Share this article on:

The Effects of Subarachnoid Administration of Preservative-Free S(+)-Ketamine on Spinal Cord and Meninges in Dogs

Rojas, Alfredo Cury MD*; Alves, Juliana Gaiotto MD*; Moreira e Lima, Rodrigo PhD*; Esther Alencar Marques, Mariângela PhD; Moreira de Barros, Guilherme Antônio PhD*; Fukushima, Fernanda Bono PhD*; Modolo, Norma Sueli Pinheiro PhD*; Ganem, Eliana Marisa PhD*

doi: 10.1213/ANE.0b013e31823a5d1b
Analgesia: Research Reports

BACKGROUND: The N-methyl-D-aspartate receptor antagonist ketamine and its active enantiomer, S(+)-ketamine, have been injected in the epidural and subarachnoid spaces to treat acute postoperative pain and relieve neuropathic pain syndrome. In this study we evaluated the effects of a single dose of preservative-free S(+)-ketamine, in doses usually used in clinical practice, in the spinal cord and meninges of dogs.

METHODS: Under anesthesia (IV etomidate (2 mg/kg) and fentanyl (0.005 mg/kg), 16 dogs (6 to 15 kg) were randomized to receive a lumbar intrathecal injection (L5/6) of saline solution of 0.9% (control group) or S(+)-ketamine 1 mg/kg−1 (ketamine group). All doses were administered in a volume of 1 mL over a 10-second interval. Accordingly, injection solution ranged from 0.6% to 1.5%. After 21 days of clinical observation, the animals were killed; spinal cord, cauda equina root, and meninges were removed for histological examination with light microscopy. Tissues were examined for demyelination (Masson trichrome), neuronal death (hematoxylin and eosin) and astrocyte activation (glial fibrillary acidic protein).

RESULTS: No clinical or histological alterations of spinal tissue or meninges were found in animals from either control or ketamine groups.

CONCLUSION: A single intrathecal injection of preservative-free S(+)-ketamine, at 1 mg/kg−1 dosage, over a concentration range of 6 to 15 mg/mL injected in the subarachnoid space in a single puncture, did not produce histological alterations in this experimental model.

Published ahead of print December 13, 2011 Supplemental Digital Content is available in the text.

From the *Anesthesiology Department, and Pathology Department, São Paulo State University–UNESP, Botucatu/SP, Brazil.

This study was funded by the Brazilian Federal Agency for the Support and Evaluation of Graduate.

The authors declare no conflicts of interest.

Reprints will not be available from the authors.

Address correspondence to Fernanda Bono Fukushima, PhD, Departamento de Anestesiologia–FMB/UNESP, Rubião Jr. s/n Caixa postal 530 Botucatu-SP CEP 18618–970 Brazil. Address e-mail to

Accepted September 28, 2011

Published ahead of print December 13, 2011

Ketamine was synthesized in 1962 as a hypnotic anesthetic.1 Its cardiovascular and psychotomimetics effects narrowed its use and, in the late 1980s, ketamine use was significantly decreased.2 However, the discoveries concerning the pathophysiology of pain, specifically related to N-methyl-D-aspartate (NMDA) receptors, have brought attention to potential uses of ketamine in pain management.3 By blocking NMDA receptors, ketamine reduces central sensitization, pain memory, and spinal facilitation. Because of these properties, ketamine has been used as a preemptive analgesic, to prevent opioid tolerance, and as an opioid-sparing drug.2,4,5

Because of the racemic structure, ketamine and the S(+) stereoisomer [S(+)-ketamine] have a greater affinity for the NMDA receptor, 3 to 4 times greater potency, and faster clearance compared to the R− stereoisomer [ketamine R (−)].6 S(+) ketamine solution is available without preservatives. Ketamine and S(+)-ketamine have both been administered by epidural and subarachnoid routes to treat acute postoperative pain and chronic pain of neuropathic syndromes.4,5 Nevertheless, despite the wide clinical experience, the safety of using ketamine and S(+)-ketamine by the epidural and subarachnoid routes remains controversial.7,8

Inconsistent results of the effects on neural tissue and meninges of ketamine or S(+)-ketamine with and without preservatives using different neuraxial delivery protocols have been reported in various species, including rat, rabbit, dog, and baboon.9 14 One study demonstrated that a continuous long-term intrathecal infusion of ketamine and other NMDA antagonists in a canine model induced significant histopathology,14 indicating that long-term exposure could be damaging.

