Article
Reference
Influence of the sympathetic nervous system in the development of abnormal pain-related behaviours in a rat model of neuropathic pain
DESMEULES, Jules Alexandre, et al.
Abstract
This study evaluated the effect of surgical sympathectomy on pain-related behaviours in a well established model of peripheral mononeuropathy produced by loose ligatures around the common sciatic nerve in the rat. Behavioural abnormalities include spontaneous abnormal position of the hindpaw after the nerve constriction, indicative of "spontaneous pain", and changes in responses to mechanical or thermal stimuli applied to this paw. These changes are usually maximal at week 2 after the surgery, stable until weeks 3-4, and disappear between weeks 8 and 12. To assess the role of the sympathetic nervous system in the development and persistence of these abnormalities, four groups of rats were behaviourally tested: (i) rats receiving a complete sham surgery, (ii) rats with a sciatic nerve constriction produced by loose ligatures around the common nerve trunk plus a sham sympathectomy, (iii) rats receiving a lumbar sympathectomy with a sham nerve ligature, and (iv) rats receiving a simultaneous surgical lumbar sympathectomy and a sciatic nerve constriction. The efficacy of the sympathectomy was assessed by the measure of the [...]
DESMEULES, Jules Alexandre, et al . Influence of the sympathetic nervous system in the development of abnormal pain-related behaviours in a rat model of neuropathic pain.
Neuroscience , 1995, vol. 67, no. 4, p. 941-951
PMID : 7675215
Available at:
http://archive-ouverte.unige.ch/unige:90120
Disclaimer: layout of this document may differ from the published version.
1 / 1
Pergamon 0306-4522(95)00098-4
Neuroscience Vol. 67, No. 4, pp. 941-951, 1995 Elsevier Science Ltd Copyright 0 1995 IBRO Printed in Great Britain. All rights reserved 0306-4522195 $9.50 + 0.00
INFLUENCE OF THE SYMPATHETIC NERVOUS SYSTEM IN THE DEVELOPMENT OF ABNORMAL
PAIN-RELATED BEHAVIOURS IN A RAT MODEL OF NEUROPATHIC PAIN
J. A. DESMEULES,* V. KAYSER, J. WEIL-FUGGAZA, A. BERTRAND and G. GUILBAUDt
INSERM U 161, Unit& de Recherches de Physiopharmacologie du SystZme Nerveux, 2 rue d’Al&sia, 75014 Paris. France
Abstract-This
study evaluated the effect of surgical sympathectomy on pain-related behaviours in a well established model of peripheral mononeuropathy produced by loose ligatures around the common sciatic nerve in the rat. Behavioural abnormalities include spontaneous abnormal position of the hindpaw after the nerve constriction, indicative of “spontaneous pain”, and changes in responses to mechanical or thermal stimuli applied to this paw. These changes are usually maximal at week 2 after the surgery, stable until weeks 3-4, and disappear between weeks 8 and 12.To assess the role of the sympathetic nervous system in the development and persistence of these abnormalities, four groups of rats were behaviourally tested: (i) rats receiving a complete sham surgery, (ii) rats with a sciatic nerve constriction produced by loose ligatures around the common nerve trunk plus a sham sympathectomy, (iii) rats receiving a lumbar sympathectomy with a sham nerve ligature, and (iv) rats receiving a simultaneous surgical lumbar sympathectomy and a sciatic nerve constriction.
The efficacy of the sympathectomy was assessed by the measure of the noradrenaline level in the sciatic nerve. Sympathectomy reduced selectively or even prevented the abnormal reaction to cold temperature and to heat (45°C) in rats with a peripheral mononeuropathy. In contrast, the abnormal reaction to mechanical pressure was not influenced, and the behavioural abnormalities indicating spontaneous pain were still present. Sympathectomy alone resulted in a reduction of the vocalization threshold to pressure on both hindpaws, but also a short-lasting increased tolerance to cold immersion.
This study confirms the selective role of the sympathetic nervous system in affecting the development and maintenance of some abnormal pain-related behaviours to thermal stimuli in rats with a moderate, but persistent, constriction of one sciatic nerve.
A rat model of neuropathic pain, generated by four loose ligatures around the common sciatic nerve in the Sprague-Dawley rat strain, replicates some pain-related disorders seen in human peripheral neuropathy,‘,’ and is now commonly used in several laboratories with different approaches. In particular, these animals develop abnormal pain-related behaviours, which, according to the taxonomic classification of the IASP, may be named
“allodynia” and “hyperalgesia” to mechanical and thermal (hot and cold) stimulation of the nerve- injured paw. They also exhibit an abnormal position of this paw, possibly reflecting the reluctance of the animal to place it on the floor. The abnormal sensi- tivity is maximal between two and three weeks after the beginning of the constriction, with a recovery at
*Present address: Division de Pharmacologic Clinique et Consultation de la Douleur HBpital Cantonal Universitaire, 24 rue Micheli-du-Crest, 1211 Genive 14, Switzerland.
tTo whom correspondence should be addressed.
Abbreviations: EDTA, ethylenediaminetetra-acetate; SMP, sympathetically maintained pain.
eight to 10 weeks. Some pain-related behaviours have also been observed for the contralateral limb, and are clearly related to the intensity of abnormal reactions observed from the nerve-injured paw.*,’
Several aspects of neuropathic pain often seem to
be related to a pathological control exerted by the
sympathetic nervous system.“,26,34,53 In this respect,
the abnormal thermal sensitivity, particularly marked
for cold stimulus in rats with the ligated sciatic
nerve,30~5’.54 may reflect an involvement of the sym-
pathetic system. Indeed, cold “allodynia” is some-
times used to assess the presence of sympathetic
maintained pain in humans.9,26,34 It has been demon-
strated that sciatic nerve constriction in rats induces
dramatic axonal degeneration not only of myelinated,
but also unmyelinated fibres.5~‘4~‘6~27~32.50,52 It is not
known to what extent damaged unmyelinated fibres
are afferent or efferent. Initial findings, however,
suggest the involvement of the sympathetic system in
some of the behavioural signs observed in this
mononeuropathic model: (i) the mononeuropathy is
associated with abnormal variations in paw tem-
perature6,63 and (ii) sympatholytic treatment with
941942 J. A. Desmeules et al.
guanethidine selectively prevents or suppresses the exaggerated sensitivity to thermal stimuli, mainly to cold stimuli.5’.54
To further analyse the influence of the sympathetic system in the development of the various behavioural abnormalities seen in mononeuropathic rats, the effects of surgical sympathectomy on the emergence of pain-related behaviours occurring spontaneously, or elicited by mechanical or thermal stimulation in the model of neuropathy in the rat, were evaluated in the present study. Preliminary data have been reported.”
EXPERIMENTAL PROCEDURES Animals
Albino male Sprague-Dawley rats (Charles River, France), all eight weeks old, weighing 175-200 g on arrival, were used: n = 36 for the behavioural study, n = 25 for the biochemical assessment of the sympathectomy. The animals were allowed to habituate to the colony room for at least one week before the beginning of the experiment.
