Does botulinum toxin type A decrease pain and lessen disability in hemiplegic survivors of stroke with shoulder pain and spasticity?: a randomized, double-blind, placebo-controlled trial.

OBJECTIVE

The aim of this study was to assess the efficacy of botulinum toxin type A injections in reducing pain, impairment, and disability in patients who have had a stroke with shoulder pain and spasticity.

DESIGN

In this prospective randomized, double-blind, placebo-controlled trial, adults (n = 37) with post-stroke shoulder spasticity were screened for preinjection spasticity, rated 3 or 4 on the Modified Ashworth Scale for the shoulder adductors/internal rotators and shoulder pain. After the baseline screening, 21 subjects were randomized to receive either onabotulinumtoxinA (Botox; 140-200 units), into the pectoralis major with or without injections to the teres major, or placebo (saline) injections. Daily pain ratings using visual analog scales of best and worst pain and Disability Assessment Scale for dressing, hygiene, pain, and cosmesis; McGill Pain Questionnaire-Short Form; Fugl-Meyer Scale; upper limb range of motion; and Modified Ashworth Scale scores were assessed at baseline and 2, 4, and 12 wks after injection. Primary outcomes were assessed at week 4.

RESULTS

The subject groups were well matched at baseline. Both the botulinum toxin type A and placebo groups showed decreased pain scores at 4 wks (P's < 0.05), with no significant differences between the groups found for any of the daily pain ratings (P's > 0.05). Significant improvement (P < 0.05) in change scores for hygiene on the Disability Assessment Scale was found in the botulinum toxin type A group compared with the placebo group at week 4, and there was a similar trend toward significance for improvement on the Disability Assessment Scale dressing scale (P = 0.061).

CONCLUSIONS

Although botulinum toxin type A shoulder muscle injections in patients who have had a stroke with spasticity and shoulder pain resulted in improvement in selected disability measures, the observed pain reduction was not greater than that found for placebo.