- Email: [email protected]
Lumbar catheterization of the spinal subarachnoid space in the rat
Journal of Neuroscience Methods 65 ( 1996) 167- 172
Lumbar catheterization of the spinal subarachnoid space in the rat Rolf Vilhelm Starkson * , Anne Kjplrsvik, Ame Tjdsen, Kjell Hole Depurtment
of Physiology,
University
of Bergen.
Aarstudueien
19. N-5009
Bergen.
Norwy
Received 31 May 199.5;revised 13 November 1995; accepted 14 Novcmkr 1995
Abstract The method commonly used for catheterization of the lumbar subarachnoid space in the rat implies inserting the catheter through the atlanto-occipital (A-O) membrane and moving the catheter caudally along the spinal cord. The method is associated with a considerable morbidity. A method for direct catheterization of the lumbar subarachnoid space was therefore developed. Major surgery was avoided by using a catheter-through-needle technique. Of 32 rats, none died. There were no signs of neurological disturbances, and all animals gained weight as normal the first week after implantation. Data from rats catheterized by the A-O method were used for comparison. Of 40 animals. 2 died, 11 showed signs of neurological disturbances, and the mean weight was reduced during the first week after catheterization. The two groups of animals showed different behavioural responses to intrathecal injection of N-methyl-D-aspartate (NMDA, 0.1 -I .6 mM, 15 ~1) which is thought to stimulate afferent pathways mediating nociception. Animals with u lumbar catheter showed licking, biting and scratching behaviour in a dose-related manner for concentrations up to 1.6 mM. The animals with A-O catheters showed a maximum level of this behaviour already at 0.4 mM, while 0.5 mM induced convulsions. A possible explanation of this difference in response to NMDA could be a long-lasting pain state in the A-O group, caused by catheter-induced changes in the spinal cord and by the extensive surgery. It is concluded that the direct lumbar catheterization has several advantagescompared to the A-O method, decreasing the suffering of the animals, the neurological disturbances and the interference with nociceptive functions of the spinal cord. Keyw0rd.s: Catheterization; Subarachnoid space; Spinal cord; Lumbosacral region; N-methyl-D-aspartate; Pain; (Rat) _-
1. Introduction
after implantation (Kooistra et al.. 1986; Serpell et al., 1993).
Injecting drugs into the spinal fluid is a useful tool in the research on functions of the rat spinal cord. The method described by Yaksh and Rudy (19761, modified by LoPachin et al. ( 19811, has become the most widely used technique of catheterization of the rat spinal subarachnoid space. The method implies freeing neck muscles from the occipital crest and sliding the catheter through a slit in the exposed atlanto-occipital (A-0) membrane and caudally along the spinal cord. The method has some disadvantages. A few animals (3-5%) die during the first days after catheterization (personal observations); lo-30% of the animals show signs of neurological impairment (Schoeffler et al.. 1991; Igawa et al.. 1993; Kristensen et al., 1993) and the mean body weight is reduced during the first week
* Corresponding author. Tel.: (47) 55-ZO-63-38/55-20-63.90/55-2063-00; Fax: (47) 55.20-64- 10: E-mail: [email protected]. 0165-0270/96/$15.00 Q 1996 Elsvier Science B.V. All rights reserved SSDl 0165.0270(95100164-6
The impaired general condition and the neurological disturbances may seriously influence the modulation of pain. Therefore, an alternative intrathecal injection technique has been sought. Lumbar intrathecal injections in the rat by acute needle puncture has been described by others (Yaksh and Rudy, 1976; Papir-Kricheli et al., 1987; Long et al.. 1988; Mestre et al., 1993). Also, rhoracic and lumbar techniques of subarachnoid catheterization involving surgery of muscular tissue and bone, have been described by others (Dib, 1984; Bahar et al., 1984; Martin et al., 1984; Hughes and Barr, 1988; Gonzdlez-Darder et al., 1989). However, a method of lumbar, intrathecal catheterization in the rat, without the need of extensive surgery, has to our knowledge not previously been described. A method for direct lumbar catheterization of the spinal subarachnoid space in the rat is described. For comparison, a group of rats implanted with A-O catheters was included. Comparisons were made with regard to practical aspects of
168
R.V. Stpkson
et al./
Journal
of Neuroscience
catheterization, influence of the procedure on the animals’ general and neurological conditions, localization and spread of the injected material, and the influence of the technique of catheterization on the behaviour induced by intrathetally administered NMDA. A preliminary report of this study has previously been presented (Storkson et al., 1993).
2. Materials and methods 2.1. Drugs NMDA (Sigma Chemical, USA) was dissolved in 9 mg/ml NaCl. Lidocaine 20 mg/ml (Xylocain, Astra) was used undiluted. 99mTechnetium-labelled diethylene-triamine-pentaacetic acid ( 99mT~-DTPA) used for the scintigrams was delivered by Institute for Energy Technology, Kjeller, Norway. 2.2. Animals Male Sprague-Dawley rats (Mol:SPRD; Mollegaard, Denmark) were used, weighing 290-340 g at the time of catheterization. The animals were housed at 21-24°C three to a cage, with free access to food and water. They were kept on a 12-h light/dark cycle with lights on at 07:OO h. In total, 40 animals were catheterized by the A-O route and 32 by the lumbar route. After catheterization the animals were kept in individual cages. Each animal was used for one experiment only. 2.3. Ethical approval The experiments described in this article have been performed in accordance with Ethical Guidelines for Znvestigations of Experimental Pain in Conscious Animals (Zimmermann, 1983). The experiments were approved by the Norwegian State Commission for Experiments in Animals. 2.4. Lumbar catheterization The rat was anaesthetized by inhalation of 2 ~01% halothane in air, after a short period (30-90 s) of induction with 4 ~01%. A 5 mm longitudinal incision was made through the skin a few millimetres lateral to the midline and IO-15 mm caudal to a line between the two ventral iliac spines. A guide-cannula (Asik, 20 ga, 0.9 X 38 mm) was inserted through the incision and directed cranially and slightly medially. Sliding the cannula along the dorsal surface of vertebra L6, an increased resistance was felt when reaching the point where vertebra L5 overlaps L6. By raising the ventral iliac spines, the lumbar part of the spinal column was made kyphotic, and the needle could be
Methods
65 (1996)
167-172
