Whereas loss of motor function after spinal cord injury (SCI) is generally perceived as the major source of functional impairment, postinjury pain and spasticity are additional causes of disability and distress (1). Spasticity is characterized by an increase in tone (hypertonus) and a velocity-dependent increase in resistance to muscle stretch, believed to result from decreased inhibition of α motor neurons. Significant spasticity is associated with numerous problems, including poor perineal hygiene resultant upon adductor spasm, contractures, and decubitus ulceration, which all adversely affect on the individual's quality of life. Several pharmacological and surgical options may be used in the treatment of patients with severe SCI-induced spasticity. Surgical options have included ablative procedures, such as anterior or posterior root rhizotomy, myelotomy, and myotomy; however, these procedures are associated with significant morbidity, including therapeutic failure, loss of residual neurological function, and muscle atrophy. A number of clinical reports suggest that spinal cord stimulation may have a role in ameliorating spasm post-SCI; however, this therapeutic modality has thus far failed to gain widespread acceptance (2).
A number of drugs have been used in the management of SCI-associated spasm, including diazepam, dantrolene, and botulinum toxin A. Potential roles have also been suggested for the a2-adrenergic agonist, clonidine, the gamma-aminobutyric acid (GABA) analog, gabapentin, and the imidazole derivative, tizanidine. Despite extensive research into all of these therapeutic options, the GABAB receptor agonist baclofen remains pivotal in the management of post-SCI spasticity. Baclofen seems to mediate its effect by binding to presynaptic GABAB receptors, inhibiting calcium influx, and thus excitatory neurotransmitter release. It has been shown to be extremely effective in reducing both increased flexor tone and the frequency and severity of flexor and extensor spasms and, in addition, has been effective in both complete and incomplete cord transection (3). Baclofen is commercially available as both 10- and 20-mg tablets, and careful titration is required to identify the optimal dosage for each individual patient. The initial dose is often 10–15 mg/d gradually increasing to a maximal dose of 80 mg/d, although occasionally much larger doses may be required. Because baclofen predominantly undergoes direct renal elimination, it must be administered with caution in the setting of renal impairment to avoid excessive neurological side effects including, drowsiness, dizziness, insomnia, confusion, and ataxia. Withdrawal phenomena including anxiety, convulsions, tachyarrhythmias, and both auditory and visual hallucinations have been reported after abrupt discontinuation of the drug. Overdose has been associated with respiratory depression, coma, and seizure activity, and the seizure threshold may be reduced in epileptic patients. When effective symptom control cannot be achieved using oral baclofen administration, intrathecal (IT) delivery can offer improved symptom control and a lower side effect profile.
A 57-year-old woman was rendered quadriplegic as a result of a traumatic complete C6 transection. Her recovery was complicated by severe lower limb spasms and recurrent urinary tract infections (UTI). At no time did she complain of SCI-associated pain. Oral baclofen therapy was commenced postinjury and gradually titrated to a dose of 80 mg/d. Although she did initially derive a degree of benefit from this therapy, side effects rapidly developed, including sedation and somnolence, gastrointestinal upset as manifest by severe nausea, and cognitive impairment resulting in memory loss. As a result of these intolerable side effects, placebo-controlled IT baclofen testing was performed over a 5-day period, with test dosages of 25, 50,100, and 200 μg of preservative-free baclofen administered by a single-shot IT injection (total volume injected, 3 mL). After successful IT testing, a SynchroMed EL 10-mL (Medtronic, Inc., Minneapolis, MN) IT drug delivery system was implanted. An InDura® multiple side-hole catheter (Medtronic) was sited with the distal tip located at L1. An initial daily baclofen dose of 200 μg was used (2000 μg/mL). Approximately 2 wk after the initial implantation, spasm control significantly deteriorated. Interrogation of the pump indicated that the programmed dose of IT baclofen had been delivered. To assess the integrity of the system, a catheter myelogram and a subsequent computerized tomography (CT) was performed. These studies demonstrated that the IT catheter had become displaced and had migrated into the epidural space. The catheter was reimplanted with its tip at the level of the first lumbar vertebrae. The patient made an uneventful recovery. Over the next 3 mo, the IT baclofen dosage was gradually titrated upwards to achieve full control with a daily dose of 400 μg. She remained symptom-free on this