The potential power of robotics for upper extremity stroke rehabilitation

Introduction

In the mid-1990s, robotic stroke rehabilitation began to emerge. ¹ Small, uncontrolled case studies eventually led to larger randomized controlled trials. Now, a Cochrane review ² and some national guidelines^3,4 recommend robots for upper extremity treatment following stroke.

A means to more intense rehabilitation and beyond

Most front-line therapists will tell you that robots are a method of increasing the number of “reps”. Repetition drives neuroplasticity and the current number of repetitions observed in practice ⁵ are simply insufficient to promote the functional recovery seen in animal stroke studies. ⁶ Users of robotic devices can make ∼1000 movements/hour, about 31 times that reported in typical inpatient rehabilitation settings. ⁵

The story on robotics, however, runs much deeper. Despite significant differences between devices and the therapy they deliver, authors continue to collapse and combine them as a single intervention during meta-analyses. ² Differences between devices may be key to understanding differing results, yet head to head comparisons are essentially non-existent. Many devices provide assistance to those who struggle to move or resistance to those with inappropriate trajectories. Some robots can provide progressive gravitational support which impacts the quality of movement. ⁷ While these two concepts are intuitively applied by therapists, a robot can accurately measure progress within and across sessions. Robotic assessments correlate with clinical scales like the Fugl-Meyer upper extremity assessment ⁸ or the Chedoke McMaster Impairment Inventory. ⁹ In our experience, however, robotic assessment fidelity is better than these simple ordinal scales.

The importance of assessment

In clinical practice, we see many patients with visual or sensory dysfunction being treated with motor-based therapies. The underlying assumption is that deficits in the other domains will improve as well. Our experience is that the recovery is often sub-optimal with this approach.

The ability of robotic tools to perform highly accurate assessments is incompletely integrated into many robotic rehabilitation protocols. Many devices carry out relatively simple unimanual motor assessments. However, stroke recovery is not simply a product of the motor system. Sensory, visuospatial and cognitive impairments can substantially impact recovery. Robotic assessment tools have been developed to quantify dysfunction in these other domains. These tools will not only help us to develop better prediction models for recovery but also understand the neural mechanisms that drive recovery.

Robotic rehabilitation will likely benefit from effectively integrating careful neurologic assessments of impairments across several domains to determine treatment plans for patients. By the very nature of their accuracy and precision, robotic devices have the potential ability to better personalize treatment than many standard rehabilitation techniques. In theory, this will lead to better outcomes, but this hypothesis needs testing.

The road ahead

At present, robots are not a panacea. Most devices lack portability. Engineering robotic devices for the hand has moved at a slower pace than the upper arm, in part, because of design complexity. Critics often challenge the cost of robotic rehabilitation. However, the largest study to date has shown a significant costs savings with their use. ¹⁰

Robotic rehabilitation is still very far from its full potential. Engineering advances are required to integrate assessment metrics into a package that can both assess across multiple neurologic domains and treat. Integration of personalized treatment that also targets visuospatial, sensory and cognitive function, rather than just motor function will likely result in better outcomes. A better understanding of the neurologic processes underlying recovery is needed to guide the appropriate development of such rehabilitation interventions. Despite nearly 20 years of progress, there is still much work to be done.