Biofeedback and Virtual Reality in Gait Rehabilitation

Biofeedback may be defined as a process that enables an individual to observe and subsequently learn how to change physiological activity for the purpose of improving health and performance (Schwartz 2010). Precise instruments measure physiological activity and rapidly and accurately ‘‘feed back’’ information to the user (Schwartz 2010). The presentation of this information supports desired physiological changes (Schwartz 2010), specifically gait movements in this chapter. Biofeedback techniques in publications include those based on Electromyography (EMG) recordings of muscle activation or position, or force sensors that provide feedback on joint motion or functional attributes such as weight-shifting.

In recent years, technological advances such as virtual reality (VR) are being used as a therapeutic tool in SCI rehabilitation (Abou et al. 2020). VR is a computer-based technology that allows users to interact in a computer-generated environment, allowing the practice of rehabilitation exercises in a safe, standardized, reproducible, and controlled environment (Abou et al. 2020). VR comprises two types of systems according to the immersion level: (i) semi-immersive or non-immersive systems, and (ii) immersive systems (De Miguel-Rubio et al. 2020). Semi-immersive and non-immersive systems use a screen to display the environment with a low level of immersion (e.g., commercial videogame consoles), and immersive systems offer full integration of the user into the virtual environment, providing sensory inputs to the patient (e.g., VR caves, large-screen projections, and head-mounted displays; De Miguel-Rubio et al. 2020; Henderson et al. 2007). VR therapy can help practitioners provide external feedback to their patients about their performance and increase adherence to intensive and repetitive exercise training in acute or chronic SCI treatment (Levin et al. 2015; Ionite et al. 2022; Leemhuis et al. 2021).

Discussion

Several studies have been conducted including patients with chronic SCI, who performed different training programs using virtual reality (VR) (An & Park 2018, 2022; Donati et al. 2016; Duffell et al. 2019; van Dijsseldonk et al. 2018; Villiger et al. 2017; Zwijgers et al. 2024), or standing and walking programs coupled with biofeedback (Amatachaya et al. 2023; Cheung et al. 2019; Govil & Noohu 2013; Mollà-Casanova et al. 2024; Nithiatthawanon et al. 2020; Sayenko et al. 2010; Tamburella et al. 2013). Though VR training has been incorporated more often in studies training sitting balance or standing balance in people with SCI, some studies show that VR and biofeedback training can improve walking as well.

An and Park (2022) compared a rehabilitation protocol that included kicking motions to participants in a virtual soccer game in 40 people with incomplete tetraplegia. At the end of the intervention, both groups showed improvements in walking speed (10MWT), but the VR group improved significantly more than the control group (reduction in time to complete test from before to after intervention: 13.75 seconds vs. 9.45, P < .01; An & Park 2022).

Another type of walking training with VR biofeedback, Gait Real-time Analysis Interactive Lab (GRAIL), has been tested in people with SCI (Zwijgers et al. 2024; van Dijsseldonk et al. 2018). GRAIL consists of an instrumented dual belt treadmill with two embedded force plates capable of moving in multiple directions and of generating mechanical perturbations with an eight-camera VICON motion capture system (van Dijsseldonk et al. 2018). In a multicenter RCT, Zwijgers et al. (2024) randomly assigned 35 participants with incomplete, chronic SCI to either 11 hours of walking training using the GRAIL system or to conventional walking and strength training. After six weeks, both groups’ maximal walking speed increased by 0.07 m/s at post-intervention and by 0.10 m/s at follow-up relative to baseline. A small pre-post study found slightly higher gains in walking speed (0.14-0.19 m/s) after GRAIL training (van Dijsseldonk et al. 2018).

In a longer-term approach to assessing walking training with VR, Donati et al. (2016) enrolled eight people with SCI into a year-long gait neurorehabilitation program that included immersive VR, enriched visual-tactile feedback, and custom-designed exoskeletons. Among other outcome measures assessed, all patients experienced a three-to-six-point gain in WISCI scores (on a scale of 1-20, greater than 1 SD of 3.4 points; Donati et al. 2016).

The RCT of Cheung et al. (2019) included 16 participants with incomplete SCI who received standard physiotherapy; in addition, participants either received BWSTT plus biofeedback three times per week for eight weeks (active group, n=8) or passive lower limb mobilization (control group, n=8). BWSTT was performed with the Lokomat system and an EMG biofeedback system was applied to the bilateral vastus lateralis with generated audio signals if the muscle activation was less than 30% of maximal recruitment, encouraging active participation during the stance phase of the gait cycle (Cheung et al. 2019). The experimental group improved more in WISCI II and SCIM-III mobility scores, however, it should be noted that they received 40% more walking training hours than the control group; it is therefore difficult to determine the effects of the EMG biofeedback system for these participants.