advanced 3-5 mm in the narrow space between L5 and L6. The tip of the needle should now be exactly in the midline. The correct intrathecal localization was confirmed by a tail-flick or a paw retraction, by easy insertion of a catheter through the cannula, and occasionally by backflow of spinal fluid. The catheter was made of a polyethylene tube, outer diameter 0.6 mm (PE-10, Clay Adams). Submerged in water of 6O“C, one end of the catheter was reduced in diameter by stretching it to about 150% of the original length. The final catheter was cut to 14 cm, including 3 cm of the stretched part. The tapered end of the saline-filled catheter was inserted through the needle. To simplify the further advancement of the catheter, the direction of the needle was adjusted to arrange the needle as close as possible in line with the length axis of the spinal column. The catheter was inserted about 3 cm beyond the tip of the guide cannula to reach the level of the caudal ribs, which correspond to the lumbar enlargement of the spinal cord. The needle was then carefully removed, avoiding displacement of the catheter. A thin steel wire was inserted into the catheter. A bead was made on the catheter, 9 cm from the tip, by passing the flame of a lighter briefly underneath (Fig. la). The wire prevented occlusion of the melted catheter. With the wire still in place, the position of the catheter was easy to adjust if necessary. After the removal of the wire, 5 ~1 of saline were injected through the catheter to control for leakage. The catheter was finally tunnelled under the skin, appearing close to the base of the tail, 9 cm from the caudal ribs. The bead prevented the catheter from being dislocated. As the skin by the tail is firmly attached to the deeper tissues, the fixation close to the tail gave the catheter sufficient stability. Ten microlitres of saline were injected and the catheter was sealed by melting the end. The deadspace of the catheter was about 6 ~1. The method required about 10 min for insertion of the catheter and 20-40 min for recovery from anaesthesia. Behavioural studies could be performed immediately after complete recovery. After finishing the experiments, lidocaine (15 ~1, 20 mg/ml) followed by saline (7 ~1) was given through the catheters to verify an intraspinal location. An immediate motor paralysis of the hind part of the animal (within 15 s) lasting for 20-30 min indicated a correct intraspinal location. Probably because there was no surgical incision at the exit of the catheter, the rats did not pay attention to the polyethylene tubes the first days after insertion. But as the outer part of the catheter was placed within reach of the teeth of the animal, the percentage of undamaged catheters was acceptable only the first 5-7 days after catheterization. If the catheter is needed for a longer period of time, the following modifications are suggested. The catheter should be cut to a length of 28 cm, inserted intrathecally as
lheca&dribs rrR.V. Sr#rk.wn
(a)
The wntrd
rt al. /Journal
of‘ Neuroscience
iliac spires
Method
65 f 1996)
The caudal ribs
/-,+->
The incision
assessment
The animals were observed with respect to dysfunction of the limbs whiie walking. If one of the limbs showed reduced ability to bear weight but the movements otherwise appeared normal. the term ‘minor neurological symptom’ was used. If two or more limbs were affected there was usually a more severe paresis or a paralysis of one of the limbs and the term ‘major neurological impairment’ was used. 2.7. NMDA-induced
H
6Crn
+I
Fig. I. a: The lumbar catheter tunnelled to the base of the tail. The external part of the catheter is drawn with a solid line, the internal part with a dashed line. The bead, represented by the filled circle, is located subcutaneously and prevents dislocation of the catheter. b: The lumbar catheter tunnelled to the occipital region. The external part of the catheter is drawn with a solid line, the internal part with a dashed line. One bead is located subcutaneously and prevents dislocation of the tip of the catheter, another bead is located externally and prevents the external part of the catheter from sliding back under the skin.
described, and fixed through a layer of superficial muscle rather than through the skin. Then the catheter should be tunnelled rostrally to allow appearance through the skin in the occipital region (Fig. lb). Finally, a bead should be made 3 cm from the end of the catheter to prevent it from disappearing to a subcutaneous location. The deadspace of the catheter will be increased to about 12 ~1. A reduction of the deadspace with a few microlitres is possible by stretching the part of the catheter located between the two beads. 2.5. Atlanto-occipital
catheterization
The rat was anaesthetized by subcutaneous injection of a mixture of fentanyl 0.1 mg/kg and fluanison 5 mg/kg (Hypnorm, Janssen) with the addition of midazolam 2.5 mg/kg (Dormicum, Roche). Halothane was not used as
I h9
the anaesthetic for this method of catheterization because of the more extensive surgery. The preparation and insertion of the catheter through the A-O membrane was done as previously described (Yaksh and Rudy, 1976; LoPachin et al., 1981). For the catheter to reach the lumbar enlargement, an intraspinal length of 8.5 cm was used. The deadspace of the catheter was about 4.5 ~1. The A-O method required about 20-30 min for insertion of the catheter and 5-7 days for recovery. To verify an intraspinal location, lidocaine was used as described above.
2.6. Neurological
W
167- 172
behaviour
The NMDA-receptor, a subtype of the excitatory amino acid receptors, is involved in nociception, neural plasticity and motor function. NMDA injected intrathecally induces pain-related behaviour, with licking, scratching and biting of the affected part of the body, both in the mouse (Aanonsen and Wilcox, 1986, Aanonsen and Wilcox, 1987; DeLander and Wahl, 1988; Mjellem-Joly et al., 1991) and in the rat (Raigorodsky and Urea, 1987: Mjellem et al., 1992; Bjbrkman et al., 1994). In our experiments with implanted catheters, intrathecal administration of NMDA (15 ~1, O.l- 1.6 ntM1 followed by saline (7 ~1) induced pain-related behaviour lasting for l-6 min. The time spent licking, scratching and biting the caudal part of the body was recorded. Licking and biting the hind paws, and scratching the hind paws against the cage or against the skin, was also recorded as pain-related behaviour. The NMDA injections were performed after recovery from anaesthesia, as judged by the return of a pattern of normal motor behaviour. Thus, rhe lumbar group received their NMDA injections 2-4 h after catheterization, while the NMDA injections were done 667 days after the catheterization in the A-O group. 2.8. Scintigraphic investigation
Fifteen microlitres of the water-soluble 9YmTc-DTPA (37 kBq/pl) were injected intrathecally in four anaesthetized rats. The initial localization of the injected material and the subsequent spread were followed for 25 min by dynamic acquisition with a gamma-camera. Markers were introduced into some of the scintigrams to enable reconstruction of the outline of the animals.
170
R.V. Stwkson
et al./Journal
of Neuroscience
Methods
65 (1996)
167-172
2.9. Statistics
The data on pain-related behaviour did not show a normal distribution. However, according to the central limit theorem, the sample mean for a sufficient large sample size of such data will follow a normal distribution (Rosner, 1986). Based on this approximation, a parametric statistical method, analysis of variance (ANOVA) was used for evaluation of dose-response results. Fisher’s exact test was used when looking for differences in the number of complications in the two groups of catheterized animals. Statistical significance was accepted at the 5% level.
3. Resdts After the implantation of lumbar catheters none of the 32 animals died, there were no detectable signs of neurological impairment, and the animals showed a normal increase in body weight (Table 1). In contrast, 38 of 40 animals survived the A-O catheterization, 7 animals had major neurological problems (flaccid paresis/paralysis of two or more limbs) and 4 had minor neurological symptoms (affecting only one limb). The mean body weight was reduced during the first week after A-O catheter implantation (Table 1). The difference in survival did not reach a significant level, but the difference in neurological complications was statistically significant (P < 0.005, Fisher’s exact test). The scintigrams showing radioactivity after injection did not reveal any differences between the two methods of catheterization, neither for the initial localization of the injected material, nor for the distribution after 25 min (Fig.