For the behavioural study, rats were randomly assigned to one of four different surgical groups:
Sham-operated rats in which the medial portion of the sympathetic chain and both sciatic nerves (at the mid- thigh level) were exposed but not injured (n = 9).
Sciatic nerve-constricted rats, in which the left sciatic nerve was ligated and a sham surgical procedure performed to expose the right sciatic nerve and the portion of the sympathetic chain (n = 9).
Sympathectomized rats in which a transperitoneal sympath- ectomy was performed and both sicatic nerves were exposed but not injured (n = 9).
Sympathectomized rats with a sciatic nerve constriction in which (i) a transperitoneal sympathectomy, (ii) a loose ligation on the left sciatic nerve and (iii) a sham surgical procedure exposure of the right sciatic nerve were performed (n = 9).
For the high-performance liquid chromatography analysis of noradrenaline level (see below), rats were assigned in five groups: four groups in which rats received surgical pro- cedures similar to those described above and one group in which the rats, considered as control rats, did not receive any surgical procedure (n = 5 in each group).
Surgical procedures
The neuropathy was produced according to the method described in detail by Bennett and Xie’ and Attal et a1.j Briefly, rats were anaesthetized with sodium pentobarbital (50 mg/kg, i.p.) and four ligatures (5.0 chromic gut, about 1 mm spacing) were tied loosely around the common sciatic nerve at the level of the mid-thigh, so that circulation through the epineural vasculature was not interrupted. A sham procedure consisted of exposing the sciatic nerve without placing ligatures.
A lumbar surgical sympathectomy was performed under deep pentobarbital anaesthesia (50-60 mg/kg, i.p.). The left sympathetic abdominal chain was exposed using a medial transperitoneal approach. A complete dissection of the sympathetic chain from L2 to L5 was performed with the aid of a dissection microscope (lo-40 times magnification).
After segmental identification of the ganglia according to the criteria of Baron et al. (Fig. 1 in Ref. 4), and Koltzen- burg et a[.,# the left ganglia L22L5 were cut out. It was always possible to remove the appropriate section of the sympathetic chain in one piece. Due to the fusion of lumbar
ganglia, in 80% of Sprague-Dawley rats4 the sympathec- tomy often involved the contralateral part of the sympath- etic chain. As a result, the sciatic nerve ligation was performed on the left side. The wound was washed with saline and closed in layers (chromic gut 4.0). A sham sympathectomy consisted of a similar surgical procedure in which the sympathetic chain was exposed but not removed.
Since these experimental conditions induce a certain amount of suffering, the guidelines on ethical standards for experimental pain in animals proposed by the Committee for Research and Ethical Issues of the IASP were strictly followed.6’ In particular, the experimental duration was as short as possible and the number of animals used was kept to a minimum. In addition to the general housing conditions used for normal rats (in an animal room at a constant temperature of 22°C with a 12 h alternating light-dark cycle), special precautions were taken: after surgical pro- cedure, rats were housed in individual cages, then three to four days later, a maximum of three per cage. The floors of the cages were covered with sawdust in order to minimize the possibility of painful mechanical stimulation. Food and water were available ad libitum. Operated animals were able to eat and drink unaided.
Biochemical measurements
In the five groups of rats used for this part of the study (with or without a sciatic nerve constriction), a portion of the sciatic nerve from 5 mm above the level of the constric- tion to 5 mm below this level (equivalent to 150-200 ng protein) was taken at two weeks following surgical pro- cedure, the time of maximum abnormal behaviour. The sciatic nerve segments were homogenized by sonification in 100 ~1 0.1 M HClO, containing 0.1 mM EDTA and 0.01%
cysteine. After centrifugation at 35,000 g (4°C) for 40 min, the filtered supernatant was frozen to -80°C until it was used for analysis. The pellet was dissolved in 0.4 M NaOH at 4°C overnight and used for the measurement of proteins.
The level of noradrenaline was measured by using high- performance liquid chromatography with electrochemical detection, as described previously.29 The column used for separation of the compounds was an ESA HR80 (8 cm, packed with 3 pm spherical octadenylsilane; ESA, Inc., Bedford, MA, U.S.A.) and the buffer solution was the ESA CAT-A-Mobile-Phase. The flow rate was set at 1.5 ml/min;
the analysis was performed at 25°C. The electrochemical detection was made with a Coulochem 5100 (ESA) detector equipped with an analytical cell model 5014 (serial coupled cells). The potential of the first cell (coulometric cell) was set at f0.05 V to reduce the background level compared to the mobile phase, and the second at 0.35 V for noradrenaline measurement. The retention times, peak heights, calibration curves and concentrations were obtained from a data pro- cessor (Merck D-2000). In preliminary experiments, a con- trol of the identity of noradrenaline in the sciatic nerve extracts was made by verifying the identity of the peak height ratios observed at different potentials for the sciatic extracts and standard compounds.
Behavioural test procedures
Measurement of the rat paw temperature. The cutaneous hind paw temperature was measured on the external part of the plantar part in the territory of the sciatic nerve, using a thermistor probe (Thermocouple Digi-Sense Thermometer, Model No. 8528-10; Cole-Parmer Instrument Company, Chicago, IL, U.S.A.; 15mm diameter). To avoid possible changes due to the different thermal stimulations measure- ments were done at the beginning of the test session at 09.00 in a quiet room maintained at constant temperature (23-25°C). To avoid stress induction, paw temperature was measured without straining the rats. Rats were left in their cage and the thermocouple was gently placed on the external part of the plantar surface. Stable temperature readings were obtained after 20 s with a precision of O.l”C. For each
Influence of the sympathetic nervous system 943 rat, two consecutive measurements were averaged and both
sides were successively measured in random order.
Nociceptive behavioural tests. Four different behavioural tests, applied randomly by one experimenter with a delay of at least 4 h between each series of tests, were systematically performed in all rats before and at different time points after the surgical procedure.
Post-surgery testing sessions started two days after sur- gery and were performed every three days until the third week, then once a week for up to two months. Since rats with sciatic nerve constriction exhibit abnormal positions of the operated hind paw, even a blind experimenter initially unaware of the conditions of the experiments would rapidly discern experimental from control rats: therefore, it was impossible to perform the behavioural studies blindly.
However, rats were randomly assigned in groups of four for a series of tests, and before each testing session the exper- imenter was unaware of the previous results obtained by each of the individual animals.
To avoid sensitization and experimenter effects, the behavioural tests were performed in the same order and on two consecutive days as follows: (i) the score of “spon- taneous pain”, (ii) measure of the vocalization threshold to paw pressure, (iii) measure of the struggle latency to paw immersion in a 45’C water bath, on the first day, and (iv) measure of the struggle threshold to paw immersion in a cold water bath, on the second day.
Score qf spontaneous pain. The method described for the formalin test”.” and extensively used’,j6 was adapted as described initially (see Refs 2, 3, 36 and 54). Each rat was placed in a Plexiglas cylinder (19 x 31 cm), allowed to habituate for a few minutes before the test and observed for three consecutive 5-min periods. Six different positions of the nerve-injured paw, reflecting an increasing reluctance of the rats to place the paw normally on the floor, were rated:
from 0, the normal position, to 5, the most abnormal position when the rat was licking its paw. For each 5-min period the time spent in each position was noted, expressed in seconds, and the score was calculated according to the formula (T,xI+T2~2+T~~3+T~~4+Tsx5)/300
(T, , T2. are the times expressed in seconds, spent in positions I, 2. over a total of 5 min). Then, using three consecutive score values, the mean score was calculated for each particular rat.
