© U.S. Cancer Pain Relief Committee, 1999 0885-3924/99/$–see front matterPublished by Elsevier, New York, New York PII S0885-3924(99)00069-X

Vol. 18 No. 3 September 1999 Journal of Pain and Symptom Management 223

Clinical Note

Histological Findings After Long-Term Infusion of Intrathecal Ketamine forChronic Pain: A Case Report

Martin Stotz, MD, Hans-Peter Oehen, MD, and Helmut Gerber, MD

Institute of Anesthesiology (M.S., H.G.) and Pathology (H-P.O.), Kantonsspital, Basel, Switzerland

Abstract

Ketamine, a selective, noncompetitive N-methyl-

D

-aspartate (NMDA)-receptor antagonist, is able to alter pain perception at the spinal level. Little clinical data exist on the intrathecal and epidural use of ketamine in chronic pain. Histopathologic findings after intrathecal injection of ketamine with and without preservatives are rarely reported. This outcome was evaluated in a 72-year-old woman with abdominal pain due to cancer who was treated with the intrathecal application of bupivacaine, clonidine, and morphine. We reached satisfactory pain relief with the addition of ketamine to the mixture for 7 days. On postmortem, focal lymphocytic vasculitis close to the catheter injection site was found. This finding has not been described previously after long-term application of ketamine intrathecally. The intrathecal infusion of ketamine with preservative, or the mixture of ketamine, clonidine, morphine, and bupivacaine resulted in isolated lymphocytic vasculitis of the spinal cord and leptomeninges without any clinical signs of neurological deficit.

J Pain Symptom Manage 1999;18:223–228.

© U.S. Cancer Pain Relief Committee, 1999.

Key Words

NMDA-receptor antagonist, ketamine, chronic pain, intrathecal infusion, histopathology,

side effects

Introduction

Ketamine, a selective noncompetitive

N

-methyl-

D

-aspartatic acid (NMDA)-receptor antagonist,is able to alter pain perception at the spinallevel, regulating the spinal wind-up and recep-tive field size of the dorsal horn neurons.

1

Ke-tamine inhibits the NMDA receptor by block-ing the open channel of neurons and therebyreducing the channel mean open time and thefrequency of channel opening due to an allo-steric mechanism.

2

Ketamine has been administered epidurallyfor pain relief in humans without side effectssuch as respiratory depression, urinary reten-tion, or pruritus.

1,3

Intrathecally administeredketamine has been shown to have some anes-thetic effect.

4

Although no detrimental neuro-logical symptoms have been noted after spinaladministration of ketamine, the neurotoxicityof intrathecal ketamine administration has notbeen studied clinically.

5–9

Few data exist on the epidural and intrathe-cal use of ketamine in chronic pain patients. Inone case series, the addition of ketamine to amorphine–lidocaine–clonidine mixture intra-thecally provided sufficient analgesia in fourpatients with terminal cancer pain without thedevelopment of tolerance.

10

Yang et al.

11

also

Address reprint requests to:

Martin Stotz, MD, Instituteof Anesthesiology, Kantonsspital, 4031 Basel, Swit-zerland.

Accepted for publication: November 30, 1998.

224 Stotz et al. Vol. 18 No. 3 September 1999

found a significant reduction of intrathecalmorphine dosage and pain scores when ke-tamine was added to the morphine mixture.

We report the use of intrathecal ketamine inthe management of a patient with intractablecancer pain of the lower abdomen. Postmor-tem histological

evaluation was obtained afterlong-term intrathecal infusion with commer-cially available ketamine.

Case Report

HM was a 72-year-old Caucasian female witha history of peritoneal malignant mesothe-lioma diagnosed in 1996. She underwent sev-eral laparotomies for tumor debulking, partialresection of the small intestine, and lysis of ad-hesions. No adjuvant chemotherapy or radia-tion therapy was undertaken.

Pain was initially managed with oral opioidanalgesics, a nonsteroidal anti-inflammatorydrug, and a psychotropic. She presented in ourpain unit with pain consisting of constantburning sensation in the right lower abdomenand intermittent sharp pain in the right in-guinal region, probably due to infiltration ofthe abdominal wall. Her pain scores were in-creasing and reached a score of 8 on a scale of10 before hospitalization (day 0). Pain scoresand management are detailed in Fig. 1.