In the lumbar catheterized animals, the threshold for inducing pain-related behaviour was lower than 0.1 mM. There was a statistically significant dose-related effect of increasing concentrations of NMDA up to 1.6 mM (Fig. 3) as judged by ANOVA (F4,z5 = 11.02, P < 0.001). No motor disturbance was seen. In the A-O group, the threshold for inducing pain-related behaviour was 0.1-0.2 mM NMDA. The highest level of pain-related behaviour was reached already at 0.4 mM of NMDA (data not shown). Increasing the NMDA ScKt-
2).
The A-O catheters could be used for several weeks. The lumbar catheters had to be used within 5-7 days when tunnelled to the base of the tail. Two of these catheters (6%) were bitten off on the third postoperative day and four catheters were lost during the following 2 days, leaving 26 of 32 catheters (81%) in place after 5 days. After 7 days 21 catheters (65%) were functioning. hmathecal injection of NMDA-induced pain-related behaviour in a dose-dependent manner.
Table I Complications after spinal subarachnoid catheterization A-O method Lumbar method Deaths during the first 2 days after catheterization (n of animals) Detectable neurological complications (n of animals) Weight gain (g) 8 days after catheterization ( f SD)
Fig. 2. Scintigram 25 min after intrathecal injection of 99mT~-DTPA. The A-O catheter is shown to the left, the lumbar catheter to the right. The recorded activity is shown in black. The outline of the animals, the caudal ribs and the iliac spines are added in grey, based on markers in the scintigrams.
2 (of 40)
0 (of 32)
11 (of 40)
0 (of 32)
-18ff22)
+47(f16)
n=4
250 ZT 5 ki 200 z 5 .Q 2 1502 00 ; IOO.c a” 50 -
J-/l n=6
n=8 n=6
n=a i-.i
oI
I
I
I
1
0.1
0.2
0.4
0.8
1.6
NMDA concentration
(mM)
Fig., 3. Pam-related behaviour after mtrathecal injection of NMDA (I5 ~1) through lumbar catheters. The amount of time the animals spent licking, scratching or biting the hind part of the body during a 5 min observation period. Mean f SEM.
R.V. Sterkson
et al./Journal
of Neuroscience
concentration to 0.5 mM induced convulsions. For ethical reasons, only two animals with A-O catheters were given 0.5 mM NMDA.
4. Discussion
4.1. kxuiization
of the injected solution
When the radioactive 99mTc-DTPA was injected through the catheters, the A-O and the lumbar method appeared to result in similar localization and spread of the injected solution. Also, injection of NMDA through the two different types of catheters induced pain-related behaviour affecting the same regions of the animals. Thus the two methods seem to be comparable regarding drug deliverance. 4.2. Functional duration of the catheters For reasons discussed below, 32% of the originally implanted A-O catheters were not available for experiments. The remaining catheters (68%) functioned for several weeks. The lumbar catheters, on the other hand, had to be used within 5-7 days. The animals bit off the outer part of the lumbar catheters after some days, leaving 65% of the catheters in place after 7 days. For experiments requiring catheterization for more than 1 week, the lumbar catheters should be tunnelled to the occipital region as described. When connected to subcutaneous minipumps for intrathecal administration of drugs, there would be no difference in duration of the two types of catheters. The lumbar catheters should then be fixed through a layer of superficial muscle rather than through the skin. For details on connection to minipumps see Wiesenfeld and Gustafsson (1982) and Loomis et al. (1987). 4.3. Practical aspects of catheterization by the two meth0d.S
The lumbar route of catheterization had several advantages compared to the A-O route. (a) The catheterization was performed in about half the time. (b) Recovery was reduced from several days to less than 1 hour. The different anaesthetic techniques may have contributed to this reduction. cc) Instead of a decrease in weight, the animals showed a normal increase in weight after catheterization. 4.4. A-O catheters induce neurological disturbances
and functional
Neurological problems, affecting one or more limbs, appeared in 27% of the A-O catheterized animals. No neurological limb affection was observed after the lumbar catheterization. It cannot be excluded, however, that minor
Metho&
65 (1996)
167-172
171
disturbances, too small to be detected by observation of the animals, might exist also after lumbar catheterization. When it is essential to avoid even small disturbances, the lumbar catheterization method may be refined by using thinner catheters and thinner, short bevelled needles. The A-O catheterization results in an inhomogeneous group of rats to study. The neurological impairments that occur may be a result of the insertion of the catheter producing oedema, bleeding or mechanical damage affecting the medulla oblongata, the spinal cord or the nerve roots. Loss of body weight the first postoperative week may be due to the physiological stress response following surgery, neurological problems following the insertion of the catheter, and difficulties in eating because of the cervical wound. The death of some of the animals during the first few days, may be caused by one or more of the problems mentioned above. The A-O catheter has been found to induce changes in the behavioural response to neurogenic pain (Serpell et al., 1993) and changes in the pharmacological responses to substances active at the K and S opioid receptors (Long et al., 1988). We propose that the A-O method of catheterization leads to different degrees of long-lasting pain. The contributing factors may be the spinal cord lesions, the substantial neck wound and a possible chronic, mechanical irritation of the spinal cord both immediately below the A-O membrane and at the tip of the catheter, due to movements of the catheter induced by movements of the head (Bahar et al., 1984). The observed inhomogeneity regarding body weight and neurological symptoms and the possible pain problems in the A-O catheterized animals may interfere with the mechanisms of nociception in an unpredictable way. It seems likely that the use of a lumbar catheter will reduce this kind of interference, and thus be a more suitable method when studying spinal mechanisms of nociception. 4.5. Differences in the behavioural response to inrrathecal NMDA The present investigation of NMDA-induced pain-related behaviour supports that rats with A-O catheters and those with lumbar catheters are functionally different. The rats with lumbar catheters showed a usual dose-response to intrathecally injected NMDA for a wide range of concentrations (O.l- 1.6 mM). The rats with A-O catheters responded to a more narrow range of concentrations, with the threshold at 0.1-0.2 mM and the maximum response at 0.4 mM NMDA. Convulsions were induced when the concentration of NMDA was further increased (0.5 mM). Similar results have been obtained by others using the A-O catheterization technique for intrathecal injection of NMDA (Raigorodsky and Urea, 1987). The reason for the different response to NMDA is not known. One possibility is that the A-O catheterization. inducing different degrees of
172
R.V. Stwkson
et al./Journal
ofNeuroscience
long-lasting pain, causes a change in the functional state of the spinal cord with an increase in the sensitivity of the NMDA receptor to higher concentrations of NMDA. For the lumbar catheterization only a light halothane anaesthesia was used. For the A-L) catheterization, a deeper and more long-lasting anaesthesia had to be used due to the more extensive surgery. Since the animals were allowed complete recovery before testing, it is unlikely that the difference in anaesthesia was of importance for the difference in response to NMDA. However, the shorter time interval between catheterization and the NMDA injection in the lumbar group, may have contributed to the difference in NMDA response between the two groups. 4.6. Conclusion The lumbar catheterization is a reliable and time-saving method that minimizes the neurological disturbances and reduces the suffering of the animals. The advantages of the technique are important from an ethical point of view, and they may also be important by reducing unpredictable interference when studying pain mechanisms.