Other studies, it would appear, provide evidence that biofeedback may be helpful in muscle activation when attempting walking in people with SCI. In the study by Govil and Noohu (2013), biofeedback was provided in the form of EMG from the gluteus maximus muscle. Participants (N=30) were randomized into two groups, either receiving biofeedback and gait rehabilitation or just gait rehabilitation. Both groups significantly improved from baseline in EMG amplitude, walking velocity and step length, but the group receiving biofeedback improved cadence too. In an RCT, Amatachaya et al. (2023) tested a lower limb loading training (LLLT) program with or without biofeedback over four weeks (30 min per day, 5 days per week) in participants with incomplete and chronic SCI. At the end of the intervention, the mobility improvement (10MWT, 6MWT, and FTSTS test) of participants in the experimental group was significantly greater than that of the participants in the control group (p<0.05); however, this difference was not found at six months after the training programs (Amatachaya et al. 2023).

VR and biofeedback-based physical therapy have been employed in home-based rehabilitation in order to improve access, increase adherence and uniformity in interventions, and the ability to remotely monitor and solve problems (Reilly et al. 2021). Exercise through home-based video games, such as Nintendo Wii (Nintendo, Kyoto, Japan) and Xbox Kinect (Microsoft^®, Redmond, WA, USA) have been used by clinicians to offer moderate intensity exercises (Mat Rosly et al. 2017), task-oriented training and high repetition to maximize motor learning and neuroplasticity (Levin et al. 2015), along with increased motivation and enjoyment, and the ability to be used independently by the patient (Perrochon et al. 2019). Villiger et al. (2017) tested the feasibility of a home-based (i.e., unsupervised) VR-augmented training intervention (in sitting and standing positions) for 4 weeks in participants with motor-incomplete SCI. The intervention was well-accepted by participants, and the results revealed significant improvements in muscle strength (LEMS) at short-term assessment (Villiger et al. 2017).

There remains a strong need for further well-designed RCTs investigating the effect of VR therapy on different mobility outcomes among people with SCI and providing information about VR long-term effects in order to develop robust and detailed recommendations (Abou et al. 2020; Yeo et al. 2019).

Conclusions

There is level 1 evidence (from 1 RCT: An & Park 2022) that participants who performed kicking motions in a virtual soccer game improved their walking speed (10MWT) more than the participants who performed the same number of kicking motions without the VR component.

There is level 1 evidence (from 1 RCT: Cheung et al. 2019) that BWSTT (performed with the Lokomat system and an EMG biofeedback system) provided greater improvements in WISCI II walking scores and SCIM-III mobility subscores than passive lower limb mobilization in participants with incomplete SCI.

There is level 2 evidence (from 1 RCT: Zwijgers et al. 2024) and level 4 evidence (from 1 pre-post study: van Dijsseldonk et al. 2018) that individualized VR gait training on an instrumented dual belt treadmill with the capacity to move in several directions to generate mechanical perturbations and the utilization of the GRAIL system for 6 weeks, provides significant improvements in walking speed (2MWT) and functional ambulation (SCI-FAP); however, these improvements are not superior to the ones obtained after a conventional locomotor and strength training program; in participants with incomplete and chronic SCI.

There is level 2 evidence (from 1 longitudinal study: Villiger et al. 2015) and level 4 evidence (from 1 pre-post study: Villiger et al. 2013) that lower limb training augmented by biofeedback of ankle and knee movements can improve gait and muscle strength.

There is level 1 evidence (from 2 cross-over RCTs: Amatachaya et al. 2023; Nithiatthawanon et al. 2020) that adding visual feedback relating to the amount of lower limb loading during bodyweight shifting, stepping, and OWT provides improvements on 10MWT, 6MWT, and FTSTS test in participants with chronic SCI.

There is level 1 evidence (from 1 RCT: Mollà-Casanova et al. 2024) that six weeks of visual illusion therapy based on virtual walking (quiet standing watching a video projection of legs walking on a treadmill) plus a therapeutic exercise program (including gait and multicomponent training exercises) provides significant improvements in tibialis anterior strength and in walking speed (10MWT) and walking ability (WISCI II) than a placebo visual illusion intervention plus the same therapeutic exercise program, in participants with incomplete and chronic SCI.

There is level 2 evidence (from 1 RCT: Govil & Noohu 2013) that EMG biofeedback may improve gait outcomes in patients with SCI.

There is level 4 evidence (from 1 pre-post study: An & Park 2018) that a semi-immersive VR intervention for 6 weeks provides significant improvements in walking ability (WISCI II) in patients with incomplete and chronic SCI.