Vocalization threshold to paw pressure. The vocalization threshold to paw pressure was determined by a method adapted from the RandalllSelitto description, with a Basile Analgesymeter (Apelex) as currently used in our group?’ a constant increasing pressure is applied to the hind paw until the rat squeaks with the 1 mm tip diameter of the pusher of the Basile (Apelex). The probe was positioned between the third and fourth metatarsus of the dorsal part of the paw, in the sciatic nerve territory. For each rat, two stable consecutive threshold determinations (expressed in
grammes) were carried out and averaged. In these con- ditions and according to the taxonomy proposals of the IASP,4R a decrease in thresholds for vocalization would reflect mechanical “allodynia”: a nociceptive reaction for a non-nociceptive stimulus in normal conditions.
Struggle latency to 45°C. The struggle latency to paw immersion in a hot water bath was determined by immersion of one hind paw in a 45°C water bath (in the noxious range), as described previously.3 Each rat was carefully handled and wrapped in a towel so that only the limbs and the head were free. The hind paw, including the heel and the digits, was immersed in the 45°C water bath until the rat struggled. For each rat, two consecutive latency determinations (expressed in seconds) were carried out (with an interval of at least 30 min) and averaged. According to this procedure and the taxonomic proposals of the IASP,4S a decrease in latency for struggle reflects thermal “hyperalgesia”: an exacerbated reaction to an initially nociceptive stimulation.
Struggle threshold to cold. To evaluate the struggle threshold to cold, one hind paw was immersed in successive cold water baths (10, 8, 6, 4, 2°C) until the rat struggled (with a cut-off at 15 s to avoid stress reaction or tissue damage). An interval of at least 30 min was allowed before applying the next cold stimulation. For each rat, two stable consecutive threshold determinations (expressed in
“C) were carried out and averaged. An increase in the cold temperature threshold was indicative of cold “allodynia”
according to the taxonomic proposals of the IASP!”
a nociceptive reaction for a non-nociceptive stimulus in normal conditions.
Statistical procedure
To evaluate the influence of surgical procedures, ANOVA was used, taking the control threshold prior to the surgical procedure as the co-variable and the values obtained at different times after surgery as the variable. For com- parison of each time point, a Tukey test was used. A paired Student f-test was employed to compare the pre- and the maximally modified post-operatory values, as well as for the comparison between the nerve-injured and the contralateral paw.
RESULTS
The mean weight of all animals increased over the two month observation, reaching 400&.5OOg at the end of the observation period without any significant differences between normal, neuropathic and sympa- thectomized rats. The biochemical measurement of noradrenaline in the common sciatic nerve, per- formed in 25 additional rats by high-performance liquid chromatography with coulometry detection2”
Table 1. Biochemical measurement (high-performance liquid chromatography with electrometry detection) of noradrenaline in the common sciatic nerve in control rats without surgery, and in rats two weeks after sham surgery, a loose sciatic nerve constriction, a surgical lumbar sympathectomy or both lesions (n = 5 in each group);
results are expressed in pmol/mg protein
Sham Nerve Nerve constriction
Control operated constriction Sympathectomy + sympathectomy
5.9 5.9 5.2 1.9 n.d.
6.0 5.0 n.d. n.d. n.d.
9.0 10.2 n.d. nd. n.d.
3.4 14.0 nd. n.d. n.d.
7.5 17.3 n.d. n.d. nd.
With the unexplained exception of one rat with sciatic nerve constriction, there was a dramatic drop in noradrenaline after the three types of surgical lesion; n.d., non detected.
944 J. A. Desmeules et al.
A NERVE INJURED PAW 34.
$I 32s g 30- 5
e:
26- E s
26.
1 24.
a
p 22- ,
0 2 4 6 a 10
weeks
CONTftALATERAL PAW
e
= 30- F?e: 26- E d 26- ii 24.
li 221 , , 1 ,
I0 2 4 6 6 10
weeks
Fig. 1. Skin temperature of the experimental (A) and contralateral (B) paw at different times before (point 0) and after surgical procedure: sciatic nerve constriction (open squares) sympathectomy (open diamonds) or sciatic nerve
constriction and sympathectomy (filled circles).
showed a dramatic drop in noradrenaline
in rats with either a sympathectomy, a sciatic nerve constriction or both lesions at week 2 after surgery, as compared to normal and sham-operated rats (Table 1).
The basal skin temperature was slightly different in the various groups of rats (Fig. l), but was comparable for both hindpaws in each rat and in each group. The vocalization threshold to paw pressure, the struggle latency to 45°C stimulus and the struggle threshold to cold were similar for both hind paws in each rat and in each group (Table 2).
Sham animals
As expected, the vocalization threshold to paw pressure, the struggle latency to 45°C and the struggle threshold to cold from both hind paws remained steady without significant differences from pre- operative values [ANOVA, all F(l3,112) = 1.16, ns].
In addition, the animals exhibited no abnormal position of the paws (score of spontaneous pain = 0).
I I
2 4 6 I I 1
6 10
t
Weeks
Fig. 2. Time course of the “spontaneous pain-related behaviour” rated according to the position of the nerve- injured paw at different time points of the study after sciatic nerve constriction (filled circles) or after sciatic nerve con- striction and sympathectomy (open squares). The two groups of animals exhibited a similar evolution, significantly different from the zero score, observed in sham rats at P < 0.01 (**). Sympathectomy alone did not induce
modification in the paw position.
Mononeuropathic animals
Paw temperature.
After an initial increase in the temperature of the nerve-injured paw at post- operative days 2-3, the temperature decreased and remained steady between days 16 and 28 (when the rats had clear abnormal pain-related behaviours;
Fig. lA, open squares). A comparable time course was seen for the contralateral paw (Fig. lB, open squares).
Score of spontaneous pain-related behaviour.
In contrast to the sham rats, the sciatic nerve-ligated rats never placed their paw normally on the floor, as described previously.3 The mean score of spon- taneous pain-related behaviour, using the abnormal hind paw position as a behavioural parameter, was maximal two weeks after sciatic nerve constriction (mean score of spontaneous pain = 1.71 f 0.23;
Tukey test, P < 0.01) still significant at week 3, then regularly decreased over the observation period with normalization at week 8 [ANOVA, F(13,112) = 14.9, P < 0.001; Fig. 21.
Vocalization threshold to paw pressure.
As described previously,3 the sciatic nerve constriction produced a significant modification of the vocaliza- tion threshold to paw pressure on the nerve-injured paw compared to the pre-operative value [ANOVA,
Table 2. Preoperative values of behavioural tests in the various groups of ratsVocalization threshold 45°C latency Cold threshold
(g, kS.E.M.) (s, +S.E.M.) (“C, kS.E.M.)
