Skip navigation
Link to Back issues listing | Back Issue Listing with content Index | Subject Index

Intrathecal baclofen for severe spasticity

Systematic review

Health service purchasers are often faced with the problem of making decisions in the absence of evidence on benefit and cost. Evidence is most often lacking in rare but severe conditions, and in early applications of new technology. The example of intrathecal baclofen for severe spasticity [1] shows that even in these difficult circumstances sufficient evidence may be garnered to inform decision-making.

Intrathecal baclofen needs to be used on a continuous basis. This involves not just demonstrating that it is effective in the individual patient, but complex neurosurgery to implant a continuous infusion device delivering baclofen to the head, or cord, or both. The device needs to be refilled several times a year, and the operation re-done every five or six years when batteries need replacing.

Systematic review

This set out to gather several different types of information. Information on benefits of continuous intrathecal baclofen was sought through a literature search (to end 1999) using four electronic databases, including the Cochrane Library. Any type of study was eligible, if patients had one of five conditions (cerebral palsy, multiple sclerosis, hypoxic brain injury, traumatic brain injury, or spinal cord injury). Studies had to describe some functional benefit, like bed-bound patients being able to sit in a wheelchair, or improved ability to perform activities of daily living, or reduction in spasm-related pain. In all the included trials patients had to have severe disabling spasticity refractory to oral medicines, and in addition they must have shown a response to a bolus dose of baclofen.

A separate search was for economic analyses or cost studies. Cost information identified from the literature was supplemented by semi-structured interviews with clinicians. Quality of life was estimated from the evidence review, supported by clinical opinion. Three simplified scenarios were used, which concentrated on mobility and pain as the criteria most likely to be affected. The scenarios were:

  1. Bed-bound patients with severe spasm-related pain
  2. Bed-bound patients not in pain
  3. Wheelchair users with moderate spasm-related pain.


There were 17 studies published between 1985 and 1997, with information on between 7 and 70 patients. Follow up was between about six months and six years. A summary of the outcomes is in Table 1.

Table 1: Outcomes after intrathecal baclofen for severe spasticity


Number benefiting/total with complaint

Percent benefiting

Improved ease of nursing care 83/90 92
Bedridden patients able to sit in a wheelchair 50/76 66
Reduction in spasm-related pain 55/62 89
Improved ability to perform ADL 45/62 73
Ambulatory patients improving ability to walk 18/45 40
Improved ability to sit comfortably in a wheelchair 31/36 86
Wheelchair bound becoming ambulatory (assisted) 4/36 11
Improved ability to transfer 25/26 96
Improved skin integrity 19/23 83
Improved wheelchair mobility 13/18 72

Quality of life improvement estimates were 0.27 for a bed-bound patient not in pain, to 0.5 for a bed-bound patient experiencing severe spasm-related pain. The cost was estimated at about £12,000 for assessment, test dose and implantation procedure, with follow up costs of up to £1,200 a year for refills. Over five years, the total discounted cost was £15, 400. The cost per quality adjusted life year ranged from £6,900 to £12, 790 for the three scenarios (Table 2).

Table 2: Cost/QALY estimates


Cost/QALY (£)

Bed-bound patients experiencing severe spasm-related pain 6,900
Bed-bound patients not in pain 12,790
Wheelchair users with moderate spasm-related pain 8,030


The authors are justifiably cautious, but their conclusion was that intrathecal baclofen produces functional benefits and is likely to be an appropriate use of resources in carefully selected patients. They specified that patients had to have severe disabling spasticity refractory to oral medicines, and have shown a response to a bolus dose of baclofen. Methods employed included systematic searches for evidence of effectiveness and cost, backed up with sound clinical opinion, and restricted to a particular scenario. The process results in a reasonable quantification of cost/benefit, which we can compare with other things we purchase.

The benefits and costs are still estimates, though. They can be modified as more evidence emerges on benefit, on cost, or, crucially, on the quality of life improvements obtained.


  1. FC Sampson et al. Functional benefits and cost/benefit analysis of continuous intrathecal baclofen infusion for the management of severe spasticity. Journal of Neurosurgery 2002 96: 105201057.
previous or next story in this issue