Troubleshooting for Intrathecal Baclofen Therapy

US Neurology, 2007;4(1):
Received: January 02, 2006 Accepted December 16, 2006 Citation US Neurology, 2007;4(1):

Intrathecal baclofen (ITB) administration via a programmable implanted delivery system has become standard of care in the treatment of disabling, severe spasticity resulting from a wide range of diseases of the brain or the spinal cord. ITB improves severe spasticity of either spinal or cerebral origin in the majority of cases.1,2 Treatment outcomes are measured by gains in pre-determined functional goals, as well as by the modified Ashworth spasticity scale (MAS).3

Baclofen exerts its effect at the gamma-aminobutyric acid (GABA)B receptor locus of spinal interneurons, resulting in suppression of the disinhibited mono- and polysynaptic spinal reflexes. This is believed to generate hyperpolarization of neurons, resulting in prevention of calcium influx, facilitation of potassium conductance, and inhibition of excitatory neurotransmitter release.4 A differential agonist-induced desensitization of pre- and post-synaptic responses mediated by GABAB receptors has been demonstrated. It is suggested that such desensitization may contribute to the physiological tolerance to baclofen that may occur during initial treatment.5

Baclofen administered directly into the cerebrospinal fluid (CSF) achieves high concentrations in the spinal CSF and in the spinal cord. It penetrates into the superficial layers of the cord to bind to GABAB receptors. The result is that a small dose of baclofen, measured in micrograms per day, is needed to achieve the desired clinical effects. The CSF concentration of baclofen is 200 times higher than that in the plasma of patients receiving intrathecal therapy. A gradient of concentration exists, with the lumbar CSF concentration being four times higher than cisternal concentrations.6,7 This phenomenon is thought to be responsible for the greater effect on lower limb spasticity than on the upper extremities.8 Although baclofen is available in oral form, its poor lipophilic qualities prevent effective penetration of the blood–brain barrier (BBB), resulting in the need for large doses to treat severe spasticity.8 Orally administered baclofen is distributed throughout the CNS, often resulting in impaired cognition and alertness.9 Patients treated intrathecally have excellent clinical outcomes; combined with a drastic reduction of dose-related adverse effects, this results in an enhancement in quality of life.10

Prior to pump implantation, a screening bolus of baclofen (usually 50–100mcg) is administered via lumbar puncture to test the patient’s response.11 The response is measured by a reduction in the MAS and an improvement in individual pre-determined functional goals, such as speed and ease of ambulation and transfers or subjective pain perception. Once an infusion system is implanted, treatment typically starts at a daily dose twice that of the successful bolus test dose.11 The therapy is gradually titrated to the desired individualized functional clinical outcome, while oral antispastic medications are concomitantly tapered off. Due to tolerance, the dose of ITB needs to be gradually increased over a period of six to 18 months until stability is achieved. A steady course can then be maintained, unless the patient has a progressive neurodegenerative disease or a destabilizing event.12

The Synchromed® II infusion system, the implantable pump used to deliver ITB therapy, has a refillable reservoir of 20 or 40cc, holding a drug supply for up to six months. The system allows for individual programming of ITB to deliver a simple continuous infusion or complex continuous infusion for managing circadian tone changes noted in some patients and/or periodic drug boluses. The battery life is estimated to be seven years.13 Patients implanted with Synchromed II can safely undergo scans in magnetic resonance imaging (MRI) machines with a field strength of ≤1.5 Tesla. During the scan, drug delivery temporarily halts, but normal pre-programmed function is resumed when the test is completed.13,14

ITB complication rates range from 20 to 50%.15,16 Complications include drug (baclofen) adverse events, system-related malfunction (catheter or pump), human error (programming or refill mistakes), or host events (infections, intercurrent illness).11

Drug-related Adverse Events

Although rare, withdrawal from baclofen or a baclofen overdose can occur. The clinician must be aware of the manifestations of both and be familiar with the interventions required to treat these potentially life-threatening conditions.17,18 Most cases of withdrawal from baclofen are mild and can be safely and successfully managed in the outpatient setting. Occasionally, patients experience acute withdrawal, which can potentially be fatal.