Acknowledgements The authors thank M. Foiling, M.D., for preparing the scintigrams. We also thank L.J. Rygh for practical assistance and professor T.I. Kanui for useful suggestions. This study was supported in part by the Norwegian Research Council and by the European Commission.
References Aanonsen, L.M. and Wilcox, G.L. ( 1986) Phencyclidine selectively blocks a spinal action of N-methyl-D-aspartate in mice, Neurosci. Lett., 67: 191-197. Aanonsen, L.M. and Wilcox, G.L. (1987) Nociceptive action of excitatory amino acids in the mouse: effects of spinally administered opioids, phencyclidine and sigma agonists, J. Pharmacol. Exp. Ther., 243: 9-19. Bahar, M., Rosen, M. and Vickers, M.D. (1984) Chronic cannulation of the intradural or extradural space in the rat, Br. J. Anaesth., 56: 405-410. Bjorkman, R., Hallman, K.M., Hedner, J., Hedner, T. and Henning, M. (1994) Acetaminophen blocks spinal hyperalgesia induced by NMDA and substance P, Pain, 57: 259-264. DeLander, G.E. and Wahl, J.J. (1988) Behavior induced by putative nociceptive neurotransmitters is inhibited by adenosine or adenosine analogs coadministered intrathecally, J. Pharmacol. Exp. Ther., 246: 565-570. Dib, B. (1984) Intrathecal chronic catheterization in the rat, Pharmacol. B&hem. Behav., 20: 45-48. Gom&ez-Darder, J.M., G6mezCBrdenas, E. and Gil-Salt, J.L. (1989) Microsurgical catheterization of the intrathecal space in the rat for chronic infusion of drugs, Rev. Esp. Anestesiol. Reanim., 36: 153156. Hughes, H.E. and Barr, G.A. (1988) Analgesic effects of intrathecally
Methods
65 (19961167-172
applied noradrenergic compounds in the developing rat: differences due to thermal vs. mechanical nociception, Brain Res. Dev. Brain Res.. 41: 109-120. Igawa, Y., Andersson, K.E., Post, C., Uvelius, B. and Mattiasson, A. ( 1993) A rat model for investigation of spinal mechanisms in detrusor instability associated with infravesical outflow obstruction, Ural. Res., 21: 239-244. Kooistra, K.L., Rodriguez, M., Powis, G., Yaksh, T.L., Harty, G.J., Hilton, J.F. and Laws, E.R., Jr. (1986) Development of experimental models for meningeal neoplasia using intrathecal injection of 9L ghosarcoma and Walker 256 carcinosarcoma in the rat, Cancer Res., 45: 3 17-323. Kristensen, J.D., Post, C., Gordh, T., Jr. and Svensson, B.A. (1993) Spinal cord morphology and antinociception after chronic intrathecal administration of excitatory amino acid antagonists in the rat, Pain, 54: 309-316. Long, J.B., Mobley, W.C. and Holaday, J.W. (1988) Neurological dysfunction after intrathecal injection of Dynorphin A(1 - 13) in the rat. 1. Injection procedures modify pharmacological responses, J. Pharmacol. Exp. ‘lher., 246: 1158-I 166. Loomis, C.W., Milne, B. and Cetvenko, F.W. (1987) Determination of cross tolerance in rat spinal cord using intrathecal infusion via sequential mini-osmotic pumps, Pharmacol. B&hem. Behav., 26: 13l- 139. LoPachin, R.M., Rudy, T.A. and Yaksh, T.L. (1981) An improved method for chronic catheterization of the rat spinal subarachnoid space, Physiol. Behav., 27: 559-561. Martin, H., Kocher, L. and Chery-Croze, S. (1984) Chronic lumbar intrathecal catheterization in the rat with reduced-length spinal compression, Physiol. Behav., 33: 159-161. Mestm, C., Pelissier, T., Wilcox, G.L., Eschaher, A. and Lavarenne, J. (1993) Intrathecal administration of drugs in unanesthetized rats: technical and pharmacological aspects, IASP Publ., Congress Abstracts, Abstr. 217. Mjellem, N., Lund, A., Eide, P.K., Storkson, R. and Tjslsen, A. (1992) The role of 5-HTla and 5-HTlb receptors in spinal nociceptive transmission and in the modulation of NMDA induced behaviour, NeuroReport, 3: 1061-1064. Mjehem-Joly, N., Lund, A., Berge, O.G. and Hole. K. (1991) Potentiation of a behavioural response in mice by spinal coadministration of substance P and excitatory amino acid agonists, Neurosci. Lett., 133: 121-124. Papir-Kricheli, D., Frey, J., Laufer, R., Gilon, C., Chorev, M., Selinger, Z. and Devor, M. (1987) BehaviouraI effects of receptor-specific substance P agonists, Pain, 3 1: 263-276. Raigorodsky, G. and Urea, G. (1987) lntrathecal N-methyl-D-aspartate (NMDA) activates both nociceptive and antinociceptive systems, Brain Res., 422: 158- 162. Rosner, B. (1986) Fundamentals of Biostatistics, 2nd edn., PWS Publ., Boston, MA, pp. 151-153. Schoeffler, P., Auroy, P., Bazin, J.E., Taxi, J. and Woda, A. (1991) Subarachnoid midazolam: histologic study in rats and report of its effect on chronic pain in humans, Reg. Anesth., 16: 329-332. Serpell, M.G., DeLeo, J.A., Coombs, D.W., Colbum. R.W., Twitchell, B.B., Willenbring, S. and Fromm, C. (1993) Intrathecal cadteterization alone reduces autotomy after sciatic cyroneurolysis in the rat, Life Sci., 53: 1887-1892. Storkson. R.V., Kjwsvik, A., Tjdsen, A. and Hole, K. (1993) Catheterization of the lumbar spinal subarachnoid space in the rat, IASP Publ., Congress Abstracts, Abstr. 218. Wiesenfeld, Z. and Gustafsson, L.L. (1982) Continuous intrathecal administration of morphine via an osmotic minipump in the rat, Brain Res., 247: 195-197. Yaksh, T.L. and Rudy, T.A. (1976) Chronic catheterization of the spinal subarachnoid space, Physiol. Behav., 17: 103 I- 1036. Zimmertitann, M. (1983) Ethical guidelines for investigations of expetimental pain in conscious animals (editorial). Pain, 16: 109-10.