Although long-term exposure may have deleterious effects, acute perioperative use is an important application of spinal ketamine. The long-term studies may thus overestimate the risk associated with acute single use exposure. Moreover, while many clinicians may use ketamine epidurally and it could be argued that this route yields less spinal exposure, we note that the general risk of subarachnoid delivery after epidural injection is 5%.15 Accordingly, the aim of this study was to evaluate the acute neurotoxic effect to the spinal cord and the meninges of dogs after a single subarachnoid dose of preservative-free S(+)-ketamine in a dose and concentration range resembling that used in patients.

Back to Top | Article Outline


After approval by the Sao Paulo State University Ethics Committee on animal experiments, adult mongrel dogs (n = 16) of both sexes were obtained from the Experimental Animal Center at the State University of Sao Paulo at Botucatu Campus. The mean and SD weight of the animals was 9 ± 2 kg (7 to 12 kg) for the control group and 11 ± 3 kg (6 to 15 kg) for the ketamine group. The mean length of the vertebral column was 60 ± 5 cm for the control group and 58 ± 4 cm for the ketamine group. The length of the vertebral column was measured from the occipital protuberance (C1) to the lombrosacral space (L7).16 All tests were performed in accordance with the guidelines of the International Association for the Study of Pain.17

The dogs were randomized into 2 experimental groups (control and ketamine) according to the type of solution injected into the subarachnoid space. Eight dogs were allocated to each group. The control group received 0.9% normal saline solution intraspinally. The ketamine group received 1 mg/kg−1 of preservative-free S(+)-ketamine solution. The researchers administering the solution were blinded to group assignment.

The animals were fasted 12 h before the procedure, with water ad libitum. All dogs were submitted to the same anesthetic technique, etomidate (2 mg/kg) and fentanyl (0.005 mg/kg) administered IV. A 10-cm area around the site of the spinal puncture at the L6 to L7 intervertebral space level was washed with water and soap, followed by hair removal and skin cleansing with 0.9% normal saline. The depilated skin was submitted to antisepsis with a 2% chlorhexidinegluconate solution, and sterile fields were appropriately positioned. Subarachnoid puncture was performed through the median line, approximately 45° to the skin with a 22-gauge Quincke needle. Difficulties during the procedure and the characteristics of the cerebrospinal fluid were recorded. If a traumatic spinal tap was identified, as defined by the presence of blood in the cerebrospinal fluid or the need for more than one attempt of puncture, the animal was immediately excluded from the study and no solution was administered into the subarachnoid space. Once the needle was properly located and clear cerebrospinal fluid could be identified, the 1-mL of solution [saline or S(+)-ketamine] was injected over 10 seconds through 1-mL disposable syringes.

The S(+)-ketamine sterile preservative-free solution was synthesized by Cristália (Sao Paulo, Brazil) at pH 4.5. The solutions were supplied in individual ampoules. The ampoules contained a fixed mass of liquid ketamine that was reconstituted to the appropriate dose in 1 mL of saline. Thus, a 10-kg dog would receive 10 mg of ketamine in 1 mL of solution (10 mg/mL) whereas a 15-kg dog would receive 15 mg of ketamine in 1 mL of solution (15 mg/mL). The 0.9% saline solution (Baxter Healthcare Corp., Sao Paulo, Brazil) administered to the control group had a pH of 5.0.