The most common cause of acute withdrawal from baclofen is catheter malfunction. Early symptoms include pruritus, hypertonia, hyperthermia, and painful spasms. Autonomic dysreflexia can be triggered. In severe cases, delirium, seizures, coma, or death may occur.19 All patients treated with ITB and their care-givers must be educated about the early symptoms of baclofen withdrawal. An emergency supply of oral baclofen must be prescribed to be used when patients or their caregivers recognize early symptoms of withdrawal. A practice management system that keeps track of patient refill dates and drug doses should be part of the ITB team’s standard operating procedures.11 Local emergency department staff should be educated and familiar with this syndrome.17–21 The treatment of baclofen withdrawal includes a high dose of baclofen via an oral or enteral route, resumption of ITB as soon as possible, intravenous benzodiazepines, or cyproheptadine via an oral or enteral route.22

The main cause of baclofen overdose is human error, such as using an incorrect drug concentration or programming mistakes.11 The symptoms include decreased cognition and alertness, nausea, vomiting, hypotension, hypoventilation, hypercapnia, and, in severe cases, circulatory and respiratory failure requiring fluid resuscitation, vasopressors, and airway management. The treatment of a baclofen overdose should include stopping the pump and emptying it, removing the drug from the catheter via the side port. Alternatively, an attempt to remove a large volume of drug-containing CSF should be made via a lumbar puncture, restarting ITB therapy at the appropriate time to avoid baclofen withdrawal later.

Other Drug-related Adverse Events

Our institution encountered a few severe psychiatric events that we believe are related to ITB and that resolved with discontinuation of baclofen administration or with dose reduction. One case of frank psychosis resolved with discontinuation of ITB. Several patients reported severe unrelenting depression and others reported insomnia. These appear to be ITB-doserelated and improved with dose reduction.

Host Complications

All patients who exhibit a recurrence of spasticity after a period of stable treatment regimen need to be carefully evaluated for the following medical conditions:11 disease progression; intercurrent illness, such as infection (especially urinary tract and respiratory infection), skin breakdown, bone fractures, or other injuries; concomitant medication use, especially the use of antidepressants23 and stimulants; and severe constipation.

An elevation of body temperature above 39°C may cause excessive drug delivery. Prompt treatment of fevers is needed to avoid this complication. Changes in pressure such as in traveling to high altitudes, scuba diving, or airplane travel can cause temporary changes in the rate of drug delivery with relatively minor, transient changes in the patient’s tone. Typically, all that is needed to address this problem is patient education and awareness.13

Patients who once had a favorable response to ITB but now exhibit resistance to increasing doses of baclofen require careful work-up to identify either a reason for the increased tone or evidence of system failure. If the work-up is negative, two options should be considered:

  • switch the patient to intrathecal morphine (via the same infusion
    system), either as a temporary measure or permanently—this may offer
    an alternative in managing the patient’s spasticity;24 or
  • consider significant dose reduction but continue ITB therapy.25

Implementation of either of these measures should prompt the clinician to consider medicating the patient with oral baclofen, benzodiazepines, and cyproheptadine22 to prevent acute baclofen withdrawal.

System Malfunctions

Catheter Complications

Catheter-related complications are the most common problems associated with ITB therapy and often result in either sudden or gradual baclofen withdrawal. Most complications need not result in permanent cessation of therapy. Identification of a catheter malfunction will usually lead to surgical revision. Complete or partial fractures or disconnects, kinks, catheter migration, arachnoiditis, or fibrin plugs are all possible.1,15,26 Partial disconnects or micro-fractures can result in variable responses, which can be positional. Catheter migration out of the subarachnoid space into the subcutaneous tissue usually occurs early in the course of ITB and is the result of suboptimal anchoring. This complication usually occurs early following system implantation. The clinical benefit experienced during bolus test dose is not achieved. Complete catheter fracture or disconnection, which can happen at any time during the course of ITB therapy, results in complete cessation of therapeutic effect. In rare cases, penetration of the catheter tip into the subdural space may occur with drug puddling in a subdural pocket and spilling into the subarachnoid space intermittently. This results in an inconsistent, random response to ITB (periods of spasticity alternating with periods of hypotonia).

