activity evoked in Aδ and C fibers of dorsal root fascicles in normal rats using a similar recording procedure. In the sciatic nerve neuroma model, 2–6% of C fibers are spontaneously active, and this level is maintained for a long period of time after sciatic nerve ligation (Welk et al., 1990; Devor, 1994). Furthermore, the incidence of spontaneously active A fibers gradually declines as time passes so that active C fibers predominate after the first PO month (Devor, 1994). Nevertheless, the number of spontaneously active small-caliber fibers in neuropathic rats is relatively small. Kajander and Bennett (1992) also found a relatively small number of spontaneously active small-caliber fibers in the CCI model of neuropathic pain. Is input from such a small number of fine fibers sufficient to maintain central sensitization? One possibility is that, after initial triggering, a low level of persistent input from a relatively small number of C fibers is sufficient to maintain central sensitization. Alternatively, although spinal input from small-caliber fibers is critically important for the initiation of central sensitization, large fiber input may play an important role in maintaining or enhancing central sensitization (Andersen et al., 1996; Cervero and Laird, 1996; Ma and Woolf, 1996a,b).
The sympathectomy group (spinal nerve ligation followed by surgical sympathectomy) showed a lower level of ectopic discharges along with a smaller magnitude of pain behaviors, when compared with the one week PO group (spinal nerve ligation alone). This suggests that the sympathetic nervous system contributes to the generation of ectopic discharges as well as to the production of pain behaviors. One problem with comparing these two groups, however, is that there is a difference in timing between the two groups after spinal nerve ligation. Rats in the one week PO group had spinal nerve ligation 7–10 days before the recording. On the other hand, two weeks had passed since the nerve ligation by the time of recording in the sympathectomy group. Therefore, it is possible that the difference between these two groups was due to spontaneous reduction in pain behaviors and the rate of ectopic discharges between 1 week (or 10 days) and 2 weeks. However, a previous study (Choi et al., 1994) showed that pain behaviors in our model are maintained at a stable level for about 6 weeks PO. Furthermore, the rate of ectopic discharges in the sympathectomy group is actually lower than in the 10-week PO group. Therefore, a lower discharge level and a smaller magnitude of pain behaviors in the sympathectomy group, when compared with the one-week PO group, most likely reflect the effect of sympathectomy. However, the fact that a substantial level of activity still remains after sympathectomy suggests that a component or components of ectopic discharge generator mechanisms exist independent of the sympathetic nervous system. Therefore, it is likely that there are at least two separate mechanisms that are involved in the generation of ectopic discharges by injured afferent nerves: one depends on the sympathetic nervous system and the other does not.
In summary, the present study examined various aspects of ectopic discharges from injured afferent neurons and multiple components of pain behaviors after spinal nerve injury. As postinjury time progressed, the level of ectopic discharges and the magnitudes of the various components of pain behavior declined. A similar decline of both ectopic discharges and behaviors was seen after surgical sympathectomy. The data indicate that the level of ectopic discharges is well correlated with behavioral outcome in a rat neuropathic pain model. The results suggest that there are both sympathetically dependent and sympathetically independent components of ectopic discharge generator mechanisms.
Acknowledgements
This study is supported by NIH Grants NS 31680 and NS 11255. HCH was supported, in part, by the Kil Chung Hee Fellowship Fund.
References
Andersen OK, Felsby S, Nicolaisen L, Bjerring P, Jensen TS, Arendt-Nielsen L. The effect of Ketamine on stimulation of primary and secondary hyperalgesic areas induced by capsaicin-a double-blind, placebo-controlled, human experimental study, Pain 1996;66:51-62.
Attal N, Jazat F, Kayser V, Guilbaud G. Further evidence for ‘pain-related’ behaviours in a model of unilateral peripheral mononeuropathy. Pain 1990;41:235-251.
Bennett GJ, Xie Y-K. A peripheral mononeuropathy in rat that produces disorders of pain sensation like those seen in man. Pain 1988;33:87-107.
Blumberg H, Jänig W. Discharge pattern of afferent fibers from a neuroma. Pain 1984;20:335-353.
Bonica JJ. Causalgia and other reflex sympathetic dystrophies. In: Bonica JJ, editor. The management of pain. Philadelphia: Lea and Febiger, 1990. pp. 220-243.
Campbell JN, Meyer RA, Raja SN. Is nociceptor activation by alpha-1 adrenoreceptors the culprit in sympathetically maintained pain? Am Pain Sol J 1992;1:3-11.
Cervero F, Laird JMA. Mechanisms of touch-evoked pain (allodynia): a new model. Pain 1996;68:13-23.
Choi Y, Yoon YW, Na HS, Kim SH, Chung JM. Behavioral signs of ongoing pain and cold allodynia in a rat model of neuropathic pain. Pain 1994;59:369-376.
DeLeo JA, Coombs DW, Willenbring S, Colburn RW, Fromm C, Wagner R, Twitchell BB. Characterization of a neuropathic pain model: sciatic cryoneurolysis in the rat. Pain 1994;56:9-16.
Desmeules JA, Kayser V, Weil-Fuggaza J, Bertrand A, Guilbaud G. Influence of the sympathetic nervous system in the development of abnormal pain-related behaviours in a rat model of neuropathic pain. Neuroscience 1995;67:941-951.
