The neural correlates of bilateral upper limb training after stroke: a systematic review

Background: Recovery of upper limb (UL) function after stroke is poor with only a third of stroke survivors regaining some functional use. Bilateral training (BT) of the Uls is a promising intervention but its underlying mechanisms are unclear. BT involves practicing identical movements with both UL (e.g. carrying a box, lifting two cups) and is not to be confused with bimanual training where both limbs perform different movements (e.g. tying shoelaces). It is important to understand how BT affects both UL motor recovery and neuroplasticity in order to assess its potential as a therapeutic intervention. This is the first review of the effects of BT on motor recovery and neuroplasticity together.
Purpose: This systematic review aimed to elucidate the mechanisms underlying BT by determining (i) the effects of functional changes by BT on neuroplasticity and (ii) the relationships between changes in UL function and changes in neuroplasticity as a result of BT.
Methods: The complete holdings of 11 databases were searched up until December 2014.Trial registers and reference lists of included studies and reviews were checked. Quantitative studies of any design employing BT with both UL motor and neurophysiological outcomes involving adult stroke survivors were included. Studies not available in English or in full text were excluded. Two independent reviewers selected studies, extracted data and reviewed methodological quality using the Effective Public Health Practice Project (EPHPP) tool. The EPHPP tool was selected from four reviews. Where a decision could not be made between the two reviewers, a third reviewer was sought and consensus reached through discussion.
Results: From 41,438 records, eight studies comprising164 participants were included. The global rating for methodological quality was ‘strong’ for two studies, ‘moderate’ for two studies and ‘weak’ for the remaining four. Consider-able heterogeneity of participants, modes of BT, comparators and measures meant pooled outcome analysis was not possible. The variation in modes of BT included: in-phase and anti-phase movements, functional tasks involving objects and movements only, mechanically coupled and free movement, auditory cued and self-paced movements. The most common outcome measure used was the UL section of the Fugl–Meyer which was used in six studies; however some studies used different subsections. Functional magnetic resonance imaging(fMRI) and transcranial magnetic stimulation (TMS) were used in three and five studies respectively, employing a wide range of measures. Only five studies reported neural correlates of BT, but findings were inconsistent. No clear pattern of neuroplasticity associated with BT was apparent from this review.
Conclusion(s): Current evidence from this systematic review indicates that the neural correlates of BT after stroke are currently unknown and need to be examined systematically to understand the therapeutic potential of different modes of BT. Future research needs to identify effective training modes of BT, match the optimal mode to people with specific stroke lesions and use appropriate outcome measures.
Implications: Physiotherapists should use their clinical reasoning when using BT, considering various modes of delivery in conjunction with patients’ UL impairments and functional goals.