-
Left paw Right paw Left paw Right paw Left paw Right paw Sham-operated 329 k 19 333 * 11 11 + 0.9 11.5*0.9 5.6 & 0.8 5.2 * 0.7 Nerve constriction 322 + 19 328 & 19 9.1 +0.5 9.5 + 0.5 5.5 * 0.5 5.5 & 0.6 Sympathectomy 333 * 15 330 + 6 9.0 & 1.1 9.4 k 0.8 5.5 f 1 5.9 * 1 Nerve constriction
+ sympathectomy 338 * 20 345 + 12 10.2 f 0.9 10.7 k 0.8 5.4 rf: 0.3 5.5 * 0.3
Influence of the sympathetic nervous system 945 F(l3,112) = 6.63, P <
O.OOl].
Many animals devel-oped a transient hypoalgesia from the nerve-injured paw in the first five post-operative days, but the mean vocalization threshold did not differ significantly from the mean control value (398 f 63 g) on post- operative day 3 (Fig. 3A). Subsequently, a marked decrease in this threshold, which may reflect
“allodynia” to mechanical stimulation, was observed for the nerve-injured paw. This maximal change in threshold was reached between the second and third weeks (mean vocalization threshold = 180 f 12.5 g on week 2; Tukey test,
P <0.01); recovery to pre- operative values were observed between weeks 8 and 10.
For the contralateral paw, the mean vocalization threshold was 28 1 f 10 g at week 2, but this decrease was not significant compared to the mean pre- operative threshold [ANOVA, F( 13,112) = 1.63,
P = 0.08).However, since there was an inter- individual variability of the time point at which the maximal mechanical sensitivity was reached, the mean value obtained at a specific time point (for instance at post-operative week 2) could minimize the degree of the abnormal behaviour. Thus, the peak values obtained for each paw in each rat were also analysed irrespective of the time point of their occurrence. The minimal mean vocalization threshold to paw pressure was 170 k 16 g on the nerve-injured paw and 243 + 11 g on the contra- lateral paw (representing 53% and 74% of pre- operative values, respectively), both significant as compared to the pre-operative values (Student’s paired t-test,
P <0.01).
Struggle latency to 45’C.
Compared to the pre- operative value, the latency for a struggle reaction to 45C regularly decreased for both paws over the observation period, with struggle reaction peaking between weeks 2 and 3 after surgery [ANOVA, F(l3,112) I 6.24,
P <0.01; Fig. 3B]. The mean struggle latency was 4.0 k 0.8 s for the sciatic nerve- injured paw and 7.0 k 1.0 s for the contralateral paw at week 2; recovery to the pre-operative values occurred between six and eight weeks.
Struggle threshold to cold.
Compared to the pre- operative value, the threshold of the struggle reaction to cold for the nerve-injured paw increased over the observation period [ANOVA, F(l3,112) = 3.67, P < O.OOl]. The maximal change in threshold tem- perature peaked at week 2 and remained stable until week 4: the mean struggle threshold was 9.33 + 0.44”C at day 18 (Tukey test,
P <0.01;
Fig. 3C). Recovery to the pre-operative values was observed at week 9. The mean threshold for a struggle reaction to cold for the contralateral paw was non- significantly increased [7.67 f 0.37”C at day 18;
ANOVA, F(l3,lll) = 0.87, P =
0.571.
Sympathectomized animals
Paw temperature.
The time course of the skin temperature was comparable for the two paws, and
comparable to that obtained in animals with a ligated sciatic nerve. After an initial warming following the sympathectomy, the temperature returned to the pre-surgery value: at week 2, the temperature was not
A
g
IOJ
B
lo-
IJ-
6-
4-
Weeks Sciatic constriction
Fig. 3. Time course of the three phasic behavioural tests at different times after a sciatic nerve constriction (filled symbols) and in control, contralateral paw (open symbols).
(A) The mean vocalization threshold to paw pressure (expressed in grams) was significantly decreased as com- pared to the pre-operative value for the nerve-injured paw.
(B) The struggle latency to paw immersion in a 45-C water bath (expressed in seconds) was also significantly decreased as compared to the pre-operative value for the nerve-injured paw. (C) The struggle threshold to paw immersion in a cold water bath (expressed in “C) was significantly increased compared to the pre-operative value for the nerve-injured
paw (*P < 0.05, **P < 0.01).
946 J. A. Desmeules ef al.
significantly different from the initial value (Fig. 1, open diamonds). With sympathectomy alone, there was no spontaneous pain-related behaviour (mean score of spontaneous pain = 0).
~0calization threshold to paw pressure. Overall, sympathectomy alone produced a significant decrease in the vocalization threshold to paw pressure for both hind paws as compared to the pre-operative values [ANOVA, F( 13,112) = 5 and 3.6 for each paw re- spectively; P < O.OOl], which was more pronounced between weeks 2 and 3 after surgery (Fig. 4A). At this time point the mean vocalization threshold was 220 + 19 g for the left paw (Tukey test, P < 0.01) and 249 k 24 g for the right paw (Tukey test, P < 0.05).
Recovery to pre-operative values occurred at week 6.
Taking into account the maximal individual threshold decrease, the mean vocalization threshold to paw pressure reached 198 k 11 g for the left paw (i.e. 59% of the pre-operative value) and 218 f 15 g for the right paw (i.e. 66% of the pre-operative value) (differences significant for the two paws; Student’s paired t-test, P < 0.01).
Struggle latency to 45°C. Compared to the pre- operative value, the latency of the struggle reaction to 45°C was slightly decreased, significant for the left paw [mean value 11.5 f 0.9 s at day 9; ANOVA, F(13,112) = 2.35, P < 0.011 but not significant for the right paw [ANOVA, F(13,112) = 0.99, P < 0.46;
Fig. 4B].
Struggle threshold to cold. Compared to the pre- operative value, the threshold for struggle reaction to cold was significantly although transiently decreased for both hind paws [ANOVA, F(13,112) =2.73, P < 0.01; Fig. 4C]. The maximum change was seen at day 12: at this time point, the mean struggle thresholds were 2.7 + 0.3 and 3.1 f 0.3”C for the left and right paws, respectively.
Mononeuropathic rats with a sympathectomy Paw temperature. The time course of the paw temperature was comparable for the two paws, exhibiting a similar pattern as the other groups (Fig. 1, filled circles). In particular, between weeks 2 and 4 (time of the maximal pain-related behaviours in neuropathic rats), the paw temperature was approximately 26°C both in neuropathic rats (open squares) and in neuropathic animals with a sympathectomy (open diamonds).
Score of spontaneous pain-related behaviour. The score of spontaneous pain [ANOVA, F(13,112) = 9.50, P < O.OOl] was not modified by sympathectomy as compared to the score observed in animals with the sciatic nerve constriction (Fig. 2, open squares).
However, the peak of spontaneous pain was reached more rapidly, at day 9 vs day 15, with the mean score of spontaneous pain at day 9 being 1.8 f 0.2 (Tukey test, P < 0.01).
