Study designs of evaluations included in the review
Randomised double-blind placebo-controlled crossover trials, prospective open trials, and studies of retrospective open design were included; some studies used a combination of methodologies. The studies had to report either the pre- and post-intervention Ashworth and Penn spasm scores, or the number of patients treated successfully.
Specific interventions included in the review
IB delivered as bolus or continuous infusion. Most of the included studies were uncontrolled, but where control participants were recruited, they received placebo.
Participants included in the review
Patients suffering from severe spasticity due to spinal cord injury, multiple sclerosis, cerebral palsy and other central nervous system disorders (stroke, traumatic brain injury, hereditary spastic paraparesis, syringomyelia and unspecified trauma). The mean age of the participants was 36.5 years.
Outcomes assessed in the review
1. The change in the Ashworth scores and Penn spasm scores associated with the use of IB. The Ashworth score measures muscle tone on a 5-point ordinal scale. This ranges from 1, denoting no increase in tone, to 5, denoting limb rigid in flexion or extension. The Penn spasm score measures spasm frequency on a 5-point ordinal scale. This ranges from 0, denoting no spasms, to 4, denoting spasms occurring more than 10 times per hour.
2. Successful response to treatment, overall, and categorised into 3 phases. A positive response to a bolus dose of baclofen was defined as a 2-point or more decrease in Ashworth or Penn spasm score (phase 1), the number of participants implanted (phase 2), and the total number still implanted at 1 year without any significant decrease in effectiveness or unacceptable side-effects (phase 3).
3. The changes in dosage levels of IB over time were assessed.
How were decisions on the relevance of primary studies made?
Consensus of two reviewers was required when selecting the studies for the review.