Lumbar catheterization of the spinal subarachnoid space in the rat Rolf Vilhelm Starkson * , Anne Kjplrsvik, Ame Tjdsen, Kjell Hole Depurtment
of Physiology,
University
of Bergen.
Aarstudueien
19. N-5009
Bergen.
Norwy
Received 31 May 199.5;revised 13 November 1995; accepted 14 Novcmkr 1995
Abstract The method commonly used for catheterization of the lumbar subarachnoid space in the rat implies inserting the catheter through the atlanto-occipital (A-O) membrane and moving the catheter caudally along the spinal cord. The method is associated with a considerable morbidity. A method for direct catheterization of the lumbar subarachnoid space was therefore developed. Major surgery was avoided by using a catheter-through-needle technique. Of 32 rats, none died. There were no signs of neurological disturbances, and all animals gained weight as normal the first week after implantation. Data from rats catheterized by the A-O method were used for comparison. Of 40 animals. 2 died, 11 showed signs of neurological disturbances, and the mean weight was reduced during the first week after catheterization. The two groups of animals showed different behavioural responses to intrathecal injection of N-methyl-D-aspartate (NMDA, 0.1 -I .6 mM, 15 ~1) which is thought to stimulate afferent pathways mediating nociception. Animals with u lumbar catheter showed licking, biting and scratching behaviour in a dose-related manner for concentrations up to 1.6 mM. The animals with A-O catheters showed a maximum level of this behaviour already at 0.4 mM, while 0.5 mM induced convulsions. A possible explanation of this difference in response to NMDA could be a long-lasting pain state in the A-O group, caused by catheter-induced changes in the spinal cord and by the extensive surgery. It is concluded that the direct lumbar catheterization has several advantagescompared to the A-O method, decreasing the suffering of the animals, the neurological disturbances and the interference with nociceptive functions of the spinal cord. Keyw0rd.s: Catheterization; Subarachnoid space; Spinal cord; Lumbosacral region; N-methyl-D-aspartate; Pain; (Rat) _-
1. Introduction
after implantation (Kooistra et al.. 1986; Serpell et al., 1993).
Injecting drugs into the spinal fluid is a useful tool in the research on functions of the rat spinal cord. The method described by Yaksh and Rudy (19761, modified by LoPachin et al. ( 19811, has become the most widely used technique of catheterization of the rat spinal subarachnoid space. The method implies freeing neck muscles from the occipital crest and sliding the catheter through a slit in the exposed atlanto-occipital (A-0) membrane and caudally along the spinal cord. The method has some disadvantages. A few animals (3-5%) die during the first days after catheterization (personal observations); lo-30% of the animals show signs of neurological impairment (Schoeffler et al.. 1991; Igawa et al.. 1993; Kristensen et al., 1993) and the mean body weight is reduced during the first week
* Corresponding author. Tel.: (47) 55-ZO-63-38/55-20-63.90/55-2063-00; Fax: (47) 55.20-64- 10: E-mail: [email protected]. 0165-0270/96/$15.00 Q 1996 Elsvier Science B.V. All rights reserved SSDl 0165.0270(95100164-6
The impaired general condition and the neurological disturbances may seriously influence the modulation of pain. Therefore, an alternative intrathecal injection technique has been sought. Lumbar intrathecal injections in the rat by acute needle puncture has been described by others (Yaksh and Rudy, 1976; Papir-Kricheli et al., 1987; Long et al.. 1988; Mestre et al., 1993). Also, rhoracic and lumbar techniques of subarachnoid catheterization involving surgery of muscular tissue and bone, have been described by others (Dib, 1984; Bahar et al., 1984; Martin et al., 1984; Hughes and Barr, 1988; Gonzdlez-Darder et al., 1989). However, a method of lumbar, intrathecal catheterization in the rat, without the need of extensive surgery, has to our knowledge not previously been described. A method for direct lumbar catheterization of the spinal subarachnoid space in the rat is described. For comparison, a group of rats implanted with A-O catheters was included. Comparisons were made with regard to practical aspects of
168
R.V. Stpkson
et al./
Journal
of Neuroscience
catheterization, influence of the procedure on the animals’ general and neurological conditions, localization and spread of the injected material, and the influence of the technique of catheterization on the behaviour induced by intrathetally administered NMDA. A preliminary report of this study has previously been presented (Storkson et al., 1993).
2. Materials and methods 2.1. Drugs NMDA (Sigma Chemical, USA) was dissolved in 9 mg/ml NaCl. Lidocaine 20 mg/ml (Xylocain, Astra) was used undiluted. 99mTechnetium-labelled diethylene-triamine-pentaacetic acid ( 99mT~-DTPA) used for the scintigrams was delivered by Institute for Energy Technology, Kjeller, Norway. 2.2. Animals Male Sprague-Dawley rats (Mol:SPRD; Mollegaard, Denmark) were used, weighing 290-340 g at the time of catheterization. The animals were housed at 21-24°C three to a cage, with free access to food and water. They were kept on a 12-h light/dark cycle with lights on at 07:OO h. In total, 40 animals were catheterized by the A-O route and 32 by the lumbar route. After catheterization the animals were kept in individual cages. Each animal was used for one experiment only. 2.3. Ethical approval The experiments described in this article have been performed in accordance with Ethical Guidelines for Znvestigations of Experimental Pain in Conscious Animals (Zimmermann, 1983). The experiments were approved by the Norwegian State Commission for Experiments in Animals. 2.4. Lumbar catheterization The rat was anaesthetized by inhalation of 2 ~01% halothane in air, after a short period (30-90 s) of induction with 4 ~01%. A 5 mm longitudinal incision was made through the skin a few millimetres lateral to the midline and IO-15 mm caudal to a line between the two ventral iliac spines. A guide-cannula (Asik, 20 ga, 0.9 X 38 mm) was inserted through the incision and directed cranially and slightly medially. Sliding the cannula along the dorsal surface of vertebra L6, an increased resistance was felt when reaching the point where vertebra L5 overlaps L6. By raising the ventral iliac spines, the lumbar part of the spinal column was made kyphotic, and the needle could be
Methods
65 (1996)
167-172