Devor M. The pathophysiology of damaged peripheral nerve. In: Wall PD, Melzack R, editors. Textbook of Pain. New York: Churchill Livingstone, 1994. pp. 79-100.
Devor M, Jänig W. Activation of myelinated afferents ending in a neuroma by stimulation of sympathetic supply in the rat. Neurosci Lett 1981;24:43-47.
Devor M, Jänig W, Michaelis M. Modulation of activity in dorsal root ganglion neurons by sympathetic activation in nerve-injured rats. J Neurophysiol 1994;71:38-47.
Dubuisson D, Dennis SG. The formalin test: a quantitative study of the analgesic effects of morphine, meperidine, and brain stem stimulation in rats and cats. Pain 1977;4:161-174.
Gracely RH, Lynch SA, Bennett GJ. Painful neuropathy: altered central processing maintained dynamically by peripheral input. Pain 1992;51:175-194.
Han HC, Lee DH, Chung JM. Correlation between pain behaviors and ectopic discharges in a rat neuropathic pain model. Neurosci Abst 1995;21:896.
Harper AA, Lawson SN. Conduction velocity is related to morphological cell type in rat dorsal root ganglion neurones. J Physiol 1985;359:31-46.
Häbler H-J, Jänig W, Koltzenburg M. Activation of unmyelinated afferents in chronically lesioned nerves by adrenaline and excitation of sympathetic efferents in cat. Neurosci Lett 1987;82:35-40.
Kajander KC, Bennett GJ. Onset of a painful peripheral neuropathy in rat: A partial and differential deafferentation and spontaneous discharge in Aâ–¡ and Aâ–¡ primary afferent neurons. J Neurophysiol 1992;68:734-744.
Kim SH, Chung JM. Sympathectomy alleviates mechanical allodynia in an experimental animal model for neuropathy in the rat. Neurosci Lett 1991;134:131-134.
Kim SH, Chung JM. An experimental model for peripheral neuropathy produced by segmental spinal nerve ligation in the rat. Pain 1992;50:355-363.
Kim KJ, Yoon YW, Chung JM. Comparison of three rodent neuropathic pain models. Exp Brain Res 1997;113:200-206.
Kim SH, Na HS, Sheen K, Chung JM. Effects of sympathectomy on a rat model of peripheral neuropathy. Pain 1993;55:85-92.
Kinnman E, Levine JD. Sensory and sympathetic contributions to nerve injury-induced sensory abnormalities in the rat. Neuroscience 1995;64:751-767.
Korenman EM, Devor M. Ectopic adrenergic sensitivity in damaged peripheral nerve axons in the rat. Exp Neurol 1981;72:63-81.
Ma Q-P, Woolf CJ. Basal and touch-evoked fos-like immunoreactivity during experimental inflammation in the rat. Pain 1996a;67:307-316.
Ma Q-P, Woolf CJ. Progressive tactile hypersensitivity: an inflammation-induced incremental increase in the excitability of the spinal cord. Pain 1996b;67:97-106.
Mao J, Price DD, Mayer DJ, Lu J, Hayes RL. Intrathecal MK-801 and local nerve anesthesia synergistically reduce nociceptive behaviors in rats with experimental peripheral mononeuropathy. Brain Res 1992;576:254-262.
Na HS, Leem JW, Chung JM. Abnormalities of mechanoreceptors in a rat model of neuropathic pain: possible involvement in mediating mechanical allodynia. J Neurophysiol 1993;70:522-528.
Ritter AM, Mendell LM. Somal membrane properties of physiologically identified sensory neurons in the rat: effects of nerve growth factor. J Neurophysiol 1992;68:2033-2041.
Roberts WJ. A hypothesis on the physiological basis for causalgia and related pains. Pain 1986;24:297-311.
Scadding JW. Development of ongoing activity, mechanosensitivity, and adrenaline sensitivity in severed peripheral nerve axons. Exp Neurol 1981;73:345-364.
Seltzer Z, Dubner R, Shir Y. A novel behavioral model of neuropathic pain disorders produced in rats by partial sciatic nerve injury. Pain 1990;43:205-218.
Sheen K, Chung JM. Signs of neuropathic pain depend on signals from injured nerve fibers in a rat model. Brain Res 1993;610:62-68.
Shir Y, Seltzer Z. Effects of sympathectomy in a model of causalgiform pain produced by partial sciatic nerve injury in rats. Pain 1991;45:309-320.
Waddell PJ, Lawson SN. Electrophysiological properties of subpopulations of rat dorsal root ganglion neurons in vitro. Neuroscience 1990;36:811-822.
Wall PD, Gutnick M. Ongoing activity in peripheral nerves: the physiology and pharmacology of impulses originating from a neuroma. Exp Neurol 1974;43:580-593.
Welk E, Leah JD, Zimmermann M. Characteristics of A- and C-fibers ending in a sensory nerve neuroma in the rat. J Neurophysiol 1990;63:759-766.
Yoon YW, Na HS, Chung JM. Contributions of injured and intact afferents to neuropathic pain in an experimental rat model. Pain 1996;64:27-36.
