OBJECTIVES: To identify and summarise the economic evidence for botulinum toxins in post-stroke spasticity and spasticity associated with cerebral palsy. METHODS: Cost-effectiveness and cost-analysis studies of interventions for treatment of postictal and cerebral-palsy-related spasticity, in which at least one arm consists of a botulinum toxin, were considered for inclusion in the review. Medline, Embase, NHSEED and the proprietary Allergan Botulinum Database were searched up until February 16, 2004 for relevant studies. Additionally, conference proceedings of seven clinical and pharmacoeconomic organisations were hand searched for the period of 2001 to 2004. RESULTS: One cost-effectiveness (Wallesch 1997), two cost-consequence (Houltram 2001, Loaiza 2000) and two cost studies (Balkrishnan 2002, Radensky 2001) met the criteria for inclusion in the review. Wallesch presented the incremental cost per unit improvement in Ashworth scale, Loaiza presented quality adjusted life years and Houltram presented Modified Ashworth Scale together with several other functional measures. The cost analyses found that overall treatment costs were lower in treatment plans that included botulinum toxin A (Btx-a). Although Btx-a increased drug costs by between $750 and $1000 annually, overall treatment costs were lower due to the reduced hospitalisation and nursing facility admissions associated with Btx-a treatment. Btx-a treatment was also associated with fewer co-administered treatments, and resulted in lower treatment costs than treatment plans that did not include Btx-a. CONCLUSION: The addition of botulinum toxin to a treatment regimen appears to be cost neutral. There is a need for cost-effectiveness analyses using outcome measures with greater external validity than those identified in the studies included in the review. A cost-utility analysis based on clearly derived utilities is required.