advanced 3-5 mm in the narrow space between L5 and L6. The tip of the needle should now be exactly in the midline. The correct intrathecal localization was confirmed by a tail-flick or a paw retraction, by easy insertion of a catheter through the cannula, and occasionally by backflow of spinal fluid. The catheter was made of a polyethylene tube, outer diameter 0.6 mm (PE-10, Clay Adams). Submerged in water of 6O“C, one end of the catheter was reduced in diameter by stretching it to about 150% of the original length. The final catheter was cut to 14 cm, including 3 cm of the stretched part. The tapered end of the saline-filled catheter was inserted through the needle. To simplify the further advancement of the catheter, the direction of the needle was adjusted to arrange the needle as close as possible in line with the length axis of the spinal column. The catheter was inserted about 3 cm beyond the tip of the guide cannula to reach the level of the caudal ribs, which correspond to the lumbar enlargement of the spinal cord. The needle was then carefully removed, avoiding displacement of the catheter. A thin steel wire was inserted into the catheter. A bead was made on the catheter, 9 cm from the tip, by passing the flame of a lighter briefly underneath (Fig. la). The wire prevented occlusion of the melted catheter. With the wire still in place, the position of the catheter was easy to adjust if necessary. After the removal of the wire, 5 ~1 of saline were injected through the catheter to control for leakage. The catheter was finally tunnelled under the skin, appearing close to the base of the tail, 9 cm from the caudal ribs. The bead prevented the catheter from being dislocated. As the skin by the tail is firmly attached to the deeper tissues, the fixation close to the tail gave the catheter sufficient stability. Ten microlitres of saline were injected and the catheter was sealed by melting the end. The deadspace of the catheter was about 6 ~1. The method required about 10 min for insertion of the catheter and 20-40 min for recovery from anaesthesia. Behavioural studies could be performed immediately after complete recovery. After finishing the experiments, lidocaine (15 ~1, 20 mg/ml) followed by saline (7 ~1) was given through the catheters to verify an intraspinal location. An immediate motor paralysis of the hind part of the animal (within 15 s) lasting for 20-30 min indicated a correct intraspinal location. Probably because there was no surgical incision at the exit of the catheter, the rats did not pay attention to the polyethylene tubes the first days after insertion. But as the outer part of the catheter was placed within reach of the teeth of the animal, the percentage of undamaged catheters was acceptable only the first 5-7 days after catheterization. If the catheter is needed for a longer period of time, the following modifications are suggested. The catheter should be cut to a length of 28 cm, inserted intrathecally as
lheca&dribs rrR.V. Sr#rk.wn
(a)
The wntrd
rt al. /Journal
of‘ Neuroscience
iliac spires
Method
65 f 1996)
The caudal ribs
/-,+->
The incision
assessment
The animals were observed with respect to dysfunction of the limbs whiie walking. If one of the limbs showed reduced ability to bear weight but the movements otherwise appeared normal. the term ‘minor neurological symptom’ was used. If two or more limbs were affected there was usually a more severe paresis or a paralysis of one of the limbs and the term ‘major neurological impairment’ was used. 2.7. NMDA-induced
H
6Crn
+I
Fig. I. a: The lumbar catheter tunnelled to the base of the tail. The external part of the catheter is drawn with a solid line, the internal part with a dashed line. The bead, represented by the filled circle, is located subcutaneously and prevents dislocation of the catheter. b: The lumbar catheter tunnelled to the occipital region. The external part of the catheter is drawn with a solid line, the internal part with a dashed line. One bead is located subcutaneously and prevents dislocation of the tip of the catheter, another bead is located externally and prevents the external part of the catheter from sliding back under the skin.
described, and fixed through a layer of superficial muscle rather than through the skin. Then the catheter should be tunnelled rostrally to allow appearance through the skin in the occipital region (Fig. lb). Finally, a bead should be made 3 cm from the end of the catheter to prevent it from disappearing to a subcutaneous location. The deadspace of the catheter will be increased to about 12 ~1. A reduction of the deadspace with a few microlitres is possible by stretching the part of the catheter located between the two beads. 2.5. Atlanto-occipital
catheterization
The rat was anaesthetized by subcutaneous injection of a mixture of fentanyl 0.1 mg/kg and fluanison 5 mg/kg (Hypnorm, Janssen) with the addition of midazolam 2.5 mg/kg (Dormicum, Roche). Halothane was not used as
I h9
the anaesthetic for this method of catheterization because of the more extensive surgery. The preparation and insertion of the catheter through the A-O membrane was done as previously described (Yaksh and Rudy, 1976; LoPachin et al., 1981). For the catheter to reach the lumbar enlargement, an intraspinal length of 8.5 cm was used. The deadspace of the catheter was about 4.5 ~1. The A-O method required about 20-30 min for insertion of the catheter and 5-7 days for recovery. To verify an intraspinal location, lidocaine was used as described above.
2.6. Neurological
W
167- 172
behaviour
The NMDA-receptor, a subtype of the excitatory amino acid receptors, is involved in nociception, neural plasticity and motor function. NMDA injected intrathecally induces pain-related behaviour, with licking, scratching and biting of the affected part of the body, both in the mouse (Aanonsen and Wilcox, 1986, Aanonsen and Wilcox, 1987; DeLander and Wahl, 1988; Mjellem-Joly et al., 1991) and in the rat (Raigorodsky and Urea, 1987: Mjellem et al., 1992; Bjbrkman et al., 1994). In our experiments with implanted catheters, intrathecal administration of NMDA (15 ~1, O.l- 1.6 ntM1 followed by saline (7 ~1) induced pain-related behaviour lasting for l-6 min. The time spent licking, scratching and biting the caudal part of the body was recorded. Licking and biting the hind paws, and scratching the hind paws against the cage or against the skin, was also recorded as pain-related behaviour. The NMDA injections were performed after recovery from anaesthesia, as judged by the return of a pattern of normal motor behaviour. Thus, rhe lumbar group received their NMDA injections 2-4 h after catheterization, while the NMDA injections were done 667 days after the catheterization in the A-O group. 2.8. Scintigraphic investigation
Fifteen microlitres of the water-soluble 9YmTc-DTPA (37 kBq/pl) were injected intrathecally in four anaesthetized rats. The initial localization of the injected material and the subsequent spread were followed for 25 min by dynamic acquisition with a gamma-camera. Markers were introduced into some of the scintigrams to enable reconstruction of the outline of the animals.
170
R.V. Stwkson
et al./Journal
of Neuroscience
Methods
65 (1996)
167-172
2.9. Statistics
The data on pain-related behaviour did not show a normal distribution. However, according to the central limit theorem, the sample mean for a sufficient large sample size of such data will follow a normal distribution (Rosner, 1986). Based on this approximation, a parametric statistical method, analysis of variance (ANOVA) was used for evaluation of dose-response results. Fisher’s exact test was used when looking for differences in the number of complications in the two groups of catheterized animals. Statistical significance was accepted at the 5% level.
3. Resdts After the implantation of lumbar catheters none of the 32 animals died, there were no detectable signs of neurological impairment, and the animals showed a normal increase in body weight (Table 1). In contrast, 38 of 40 animals survived the A-O catheterization, 7 animals had major neurological problems (flaccid paresis/paralysis of two or more limbs) and 4 had minor neurological symptoms (affecting only one limb). The mean body weight was reduced during the first week after A-O catheter implantation (Table 1). The difference in survival did not reach a significant level, but the difference in neurological complications was statistically significant (P < 0.005, Fisher’s exact test). The scintigrams showing radioactivity after injection did not reveal any differences between the two methods of catheterization, neither for the initial localization of the injected material, nor for the distribution after 25 min (Fig.