dose for a period of 35 mo. As pain had never been a component of postinjury neurological state, she was not administered any other drug (morphine, hydromorphone, or fentanyl) by the IT route. Subsequent to this, symptom control began to deteriorate with reappearance of severe lower limb spasticity; however, she did not at any time develop altered neurological function or pain. Again, the drug delivery system was interrogated and did not reveal evidence of failure to deliver drug. It was postulated that a series of recurrent UTIs were responsible for the deterioration in symptom control. A series of incremental upward dosage adjustments was performed, resulting in a total daily dose of 500 μg/d. Despite this increase in dosage and resolution of the UTIs, her symptoms did not improve. Thirty-eight months after the implantation, a catheter myelogram was again performed to assess if, as had previously been the case, catheter migration was the cause of this deterioration in symptom control. Injection of the contrast showed no evidence of leakage along the course of the catheter, and filling of the thecal sac was identified several millimeters proximal to the catheter tip. Magnetic resonance imaging (MRI) myelography was subsequently performed, suggesting the presence of catheter tip-associated granuloma (Fig. 1B). As the mass did not cause any significant compression of neural structures, IT catheter removal and reimplantation was performed via reopening of the original lumbar wound site. Her spasticity is currently well controlled after a return to the previously stable daily IT baclofen dose of 400 μg.
When adequate symptom control cannot be achieved by oral administration of baclofen, or an excessive side effect profile is associated with an effective dose, direct IT administration is an effective method for controlling spasticity and reducing total baclofen dosage (1). IT administration of baclofen has been shown to reduce both clinical symptoms (2,3) as well as electromyographic activity in affected muscle groups (4). IT baclofen provided long-term control of spasticity in a multicenter study demonstrating significant reduction in the Ashworth score for rigidity and spasms over a mean follow-up period of 19 months (5–41 months) in 75 patients suffering from spasticity secondary to SCI and multiple sclerosis (5). Although gradual dose increases may be required in this patient population to maintain optimal control, drug tolerance does not seem to be a limiting factor. The only complication noted in this study was late pump pocket infection, and the authors concluded that IT baclofen delivery was a safe and effective method for controlling spasticity secondary to SCI or multiple sclerosis.
The efficacy of IT baclofen administration in patients who have failed oral therapy has been demonstrated in a double-blind, randomized, crossover study (6). Twenty patients suffering from spasticity secondary to SCI or multiple sclerosis were followed for a mean period of 19.2 months after commencement of IT baclofen administration. Significant reductions in muscle tone and spasm severity were demonstrated. The study was complicated by two episodes of catheter displacement and one episode of pump failure; however, no significant side effects attributable to baclofen were noted. The authors concluded that IT administration is an effective and safe long-term therapy in the management of patients with spasticity of spinal origin that is unresponsive to oral baclofen therapy. In addition, IT baclofen administration has a potential role in the management of central pain (7,8) and has been classified as a fourth-line drug when used in a pain management capacity (9).
In 1991, North et al. (10) reported the first case of an IT catheter tip inflammatory mass in a patient receiving morphine. Since then, there have been multiple reports of catheter-associated masses (11–16). A review of 41 cases compiled from the medical literature (16 cases) (as of November 2000) and from clinical reports (25 cases) to Medtronic Inc or to the U.S. Food and Drugs Administration revealed that in all cases, the IT drug delivery system was used in the management of a chronic pain condition. Drugs used included morphine or hydromorphone, either alone or in combination with another drug. The mean duration over which patients had been receiving IT drugs was 24.5 months. Importantly, catheter-associated masses were not described in any patient who had received baclofen as their only IT drug (17). Of the 41 patients reported, 30 underwent surgery to relieve spinal cord or cauda equina compression, and 11 were rendered nonambulatory. Increased physician awareness and a postal survey in 2001 resulted in an additional 51 cases being reported (18). These reports described noninfectious chronic inflammatory masses surrounding the tip of the IT drug delivery catheter; clinical indications for therapy and IT drugs used closely paralleled those described by Coffey and Burchiel (17).
