NETWORK META-ANALYSIS OF UPPER LIMB EXERCISE INTERVENTIONS ON MOTOR FUNCTION, ACTIVITIES AND PARTICIPATION IN INDIVIDUALS WITH (SUB-)ACUTE STROKE

S. Tenberg^1,2, S. Müller¹, L. Vogt², C. Roth², K. Happ², M. Scherer¹, M. Behringer², D. Niederer³

¹University of Applied Sciences Trier, Department of Computer Science / Therapy Sciences, Trier, Germany, ²Goethe University Frankfurt, Institute of Sports Sciences, Sports Medicine and Exercise Physiology, Frankfurt, Germany, ³Goethe University Frankfurt, Institute of Occupational, Social and Environmental Medicine, Department of Sports Medicine and Exercise Physiology, Frankfurt, Germany

Background: Upper limb paresis is a leading functional impairment after stroke. Various exercise interventions are effective to treat these impairments. Up-to-date, there is no conclusive evidence on which upper limb exercise intervention is most effective for individuals with (sub-)acute stroke.

Purpose:​To investigate the comparative effectiveness of upper limb exercise interventions on motor function, activities of daily living and participation in individuals with (sub-)acute stroke.

Methods: This systematic review with network meta-analysis included randomized controlled trials examining individuals with (sub)acute stroke (<6 months post-stroke), active upper limb exercise interventions, and any kind of control intervention. Both main outcome (upper limb motor function) and secondary outcomes (ADLs; activities of daily living and participation) were assessed at post-intervention and follow up. Risk of bias was assessed using the Cochrane risk-of-bias tool for randomized trials. Standardized mean differences, i.e. Hedge’s g, were used as effect size estimators. We calculated frequentist-based network meta-analysis for the comparative effectiveness calculations using R package netmeta. Nonspecific/multimodal active upper limb therapy was used as comparator. Network plots were conducted for geometry of the network. P-score was used to summarize intervention hierarchy with lower values indicating a higher likelihood of being most effective. Contrast-based forest plots with standardized mean differences, 95% confidence intervals and ordered treatments by ascending P-score were generated. Results were derived from direct and indirect evidence comparisons.

Results: This review included 145 randomized controlled trials with 6432 patients and 46 different treatment classes (119 trials, 5553 patients and 41 treatment classes for network meta-analysis). Electrical stimulation combined with task-specific training (1.03 [0.51; 1.55], p<0.0001), high-volume constraint-induced movement therapy (0.86 [0.4; 1.32], p=0.0003) and strengthening [0.65 [0.17; 1.13], p=0.0114) achieved robust significant results (k = 107) when considering the sensitivity and subgroup analyses and were P-score-ranked best (0.11, 0.18, 0.28). No clear and robust beneficial effects were found for motor function at follow up (k = 38), ADL and participation at post-intervention (k_ADL= 64, k_{participation}= 14) and at follow up (k_ADL= 25, k_{participation}= 0) when considering sensitivity- and subgroup analyses.

Conclusions: This review found that electrical stimulation combined with task-specific training, high-volume constraint-induced movement therapy and strengthening were the most effective active upper limb interventions for improving upper limb motor function in individuals with (sub-)acute stroke. Future studies should address additional domains of the International Classification of Functioning, Disability and Health (like ADLs and participation) for outcome assessment.

Implications: The effectiveness of electrical stimulation with task-specific training supports the existing evidence for the beneficial effects of functional electrical stimulation as a superior intervention when compared to other types of electrical stimulation and conventional therapy. The results are valid on a group, but not on an individual level. However, if resources are available, electrical stimulation in combination with active upper limb interventions should find more application in stroke rehabilitation. Therapy alternatives may be high constraint-induced movement therapy and strengthening.

Funding acknowledgements: This work was supported by the Federal Ministry for Economic Affairs and Energy (Germany; BMWi) under Grant ZF4478606TS9.

Keywords:
Stroke
Exercise interventions
Upper limb rehabilitation

Topics:
Neurology: stroke
Disability & rehabilitation
Research methodology, knowledge translation & implementation science

Did this work require ethics approval? No

Reason: This work did not require ethics approval since it is a systematic review with network meta-analysis.