Effectiveness of Botulinum Toxin Treatment for Upper Limb Spasticity Poststroke Over Different ICF Domains: A Systematic Review and Meta-Analysis

Objective: To provide a comprehensive overview of reported effects and scientific robustness of botulinum toxin (BoNT) treatment regarding the main clinical goals related to poststroke upper limb spasticity, using the International Classification of Functioning, Disability and Health. Data Sources: Embase, PubMed, Wiley/Cochrane Library, and Ebsco/CINAHL were searched from inception up to May 16, 2018. Study Selection: We included randomized controlled trials comparing upper limb BoNT injections with a control intervention in patients with a history of stroke. A total of 1212 unique records were screened by 2 independent reviewers. Forty trials were identified, including 2718 patients with history of stroke. Data Extraction: Outcome data were pooled according to assessment timing (ie, 4-8wk and 12wk after injection), and categorized into 6 main clinical goals (ie, spasticity-related pain, involuntary movements, passive joint motion, care ability, arm and hand use, and standing and walking performance). Sensitivity analyses were performed for the influence of study and intervention characteristics, involvement of pharmaceutical industry, and publication bias. Data Synthesis: Robust evidence is shown for the effectiveness of BoNT in reducing resistance to passive movement, as measured with the (Modified) Ashworth Score, and improving self-care ability for the affected hand and arm after intervention (P<.005) and at follow-up (P<.005). In addition, robust evidence is shown for the absence of effect on arm-hand capacity at follow-up. BoNT was found to significantly reduce involuntary movements, spasticity-related pain, and caregiver burden, and improve passive range of motion, while no evidence was found for arm and hand use after intervention. Conclusions: In view of the robustness of current evidence, no further trials are needed to investigate BoNT for its favorable effects on resistance to passive movement of the spastic wrist and fingers, and on self-care. No trials are needed to further confirm the lack of effects of BoNT on arm-hand capacity, whereas additional trials are needed to establish the suggested favorable effects of BoNT on other body functions, which may result in clinically meaningful outcomes at activity and participation levels. (C) 2019 by the American Congress of Rehabilitation Medicine