Animals in the study were evaluated 1 hour after the intrathecal injection and at 3-day intervals for 21 days after the injection. Each animal was assessed regarding the following secondary outcomes: motor deficit, anal sphincter tonus, and response to nociception. Motor deficit was determined by the ability to walk, jump, and sustain the tail in an upward position. Anal sphincter relaxation was ascertained through visual inspection [normal (totally closed) or abnormal (sphincter open with spontaneous fecal loss)]. Nociception was assessed by reaction to painful pressure. To control for possible interference due to visual perception of the stimuli by the animals, one researcher was responsible for masking the animals with a nontransparent cloth comfortably positioned around their neck. Pressure nociceptive stimuli were elicited by the bilateral pinch of a skin fold over sacral, lumbar, and thoracic dermatomes, as well as the interdigital membranes of hindlimbs by a surgical clamp. The presence of pain was defined by the following: limb withdrawal, vocalization, and facial expression. Motor deficit, anal sphincter relaxation, and nociception were classified dichotomously into absent or present. If the slightest deficit in any of these dimensions was observed at clinical assessment, the animal would be classified as positive for deficit.

Finally, histological analysis of the spinal cord and meninges of the dogs was performed after the 21-day observation period. To obtain tissue samples, animals were first given sodium pentobarbital IV then euthanized by electroshock. Thereafter, the lumbar and sacral segments of the spinal cord with the surrounding meninges were quickly removed within 3 minutes to minimize the risk of injuries to those tissues from ischemia and apoptosis. The anatomical pieces were fixed in a 10% formalin solution. After a 7-day incubation period, 3000-micron thick histological sections were prepared starting 10 cm above the level of the spinal puncture to the end of the cauda equina. The histological sections were stained by hematoxylin and eosin, Masson trichrome, and glial fibrillary acidic protein (GFAP) techniques and examined by optical microscopy. Three researchers (JGA, EMG, and MEAM) experienced in histological neurotoxicity assessment unanimously classified each of the sections according to the presence or absence of histological injury. If any kind of lesion was identified, it was further specified. To investigate the possible dose-related gradient effect, injuries were stratified according to severity and extent as ascertained by consensus. Researchers blinded to the experimental groups performed all clinical and histological evaluations.

For the purpose of measuring the number of cells, ImageJ (version 1.43), a stand-alone Java-based image analysis program developed at the National Institutes of Health (available as open source at, was used to recognize and count the nuclei.18 GFAP-stained images were captured by Nikon Coolscope II microscope and processed with a heterogeneous correction filter to address the uneven illumination of the microscope field (Fig. 1). For each animal the histological section closest to the puncture site was selected, and 4 images of the dorsal horn and 2 images from the ventral horn of those spinal cord sections were collected (Fig. 2). The software ascertained the number of cells in each of those images.

Figure 1

Figure 1

Figure 2

Figure 2

The sample size was calculated according to Fleiss et al.19 estimating a proportion of histological neurotoxicity of 1% and 80% in the control and ketamine group, respectively, so as to obtain a power value of 90% while setting the one-sided α level for statistical significance at 0.05.

The R software Version 2.3.4 was used for the performance of statistical analysis.20 To evaluate the effectiveness of the randomization procedure and the comparability of the 2 study groups, we performed one-way ANOVA comparing group differences regarding animals' weights and the length of their vertebral column. One-sided Fisher's exact test was selected to compare the frequencies of the findings on primary and secondary outcomes between the ketamine and the control groups. Differences among mean numbers of cells were analyzed by means of Student's t-test. Alpha level for statistical significance was set at 0.05.

Back to Top | Article Outline


Both groups were similar in weight (P = 0.08) and length of the vertebral column (P = 0.28). None of the animals was excluded due to traumatic puncture or deaths. The time of recovery from anesthesia in both groups was approximately 30 minutes. During the 21 days of observation, none of the animals had impaired motor function, anal sphincter relaxation, or decreased nociception.

None of the animals showed macroscopic or microscopic signs of direct injuries to neural tissue and meninges such as hemorrhage, infarct, vacuolization, necrosis, or meningeal thickness during the necropsy. Histological analyses indicated no differences in GFAP and hematoxylin and eosin staining (Figs. 3 and 4) between the 2 comparison groups. The mean number of cells in all 4 images selected from the dorsal horn was 1111 in the ketamine group and 1021 in the control group (P = 0.86). As for the 2 images selected from the ventral spinal cord, the mean number of cells was 928 in the ketamine group and 1138 in the control group (P = 0.07). No correlation or tendency between the cell count and S(+)-ketamine solution concentration was found (Fig. 5). To minimize confounding factors related to the vertebral column length, the data were recalculated and the number of cells was compared to the ratio of the concentration of the S(+)-ketamine and the length of the spinal cord (Fig. 6). Again, no relationship or tendency between cell counts and S(+)-ketamine concentration adjusted for the spinal cord length was found.