In the lumbar catheterized animals, the threshold for inducing pain-related behaviour was lower than 0.1 mM. There was a statistically significant dose-related effect of increasing concentrations of NMDA up to 1.6 mM (Fig. 3) as judged by ANOVA (F4,z5 = 11.02, P < 0.001). No motor disturbance was seen. In the A-O group, the threshold for inducing pain-related behaviour was 0.1-0.2 mM NMDA. The highest level of pain-related behaviour was reached already at 0.4 mM of NMDA (data not shown). Increasing the NMDA ScKt-
2).
The A-O catheters could be used for several weeks. The lumbar catheters had to be used within 5-7 days when tunnelled to the base of the tail. Two of these catheters (6%) were bitten off on the third postoperative day and four catheters were lost during the following 2 days, leaving 26 of 32 catheters (81%) in place after 5 days. After 7 days 21 catheters (65%) were functioning. hmathecal injection of NMDA-induced pain-related behaviour in a dose-dependent manner.
Table I Complications after spinal subarachnoid catheterization A-O method Lumbar method Deaths during the first 2 days after catheterization (n of animals) Detectable neurological complications (n of animals) Weight gain (g) 8 days after catheterization ( f SD)
Fig. 2. Scintigram 25 min after intrathecal injection of 99mT~-DTPA. The A-O catheter is shown to the left, the lumbar catheter to the right. The recorded activity is shown in black. The outline of the animals, the caudal ribs and the iliac spines are added in grey, based on markers in the scintigrams.
2 (of 40)
0 (of 32)
11 (of 40)
0 (of 32)
-18ff22)
+47(f16)
n=4
250 ZT 5 ki 200 z 5 .Q 2 1502 00 ; IOO.c a” 50 -
J-/l n=6
n=8 n=6
n=a i-.i
oI
I
I
I
1
0.1
0.2
0.4
0.8
1.6
NMDA concentration
(mM)
Fig., 3. Pam-related behaviour after mtrathecal injection of NMDA (I5 ~1) through lumbar catheters. The amount of time the animals spent licking, scratching or biting the hind part of the body during a 5 min observation period. Mean f SEM.
R.V. Sterkson
et al./Journal
of Neuroscience
concentration to 0.5 mM induced convulsions. For ethical reasons, only two animals with A-O catheters were given 0.5 mM NMDA.
4. Discussion
4.1. kxuiization
of the injected solution
When the radioactive 99mTc-DTPA was injected through the catheters, the A-O and the lumbar method appeared to result in similar localization and spread of the injected solution. Also, injection of NMDA through the two different types of catheters induced pain-related behaviour affecting the same regions of the animals. Thus the two methods seem to be comparable regarding drug deliverance. 4.2. Functional duration of the catheters For reasons discussed below, 32% of the originally implanted A-O catheters were not available for experiments. The remaining catheters (68%) functioned for several weeks. The lumbar catheters, on the other hand, had to be used within 5-7 days. The animals bit off the outer part of the lumbar catheters after some days, leaving 65% of the catheters in place after 7 days. For experiments requiring catheterization for more than 1 week, the lumbar catheters should be tunnelled to the occipital region as described. When connected to subcutaneous minipumps for intrathecal administration of drugs, there would be no difference in duration of the two types of catheters. The lumbar catheters should then be fixed through a layer of superficial muscle rather than through the skin. For details on connection to minipumps see Wiesenfeld and Gustafsson (1982) and Loomis et al. (1987). 4.3. Practical aspects of catheterization by the two meth0d.S
The lumbar route of catheterization had several advantages compared to the A-O route. (a) The catheterization was performed in about half the time. (b) Recovery was reduced from several days to less than 1 hour. The different anaesthetic techniques may have contributed to this reduction. cc) Instead of a decrease in weight, the animals showed a normal increase in weight after catheterization. 4.4. A-O catheters induce neurological disturbances
and functional
Neurological problems, affecting one or more limbs, appeared in 27% of the A-O catheterized animals. No neurological limb affection was observed after the lumbar catheterization. It cannot be excluded, however, that minor
Metho&
65 (1996)
167-172
171
disturbances, too small to be detected by observation of the animals, might exist also after lumbar catheterization. When it is essential to avoid even small disturbances, the lumbar catheterization method may be refined by using thinner catheters and thinner, short bevelled needles. The A-O catheterization results in an inhomogeneous group of rats to study. The neurological impairments that occur may be a result of the insertion of the catheter producing oedema, bleeding or mechanical damage affecting the medulla oblongata, the spinal cord or the nerve roots. Loss of body weight the first postoperative week may be due to the physiological stress response following surgery, neurological problems following the insertion of the catheter, and difficulties in eating because of the cervical wound. The death of some of the animals during the first few days, may be caused by one or more of the problems mentioned above. The A-O catheter has been found to induce changes in the behavioural response to neurogenic pain (Serpell et al., 1993) and changes in the pharmacological responses to substances active at the K and S opioid receptors (Long et al., 1988). We propose that the A-O method of catheterization leads to different degrees of long-lasting pain. The contributing factors may be the spinal cord lesions, the substantial neck wound and a possible chronic, mechanical irritation of the spinal cord both immediately below the A-O membrane and at the tip of the catheter, due to movements of the catheter induced by movements of the head (Bahar et al., 1984). The observed inhomogeneity regarding body weight and neurological symptoms and the possible pain problems in the A-O catheterized animals may interfere with the mechanisms of nociception in an unpredictable way. It seems likely that the use of a lumbar catheter will reduce this kind of interference, and thus be a more suitable method when studying spinal mechanisms of nociception. 4.5. Differences in the behavioural response to inrrathecal NMDA The present investigation of NMDA-induced pain-related behaviour supports that rats with A-O catheters and those with lumbar catheters are functionally different. The rats with lumbar catheters showed a usual dose-response to intrathecally injected NMDA for a wide range of concentrations (O.l- 1.6 mM). The rats with A-O catheters responded to a more narrow range of concentrations, with the threshold at 0.1-0.2 mM and the maximum response at 0.4 mM NMDA. Convulsions were induced when the concentration of NMDA was further increased (0.5 mM). Similar results have been obtained by others using the A-O catheterization technique for intrathecal injection of NMDA (Raigorodsky and Urea, 1987). The reason for the different response to NMDA is not known. One possibility is that the A-O catheterization. inducing different degrees of
172
R.V. Stwkson
et al./Journal
ofNeuroscience
long-lasting pain, causes a change in the functional state of the spinal cord with an increase in the sensitivity of the NMDA receptor to higher concentrations of NMDA. For the lumbar catheterization only a light halothane anaesthesia was used. For the A-L) catheterization, a deeper and more long-lasting anaesthesia had to be used due to the more extensive surgery. Since the animals were allowed complete recovery before testing, it is unlikely that the difference in anaesthesia was of importance for the difference in response to NMDA. However, the shorter time interval between catheterization and the NMDA injection in the lumbar group, may have contributed to the difference in NMDA response between the two groups. 4.6. Conclusion The lumbar catheterization is a reliable and time-saving method that minimizes the neurological disturbances and reduces the suffering of the animals. The advantages of the technique are important from an ethical point of view, and they may also be important by reducing unpredictable interference when studying pain mechanisms.