In a review of the 92 cases identified by Yaksh et al. 18 between 1990 and 2002, most patients (74 cases) received their IT drug delivery system for the management of chronic noncancer pain. The main duration of therapy was 29 months; 46 patients received morphine alone, whereas a further 11 received it in combination with other drugs (bupivacaine, tetracaine, or clonidine). Dosage information data were available in 40 patients, 12 of whom were treated with the dose <10 mg/d, with 2 groups of 14 patients each receiving 10–14 mg/d and >15 mg/d of morphine. The IT morphine concentration was >50 mg/mL in 18 of the 30 patients in whom the concentration was reported. Other drugs that were in use at the time of catheter tip mass diagnosis include hydromorphone, either alone (seven cases) or in combination (eight cases), tramadol (one case), and fentanyl (one case). No drug information was available in 18 cases. There was, however, no report of a catheter-associated inflammatory mass in a patient who had received baclofen as the only IT drug. The presenting symptoms that led to the diagnosis of catheter-associated mass were reported as loss of drug efficacy or onset of new neurological symptoms, including radicular pain, para/monoparesis, and cauda equina syndrome.
Radiological confirmation was obtained with MRI in 45 cases, myelogram with or without CT in 17 cases, and MRI, myelogram, and CT in 4 cases. Patients in whom the catheter-associated mass did not fill the spinal canal or cause neurological impairment were effectively treated by discontinuing the IT infusion or replacing it with sterile normal saline infusing at a minimal infusion rate, with resultant stabilization or resolution of the mass. In patients demonstrating neurological sequelae, prompt neurosurgical intervention was warranted. Using data obtained from these cases, the estimated cumulative risk of developing an IT mass in the chronic pain population has been reported as 0.04% over one year, 0.12% over two years, and extending to 1.15% over six years. Because no cases have been reported that involved patients treated with IT baclofen for spasticity, incidence and relative risk rates were only calculated for patients treated using IT drug pumps for chronic pain. These studies concluded that patients receiving large-dose IT opioid either as a sole drug or in an admixture, or those receiving drugs not approved for IT administration, should be monitored closely for evidence of IT mass formation and prompt action should be taken if required.
Potential mechanisms for catheter tip granuloma formation have included catheter-based mechanisms, such as catheter tip design, placement-associated trauma, final tip position, infection, and silicone hypersensitivity, and drug-related mechanisms, such as impure/contaminated drugs and the action of opioid agonists on immunological function and the blood-central nervous system barrier. Chronic infection has effectively been discounted as a potential cause because of the absence of positive microbial staining and culture in human reports and experimental animal models (10,11,18). Endotoxin-induced aseptic meningitis-like mechanisms (19) have been proposed; however, IT morphine sulfate has been shown to be a nonendotoxin inducer (20). Initial case reports involved a catheter design in which a single hole was used; however, more recent reports have involved newer closed-end multi-hole catheters, a design which was considered less likely to provoke granuloma formation. Silicone allergy is also an unlikely mechanism, given that granuloma formation only occurs at points of active drug infusion and has not been demonstrated on silicone lumbar shunts or silicone breast implants (21–23).
In their sentinel paper, North et al. (10) raised the possibility of arachnoiditis as an important etiological factor; however, the reports of granuloma formation in the absence of previous spinal surgery argue against this. Previous spinal surgery can lead to the formation of blind ending pouches, which by influencing local drug concentration, may modulate the process of granuloma formation. Implantation-associated trauma has been suggested as a possible cause; however, the prolonged latency period would seem to argue against this (17,24). An outbreak of granuloma formation was reported consequent upon drug contamination; however, this seems to be an isolated event (25). The most plausible explanations for IT granuloma formation synthesized from animal and human data include the properties of the IT drug (particularly morphine), cerebrospinal fluid (CSF) flow dynamics, and possibly catheter tip location (17). Large-dose IT morphine administration has been demonstrated to induce catheter tip granuloma formation in a canine model (26). These masses consisted of multifocal accumulations of neutrophils monocytes, macrophages, and plasma cells. Whether this is a μ receptor-mediated response, or a local effect of morphine glucuronide metabolites, remains to be elucidated. In an animal study (26), a mild local pericatheter response consistent with foreign body reaction was demonstrated in all animals including the saline-treated control group. This local pericatheter reaction was no more florid in those animals that later developed a catheter tip-associated granuloma than in the control group; however, it has been proposed that this local foreign body response may be a required continuing stimulus for granuloma formation (26). Experimentally, in a canine model, chronic IT baclofen delivery (2 mg/mL) for a 28-day period was not shown to result in granuloma formation (18), but this does not exclude the -potential for this complication in long-term infusion. The possibility of a role for regional CSF flow dynamics, which determine local drug concentration, must be considered.