Figure 3

Figure 3

Figure 4

Figure 4

Figure 5

Figure 5

Figure 6

Figure 6

Back to Top | Article Outline


Good pain control after surgery is important to prevent negative outcomes such as neural sensitization. Clinical windup occurs from the processes of NMDA activation, windup central sensitization, long-term enhancement of pain, and transcription-dependent sensitization. Ketamine has been used in perioperative pain management, including the spinal route. Its use is justified by the fact that it exerts a specific NMDA blockade and, hence, modulates central sensitization.21

Other studies have investigated some aspects of the safety and toxicity of continuous and single intrathecal doses of ketamine, S(+)-ketamine with or without preservative (Table 1); however, the safety and toxicology of ketamine with regard to the spinal cord and meninges remain unclear. Although some case reports have indicated that the chronic use of intrathecal ketamine in human patients has depicted serious pathologic observations at autopsy,8 the results of this study in dogs demonstrate that the acute dose of 1 mg/kg−1 of preservative-free S(+)-ketamine administered in the subarachnoid space did not cause toxicity to the nervous tissue or the meninges.

Table 1

Table 1

To our knowledge, this is the first investigation to specifically examine the neurotoxicity profile of a single intraspinal doses of preservative-free S(+)-ketamine, both clinically and histologically, while measuring outcomes for longer than 1 week. Contrary to studies using preservative-free S(+)-ketamine with daily injections by catheters or used with a single injection9,13,22 that showed damage to the nervous tissue, in the present study we did not find any lesions in the nervous tissue or meninges 21 days after the injection (Table 1).8 14,22

The present study was a blinded, randomized, controlled trial in which potential confounding issues due to lesions induced by the spinal puncture procedure in the ketamine group were controlled by comparison to the control group. The experimental model of a single intraspinal dose is similar to the usual spinal anesthesia procedures used in humans. This technique has less risk of complications than other models in which implantable intrathecal catheters are used.

The doses of ketamine administered in the epidural space to manage postoperative pain range from 0.25 mg/kg−1 23,24 to 1 mg/kg−1 in humans. The 1 mg/kg−1 dose was chosen because it is the largest dose reported in the literature.25 In animals, this dosage is used to study the analgesic and neurotoxic effects of ketamine.9,11,12

The prolonged clinical observation (21 days) enabled the evaluation of chronic effects that S(+)-ketamine could also cause to the meninges. Results of previous studies with this same methodology showed that the drug primarily damages the nervous tissue, where clinical and histological alterations are immediately observed after drug administration.26,27 However, when the meninges are damaged, a longer interval allows for the inflammatory reaction, which later causes the nervous tissue lesion. Nevertheless, another study using the same methodology with subarachnoid amitriptyline28 depicted extensive adhesive arachnoiditis 21 days after spinal injection.

There are reports that repeated doses of S(+)-ketamine can cause excessive antagonism in NMDA receptors. This antagonism can produce neurotoxicity due to inactivation of the inhibiting mechanism (comprising the NMDA receptor blockage in the interneuron mediated by γ-aminobutyric acid which is responsible for the tonic inhibition of excitatory pathways) resulting in excitotoxic lesion with cellular necrosis and apoptosis.29 Moreover, prolonged NMDA receptor antagonism prevents the endogenous mechanisms of surviving and regenerating neurons.30 This can explain some case reports in human patients who have received chronic intrathecal ketamine and showed serious pathological alterations after autopsy.8,31,32

Benzothonium chloride and chlorobutamol are both preservatives found in the racemic ketamine preparation. Benzothonium chloride is thought to cause severe histological lesions that can spread in the dorsal and ventral regions of the medulla,11 and chlorobutamol was associated with lesions in the posterior dorsal root in 100% of the animals studied.13 The drug solution used in our study was free of chemical preservatives so as to exclude the possibility that the histological findings were reactions to a substance other than ketamine. Fast removal and fixation of the anatomical piece, as well as the comparison to controls submitted to the same procedure, make our findings unlikely because of cord extraction-related ischemic injuries or other procedure-related mechanisms.