Vocalization threshold to paw pressure. The post- operative vocalization threshold to paw pressure was significantly lower than before surgery [ANOVA,
F(13,112) = 6.25, P < 0.001; Fig. 5A]; the threshold decrease was also observed for the opposite paw, although to a lesser extent [ANOVA, F( 13,112) = 3.88, P < O.OOl]. After sympathectomy, the mean minimal vocalization threshold to paw pressure declined to 153 & 13 g (45% of the pre-surgery value) and 211 k 15 g (61% of the pre-surgery value) for the
A
400
300.
200-
ioo-
B
14-
12-
lo- 8-
6-
4-
2-
C 10 1
r I I I I I
0 2 4 6 6 10
t
I I I I I 1
0 2 4 6 6 10
t
I I I I I I
i Sympathectomy
2 4 6
‘deks 10Fig. 4. Time course of the three phasic behavioural tests for both hind paws at different time points of the study after a sympathectomy (filled symbols). (A) The mean vocalization threshold to paw pressure was significantly decreased for both hind paws as compared to the pre-operative values.
(B) The struggle latency to 45°C was only slightly decreased at day 9. (C) The struggle threshold to cold was also
transiently decreased at day 12.
Influence of the sympathetic nervous system 941 nerve-injured and the contralateral paw, respectively
(significant differences for both paws; Student’s paired t-test,
P <0.01).
Struggle latency to 45°C and to cold.
The sympath- ectomy partially but significantly prevented the abnormal responses to heat stimulation for the nerve- injured paw [ANOVA, F( 13,112) = 0.99,
P =0.45;
Fig. 5B]. In contrast, the decrease of the struggle
A400 -
300 -
200 -
IOO-
B
14-
12-
lo-
6-
6-
4-
2-
C
IO-
a-
6-
.I-
2-
j
1
I I I I I I
0 2 4 6 6 10
t Weeks
Sciatic constriction + Sympathectomy
Fig. 5. Time course of the three phasic behavioural tests for both hind paws at different time points of the study after a sciatic nerve constriction and a sympathectomy (filled symbols). (A) The vocalization threshold to paw pressure was significantly decreased for both hind paws. (B) In contrast, there was no decrease in the struggle latency to 45°C. (C) There was no change in the struggle threshold
to cold.
latency for the contralateral paw was still significant compared to the pre-surgery value [ANOVA, F(13,112) = 1.92,
P <0.051. The sympathectomy completely prevented the abnormal reactions to cold stimulation for the two paws (Fig. 5C); the struggle threshold did not significantly differ from the pre- surgery values [ANOVA, F(13,112) = 1.07, ns].
DISCUSSION
In the present study we confirmed that using highly integrated behavioural tests,2,3,36.37*53 well quantified behavioural abnormalities indicative of painful neuropathy develop following loose con- strictive ligation of the sciatic nerve in rats. These abnormal behaviours reached their peak between the second and third weeks after surgery and were spontaneously reversible eight to 10 weeks after surgery. These behaviours and their time course, repeatedly observed and highly reproducible in our group, either with 4-O and 5-O chromic gut or silk, sharply contrast the short latency and duration abnormal withdrawal threshold to high thermal stimulus (three to five days) described as a conse- quence of a toxic effect of chromic gut.46 In our study, following sciatic nerve constriction alone, there was a progressive development of a decrease in the vocalization threshold to mechanical stimulation, the presence of spontaneous pain-related behaviour implicating the nerve-injured paw, and a decrease in the struggle latency to 45°C by several seconds.
We also observed a clear increase in the struggle threshold to mild cold thermal stimulation, a measure not previously studied, and which emphasizes the importance of the abnormal reaction to cold in this model. Another important factor is that with all of the phasic tests, abnormal reactions were also detected on the contralateral paw, not only with mechanical stimuli, as repeatedly observed in this mode12s3 and in inflammatory pain models3’ but also with thermal stimuli. Such phenomena are reminis- cent of mirror pain commonly seen in human causal- gias.” The putative mechanisms responsible for a
“secondary distant hyperalgesia” are likely to be due to cross-talk between neuronal populations at the spinal and/or the supraspinal level, and have been extensively discussed elsewhere.3’
In this study, surgical sympathectomy, the efficacy of which was verified by the biochemical analysis of noradrenaline in the injured sciatic nerve, prevented the abnormal reactions to cold and heat. In clear contrast, the abnormal reaction to mechanical stimuli (on both sides) and “spontaneous pain” persisted.
These data are in agreement with previous studies in
the same model using chemical sympathectomy
induced by repeated injections of guanethidine.5’.54 In
these first experiments, abnormal reactions to 10 and
46°C were significantly reduced or suppressed by the
chemical depletion of noradrenaline, while abnormal
reactions to mechanical stimulation or scores of
948 J. A. Desmeules ef al.
spontaneous pain remained unchanged. Our results are also in full agreement with those obtained after guanethidine treatment in another rat model of neuropathic pain in which the sciatic nerve was partially ligated transversally,59 but contrast with studies of rats with ligation of the two spinal nerves L5 and L6,4’ in which surgical and chemical sympath- ectomy relieved pain-related behaviours to heat and mechanical stimulus as we11.42,43 These differences may be due to the different types of nerve injury, but also to the different tests used to assess the nociceptive behaviours. The use of paw withdrawal tests, which are obviously under supraspinal controls, but still present in spinalized animals2’ are extremely different to the tests used in our present study, which are directly integrated at the supraspinal level.
Furthermore, the differential effects of several manipulations (morphine injection, spinothalamic tract or thalamic lesions) have strongly emphasized functional differences between the paw withdrawal and the vocalization thresholds to pressure.37.40.55 Moreover, the struggle latency to thermal stimulus is thought to be a more complexly organized reaction than the paw withdrawal.22 The differences between our data and those of Kim et aL4* might also be related to the relatively older age and higher weight of the rats used in our study, since in younger rats pain-abnormal reactions are considered to be more pronounced.60
Our present data are in agreement with clinical observations of sympathetically maintained pain (SMP) in which cold “allodynia” has been considered as a significant clinical sign of sympathetic dysfunction.‘0~“~26~49
The distinctive consequences of sympathectomy on thermal and mechanical abnormal reactions in mononeuropathic rats remain to be elucidated; how- ever, our study of the effect of sympathectomy alone on test behaviour further substantiates the understanding of the systems involved.
Why does sympathectomy relieve abnormal thermal reactions in neuropathic rats?
Role of a putative paw re-warming? A putative paw re-warming may occur after sympathectomy via an increase in cutaneous blood ~Iow,~‘,~~ thus being responsible for the temperature change of the struggle threshold to cold stimuli in sympathectomized rats (Fig. 5B), and consequently in mononeuropathic rats with a sympathectomy (Fig. SC). However, several arguments can be put forward to reject the major role of this re-warming in the present study: (i) in fact this re-warming did not appear to be specific for the sympathectomy and in any case was not durable, since it was not detected between weeks 2 and 4 after surgery (the time of the maximum pain-related behaviours in rats with sciatic nerve constriction);
(ii) the struggle threshold to cold or the struggle latency to 45°C were comparable in the various groups of rats, even with different basal skin tempera-
tures (Fig. 1, Table 2); (iii) sympathectomy alone did not induce a decrease in the struggle latency to hot temperatures; and (iv) on the contrary, when associated with the sciatic nerve constriction, sym- pathectomy dramatically reduced the pain behaviour elicited by immersion of the paw in a 45°C water bath (struggle latency increased by several seconds).4
Furthermore, the paw temperature may be biased by the test, since we measured the skin temperature and not the deep tissue temperature, in particular, the cutaneous vasculature within the dermis, where noradrenergic sympathetic innervation is predomi- nant.2* However, with the same measuring technique it is possible to detect a clear paw re-warming in neuropathic rats with 100 pg/kg i.v. of clonidine,38,39 but lower doses of clonidine (30-50 pg/kg, i.v.) relieve or suppress abnormal reactions to thermal stimulation (without influence on skin temperature).