A systematic approach to identifying a catheter problem includes the following steps:

  • Program a bolus dose equal to the patient’s original, effective test dose
    via the pump–catheter system. If the patient has a good response to such
    a bolus, this may indicate a partial disconnect or tear in the catheter. Here,
    the relatively large bolus volume can penetrate where a slow continuous
    infusion cannot. In this situation, a decision needs to be made to either
    revise the catheter or to maintain the patient on intermittent boluses of
    baclofen as either a permanent or interim solution.
  • X-ray the catheter course to localize the radio-opaque catheter
  • Evaluate the catheter patency by aspirating the catheter port, as CSF flows
    easily from an intact system.
  • Inject the catheter port with contrast dye and perform an examination of
    the intrathecal penetration using fluoroscopy. This must be carried out with
    great caution as the CSF must first be aspirated. If CSF cannot be aspirated,
    the injection should be aborted to avoid the risk of overdose from the
    baclofen situated in the catheter. In cases of subdural drug pooling in the
    subarachnoid space, this maneuver can cause serious baclofen overdose;
    therefore, if this is suspected, a catheter port injection should be avoided.

Further ways to identify catheter complications include the following:

  • Carry out spiral computed tomography (CT) of the catheter course to
    assist in localizing the catheter tip and identifying fractures.
  • Use a radioisotope—technetium diethylene triamine pentaacetic acid
    (DTPA) or indium—injected into the pump reservoir. The appearance of
    an isotope at the spinal level is compatible with, at least, partial catheter
    patency, while its absence suggests complete catheter malfunction.27
  • Deliver a bolus dose via lumbar puncture, equal to the patient’s original,
    effective test dose or at a higher dose. A good response to direct
    intrathecal delivery of drug via lumbar puncture, when there was no
    response to a similar bolus programmed through the pump, definitively
    proves some kind of catheter malfunction even in the absence of
    conclusive evidence from any of the above tests.

Pump Complications

Rare unexpected battery failure or rotor stall can occur.13 The identification of pump failure always results in replacement of the device. Battery failure is expected after seven years of use and should be anticipated in most patients, leading to a planned, elective replacement of the pump. In the rare cases in which pump malfunction is suspected, two steps should be taken to evaluate it. First, aspirate the content of the reservoir and compare it with the data from the system telemetry. Additionally, a calculation of the volume of drug in the reservoir should be performed based on reservoir volume and delivery rate. All of these numbers should match. A mismatch suggests a pump malfunction. Second, visualize rotor function using serial X-rays of the pump. The X-ray must be performed in the anterior posterior direction and must be underexposed to visualize the rotor. Once the first X-ray has been taken, the pump should be programmed to deliver a bolus of drug that will generate a 60–90° revolution of the rotor.

Operator-generated Complications

A systematic approach to refills and pump programming prevents most human errors.11 If a mistake has occurred, such as a refill with the wrong concentration of drug, steps should be taken to correct this as soon as possible to avoid withdrawal or an overdose.

Prevention of Complications in the Post-operative Period

Surgical replacement of the pump at the end of battery life must be performed by experienced providers who are familiar with the necessary precautions to correctly empty and re-prime the catheter at the time of surgery. In order to avoid complications related to surgical technique, the clinician must be familiar with the pitfalls associated with system implantation and follow best practice guidelines for device and catheter handling and infection control.28,29

Special Populations

Multiple Sclerosis

Managing severe spasticity in the patient with multiple sclerosis (MS) via the use of ITB is well established.30,31 The patient with MS who is treated with ITB has a progressive disease, and may require frequent regulation of the ITB dose depending on disease progression. MS patients appear to be particularly sensitive to intrathecal baclofen, and often need only minute dose adjustments. Also, ultimate effective treatment doses can be low (despite poor response to oral baclofen). The test dose in these patients may need to be considerably lower than the typical test dose. Doses of 25mcg or lower often achieve the desired effect. It is not unusual for the patient with MS to remain hypotonic for many hours, in contrast to other patient populations in whom baseline spasticity returns within about eight hours after the administration of the bolus.32

Patients with Ventriculo-peritoneal Shunts

The interactions between a ventriculo-peritoneal (VP) shunt and an intrathecal drug delivery system must be considered. A VP shunt diverts CSF from the ventricle, causing a decrease in CSF volume and a relatively higher concentration of baclofen in the CSF. When a patient on a stable dose of ITB requires a VP shunt installation, the dose of ITB likely needs to be reduced. In contrast, patients whose VP shunt malfunctions, causing an expansion of expansion of the CSF volume, are likely to experience an increase in their spasticity.33,34


The treatment of severe spasticity with intrathecal baclofen has revolutionized the approach to the spastic patient and has improved the quality of life of thousands of individuals. The ITB team should be familiar with the systematic approach to the resolution of any suboptimal treatment response so that patients can be safely and continuously maintained on this effective therapy.