Previous studies found a strong correlation between the drug concentration and neurotoxicity. Our main goal was to evaluate the effect of drug doses used for acute pain treatment in humans; thus, the concentrations used were low and variable between the animals. This can explain the absence of histological damage. Moreover, the purpose of our study was to evaluate the possible clinical effects in the nervous tissue and meninges. As such, we did not evaluate the possible analgesic effects of intrathecal ketamine but rather used a dichotomous clinical evaluation instead of a scalar measure for the comparison of motor function, anal sphincter tone, and sensibility to painful stimulation. Because the classification scheme used was quite stringent (the slightest deficit would be classified as positive), it is likely the comparisons were even more rigorous than scalar comparisons. Also, the assessment of anal sphincter tone was made exclusively by visual inspection without the use of a sphincter manometer. Although the observation of sphincter relaxation is usually straightforward, the assessment method used would not be able to detect minor deficits regarding anal sphincter pressure.

On the basis of the present results, preservative-free S(+)-ketamine administered in the subarachnoid space in a single dose of 1 mg/kg−1 and 0.6 to 1.5% concentrations did not trigger toxicity, neither on the nervous tissue, nor on the meninges, in this experimental model in dogs. Despite the negative results observed, more research is necessary to determine the safety of ketamine in humans.

Back to Top | Article Outline


All authors participated in the study, wrote and approved the manuscript.

This manuscript was handled by: Tony L. Yaksh, PhD.