On presentation, we first used intravenous(IV) patient-controlled analgesia (PCA) withmorphine (morphine 2 mg/ml; PCA program:1 ml/demand, lockout interval 7 min, 4-h limit20 mg) in addition to continuing the oral anal-gesic management (pain score 7). On day 8, aCT-guided neurolysis of the celiac plexus wasperformed without beneficial effect (score 6).Then, a thoracic epidural catheter was placed atT6 and a continuous epidural infusion of bupi-vacaine 0.125% and morphine (0.04 mg

?

ml

2

1

)was started at a rate of 6–12 ml

?

h

2

1

(day 14)with minor success (score 6–8). On day 22, weplaced an intrathecal catheter (Spinocath, B.Braun Co., Germany) with continuous infusionof a mixture of bupivacaine 0.25% and mor-phine (0.12 mg

?

ml

2

1

) at a rate of 2–2.5ml

?

h

2

1

(score 6). The tip of the catheter was placed atthe level of T7. Because of unsatisfactory painrelief, we first increased the infusion rate to3–3.5 ml

?

h

2

1

, which resulted in a sensory levelC4 to L2, then increased the morphine dose to0.3 mg

?

ml

2

1

and added clonidine to the mix-ture (3

m

g

?

ml

2

1

) with still unsatisfactory painrelief (for detailed management, see Fig. 2). Atthis point, the patient experienced some mo-tor weakness of the upper extremities and theabdomen without apparent discomfort.

Finally, pain relief (score 2) was obtainedwith the addition (1 mg

?

ml

2

1

) of commercially

Fig. 1. Management trials and related pain scores on pain scale during hospitalization in a 72-year-old patient withpain due to peritoneal malignant mesothelioma.

Vol. 18 No. 3 September 1999 Intrathecal Ketamine for Chronic Pain 225

available ketamine (preservative: benzethoniumchloride) to the intrathecal mixture. The meandaily dose of ketamine administered intrathe-cally was 67.2 mg. After 7 days of intrathecal in-fusion, the patient developed acute psychoticalterations that were thought to be related toketamine; no abnormal neurological signswere found. Ketamine was removed from theinfusion scheme and i.v. PCA (morphine 2mg/ml; PCA program: background infusion0.5 ml/min, lockout interval 5 min, no 4-h limit)was added to the pain management (score 2).The next day, the patient developed a stiff neckwith normal peripheral reflexes. The cerebrospi-nal fluid (CSF) was clear and the analysis showeda cell count of 16

?

ml

2

1

(normal 0–3), normalglucose, protein, and chloride; no microorgan-isms were found in microscopic examinationand culture. Restitutio ad integrum occurredafter 18 hours without treatment.

Ten days later, the patient succumbed to herdisease (before death, score 3) and autopsy wasperformed within 41 hours after death withspecial attention to the spinal cord and lep-tomeninges. Histological sections were stainedwith hematoxylin–eosin and Klüver-Barrera.

Histological Findings

Histological examination showed focal lym-phocytic vasculitis in the medullary tissue, in

the nerves, and in the leptomeninges (Fig. 3)of the thoracic and lumbar spinal cord. Themononuclear infiltrate consisted mainly oflymphocytes. The percentage of vessels show-ing signs of inflammation was 10% in the med-ullary tissue, 25% in nerves, and 30% in theleptomeninges of the caudal thoracic segment;others did not show signs of alteration.

No other histological changes were observedin medullary tissue. There was no necrosis,hemorrhage, or signs of demyelinization orvacuolization (Fig. 4). There was no vasculitisin other organs or tissues, neither in the brainnor its meninges.

Discussion

Neuraxial opioid therapy is often effective intreating cancer pain that has not been ade-quately controlled by enteral or parenteral treat-ment.

12

The addition of epidural clonidine toopioids has been shown to be particularly effi-cacious in the treatment of neuropathic pain.

13

In this case, we were unable to reach satisfac-tory pain relief with an intrathecal mixture ofbupivacaine, morphine, and clonidine, despitean adequate high sensory level.

Fifty mg of intrathecally administered keta-mine can produce surgical anesthesia with asensory blockade lasting for 1 hour.

4

The com-bination of intrathecal and/or epidural keta-

Fig. 2. Detailed infusion scheme for intrathecal treatment with ketamine.