Thus, the re-warming of the paw by the sympathetic lesion does not appear to be the major and certainly not the only mechanism of its beneficial effect on the hyper-responsiveness to thermal stimulation exhib- ited by the neuropathic rats. Furthermore, in patients suffering from SMP, the positive effect of phento- lamine also appears to be unrelated to changes in skin temperature,49x62 even if opposite vasomotor effects of the sympathetic and somatic systems observed in normal human subjects have a role in the frequent temperature skin fluctuations of SMP patients.53
Role of noradrenaline depletion? That the total depletion in noradrenaline in the sciatic nerve improves the abnormal reactivity to thermal stimu- lation could appear paradoxical, since in the mononeuropathic rats there is already a profound alteration in the peripheral content of noradrenaline, as initially shown63 and confirmed by the biochemical controls performed in the present study. However, it has been extensively demonstrated that, under normal conditions, the influence of sympathetic fibres on afferent fibres is usually minimal in rodents, but dramatically increases after a nerve injury as demon- strated by the use of adrenegic agonists.20.34,35.5628.66 Increased sensitivity to adrenergic substances has also been observed in patients suffering from causalgia or SMP after direct application of noradrenaline or derivatives,‘7.65 while the plasma concentration of noradrenaline seemed to be decreased in the affected limb.2’ The increased sensitivity may be due to several mechanisms, including a recently described noradrenergic sprouting within the dorsal root ganglion itself; 34,35,47 however, it should be stressed that this particular phenomenon has, so far, been observed for large afferent fibres, and it is not known whether it exists for cold and warm receptors. Thus, receptor dysfunction may underlie the pathological process, which is then suppressed by sympathectomy.
However, the dissociative effects depend on the stimulus modality inducing the pain-related behaviour, i.e. the lack of effect on mechanical abnormal reactions was unexpected.
Influence
of thesympathetic nervous system
949 Why sympathectomy does not relieve abnormalreactions to paw pressure nor the “spontaneous” pain- related behaviour in neuropathic rats
In addition to the lack of effect on abnormal reactions to paw pressure, there was no change in scores of “spontaneous” pain-related behaviour. This similarity further substantiates the hypothesis that this behaviour reflects the reluctance of the rats to place their paw normally on the floor. As mentioned above, similar effects have been observed previously with a chemical sympathectomy in this rat mode15’.54 and in another model of neuropathic pain.” The abnormal reactivity to paw pressure was enhanced by sympathectomy, and significant effects were observed for the paw contralateral to the injured nerve. In fact, these data fit well with those obtained in rats with a sympathectomy alone. Our results with sympathec- tomy in nerve-injured rats are in keeping with our results in rats with sympathectomy alone, in which there was a significant bilateral decrease in the vocal- ization threshold to paw pressure, likely responsible, in our opinion, for the lack of relief by sympathec- tomy of abnormal mechanical responses in neuro- pathic rats. The time course of the effect of the sympathectomy on mechanical reactions (maximum reached rapidly and recovery at the third week after surgery) fits well with the morphological degener- ation and regeneration signs reported after sympath- ectomy in the cat.’
The abnormal mechanical reaction after sympath- ectomy is also reminiscent of early experimental reportP4 and numerous clinical observations,25,45,6’,64 and the pathogenesis remains obscure. However, a transitory ephaptic connection between low- threshold mechanoreceptors and polymodal nociceptors, consecutive to degeneration and regener- ation of Schwann cells, expression of adrenoreceptors on peripheral afferents and/or a central link between the autonomic nervous system and primary afferents, may be possible mechanisms (references in Refs 12,
13, 20, 35 and 66).
Nevertheless, the differences in the modulation exerted by the sympathetic system according to the stimulus modality cannot be explained by these vari- ous data so far; it can only be proposed that different sensory receptors possibly differentially located in the skin or dermis are differentially innervated and modulated by the sympathetic system.**
Our findings emphasize the importance of using a variety of nociceptive tests based on various stimulus modalities and intensities in analysing the significance of tonic and phasic activity of the sympathetic system for each abnormal pain behaviour.” Moreover, since sympathectomy alone produced small transient abnormal behavioural reactions to mechanical stimu- lation, a study of the consequences of this procedure alone could be relevant for the pathophysiological investigation of post-sympathectomy pain syndrome.
CONCLUSION
Surgical sympathectomy selectively reduced or even prevented the abnormal reaction to cold tem- perature and to heat (45C) in rats with a peripheral mononeuropathy. In contrast, the abnormal reaction to mechanical pressure was not influenced, and the behavioural abnormalities indicating spontaneous pain were still present. Sympathectomy alone resulted in a reduction of the vocalization threshold to pressure on both hind paws, but also a short-lasting increased tolerance to cold immersion. This study confirms the selective role of the sympathetic nervous system in affecting the development and maintenance of some abnormal pain-related behaviours to thermal stimuli in rats with a moderate, but persistent, constriction of one sciatic nerve.
Acknowledgements-The authors wish to thank
Dr B. Vos, A. H. Dickenson and V. Chapman for useful suggestions and English revision. This research was partly supported by a grant program from the Bristol-Myers Squibb Company.and J.A.D. received a grant from the Fonds National Suisse pour la Recherche Scientifique.
REFERENCES
1. Andres K. H., von During M., Janig W. and Schmidt R. F. (1985) Degeneration patterns of postganglionic fibers following sympathectomy. Anar. Embryol. 172, 133-143.
2. Ardid D. and Guilbaud G. (1992) Antinociceptive effects of acute and chronic injections of tricyclic antidepressant drugs in a new model of mononeuropathy in rats. Pain 49, 279-287.
3. Attal N., Kayser V., Jazat F. and Guilbaud G. (1990) Further evidence for ‘pain-related’ behaviours in a model of unilateral peripheral mononeuropathy. Pain 41, 2355251.
4. Baron R., Janig W. and Kollman W. (1988) sympathetic and afferent somata projecting in hindlimb nerves and the anatomical organization of the lumbar sympathetic nervous system of the rat. J. camp. Neural. 275, 460-468.
5. Basbaum A. I., Gautron M., Jazat F., Mayes M. and Guilbaud G. (1991) The spectrum of fiber loss in a model of neuropathic pain in the rat: an electron microscopic study. Pain 47, 247-380.
6. Bennett G. J. and Ochoa J. L. (1991) Thermographic observations on rats with experimental neuropathic pain. Pain 45, 61-67.