In the setting of significant spinal cord trauma, spinal cord trauma syringomyelia or myelomalacia have been suggested to be associated with altered CSF flow dynamics, which may be demonstrable on MRI (27). It is possible that significantly altered CSF dynamics may result in increased local drug concentrations, thereby increasing the potential for granuloma formation. Additionally, it is plausible that the trauma of repeated-catheter implantations may induce a localized inflammatory response that may be sufficient to induce granuloma formation irrespective of the drug administered.
A consensus paper has established guidelines for the prevention, diagnosis, and management of catheter tip-associated masses (Table 1) (28). The authors point out that baclofen-associated masses have not been reported and that, therefore, the guidelines are predominantly concerned with patients receiving IT opioids. They suggest that baseline neurological variables should be assessed and closely monitored for early signs of neural compression by an expanding IT mass.
Preventive measures suggested include maintaining the IT opioid dose and concentration at a small level and lumbar versus thoracic placement of the catheter, although this may simply reduce the potential neurological sequelae of granuloma formation. Any alteration in drug efficacy or neurological status must alert the attending physician to the possibility of a catheter-associated mass, and appropriate investigations must be undertaken. Noninvasive diagnosis may be effectively achieved using either CT-myelogram or T1-weighted MRI. A number of acceptable therapeutic options are available depending upon the patient's clinical status. Surgical removal and decompression are considered preferable in situations where there is profound or progressive neurological impairment. For small masses diagnosed during investigation of diminished analgesic efficacy, discontinuation of IT drug delivery or withdrawal of catheter to a lower IT site may be associated with significant shrinkage or disappearance of the mass (17).
IT drug delivery systems are frequently used in the management of chronic pain states and spasticity. This method of drug delivery provides an effective method for symptom control when oral administration proves ineffective, or is associated with intolerable side effects. Perhaps the most significant potential complication associated with this system is the development of a catheter-tip associated inflammatory mass. Not only can this complication reduce the efficacy of the drug delivery system, but it may also cause serious neurological sequelae. Granuloma formation has been demonstrated only in the presence of chronic IT opioid administration and, consequently, all consensus recommendations relate to delivery of this drug group only. In reporting the first case of IT granuloma associated with baclofen, we suggest that the consensus recommendations for catheter-associated granuloma prevention, diagnosis, and treatment (28) must additionally be applied to all patients receiving prolonged IT administration of baclofen, particularly in the setting of SCI and potential derangement of CSF flow dynamics.
1. Burchiel KJ, Hsu FPK. Pain and spasticity after spinal cord injury. Spine 2001;26:146–60.
2. Nance P, Schryvers O, Schmidt B, et al. Intrathecal therapy for adults with spinal spasticity: therapeutic efficacy and effect on hospital admissions. Can J Neurol Sci 1995;22:22–9.
3. Lazorthes Y, Safferin-Caute B, Verdie JC, et al. Chronic intrathecal baclofen administration for control of severe spasticity. J Neurosurg 1990;72:393–402.
4. Müller H, Zierski J, Dralle D, et al. The effect of intrathecal baclofen on electrical muscle activity in spasticity. J Neurol 1987;243:348–52.
5. Coffey JR, Cahill D, Steers W, et al. Intrathecal baclofen for intractable spasticity of spinal origin: results of a long-term multicenter study. J Neurosurg 1993;78:226–32.