226 Stotz et al. Vol. 18 No. 3 September 1999

mine with opioids is intended to avoid sideeffects by either drug. Two authors have re-ported the successful addition of commerciallyavailable ketamine for treatment of chronic

cancer pain.

10,11

The sedative effect after largedoses of intrathecal ketamine, which was alsoseen in our patient, is apparently due to sys-temic uptake.

3,4,14

Psychotic alterations, as seen

Fig. 3. Severe lymphocytic vasculitis of spinal vessel after long-term intrathecal infusion of ketamine with preserva-tive, normal medullary tissue; stained hematoxylin–eosin.

Fig. 4. Normal nerve sheaths with central axon, no signs of demyelinization or vacuolization in the same patient;stained Klüver-Barrera.

Vol. 18 No. 3 September 1999 Intrathecal Ketamine for Chronic Pain 227

in our patient, are common after i.v. or i.m.ketamine administration.

15

The histological findings of an isolated lym-phocytic vasculitis of the spinal cord and lep-tomeninges near the catheter tip are surpris-ing. Isolated vasculitis of the central nervoussystem is rare and may be due to a primary dis-ease such as granulomatous angiitis or due to asecondary manifestation of a group of underly-ing diseases such as lymphoma, leukemia, ormetastatic small cell lung cancer.

16

With nega-tive microscopic examination and culture ofCSF analysis, an infectious angiitis seems im-probable in our case. In addition, no other sys-temic underlying disease was found at autopsy.Either a drug-related origin of angiitis or im-munocomplex-mediated reaction of spinal ves-sels must be suspected.

The intrathecal administration of drugs al-ways carries the potential for neurotoxicity.Comparing magnesium sulfate, saline, andimplanted catheter alone, no histological dif-ferences were shown in rats after 1 month andrepeated injections.

17

Low pH and high osmo-lality can produce parenchymatous necrosisand/or hemorrhage of the spinal cord aftersingle-shot spinal anesthesia with hyperbaricbupivacaine.

18

With the pH of the intrathecalmixture infused in our case being 5.46 and theosmolality 299 mosmol/kg, they seem unlikelyto be responsible for the observed angiitis.

Mild deformation of the spinal cord insheep was found after repeated injection ofclonidine through intrathecal catheters, withsigns of local demyelinization at the cathetersite, but no leptomeningeal or cord inflamma-tion or other histological changes were seen.

19

After single-shot intrathecal administrationof ketamine, with and without preservative(benzethonium chloride), in monkeys, Brock-Utne et al.

5,6

found no inflammatory reactions.The same negative findings are reported byBorgbjerg et al.,

8

and they found no histologi-cal difference between preservative-free keta-mine 1% and saline after repeated intrathecalinjections in rabbits. Others have observedradicular demyelinization in rats after repeatedintrathecal injections of ketamine 5%, withbenzethonium chloride as preservative.

20

Mali-novsky

et al.

7

found mild vascular lesions andloss of vessel outline with the preservative chlo-robutanol, but not with lidocaine or preserva-tive-free d-ketamine 1% after a single-shot ad-

ministration. Recently, a case report has beenpublished noting a subpial vacuolar

myelopa-thy after intrathecal administration of keta-mine with preservative.

21

Preservative-free ketamine as a racemate orisomers may even have beneficial effects

invitro.

Cytoprotective properties in hippocampalneurons

22

and nephroprotective effects inhibit-ing the mesangial cell proliferation at clinicallyrelevant concentrations have been observed.

23

In conclusion, satisfactory pain relief was ob-tained in a patient with intractable cancer painunresponsive to conventional pain therapy withthe addition of commercially available keta-mine to an intrathecal mixture of bupivacaine,morphine, and clonidine. Clinically, no neuro-logic deficits were observed. Postmortem find-ings of focal lymphocytic vasculitis close to thecatheter injection site, without other histologi-cal changes after long-term intrathecal infu-sion, may be related to the preservative benze-thonium chloride or the toxicity of the mixtureitself.

References

1. Yaksh T. Epidural ketamine: a useful, mechanis-tically novel adjuvant for epidural morphine? RegAnesth 1996;21:508–513.

2. Orser B, Pennefather P, MacDonald J. Multiplemechanisms of ketamine blockade of

N

-methyl-

D

-aspartate receptors. Anesthesiology 1997;87:903–917.