7. Bennett G. J. and Xie Y. K. (1988) A peripheral mononeuropathy in rat that produces abnormal pain sensation like those seen in man. Pain 33, 87-107.
8. Bernard C. (1851) Influence du grand sympathique sur la sensibilite et la calorification. C. r. Sot. Biol.. Pari.v 163-164.
9. Besson J. M. and Guilbaud G. (1991) Lesions of Primary Afferent Fibers as a Tool for the Study of Clinical Pain.
Excerpta Medica, International Congress Series. Elsevier, Amsterdam.
950 J. A. Desmeules et al.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
37.
38.
39.
40.
41.
Blumberg H. and Janig W. (1993) Clinical manifestations of reflex-sympathetic dystrophy and sympathetically maintained pain. In Textbook of Pain (Eds Wall P. D. and Melzack R.), pp. 685-698. Churchill Livingstone, Edinburgh.
Bonica J. J. (1979) Causalgia and other reflex sympathetic dystrophies. In Advances in Pain and Research Therapy (eds Bonica J. J., Liebeskind J. C. and Albe-Fessard D. G.), Vol. 3, pp. 141-146. Raven Press, New York.
Campbell J. N. (1994) Can altered sympathetic function contribute to allodynia and ‘hyperalgesia’? In Wenner-Gren Center International Symposium, Touch, Temperature and Pain in Health and Disease, October 1993. IASP Publications.
Campbell J. N., Meyer R. A. and Raja S. N. (1992) Is nociceptor activation by alpha-l adrenoreceptors the culprit in sympathetically-maintained pain? Am. Pain S. J. 1, 3-l 1.
Carlton S. M., Dougherty P. M., Pover C. M. and Coggeshall R. E. (1991) Neuroma formation and numbers of axons in a rat model of peripheral neuropathy. Neurosci. Left. 131, 88-92.
Coderre T. J., Abbott F. V. and Melzack R. (1984) Effects of peripheral antisympathetic treatments in the tail-flick, formalin, and autotomy tests. Pain 18, 13-23.
Coggeshall R. E., Dougherty P. M., Pover C. and Carlton S. M. (1993) Is large-myelinated fiber loss associated with hyperalgesia in a model of experimental peripheral neuropathy in the rat? Pain 52, 233-242.
Davis K. D., Treede R. D., Raja S. N., Meyer R. A. and Campbell J. N. (1991) Topical application of clonidine relieves hyperalgesia in patients with sympathetically-maintained pain. Pain 47, 309-317.
Dennis S. G. and Melzack R. (1979) Comparison of phasic and tonic pain in animals. In Advances in Pain Research and Therapy (eds Bonica J. J. et al.), Vol. 3, pp. 7477759. Raven Press, New York.
Desmeules J. A., Kayser V. and Guilbaud G. (1992) Influence clinique dune sympathectomie chirurgicale sur un modele de douleur neuropathique chez le rat. II&me reunion de la Societe Francaise des Neurosciences, Strasbourg.
Devor M. and Wall P. D. (1990) Cross-excitation in dorsal root ganglia of nerve-injured and intact rats. J. Neurophysiol.
64, 173331746.
Drummond P. D., Finch P. M. and Smythe G. A. (1991) Reflex-sympathetic dystrophy: the significance of differing plasma concentrations in affected and unaffected limbs. Brain 114, 202552036.
Dubner R. (1994) Methods of assessing pain in animals. In Textbook of Pain (eds Wall P. D. and Melzack R.), pp. 2933302. Churchill Livingstone, Edinburgh.
Dubuisson D. and Dennis S. G. (1977) The formalin test: a quantitative study of the analgesic effect of morphine, meperidine and brain-stem stimulation in rats and cats. Pain 4, 161-174.
Dusser de Barenne J. G. (1931) L’influence du systeme nerveux autonome sur la sensibilitt de la peau. J. Psychol.
Norm. Path. 1777182.
Farcot J.-M., Grasser C. and Muller J.-F. (1990) Post-sympathectomy pain In Treatment of Chronic Pain (eds Mumenthaler M., Van Zwieten P. A. and Farcot J.-M.), pp. 134-139. Harwood Academic, New York.
Frost S. A., Raja S. N., Campbell J. N. and Khan R. A. (1988) Does hyperalgesia to cooling stimuli characterize patients with sympathetically-maintained pain (reflex sympathetic dystrophy)? In Proceedings of the Vth World Congress on Pain (eds Dubner R., Gebhart G. F. and Bond M. R.), pp. 151-156. Elsevier Science (Biomedical Division), Amsterdam.
Gautron M., Jazat F., Ratinahirana H., Hauw J. J. and Guilbaud G. (1990) Alterations in myelinated fibres in the sciatic nerve of rats after constriction: possible relationships between the presence of abnormal small myelinated fibres and pain-related behaviour. Neurosci. Lett. 111, 28833.
Gibbins I. (1990) The final common sympathetic pathways. In Reflex-Sympathetic Dystrophy: Pathophysiological Mechanisms and Clinical Implications (eds JPnig W. and Schmidt R. F.), pp. 1777196. VCH Verlagsgesellschaft mbH, Weinheim.
Godefroy F., Bassant M. H., Lamour Y. and Weil-Fugazza J. (1991) Effect of aging on dopamine metabolism in the rat cerebral cortex: a regional analysis. J. Neural. (Gen. Sect) 83, 18824.
Guilbaud G., Benoist J. M., Jazat F. and Gautron M. (1990) Neuronal responsiveness in the ventrobasal thalamic complex of rats with an experimental peripheral mononeuropathy. J. Neurophysiol. 64, 1537-1554.
Guilbaud G., Gautron M., Jazat F., Ratinahirana H., Hassig R. and Hauw J. J. (1993) Time course of degeneration and regeneration of myelinated fibres following chronic loose ligatures of the rat sciatic nerve: can nerve lesions be linked to the abnormal pain-related behaviours? Pain 53, 147-158.
Guilbaud G., Kayser V., Attal N. and Benoist J. M. (1992) Evidence for a central contribution to secondary hyperalgesia. In Hyperalgesia and Allodynia (ed. Willis W. D. Jr), pp. 187-201. Raven Press, New York.
JHnig W. (1988) Pre- and post-ganglionic vasoconstrictor neurons: differentiation, types and discharge properties.
A. Rev. Physiol. 50, 525-539.
J&rig W. and Schmidt R. F. (1990) Reflex sympathetic dystrophy. In Pathophysiological Mechanisms and Clinical Implications (eds Jlnia W. and Schmidt R. F.). VCH Verlaeszesellschaft mbH. Weinheim.
J&g W., Devor M., McLachlan E. M. and Michaelis M. (19j4? A new site of coupling between sympathetic neurones and afferent neurones following peripheral nerve lesions. In Pain Research and Clinical Management, Proceedings of the International Symposium on Pharmacological Aspects of Peripheral Neurones Involved in Nociception (eds Besson J. M., Guilbaud G. and Ollat H.), pp. 89-107. John Libbey, London.
Jazat F. and Guilbaud G. (1991) The ‘tonic’ pain related behaviour seen in mononeuropathic rats is modulated by morphine and naloxone. Pain 44, 977102.