Acknowledgements The authors thank M. Foiling, M.D., for preparing the scintigrams. We also thank L.J. Rygh for practical assistance and professor T.I. Kanui for useful suggestions. This study was supported in part by the Norwegian Research Council and by the European Commission.
References Aanonsen, L.M. and Wilcox, G.L. ( 1986) Phencyclidine selectively blocks a spinal action of N-methyl-D-aspartate in mice, Neurosci. Lett., 67: 191-197. Aanonsen, L.M. and Wilcox, G.L. (1987) Nociceptive action of excitatory amino acids in the mouse: effects of spinally administered opioids, phencyclidine and sigma agonists, J. Pharmacol. Exp. Ther., 243: 9-19. Bahar, M., Rosen, M. and Vickers, M.D. (1984) Chronic cannulation of the intradural or extradural space in the rat, Br. J. Anaesth., 56: 405-410. Bjorkman, R., Hallman, K.M., Hedner, J., Hedner, T. and Henning, M. (1994) Acetaminophen blocks spinal hyperalgesia induced by NMDA and substance P, Pain, 57: 259-264. DeLander, G.E. and Wahl, J.J. (1988) Behavior induced by putative nociceptive neurotransmitters is inhibited by adenosine or adenosine analogs coadministered intrathecally, J. Pharmacol. Exp. Ther., 246: 565-570. Dib, B. (1984) Intrathecal chronic catheterization in the rat, Pharmacol. B&hem. Behav., 20: 45-48. Gom&ez-Darder, J.M., G6mezCBrdenas, E. and Gil-Salt, J.L. (1989) Microsurgical catheterization of the intrathecal space in the rat for chronic infusion of drugs, Rev. Esp. Anestesiol. Reanim., 36: 153156. Hughes, H.E. and Barr, G.A. (1988) Analgesic effects of intrathecally
Methods
65 (19961167-172
applied noradrenergic compounds in the developing rat: differences due to thermal vs. mechanical nociception, Brain Res. Dev. Brain Res.. 41: 109-120. Igawa, Y., Andersson, K.E., Post, C., Uvelius, B. and Mattiasson, A. ( 1993) A rat model for investigation of spinal mechanisms in detrusor instability associated with infravesical outflow obstruction, Ural. Res., 21: 239-244. Kooistra, K.L., Rodriguez, M., Powis, G., Yaksh, T.L., Harty, G.J., Hilton, J.F. and Laws, E.R., Jr. (1986) Development of experimental models for meningeal neoplasia using intrathecal injection of 9L ghosarcoma and Walker 256 carcinosarcoma in the rat, Cancer Res., 45: 3 17-323. Kristensen, J.D., Post, C., Gordh, T., Jr. and Svensson, B.A. (1993) Spinal cord morphology and antinociception after chronic intrathecal administration of excitatory amino acid antagonists in the rat, Pain, 54: 309-316. Long, J.B., Mobley, W.C. and Holaday, J.W. (1988) Neurological dysfunction after intrathecal injection of Dynorphin A(1 - 13) in the rat. 1. Injection procedures modify pharmacological responses, J. Pharmacol. Exp. ‘lher., 246: 1158-I 166. Loomis, C.W., Milne, B. and Cetvenko, F.W. (1987) Determination of cross tolerance in rat spinal cord using intrathecal infusion via sequential mini-osmotic pumps, Pharmacol. B&hem. Behav., 26: 13l- 139. LoPachin, R.M., Rudy, T.A. and Yaksh, T.L. (1981) An improved method for chronic catheterization of the rat spinal subarachnoid space, Physiol. Behav., 27: 559-561. Martin, H., Kocher, L. and Chery-Croze, S. (1984) Chronic lumbar intrathecal catheterization in the rat with reduced-length spinal compression, Physiol. Behav., 33: 159-161. Mestm, C., Pelissier, T., Wilcox, G.L., Eschaher, A. and Lavarenne, J. (1993) Intrathecal administration of drugs in unanesthetized rats: technical and pharmacological aspects, IASP Publ., Congress Abstracts, Abstr. 217. Mjellem, N., Lund, A., Eide, P.K., Storkson, R. and Tjslsen, A. (1992) The role of 5-HTla and 5-HTlb receptors in spinal nociceptive transmission and in the modulation of NMDA induced behaviour, NeuroReport, 3: 1061-1064. Mjehem-Joly, N., Lund, A., Berge, O.G. and Hole. K. (1991) Potentiation of a behavioural response in mice by spinal coadministration of substance P and excitatory amino acid agonists, Neurosci. Lett., 133: 121-124. Papir-Kricheli, D., Frey, J., Laufer, R., Gilon, C., Chorev, M., Selinger, Z. and Devor, M. (1987) BehaviouraI effects of receptor-specific substance P agonists, Pain, 3 1: 263-276. Raigorodsky, G. and Urea, G. (1987) lntrathecal N-methyl-D-aspartate (NMDA) activates both nociceptive and antinociceptive systems, Brain Res., 422: 158- 162. Rosner, B. (1986) Fundamentals of Biostatistics, 2nd edn., PWS Publ., Boston, MA, pp. 151-153. Schoeffler, P., Auroy, P., Bazin, J.E., Taxi, J. and Woda, A. (1991) Subarachnoid midazolam: histologic study in rats and report of its effect on chronic pain in humans, Reg. Anesth., 16: 329-332. Serpell, M.G., DeLeo, J.A., Coombs, D.W., Colbum. R.W., Twitchell, B.B., Willenbring, S. and Fromm, C. (1993) Intrathecal cadteterization alone reduces autotomy after sciatic cyroneurolysis in the rat, Life Sci., 53: 1887-1892. Storkson. R.V., Kjwsvik, A., Tjdsen, A. and Hole, K. (1993) Catheterization of the lumbar spinal subarachnoid space in the rat, IASP Publ., Congress Abstracts, Abstr. 218. Wiesenfeld, Z. and Gustafsson, L.L. (1982) Continuous intrathecal administration of morphine via an osmotic minipump in the rat, Brain Res., 247: 195-197. Yaksh, T.L. and Rudy, T.A. (1976) Chronic catheterization of the spinal subarachnoid space, Physiol. Behav., 17: 103 I- 1036. Zimmertitann, M. (1983) Ethical guidelines for investigations of expetimental pain in conscious animals (editorial). Pain, 16: 109-10.