I am grateful to Barbara Ridley, RN, NP, for her review of the manuscript and her comments. I would like to acknowledge the Spasticity Management team members at Alta Bates Summit Medical Center in Berkeley, California, for the collaborative work in the treatment of patients with severe spasticity. The team comprises numerous individuals, too many to mention each by name, from the departments of Neurology, Rehabilitation, Neurosurgery, and Radiology.

1. Albright AL, Gilmartin R, Swift D, et al., J Neurosurg, 2003; 98:291–5.
2. Ordia J, Neuromodulation, 2002;5(1):16–24.
3. Bohannon R, Phys Ther, 1986;67:206–7
4. Muller H, Local-spinal therapy of spasticity, Berlin: Springer-Verlag, 1988;223–6.
5. Wetherington J, J Phys, 2002;544(2):459–67
6. Knutsson E, Lindblom U, Mårtensson A, J Neurol Sci, 1974;23: 473–84.
7. Kroin J, Clin Pharmacokinet, 1992;22:319–26.
8. Kroin J, Parenteral Drug Therapy in Spasticity and Parkinson’s Disease, Park Ridge, New Jersey: Parthenon, 1991;67–77.
9. Francisco G, Top Stroke Rehabil, 2001;8(1):36–46.
10. Gianino JM, York MM, Paice JA, Shott S, J Neurosci Nurs, 1998; 30(1):47–54.
11. Ridley B, Rawlins PK, J Neurosci Nurs, 2006;38(2):72–82.
12. Nielsen JF, Hansen HJ, Sunde N, Christensen JJ, Clin Neurol Neurosurg, 2002;104:142–5.
13. Medtronic Inc., Clinical Reference Guide, 2007.
14. Medtronic Inc., MRI guidelines for Neurological Products, 2005.
15. Follet K, Neuromodulation, 2003;6(1):32–41.
16. Teddy P, Jamous A, Gardner B, et al., Br J Neurosurg, 1992; 6(2):115–18.
17. Lioresal® Intrathecal. Package insert, 2005.
18. Medtroic Inc. Product monograph, 2002.
19. Coffey RJ, Edgar TS, Francisco GE, et al., Arch Phys Med Rehabil, 2002;83:735–41.
20. Raphael J, Psychosomatics, 2005;46:503–7.
21. Samson-Fang L, Gooch J, Norlin C, Dev Med Child Neurol, 2000; 42:561–5.
22. Meythaler JM, Roper JF, Brunner RC, Arch Phys Med Rehabil, 2003; 84:638–50.
23. Stolp-Smith KA,Wainberg MC, Arch Phys Med Rehabil, 1999; 80(3):339–42.
24. Soni BM, Mani RM, Oo T, Vaidyanathan S, Spinal Cord, 2003; 41:586–9.
25. Cooper JA, Ridley B, Neurol, 2006;67:1495–6.
26. Dickerman RD, Schneider SJ, J Pediatr Surg, 2002;37:17–19.
27. Hicks RJ, Kalff V, Brazenor G, Clin Nucl Med, 1989;14:275–7.
28. Albright AL, Turner M, Pattisapu JV, J Neurosurg, 2006;104 (Suppl. 4):233–9.
29. Follet K, Neuromodulation, 2003;6:32–41.
30. Haselkorn JK, Balsdon Richer C, Fry Welch D, J Spinal Cord Med, 2005;28:167–99.
31. Jarrett L, Int J MS Care, 2001;3:1–11.
32. Ridley B, Rehabil Nurs, 2006;4:158–64.
33. Fulkerson H, Child’s Nerv Sys, 2007;7:733–8.
34. Turner M, Conference: Spasticity Management in the Patient With Stroke & MS, San Francisco, 2006.