Kayser V. and Guilbaud G. (1990) Differential effects of various doses of morphine and naloxone on two nociceptive test thresholds in arthritic and normal rats. Pain 41, 353-363.
Kayser V., Desmeules J. A. and Guilbaud G. (1992) Effects of noradrenergic agonists on chronic pain models: arthritic and mononeuropathic rats. In Towards the Use of Noradrenergic Agonists for the Treatment of Pain (eds Besson J. M.
and Guilbaud G.), pp. 151-167. Elsevier Science, Amsterdam.
Kayser V., Desmeules J. A. and Guilbaud G. (1995) Systemic clonidine differentially modulates the abnormal reactions to mechanical and thermal stimuli in rats with peripheral mononeuropathy. Pain 275-285.
Kayser V., Peschanski M. and Guilbaud G. (1985) Neuronal loss in the ventrobasal complex of the rat thalamus alters behavioural responses to noxious stimulation. In Aduances in Pain Research and Therapy (eds Fields H., Dubner R.
and Cervero F.), Vol. 9, pp. 277-284. Raven Press, New York.
Kim S. H. and Chung J. M. (1992) An experimental model for peripheral neuropathy produced by segmental spinal nerve ligation in the rat. Pain 50, 3555363.
Influence of the sympathetic nervous system 951 42. Kim S. H., Na H. S., Sheen K. and Chung J. M. (1993) Effects of sympathectomy on a rat model for peripheral
neuropathy. Pain 55, 85-92.
43. Kim S. 0. and Chung J. M. (1991) Sympathectomy alleviates mechanical allodynia in an experimental animal model for neuropathy in the rat. Neurosci. Left. 134, 131-134.
44. Koltzenburg M., Kress M., and Reeh P. (1992) The nociceptor sensitization by bradykinin does not depend on sympathetic neurons. Neuroscience 46, 465-473.
45. Leriche R. (1949) De la causalgie experimentale. In La Chirurgie de la Douleur, 3rd edn, pp. 1577159. Masson, Paris.
46. Maves T. J.. Pechman P. S.. Gebhart G. F. and Meller S. T. (1993) Possible chemical contribution from chromic cut sutures produces disorders of pain sensation like those seen in man. Pain 54, 57-69.
47. McLachian E. M., Janig W., Devor M. and Michaelis M. (1993) Peripheral nerve injury triggers noradrenergic sprouting within dorsal root ganglia. Nature 363, 5433546.
48. Merskey H. (1986) Classification of chronic pain-Descriptions of chronic pain syndromes and definition for terms.
Prepared by IASP Sub-Committee on taxonomy. Pain, Suppl. 3, ‘S-S225
49. Meyer R. A., Raja S. N., Treede R. D., Davis K. D. and Campbell J. N. (1990) Neural mechanisms of sympathetically maintained pain. In Rejex Sympathetic Dystrophy: Pathophysiological Mechanisms and Clinical Implications (eds Janig W. F. and Schmidt R. F.), pp. 57-66. VCH Verlagsgesellschaft mbH, Weinheim.
50. Munger B. L. G., Bennett G. J. and Kajander K. C. (1992) An experimental painful neuropathy due to nerve constriction. Axonal pathology in the sciatic nerve. Expl Neural. 118, 204-214.
51, Neil A., Attal N. and Guilbaud G. (1991) Effects of guanethidine on sensitization to natural stimuli and self-mutilating behaviour in rats with a peripheral neuropathy. Bruin Res. 565, 2377246.
52. Nuytten D., Kupers R., Lammens M., Dom R., Van Hees J. and Gybels J. (1992) Further evidence for a myehnated as well as unmyelinated fibre damage in a rat model of neuropathic pain. Expi Brain Res. 91, 73-78.
53. Ochoa J. L., Yarnitsky D., Marchettini P., Dotson R. and Cline M. (1993) Interactions between sympathetic vasoconstrictor outflow and C-nociceptor-induced antidromic vasodilatation. Pain 54, 191-196.
54. Perrot S.. Attal N., Ardid A. and Guilbaud G. (1993) Are mechanical and cold allodynia in mononeuropathic and arthritic rats relieved by systemic treatment with calcitonin or guanethidine? Pain 52, 41-47.
55. Peschanski M., Kayser V. and Besson J. M. (1986) Behavioural evidence for a crossed ascending pathway for pain transmission in the anterolateral quadrant of the rat spinal cord. Brain Res. 376, 164-168.
56. Sanjue H. and Jun Z. (1991) Sympathetic facilitation of sustained discharges of polymodal nociceptors. Pain 38, 85S90.
57. Sato J. and Per1 E. R. (1991) Adrenergic excitation of cutaneous pain receptors induced by peripheral nerve injury.
Science 251, 160881610.
58. Seltzer Z.. Dubner R. and Shir R. A. (1990) A novel behavioural model of neuropathic pain disorders produced in rats by partial sciatic nerve injury. Pain 43, 205-218.
59. Shir Y. and Seltzer Z. (1991) Effects of sympathectomy in a model of causalgiform pain produced by partial sciatic nerve injury in rats. Pain 45, 309-320.
60. Tanck E. N., Kroin J. S., McCarthy R. J., Penn R. D. and lvankovich A. D. (1992) Effects of age and size on development of allodynia in a chronic pain model produced by sciatic nerve ligation in rats. Pain 51, 3 13 316.
61, Tasker R. R., Buda T. and Hawrylyshyn P. (1991) The history of central pain syndromes, with observations concerning pathophysiology and treatment. In Deufferentu’tion Puin Syndromes: Puthophysiology and Treatment (eds Nashold B. S. Jr and Ovelmen-Levitt J.), pp. 31-58. Raven Press, New York.
62. Treede R. D., Raja S. N., Davis K. D., Meyer R. A. and Campbell J. N. (1991) Evidence that peripheral alpha-adrenergic receptors mediate sympathetically maintained pain. In Proceedings ofthe Vlth World Congress on Puin (eds Bond M. R., Charlton J. E. and Woolf C. J.), pp. 377-382. Elsevier, Amsterdam.
63. Wakisaka S., Kajander K. C. and Bennett G. J. (1991) Abnormal skin temperature and abnormal sympathetic vasomotor innervation in an experimental painful peripheral neuropathy. Puin 46, 2999314.
64. Wall P. D., Scadding J. W. and Tomkiewicz M. M. (1979) The production and prevention of experimental anesthesia dolorosa. Pain 6, 175-182.
65. Wallin G.. Torebjork E. and Hallin R. (1976) Preliminary observations on the pathophysiology of hyperalgesia in the causalgic pain syndrome. In Sensory Functions qf the Skin in Primates (ed. Zotterman Y.). pp. 489-499. Pergamon Press, Oxford.
66. Wiesenfield-Hallin 2. (1987) lntrathecal noradrenaline has a dose-dependent inhibitory or facilitatory effect on the flexion reflex in the rat. Actu physiol. stand. 130, 507-51 I.
67. Zimmerman M. (1983) Ethical guidelines for investigations of experimental pain in conscious animals. Pain 16, 109 -1 IO.
(Accepted 6 February 1995)