3. Ravat F, Dorne R, Baechle J, Beaulaton A, Le-noir B, Palmier B. Epidural ketamine or morphinefor postoperative analgesia. Anesthesiology 1987;66:819–822.

4. Bion J. Intrathecal ketamine for war surgery: apreliminary study under field conditions. Anaesthe-sia 1984;39:1023–1028.

5. Brock-Utne J, Mankowitz E, Maharaj R, Drown-ing J. Intrathecal ketamine with preservative-histo-logical effects on spinal nerve roots of baboons. SAfr Med J 1982;61:440–441.

6. Brock-Utne J, Mankowitz E, Kallichurum S,Drowning J. Effects of intrathecal saline and ketaminewith and without preservative on the spinal nerveroots of monkeys. S Afr Med J 1982;61:360–361.

7. Malinovsky J, Lepage J, Cozian A, Mussini J,Pinaud M, Souron R. Is ketamine or its preservativeresponsible for neurotoxicity in rabbit? Anesthesiol-ogy 1993;78:109–115.

8. Borgbjerg F, Svensson B, Frigas C, Gordh T. His-topathology after repeated intrathecal injections ofpreservative-free ketamine in the rabbit: a light and

228 Stotz et al. Vol. 18 No. 3 September 1999

electron microscopic examination. Anesth Analg1994;79:105–111.

9. Abram S. Spinal cord toxicity of epidural andsubarachnoid analgesics. Reg Anesth 1996;21:84–88.

10. Muller A, Lemos D. Cancer pain: beneficial ef-fect of ketamine addition to a morphine–clonidine–lidocaine mixture administered by intrathecalroute. Ann Fr Réanim 1996;15:271–276.

11. Yang C, Wong C, Chang J, Ho S. Intrathecal ke-tamine reduces morphine requirements in patientswith terminal cancer pain. Can J Anaesth 1996;43:379–383.

12. Ballantyne J, Carr D, Berkey C, Chalmers T, Mos-teller F. Comparative efficacy of epidural, subarach-noid, and intracerebroventricular opioids in patientswith pain due to cancer. Anesth 1996;21:542–556.

13. Eisenach J, DuPen S, Dubois M, Miguel R, AllinD. Epidural clonidine analgesia for intractable can-cer pain. Pain 1995;61:391–399.

14. Nagibu M, Adu-Gyamfi Y, Absood G, Farag H,Gyasi H. Epidural ketamine for postoperative anal-gesia. Can Anaesth Soc J 1986;33:16–21.

15. Baer G, Parkas P. Ketamine-induced psycho-pathological changes in normal volunteers duringconditions used for experimental psychoses. Anaes-thetist 1981;30:251–256.

16. Lie T. Angiitis of the central nervous system.Curr Opin Rheumatol 1991;3:36–45.

17. Chanimov M, Cohen M, Grinspun Y, et al. Neu-rotoxicity after spinal anaesthesia by serial intrathe-cal injections of magnesium sulphate. Anaesthesia1997;52:223–228.

18. Ganem E, Vianna P, Marques M, Castiglia Y,Vane L. Neurotoxicity of subarachnoid hyperbaricbupivacaine in dogs. Reg Anesth 1996;21:234–238.

19. Eisenach J, Dewan D, Rose J, Angelo J. Epiduralclonidine produces antinoception, but no hypoten-sion, in sheep. Anesthesiology 1987;66:496–501.

20. Amiot J, Pallaci J, Vedrenne C, Pellerin M. Spi-nal toxicity of lysine acetylate and ketamine hydro-chloride given by the intrathecal route in the rat.Ann Fr Anesth Réanim 1986;5:462.

21. Karpinsky N, Dunn J, Hansen L, Masliah E. Sub-pial vacuolar myelopathy after intrathecal ketamine:report of a case. Pain 1997;73:103–105.

22. Himmelseher S, Pfenninger E, Georgieff M.The effects of ketamine-isomers on neuronal injuryand regeneration in rat hippocampal neurons.Anesth Analg 1996;83:505–512.

23. Jimi N, Segawa K, Minami K, Sata T, ShigematsuA. Inhibitory effect of the intravenous anesthetic,ketamine, on rat mesangial cell proliferation.Anesth Analg 1997;84:190–195.