6. Penn RD, Savoy SM, Corcos D, et al. Intrathecal baclofen for severe spinal spasticity. N Engl J Med 1989;320:1517–21.
7. Herman RM, D'Luzansky SC, Ippolito R. Intrathecal baclofen suppresses central pain in patients with spinal lesions: a pilot study. Clin J Pain 1992;8:338–45.
8. Lobser PG, Akman NM. Effects of intrathecal baclofen on chronic spinal cord injury pain. J Pain Symptom Manage 1996;12:241–7.
9. Bennett G, Serafini M, Buchiel K, et al. Evidence-based review of the literature on intrathecal delivery of pain medication. J Pain Symptom Manage 2000;20:S12–36.
10. North RB, Cutchis PN, Epstein JA, et al. Spinal cord compression complicating subarachnoid infusion of morphine: case report and laboratory experience. Neurosurgery 1991;29:778–84.
11. Aldrete JA, Vascello LA, Ghaly R, et al. Paraplegia in a patient with an intrathecal catheter and a spinal cord stimulator. Anesthesiology 1994;81:1542–5.
12. Blout JP, Remley KB, Yue SK, et al. Intrathecal granuloma complicating chronic spinal infusion of morphine: report of three cases. J Neurosurg 1996;84:72–6.
13. Bejjani GK, Karim NO, Tzortzidis F. Intrathecal granuloma after implantation of a morphine pump: case report and review of the literature. Surg Neurol 1997;8:288–91.
14. Cabell KL, Taren JA, Sagher OS. Spinal cord compression by catheter granulomas in high dose intrathecal morphine therapy: case report. Neurosurgery 1998;42:1176–81.
15. Anderson SR, Orbegozo M, Racz G, et al. Intrathecal granuloma in patients receiving high-dose intrathecal morphine therapy: a report of two cases. Pain Practice 2001;1:61–7.
16. McMillan MR, Doud T, Nugent W. Catheter-associated masses in patients receiving intrathecal analgesic therapy. Anesth Analg 2003;96:186–90.
17. Coffey RJ, Burchiel K. Inflammatory mass lesions associated with intrathecal drug catheters: report and observations on 41 patients. Neurosurgery 2002;50:78–87.
18. Yaksh TL, Hassenbusch S, Burchiel K, et al. Inflammatory masses associated with intrathecal drug infusion: a review of preclinical evidence and human data. Pain Med 2002;3:300–12.
19. Cooper JF, Harbert JC. Endotoxin as a cause of aseptic meningitis after radionucleotide cisternography. J Nucl Med 1975;16:809–13.
20. Cooper JF, Thoma LA. Screening extemporaneously compounded intraspinal injections with the bacterial endotoxin test. Am J Health Syst Pharm 2002;59:2426–33.
21. James HE, Tibbs PA. Diverse clinical applications of percutaneous lumboperitoneal shunts. Neurosurgery 1981;8:39–42.
22. McIvor J, Krajbich JI, Hoffman H. Orthopaedic complications of lumboperitoneal shunts. J Pediatr Orthop 1988;8:687–9.
23. Angell M. What's really behind the attack on silicone breast implants? Med Econ 1996;73:131–3,136,139–40.
24. Dahm P, Nitescu P, Appelgren L, et al. Efficacy and technical complications of long-term continuous spinal infusions of opioid and/or bupivacaine in refractory non-malignant pain: a comparison between the epidural and intrathecal approach with externalized or implanted catheters and infusion pumps. Clin J Pain 1998;14:4–16.
25. Jones TF, Feler CA, Simmons BP, et al. Neurological complications including paralysis after a medication error involving implanted intrathecal catheters. Am J Med 2002;112:31–6.
26. Yaksh TL, Horais KA, Tozier NA, et al. Chronically infused intrathecal morphine in dogs. Anesthesiology 2003;99:174–87.
27. Freund M, Adwan M, Kooijman H, et al. Measurement of CSF flow in the spinal canal using MRI with optimized MRI protocol: experimental and clinical studies. Rofo 2001;173:306–14.
28. Hassenbusch S, Burchiel K, Coffey RJ, et al. Management of intrathecal catheter-tip inflammatory masses: a consensus statement. Pain Med 2002;3:313–23.