Back to Top | Article Outline


1. Arditti J, Spadari M, de Haro L, Brun A, Bourdon JH, Valli M. Ketamine–dreams and realities. Acta Clin Belg 2002:Suppl:31–3
2. Subramaniam K, Subramaniam B, Steinbrook RA. Ketamine as adjuvant analgesic to opioids: a quantitative and qualitative systematic review. Anesth Analg 2004;99:482–95
3. Petrenko AB, Yamakura T, Baba H, Shimoji K. The role of N-methyl-D-aspartate (NMDA) receptors in pain: a review. Anesth Analg 2003;97:1108–16
4. Hocking G, Cousins MJ. Ketamine in chronic pain management: an evidence-based review. Anesth Analg 2003;97:1730–9
5. Walker SM, Goudas LC, Cousins MJ, Carr DB. Combination spinal analgesic chemotherapy: a systematic review. Anesth Analg 2002;95:674–715
6. Oye I, Paulsen O, Maurset A. Effects of ketamine on sensory perception: evidence for a role of N-methyl-D-aspartate receptors. J Pharmacol Exp Ther 1992;260:1209–13
7. Eisenach JC, Yaksh TL. Epidural ketamine in healthy children–what's the point? Anesth Analg 2003;96:626
8. Vranken JH, Troost D, Wegener JT, Kruis MR, Van Der Vegt MH. Neuropathological findings after continuous intrathecal administration of S(+)-ketamine for the management of neuropathic cancer pain. Pain 2005;117:231–5
9. Ahuja BR. Analgesic effect of intrathecal ketamine in rats. Br J Anaesth 1983;55:991–5
10. Borgbjerg FM, Svensson BA, Frigast C, Gordh T Jr. Histopathology after repeated intrathecal injections of preservative-free ketamine in the rabbit: a light and electron microscopic examination. Anesth Analg 1994;79:105–11
11. Errando CL, Sifre C, Moliner S, Valia JC, Gimeno O, Minguez A, Boils P. Subarachnoid ketamine in swine—pathological findings after repeated doses: acute toxicity study. Reg Anesth Pain Med 1999;24:146–52
12. Malinovsky JM, Cozian A, Lepage JY, Mussini JM, Pinaud M, Souron R. Ketamine and midazolam neurotoxicity in the rabbit. Anesthesiology 1991;75:91–7
13. Malinovsky JM, Lepage JY, Cozian A, Mussini JM, Pinaud M, Souron R. Is ketamine or its preservative responsible for neurotoxicity in the rabbit? Anesthesiology 1993;78:109–15
14. Yaksh TL, Tozier N, Horais KA, Malkmus S, Rathbun M, Lafranco L, Eisenach J. Toxicology profile of N-methyl-D-aspartate antagonists delivered by intrathecal infusion in the canine model. Anesthesiology 2008;108:938–49
15. Bogduk N. Epidural Use of Steroids in the Management of Back Pain and Sciatica of Spinal Origin. Canberra: Commonwealth of Australia, 1994
16. Hendrix PK, Raffe MR, Robinson EP, Felice LJ, Randall DA, Epidural administration of bupivacaine, morphine, or their combination for postoperative analgesia in dogs. J Am Vet Med Assoc 1996;209:598–607
17. Zimmermann M. Ethical guidelines for investigations of experimental pain in conscious animals. Pain 1983;16:109–10
18. Collins TJ. ImageJ for microsopy. Biotechniques 2007;43:25–30
19. Fleiss JL, Levin BA, Paik MC. Statistical Methods for Rates and Proportions, 3rd ed. Hoboken, NJ: J. Wiley, 2003
20. Team RDC. R: A Language and Environment for Statistical Computing. In: Computing RFfS ed. Vienna, Austria, 2006
21. Vadivelu N, Mitra S, Narayan D. Recent advances in postoperative pain management. Yale J of Biology and Med 2010;83:11–25
22. Walker SM, Westin BD, Deumens R, Grafe M, Yaksh TL. Effects of intrathecal ketamine in the neonatal rat: evaluation of apoptosis and long-term functional outcome. Anesthesiology 2010;113:147–59
23. Himmelseher S, Durieux ME. Ketamine for perioperative pain management. Anesthesiology 2005;102:211–20
24. Himmelseher S, Ziegler-Pithamitsis D, Argiriadou H, Martin J, Jelen-Esselborn S, Kochs E. Small-dose S(+)-ketamine reduces postoperative pain when applied with ropivacaine in epidural anesthesia for total knee arthroplasty. Anesth Analg 2001;92:1290–5
25. Subramaniam K, Subramaniam B, Pawar DK, Kumar L. Evaluation of the safety and efficacy of epidural ketamine combined with morphine for postoperative analgesia after major upper abdominal surgery. J Clin Anesth 2001;13:339–44
26. Pires SR, Ganem EM, Marques M, Castiglia YM. Effects of increasing spinal hyperbaric lidocaine concentrations on spinal cord and meninges: experimental study in dogs. Rev Bras Anestesiol 2006;56:253–62
27. Ganem EM, Vianna PT, Marques M, Castiglia YM, Vane LA. Neurotoxicity of subarachnoid hyperbaric bupivacaine in dogs. Reg Anesth 1996;21:234–8
28. Fukushima FB, Barros GA, Marques ME, Vidal EL, Ganem EM. The neuraxial effects of intraspinal amitriptyline at low concentrations. Anesth Analg 2009;109:965–71
29. Farber NB, Kim SH, Dikranian K, Jiang XP, Heinkel C. Receptor mechanisms and circuitry underlying NMDA antagonist neurotoxicity. Mol Psychiatry 2002;7:32–43
30. Hoyte L, Barber PA, Buchan AM, Hill MD. The rise and fall of NMDA antagonists for ischemic stroke. Curr Mol Med 2004;4:131–6
31. Vranken JH, van der Vegt MH, Kal JE, Kruis MR. Treatment of neuropathic cancer pain with continuous intrathecal administration of S +-ketamine. Acta Anaesthesiol Scand 2004;48:249–52
32. Kozec SA, Sator-Katzenschlager S, Kress HG. IntrathecalS(+) ketamine in refractory neuropathic cancer pain. Pain 2006;121:283–4
33. Vranken JH, Troost D, de Haan P, Pennings FA, van der Vegt MH, Dijkgraaf MG, Hollmann MW. Severe toxic damage to the rabbit spinal cord after intrathecal administration of preservative-free S(+)-ketamine. Anesthesiology 2006;105:813–8
    © 2012 International Anesthesia Research Society